ML20134J295
| ML20134J295 | |
| Person / Time | |
|---|---|
| Issue date: | 03/27/1996 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20134J282 | List: |
| References | |
| FOIA-96-402 NUDOCS 9611150140 | |
| Download: ML20134J295 (77) | |
Text
- _ . . _ .... __ _ .__-. .. _ . _ _. . . . _ . _ _ _ _ _ . _ _ -_ _ -__ __ _ _
l 1
InThe Matter Of:
l INRE:
PREDECISIONAL ENFORCEMENT CONFERNECE l .9 PROCEEDINGS BEFORE STEWARTEBNETER, CHAIRMAN l March 27,1996 i
i l
i i l BROWNREPORTING, INC.
ATLANTA, ATHENS, AUGUSTA, CARROLLTON ROME l 1740 PEACHTREE STREET ATLANTA, GA USA 30309 (404) 876-8979 or (800) 637-0293 On'ginal File 0327nuciasc, 299 Iages Min-U4cript@ File ID:0841602638 Word Index included with _this Min-U-Scripte information in this record was deleted in accordance with the freedom of Information Act, exemptions E
/ p _ 9 G -1]- O Q jG' ' ~~
9611150140 961112 - . _ . . . _ _ . . _ . . -- .
vi m exus nonos utsunt s a tw Axl tant 1LM, WiA1RW
, j , PREDECISIONAL ENPORCEMENT CONFERNECE March 27,1996 1
j Page1
{
p) UNfrED STATES NUCLEAR REGULATORY COMM6SSON Page 2 REGION li pg p] APPEARANCES pl pl On behas of sis NRC:
141 (5) IN RE: PREDECISONAL [q R. Butcher )
)
ENFORCEMENT CONFERENCE, ) K. Lanes l lsi CRYSTAL RfVER. ) '
p) id) E. Morschoff l
131 A.Otioon l M
pol PROL cDINGSBEFORE ISI S Ebneter, Chairman lesj STEWART EBNETER, CHAIRMAN j,
lia l March 21,1996 M B. Uryc U"
,,,, S. Rkmerds
, 04) pl G. Irrero (15) 20th Floor 101 h Street ist On beher of CrystalRWor:
g liej Allerta. Georgia M B. Hk*te pF1 psi P. Board 0 81 I (tal G. Boldi 9 01 pig P.Tanguay Kath A. W5kerson, CCR.B.1381, RPR g ,, tg 12 2]
pq B. Gutherman BROWN REPORTING,INL.
p2] G. Heinon p4) 1100 SPRING STREET, SulTE 750 ATLANTA. GEORGIA 30300 D4 i 9 51 (404) 878-8979 l Del M. Weteerg, Esq.
ps) D. Stennger, Esq R. Gaddy, Esq pej M.Jacobs G. Becker
- 07) J. Vorse A. Boennd pet R.Caldwas C.Rapp
[1st D.Lanyl G. Tracy .j poj L Raghavan j J.Beas l
pt] C. Evans, Esq.
K. Clark A2) W. MP,Nuty B. Morns. Es4 pm B. Keetng L Cierk p4) 9 51 naOWN REPOR11NG, INC. (404) 876-8979 Mim-U-Scripte O) Page 1 - Page 2 4 l
,.%, .-,u.- .
4 -
.g
PROCEED 1h65 BElbRE 5TEWART EHNETER, CilAIRMAN IN RE: l
. . March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE 4
i j Page 3 Page5
- pl MR. EBNETER
- It's a couple minutes after l gy This case is a very complex one.It i
ni 8:00.We'd like to get staned. Pat,is your staff pi involves not only the facility license but the pt here? pi license of several of your operators.All of those
' j pi MR. BEARD: Yes. Mi are under our purview also, and we will deal with l re MR. EBNETER: Gibson and Ellis,all your is; both of them partly in combination in this meeting I ist staff's here? . Ist and in individual sessions with the operators. !
ga MR. GlBSON: Yes. pj This meeting, however,is focused on l isi MR. EBNETER: Good rnorning.1 want to mi Flonda Power Corporation's license and the l pl welcome you to Region H.I'm Stu Ebneter, regional pi management of the facility and its impact on the i om administrator for Region H of the Nuclear pq actions that the operators took in their licensed py Regulatory Commission.Of course Region H has on activities. Facility management of course is na jurisdiction over licensed nuclear activities in the pri ultimately trsponsible to assure that operations and psi southeast United States,and Crystal River falls psi activities comply with the regulations and any pq within the purview of that region. pq commitments that you have made to us applicable to pq This is a meeting this morning between pq Crystal River 3's license,and that is the main A ps; Florida Power Corporation and Region H,and the ps; focus nere: Florida Power's organization apparent on purpose of the meeting is to discuss some apparent on inability to comply overall with these regulations.
Os) noncompliances with the Agency's rules and psi We have numerous inspection reports and poi regulations is.This is a predecisional enforcement poi investigation reports on the docket,and there are
- pq conference,and the purpose,again,is to discuss pa still some issues with those.The staff has pq with you some events over the past year to two years pu formulated some questions which will come out as na that have occurred at Crystal River 3,to discuss na part of the presentation.Mr. Al Gibson,our -
ps) the facts and circumstances surrounding those,and na Director of the Division of Reactor Safer" will pq to try to resolve some conflicting issues that we pq essentially run the conference, but I can't pq still have as a result of some of our inspections psj overemphasize to you the importarr of this Page 4 Page 6 01 and investigations. tu conference.
pi This is a closed meeting,and it is being p) We also have today with us the Director pi transcribed. I should caution you,if you get pi of the Office of Enforcement, Jim I.ieberman,we have 3 si questioned by the staff and if you don't know the si representatives from NRR,and they'll allintroduce 3
rsi answer, you should say you don't know.You should tsi themselves.We have a member of the EDO staff, we (si not try to get into a situation - you know what Fm p) have the Office of Congressional Affairs represented gi saying. lf you don't know, we'll tell you to put it in and the Office of General Counsel,to say nothing of tel on the record and request you to put it in writing, - ist Region 11 counsel, Ms. Evans.
pi of course, but don't get yourself into a situation pi The it. sues before us are broad,and my og where you're trying to talk about something you pq view is basically these actions that the operators
, nu don't know about.Those issues frrquently do come ny took were unacceptable.These operators were 02: up. Da functioning as your representatives.They worked 4
psi This meeting is predecisional.That is, psi for you even though they are individually licensed.
pai there has been no specific decision made yet with pq Their actions represent your philosophy,and this is os regard to enforcement,and the meeting is for ps) my summary because they work for you.Either they psi Crystal River to present any facts,arn of their psi were working outside of your expectations or you d
on reasoning regarding the issues that a:e on the on have as managers created an environment that a.; owed ps; table.You should also point out any mistakes or pai them to do this,and we would like to hear about pq errors that you may have seen that are in the NRC poi that,and I'm sure you're going to address it, pg documentation and inspection reports or other pq Those practices,to violate the rules and pu material that you may have presented to us in the pu regulations of the Commission to prove a safety
, na past.You should also provide any new information pa point,were not acceptable,and there's no excuse to psi that we have not had access to,anything that would psi violate the law to prove a point to someone.The 9q help us reach a fair and accurate determination of p,i operators had many options available to them other psi what actions we should take. ps than pursuing this course of action.lf they didn't Page 3 - Page 6 (4) Miss-U-Scripte BROWN REPORTING, INC. (404) 876-8979
e e. raut.42.vaava ma usu. s a c un4 4.auca ca, tatattute - l
, , PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 7 Page 9 {
In know those other courses of action,that's your gy MR. BEARD: Pat Beard, senior vice I pi responsibility,too.But there's clearindication m president,nuclearoperations.
}
pi to me that rnanagement,if you didn't know,that you pi MR. HICKLE: Bruce Hickle, director of p1 should have known of these types of practices.It's pj plant operations.
rst your job as managers to perform the oversight in is MR. BOLDT: Gary Boldt, vice president of pi that control room and to make sure that your staff pi nuclear production.
m is complying with the rules and regulations of the m MR.TANGUAY: PaulTanguay, director of tsj Commission.
, p1 nuclear engineering and projects.
Ist That's enough of my lecture IW yct p MR. KELLEY: Iarry Kelley, director of og want to get on with the rnecting.We'll go through pq nuclearoperation and site support.
og some introductions.I'd commented that the meeting on MR. GUTHERMAN: Brian Gutherman, manager, na is being transcribed. Al and I have discussed na nuclearlicensing. 1 psi briefly the conduct of the meeting.It's probably pq MR. HALNON: Greg Halnon, also licensing pq going to be long,and I want to give you the pq rnanager.
pq opponunity to make your story, your presentation, eq MR. BUTCHER: Ross Butcher, senior l psi so what we've decided is we've got it broken into psi resident inspector on Crystal River. !
on these apparent noncompliances.We'lllet you get on MR. LANDIS: Kctry Iandis, chief of pg through each one iry.lividually,the staff should hold l og reactorprojects, branch three. l psi their questions to the end of that one violation, og MR.JAUDON:JohnsJaudon, deputy pm and then we'll get to the questions and answers. pq director,dhision of reactor safety.
pq They're interrelated, those things, so pq MR. MERSCHOFF: Ellis Merschoff, director pa there will probably be some redundancy in the pa of reactorprojects. .
psi questions.If we just let you make the presentation pg MR. GlBSON: Al Gibson, director of the pq for an hour or two and then go on,I don't think it pq division of reactor safety. I pq will be effective, but it's also not going to be pq MR. MORRIS: I'm just a general observer, Page 8 Page 10 in effective to have everybody jump in and say,What in Bruce Morris.
mi about this or that. So I think that's how we'll m MS. EVANS: Casulyn Evans,regionai m proceed. pi counsel.
pi As we get toward the end of the pi MR. BEALL: I'mJim Beall for the Office tsi conference the staff will probably want to take a (51 of Enforcement.
tal caucus.The standard procedure that they give me is pi MR.TRACY: GlennTracy, EDO staff.
m a ten minute caucus,and my guess is it won't be a m MR. CALDWELL: Bob Caldwell, project pi ten minute caucus,it will pobably take longer, but p engineer, si we'll break and come back and have any additional pi MR. RAPP: Curt Rapp, regional inspector.
~
pm questions and then close the meeting.I do pq MR. LANYl: David Lanyi, project pu appreciate your coming in. I know this has been a 99 engineer.
pa very large job for your staff and the NRC staff. pa MS.BOLAND: Anne Boland, senior om Everybody involved has had a great burden to bear on pm enforcement specialist.
pq this. pq MS. KEELING: Betsy Keeling, Office of psi With that,we'll go through the 04 Congressional Affairs.
psi introductions, and then I'll turn it over to Bruno, psi MS. CLARK: Lisa Clast, Office of the on and then we will discuss with you the enforcement pn GeneralCounsel.
psi policy. nsj MRIWEINBERG: SandyWeinberg.I'm an esi MR. LIEBERMAN: !'m jim Lieberman, the poi attorney. ;
pm Director of the NRC Office of Enforcement. pg MR. STENGER: Dan Stenger ofWinston &
99 MR. URYC: Bruno Uryc, Director of Region pq Strawn representing Florida Power.
ria II Enforcement Staff. pa MR. GADDY: Rodney Gaddy,corporste pai MR. RICHARDS: Stuart Richards. psi counsel for Florida Power.
pq MR. IMBRO:I'm Gene Imbro. Project pq MR.JACOBS: MarkJacobs, manager of psi Director for the NRR. pq strategic communications.
BROWN REPORTING, INC. (404) 876 8979 Min.U. scripts (5) Page 7 - Page 10 4-
l PMOt4 MilNGS BEFORE STEWART EBNETER, CHAIRMAN IN RE:
{, . March 27,1996 PREDECISIONAL ENPOEdFFMRMT CONFERNECE Page 11 Page 13 pl MR. SECKER: Gary Becker, make-up tank pl and a common understanding of the agmnenner of pi issue manager.
pi the issues and the need for lastmg comprehenssve pi MR.VORSE: Jim Vorse,special agent, m corrective action.
pi Omce ofInvestigations. pi The apparent violations discussed at this 3' (q MR.RAGHAVAN:I Raghavan, project m conference are subject to further review,and they M rnanagerforNRR. m may be subject to change prior to any resukang m MR.MCNULTY: Bill McNulty, director of m en'orcement action.It's important to note that the m investigations,RegionII. m decision to conduct this conference does not mean m MR. EBNETER: Mr. Morris,would you m that the NRC has determined that a violation has pq identify who you're amhated with? pg occurred or that enforcement action will be taken.
py MR. MORRIS: I represent the operators at pq l'd also like to note at this time that the na the plant,actually four operators.I'm an attorney na statement of views or expression of opinion made by pg in Atlanta. pg the NRC staff at this conference or the lack thereof pq MR.EBNETER:Bruno? pg are not intended to represent final Agency pq MR.URYC:Thank you,Mr.Ebneter.I'd ne determinations or beliefs regardag this saatter
^d~ s pg like to take just a few minutes to 30 over the NRC pg Following this conference the regional on enforcement policy and procedure for you. pa administrator,in conjuncuon with the NRC Office of pai AAer an apparent violation is pq Enforcement and other NRC offices, will reach an pg identified,it's assessed in accordance with the og enforcement decision,and this process should take pq Commission's enforcement policy which is published pq about four to six weeks to accomplish, an as NUREG 1600.The assessment of an apparent pg Finally,if the enforcement action pa violation involves categorizing the apparent pq involves a proposed civil penaky,the NM will ' {
i pg violation into one of four severity levels based on pg issue a press release 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after the -nforcement pq safety and regulatory sigruficance. For cases where pq action is issued.Thank you.
pq there is a potential for escalated enforcement pq MR. I.JEBERMAN: Good morning. My name is Page 12 Page 14 pi action, that is, where the severity level of the vi Jim I.ieberman, and I'm attending this conference 1 pi apparent violation is categorized as severity level ta today because it involves significant regulatory I pi one,two or three,a predecisional enforcement m issues.1 want to c.=i'=h at the outset,as Mr.
Mi conference is held. p1 Ebneter did,that the performance of your operators tsi There are three primary enforcement tsi on the evenings of September 4 and 5, which we m sanctions available to the NRC,and they are notices m recognized did not perform acceptably,is not the m of violation, civil penshics,and orders. Notices m primary focus of this predecisional enforcement m of violation and civil penalties are issued based on m conference.While we're concerned about them,our 3 m identified violations. Orders may be issued for m main focus today is on the pesformance of Florida og violations or,in the absence of a violation, pq PowerCorporation.
pq because of a significant public heakh and safety pq As Mr.Ebneter rnentioned,we have a 02 issue, pai number ofissues to discuss today.I want you to pg Now, this predecisional enforcement usi know six of the concerns that I'm especially 1
j og conference is essentially the last step of the pq interested in. J pst inspection and investigation process before the psi My first area of concern focuses on the psi staff makes its final enforcement decision.The og performance of the operations department that on purpose of this conference is not to negotiate a un permitted routine exceeding of operating limits in og sanction.Our purpose here today is to obtain og several cases for more thar an hour;that insisted pg information that will assist us in determamng the og the plant operate on a curve knowing operators pq appropriate enforcement action. pq questioned the safety in so doing and,rather than en We would expect to develop information an insisting the matter be promptly resolved, taat that would provide us with a common understanding of pe instituted frequent monitoring of true performance l psi the facts, root causes and missed opportunities pg to see who did not operate on the curve;that did j 941 associated with the apparent violations,a common pq not investigate the root causes of the safety usi understanding of corrective action taken or planned, psi evolutions and perform a human factors review of the Page 11 - Page 14 (6) BElm-U4csipte BROWN REPORTING, INC. (404) 876-8979
. _ _ _ _. _ __ ___ _ y
m at: raveccutaus uctvan 3 A twMu tuaL1tu, u1Auu1AA
, , PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 15 Page 17 to events;that did not interview each member of the p Ilook forward to your explanations m operating crew to understand what happened and why; pi today,because I hope they'll be helpful for the NRC pi and that apparently did not investigate to determine ni to make an appropriate enforcement decision.With pi if there were other unauthorized evolutions to pl that,let me turn it over to Al Gibson.
j te obtain data with either the make-up tank or other to MR.GIBSON: What rd like to do todayis 1
m systems. pi first call your attention to a handout that you m My second area of concern is with the m should have been provided that includes the agenda mi communications between engineering and operations m that you see and a brief summary of the notices of in that led to the failure of the engineering pi violation that we are considering and then a more pq organization to bring forward knowledge of the og detailed descripdon of that apparent violation og September 4th evolution. pg which really has more information than we could get pa My third area of concern is the na up on a slide that could be viewed throughout the na environment for raising concerns such that a manager na room. I would like to go through each of the pq involved in licensed activities felt compelled to pq apparent violations in summary with you.
pq use the employee concern program rather than bring pq Td like first,though,to reiterate sorne j og his knowledge of the September 4th evolution to the
' psi of the thmgs the others have said. I would like to pn attention of operations and that Florida Power na encourage you,if you believe that there is a pai Corporation did not prompdy and aggressively pq mistake or an error in any of the repons that we pq resolve a safety concern raised by a senior reactor pq have sent to you or in the proposed notices that pq operator.
pg we're discussing today,to bring that to our pq In that regard,I have to presume that a pq attention.
na licensed operator does not normally raise concerns ra I would offer that,although rm going -
pa that operating limits are non-conservative,and when psi through the apparent violations in a particular pq that occurs I would have presumed that a licensee pq order, you can change the order as you wish.We pq would consider it a high safety priority to pq will hold our questions and give them to you after Page 16 Page 18 gg resolve.However,in this case it appears that the pl each of the four apparent violations have been pj issue percolated without senior managernent assisting gj discussed.We will caucus, as Mr. Ebneter said, pi and getting it promptly resolved. pi when the presentations are complete and the p1 Fourth is the performance of the pj questions are complete.
to engineering department.Notwithstanding repeated tg I would like to remind the NRC staff that pi opportunities,they did not aggressively pursue the mi if you believe there are errors,if you believe m matter, promptly resolve it,and recognize it was a m something that FPC is stating is not correct,we pi design basis curve. Even after it was recognized as ist would like for you to raise th t before we caucus.
pi a design basis curve,it was not initially properly pi I don't want to go to the caucus and then somebody pm corrected. pq say,Well,I don't agree with that,that wasn't 99 My 5th concern has to do with the I.ER 0g really correct. rd rather hear it here so that FPC pa preparation,which was done with engineering's na has an opponunity to hear it.
ps) assistance but did not describe the evolution on the psi MR. EBNETER: Ict me stop a minute. Jim t g 4th nor stated that every time the plant operated on og 1.ieberman just mentioned something to me, and I want ne or exceeded Curve 8 the plant exceeded its design pq him to rnake the comment right now.
pq basis. psi MR.LIEBERMAN: AfterAl gets through pn Finally,rm concerned why Florida Power on with describing the issues of his concern,if psi Corporation did not begin a realinvestigation of psi Florida Power wants to caucus to make sure that og this matter until after the originally scheduled poi they're satisfied with their presentation based on pq conferences were canceled almost a year after the sq any of our opening remarks, feel free to do so.
pq issue arose. frankly,in my view,which I sincerely pq MR. BEARD: Thank you.
pa hope is wrong,the actions of this company seem to pa MR. GIBSON: With that, r11 proceed and pst dernonstrate an intent to place the focus on Mr. psi go through what we see as apparent violations at pq Fields and his crew rather than on the significant pg this point.
pq shortcomings of Florida Power Corporation. psi The first item Til discuss is an BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripte (7) Page 15 Page 18 4
eRotttu1%d littORh 51LWARE EHNEILR, C11 AIRMAN IN RE:
March 27,1996 PREDECISIONA1. ENFORCEMENT CONFERNECE PaDe 19 Page 21 pi apparent violation ofTech Spec 5.6.1.1 for failure p1 We believe that, had you performed this si to follow procedures on nine occasions.During the pi review, you might have identified a violation that pi period ofJune I through September 5,1994 OP-103B, pi r11 discuss next,an example of failure to si Curve 8 was exceeded by more than one-half psig for pi translate design requirements into plant rsi continuous periods of more than 30 minutes on these isi procedures. Also, an NRC inspection in April of '95 l m nine occasions.Furthermore,the alarm response m found the same problem existed in a fire water m procedure, AR 403, was not followed in that tn storage tank.
m operators failed to take action to reduce make-up m With that,rd like to move to the final m tank pressure on these nine occasions. m violation, which is a violation of Criterion HI,an pq The second violation that we would like pq apparent violation of Criterion III for design vil to discuss is an apparent violation of pu control.This apparent violation includes four na 10 CFR 50.59.On September 4th and 5th,1994 na examples where you apparently did not adequately on operators failed to comply w th 10 CFR 50.59 when na translate plant design information into plant pq they conducted tests not described in the FSAR of 941 procedures.
psi make-up tank level and pressure without a written as The first example is Curve 8,which we've pai safety evaluation to determine if an unreviewed pai talked about previously.It did not provide control j pn safety question existed.The limits of OP-103B,the on to prevent gas binding of make-up pumps.The second 1 pm requirements of OP-402, and the requirements of pai example is that as late as February of 1995 EOPs. I poi AR-403 were violated during these tests. Subsequent pai Emergency Operating Procedures, directed operators pq analysis revealed that an unreviewed safety question pq to swap low pressure injection suction from the an did in fact exist. pq borated water storage tank to the reactor building pa T: e third violation is a violation of pa at a borated water storage tank level of five icet. '
psi Criterion XVI. Criterion XVI of 10 CFR 50, Appendix pa This level was too low to prevent gas Isinding of the pq B, requires corrective action for conditions adverse pq HPI pumps due to vortexing of water in the borated ,
psi to quality.The first example is Problem Report psi water storage tank. I i
Paga 20 Page 22 pj 944149 dated May 10,1994 which identified a pj The third example of the apparent ra concern with the accuracy of Curve 8. An pi violation ofAppendix B, Criterion IIIis that from l pi engineering evaluation datedJune 4th,1994 found no pi April 8,1993 through March the 22nd,1995 a pl errors in the curve or its supporting calculations. 91 condition existed such that during post IOCA j isi and subsequent evaluations in response to a later tsi operation with one 1.PI pump supplying two HPI pumps j m problem report,944267, dated September 7, '95 tai and with the HPI crosstic valve open the reactor i m found significant errors in the curve and the a building sump water level could be insufficient to j isi calculations. m provide adequate net positive suction heads to the I m The second example of Criterion XVI m 1.PI pumps. '
'- pq refers to corrections to the curve that were pq Finally,the fourth example is procedure on subsequently issued. Revised guidance and curves py sp.300, requiring 35 feet of water to be maintained pm issued on three occasions between September 9,1994 na in the fire water storage tanks.Because of the psi andJanuary 30,1995 were not adequate to prevent psi elevated suction point in these tanks,35 feet of a
pq operation outside the plant's design basis, pq water was not enough to meet design basis psi The final example of Criterion XVI is for psi requirements.
pq failure to take timely corrective action on an event ps In sumrnary,that's the four violations on that you had previously described in 1.icensee Event pn that are apparent to us that we would like you to 0:1 Report 92003.You may recall this was a licensee psi comment on.With that,ril turn it over to you, !
voi event report that described your finding that some nel Pat.
pm of the fuel in the emergency diesel fuel oil tanks pq MR. BEARD:I have one thing rd like to pq was not usable because it was below the elevation of pq caucus with my peopic on, probably Gary Becker and pa the suction pipe in the tank,and as part of your pa PaulTanguay.
- zst corrective action plan you stated that you would psi (Discussion ensued off the record.)
pq review other tanks for this same condition.The pq (A recess was taken.)
ps borated water storage tank was specifically listed. psi MR. BEARD: let me start off by saying l Page 19 Page 22 (8) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
.- ,l l
l
. ,4.u.
y 4 csus.o Losa.tuJ ud-4 vau, .S A AAuu LD,%L A AA, LI AM11.WLN l PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 i Page 23 Page 25 pl that we're glad to be here.You'd probably say,Why ty in a recent letter to DOL.We realize that pi is he saying that,and the answer is we're glad to m complicates this wnole thing,but as we go through pi be here and settle this thing with you,take our pi this we will present management's position and side pj medicine and get it behind us.We're not glad the pj of the story.
(q apparent violations occurred,but it's been a I'd like to briefly review the agenda; is pi catalyst for change, we've learned from the si you have a copy in front of you. I'm going to go m experience,and we're better off As we go through pi through the chronology next. I think it's important si this,we intend to demonstrate that.
pi right up front - Mr. Ebneter has said it's complex pi We heard very carefully the words of Mr. m and it's been ongoing for a while - and I think it pq Ebneter and Mr.l.iebennan.We recognize that the pq would be beneficial right up front to go through the pq overriding issue here is not what the operators did pq chronology of the key things that happened and when na or didn't do or what calculations were right or na they happened and,in so doing,I'll also give a few psj weren't right or why they weren't right.The issue pai key facts that will be elaborated on further by Mr.
pg here,as you've said,and we recognize it and we pg Hickle and Mr.Tanguay as they discuss the issues, pq intend to address it,is what was our organization pq including management oversight.
- pst doing then,what was management doing,and we will pq Right after I finish here, though, we're pn address that. i on going to have Greg Halnon just briefly show you the ps; In fact, we plan to demonstrate that it ps; make-up tank system and a couple of things that om was not an environment where people were expected or pq might be useful,and then of course I'll give a pq allowed to do what they wanted to.Mr.Lieberman, pq conclusion.We have copies of our overheads, and pg you made the statement that you hoped that your ;rn we'll provide those to you at the end of our j pai perceptions were incorrect, and we think that they pa presentation. I also have a Ivek that I used in .
psi are to a large extent,and we intend to demonstrate paj preparing,and I'llleave a copy with Mr.Ebneter. j pq that.We accept the apparent violations, that's not pq It's that thick binder over there of correspondence pq an issue:of course management accepts ps in chronological order that reflects what I'm going Page 24 Page 26 pi responsibility for everything that happens at our ni to talk about next.We also have,although I can pi plant,and that's not an issue. Instead,our focus pi see now that the print is not large enough for you pi will focus on the root causes,particularly the pi to see it at the back of the room, but we'll leave pi management issues,and corrective actions to address si it up here,the chronology.
isi and prevent recurrence. !
tg First,a little background.In the pi We will of course include the concerns pi '91.'92 time frame we received two IMPO {
m that you have with management,and we will also m evaluations,and in both cases they arcommended in a re address what we had ah.ady in our rninds felt were si finding, which is stronger than recommending, that l pi three key questions:One, why did the make-up tank pi we follow chemistry guidelines with respect to pq test happen; secondly,why didn't FPC management og hydmgen concentration reactor coolant.which was 22 pq know of the Septernber the 4th test earlier and tell on to 45 cc's per kg.We had been running somewhere na you;and third, why did it take so long to fix the pri between 10 to 15.
psi technicalissues associated with the make-up tank. l psi During this same time frame you'll recall pq As we go through our presentation,I think those pq that the industry had a lot of focus on fuel pq questions will be answered, but just in case,I will psi failures.The reasons for that,one of the reasons pq specifically answer them at the conclusion of our ps) was corrosion due to oxygen,and therefore this on presentation. vn issue was neceiving a lot of focus, and pai This conference of course is complicated, og apprtipriately so.
pq and I think it's reflected in some of the remarks pq We committed in '92 to meet the guideline pq that Mr. Ebneter rnade and Mr. l.ieberman made by the pq recommendations, and in '92 and '93 we had a series pu fact that we have terminated two operators,and now pq of meetings with engincenng operations and others pa we have an adversarial condition between the company pri on how to do this.At our plant, due to our design, psi and these operators. As a result of that, they've psi which is somewhat unique compared to the other B&W pq made a number of statements reflected in pq plants with a make-up system,it wasn't an easy pq correspondence to you and in inspection reports and psi thing to do. And as we went through this we found BROWN REPORTING, INC, (404) 876-8979 Min-U-Scripts (9) Page 23 Page 26 4
PROLLEDL%511El ORE 6thWARI LiiNETER, C11 AIRMAN IN RE:
, Atarch 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 87 Page 29 I
pl out that, every time we came back and wrestled with pl gave them.That's wrong.We should have given them p1 it,it was harder than we thought.But we're not gi from the belnnning an operating curve with plenty of pi using that as an excuse. pi margin such that, as long as they stayed there or p1 Anyway,the results of those efforts ni below and during normal evolutions the tank drains m culminated in April of '93 - wc!!,I'll just use m down,that it didn't cross the design basis curve.
- si this one.We provided the operators the curve,the tsj But we didn't do that.
m variabic pressure temperature curve.'lliere was a m As a resuk of these anomalies a problem j m curve prior to that,but it was sort of up here.We m report was written, Problem Report 94-149.During m just had a flat line at 15 pounds psi, so just keep m the period ofJune to September there were a number ne pressure below that.Very simple, period.That's pq of management operations and engineering
- og all their guidance was,and it was easy to follow. pq interactions.In some areas I've heard that no one na flowever,we were only keeping hydrogen in the va was involved in addressing this.That's not na coolant between 10 and 15 cc's per kg, so we gave in na correct.There was a lot ofinvolvement.Was it pg the alarm.We asked that they strive to meet the 04 effective in fixing the issue before Septernber the pq higher hydrogen pressure and therefore to get more pq 4th? No.It wasn't effective,but there was plenty ps; cc's per kg. psi ofinvolvement,and you'll hear some more about on Now, that did impose an additional on this, pq operator burden.We had to manipulate things about og in fact,as we now know, the curve was og five times a day to do that.Yes,it was a burden, poi reviewed inJune '94 and erroneously reported by pq but it was achievable,it was doable. Particularly pq system engineering that it was okay,and it wasn't.
99 starting aroundJanuary '94 we put increasing 99 I mean, at that point there were two things wrong.
pa emphasis on it.The operations manager was pa We didn't know it was a design basis,an ' Jiere were '
psi involved,and we felt that it could be done,and it psi some terms and what have you that were not correct.
pg was being done,and we had started as we began to pg There were in fact a series of documented meetings psj discuss what else we could do to alleviate the pq between operations and engineering.~niere was a Page 28 Page 30 pi operator burden.These discussions started as earlY p1 formal meeung on August the 8th,1994 with m asJanuary '94. pi operations and engineering followed by a management pi The next, I think, thing of interest is pl meeting with the plant manager,the manager of pj in May of '94. During a refuel we ran a 81 operations, Mr. Halnon, and a manager of nuclear tsi surveillance procedure which called for a full flow sq plant technical support. Also,during these tai test of our high pressure injection, which is also tai meetings and as we were addressing Problem Report m our make-up tank system, so we have a system that g 94149,we more and snore realized the complexity of m has two purposes,again unique to Crystal River,and Isl the system interactions,the competing operation m during that test several anomatics were noted.The m requirements.You'll hear a little bit more about pq level at the beginnmg dropped rnore quickly than was og that.
pq expected.There were some evidence that the make-up 09 The next thing 1 think,of significance na pump,or high pressure injection pump. same thing, na occurs on September the 2nd.The manager of nuclear na were cavitating.and as the system level was going pa plant technical support, Mr. Campbell, sent a memo og down in the make-up tank and the pressure was og to Mr.Hickle,the plant manager,and this memo, psi plotting, the operating crew noted that the points pq contrary to what the operators that we terminated psi as they were plotting didn't appear to follow the psi would lead you to believe, was not an ultimatum.
pn shape of this curve. on That was absolutely not the case.Nor did anyone psi l'Il note at this point, hot to belabor psi else tell this crew that it's going to be closed poi the point, that this curve is a locus of calculated poi out,you have one last chance to do somethmg.
pq points per design basis as we now know design pq That's not true,and we can document that.
pg basis. Any time you drain it down other than for 99 Instead,the memo focused on how could we pa I.OCA design action conditions,the fact that it pa further relieve the operator burden through plant asi doesn't follow the shape of this curve exactly, psi modifications. And I might add that most of the pg there's no reason why it should.The real issue is sq memo was focused on the operations suggestions from l psi and was that this was a design basis curve that we pq the August 8th rnecting.llowever,the memo again l l
Page 27 - Page 30 (10) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979 I I
1
.%a.. -w . *~ .%-a.-e 1
i
)
a tua exocteutwa uuunt 3 t emuu tratit (, utulotu RREDECISIONAL ESTORCEMENT CONFERNECE March 27,1996 Pago 31 Page 33 pi stated erroneously that the operating cutre appeared ni thought it was an unauthorized test. Bruce and 1 m to be okay,and that was wrong.I'd also note that m were so angry, we wanted an ameliorating influence pi the rnemo also reflected that there was another p1 if that was needed.
p1 remaining corrective action to Problem Report 149 pj The purpose of the committee was to Isj which said the curve would be reviewed again by Isi review the evolution,to address crew perfortnance, m Septemberthe 30th. Is) and provide recommendations thereto.They concluded m Nevertheless,the make-up tank evolutions m that it was indeed an unauthorized test.Now, rei occurred on September the 4th and 5th.The crew,as rei remember,on September the 5th we didn't know about m a result of that, presented data that culminated in m the 4th,and I'll get to why we didn't,and they po) Problem Report 94-267. At this point I wasn't going na developed recommendations to address the crew 09 to talk at length on why they did it,but let me say og performance and the issue,and Bruce will talk about va a few preliminary words because we're going to get prj those,and to some extent,because the psi back to that.In our view,there was no excuse for ps) recommendations were very good and we completed all 041 doing it,it was not in accordance with our p,i of them, and address the principle issue of human ps) expectations,and none of the other crew shift psi performance and the principles involved.
psi supervisors would have done it.We're convinced of psj Now,in hindsight, and I would agree with on that;they told us.We don't think that management pn you or anybody who says it, our own investigation paj had any inkling that they were going to do it.It psi later came up with this conclusion.lt was a good poi was done on a weekend. Hydrogen pressure was Ipe) initiative, but there's one thing wrong. In our pq normal. In fact,it was 30 cc's per kg.There was lpq zeal to deal with this thing,the human performance p9 no reason to be doing anything,but they did it,and 99 issue, we circumvented our regular pmcess of pa we'll come back again to this later.I'm going to pa writing a problem report and doing a root cause,and .
maj answer that question fully in the end. psi we should have done that, written a problem report p41 On September the 13th the plant manager p ) about why did they do it,and we've subsequently psi talks to the system engineer that's been involved in ps) done that.If we had done it then,rnaybe -I mean, 1
pj this,and the system engmeer was complaining that, ni you can hypothesize all you want to - we'd have m " Hey,the operators ran this evolution to get data I m found out about the 4th.But we Udn't.and the l pi and they should have had me involved",and that I ai t reason in hindsight,is because at the time the p) keyed the plant manager that,"What,they did what?" pi operators appeared to be very forthcoming, sincere, ist Now,why didn't we ask on September the 7th,"How y "Yes,we screwed up,we're sorry,w: meant well,we m did you get the data"? We can't figure that out. I ej t were trying to help resolve this issue',and we m But on September the 13th the plant manager said, l m accepted that.We beheved them.
m " Wait a minute,this may be an unauthorized test." l Isj Problem Reports 149 and 267 were still in m He notified the senior resident inspector and the m place to address the technicalissues,and so we pq branch chief and he convened a management review pq mistakenly did not see the need to do more at that pu cornnuttee on the 15th.This is a new initiative. pq time.
pa We had not had management review committees before. 'p ri As we proceed,from September thmugh ps) Bruce Hickle told me that pmbably on the losj December the corrective actions from the management 041 14th or so, he told me that he was going to review !n 1 committee, Problem Reports 267 and 149,there were ps) this management review committee. Gary Boldt at l psi still some things there.We were working on it.
pej this time was attending an IMPO meeting,and I said. 'psj The corrective actions from the MRC included that we pn "Yes,I like that idea,let's do it."We realized !nn counsel all the operators.1 personally talked to pai the principle involved. In fact,as I recall- inst the shlft supervisors.We reviewed other operating poi pmbably on the 14th,in talking with Mr.Hickle,I lpej curves to see whether we were operating close,and poi used the word Chernobyl myself.We said,"We've got laq there might be an issue with operators getting in 94 to do something different on this, we've got to pu trouble.We looked at that.We did a number of pa express management attention early, we need to focus pa things.We were continuing to look at the technical ps) on crew performance and the human performance 1931 issues, working on the calculations and what have p41 issues."We wanted the management review committec !p41 you.
ps) to do that and also confirm whether or not they lps) in fact,as a result of those efforts,on BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (11) Page 31 - Page 34
~^ ' ~ '^
PRO ( 1 I li1NGS 111ilDRE STEWART liBNETiiR, C11AIRMAS IN RE:
4
. March 270 1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 35 Page 37 pl November the 16th at a meeting which I attended a vi Two things came out of that.As part of l m presentation was made,and we determined that Curve a our evolving corrective action to have more i
la 8 was in fact a design basis,and that resulted in pi efficient management processes, we had canbarked on a p) our 1.ER 94009. Mr.Ueberman made a comrnent 7 pi process where key issues were so designated,an m cartier about somethmg along the lines of,"Well, ist issue manager assigned,and the key issues generally mi why in 149 didn't you recognize that there'd been pl were those that involved complex interactions.
4 m other possible design violations" nd we did.1 la usually between cross organizations,an elemciu of d
ist just read the words in our caucus.We realized pl management that's a challenge and that we needed to in there may have been other- we in fact were on the pi do better.Then,in addition to the issue manager, og curve in other occasions and we had evidence that pq there was an issue sponsor who was a director of Mr.
- og there were other possibic violations of design 09 Boldt or myself.
pa basis.
na We made Mr.Becker the issue manager for j
! pg We of course notified you within a one og all the make-up tank techrucal issues, pulled it all v41 hour report that same day and wrote the II.R.And
- 04) together in one action plan, and I believe on the pq let me say this, but this is something we're sq first of February '95 we further modified the
! pq certainly not going to harp on.The safety pg operating curve,which we had moddied previously, on consequences of September the 5th were something on once on September the 9th and once on September the i psi like three times ten to the minus eleventh core psi Sch,and we gave the operators a reduced curve and, og damage frequency. pq as has been pointed out,we don't know that was pq Now,the next thing that I think is pq still good enough.But finally.on February the 1st l pg relevant is that on November the 22nd there was an
' ny we gave them a seven pound offset,and that has pa FPC/NRC management meeting here in Atlanta. Gary pa subsequently proved to be conservative.
gs: Boldt represented us. Among other things, make-up pg Also,1 will note for the record that i
pq tank evolution was discussed. And by the way, back pq after September the 9th there had been no further psi when I said when we learned on September 13th of an psi alarm violations of ariy limit or cur, or any alarm Page 36 Page 38 j pi unauthorized test and we notified our senior and our ei with regard to make-up tank evolution, September the l m branch,there's one other thing that we, Florida m 9th,1994.In February we also established what we l 4
p1 Power, failed to do.1 failed to call Mr. lbneter, pj call our 49 step management corrective action plan pi and that was my failure. Mi as follows on to our self-assessment that was
~
ist Anyway, after the November 22nd meeting sq conducted in December of '94 that I mentioned i si we sent a letter to you of December the 2nd pl earlier.1 mention this because we think it's very
- m documenting what we did in our management review m pertinent that these actions address the broader im comnuttee meetings,what actions we were taking and pi underlying factors reflected in the apparent pl had taken and were taking. At the same time,in pi violations.
pq fact on that same day,I signed out a memo pq As you know, the plan focuses on safety j pq iriitiating a management self-assessment to look at 99 and human performance.In March we had our first va the issues that had been discussed in our meeting of pri meeting with the NRC,the region,to discuss this ps: 1122 and other issues that we'd identified,and psi management corrective action plan.There's been a pq this self-assessment was led by a member of our sq series of periodic meetings, four to date,and those psi off-site safety review committee. ps) will continue.As we proceed,onJuly 7th og During December '94 to March '95,the psi inspection Report 9513 established the original l pn first 01 investigation was being conducted,and the nn predecisional enforcement conference and,as we were t pm crew that performed the evolution was interviewed as pq preparing for that,on July the 13th we became aware i og well as our manager of operations.As we continued ne that there'd been an evolution the day before,and pm to follow the corrective actions from Problem sa we became aware thmugh two channels.
pq Repons 149 and 267,we continued to realize the pq One,the bargaining unit operators had pa complexity of the interactions more and rnore,and we pri raised the issue through their attorney.1 guess ;
psi were discovering some other things wrong,like the pg they had read the inspection report,I believe
- pq borated water swapover point and those kinds of 94 that's correct,and realized that September the 4th psi things. pq was not discussed in it and property brought it up.
Page 35 - Page 38 (12) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
_ _ - - - - - = ~ - -
PRou II4hG5 BEFORE STEWART EILNETER, CliAIRMAN
- . PREDECISIONAL ENPORCEME!G CONFERNECE March 27,1996 !
j i
l Page39
~
Page 41 pi and we became aware through that channel. '
' to about that?
ca Concurrently and at the same time,the system pi MR. BEARD: Has anyone asked therre pi engineer who was involved back in September in pi MR. KELLEY: It may have been asked -
{ pi resolving and addressing Problem Reports 149 and pj MR. BEARD: Do we want to table that and
.l In 267 - and yes,this is system engineer - and two te get back to you? Would you rite that down as I
m other system engineers back in the September time m something to get back to thend i m frame did in fact know that it had been conducted in MR. KELLEY:We't! do that.
pi twice. pj MR. BEARD: My understanding was that, m However, back in that time frame and up m again,the LER was written with regant to what pq untilJuly they didn't put any importance on that. pg happened on the 5th.Now, why the engineer involved pq They were focused on resolving the technical issues. pu didn't think about the 4th or come forward,I can't na and to them the data and everything,it was all the na say at this point,and we'll get back to you with pai same.They were assurrung that we all knew or that pai our understanding of that.
1 pg management knew that it had been done twice. og MR.UEBERMAN: The otherissue is you
] pq Now, we personally talked to them, every pq just said that the person who fded the concern,the pai one of them.We'rc convinced that in their case psi important thing is to raise the issue, that it's pn there was no intent not to tell,and we could say, on better that he raise it that way than not at all,
- j. gisi "Well, you didn't use good judgment."Well, okay, psi obviously,but he was concerned,you say, that
, poi fine. But I will say that when the system engineer 9e because the bargaining unit people may be angry.
pq read the inspection report he noted also that," Gee, pg MR. BEARD: Yes.
pq ft didn't say anything about the 4th",and properly an MR. UEBERMAN:Is that a normal concern, na he came forward.He told his boss,who told his na that,when a manager raises : issue that may be -
pai boss,who was a maintenance manager, who told Mr. pai relevant to safety or relevant m how the plant's pq Hickle. All this was within a day or two,and Mr. pq operating, that workers respond by becoming angry?
psi Suven also went through our employee concerns pq MR. BEARD: No.1 think that is not =
Page 40 Page 42 pi program, so I got it through that channel and I got p: expected response,and if anyone fects free to talk
- pi it fmm Mr. Hickle.
m about. )
) pi Why did Mr.Suven use our employee pi MR. BOLDT: Let me add one other thmg.
{
- pi concerns program channel? Because he was the si Brian at that time, Mr.Suven, was actually manager si manager of a large group of bargaining unit peopic 18 of system engineering,was he not, manager of pi and realized that four of our operating crew were mi primary systems? ,
m bargaining unit people,and he was a little bit m MR. GUTHERMAN: He was the supervisor of !
pi concerned about,* Hey,you know,the union folks tai system engineering, primary plant.
1 pi were going to be angry with me for bringing m MR.BOLDT: At the time of the test he pq something up
- Now,he then came forward within a pq was the supervisor of the mechanical primary pu day and said,"I wrote it".We then found out,and 99 systems, so system engineering reported to him. At j na we told you immediately with a phone call. na the time of theJuly discovery he had been pai MR.UEBERMAN: Could Iinterrupt? You on transferred for development purposes to manager of pq said several things which I want to ask about.Were pq the mechanical maintenance shops.He was shop I
psi any of these engineenng system managers involved in psi manager. l pai reviewing the LER, developing the LER response? psi This was his first experience with on MR. BEARD: Yes.
on bargai,ning unit personnel, his very first, to my psi MR.UEBERMAN: Who was involved? psi knowledge,and I think he knew that his performance, psj MR. H ALNON: Phil Salzman and I believe poi his personal performance depended on his ability to og Pat Hinman.
pg get good performance from his people,and I think he pq MR.UEBERMAN:That LER describes the en had a genuine concern and rnaybe some naivete,but he na event,but at that time the 4th wasn't discussed. pa was genuinely concerned that he might lose the pai MR. BEARD: I can't answer the question na support of his personnelin the shop,but there was pq as to what they thought about that.Can anyone? pq no such reaction.
psi MR. UEBERMAN: Has anyone asked them pq MR. UEBERMAN: Have you given counseling BROWN REPORTING,INC. (404) 876 8979 Min-U-Scripte (13) Page 39 Page 42 '
~ ^' '~
l iROCEEDINGS bel iORE STEWART EENETER, CllAIRMAN IN RE:
, , March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page G3 Page 45 pi to other first line supervisors as part of your pj had told somebody.That's where we found out that pi supervisory training program to counselindividuals ni there were three engineers who knewof the test.
i ta on how to go about raising these types ofissues and pj There's no evidence that management knew of the I
pi to address these rnatters?
pi test.
(si MR. BEARD: Well, we've provided tramma te Then there were some other issues about ist and counseling with regard to the employee concerns asi we should have done a full root cause,which v.e m program, how it works, what the intent is, m agree with,and that the shift supervisor's im encouraging people to come forward.SoJim,I think 14 authority as to what he could or couldn't do was in in the broad sense my answer would be yes.1 think sq adequately defined in September,and we provided og Mr.Iloidt put Mr.Suven in the proper frarne of pq that investigation to 01.
d py mind. pq MR. GIBSON: Could you clanfy that last na Also, realize that management, that we na statement? I didn't understand.You said that na had already made an issue with the operaters back in psj there was evidence that the shift supervisor's pq September.We had the NRC now, and there was a big pq authority was not adequately defined.
pq cffect around our plant,a lot of concern,
-' sq MR. DEARD: No.The conclusion was that pq particularly the operators were edgy,the bargauung og it was adequately defined. it's in our report.
on unit peopic.Thank God the people came forward. On We took additional corrective action as a psi Mr. Suven, the engineer that read the report and pai result of that investigation, some of which was en brought it up, we're so very happy that it came og discipline.We terminated the shift supervisor and pg forward,that it did come forward,and in that way pq assistant shift supervisor,we revoked the license
, p9 I'd say our system worked. py of the chief nuclear operator,gave a reprimand to pa MR.LIEBERMAN: Have you given any pa one of the operators,and verbal counsung for one '
pa counseling to the enginects that didn't bring it pm of the others who was not involved in Septernber the 941 up? pq 4th.
psi MR. BEARD:I personally haven't. psi And you might say,"Well, gee, why did Page 44 Page 46 pi They've been talked to by you guys.Do you want to vi you do this now and not back in September",and the ta answer that now or later? Who has talked with ta answer is very sunple.it's a matter ofintegrity, pi Hinman and Salzman about this? ta not coming forward when you knew you should have.
pi MR. BOLDT: Jerry Campbell would have 81 It's an entirely different issue.Now,had we known rsi done that,but he's not here for me to verify that tsi about both in the same time in September would we Isi with. is) have donc differently with the operators then?
m MR. BEARD: Take that down, and we'll get m That's hypothetical.Maybe not much more than with tal back to you. Iai respect to the counseling and the things we did do.
re As a result of learning this, we launched im The issue here is the time lapse and that people pq into our investigation which was headed by,again,a pq who, through their training, experience,and in fact on member of our on-site safety review committee.We py the basis for theirlicense,sliould have known pa had a member of our security group on it.We had na better and should have told us,and they didn't.
psi our safety concerns rep and a technical rep.We did psi MR. LIEBERMAN: Pat,are you gomg to pq an extensive investigation.We talked to manY pq further discuss this investigation later on today?
pq people.And the conclusion of the report was yes, pq MR. BEARD:In what sense?
poi an unauthorized test was conducted on the 4th,and pei MR. LIEBERMAN: There was an issue called on there was no other unauthorized test conducted on pri developed issue number three involving correcth e om the make-up tank that they could find in 1994. pai action which the investigators didn't have time to em There was evidence that the shift og pursue.1 was wondering if that issue was ever pq supervisor and assistant shift supervisor had pq pursued.
pg withheld information.There were something like pg MR. BEARD: Yes,it was.1 think as we pri twelve occasions where they had opportunity to come pa go through some of the details that you'll see that, na forward with that.There was evidence that the psi and we can come back to that if you have further pq bargaining unit operators were not entirely pq questions.
pq forthcoming but some evidence that they thought they psi in addition to the discipline,we took Page 43 - Page 46 (14) Miss-U-Scripte BROWN REPORTING,INC. (404) 876 8979
~
4a 4a. rautLLutau31)La UAW 3'A'ANh A LadL11dhlhlhU AM PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 PIge 47 Page 49 pi six additional corrective actions such that we now pi further talk about some of these things as we get pi call our corrective action plan the 49 plus six. m around to answering the key questions and your other pj including such things as broadening the focus of p) questions.This is just sort of a summary of what p) what situations could be defined as infrequently pl are some of the key things I've said.Certainly the gi performed tests, more guidance to our shifts a isj issues involved are complex.We've learned a lot pi checklist requiring plant manager approval, tog lej about how this system interacts with something like m keeping practices.We expanded the management m nine competing operation requirements.Our si review committee,We now callit the review pi corrective actions have been ongoing since pl management panel process,and that's now documented pi September '94, including our management corrective pq in a directive and how it's conducted and a number pq action plan, which does address the underlymg vil of other things. pij factors.
pri Now,in August thmugh December the na We've had interactions ongoing with you pai second OIinvestigation was conducted,and a number pai since November,and we feel that we've kept you p4j of other people were interviewed that were not 04) informed with the exception that I didn't call Mr.
Ost interviewed the first time, including management. Usi Ebneter after September 5th. Again, not dwelling on pst in December of 1995 we became aware that there were pe) the safety consequences, what's important, as you've nn additional make-up tank alarm events,and we were on said,and we recognize,is the human performance pai provided through a third step grievance some data vai issues. including management.
ps from the computer that indicated that on other poi l'd like Greg now to just very briefly pq occasions in July through September during our pq talk about the system.and then we'll go to the pq normal evolutions of five a day to keep the pressure p,j plant manager.
pa up that it appeared that alarms had been exceeded pa Mr. Ebneter, you bmught up the fact that .
psi and maybe for some period of time, pal you'd like to after each violation then have p41 Now,you could say,"Well,se,why p41 questions,and we appreciate that.
psi didn't you know about that sooner",and what can you psj MR. EBNETER: I recognize that it may not Page 48 Page 50 pl say other than,"Well,yes, we should have." I wish p1 fit your agenda.
pi we had.Well, why didn't we? The investigation of pi MR. BEARD: It may not fit exactly, but '
p) August '94 went back and looked at the strip chart pj we'll work with it.We planned to have Mr. Hick!c p1 recorders which the operators used to manipulate the p) talk about the first two violations and then talk to tsi plant.They talked to the operators.They did si the management oversight that relam to those.
mi everything but one thing that was subsequently done mi Now,I would assume that probably, say, m by one of the operators we terminated,and that's to m if we stopped after A that your questions would mi go back and pull off all the computer data,which iai pmbably focus on management oversight,which he's si the computer generates the alarm,and see what that p) going to talk about.We'll stop after A and then pq says. pq see howit goes.
pq We investigated this information,and Mr. pq MR. GlBSON: Let's try it.
pa lialnon conducted that investigation,and we l na MR. EBNETER: I'll let Al sort of control pai confirmed that in fact in nine instances between p3) us.
p41 July '94 and September the 9th that the alarm was p.3 MR. GlBSON: After each violation let's psi received and was not dealt with in a timely manner, psi stop, and if you're going to cover our question pe) and we reported that to you in our letter of psi later,tell us so and we'll relate it after.
On December the 27th. As a result of that on I'd~ like to maybe ask a question or two psi investigation, we've taken some additional pai about your presentation at this point,if I may.
poi corrective actions with regard to alarm response poi You at the outset described your riq training,further strengthening the procedures and pq expectation that the operators should stay below pu things like that,really a reinforcement of the pq Curve 8,but that's a little different than the na things we already had under way.Then of course we pa understanding we had.We understopd that your ;
p31 received the March 8th report,and that brings us to pai expectation was that they stay on Curve 8.n close 941 today. g41 to or on Curve 8 as possible, !
psi So that's what happened,and we will ,psj MR. BEARD: Al 1 believe that3 BROWN REPORTING, INC, (404) 876-8979 Min-U-Scripte (M) Page 47 - Page 50
I ROCEEDINGS BEFORE STEWART EHNETERo C11AllB1AN IN RE:
March.27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 51 I Page 53 ty incorrect,but I would let Mr.Hickle and Mr.- i vi appreciation before September.1 did after that m MR.GIBSON: What we understood from {
m because I talked personally with him after !
pi interviews with operators and discussions with pl September.I was aware of the fact that there was pl others is that Pat Beard's expectation was for them Hj additional operator burden,that they were pushed in Isl to stay on Curve 8-ts) a narrow box.1 was not aware before September that
- q MR. BEARD
- That's totally incorrect. My Isj there was an issue about the curve being wrong.1 m expectation was from the beginning and has remained m think that was being discussed with Mr. Van Sicklen ;
tai to strive to maintain hydrogen concentration within tal and a few other people.
[9] the re ommended chemistry guidelines.That's the p) l MR. LIEBERMAN:Is it unusual for pq only expectation that I've ever expressed.
pq operators to have the view that various operators' oy MR. BOLDT:I would go on the record and o g limits or administrative limits are pri say, too, AI, having been in many meetings with Mr-
- 02) non-conservative? Is that a common occurrence.
psj Beard,that the only conversation I recall him MR. BEARD: No,it's not common.
03) pq having dealt with cc's per kg,which is not an pq MR. LIEBERMAN: Are you disappointed or psj clement of the curve at all.Only pressure and psi do you feel that your managers met your expectations psj level are. He was strictly talking about the nel in not bringing it forward to you or should they on maintenance of the dissolved hydrogen level. on have brought it forward to you,or is that something j pai MR. GIBSON: So you expected them to meet pai that they should handle themselves.
pa the objective of 25 cc's per kg.That was your psi MR. BEARD: That's a very difficult pq expectation. pq question to answer at this point in time.To give pq MR. BEARD: Yes.1 expected our 99 you a candid answ cr,I would almost have to go back p2) organization to operate the plant such that the p21 and hypothesize. Let me say that I'm quite '
pal hydrogen chemistry was maintained in the band. psj confident, and I was quite confident in September of 94 MR. GIBSON: In the time frame that we're pq '94 with regard to the fact that we had a very psi discussing,the fall of '94, did you have an psi effective,in my view then and now,an effective Page 52 Page 54 pi understanding and appreciation of the burden that vi employee concerns program.1 can give you many m was on the operators for maintaining this hydrogen m examples where it worked, where people felt free to pi concentration?
pl bring things forward.They weren't blocked and, si MR. BEARD: I didn't have as detailed an si they were promptly dealt with. I personally read tsi understanding as I have today.I recog:.2ed that it is; evcry employee concern.
to placed an additional burden,that they were having p) My managers and myselfin September,and m to manipulate the tank more often,that increased m since,I think,are involved in the plant.1
- al operator attention was required,and that we were tal frequently go to the plant.1 walk around the (9} Iooking for ways to further alleviate this burden, si plant,l'm in the contml room. People know they pq We talked about things from relocating the make-up sq can talk to me. I think it was, first of all, pq tank to give it a different elevation,which really ny unusual for operators to have,you know, concerns pri wasn't very feasible,to how to use the hydrogen p2l with real safety concerns with things that they were val regulator, setting it higher and letting it be ps) trying to operate. I feel that in this case, and I pq automatic,and what kept us from that was Appendix 04 would say that you can ask Mr. Van Sickten whenever psi R,one of those competing requirements.1 was aware psi he's up here,but in hindsight,if you were that nel that there was additional operator burden,and we pe) concerned,why didn't you use the employee concerns on were taking steps to address it- pa progtum? And I think in hindsight I'm not sure he pai MR. LIEBERMAN
- Were you aware that the ps) could answer that,in all candor.
pq operators had a concern that this curve was poi MR. GIBSON: Pat, we've talked about nine pq non-conservative? pq occasions where Curve 8 was violated.I'm sure that 99 MR. BEARD: I'd say yes. I didn't no now you're aware and in fact the alarm for exceeding pri appreciate the depth of the concern of one chief p21 Curve 8 occurred on many more occasions than that.
pal operator who was the one that really pushed this, pai We selected the nine occasions that were the most pq although I had talked with him personally.1 will r24 significant.
psi say in all candor that I didn't have a full ps; MR. BEARD: Yes, Page 51 - Page 54 (16) Min-U-Scripts BROWN P.EPORTING, INC. (404) 876-8979
hvU.wnda nu vat u touu matiuCuoniotAA
~
a at:
, PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 P ge 55 Pcge 57 19 MR. GIBSON: Were you personally aware pl You're getting mixed up with what we have today and ,
l wi that that curve was frequently violated? p1 back then.Back then we had an annunciator alarm.
p; MR. BEARD: No. pj On occasion I knew that the tank was coming alarm.
pi MR. GIBSON: Were you, Gary? 14j Mark had talked about whe9 we get up near the curve
- 15) MR. BOLDT: No. isi the alarm comes in.We never talked duration.We ts; MR. GIBSON: Were you, Bruce? is; never talked about that this was violating the ;
i m MR. HICKLE: No. m curve.What we talked about was that the alarm came tai MR, GIBSON: That doesn't seem consistent isj in,it added concern,and my expectations was the im with management's presence in the control tuom, the pi alarm was being dealt with.
pm tours that you say you took and that you addressed pq Now,in hindsight should I have gone up 4 pu earlier, Pat,the fact that yes,the investigation og and reviewed the annunciator printouts? I didn't l pa that we did,the internalinvestigation failed to pa think ofit at the time.My expectation was the ps) pick up on it.But of more significance to me is ps) alarms were being dealt with like they normally ,
I pal why didn't managernent know that these numerous p41 would be dealt with.When the alarm comes in,the l psi alarms were occurrmg during normal plant operation. psj horn sounds and you get an annunciator window that pa MR. BEARD: I will answer that. Bruce is pej says make-up tank high or low pressure and you look pn going to cover this.That's a question that we ask pn up on a screen.At that point the operators are pai ourselves.First of all,there were 669 make-up pai expected to take the LER out.I'm not sure where l ps! tank evolutions during the penod of the pq you were leading.
pm investigation,two of which were the 4th and 5th, pq MR. BEARD: Well,my point is if you go my And -let me get might math righ.-on 49 of those nu in a control room at any time, say I go in there and pa an alarm was received. Im I maybe spend 15 minutes and I would say you spend .
pa) MR. HALNON: 21 of the evolutions had an in more time,and Bruce will talk about this more than
); pai alarm.Some of them were multiple alarms. p41 I do,and Gary. At any time there may be a number
, psj MR. LIEBERMAN: We can accept it was a usj of alarms in,and we can talk about them. I don't
$ Page 56 Page 58 p1 small percentage, but for that small percentage a pj want to go over every one. So the point is, you i
pi number of them were for a substantial e .c Anod. p) know,in this one percent of the time somebody other l pi MR. BEARD: I'll get to that.1 p1 than a shift supervisor noted that," Gee,there's pl understand.Let me finish :r a period of how p) been one that's in alarm for 55 minutes."
is) many months are we talking about? Isj You can say that,but it comes down to if tej MR. HALNON: About three months. tal there's an issue here the person who is in the m MR. BEARD: July, August, September. m control room all the time,he's not there all the lei There were 21 alarms. (a) time, he shouldn't be,and I'm not.We depend on pi MR. HALNON: 21 situations the alarm came pi that person,the shift supervisor, to say or tell us om in,some multiple times. pq we've got a problem here,we can't clear this alarm pq MR. BEARD: 21 times the alarm was pq in a timely manner.Should we have known?Yes.
Da received, and nine times it was clearly too long. pa They should have told us.
pal Let's just first of alllook at the total time that psi MR. LANDIS: Was that ever raised, Pat, p41 the nine alarms were in for a substantial period of v4) that question,that we have alarms in for a along ps) time over the total time period of three months,and ps) period of time and it's getting difficult to keep it psj it's like one percent.Now, we've got to talk about nei out of alarm? Was that issue ever raised?
pn where does this alarm come up,and it's a computer pn MR. BEARD: It wasn't raised to me.Was pal generated alarm.First there's a screlling -let pe) it raised to you,Greg?
09] me make sure I'm right.There's a scrolling- pe) MR.HALNON: No,not the period of time, pm first there's alarm annunciators around the plant, jpq just the fact that it happened.
99 you get a horn.and the thing comes in.In this pq MR. BEARD:We're sort of getting ahead pa case,though,there are other alarms that are pa of ourselves.We were searching our souls,again, paj computer generated that show up on a scrolling CRT. psi asking ourselves if we should have known. How can pai and if one comes up 't's orange or red. p4) you say otherwise in hindsight? Could we have ps) MR. HALNON: Let me stop you there. psi reasonably expected it to be known?That does lead BROWN REPORTING, INC. (404) 876-8979 Min U-Scripts (17) Page 55 - Page 58
PROLLEDIN63 BEFORE 5IEWART EBNEIER, CIIAIRMAN IN RE:
March.27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 59 Page 61 p) to one thing we're going to talk about, rnaybe one of pj MR. BEARD: You might funher discuss, ia those sort of underlying things, gj though -
p) As we've gone through this we've realized p) MR. BOLDT:It would be the type oflog, vi thit our plant operauons manager, the way we were p) though,that would be reviewed by shift supervision, to organized at that time,had a tremendous load that rg especially internal,if there were some concerns.
tsi he was trying to cope with and he needed help.We sq MR. BEARD: log keeping is one of the m now have two peopic.We have an operations manager il corrective action plans in our 49 plus six that tai end an operations superintendent who the shift pl we're looking at.
pi supervisors repon to directly.The operations 16, MR. HALNON: Keep in mind we're dealing sq manager,he's licensed,he has to go through pq Mih two errors here, what we expected back then, on training, he has to go through the simulator and pq and what we expect now is a much higher standard.
pa watch everybody else.He's always the focal point pri Ifit was an abnormal and unusual thing that met the psi of licensing issues with operations. p3) threshold ofloggable things back then,I think v43 So as part of our broad corrective pq we've seen that it had been inconsistent back in (Se i.sj action,in fact one of the specific ones after ps) carly '90s and mid '90s.The same thing with the psi December, when we looked at this,was to create this psi spent fuel allegation we had one time where we had pq edditional position of an operations superintendent, on an inconsistent logging issue. it was still an poi this is one of our broader corrective actions, who poi issue, pe) has the time to spend more time in the control room MR. MERSCHOFF: So you view this, the log ve) pq to ask these types of questions. pq keeping, as an opponunity for management to become pq MR. HICKLE: It's an operations manager pq aware of this alarm that was missed?
pri positSn. pa MR. BEARD: Well, certainly, Ellis,it '
pai MR. BEARD: Well,1 didn't get the titles p3) would have been an opportunity.But I think,again, p41 right, but we now have two people. If you want to pq if you look in the control room,in any control room psi talk about a broad ps) in this country on a day-to-day basis,in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> Page 60 Page 62 pj where-we ought-to-be-held-responsible,yes,we 93 how many alarms do you get? No one logs all those pi should have realized sooner that Mr.Halnon needed pi things in their log that they personally keep. It's pi more help. pi casy to say that,and we certainly are strengthening j gi MR. LIEBERMAN:IfI could go back to the p) our log keeping practices,and you'll see that as ist issue we were talking about just before Al raised {
153 one of our corrective actions.Certainly,in isi the question that we're on now- 1s1 hindsight,you can say if they'd logged every alarm pi MR. MERSCHOFF: Before we leave this, m and all that maybe we would have seen it.
- sj would you expect an alarming condition like this to taj MR.GiBSON:Irt me see ifI can te be logged with the alarm corning in for an extended p; summarize what I see as a concern.What I see here, !
pq penod of time? pq and I'm going to say some things that maybe haven't pq MR. H ALNON: No. If it's an abnormal pn been said yet,but based upon the interviews of the pri event,yes.1 mean.if the operators perceived it. pri operators that have taken place and all the various pai MR. MERSCHOFF: For the multiple pai correspondence on this subject it appears that the pq conditions where the alarm came in, would that be pq operators in the control room had the clear message psi somethirig you'd expect to show up in the operator's psi that you and the other FPC managers expected them to pej log? pq operate on or near Curve 8 in order to maintain the on MR. HALNON: Typically it does not. Do on hydrogen concentration to meet the objective on psi we expect it? I would say no. pai hydrogen concentration.
poi MR. HICKLE: All of the alarms are logged poi in fact,there was es en one occasion pq on an alarm printer,so we have a paper record. pq where operators were reluctant to do that and -
29 MR. MERSCHOFF:Is that reviewed by py someone picked up the phone and threatened to dial pa management? pri Pat Beard's number and hand the phone to the pai MR. HICKLE: It's reviewed by the OTA in pai operator in order to get it up to the curve. Faced rq certain circumstances.It's not routinely reviewed pq with that, operators were having some difficulty ps; by management. psj maintainmg the plant on or near that curve without Page 59 - Page 62 (18) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979 ,
g PREDECI!ilONAL ENFORCEMENT CONFERNECE March 27,1996 Page 63 Page 65 1
- p) violating the procedure,and yet management of the p1 find themselves trapped in falsehoods.
] pi station wasn't aware apparently of this difficulty. p) MR. LIEBERMAN: Does your management
- pj You v tre not aware of the frequency of alarms and gj staff agree that operating crews were not being I
gj the '.lifficulty that operators were having meeting pi tracked on how well they met the curve or came close
[si yoarobjective. isj to the curve? i j tai It would appear that you should have been isi MR. HALNON:I agree,and I'm probably m aware, particularly in view of an earlier statement m the closest one to it.We tracked chemistry p) you made,that managers were in the plant and in the inj parameters on the whole.If we saw chenustry ,
, pj control room frequently monitoring what was going pi parameters trending the wrong way we took
- pq on.It would seem to me that you should have been poi appropriate action to get them trending back.
pq aware. 09 You'll never find anything I wrote down or anything pa MR. BEARD: A1,I can understand why you na that said Crew A,B,C or D was not doing a good job psi would,just reading all the rnaterial that's been ps) meeting chemistry parameters.They rotated on a no unearthed,why you would ask that question. And let pq weird schedule during the week,and any one week you psi me just say,again,that I don't think it's psi could have had four or five crews affecting it. ]
ps) reasonable to assume,first of all,that management psi MR. BEARD: At our8:30 management ;
pn should have known.You really have to go up and go on meeting, which is something that we instituted m j ps) in the control room environment and look at ps) probably '83 again where we wanted to focus on po' everphing that's going on,look at all the things pq maintaining pmper chemistry, we instituted, and I ga that happen on any given day,all the alarms that pm think it's everyThursday because I'm there every pq come in and out,and I've been there and asked 99 Thursday morning,a graph of various chemistry pa questions, particularly ones that are showing red on pa levels, reactor coolant pH, steam generator oxygen ,
i par the annunciator panet,and in fact I've always psi chlorides,that sort of thing,and one of them is pq gotten a good answer on w: , it was in.1 just don't pq hydrogen pressure in the make-up tank,and that's psi think in hindsight in our view it's reasonable to psi all we look at.That's allI've looked at.1 mean, g Page 64 Page 66 pj make that assumption. pi even to suggest that there's a report of which pi Now, again,I want to get back to this pi operating crew didn't maintain - ridiculous. Not pj issue,and it's something i said earlier that's piso.
pi complicating this whole thmg for you as the NRC and p1 MR. LANDIS: Let me back up just for a p) for us,and that is that we have an adversarial . isj minute.Was one of the graphs hydmgen pressure in isi position now with two operators. In fact,when I p) the make-up tank?
m read the enforcement policy,it's very clear, and I m MR. GUTHERMAN: It was concentration of 1 pl don't know what they're going to do but I think I get hydrogen in the RCS. !
pi know,and that is that if you read the enforcement pi MR. GIBSON: We're going to have to move I pq policy against the individuals their main line of om on. Jim,did you have a question) l pu defense would appear to be management made us do pq MR.LIEBERMAN:I just have one point and {
pa it", and I'm cominced that they're totally focused na then we can go to the violations because I think l na on making a case after the fact with all the psi this is important.The issue has to do with your pq rationahzation that they can, short oflying,and ny saying that it's somewhat unusual for licensed psj l'm not even sure that's not occurred,with two ps) operators to raise concerns about curves not being psi attorneys to make that case, and I realize that psi conservative.lt was being addressed by various on makes your position difficult. On engineering and operations staff with the company. I psi l'Il just say right here,and I'll go on ps) You weren't aware of that issue, p9j the record that I totally reject anything that says pe; MR. BEARD: Of what, Jim?
poi that I personally was trying to take action against 'pm MR. LIEBERMAN: That the licensed pq individuals,in fact that we were tracking pq operator had a concern that the curve was not na individuals who didn't or anything of that sort, na conservative.
psj That is not true,and I don't know how to make it psi MR. BEARD: Not before September.1 was pq any clearer.Now,if they want to rationalize that pq not aware of that particular.What I was aware of ps1 aher the fact, fine,but I think they're going to ps) was that there was concern with having to operate BROWN REPORTING, INC. (404) 876-8979 Misa-U-Scripte G9) Page 63 - Page 66
PRut.ELDiMA llEl ORE 51LWARf EliNETER, CllAIRMAN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 67 Page 69 p) the tank,that it was an additional burden and so pj was a concern with the way the curve was.
m forth. )
pi MR. BEARD: Now, clearly on September ;
pj MR. LIEBERHAN: But your staff was aware pj 4th,when they ran the test and Mr. Van Sicklen was !
M1 that the licensed operators were concerned that the p) part of that crew, when they noted what they thought m curve was not conservative. {
si was possible cavitation and again the system tg MR. BEARD: I think Mr. lialnon was aware gj response didn't track,I think my feeling is at that I m that Mr. Van Siciden was. m time, really because of the cavitation thing,that si MR. HALNON:I took over operations in tai it now really was a safety concern. I'm not trying n theJanuary of '93 time frame after coming through pi to gild the lily because I'm going to answer the !
pq license class and had a series of shift meetings.1 pq question.I'll wait until he finishes talking.
pu came in usually at four o' clock Sunday mornings to on MR. HALNON: It wasn't a hot issue.1 pri talk to the crews when it was quiet,and that's when na may have talked to them once.
pa it was first brougit up that they had a concern with pq MR. BEARD: I think it became a hot issue pq the curve the way it was.
pq when they ran the flow test.We noticed possibic pq MR. EBNETER: His question is were you os cavitation,and then again,you know,this thing pai aware of the non-conservatism of the curve.
psi about the pressure doesn't sectn to fonow this pn MR. HALNON:I snew of the curve,and I on curve,which it shouldn't necessarily have.1 think eq don't know if this answers your question,Stu.I pai that's when it became a real safety concern. In pq knew of the conct rn that they were feeling that the nel fact,I can recall talking to Mr. Van Sicklen pq curve slope did nr t match the curve or tank pq sometime after the problem report and all of that, my response, my and I remember what sticks in my mind is the pa MR. LIEBERMAN: Were you concerned fmm a pa cavitation thing.That's my view.
pa safety point of view at the potential of gas binding pai Let me answer your question.To the pq the pump? p41 extent that it was a safety concern prior to ps3 MR. HALNON: Excuse me? ps: September '94,to the extent that it was and say it Page68 Page 70
- q MR. L!E3ERMAN: Were you concerned the el didn't come to me, am I concerned? Yes. Fm pi hydrogen - m concerned.I don't like that.
p) MR. HALNON:I knew the curve was there p1 MR. HALNON: Can I correct the record for p) to prevent the gas binding of the pumps and i knew in a minute? My time frarnes are getting mixed up The si that they were trying to get cc's per 5 on that. m concern inJanuary was the higher pressure in the ;
pj MR. LIEBERMAN: My question for Pat is: pj make-up tank versus the potential, not the potential m Are you concerned that this issue did not get up to m but the emergency boration issue of getting water isi your attention! si from the BWST into the RCS quick enough.That was si MR.SEARD: Sooner? pi the concern inJanuary.We didn't know the curve pq MR. LIEDERMAN: That's my question.What pq issue with the slope until after the test SP430, on have you done to make sure your expectation is now pq and that was the one issue that I believe - and
.pa known to your management staff so we won't have an na again,it's so far along the way.
pai event hke this happening tomorrow?That's the om QUESTION FROM THE FLOOR:Could you go oci issue. pq over again when the SP430 test was? l pq MR. BEARD:I'll address that.Let me pq MR. BEARD: May of '94.
em first get back to Greg.Your question that Greg pq QUESTION FROM THE FLOOR:So that was on ined to answer was did the operator or operators on well before6eptember. I believe Mr. Lieberman's psi prior to September the 4th have a safety concern. ps; question was what about between May of '94 and the pol Now,that's different from we noticed that as the po; Septernber time frame and was the issue raised to pq tank level comes down that pressure doesn't seem to pq your attention during that time frame.
129 follow this curve.I've already said it doesn't py MR. BEARD: Yes. Jim,is that your
- 2a necessarily and shouldn't.Did they have just that pa question,between May and September?
pm concern or did they have a safety concern? pa MR. LIEBERMAN: Yes.That's the time pq MR. HALNON: I don't recall exactly how pq period, ps) they couched the concern.I just recall that there ps) MR. BEARD:Well,first of all,yes,I Page 67 - Page 70 (20) Min-U-Scripts BROWN REPORTING, INC. (404) 876-8979
m m.. ratut.cturma ou vitt 3 A c wn A uut a tu, u1AumAA PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 71 Pege 73 ty was aware of it.1 read every problem report. For pj he was,why didn't he bring out September the 4th m exampic,1 read 149.Yes,I was aware ofit.I m since he knew about it. Sort of as a followen to pi read the corrective action for 149,which was a good pi that, as I recall, you made a statement or question g) corrective action plan, reading the words. At that si along the line that said has anybody counseled the is point I had no reason to feel that safety concerns ist system engineers,and we didn't give you a very good l
- inj were not being elevated to my attention or being isj answer.1 saw you shake your head.So I'm going to L 3
m dealt with.I mean, we now know that we didn't deal m address those two things.
- tal with it very effectively.1 mean,certainly there's inj First of all,let me get back to the fact
)
181 no doubt about that.But it was brought to my goi that yes,we know that the system engineers,there i 4
sq attention,as was Problem Report 267 and the LER, pq was a Mr.Hmman and Mr.Salzman who were involved on and I think I signed that LER. pq in addressing the prob!cm report which of course was
- pa But what I have donc since then? I've na the basis for the LER,and in fact he was aware that pai talked to the shift supervisors on at least two paj the September 4th thing happened.However,they're j pq occasions,and I think they're documented,about the pq engineers,and I am not saying this in a disparaging I pu lessons learned from the make-up tank event,what sq way,but engineers particularly don't necessarily 1 pai our expectations are. In fact, you know,it's poi think in terms of human performance, operator
, na ironic that back in September they gave an attaboy va performance.We hadn't even had the management
- pai for a questioning attitude, and we've maintained pa; review committee meeting at that time.In caucusing poi that.
- poi with my people,we did ask the question in our 4
pq In fact,one of the things,as we pq investigation and ourselves many times,"Why didn't pu consider the MRC recommendatic as and our reaction to pq you say something in September",and the answer is
- pa that,and I think it's reflected in some of his na we were focused,I was focused.I,the system -
pai memos,is we wanted to strike the right balance.We paj engineer,on the technicalissues.
i pq didn't want to turn off the information and create a pq September 4th and 5th,it was all one k pq chilling effect by overreaction, so there was that pq thing,there was data,there was a technicalissue, t
Page 72 Page 74 pi balance to encourage them to come forward.But with pj and we focused on that.1 think why that holds pi respect to the event," Guys, you did it the wrong m water is when they read the inspection report and at pi way"That was always the issue.The end doesn't pi that time,you know,in July '95, everybody was very !
pi justify the means.And since September - well,let pl sensitive to the issue as an operations / human I is me tell you,I mean,this event has been in te performance / management issue.That's what's ici everybody's mind every day, every conversation,in tel important now.They were sensitized to that, they m meetings with management,one on one with my team, m read the report,and they saw it wasn't there and {
isi soul-searching further as we prepared for this. Let is) came forward,to their credit. !
isi me tell you.This isn't ever going to happen again tsj Nevertheless, your question was has
{
poi at our plant. pq anybody counseled them.1 guess we thought about ny MR.LIEBERMAN:I hope so. p q that,and counsel them about what? What did they do na MR. GlBSON:I think that unless we're na wrong as engineers back in the September time paj prepared to spend the night we'd better move on. pai frame? I mean they came -
pq MR. BEARD:I could just go to the end pq MR. LIEBERMAN: Do you want me to answer pq and answer the key questions and summarize, but pq that?
pq since you put so much work into it, Bruce and Paul, pq MR. EBNETER: The question is not why you pn we'll let you talk. pa should counsel them,it's did you or didn't you.
poi (A recess was taken.) pel MRILIEBERMAN: In November when the LER poi MR. BEARD: Before I turn it back over to poi was issued there's an event desenption.On pm Greg,I'd like to come back to two points in pq September 5,'94,conuel room license operators pu response to your quesuons. pu suspected the curve to be inaccurate,they did pa First of all,to Mr.Lieberman's pa whatever they did,and I presume the various paj question,I think there were two parts.One was pai engineers reviewed this LER in November, not pq were your system enpneers or was your system pq September,and this is after the management review pq engineer involved in writing the LER, and then,if psj meeting.They wouldn't have recognized that that's BROWN REPORTING, INC. (404) 876-8979 Min-U Scripts (21) Page 71 - Page 74
PROCEED 1NUS BEFORE STEWART EBNETER, CIIAIRMAN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE '
Page 75 Page 77 p1 not how the event happened.It was September 4th, ej you're darn right,and we're say not saying it was. ,
m suspecting it was inaccurate,that they started. gj Would we have rewritten it differently?
p) MR. BEARD: ButJim,the ER p) MR. EBNETER: Do we want an answer to the I pi constructively was focused on we had a curve that 81 counseling?
l re was design basis and we violated the design basis. p) MR. LIEBERMAN: The answer is they p) The ER was not written on the event and the ist haven't counseled and they didn't see a need to do m operators did wrong on the 4th and 5th,and that's m 50.
tai the answer. I mean, after the ER what would we is; MR. BEARD:I didn't say that last one, gei have counseled on? Would they say," Man,I wish pi Jim. I said at this point we were trymg to think '
pq we'd told everybody sooner"? Sure,they're going to pq what we should have counseled them on.
pq say that. 99 MR. LIEBERMAN: Up to now.
pa) Let me get to another point,though. pzj MR. EBNETER: Someone may have counseled pal What about what have we done in engineering which ps) them, though,that you're not aware of.
, poi includes these guys? As part of event-free 0 41 MR. BOLDT: I want to askJerry Campbell pq operations,which is the cornerstone of our pq that question. '
ps) management corrective action plan,we've made a psi MR. EBNETER: Let me put it this way.
i on clear case that mistakes, whether they be on You send them a letter telling us that you checked psi calculational or human performance or whatever, vai with them and you either did or did not counsel na apply to engineers as well as operators. Nuclear pe) them,just to make sure we're not missing anything.
pq plants in the past traditionally,and in our place, pq MR. BEARD: Let me say this, and Paul, l 29 if the operators screw up we make a big deal about pq you can disagree,but since September,with our um it.What about if the engmeer makes the wrong pai focus on the calculation process,our 50.59 process, psj calculation? What do you do about thend Well,let pai our operability process, have we all been counseled p41 me tell you what.We're doing a lot of things now, pai on what you said?You bet, including me.
j psi and we've had cases where we've counseled and 9e The second thing I wanted to come back Page 76 Page 78 pj disciplined engineers since then.So in a broad pj to,and I think this was Mr. Gibson's question, was a sense and through our heightened pl why didn't management know that these alarms were pl operation / engineering interface I think in a broad pi current.1 want to come back to that one more time pl sense, Jim,they have been counseled or made more p) with the caveat that management including myscif, si aware. Iq we're responsible for everything.That's not an isj MR. LIEBERMAN:I don't want to be p) issue. But two things.When you say management,I
- m argumentative.1 thmk engineers should be aware of a think what you mean is above the shift supervisors, mi the rule of completeness, they should be aware of ist right? AmIcorrect?
P1 the rule of misconduct,and I think in the LER, pi MR. GIBSON: Yes.
pq focusing on the design basis obviomly.but it pq MR. BEARD: Because as you also said,and pu describes the evolution and what happened. And pq we agree,the shift supervisors are also pri frankly,I don't think this met the standards of pri m7nagement.They're exempt tvople.So why didn't usi completeness and accuracy that we should expect in pal we know? And we will say that at one point,to be 041 understanding an issue.Not only on this issue,but 04j defensrve,that 99 percent of the time this psi later on when it says each time the operating point pq particular alarm wasn't in what had been reasonable i pq was on or near the curve the plant could have been pq to expect, nor did we log it.Would it be
_ on operating outside the design basis,from one point on reasonable tu expect : hat management above the shift ps) that's a true statement, but when it was on the psi supervisor should note that through personal poi curve it's my understanding it was operating outside poi observation? Our answer is no.That answer is no.
l pq the design basis. And to someone in our office pq So how should we have known it? Bruce is going to l
pq who's reviewing this LER who didn't have all the pq talk about that when we address management
- pri information.they would really not have recognized pri oversight.
pai the situation,and I think that should be a pai We should have known it had we been 9c1 concern. p4) entirely successful at that point in viewing our psi MR. BEARD: Was the LER a great LER? No, ps; management below us, including shift supervisors, l Page 75 - Page 78 (22) Misa-U-Scripte BROWN REPORTING, INC. (404) 876-8979 l
l l
.., m .
ruvu.wiaua uu uno a4 cmuu Loaca u4 unutoau
, PREDECISIONAL ENTORCEMENT CONFERNECE March 27,1996 PIge 79 Page81 pi with the importance and insistence on rigid and p) mind.Everyone knows that there are competing m strict adherence to procedures,the event-free m operational requirements going on, the high pressure pi operations ptogram and precursors.When something p1 in the make-up tank, the swap.over point.
p) isn't right you wnte it down and tell us,and we p) There are a lot of different competing isj review every precursor card the next morning at the tsj requirements going on, so the operators were dealing isj management level, including Mr.Hickle.rd say isj with those things.Also,the standard ES pump,and m that we were remiss in not achieving that awareness m what I mean by that is the pump, the ES selected m and use of the precursor program sooner or at that ta) pump,the HPI pump that is not running,and we show m time.So in that sense,yes,because,had we done m this one as the C pump on this diagram and you'll pq that, they would have reported it or should have pq get a copy of this in your handout when you get py said," Hey,we're having difficulty,we're getting 09 them,but this valve right here maintains separation prj an alarm,we can't clear it sooner",and we would va f om the make-up tank at all times, so one train of pai have known and should have known.
psi the HPI system is always independent of the make-up pq Greg, do your thing and then we'll get to pq tank issue. It's not an issue,and that adds into psi liruce. psi the safety consequences.
psi MR. HALNON: Like I said,I want to osj The spare pump,your question,is it pn really drop mine down to just about a minute. First na available if it's operable, not in maintenance, but pai of all,I want to correct something I said earlier. per under normal circumstances we treat this pump,if we poi i took over in January of '94, not '93, so I was a poi get into maintenance on that, we treat that like an pq year off. I was talking to Bob and all these dates pq action statement.We work it as quick as possible, pq are starting to run together. rd like to just pq and we have a high level of attention on those ga spend a few seconds describing the control room and pa pumps, so when we get into maintenance on those .
ps) alarms. Is there anything physically with the plant pai pumps we put a lot of attention on it and work it pq you need to have described *:at you feel you're not 99 two shifts,sometimes three shifts.
ps) clear on, the three pumps,the make-up tanks or the ps) One last point that rd make before I sit Page 80 Page 82 ;
pj llWST levels? pi down is that as you're drawing down these tanks and m MR. RICHARDS:If you get a safety m putting water into the RCS which is going into the p1 injection do all three pumps start or is one of them pi reactor buildmg sump because of the IDCA,you get !
p1 locked out? pi over to where you swap over the long term isi MR. HALNON: At any one time the third tsj recirculation cooler,take the LPI pnmps, take iq pump is always spared and it takes manual operation tsi suction from the reactor building sump and, m to get that started, so at any one time two pumps m depending on the size break you might have,if you m will start,and one of the normal running pumps, tai su11 have your HPI pumps running,the dinharge of m smcc this is a multi function system,one of the m the LPI pumps feeds to suction of the make-up pumps pq normal running pumps is always one of those two pq or HPI pumps. At that point, when you provide that py pumps that will be running on HPIinjection. On what we term piggyback operation,this check valve na MR. GIBSON: Were all three pumps na in the make-up valve ceases,and that isolates the l paj operable during the test? pai make-up tank from the picture.from that point on pq MR. H ALNON: No. During the evolutions pq there's no concern with hydrogen training.So this psi during the 4th and 5th we had the spare pump,which ps) discharge pressure,when you're feeding this,when ps) at this time we had the Il pump spared, and it was in sq you're feeding the one that's normally running,you on maintenance.It was just about ready to come back on back up and seat that check point,so that again
~
pe; into service.The A pump and the C pump were our ES pai adds into the safety consequences. l ps) pumps.The A pump was the normal running pump. poi These technical issues won't be covered pq MR. RICHARDS: Do tech specs allow you to poi any further. I just want to make sure we're clear pq take one pump out? pq on those types of things.lf there's no other pa MR. HALNON: The tech spec only mentions pa questions about the physical plant, r11 go ahead psi two pumps.The third pump is a swing pump in case ps) and turn it over to Ilruce.
pq you have to do maintenance, pq MR. HICKLE: Good morning.As Pat told psi Just a couple of key points to keep in psj you, rm going to cover the two violations that had BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (23) Page 79 - Page 82
PROCEED 1NGS bel; ORE S FEWART EBNETER, CliAIRMAN March 27,1996 IN RE:
PREDECISIONAL ENFORCEMENT CONFERNECE Page 83 Page 85 p) to do with our operators and operating practices.
p) operating crew.It outlined precautions, m I'm also going to discuss the management issues m responsibihties, expectations, and on September 5th pi associated with those. I'm going to discuss the p) designated a dedicated operator to be stationed in p; second violation first,which is the tests that our p) our auxiliary building.In the event of an '
- n operators performed.Then I'll talk about the other tsj accident,that dedicated operator was to vent and
- aj nine evolutions that were the subject of the first pi bring the plant back within normal acceptable m violation and inspection report,and then finally m operating parameters as illustrated by our make-up
- s; there were three management issues or concerns that :
I tai tank . urve.The hydrogen pressure was raised to ra were identified in the inspection report that l'il pj alxne the alarm set point,the make-up tank level i pq talk about after those two violations because there pq ral,ed to the high levellimit,86 inches,and then pq is some overlap, and I think there is some ,
ny a 1,1ced from our make-up system was established pal relationship between the management issues.
Da w hich allowed the level in the make-up tank to drop na Now, Mr. Beard has already covered some ,
4 ps) to about our low level limit, and data was taken p4i of the information that I had to trview and some of p4) during this evolution.
pq your questions have already addressed some of those 04 After about 35 minutes on September 5th
- asi subjects, so if you find that I'm getting into an nel the plant was returned to the acceptable region of
, an area where you feellike you have sufficient on our operating cune.A problem report was written ps) information tell me and we'll move on.Some of the pai on September 7th by the operating crew that j poi information I'll move through rather quickly for
' psi conducted the test,and during that time the make-up 2m th:t same reason. 3 pq tank pressure exceeded the curve by approximately i 29 MR. BEARD: And we'llleave you a book of py 1.7 pounds at the low levellimit,so right about I
- n1 all the overheads and this other boog of the p21 here. '
, pai chronology that I mentioned. pal Let me show you a trace very quickly,a 9 41 MR. HICKLE:I won't spend a lot of time p4: computer trace.This is downloaded computer ps) discussing the evolutions.I'!! provide just an ps) information, so you can see over time how this .
Page 84 Page 86 l pl overview for the sake of discussing the root causes pj evolution was conducted.First,the make-up tank m and the cc vective actions.
' gj hydrogen pressure was increased with a hydrogen pi The first violation I'm discussing are pj add.That brought the tank above the alarm curve.
si the two examples of conducting an unauthorized test si This is the alarm curve which emulates the make-up
- s, or experiment without a written safet valuation ts: tank curve that we had in our plant curve book,
- ej containing the basis for the deternunation that an ist which is OP-103B.The alarm was received in the m unreviewed safety question did not exist.The m control room.At that point levelinventory was I rei inspection report talks about that there was no p) added to the make up system and the level went up l al approved procedure for the test and there was also a p) until such time as the alarm cleared and the sq failure to follow procedure- pq function, which is the combination of pressure and py Some useful facts in understanding our ny level,was right on the curve.
paj causes and corrective actions.First of all,there pri At that point our operators established a ps) were two evolutions,one performed on the 4th of ps) bleed from our make-up system allowing level to ,
p41 September and one performed on the 5th,1994, both p41 drop,and over a 35 minute period plotted data and I se occurred on the midnight shift,and both were pq reproduced this curve as part of the problem report l psi conducted with the purpose of testing the validity ps) that they wrote,and then later recovered by l pn of our rnake-up tank curve as illustrated here. pn increasing inventory to the tank. j psi There was no operational reason,and we concur,for ps) The September 4th evolution was similar poi those two evolutions to be performed.The dissolved poi but not identical. In this instance hydrogen was '
ya hydrogen in the reactor coolant system was within pq added.the alarm was received at near the high level ry specification at the time.Both tests were pu limit in the tank,and then while the alarm was in l 223 perforrned without a test procedure or a 50.59 rai the bleed was initiated and data was gathered.The
- rs) evaluation. ps! difference between the two evolutions,in this 7 41 Some key points regarding the scenario, p41 instance it lasted a little longer,about 45 minutes ysi or facts.A pre-job briefing was conducted by the psi roughly,and the alarm was not cleared prior to the Page 83 - Page 86 (24) Misa-U-Scripte BROWN REPORTING, INC. (404) 876-8979 !
1
..,m., r ,wm a .w ~ u ,L i b. - , ub i. , u@ ,. ua..u , i
, PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 '
P;ge 87 P2ge 89 19 start of the evolution. I'll come back to that vi si currently in progress to provide technical basis m point in a few minutes. pi for the BWST swapover point. During this analysis pj We've spent a lot of time looking at this p) make-up tank overpressure per curve number eight pi event or these events,and we believe that there are 83 will be reevaluated.This action is scheduled to be isj several root causes, as follows. First of all, we rsi completed by September 30,1994.
Iq believe that we had a deficiency with shift p) This memo clearly told me and our m supervisor leadership.What do we mean by that? We m operators that it wasn't a closed issue.There was pi think the strongest personality that was on that pi still technical information that had to be evaluated
)
pi shift that night, the person that was really running in that would have a bearing or potential bearing on poi the test and organizing it and planning it,was our om the final outcome of this technicalissue, ny chief nuclear operator.Our shift supervisor pn When I received this memo I conferred pri allowed our chief nuclear operator to take that pri with Greg,our operations manager,and asked him to paj control.Furthermore,we have evidence,and I'll ps) take that back through his organization to our l
pq talk about the management relationship to this in a pq operators and find out if there were any other I psi few minutes, but we have evidence that our shift pq concerns related to the make-up tank issue,any {
pq supervisor had some problems with strength and psi other concerns that they cared to articulate that we l pn leadership, and I'll discuss specifically what I on may need to address,so that we were sure that by psi mean in a few minutes. Osi this date we had everything addressed we needed to. 1 pm The existing procedures that existed at A final contributing factor is that poi the time in 1994 were either not consulted or they pq were not used correctly.We think
- hat the crew lvoi pm management's efforts to strengthen shift supervisor pu leadership were not timely.The first personnel pri made some cognitive errors,that they used some poor p21 change that we made back in the early part of 1994 .
pai judgment,that there was some rationalization that par with a specific objective of trymg to strengthen pu occurred and probably some group-think, group-think pq on-shift leadership was replacing our operations f ps; as much as they all reached as a group the wrong pq manager,who had been in the job a long time,with a )
Page 88 Page 90 '
pi conclusion that this evolution could be performed p) new operations manager,Greg llalnon.That was the m within the scope of our existing procedures, and m beginning of an overall plan that we expected to pl management was not successful finally in achieving p) create improved leadership on shift and in vi consistent adherence to procedures by operators.We pl management positions of operations with the express I is) believe that that is also a rnot cause. si intent of trying to improve human performance in the l gj We've done a lot of things to address pl organiz:. tion to try to move up to what we viewed as l m that which I'm going to talk about under the m the need to increase studards in the organization. I si management issue.but at that time procedure use was p) flowever, we didn't move fast enough on p) inconsistent from shift to shift, and the pi our shift supervisors.There were a few and also pm expectations were also not clearly understood by all om some assistant shift supervisors that eventually we f pn of our operators.There were some contnbuting lp y had plans to rotate into the organization in other I p?) factors. p2) areas into support functions so that we could bring pal First,there was no doubt the crew was v31 new talent in.Our operations organization was pq strongly motivated to prove that the make-up tank pq primarily staffed by supervision that had been there ps) curve was wrong,and they mistakenly felt that the ps) a long time,and we felt a change was necessary,a pq issue was going to be closed.We referred to that ps; change so that we could promote new ideas and pn earlier.There was a letter,a memo that I received on improve our operations.But we were somewhat pm on September 2nd which that crew reviewed pnor to !ns) constrained inasmuch as we didn't have a surplus of poi conducting the evolutions, and the letter states - om extra operating licenses.We didn't have younger poi this is an excerpt - that engineering believes that poi people - when I say younger, meaning new talent -
pu this curve is accurate and reasonably conservative 99 in the mill trained and ready to step in, so this pa to protect the high pressure injection pumps imm p2) process took time.In hindsight,we didn't move ps! hydrogen gas intrusion in the worst case large break pal fast enough.
pq LOCA.In addition, corrective action number eight pq l'm going to discuss corrective actions, ps of Problem Report 149,which Mr. Beard referred to, ps) but l'rn going to separate these. First ril talk IIROWN REPORTING, INC. (404) 876-8979 Min-U-Scripte (25) Page 87 - Page 90 l
l l
PROCEEDINGS BEFORE STEWART EBNETER, CIIAIRMAN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE i
Page 91 Page 93 !
01 about what we did after this learning about the pl strengthened program barriess to prevent similar m September 5th event, the September 5th evolution, m judgmental errors.We conducted shift meetings,and p1 recalling that,looking at our chronology,we didn't p) Pat's already talked about that.We had multiple 81 know the evolution that occurred on the 4th at that Iq shift meetings to discuss what we had learned and to
[sj time.So I'll separate the corrective actions. Isj reinforce our operating standards.
tg First these are the ones that we took im MR. LIEBERMAN: I know I'm not supposed j m over September 5th.When I became aware of the waY m to interrup, but could you give me a time frame m that the data was gathered on this evolution,I pl when those shift meetings were conducted?
p) immediately notified Pat Beard and also the same day pj MR. HALNON: It was throughout the fall. ;
poi notified the NRC resident and our branch chief,NRC poj MR. HICKLE:I'd like to give that pq branch chief. At that time we had not conducted an !
on exactlyif we can.
na investigation.We only became aware that we had a paj MR. HALNON:I don't know if we have va potential problem and we had enough information to pai exact dates.They happened throughout the fall. !
pa suspect that the right controls were not applied pq MR. BEARD:All of '94.
Os1 that evening to conduct that test. psj MR. HALNON:I think that some of them pq We formed a management review committee. pej may have had follow-up in '95 also.
On That was my idea.We have had no history of on MR. HICKLE:It came in phases.Of l pai management review cornmittees.That wasn't a part of pq course we initially met with our operating crews and 091 our formal corrective action process, but I felt poi tried to reinforce the things we felt we needed to pa that it needed elevated management attention. poi do right away,but the reinforcement came through j pq Furthermore,I was very upset about what pu the requalification cycle.
pa had occurred, and I wanted people that maybe were a pri MR. LIEBERMAN: The imtial meetings, ' ]
pa little cooler that could provide some advice to paj would that have been in September? I py myself and also to Greg so that we could evaluate pq MR. HICKLE:I can't give you a date.1 psi the circumstances and act reasonably based upon the psi believe it was,yes. l Page 92 Page 94 01 facts. I'll talk about the management review pj MR. EBNETER: Could you reconstruct that I m committee after i discuss the rest of these m and send it to us,or would you just be guessing?
m corrective actions. I'll talk specifically about pj MR. HALNON:It would be a guess. ,
pi what we looked at, what we did and what we MR. HICKLE:I think we can get pretty pi l ts) recommended. isi close.These were inunediate corrective actions.
im Crew discipline resuhed.The crew im QUESTION FROMTHE FLOOR:According to a i m discipline resulted after the management correc;ive m memo on October 4th of '94 which was written with (si action cornnuttee.I'll go back to that and describe p1 regard to this process, Bruce had already provided f gi it more fully because I know there is a concern p) counseling for the shift and talked to Fields and poi about it.First we counseled the crew.The pq Weiss, and there was already counseling for the ny counseling of the shift supervisor and assistant pq reactor operators on the shift by Mr.Halnon,so p2) shift supervisor was done by me personally, and Greg na that according to this memo all of the shifts had om attended that counseling session.Later,the rest om already been talked to in the October time frame, pq of the members of the shttt were talked to by our ny MR. HALNON: The next phase came when 01 l pq operations manager.Our operators that were ps, came.We knew it was a traumatic experience to have om involved w th this evolution received operator pq 01 come to the site,and Pat wanted to have on retraining on the lessons imrned,and I'll on communication going with the shift so they knew what poi specifically tell you what those lessons were that pa) was going on.
on were learned when I talk about the MRC. poi MR. HlCKLE: And Pat and myself attended pm We also required the crew, specifically am those meetings and met with each shift when that 24 the shift supervisor and assistant shih supervisor, pq occurred, so it occurred over a several months' l
pa to go back and write a test procedure with a 50.59 pa period of time. l pa evaluation to reinforce the expectation that we felt pm MR. GUTHERMAN: Is your question related pq was not fulfilled that evening,and they did that. p.) to when that specific crew was counseled or all of psi We reinforced management expectations and we psi the crews?
Page 91 - Page 94 (26) Min-U-Script @ BROWN REPORTING, INC. (404) 876-8979
huu.444.ub 4diUluM 4. Musi 4 iu.41.46.Eh4&L
~ ~
Aa au.:
, PR,EDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 P ge 95 Page 97 !
p1 MR. LIEBERMAN: The other crews- p) that we didn't feel that our procedure use from 1
- pi MR. GUTHERMAN: Thank you. m shift to shift was consistent.lf you recall,I pi MR. BEARD:Just for the record,and this pi called that a contributing factor, so I'm trying to pj is in this thing that I will leave you, the pi separate for you what we viewed as a generic pi chronology of stuff,but there was a memo to file tsj weakness from events that directly or conditions
- pi from Mr. Hickle that he counseled the shift get that directly caused the evolutions to occur.Therc m supervisor and assistant shift supervisor on
{
m was sufficient guidance for the conduct of these J rei September 15th.Irt me see if I can find it real pj evolutions.and I'll show you that later when I talk I pi quick. I'm not sure that I can.We'll look into pi about martagement concerns. I'll put you on that pq that for you. pq shift that night,and we'll go through the logic 09 Would somebody capture that as a on that should have occurred,that I would have pai follow-up item? pa expected to have occurred,and I'll show you the pai MR. GUTHERMAN: I've got it. psi barriers that were in place at the time that would 04 MR. HICKLE: So we conducted shift 04) have prevented these evolutions from occurring.
psi meetings.We also,in these meetings and psi MR. BEARD: Existing procedural psi separately, reinforced our procedure use nei guidance.
On expectations,and that was done through event-free on MR. HICKLE: Existing procedural pai operations,through our event-free operations psi guidance.I'll come back to that and make that voi seminars that Greg conducted himself at our training poi clearer.
pm centers for all of our operating shifts. Procedures pq We've made improvements in our training.
pn for procedure use were strengthened,and what I mean pn That's one of the things that you can do to pri there is we developed a standard which specifically tra reinforce expectations and correct human performance, psi tells our operators under what circumstances they psi problems. Lessons learned were also incorporated pq need a procedure,that they r" ed a procedure in hand p41 into our event-free operations program.Since the ps) and they need to specifically sign off the steps or. psi time of these events we've developed a very Page 96 Page 98 pi depending on the category assigned,there are some p1 comprehensive tool to use that's a program to m occasions where they're allowed to provide simple m improve human performance at the station,and I pl manipulations without a procedure, p) think most of you that have participated in our pi We then took that standard and went back pj meetings to discuss our management corrective action tsj to every single operating procedure we have and is) plan are very familiar with our event-free mi specifically delineated the steps in those by si operations program. Its main objectives are to m category so there would no longer be inconsistency m impmvc safety culture in the plant,to improve a J pi from shift to shift as to how these pmcedures were l pi questioning attitude,to cause people by these l
[9) being used, when these evolutions and manipulations pi things to do the right thing the first time,and we pq were being conducted. pq provide tools and we stress the importance of pq MR. GIBSON: You did this even though :pq procedurr use as one of those tools.
pa your internal invesugation later showed the lpa Also,we're strrngthening on-shift usi procedural guidance was adequate? lus) leadership.We've been able to take our plans j v4 MR. HICKLE: That's correct. And the 104) further. Remembering that it started with Mr.
psi reason for that is this:liuman performance / operator ! psi thinon, since that time we've replaced two of our ps; action is a combination of expected behaviors and lpei shift supervisors.We had identified a few that we on are always going to be a combination of training, inn felt eventually we'd like to see rotate.We've
~
pai experience and the tools that are in place such a5 pai changcd our shift crew compositions to try to make l pei procedures.Thile we feel the tools were adequate, vej sure that the team dynamics are right for pq if they had been consulted for these two tests that pq communications to occur,and we've established a pn were performed.one of the logical things to do to pn long term plan and recruited and done some pa prevent recurrence is to provide more defensive pri recruiting to allow us to take and train experienced psi depth, and that was the intention with respect to psi engineers to eventually come up thmugh operations, p4) developing this pmcedure. e4) receive the senior director operating license and psi 1 want to also add that I said earlier lus) become assistant and shift supervisors.We've BROWN REPORTING, INC. (404) 876-8979 Min-U. Scripts (27) Page 95 - Page 98
PROLEEDING5 BEFORE 5 DART EBNETER, CliAIRMAN IN RE: I March.270 1996 PREDECISIONAL ENTORCEMENT CONFERNECE Page 99 Page 101 pl recruited three so far.
01 judgments were made, we think some rationalization I pi The criteria for recruitment they've had pl occurred, we think there were some strong pi to have naval nuclear experience and they have to be pl motivations which may have clouded judgment, but we
- pl degreed engineers from an accredited university,and !q never felt that the operating crew, based on the Isl we're finding that there are quite a few people out tsi information that we had, intended to violate our ;
isi there we're able to recruit that meet those mi procedures. And that was a mitigating a criteria.We also screen them for what we view as m circumstance, i tsj very strong leadership capability, because that is The crew provided pmfessional, isi p) probably one of the rnore critical criteria for a pi believable responses to everything that we asked.
pq shift supervisor.
pq The crew generated a problem report that documented ny MR. BE ARD: Bruce, you might also mention on the evolution that they conducted on the 5th.Now, na that we've recruited internally. An excellent na that was the only one that we were aware of at the psi example is sitting right here.We've taken va time.There didn't appear to be any coverup on the
_ pq enginects such as Gary Becker, and Gary has stood pq part of that crew.At the time we didn't recognize pq watch with the assistant shift supervisor.Two more psi that curve as a design basis limit,and the crew psi have gone thmugh that and we have two in the psi didn't.cither, so there was no need to believe that -
on pipeline right nowinternally. on they intentionally violated the design basis curve, i pq MR. HICKLE: I mentioned that I'd discuss psi but we realize that now after the fact.
poi the management review committee a little further- poi MR. LIEBERMAN:Just to clarify that pq The purpose,the stated purpose when we convened, pq point to make sure it cornes out on the record the pn was to conduct an overview of open issues relative pu right way,I think you just said that now you feel pai to the problem report that our operators wrote about pa they intentionally violated the design basis curve. '
sai the September 5th evolution and review the test ga MR. HICKLE: No, that's not what I pq performed by the operating crew.The committee pq intended to say psi members are listed.They're fairly high level psi MR. EBNETER:I think you said at the ;
Page 100 Page 102 pl managernent, and our intention was to try to bring ni time you never felt the crew intended to violate the m senior level engineering and operational expertisc pi procedure,that at the time there didn't appear to pi to bear in assessing the issue or assessing the pi be any coverup,and at the time you didn't :
pi facts relative to the evolution. pi understand that the curve was a design basis.All Is- Our director of quality programs is aur tsj of it was precluded by "at the time." !
p1 former plant manager.Our director of nuclear iq MR. BEARD: That's what I heard.
m operations engineering and projects was in charge of m MR. HICKLE: Remember, this is a snapshot ;
!al all of our design engineering, and our rnanager of pi in history. l p1 nuclear plant maintenance was there because he was a mi MR. EBNETER: But I know where Jim's pq former shift supervisor. in fact, he was one of our pq going.I think.I just wanted to clarify.
pq best shift supervisors. pn MR. BEARD: I thought that's what he na These are the general conclusions that va sa d.
pai the management review committee arrived at,The psi MR. HICKLE:I don't want to infer that pq evolution clearly did not meet documented operating pq now we think to the contrary.
pq standards and expectations reinforced in training, psi MR. LIEBERMAN: It's important to make pq specifically those that refer to questioning ps: sure that's clear.
On attitude, procedure compliance,use of approved on MR. HICKLE:We determined that we needed psi procedures and infrequently performed tests and psi to discuss the importance of adherence to operating poi evolutions.The committee believed that a test pe1 curves and other litnits and expected response to pm procedure with a 50.59 evaluation should have been pq alarm conditions with all operating shifts,that we pu used to perform the evolution. pu should review all operating curves and OP-103,our pa We did not characterize operator actions pa plant curve book,to identify other instances where psi as an intentional violation of procedures, and if pa operating crews may be required to operate too close pq you recall the root cause,and we still wouldn't pq to the limit,i.e.where too little margin exists
. psi characterize it that way, but we think some wrong psi between the normal administrative limit and the Page 99 - Page 102 (28) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
- - =- - -.. ~. -- - - - .. _.
IN RE: PROCEEDINGS BEFORE STEWART EBNETER, CIIAIRMAN PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 ,
Page 103 S Page 105 pl operating limit,and that we should provide ni looking back in hindsight,that the operators were i pi counseling for the shift that performed the test,
' m candid about the September 5th event or what they pl stressing the importance of avenues for resolving pj did on September 5th?
p1 issues, the importance of maintaining operating pi MR. HlCKLE: That's a good question.No Isi limits, correct methods for performance of Isi is the answer.We asked the operators, and I say pi evolutions, abnormal evolutions, and the Isj the operators,but the assistant and shift 1 m consequences of repeat performance. m supervisor were the only two operators that were pj Now,all of these recommendations are )
tai interviewed by the MRC.We did not interview the m cumpleted,and one question you may have is why - p) engineers,we did not interview the other shift pq MR. BEARD: Would you pause for a pq members as a committee.
pij minute? vij Getting back to your question,I said no i na MR. HICKLE: Yes.
na because we asked them if there was anything else psj MR. BEARD: Go ahead.I just wanted to ps) that they had to tell us that had a bearing on the 1 04) make sure that Mr.Lieberman heard everything you pq issue,and I felt like that meeting was conducted
{ ps) were saying. nsj pmfessionally in a non-adversarial manner,and I
. pq MR. LIEBERMAN:1 heard him. Dej felt like they looked like they were comfortable on MR. BEARD: All of these were completed, on enough to bring that kind ofinformation forward.
pq is what he said. pq MR. LIEBERMAN: But other than the 4th, j poj MR. HICKLE: So why didn't we go nel other than the 4th,as to the 5th, what happened on i pq further? At that time we felt we had all the tm the 5th, were they candid as to what happened on the
] pij information that we needed.Obviously in hindsight pil 5th?
pri that wasn't correct.lf we had had any inkling that ga MR. HlCKLE:We felt they were,yes. -
j pa) that crew was trying to hide their actions or if ps) MR. LIEBERMAN:And even today,in l pq they had come into that meeting with a defensive p41 hindsight, when you know that much more {
psj posture or even if they had not accepted the ps) information? l l
Page 104 Page 106 p1 recommendations made by the MRC,we would have had vi MR. HICKLE: The information that they J
m cause to go further.
mi provided to us appears candid today, p1 Another factor, though. is at that time p) MR. LIEBERMAN: On your second bullet or pj that preceded our event free operations program,and pj arrow there,it says, Review all operating curves to is) the root cause evaluations that we were conducting (si identify other instances where operating crews may iq were formal structural root cause evaluations and tel be required to operate too close to the limit. Does m were not very many in terms of what we do now, m that suggest that in this instance with Curve 8 they m today,and they addressed primanly mechanical and pl were required to operate too close to the limit?
m technical or equipment performance problems and,in p) MR. HICKLE: They operated too close to pq some instances, some instances, regulatory issues. pq the limit.This is the window that they had to pii We had not developed our root cause program and pit operate within. Provided they would keep the na integrated that into our human performance na function,the level-pressure relationship in this p3) initiative so that the root causes were very rarely ps) area for a long enough period of time,they could ny done to look at deep root causes of human pq achieve 25 cc's per kg in the system.That window ps) performance problems. ps) is not large relative to the entire curve, although psi We didn't conduct a formalinvestigation ps) we felt,I felt,that it was achievable.At no time nn in the sense of an 01 investigation complete with on did 1,~ or anyone else, to my knowledge, tell any of pai transcripts and impartial panels and such because we psi our operators to operate on the curve or even near poj detected absolutely nothing in our exchange with our poi the curve.What we asked them to do was operate the pq shift supervisor and assistant that would cause us pq make-up tank so that the objective of meeting 25 pij to question the validity of the information that we pij cc's per kg or greater in the reactor coolant system p21 received.
na was achieved or tell us why they couldn't,why they psi MR. LIEBERMAN: Even today, after you Raj have a problem doing that.Now,at the same time, pq know so much more .han you knew in September, just pq and I was going to talk about this in the first ps! focusing on the September 5th event, do you feel, psi violation-BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (29) Page 103 Page 106
PROCEEDING 5 HEFORE STEWART EBNETER, C11 AIRMAN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONTERNECE .
1 l
Pige 107 Page 109 p1 MR. GIBSON:Ilruce,let me raise a couple p1 hydrogen. !
m of points.1 have an e-mail from D. Jones to D.
m MR. BEARD: Yes.
p1 Fields datedJune 27th,'94 that says,HotTopic. pi MR. LANDIS: Was that information ni it says: To mar.imize RCS hydrogen concentration pl understood by the crews?
isj ensure that make-up tank level and pressure are tg MR. BOLDT:I believe so by those ;
m bei .: maintained at the top of allowable bands, ist e-mails, yes.
m douale exclamation point. l m MR. HiCKLE: I think Greg should answer l
! m I have a memo from Halnon to Fields which Iai that. !
tel goes through an explanation of how to meet the m MR. BEARD:As long as you could keep pm objective on hydrogen concentration.it says,in {
! pq above 20 pounds.The key was how much hydrogen i 09 effect,to raise the h. ogen pressure once a day to 09 overpressure would result in 25 cc's per kg,and our 1
na the top of the band and then pressure will decay na experience indicated it was 20,and that's that 5
pai below 20 pounds. psi green line.
pq MR. EBNETER: Are you familiar with !
pq MR. HALNON: Let's walk through this, pq these?
pq How do we communicate to the operators what they i pai MR. HlCKLE:I've seen the psi needed donc,I think that's the question,and what on correspondence. On did he mean by this.One thing is we have three I paj MR. BEARD: Yes. pai e-mails.1 think that's all the communication we i poi MR. HICKLE:I saw the correspondence poi had, but that's wrong.1 was in the control room pm recently "At the top of the band" should mean in pm daily when I was on site to talk about this.We've 94 this area.I don't think that means,at least the pu talked about what we mean by keeping pressures up, na way 1 interpret it when I read it,it doesn't mean pa by keeping cc's per kg up,and discussion ensued '
psi on the curve,it means at the top of the band. pai hem these e-mails. Operators knew they had to keep pq MR. GIBSON: What does "at the top of the pq the pressure above 20 pounds consistently to get 2
pq allowable bands", double exclamation point,mean to ps)there. '
Page 108 Page 110 p1 you as an operator? p1 We have allowable bands all over the 4
m MR. HICKLE: I think these are the m place.You have an allowable band down here,a load
, p) tilowable bands right here, p) level one.You have an allowable band on the high pi MR. GIBSON: So on the top is on the pi level,and you have an allowable band on the tsi curve? rsi pressure.You put those together and you have a ist MR. HICKLE:I didn't point to the isi preferred region,and that's where we wanted them to m curve.I'm pointing to the area under the curve, m k cp it.That's where they knew they had to keep j ist MR. BEARD: Maybe we should have Mr. Inj it.Different operators had different confidence in m Halnon now -let's understand who is- m their ability to keep it there; different operators l
pq MR. HALNON: May I see the e-mails? pm were more versed to keep it there.The burden that i pq MR. BOLDT: You got two or three. 09 we talked about was the increased operator attention na MR.GlBSON: We only have two. na required to keep it there.
pai MR. BEARD: At this point on the record. pal The actual adding water,the actual 04 if I'm correct,our plant manager said that he did pq adding hydrogen is not a complex situation,and it l
) pq not tell the crews to operate on the curve,and that psj really wasn't a burden.It's the attention that ps) is a true statement.Now,let's talk about what did psi they had to do to keep it there. i on the crews to do what.That's where we're going,. :1 na MR. GIBSON: But Curve 8 didn't have,if psi one of them is right here,and we'lllet him speak pai i remember correctly,it didn't have the horizontal
- nel for himself. poi line at 20 psi.The operator had one single curve pm MR. LANDIS
- While Mr. Halnon is getting pm drawn on an axis and was told to keep it at the top pq ready to do that,can I clarify one thing? pq of the allowable band.1 would have thought that he ma Maintaining yourscifin the box that Ilruce had just ma might have thought keep it as close to the curve as
, ps; shown on the far right-hand lower corner ofit well psi possible.
. pq away from the curve for an extended period of time pq MR.HALNON: And the e-mails say,While
, ps) would have produced greater than 25 cc's per kg of pq maintaining in compliance with the curve.
Page 107 - Page 110 (30) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
4.s su.4 PROCEEDING 5 MEl ORE STEWART EBNETER, CHAIRELN
, PJtEDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 111 Page 113 pi MR. RICHARDS: You just pointed out that p) reports that we shared with you.In fact,in our pi the upper and lower levels were defined in the pi letter of December 27 we made that very clear.
pi procedure, so those numbers are available to the p3 looking back,I wish we had said,*Iook, pj operators. In defining the band, did you ever put p) we realize that this is an additional burden for now pi down in an e-mail or writing that the band is mi until we can figure out how to alleviate it forever, mi consistent with the curve in the upper and lower sq we want you to maintain hydrogen,we know it can be m limits, stay above 20 pounds?
m done." And of course you still have an alarm,and rei MR. HALNON: The 7-25 e-mail,it's from isi of course we didn't know it was design basis then.
rq me, and it's actually forwarded from DaveJones.
pj Md we didn't clearly say that, not to get the pq That was from the experience that we knew that if we pq alarm.We're not going to say that we did.
pii kept pressure above the 20 pounds. Did we ever draw pij So what have we done about that? Now, pri it on the curve for them? No.They had a laminated pai you'll also recall somewhere around about this time paj curve in front of them and they knew approximately pai we had the Salem event in the industry. Salem shut pq where they had to keep it,what their job was, pq down and all this stuff and they found out that the pq because we had normal RCS !cakage, which causes the pq operators were living with what we call operator pq tank to dmp in level until they add level again, psi workarounds.One of the corrective actions we na and that was several evolutions to add level during on talked about,and I think again it gets to this paj the day, and they knew that depends on where they pai point,when I think the early plants said, hey,and pq put it here, that if they put it down here it was poi certainly in this region, maybe we've got operator pq going to take increased attention because it very pq workarounds so let's go focus on that.This is py quickly was going to go below 20, half a shift,just mij one.We have an operstor,a senior reactor operator p21 fmm normalleakage, p21 in charge now of bird-dogging that list.We have -
paj If the operators put it way up in here p3; frequent management meetings to review it just like pq they wouldn't have to give :: s much attention for pq about every other plant does,too.
psi a period of time because it was going to take longer pq So yes,this was an operator workaround.
Page 112 Page 114 p; to get down here.They knew that if they put it up p3 We realized we were having difficulties.We never p) in here they could keep it.Again,you're hedging gi said.though,that it's okay and we never said it's p) up against two limits,and that was a burden.
pi okay to get the alarm anyway to keep the hydrogen Iq MR. RICHARDS: In the interest of making pj up. Management,to my knowledge,being Hainon and pl sure what you expected was clear to the operators, is above,never said that.
iq was there a standirig night order or standing p1 Now, we probably should have said it's m expectation written in a book that said the goalis m not okay and don't do that, but we never said keep rei to stay above 20 pounds? isi hydrogen up there and get the alarms we have to and p) MR. HALNON:Just the e mail and the pi it's okay.We did not say that.
pq discussions we were having.It was an emerging pq MR. HALNON: There's one other.The pq issue that we were having to do this.There was 09 reason I wanted to read the e mailis because p2) ongoing discussion.Different operators came up pri there's one on 7 27.
paj with different ideas on different ways to maintain 03) MR.GIBSON:I've got 6 27 and 7 25.
pq it,and we had an emphasis on RCS leakage on our pq MR.BOLDT:let me add a bit of pq make-up system to try to not keep that happening as pq characterization here because I've been looking at pq quickly. pq these very cautiously,too.There was an evo ving on MR. BEARD: Irt me jump ahead and address on understanding which are characterized in time
~
pej what I think sort of-let me go straight to the paj sequence of those two e-mails plus the third one pq point from our side. Bruce is going to get to poi that Greg's about to talk about.There was am this.In hindsight, did we give the operators,we, pq instruction to stay inside the curve limits,the pq management above the shift supervisors, clear and pq limitations of the curve.They'd established that p2 consistent guidance and expectations on the balance pri the 20 pounds for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> would be well above the pa) between maintaining hydrogen pressure in the reactor ps; 25 cc's per kg,and in a number of cases it tells pq coolant versus don't get alarms? The answer is no. pq them to raise a level,and I would say in the latter sq And that's reflected in one of our own inspection t2sj case in particular,the 7-27 memo talks about BROWN REPORTING, INC. (404) 876-8979 Mla-U-Scripte GI) Page 111 - Page 114
PROCEEDINGS BEFORE STEWART EBNETER, CliAIRMAN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 115 Page 117 pl raising the level to the upper end of the band,as ty maintain the pressure as high as they could without
{
l gi Kerry pointed out.But those e-mails were being gj violating Curve 8.
pj developed again largely by their peer group,and the pi MR.HALNON:I think that's probably
+
si instructions we had given was to stay about 25 cc's pjfair.
Isj per kg, and they established the methodology which isj MR. GiBSON: Now, several times today you !
iq typically is how operators do Gat, how do go about Iq have mentioned,"We did not know that Curve 8 was a I m doing that acceptably,and thi $ basically the m design basis limit." Bruce, you were the last ial methodology by which they had established that. Im person to say that,and the operators of course have g CR. HALNON: This is an e-mail from Jones pl said this many times also.But I've also heard you pm to Fields. Jones was essentially working as my pq say that you expect procedures to be followed j pq assistant,DaveJones.7 99 whether they are design basis limits or not.
na MR. BOLDT: By the way,1 think rr:aybe one na MR. BEARD: Yes.
psi other point. Bruce,I don't recallif you made it paj MR. GIBSON:I wanted to clarify that pq in your presentation,but we'!! talk about the pq point.
ps) interpretation contact for procedures. For example, pq MR. HICKLE: May I clarify?
ps) the OP,at that time it was DaveJones.He was the paj MR.GlBSON: Please.
ps) interpretation contact for the make-up system na MR. HICKLE: It's true that we expect our pai procedure. pai operating curves and limits to be followed pa MR. HALNON: This is an e-mail provided pq irrespective of the technical basis for these.
pq to the operating shifts talking about that where we pq That's true.So why did I make a point about the my found that raaintaining hydrogen pressure above 20 pu design basis for the make-up tank curve? I think ga pounds for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> will be maintamed in cc's per " J pa that if our operators had known the technical basis i esi kg in the desired band which is above 25 cc's per psi for that curve they would have had a greater pq kg.it goes on to say that without performing the pq questioning attitude about the alarm condition.1 psi evolutions that we're familiar with on the nine ps) think they would have recognized the curve as more ;
1 Page 116 Page 118 l pj additional evolutions that we talked about,and it ni important to safety than other curves that they ,
ni goes on to say down at the bottom that the day shift pi have. !
pi supervisor should daily add H2 via the regulator and pi MR. GIBSON: So you understand that your pi then make an equilibrium addition to the make-up p; operators considered some curve, some procedures i tsi tank, which is squeezing the bubble,ta raise g, more imponant than others,but you seem to accept tai level.This is, quote unquote, to raise level to tej that if they can judge it not to be too important m the upper level of the level band,the level band, m then they don't always comply with it.
rei this way on the curve,to maintain a higher average tej MR. HICKLE: No.The reason that's im make-up tank overpressure while complying with the si important is this. lf they had recognized that this pq make-up tank curve limits. pq was a design basis curve and that it was sometimes pq He's saying get over here because you've pu challenged during hydrogen additions to the make-up ;
paj got more room and get up here because that's going na tank they would have questioned that more.They !
031 to put you above 20 pounds.The key was to maintain pal would hve questioned whether or not that should pq above 20 pounds average make-up tank pressure,so pq have been a valid operating curve at all. In fact, ps) we're clearly talking about this window right here. pq when you look at our corrective actions,you'll see ps) And that was the addon, probably the e mail to the psi that what we ha e provided is an operating curve pa two that you have,A1.1 think they were in the na that has quite a t t of margin to the design basis paj same group. psi limit.
pq MR. GIBSON: I'd like to clarify just one pq It's certainly true that all of the l poi point.1 guess, based upon what you have said,that pq limits that we establish for our operators to follow pq it's clear that you expected Curve 8 to be met.It 39 have different technical bases.Some,such as pai was not your intent that Curve 8 be violated. pa safety limits and tech specs.have dire consequences psi MR. BEARD: That's correct, ps) and safety significance, and some may not.They tnay sq MR. GIBSON: The operators interpreted pq be equipment pintection curves.We expect our gsi your direction,I think it's reasonable to say,to psi operators to follow all of these and,when they Page 115 - Page 118 (32) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
a n: 1*RULhhDIAb311L1 ORh 51hWAh! AnNhltR, LliAliO1AA PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 P:ge 119 Page 121 p) receive an alarm,to promptly address the alarm vi that is the design basis limit.That's what they pi condition,to follow the AR and to return the plant gi should have had all along.
pi to a configuration out of the alarm condition. pi And what they thought,at least at the pi The only point is that if they had 81 time, what they thought this one curve was and we la recognized the design basis limit I think they would is thought was an administative limit curve.So the n have come forward.I think the management oversight isj question is is it okay to violate an administrative m issue that we've been discussing may not even have m limit curve.Is it okay to get an alarm for an is) been there, because I think our operators would have tsj admmistrative limit.The answer is no,it's not pi told us.They would have said," Hey,I'm worried im okay, but sometimes it happens. I'm thinking about sq about this, do you know what the basis for this pq other times in the plant.Well,ifit does,then ny curve is and do you know I get these alarms 0g you're supposed to take immediate response.I use pri sometimes and sometimes they're in for a long period pri the word immediate,and I don't know how that's psi of time." pst defined,but you're to take response to clear the p43 MR. GIBSON: I think your operators, at 04) alarm,as distinguished from if you go across the pq least this particular crew, understood this curve psi design basis curve.That's totally unacceptable any est quite well. psi time.That's an LER.
pn MR. HlCKLE: To the best of our on MR. EBNETER:I don't disagree with pai knowledge,at some point they understood the poi that.lf you're going to discuss it on the next voi technical basis, but it was never communicated to pq violation,let's finish that one up.
pq the other shifts. it was never communicated in that pq MR.TANGUAY: I'd like to clarify that 94 way to us. pq Pat didn't know there was a design basis curve.
p2l MR. LANDIS: May I make a disti nction ga That should be clarified.No one within operations .
psi here? They knew the basis for the curve, correct me psi knew it was a design basis curve; engineering knew. 1 941 if I'm wrong, but they did not understand that that pq MR.LIEBERMAN: Allthe time?
psi curve was a design basis curve. psi MR.TANGUAY: Yes.
Page120 Pago 122 p1 MR. HICKLE:What's your impression, p1 MR. LIEBERMAN: So when the problem pi Greg? pi report was written,did they understand it was a pi MR. HALNON:I think we all knew it was a pi design basis curve that had been violated?
g; curve to prevent hydrogen binding. pj MR.TANGUAY: The engineer knew.
[si MR. BEARD: We actually have progressed im MR. LIEBERMAN: Was he one of the persons is into what we were going to talk about in the next to who knew about the issue of September 5th?
m violation and where we're going to get to exactly m MR.TANGUAY: No. September 4th,you isi this, but I'd just like to make a comment. si mean?
pi First of all,no one, including the pi MR. LIEBERMAN: No,the 5th, pq operators on the 4th and 5th, knew that this was a pq MR.TANGUAY: Once the problem report p u design basis limit curve.They understood the on came out,then just about everybody would have known pri intent of the curve.What they didn't know was that pri that.
ps) if we got on it or went across it we violated the pai MR. LIEBcRMAN: So engineering knew,1 pq design basis,which is a one hour report LER and all pq guess,on September 7th that the design basis curve psi of that. And I think you said, and rightly so,that pq had been violated.
pq regardless of what the curve is,is it okay to pq MR.TANGUAY: It was a design limit pa violate curving the alarms,and the answer's no,but on curve.They knew that.
~
pq with a caveat.I guess if the operators had what pq MR LIEBERMAN: And that it had been poi they now have - and again, what we really did wrong poi violated.
pq with the operators is to give them a design basis pq MR.TANGUAY: Yes.
pq curve as an administrative openting curve.That's pq MR.LIEBERMAN: Why was the LER written pri what we did wrong.What they now have is an pri in November?
psi adrninistrative curve here. Above that is a curve pai MR. GUTHERMAN:I can explain that.1 pq where they get the first alarm, a computer alarm, as above that they get the annunciator alarm,and above lpq was basis pq design responsible for itdetermining issue.whether whether it was was operation outside BROWN REPORTING, INC. (404) 876-8979 Min-U-Suipts (33) Page 119 Page 122
PROLEEDING5 BEFORE STEWART EBNETER, CIIAIEMN " IN RE: {
March 270 1996 PREDECISIONAL ENFORCEMENT CONFERNECE !
l Page 123 Page 125 pj design basis.That was a single curve for a large pj MR. BEARD:I'm looking at the problem m break LOCA which bounds all the IOCA sizes.When j
m report in question,267,which was dated 9-7-94. !
p1 the event happened and I was assigned to describe ni It's checked as a suspected design basis issue,and pi that,I knew intuitively that high pressure p1 then you can go through and see what the corrective l tu injection was not credited for large Feak IOCA tsi actions were which culminated on November the 16th
[q mitigation. I also knew that the curo protected sq and we confirmed that it was.1 was in a meeting a the high pressure injection pumps a I went m where this was presented by Mr.Gutherman.We said tai through it,I needed to find out whether in this ist engineering knew it was a design basis curve on si specific case for that specific accident whether we pi September the 7th. llave we fully clarified what pq would have destroyed equipment necessary to mitigate pq that means?
pq a specific design basis event.
nii MR. GUTHERMAN: That means that in the pri Knowing that IIPI was not credited in a pri absence of any other curves which are LOCA break pai large break LOCA,I thought that perhaps it was pai site specific it's a bounding curve,and it must be oci simply a procedure violation but not operation pq adhered to for the entire spectrum of break sizes.
pq outside design basis. Going through the review pq That's why it's a design basis curve in the broad par cycle with my peers, having put that memo together pq sense.
071 severut times,we came up with that one scenario of pa MR. BEARD: And we confirmed that on the par a core flood line break that's desenbed in the pai 16th, poi memo,which put us outside design basis of the nei MR.GUTHERMAN:And we confirmed on the pq plant.That took two months to get done. pq 16th that there was a specific size LOCA, location ny MR. EBNETER:And who were your peers? pq LOCA and a number of other failure mechanisms where pri MR. GUTHERMAN: There was an individual pri equipment necessary to mitigate the accident would ps) - I was in design engineering at the time.There paj have been damaged by violating that curve, pq was an individual in licensing, Paul Fleming. He pq MR. LANDIS: And that was on the 16th of ps) was my primary counterpart in giving me a critical pq November?
Page 124 Page 126 p) review of what I was doing and helping me find the pi MR. GUTHERMAN: That's correct.
m B&W information which talked abvut the core flood m MR. EBNETER: Could I ask you,and it's pj line break event. pi just informational,but what took you two months? I ;
pl MR.LANDIS: Going back to clarify,on si understood you had to go to B&W to get this data;is j isi September 7th, then, you were suspect that it was a [si that correct? !
in design basis curve,and then you were doing some tq MR. GUTHERMAN: I received the assignment i m evaluation. m as one of the corrective actions on the problem l tai MR. GUTHERMAN: Yes. gal report in roughly mid September.1 was supervising j ici MR. LANDIS: And what time did you make p1 the design group at the time in St. Petersburg.
pq the conclusion that indeed you were outside design pq There were a number of other priorities that I had ;
pq basis? '
09 to attend to.1 was not dedicated to this effort.
pri MR.GUTHERMAN: The day of the memo, pri in retrospect,I perhaps should have delegated that ;
pal which was November 16th,I believe. psi to one of my engineers but chose not to do it,and pq MR. BEARD: November 16th. pq it just took me two months to get the various pq MR. LANDIS: Who made that decision? ps revisions of this memo thmugh a review cycle and pq MR. GUTHERMAN: I was the author af the pq incorporate comments.
pn memo incorporating input from a number of n v peers. p7i MR.EE ETER:I'minterested pq MR. LIEBERMAN: When you suspected it paj fundamemally, which is just a little bit outside of poi might have been a design basis,did you raise that voi the enforcement type thing, but if you had a um to operations or did you want to wait until you were um situation today where this occurred,would it take pu sure? pn you two months again to get this information from pri MR. H ALNON: I believe the problem rcport p21 B&W? Do you not have this design basis information pai actually checked the box, suspected design basis pai within the engineering department?
taa issue on it,which tells you that that's what caused pq MR.TANGUAY: We have the bulk of the
- 2s1 the design basis review to start happening. psi information in-house, but bringing this information l Page 123 - Page 126 (34) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
ta su.: enverwnwa ou unc a a o muu coat A La, s suunata
. PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 127 Pzgo 129 pj in-house and having a thorough understanding of the p) by not telling them it was a design basis curve.
p) what the content is and what it means, we don't have gi l'm trying to understand.You seem to acknowledge pi all that knowledge in-house. p1 that operators treat design basis curves differently si MR. BEARD: Let's talk about what would 141 from administrutive curves.ls that your ts) happen today,because we haven't gotten to that. (s) expectation and to what extent does that occur?
pi Today someone would write a precursor card.it p) MR. BEARD: No You're right.The m would be reviewed the ne'.t morning and someone would m problem is we get so involved in these things and we I pi take it from there in our operability process.What pj keep asking what if, what if, what if.We got into (q would happen then? p1 a mode of thinking,it's really hypothetical,and I pq MR. GUTHERMAN:In alllikelihood - pm you're right.it's not worth talking about.Our I pq MR. BEARD: What would happen? pij expectations are if you have a curve,if you have a prj MR. GUTHERMAN:It would be called a na limit,you're supposed to adhere to it. If you get ps) suspected design basis issue,and then there are paj an alarm, you're supposed to do something about it.
p4) some subjective - 04j That was our expectation then and it's our psi MR. BEARD; We would enter CP what?
ns) expectation now, no ifs, ands or buts.
ps) MR. HALNON: 150, but we report suspected poi MR. BOLDT:I've stated on the record on design basis issues as design basis issues. On that it didn't matter to me whether it was an ps) MR. BEARD:!t would be assigned a ps) operating basis curve or any other basis curve,it pei certain level of timeliness for corrective action. poj was a curve to be followed,and our AR procedure in pq and then what? pq fact said that when the alarm comes in you're in an ny MR. HICKLE: And a problem repon would pq unacceptable region of the curve. Unacceptable ,
na be generated.We would also do a formal structured pai means get out ofit. - I ps) operability assessment per CP-150 which would pai MR. EBNETER: If you said otherwise,I p41 establish timelines for resohGn of the technical 94) would be very surprised,but do you have on the ps) issue, and that entire CP-150, until closed out, ps} record any audits, oversights or anything that would Page 128 Page 130 p) becomes the custody of our shift manager, who pl show that these types of things were regular ;
p) reports to me to ensure timely closure and to p) occurrences within the Crystal River operations over p1 coordinate resources if necessary to make that p) the past two or three years?
p1 happen. p) MR. BEARD: Our investigation of August isi MR. LANDIS: Do you have timeliness 151 of '95 explored that quite thoroughly, so yes,the isj guidelines in that? pj answer is yes.We've had an audit,and that is not m MR. HICKLE: Yes. m expectation,and it never was.Really,we're into pi MR. LANDIS:llow long would that have p1 the next violation.
pi been? pi MR. EBNETER: Then let's move on.
pq MR.GUTHERMAN:The shift supervisor and om Gibson, you're in charge.1xt's keep this thing pq shift manager make a decision on the subjectiveness 09 moving, pri of the issue and safety significance of the issue. nzj MR. HICKLE: Let me clarify one thing I pai MR. GIBSON: The point that I want to get ps) thought I heard you say,at leart the way I v4j clarification on before we leave this is your p41 interpreted it. Mr. Ebneter's question,the way I usi expectations regarding pmcedural compliance. ps) understood it,was are these regular occurrences poi Whether this was a design basis issue or not,1 pej that operators are violating limits and if that was on think it's fair to say that most of the parties pn the question, the answer is no.There were no
~
pai involved understood it was an important curve, pai assessment reports or results of any kind. including pe; MR. BEARD: Yes. poi management assessments in the simulator and in the pq MR. GIBSON: That being on the wrong side pq control room,which are donc routinely,to tell us pu of i; could vsult in cavitation and loss of safety pq that there was a blatant disregard or even a casual pri initxtion. Ses cu1 times in the presentation there pri disregard for operating limits or curves.
pai have been retnarks to the effect that you believe if pai MR. BEARD: I think what you just said, p41 the operators had known it was a design basis curve p4) 13ruce,I think what you meant to say is all of the psj they might have 2cted differently,we made a mistake ps) audits, surveillances and personal observations by BROWN RFPGRTING, INC. (404) 876-8979 Min U-Scripte (35) Page 127 - Page 130
PROLLEDIMid liEFORE 61EEUU LliAE'lER, WiAIRMAN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 131 Page 133 {
14 you and others did not reveal other instances of pj Pat talked about a formal investigation that we m deliberate or inadvertent or whatever violations, ni conducted. I won't get into that any further. Some p) MR. HICKLE: Or casual disregard for p1 further crew discipline was taken based upon what we pl operating limits. si viewed as an integrity issue. Pat also discussed ta MR. BEARD: You took him to say we didn't is) that.We also further revised our procedures to do l Im have any audits to show that. I think what you said te just these things, expand the scope ofirdrcquently m was that all of the oversight audit surveillances m performed tests or evolutions,a checklist,and i im found other occasions where this was the case. tai that's the screening criteria that the operator uses (9) MR. HICKLE: That's true.There's [q to determine whether or not he's performing a test om routine management self-assessments that are done in om or an experiment,and to reqture director approval n y the simulator and also in the control room,and one pq of the checklist.When that screening is completed na specific criteria that is assessed,it's part of a na it requires my signature or my signature authority ps) checklist,is alarm response,and deficiencies have 03) to review it.
pq to be identified.And we specifically look for pq MR. LIEBERMAN: Bruce,are you satisfied psi promptness, use of AR procedures and thoroughness ne that there's adequate guidance as to what defines a pq when those procedures are invoked,and there is not psi test?
pn a record that we have in conducting those on MR. HICKLE:I believe that there was poi assessments or from any of our QA audits that we ps) adequate guidance in place,and I was going to talk pel have a casual disregard for these limits. poi about it in management issues, but I believe there pq MR. LIEBERMAN: That has to raise one pq was adequate guidance back then.We also pq more question, and that is, given that with this pq incorporated a CAPS approach in the criteria for pa particular curve there were a number of instances pa determining procedure adequacy and sL2 '
psi where action wasn't taken in what I'll call a psj supervision.It provides guidelines for pa reasonable time to get back to the curve - pq communication, approval. planning and scheduling pq MR. HICKLE:I want to show that to you psj when the shift supervisor has a question about the Page 132 Page 134 p1 and explore that.1 think you have to understand pj intent of a procedure on shift,and that's in our m how the operators were interpreting their p3 conduct of operation procedure A-I-500 now. l p) information and the options that they had to p) We also recognize that the MRC process, pl exercise for system control.I was specifically pi although not bad,didn't go far enough, so we've tsj going to talk about that when we get to the first tsj expanded that application and we've formalized it as rei violation. [e part of our noimal way of doing business to handle m MR.GlBSON: I think we really need for m corrective actions. At the level where we have a tai you to proceed through your presentation and then tej potential NRC violation or another significant issue pi we'll stop and ask our questions at the end. Im related to safety documented in a problem report, pq MR. HICKLE: Continuing, then, with the um the MRC is required to meet and to review the root pq MRC recommendations that we generate a procedure or pq cause evaluation and the corrective actions that are na work instructions as appropriate after the fact for na taken,to meet with all of the people that are pai the make-up tank overpressure test, counseling for psi involved with the problem and determine the adequacy pq the operators on shift, validate the make-up tank pq of those and provide recommendations where we need psi hydrogen overpressure curve and reissue it if ps) to go further.
pe necessary, review ph.nt modifications to ensure that nel MR. LIEBERMAN:llow is that going to on operator burden is minimized and rensit the na relate to your human factors performance review pm technical justification for the 25 cc's per kg limit pai procedures?
poi in the reactor coolant system.All of these poi MR. HICKLE:In this way.The MRP pq recommendations we established a schedule for,they pq reviews the root cause,the structured root cause pq were tracked by our quality progmms,and they were pq evaluation attendant with these problems and,in pa all completed by December of 1994. pa doing that,that structured root cause now is pa Some additional corrective actions were ps) required to address the roots of human performance am taken after we learned about the evolution that was pq issues.
psi conducted the night prior, September 4th of '94. pq MR.LIEBERMAN: But prior to the MRP Page 131 - Page 134 (36) Min-U-Scripts BROWN REPORTING, INC. (404) 876-8979
~
am imutetulw3 512vde Utmuu unta ut, u1AndtAs PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 P ge 135 Page 137 ty process?
pl tnined personnel capabic of assuming leadership d
pi MR. HiCKLE: Prior to the MRP process m positions in operations very shortly after the pi they have to bring all that information to the MRP. pj examinations that are currently being conducted.
pi Now,the MRP is chaired usually by myself,but if 81 The significance. I'll be more specific si l'm not available another director chairs it.We al on the first point in the next violation.The l ri have at least three directors present on the MRR tel safety consequence of the test was very,very low, m The intention is not to apply pressure to our m but the human performance significance and the pj people, and we've communicated this, but it's to pl regulatory significance was extremely high. It was j pi require and structure management involvement in the pj very high to us. Fundamental operating principles
] pq corrective action process. pq were violated,and similar judgmental errors,if j pq We've reinforced our tog keeping 09 they occurred under different situations, could have l' p2 practices,I think we discussed that earlier,and pri serious impact, serious safety consequences, we psj we've provided add 4ional training using examples ps) recognize that, and for that reason we tried very p4i describing shift supervisor authority. And that p4) hard to be extremely thorough and comprehensive in psi training, we provided examples for the shift pq the actions we took to address this.
l l pq supervisors to diagnose whether or not they had the ps) Now, the next thing I wanted to discuss on authority to authorize certain hypothetical on was the first violation.
l pai evolutions to occur,and then we critiqued their sq QUESTION FROM THE FLOOR:I have a couple
- pq answers. poi of questions.I just want to go back to the design i pq So what are the results that we've pq basis curve. Did operations know? By operations,!
- y achieved on this with respect to the second pu mean the operations managenmat, not the operators.
i p2j violatw?There's absolutely no doubt that there's p2) When did they know that it w .., a design basis ps; an increased sensitivity throughout operations as to p3) curve?
g4) procedure use standards. I mentioned earlier we p4) MR. HICKLE:Around November of '54.
psj developed a procedure to talk about how to use psj Well, the problem report probably preceded the LER.
Page 136 Page 138 p) procedures.That was developed by a team of our p) MR. HALNON: That was the memo that Brian m operators representing operations.We're using them m wrote.
p1 and we are developing experience with it.We're not pi MR.HICKLE:It was in November,well pi using it perfectly,I won't tell you that,but right si after the evolution.
Isj now we're seeing a lot more consistency on procedure i 151 QUESTION FROMTHE FLOOR:How do we iq usage from shift to shift.That's a positive tsj ensure that there are so many other design bases m result,in my opinion. m that they may not know the operation?
pj The initial procedure barriers we think p1 MR. HICKLE: That's a very good pi are going to help ensure that similar judgmental pl question.The first thing we did was we went back pq errors are avoided. Again, defense in depth. I can og to our plant curve book,OP-103,and we did an 09 tell you we haven't had any more unauthorized tests pq operational review of every single curve there to pri or experiments and we don't believe we will based on 021 see if there were any other situations where we ps) the training and pro edures we've got in place.But ps) might be challenging a curve,where there might be p41 we don't think that i,ur job is over by any means at p41 an issue of sufficient operating margin,and we pq all.The lessons learn:d are going to have to be psi didn't find any that gave us concern.That was the psi continued to be reinforced to ensure that these ps) first layer of corrective action,to address your pa operating principles are firmly ingrained not only pa point.
ps) in the operators we have now but in the operators pq Now,the next level's not completed.The poi we're bringing up through the ranks.To that end poi next levelis to go back and do a systematic pq these things have been incorporated into our pq engineering design review of all of our operating pu trauung program. pn curves and develop or bring forward the technical p2j Shift supervisor leadership has been p2] basis for those curves to see if there are any rm improved by some of the changes we've made.We psi others that are design basis.On the surface it pq don't think we're done yet,and we've got plans in R41 doesn't look like we have that,but until that ps) place. In fact,we believe that we'll have SRO psi systernatic review is completed - Paul, do you want BROWN REPORTING, INC. (404) 876-8979 Min-U.Scripte (37) Page 135 - Page 138
PROCEEDIN65 bel OME 51EWAR A EBNETER, CIIAIRMAN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE I Page 139 Page 141 vi to say anything about this? j pj suspected at that time. j m MR.TANGUAY: We have talked about doing m QUESTION FROM THE FLOOR:So coming back ;
pi that. I don't have an integrated plan at this point pi to the question I asked, should a certain thing be I pi in time to complete that activity,but we do intend pi reportable at that time because it's suspected to bc is) to look at it. ist design basis or only when the evaluation is isi QUEO .ON FROM THE FLOOR:The order of {
isj completed,then - becomes reportable? j m design basis,what's your reporting ability?
m MR. HICKLE: Lensing does not determine '
pj MR. HICKLE: We report suspected design pj what's reportable in the plant.That's a line m basis issues right away.We don't wait to confirm pj function.That's operations' responsibility.
pq that they are in fact design basis.lf we think pq That's my responsibility today.The wayit would on they might be,we make an immediate report. pq work today, all of our engineers,all of our pri MR. LIEBERMAN:Is that a new policy? pri technical support people have been instnicted that pai {
MR. HICKLE: That's our policy now. pai at the time they suspect an issue may be related to I v4; MR. LIEBERMAN: So since September?
p4j design basis,do not wait.They're to notify me pq MR. HICKLE: Yes.
psj personally and,in my absence, one of our assistant pej QUESTION FROM THE FLOOR:The operability pai plant managers. At that point we review the issue, j pn examination was CP-150?
On and ifit's a suspected but not confirmed design pai MR. HALNON: Right now it's CP-150. paj basis issue we report it right away.
pq QUESTION FROM THE FLOOR:Before that psi MR. EBNETER: Let me make it clear that pq point in time you make the operability examination pq suspected design basis is not reportable.If you un you may have some big conditions and you may pq choose to do that to give you margin to avoid these paj continue to operate but then you have to make a pri types of things, that's up to you,because it you
- pai reportability exattuation. pai make an error you get into the type of situation p4) MR. HICKLE: That's done separately. p41 we're talking about right here.
pq CUESTION FROM THE FLOOR:In September {
psj MR. BEARD: Our whole process, as I've i Page 140 Page 142 pj 1994,when you suspected this to be outside the ni said before, and I think the region certainly.
- 21 design basis, did you make a reportability call? m understands that,is that with yourinput we've 91 MR. GUTHERMAN:It was determined to be p) greatly strengthened our whole pmcess for si not reportable pending the evaluation that was p1 operability calls, including reportables,which
- sj ongoing to determine whether it was outside design y currently culminates in a CP-150 which I think is si basis or not. 51 quite thorough.It addresses both the design basis m CUESTION FROM THE FLOOR:So at that m operability and the reportable with that,and we're mi point in time a suspected design basis would not pl much more sensitive and we're much more likely today ini have been a reportable thing? p) to report something is suspected if we think it's pq MR. GUTHERMAN: That's correct.There pq likely or to agree on the safety significance of pq was subjectivity involved between the shift on it. 4 pri supervisor and the shift manager based on the safety pri MR. EBNETER: But that's your choice; l pai significance of the issue. pai that's not by regulation.
0 41 CUESTION FROMTHE FLOOR:But engineering p4j MR. BEARD: Yes, sir, sq made the decision in September of '94, September 7th pq MR. LANDIS: Larry Kelley, how many times
- 06) of 1994.that it was suspected to be outside the psi have you reported potential outside design basis l
pn design hasis, but then you did the analysis and came nn since Septemberof1994? '
pai to the conclusion that it was outside the design paj MR. KELLEY:I can't give you a number, nei basis and came out with an LER in November. poi but we've reported several.
pq MR. GUTHERMAN: That's correct. pq MR. LANDIS: Do you have any feel for pq QUESTION FROMTHE FLOOR:But then it was pu that?
- 22] not reported at that time. p2) MR.GUTHERMAN:The number of LERs for pal MR. GUTHERMAN: That's correct. [23; calendar year 1994 were 28,and my judgment is that
- 24) QUESTION FRCMT:iE FLOOR:Why? p4125 of them perhaps were outside design basis.
- re MR. GUTHERMAN: Bec;use it was only pq MR. BEARD: You said for '94.You meant Page 139 Page 142 (38) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
u at: -- Asutttute utt ont 31twMt1 LuAL1hK, LIIMRMAA
. PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 P:ge 143 Page 145 pj1995. pi MR. EBNETER: That's fine. If we need to gj MR. GUTHERMAN: I beg your pardon. '95. m explore why you think this is now and why it wasn't pj MR. KELLEY: He said suspected design pj ten years ago,we need to do that in a different gj basis issues.
H) forum.
Iq MR. GUTHERMAN: I'd say a large majority isi MR. BEARD: And we can do that.
is of those were suspected rather than known design isi MR. EBNETER: But we need to keep this on m basis issues. m the enforcement basis here and not all the other lei MR. HICKLE: We just feel that it's inj things.And I'm not trying to stifle the staff.If pj better to communicate the issue at the onset and,if p) you insist on doing it, you can do it, but I'm pq we need to,once we're finished with our assessment pq telling you you're not going to get through this pu to clarify it. pq meeting. If we need to have a meeting to discuss na MR. EBNETER: That's enough on that. Da some of these other issues,then we can do it, but paj MR. GIBSON: We have a ntunber of psj we can do that later.
pq questions on this violation that I would like to pq MR. IMBRO: My point was the timeliness
- ps) review with you briefly We don't need a long psi of the LER.
- pq explanation unless it's warranted,but these are psi MR. BEARD
- I think what people are on questions that are based upon - nn really trying to say here, Stu,is that somebody in vej MR. IMBRO:I just want to make one point usi design engineering in St. Pete, where they were at nn before we get into that. I'm trying to reconcile poj the time, somebody must have known that this curve pq the fact that you don't take credit for HPI for pq was based on a design basis and there was no margin pu large break LOCA,and I'm not sr e how that connects pg for operator action, and we agree. At some point ga to whether those curves were design basis and sa this curve was given to operations, and why didn't .
, ps) whether Curve 8 was a design basis curve. Clearly psi they, operations, realize, Gee, okay, therefore as the system wasn't designed to do this HPI on large pq we've got to give the operator something over here."
pq break LOCA whether you needed it or not. It seems pq If that's the issue we can address that.
s Page 144 Page 146 p) like a real straightforward question. p) MR.EBNETER: But separate from this
, m MR.GIBSON:I think we're mixing m meeting we need to have another meeting.There are p; licensing basis and design basis.1 understand your pi numerous questions raised about the availability of pi conclusion was that HPI was not part of the kj the design information and how you're doing things, is licensing basis for most accidents,but HPI was In and there have been a lot of revisions to the si clearly part of the design basis for the plant It ;q process, so we probably should naedule,and I'll m was designed to inject at high pressure and.if m leave that up to Kerry to schedule another meeting is lost. if the rnake-up pumps were lost you did not tai to discuss all these other issues.Maybe it should pi have high pressure injection as designed. isi be held at NRR.We'll talk about that.
pq MR. BEARD: At some point I think we're pq MR. BEARD: Could I suggest that we do 2
py going to take a break and recaucus.1 think we need 09 the questions that relate to this violation? I pa to clearly think about what your questions were.1 na think that a lot of the questions we had earlier may psi know that everybody here is anxious to give you as paj really reiste to the alarms and all that.Once we 4
sq answer,but I think right now I'm a little pq do that, maybe we ought to take a break, pq confused. pq MR.GlBSON:I think we should take a psi MR. EBNETER: And I agree with you.We psi break for lunch, but I would like to get through
],
pn need to keep this within the scope of the regulation on this violation before that.
~
pq and the enforcement conference here and what the p.3 we have read of course the August na issues are and not get into redesigning the plant at ps; internal investigation that you conducted, August of am this point.We all agree that this curve today is a pq '95,and in there you identified other tests that pu design basis curve. 99 had been conducted.There was a test, for example, na MR. BEARD: Yes. pa turning off the fuel cooling and monitoring the psj MR.EBNETER: And we agree that it was psj temperature rise.Have you investigated further to Iul violated by the operator, pq establish whether or not those occurred and to pq MR. BEARD: Yes, pq determine how pervasive conducting tests without BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (39) Page 143 - Page 146
~~ ' ' ~
PROC $EDtNGS lil10RE STEWARf EBNETER[C11 AIRMAN IN RE: i March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE i l
Page 147 Page 149 pi procedures and 50.59 cvaluations has been to the oj for these things,and it was specifically addressed pi station.
pi in our procedure use standards. '
pj MR. BEARD: Yes,and I'lllet Greg and p) In fact,when I talked about the CAPS j
pi Bruce talk further about that. Realizing in that pj philosophy earlier,that's exactly the kinds of tsj report the time frame in which again the team was is) issues we used as examples,and we even have si trying to be very thomugh, talking to a lot of pj examples in our operating standard.
m people,and the time frame in which some of these i
m Now what do we expect? We expect our si things were done or were not donc pmbably goes back pl operating crews to use their procedures consistent si to 1978. Go ahead, Greg and Bruce.You can talk p) with the intent that they were designed for or l pq further about that, om written for. Every procedure has a purpose pq MR. HlCKLE: Let me start out and then no statement.That procedure I referred to, that OP, ;
pai l'il turn it over to Greg.We did go back and p21 the intent is to swap tnins,to take trains out of 2 pai inspect that instarLcc that you talked about to take ps) service for maintenance,to bypass filtration !
04 a look at it.We didn't do that exhaustively v41 systems for changing filters but not to isolate all ps; because we didn't have really enough information as ps! cooling to determine a heat-up rate, because that pei far as exactly when that occurred and under what j ps) would be outside the safety analysis of that OP. ,
on conditions and who did it.It was more like on The 50.59 cvaluation uncovered that OP. ;
pej hearsay, but we went back and looked at it. ps; What I'm saying to you is yes, there was, na l'Il let Greg address the investigation, poj and that's one of the management issues,but yes am although it really wasn't a structured pm there were things that went on on the shift that !
pn investigation.There's no question that in the past py were on the ragged edge of the intent that the pai we used our procedures without as much careful paj procedure was written for,and we belicse diat we've '
ps) thought to the intent of those procedures as we do pai addressed that.That's one of the thines that we p41 today. And let me just take that evolution that 94j tightened up.That's one of the lessons that we psi you're talking about and explain to you how our psi learned.
Page 148 Page 150 p) procedures were probably used. p) MR. LIEBERMAN: So you're satisfied that p) Our operating procedure for the spent pi you've pursued these other potential unauthorized p) fuel system provides specific guidance for starting pi evolutions that were described in your report to a 81 a spent fuel cooling train, for securing a train, pi great enough extent that your corrective actions
[si for bypassing filtration systems, et cetera,and all isj encompass those issues so they won't happen again?
Im of those are sectionalized in the operating p) MR. HlCKLE: That's exactly right.
m procedure.In this instance our operators or our m MR. HALNON: Let me give you another Iai shift followed the section of the procedure which p1 reason why, Mr.lieberman.We're really talking sj talked about the normal equipment shutdown,a!! owed pi about only one other shift supervisor.There was om the spent fuel pool temperature to increase, but om only one other shift supervisor who'd been a shift pq within the limits of their procedure to determine on supervisor since just about day one that the plant pri what the heat-up rate was.Now,we believe that pri was operated,and I'm convinced that there's nobody psi occurred. paj in this world that knows more about this plant than pq What's different about it is they then 04) this guy,and he knows more about the plant because psi restarted the spent fuel train with that section of usi he did these types of things.
psi the procedure. First,let me tell you how it's pe) Another problem with this shift on different than :hese tests and then let me tell you on supervisor,and! was one of the ones that we psi what we expect.it's different than t hese tests pai identified early w. that we needed to rotate off, ne; inasmuch as we know of no other mstances where poj was his address of human performance issues.During pq decisions were rnade to use procedures outside the pm the week that these tests happened,the 4th and 5th, pq intent where that led to violation of operating pq IMPO was in doing a test on the simulator,and I was pai limits or intentional violation of operating p2) full-time with them at the training center.The
- 231 limits.That's what's dtfferent about these make-up pai shift supervisor was one of the shift supervisors I pa tank tests.Nevertheless,that's a latitude that we psi became concerned with when we went back and looked lpai had counseled psi performance before problems.We about addressing had a significant human human Page 147 - Page 150 (40) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
IN REr PROCEEDINGS BEFORE STEWART EBNETER, CIIAIRMAN PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 151 Page 153 pj performance problem with IMPO there,I addressed it vi that we shared with you,and this is one of the m to him again,and he essentially refused to address pi things that we charged the team to look into.
pi it to IMPO standards,so he was pulled off the shift pi Quote: What are the generic implications or extent p) the week after.
p) of the condition, e.g., did the crew perform other Isi When we went back and looked at the Isl unauthorized evolutions and were unauthorized pi August investigation and saw that he was irr olved ,
pj evolutions performed by other crews m with most of them,it all connected up,and that was m That was one of their charges, and here is the main corrective action.Now, going right head isi are some facts. Interviews with otheroperations pi to head with him,I asked him,"Would you have ever pj personnelindicated that several plant evolutions
, pq violated a procedure to get the data you need",he pq which primarily gathered data and which did not (nj said no.No disrespect meant, but he's looking at 99 exceed limits had been performed at various times in pa the procedure and staying in the bounds ofit but, na the past including the spent fuel pool thing which paj like Bruce said, he's on the sugged edge ofit,and pq we've talked about.Two, securing reactor cavity
, pq that's why we considered that corrective action.We pq cooling to determine reactor cavity.
pq talked to other shift supervisors in that time pq Now, except for the spent fuel cooling psi period, and I don't have all the details, but I pq thing, as I recall, as I review this data and ask pn wouldn't have done tha!.He was already pulled off
{
- on questionsghe last three exampics are things that psi shift for another reason.
psi MR.GlBSON:Is he still off shift?
ps(@ did probably before I was a senior VP ps{ ops.Is that correct?
pq MR. HALNON: Yes. pq MR. HALNON: Since I can't ret 11 the i pq MR. RICHARDS: I think he was asked if he pu time frame, Pat,I can't characterize that.
pa would perform the kind of tests that happened on the pa MR. BEARD:I can't speak to what pal 4th and 5th.I can't remember from your report psi managentent's expectations were in '88 or '89 pq whether he said yes or no. pq or '87.
psi MR. HALNON: As a matter of fact, he was pq MR. GIBSON: What were expectations in
, Page 152 Page 154 i
{
pi the shift supervisor that came up to depress the vi '94? The environment in '94 of course is based i m squeeze.He told me he never would have gone over m upon events that have transpired and communications i
pi the curve,that that was one thing he never would pi that preceded '94.As an operator in '94, might I pi have done as far as violating a limit.He was pj reasonably have concluded that conducting a test tsi pretty adamant about that.I'm rrt sure what he tsj such as was conducted on Septernber 4th and 5th would pi came up with in the report, but that was a personal Im be okay with my management? l m conversation I had with him. m MR. LIEBERMAN: This gentleman that we i is MR. GIBSON: One of the issues that we pj were just talking about said he did not see anything l pi must assess is the environment that existed in pi wrong with the evolution conducted on September um September of 1994.We have evidence here that other pq 5th. He was under the impression that the standard py unauthorized tests were conducted without 50.59 og operating procedure allowed operations to prove a na reviews. Albeit limits may not have been exceeded, 02) curve. He is now,I gather, manager of operation ;
osi tests were conducted that would have seemed to have psi support.
pq required a 50.59 review.There's some indication pq MR. KELLEY: This i. response.
psi that management had some level of awareness of these pq MR. RICHARDS: ThisTs in yourAugust psi tests.
pq investigation.He didn't say,'Td go ahead and do pn Putting yourselves back in the time frame nn it",but he didn't say-psi of the violations in 1994,were such tests within par MRIKELLEY: That's a summary.That's psi the expectations of management? Was management pq not his exact words but a summary, pq aware? Did management tolerate it,or were the pq MR. BEARD: Let's go back and answer the un tests going on and management didn't know it? pq question.In September of '94 that would not be pa MR. BEARD: Before WE answer that, pa acceptable,the make-up tank evolution was not pai there's one thing I think in answering Al's questic,n pai acce table,and thyt, was not my standard.The fact pq that has to be put in context. I think Al was as tha 6)ad done things in the past, psi referring to page nine of our investigation report psi going back to I thiTu 1980 or before,and the fact BROWN REPORTING, INC. (404) 876-8979 Min-U-ScsI (41) Page 151 Page 154 ,
PROCEEDINGS BEFORE STEWART EBNETER, CHAIRMAN IN RE:
- MarcM 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 155 Page 157 pi that we had already recognized, as Mr. Hickle has ni plant.
m said, particularly when Greg lialnon became ops A MR.HALNON: Mr.Richards,let me add pl manager,that we had work to do with regard to the pi something. I don't know ifI can ever express to in operations manager with regard to the leadership sq you the difficulty I went through in pullin m procedures and all that,I think it's reflected that iq 'ff shift.Besides this test that happened,on Iq no,that that did not meet our standards and was not isi the 4th and 5th and everything that happened since m ccceptable.
m then,that was probably one of the most traumatic im The fact that it had occurred in the isi things that happened in operations. IIe was m past,that we had at least two shift supervisors who m respected,but he was pulled off for not meeting our pq were not meeting expectations in '94,that's true, pq mndmJE and if that's not a message to the rest py and we certainly ought to be held accountable for py of the operators- he's in the position he's in now pa that,and we're taking them off shift soon.
na because of his technical knowledge,and that na MR. GlBSON: Earlier you said a few. liow na technical knowledge is a great benefit to our pq many are still on shift that should be off?
pq plant.But the message,and the message when he was pq MR. BEARD:That should be right now off, pq pulled off shift,is that human perfortnance and p1 I don't think there are any that should be off right pq procedures are key.It was a very traumatic thing, on now.There's one that's going to be replaced.
on and we've soul-searched on why that happened, na probably in terms of upgrading leadership, but not psi MR. RICHARDS: I think you're right to a na - this rson is not a Dave Fields and not psi certain degree.It's your decision whether to take pq pq him off shift or not, but how do you know. I guess py MR. HICKLE: Let me answer that more pq with the fouroperators that were on Fielc':' group, pa specifically.If we felt right now that we had an pa as far as I can tell, nobody talked to them until paj immediate need we would have taken those actions, pal August of the following year.Nine or ten months go pq even ifit meant working a five shift rotation.
pq by and nobody has talked to these people about how pq Please believe that.But that's not to infer that pq they viewed the events.
Page 156 Page 158 pl we're satisfied with our shifts the way that the oj MR. BEARD: That's not true.
pl icidership on our shifts is now.There are other pi MR. HALNON: I talked to them regularly.
pi ch:nges that we want to make that are going to go pi Mark van Sicklen was the union steward.lic was in si into the future, some near term within the next few pl my face daily on union issues and just about daily is months and some within the next few years. p) on what was happening with the make-up thing.
m MR. RICHARDS:I think the reason it's m MR. RICHARDS: So you're saying that m important to bring it up is you've got - I don't m sometime shortly after the event you sat down with m know what Mr. fields's reputation was with the rest im each of the operators and talked to them about this m of the crews,but you mentione ad been m event and addressed your expectations to them?
pq there since day one, had been respected for having pq MR. HALNON: The letter said I counseled ny the most knowledge of the plant,your words,and he ny the operators,and that is like a bone of contention na said some things summarized in your internal report na because some of the operators didn't realize they va that I got pretty anxious about.lt seems to me psi were being counseled.I couched it more as a na thit if he's been there 15 years as a shift pq coaching.When I talked with the pperators we i pq supervisor he's had a big impact on the attitude and pq talked about the reasons the test was done.the fact p2 philosophy on operating that plant for the rest of pq that the procedures weren't adhering to our on yours perators.
On standards and stuff.There was a coaching aspect pal You've stated some corrective actions to ps) about getting them to understand why that was the poi further express your positions of the operators and poi case.That was ongoing.
pq how to carry out procedures and things.I guess pq I taught a two hour class every week, two ;
pq after reading your internal report and some of his py operators at the training center,and usually an pa statements, just sitting herc listening to your pa hour of that concerns the procedural standards and ga corrective actions, op 've got a good handle am human performance standards. j pq on the impact that nd Mr.ficids may havh pq MR. HICKLE: Let me also add one point to !
psi had on the way your operat rs look at operating your ps) thoroughly answer the question.Greg did that on Page 155 - Page 158 (42) g Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
IN RE: PROCEEDINGS BEFORE STEWART EBNETER, CIIAIRMAN PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Pags159 Page 161 pi his level.1 did that personally.1 did that at 19 different from the way they operate today in terms m all-hands meetings with everyone in the plant. Pat gj of management expectations, both from the NRC and pi has done the same,and Gary.What we've done is to l pi from utility management, i pi tell them and tell them again and tell them again pi For example,in 1990,when I first came si and then build it into the training programs, so is on the job,IMPO sent me a proposed good practice si there's no question in my mind that the lessons pi procedure on how to use procedures,like which ones m learned have been communicated and that they're very m have to be in hand to be followed verbatim,which n wellunderstood. {
pj ones have to be looked at and so on,because no one pi I just wanted to answer Mr. Lieberman's pi had any.Very few plants at that time had that. I pq question,because I don't think it got completely l pq took that and said," Gee,let's give that a try." l pu answered.1 think it was your question, or maybe it pq We gave it to operators,and I was dealing with a pa was yours, A1.Back in 1994 what were management's na lot of other things and learning the ropes, and we pai expectations?Those expectations were in writing. psi said yes,it would be a good idea,but we don't need pq They were in our procedures,and they specifically pq that, and I'rn not even sure IMPO was interested in pq covered unusual evolutions, requirements for 50.59 pq good practice at that time.They are now.And then l p: evaluations, requirements for soliciting help off psi things went on.
pn shift. It was all there.1 could show that to you on I may also add, and Stu will get mad at 1
I eq if you really want me to.
pq me for saying this, that in 1990 we were a one in on MR. LIEBERMAN: That's fine.
pq operations at Alaska South.We pmbably shouldn't pq MR.HICKLE:I'm saying those pq have been,and then we were dealing with a lot of pq expectations were there.Would mat have been okay pu things. Standards continued to improve.The first rai back then if we got wind of it? No, but I don't pa time I went down and watched the crews in the pai think it's the same thing that occurred with the two psi simulator,our communications were terrible.There pq tests that were performed.That was the other point pq was no repeat backs. - l pq I was trying to make, ps What I'm saying is that yes,we had a Page 160 Page162 p1 MR. LIEBERMAN: I see lots of violations, pl Dave Fields on sh chad and the m and it's so common to get an explanation that pi fact tha ad done me of ese things pi rnanagement expectations - and we have the same p; back in 7 I sure as hell didn't know about.But ni problem in the NRC too. Management expectations si as time went on, as our standards improved, as ottr te weren't carried down to the people who have to do .si cyes got open and we looked harder and we recognized n the work.In this case it looks like a third oithe pj we needed to relieve Bill Marshall with Greg Halnon, m shift supervisors did not appreciate expectations m that's another step,that we need new cyes, we al concerning unauthorized evolutions.We might have pj needed to have higher standards,and since we put pi had the procedures in place and to be run in pi Greg in the job we've finally been focused.We had pq facilities, and I don't have to tell you how to run pq two people in there that did not meet today's 04 the facilities,you know a heck of a lot more about pu standards, nor did we think they could change,and na this than 1, but it's more than just having things na that happens with people.They'd been there a long psi written on a piece of paper.
psi time,and sometimes you can't teach an old dog new pq MR. BEARD: Of course it is. I have to pq tricks,and we realized that.We were late in pq speak out again. Based on your comments,which I pq realizing that,but w aliz pq appreciate, and yours, John,let me just go back in psi Yes,it's true that as this on time. I've been in the nuclear operations business nn title. And by the way pq since 1966.1 know a lot of you have, too, both in pm oday,if you could get him in og the Navy and IMPO and here.I can see since 1966 pq a room and sit him down and say,'Tell us honestly pq that operating standards have continued to rise both pq the way you operated back then,was that right".1 pq in the nuclear navy and the commercial nuclear power pq think he would say no. And if we didn't think that ga plants.1 came to IMPO in 1982,and then I came to pa he was contributing a positive influence today in pai this job in 1990.How navv ships were operated in psi that position he wouldn't be there.
pq 1970 and how commercial nuclear power plants were pq I wanted to try to put things in pq operated in 1970 and up until 1990 is entirely pq context.We're continuing to get better.We now BROWN REPORTING, INC. (404) 876-8979 Miss-U-Scripts (43) Page 159 - Page 162
, PROCEEDINGS BEFORE STEWART EBNETER, CHAIRMAN IN ret March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 163 Page 165 pi have a Mr. Dave deMontford backing up me operations vi believe it would have been performed.
m manager.We've got Gary Becker with an SRO who's pi MR.HALNON:It was presented as a one pi been in the contros room.We've got three other p) time only,it's going to die right here. Frankly, pj engineers getting SROs.We've brought two other l pj we didn't know how to get it through our process to I ta people into the pipeline,and all of these are going isj let it die. In hindsight, we looked back and said, Iq to be in our control room at some point, but you isj "Well, we probably would not have approved it the m can't do it instantaneously.
m way it was written, the safety evaluation, knowing tal (A recess was taken.)
tal what we know now as far as design basis and that tg MR. GIBSON: Let's resume the meeting.
pi type ofissue."
pq MR. BEARD: Are you ready for Bruce to pq MR. LIEBERMAN:I don't want to dwell on p y discuss the next violation?
pa pu this. Is engineering involved in looking at the MR GIBSON:We've discussed the pa unreviewed safety question issue?
1131 presentation thus far.We have no motr questions on ps; MR. HALNON: They're in the qualified pq violation B at this time.What I would propose we pq review process for the procedures, yes.
ps) do,in an effort to make this more efficient,is to psi MR.LIEBERMAN: Since at the time that pq allow you to go completely through your presentation psi was reviewed on October 13th there were some on on the violations,and we will use restraint.
on questions as to whether this was design basis or usi Right, Jim' per not,and that was still being evaluated,were there poi MR. LIEBERMAN: But you had one pej any weaknesses in that engineering review for that pq misstatement.We only have one question left on pq particular 50.59 issue?
pq violation B.
pq MR. HICKLE: I'm not sure at what poin*
pa MR. GiBSON:I said at this time. pa in time.
pi MR. LIEBERMAN: On violation B, one quick paj MR. BEARD:I think Greg said it w..a to pq question.On October 13th,'94 there was a pq the PRC and this was sort of okay,but we're not psi re. evaluation of the procedure and a 50.59 review {
psi going to do it it was just a disciplinary follow-up Page 164 Page 166 pl was performed for the September 5th evolution,and I pi thing.I think he said our full process weighed pi believe that concluded that was not an unreviewed gj in.1 don't think it was intended.It was our p1 safety question. Am I correct in that?
pi process then.
si MR. HICKLE: That was the procedure that p) MR. HALNON: We went through the motions tsi we asked the shift supervisor and assistant to write (si basically to get it to the ten point.
pi as part of their disciplinary action.The intention ;
to MR. LIEBERMAN: The motions did not j m was to reinforce our expectation that a procedure be m include engineering?
ial written.The way they wrote that procedure was tai MR. HALNON: It had technical supports, rej along the lines of what they had actually-it was pi the system engineers, not the design group.
pq almost a chronology of what they did that night on pq Frankly,in h!ndsight,looking at it with what we ny the 5th.That's correct.
py know now,it doesn't meet our standards now.
pa MR. LIEBERMAN:Were you satisfied with na MR. LIEBERMAN: And your corrective paj the evaluation as to whether or not there was an vai actions for this issue, does it encompass this type pq unreviewed safety question with it? pq matter?
psi MR. HICKLE: In hindsight, no. I had not usi MR. HALNON: The communication in the pq even intended that procedure to go to the PRC,but psi design review,yes.
On afterit was written i: vent to PRC. Greg was on pn MR. LIEBERMAN: Between engineermg and es the PRC, he can talk tirsthand about it, but the PRC pq operations?
pq didn't know what to do with it because it obviously poi MR. HALNON: Yes.
am wasn't the kind of procedure they would routinely pq MR. BEARD: And we'll get to that as we an approve.But because it was taken as part of a py talk about the other violations since one of the key pa disciplinary action and the test had already been ga things is corrective actions and how does this in performed.they went ahead and appmved it.But if pai design basis get translated in the plant.It's the pq it had been a procedure that had come into the PRC pq interaction between operations and design.
psi for an evolution about to be performed.1 don't psi MR. GlBSON:I'd like for you to go Page 163 - Page 166 (44) Min.U.Scripte BROWN REPORTING, INC. (404) 876-8979
AN Rt:. . - - 13Ruttt1>1Ab3 is11URL d1LWAR A L11Ni.1LR,011AllB1AA
~
PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 167 Page 169 p1 through the remainder of the violations,an3 l'd p) Now, why is this point important, that pj like to ask the staff to use restraint and hold your p) one of nine apparent violations exceeded the make-up ;
j ni questions until the end.Perhaps by that time some p) tank curve? There was no intentional disregard for p) of our questions will have been answered. pj the operating curve when these violations occurred. ;
rs; MR. HICKLE:let's move on to apparent tsi let me explain what our expectation was,what we i a violation number one in the report,the inspection si would have expected to happen,and then I'll show m report,nine examples of operation of the make-up m you what really happened.
pj tank outside the acceptable operating region while si The system would normally lese inventory l p1 adding hydrogen.The report speaks to untimely pj due to normalleakage.That's the make-up system or pq alarm response in connection with that violation. pq reactor cooling system.They're interconnected. At pq Information important in understanding ny some point the operators would add hydrogen.Now, va the violation. Violations occurred during prj the intention was,theirintention was to add pai operational evolutions conducted to establish ps) hydrogen to some point that would not put them in pa optimum hydrogen pressure.The intent of these pq the alarm condition and then to add inventory to the psj evolutions as differentiated from the Iwo tests was psi make-up system and raise the level in the tank, ps) to restore the plant to chemistry specifications,to psj which had the effect of moving the function to the pn within chemistry specifications.FrornJune I un right and down and return them to the optimum pm through September 30th there wert:669 manipulations pq operating window.Now,that's what they tried to poi of the tank icvel or pressure. 21 alann conditions pe) do.They tried to add hydrogen but stay off the pq resulted from those,and nine are characterized in um alarm and then add inventory.On most occasions pu the violation as untimely alarm response.Now, au they were able to do this withs putting the plant pa untimely means the alarm was in for greater than 30 pa into alarm. And we were satisfi~1, we being .
pai minutes, and in the examples it was as long as 190 ps; management,that 25 cc's per kg in the reactor pq minutes and greater than a half a pound over our pq coolant system was an achievable limit .md could be ps) alarm curve. ps) done operationally, safely and within legal limits.
Page 168 Page 170 pj Now,I'm telling you this net to minimize vi MR. GUTHERMAN: When they raised the gi the importance of the violations by saying that p1 level it would move it to the right and up rather si there were just a few of them I'm telling you this p1 than to the right and down.Just let me correct p) because these manipulations were conducted most of pj that.
si the time correctly, and it wasn t a normal operating g rs) MR. HICKLE: Well,it's obvious from the si expec+2 tion of any of our opertttng crews to get p1 diagram.
m this alarm.They couldn't have performr.d this many m Irt me show you what typical evolution p) manipulations successfully without an alarm,this p1 occurred in the case of the violations, what it pl minus this. pi looked like.Now,this is the one that occurred on pq Another important fact is our operators pq July 27th.In this case they were between 71 and 72 pq controlled the evolutions using a r take-up tank strip ny inches in the rnake-up tank.They added hy6 ogen.
pa chart recorder.The make-up tank chart strip pa They went into alarm. Again, this is the at:rm psi recorder on our control board tells the operators paj curve.They went into alarm. At that point
- hey pa what the pressure is and what the level is in the pq began to increase inventory and level in the make-up psi make-up tank real time. it's the only real time psi tank,which caused the function to ntigrate toward psi indication that they have that they can use to psi the acceptable region of the curve.
pn verify their alarm condition. nn MR.GIBSON:It appears they continued to ps) One of the nine apparent violations,only nei add hydrogen after the alarm occurred;is that pe) one, exceeded the make-up tank curve limit as i,sj correct?
i pq indicated by the strip chart recorder,and that was pq MR. HICKLE: No.This is 21 pounds and l 99 for a short time right after they added hydrogen, my this is 23.This is less than a half a pound.I'll
[ pri and that was for a period of fifteen minutes. rn) explain how that happened in a few minutes.
( pal During the rest of the evolutions the stnp chart p31 MR. BEARD:As the gas is compressed the t pq recorder told the operators that they were within pq pressure does go up,too.
psi the operating curve. ps) MR. HICKLE:When you raise level the BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripte (45) Page 167 Page 170 1
~ ~ ~"
PROCELDINGS lihFORE STl! WART EBNETER, CliAIRMA$ IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE i Pcge 171 Page 173 pi function actually migrates inside the acceptable pj opposed to the strip chart.
m region of the curve.That's the way that they could gj MR. HICKLE: Greg,would you address e
p) clear the alarm.The point drifts this way. p1 that,please?
pi Now.in this case they stopped their p) MR. HALNON: That was between,I think, gg addition or their level increase at about 77.5 tsi two operators, Gary Hebb and Mark van Sickten,in an tai inches.They allowed the system to stabilize,and tej answer from Pat saying it would be good to use the a that's what's happening here. Hydrogen was going m computer because it's more accurate and the alarm tai into solution and cooling down,and the function isj curve is off of that.That pretty much died with im returned to the acceptable region of the curve.
rg those two.Some of the operators talked about it, i om So what went unng> What's the root pq but I didn't see that,I didn't know about that. It 99 cause? What we wottd 1. ave expected our operators to on certainly didn't get put out as an operations pa do is to question the dauepancy between the alarm na directive in any way, so it was not consistently pai and their recorder.They looked at the recorder and paj thought through the operations department that v4 they verified from the best that they could read v41 that's what we needed to do.
! psi that recorder that they were within the acceptable psj Gary did not want to operate the plant psi region of the operating curve.When an alarm is pq for the computerfor the reasons we just told you.
On received we ask our operators to respond promptly to on It was not real time and it was not easy for the pej the alarm, verify your alann condition,and the only ps) operators to use and there was no curve display you l l pq way that they could do that was using the strip poi could see anywhere, the function that Ilruce talked pq chart recorder,there was no other real time pq to you about,so they continued to operator with the pq indication,and then to follow their annunciator pu strip chan recorder.
p2j procedure. p2) MR. HICKLE: There was insufficient '
paj MR. EBNETER:The alarm comes off the paj day-to-day management presence in the control room.
p4j computer?
p4) We think that was part of the root cause.At this ps; MR. HICKLE: The alarm comes off of a psi time, at this particular time, Greg was our Page 172 Page 174 pj computer point,and the computer point has no real in operations manager.He was spending quite a bit of m time display in the computer room. , m his time down at the simulator doing simulator
] pj MR. HICKLE: Any time you're making a p instruction, working with the crews in trying to pl hydrogen add the only thing you have to monitor to s reinforca some of the human performance improvement isj make sure that you stay away from the alarm limit is m standard , that we were emphasizing at the time,and isi the strip chart recorder.That's the only isj he was t ot spending a great deal of time in the m instrument they have for real time indication- m control wm.
ial As far as what went wrong,we felt like tsj Additi >nally Greg was the only level im our crews should have questioned that discrepancy rol that we had of operations management above the shift om and caused that to be evaluated to see which was the om supervisor. Greg reported directly to me, and we p y most conservative, the alarm or the recorder, and py believe that because of the demands on Greg at the pri given the uncertainty to respond to the alarm and na time,and he was newin his position,and because we psi follow the AR, which was our annunciator procedure pm were trying to change our operating practices and pq and return the plant promptly to the acceptable p41 bring the bar up, that he was spread too thin. lf psj region of the curve.That was our expectation- ps) we had had additional managerrent presence in the pq They didn't qtestion that.We think thit our pej contml rooms more often there may have been an on supervisors became s little complacent and failed to on opportunity not necessarily to identify this pai address the alarm condition. ps) firsthand,but there may have been an opportunity pq MR. GIBSON: firuce,l'm sorry to nel for more dialogue and maybe a little more exchange am interrupt and break our own rules,but we did have sq between management and our operating shifts,which pu an e-rnail that you may have seen where this 99 could have led to addressing this issue before we g21 discrepancy was addressed. At least it was sent to pa discovered it.
pai one operator or one crew where engineering advised psi There were some contributing factors.
poi the operator that the computer was rnore conservative p4) First, system design was not intended for fine-tuned l psi and they should rely on the computer indication as ps) control.The control board indications didn't ;
Page 171 - Page 174 (46) Min-U-Scrspie BROWN REPORTING, INC. (404) 876-8979
{
l j
4.s ,m ruvum. us ou v4u. ; 4 mm 4 w. u. 4 us, m. u,uu. , 1
, PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 l Page 175 Page 177 {
pl emulate this make-up tank curve,and what the vi fifteen minutes,but they cleared that on their m operators had to do was take their strip chart pi sinp chart report.
pl readings and put those on to the operating curve to pi MR. GIBSON: On the other hand,I would pj try to determine whether or not they were p) like to point out that you mentioned earlier that l 1st approaching their limit,and the strip chart isi Mr.Hebb was involved in many of these overshoots, pj recorder was too inaccurate for its intended use. pi and he had personally received a memorandum from m There was only resolution at best to plus or minus m engineering advising him to use the computer and not sj half a pound, and in most instances,where we p) the strip chart because the computer was more pi overshot our operating curve,it was by about a half pi conservative,and apparently he chose to ignore the om a pound, so it wouldn't have been picked up on the om advice.
py strip chart recorder. pq MR. HALNON: Gary was more involved in pa Also,these hydrogen additions were done na the ones that came out - Gary was primarily psi manually.There was no automatic system control, psi involved in the ones in the lateJuly time frame, p4j which made overshoot more likely. After hydrogen p41 and that memo came out the first part of the second ps) was added the operator would have to determine when psj week in August, so after that I can look and see how psi to stop so that he didn't challenge that curve.But ps; many times Gary was involved, but in effect, yes, he on once the hydrogen went into the make-up tank,often pa did not want to operate the plant by computer.
pai it heated up,there was thermal expansion,and that pej MR.LIEBERMAN: Are there other poj caused additional pressure to build up even after poi situations like that with strip chart versus the um the point that the hydrogen addition was stopped. pm computer on other curves?
pq All of this had to be factored in,and the control rea M1. HALNON: This is the only curve we pa just was not fine-tuned. pa nave that has an alarm generated by the computer to ,
paj Another contributing facter was that pal where the computer would be more accurate as far as p4j management guidance regarding hydrogen concentration p41 the alarm limit.
psi versus the alarm limit could have been c! carer. I psj MR. HICKLE: We've strengthened our alarm Page 176 Page 178 pi think that,if we had done a better job of p) response procedures,though, recognizing that we m explaining the basis for the 25 cc's per kg in the m don't have a crystal ball and what we get into in pj reactor coolant system to our operating crews, at pi the future in terms of modifications and what alarms pj least they may have recognized or better recognized pi off of what switches,what we want our operators to is) the relative importance of the chemistry limit (51 do is when they receive the alarm to verify it and si versus the limit of the operating curve.As we isi without any doubt at all follow *'e alarm response m said,we put a lot of emphasis on meeting the 25 m procedure immediately.
p) cc's per kg,and I think sometimes they may have p1 MR. LIEBERMAN: How have you modified pi viewed that as a higher priority,at least to the ; p) your procedure for alarm response? I gather that !
pm point that maybe they didn't question that om previous procedure addressed expected alarms versus pq discrepancy between the strip chart and their alarm pq unexpected alarms and this was viewed as an expected l pa set potnt. Da alarm so that complicated the op:rator's response.
pa MR. EBNETER: Mr.Lieberman asked me a pai MR. HICKLE:I think it may have,but I v4j question.Was the strip chart in the control room v41 think that's more of an excuse.The reason I think !
psi sensitive and accurate enough that the operators psi that is because our operators are trained to respond pai could clearly discern that there was overshoot? psi to alarms promptly,and that's reinforced by the pn MR. HICKLE: No.In the instance of the lpa assessment of their performance in the control room pai nine violations or nine instances of the violation, I nal and in'the simulator.We specifically critique om when they looked at their strip chart recorder they poi that,and it's hard for me to believe that that pq would get the alarm,they would look at the strip pq expectation isn't crystal clear.1 really think pu chart recorder and they would see on that strip pu that this contributing factor maybe had more to do pa chart recorder that they were within their operating ga with that decision-making,the fact that we weren't psj limits except for one instance,and that was the pai clear enough as to the reasons for maintaining the p41 case I talked about. For a short period after a g41 chemistry limit,and that may have created a psi hydrogen addition one day they were over about ps) priority in our operator's mind that caused them BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (47) Page 175 Page 178
PROCEEDINGS BEFORE STEWART EBNETER, CliAIRMAN IN RE:
Marcir 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 179 Page 181 pi maybe not to question the alarm as hard.1 think pj We implemented our event-free operations m this is an exceptional circumstance- m program. I talked about that earlier.We're pi MR. HALNON:I think the statement you pl getting extremely good results from that.We si made about expected alarms,there was in that bottom p1 implemented changes to reduce operutor burden.
Iq bullet the weak procedure guidance.There was ist Specifically,the make-up tank high levellimit was tai weakness as far as expected alarms because it gave iq incrrased to a huridred inches, so that moves the m the operators quite a loophole in responding to m window that makes the area larger that's acceptable is; alarms,and the procedures didn't go beyond that, tsj for operation to maintain the chemistry limit that pi and that was a weakness in it-p) we desire in the reactor coolant system.Back in om MR. BEARD: We've already talked about pq 94 it required about five manipulations a day to oti the last bullet. pu mamtain our chenustry limit, and now it's less than na MR. HICKLE: That's right. So we na two.
pai recognized we could strengthen our procedures,and paj MR. HALNON: It's gone way down.
pq we did that.What corrective actions did we put in p4) MR. HICKLE: To less than two.We pq place? Some of these corrective actions will have psi installed -
ps; an asterisk,and those are the ones that we did pej MR. EBNETER: How did you get that down on directly in response to this apparent violation, on from five to two?
pm remembering that we didn't discover this violation psi MR. HALNON: It's based on -
pq until sometime well after the time that the two og MR. HICKLE: Previously we had this am tests were performed. pq window to operate within.Now we've got this window pq MR. BEARD: 1.ike December '95- pu to operate within to maintain acceptable hydrogen in mi MR. HlCKLE: But the corrective actions pri the reactor coolant system.This is a larger
- 3) that we put in piace as a result of the tests served pal window.So by raising the level and pressure -
mi to correct this problem, too. p41 MR. EBNETER: I understand. I'm just psi On September 9th we placed an offset on psi curious. Is this still within all the chemistry Page 180 Page 182 p1 the make-up tank operating curve,and since that pl guidelines,though,I take it,by moving it off to si time we're not aware of any instances where we've m the right further?
p1 had an alarm condition. And when I say an offset,I pi MR. BEARD: Yes.We did an analysis to pl mean we established an operating limit which is pj show that raising that with all the competing (si below the design basis limit.We revised the Is) operational requirements wouldn't violate something to administrative procedures for alarm response.Now, ,,j clse, m that we did specifically,and we already talked m MR. HICKLE:We installed a pre-alarm as is) about that,directly in response to this violation is an operator aid to give warning to the operator that pi after discovering it.We reinforced expectations on p) he's appmaching the operating curve.
Da alarm response through training.I think we've om This is our computer alarm, which did not pu discussed that quite a bit this morning. On exist back in '94.We had no computer alarm This na We created an additional management na is our annunciator alarm, and this is our design esi position to focus on shift operations.That ps) basis limit.When operators add hydrogen they'll pq position is called our manager nuclear plant v41 get the computer alarm to tell them that they're psi operations with sole responsibility for management, psi approaching the annunciator alarm limit, which is pel providing management assistance and oversight and psi our operating curve, so there's plenty of margin now on assessment of our shifts that are on duty.our on to the design basis limit.
psi on-duty shifts, with no responsibilities routine ps) MR. BEARD: And that's what we should og responsibilities for operations support, om have done from day one.
pm administration.It's strictly a management pq MR. HlCKLE: We provided a conservative my function.That position now reports to an assistant un operating curve.I just showed you that.We added pri plant manager of operations and chemistry. So we've pa a chain wheel on MUC 64,or are adding.These next {
paj got another layer of rnanagement with the intention rn) three are being accomplished.MUC 64 is the shutoff l 241 of L cing able to provide a lot more oversight in the p4) valve that isolates the rnake-up pump from the as cornrol room. psi make-up tank.We restored the position indication Page 179 Page 182 (48) Min-U-Scripts BROWN REPORTING, INC. (404) 876-8979
PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 183 Page 105 pi for that same valve on the contml board so the p1 point.He's not the management position that I pi operator can see its position, and we've added a m expect to be cognizant of all of our alarms on the pl manualisolation valve for hydrugen addition in the pj control board.The first responsibility is with our q' pi turbine building. pj shift supervisor,and the oversight responsibility isj These last three things were requested by Isl for that now rests with our operations manager Isi our operators to help them in emergency operations (si position.
m in the event that we have an accident.We don't m Does that sufficiently answer your isj anticipate they'll be used for routine operations. tej question?
m So what results have we achieved? By the toi MR. LIEBERMAN: Yes. In other places the om cnd of refuchng,all of our conective actions,the na shift manager has more operations responsibility, so un last of which are the last three modifications I pg you've clarified that point.
Da discussed,will have been completed.I mentioned na MR. HlCKLE: And I gathered that was the j paj we've had no high pressure alarms identified since psi conclusion from the inspection report comments.
pq September 9th of '94.We think the questioning pq MR. BOLDT: One of the chief reasons he l psi attitude regarding expected alarms has greatly psj exists is to provide assistance to the shift j ps; improved. Now, that's a hard thing to measure. psi supervisorif he needs assistance,if he has on That's why I said we think.But that's our pn questions to pursue with the system engineer or he's pq observation.We're getting a lot more questions psi running on a timeline and he has to get some poi being asked and we've had a lot more opportunity for poi questions answered. lie would be expected to turn to pq alarms to exist with the refueling outage and our sq the shift manager to assemble the resources so that py shutdown that we're presently under,and we're 99 the shift supervisor may focus ( the control center pa seeing a very good, healthy response in that area- ,pa activities. .
pal We have also increased management lpai MR. LIEBERMAN: By the time of September I pq presence in the control room.We believe that our pq 5th should the shaft manager have been consulted?
psi shifts,and my personal observation is that our psi MR. BOLDT:In our opinion the shift l
Page 184 Page 186
- pi s;iift supervisor is much better informed of our pl manager should have been consulted,and actually in i m g lant activities,of our plant prionties.Our m our opinion this shift that was on duty should have ni operations rnanager position is providing a lot of pi known better than any other that that was the reason si help to our on<!uty shifts and very good assessment p) he was there.
1 (51 and oversight. MR. LIEBERMAN: Did the shift manager 4
isi tai MR. LIEBERMAN: What's the sole of the , tsj periodically go in the control room) m shift manager with the operations manager? Is that I m MR. HICKLE: Yes.
is) the same position or a different position? Isj MR. LIEBERMAN: Should the shift manager toi MR. HICKLE: That's a different pj have known abotit what was going on on the 5th or the pq position.The shift manager is not an operations pq 4th?
pq position.The shift manager reports to me and is !ny MR. BOLDT: They should have been va responsible for our integrated plant schedule.What va informed by shift supervision of what was occurring.
pai does he do specifically? First he's a schedule v3) MR. BEARD: They should have consulted pq enforcer,but he pulls together resources for p41 with them in advance.
psi accomplishing the scheduled items that are on our psi MR. LIEBERMAN: But from the other psi daily work schedule and also during refuelings the Insi direction,should the shift manager have been in the on same thing.He's there to be notified of evolutions on control room enough to have recognized this was !
~
paj or any other activity that the shift supervisor is pej goingon?
poi required by procedure to communicate outside the MR. HICKLE: Not in my view.They spend
.poi pq control room,and in those cases he has some Inq time in the plant,but not solely in the control ;
pq approval authonty. }py room.They're expected to evaluate work in progress pa lie also serves as our emergency !pa in the field, pai coordinator and would man our technical support pai MR. BEARD: During an outage, many plants pq centerin the event that we have an alert.And he's pq have typically what they call an outage shift psi not expected - this is a good time to clartfy this lasi manager around the clock. And we said," Gee, that BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (49) Page 183 Page 186
PROCEEDINGS bel; ORE STEWART EBNETER, C11 AIRMAN IN RE:
March *27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 187 Page 189 19 works pretty good in an outage,why don't we use pj Secondly,in terms of results, Bruce did m that concept all the time." pi mention in a corrective action our event-free pi MR. BOLDT: As it relates to this pj operations program.I'd like to stress again that a si specific shift, we had three reactor trips in late si key part that applies to the whole plant is enhanced is) 1991,orie of which was the subject of enforcement.
Is) use of what we call our precursor system, which is Isj Two of those reactor trips occurred on a midnight Isj that if anybody sees anything wrong or they're m shift which was supervised by this same shift pl worned about it, write it down and tell us. In tai supervisorinvolved in the 4th and 5th incideuts.
g j 1994 we had the program in place,but it was not pj The third reactor trip, we happened to have in place p) well implemented, as pointed out to us by IMPO.
pq an operations superintendent who was rotating.He pq A case in point.In all of '94 we only I p y stepped in and was able to assist that shift in 09 had 400 precursor cards.In 1995 we had over p2j properly mitigating that panicular reactor trip and pa 3,000.We now have implemented that.I think the pai the consequences.
pa) operators, the shift supervisors would have written pq As a result, Pat had assigned me the task l
- 04) a precursor card that they're having difficulty,and psj oflooking at the generic implications of those ps) the way the precursor program works, there's a og three reactor trips.We looked at all of them in ps) littic green carrn' hat anybody can fill out.
vn general,and we have a document that made on They're provideil to the shift manager immediately, pq recommendations,one of which was to consider the pai the nuwcar shift manager,around the clock.All pej creation of a nuclear shift manager as a resource to poi the oncy that came in in the previous 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> are j pm the shift supervisor for collecting data.One of pq discussed at 8:30 with the plant manager and other j pu the other trips $at occurred dealt with a power 99 managers anc are immediately sent out for pa range detector that was out of service.The shift l 122) disposition,aad those that are significant are '
l cai supervisor in question was approaching a limitation ps) turned in as quality repons.That's another result I pq whereby he was going to have to bypass another pq l'd like to point out.
psj channel, and he consulted only the local resources psi MR. HICKLE: This last slide just shows Page 188 Page 190 p1 on his midnight shift, consulted two of our NC pj the safety significance of these nine evolutions.I ist technicians and some other data availabic to him. pl mentioned earlier, with respect to the first pi but he never stepped outside that for example to pi violation, that the safety consequence was low.
rj call the system engineer at home or call in some 81 These are the reasons we believe this.Only onc Isj additional help. !
uj train of HPI would be affected,and our tech specs 1
- sj They constrained themselves to on-shift tsj allow an outage time of 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> per each train.
m available resources,made inappropriate decisions, m The time in violation of curve was shon relative to rol and now we tend to apply cockpit isolation to that. tai the allowed outage time.No other technical p1 Even in a cockpit you've got a radio to call to the si specification actions were in effect at the time, pq ground to get additional assistance.That's why the pq and the curve was a design basis limit only for a 09 shift manager exists, so the shift supervisor can py single accident scenario that had a core danuge pa focus on the control activities and sav,"We need o2i frequency of 3.5 times ten to the minus eight per paj help from the system engineer,you take the lead on pal year.
pq this and help me get an answer." pq The last thing I want to discuss, and I psj MR. BEARD: Although I mentioned this psi think we've picked up quite a few of the points that psi earlier today, Bruce did not mention another psi I wanted to make, so I'll go over those briefly, are on correcuve at "on that we've undertaken that also un the three mFnagement Concerns expressed in the pai addresses the root causes here and will be pej inspection repon.
pe) additional effort, and that is our structured poi MR.BE ARD: Actually,in the March 8 901 operator work around the identification and pq letter which followed the inspection report which pq resolution process. Clearly,this in a sense was an pu set up this tsactence.
pa operator workaround,and today we have a process pz) MR. HICKLE: So we know what those are, pai that identifies those, keeping them current and pa) The first one was that inadequate management pq tracking the resolution of them,and that process is pq oversight allowed recurrent cl.211enges to and psj achieving results. psi violations of operating curves
- hat were intended to Page 187 - Page 190 (50) Min-U-Scripts BROWN REPORTING, INC. (404) 876-8979
15fe: Pnoc1*ulw3 imi ont 311 maul Einh1LR, c11AnutAA PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 P2ge 191 Page 193
[q ensure that design basis limits were not exceeded. p) emerging as potential problems. Following up on p) Well,we agree, However,we do think that when we 91 deviations from standards, including intervention, p) discuss management oversight that there's many p) was necessary.That is a management oversight pi elements of that,and all of those we don't believe 81 function,and we think we've done an adequate job in tsj were inadequate, but we do feel that some were tsj that area.In summary,we think two,three and four tsj deficient,and let me show you what those are. is) were deficient.We think one and five was l m First, providing wntten guidance m adequate, !
m establishes management expectations and standards. Is) The next concern was that management did mi While we did make some changes for added defense in si not provide adequate guidance on the use of routine pq depth, overall we feel that this area of oversight na procedures for non-routine evolutions,a focused pq was adequate. Communicating operating standards, 09 issue,and we disagree.We think that the na including training on these,is a management pa procedures that were in place were adequate for the ps) oversight function,and that area we think we were ps) training and experience of our operators.Now we {
pq deficient in some areas, especially with respect to pq have strengthened our procedures to provide defense !
ps) the way that we communicated our expectations for psi in depth.This is the point where I want to explain l pej procedure use. ps) to you how the decision should have been made on I on Now, since that time I mentioned we've na September 4th and 5th,1994 on the midnight shift.
paj got a new procedure use procedure which we're using pai First of all,an operating crew pm as a tool to communicate these standards,but also poi determines a desire to perform an evolution of some. I pq our event-free operations program has gone a long sq kind.They get the idea.The next step is to l pq way to put procedure complian~ into the context of pu decide how they're going to go about that,to find j pa our need to improve overall human performance at the pa the nature of the evolution,how are we going to do . l pai plant, and we know that that expectation is cicarly ps) it.They did that.They then deternune the pq communicated and understood by our personnel, pq procedural adequacy,and this is back in 1994 using psi including our operators. ps Al-500, which is our conduct of operations Page 192 Page 194 p1 Establishing processes to identify pi procedure.
m deviations from standards.In this case we think we m If that procedure had been consulted,it I pl were deficient. Pat just talked about our precursor p1 would have referred our crew to AI-4005, description gi system. Right now dunng our outage we're receiving p) and general administration of plant procedures.In tsj around 50 precursor cards a day,and quite a few of isi consulting that procedure,they would have gone m these have been identified by operators and by shift (si through a logic matrix and they t ould have m supervisors.We're getting extremely good usage, m determined that a written procedure was required for inj and we're learning a lot about problems at a very, si the test.The next step,now that they know they
- si s ery low level,and from this information we're able p) need a written procedure,is to determine whether or sq to do trending and develop understandings of areas pq not they need a new procedure or if the existing n y that need additional management oversight. 09 operating procedure provides sufficient coverage.
pa Observing and self. assessing to ensure pa Now in answer to that question, they had pai that standards are met. in that area I already said pai five opportunities with our existing procedure 4
pq I think we were somewhat deficient,and I mentioned pq guidance to come to the correct conclusion.First, ps) that we've established a new management position to ps) they could have reviewed Operating Procedure 402, psj help assure that there's a greater degree of pq the operation of the make-up and purification on oversight of control room activitics. Also,1 on system,and they would have determined that the OP poi mentioned th2t we've got a precursor system,and we I psi does riot allow operation in the unacceptable region pe) also have catablished a formal self-assessment pe) of the curve. Remember,in both evolutions the pq process,which is done by management and by senior pq first step was to add hydrogen,to pressunze the pq management which looks not only at our precursor pq make-up tank,and that put them in alarm condition ga information but at all of the feedback that we get pa for both evolutions.That specific step in the p3) from QA reports, from inspection reports and any !ps) operating procedure refers operators to Curve 94 other source to try to understand what areas need jpy op.lo3B.
ps) additional management attention what areas rr.ay be Ipsj They also could have consulted AI-400B, BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (51) Page 191 - Page 194
PROCEEDINGS BEFORE STEWART EBNETER, QiAIRMAN IN RE:
March'27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE E
i Page 195 Page 197 pl which is originating new procedures, and determine pj for operator response to alarms, procedural
$ m that a new procedure was required for a test or an (
m precautions regarding adherence to operating pi unusual evolution.They would have gone through a pj parameters contained in administrative curves,and pl matrix and they would have determined this was a pi the responsibilities and limitations of the shift
, tu test or an unusual evolution.lf they didn't isj supervisors.
2 !
tej consult either of these,they could have consulted For each of these. guidelines for Is) m AI400A and determined that the mterpreta on m operator response to alarms, we agree.Although Ist contact must resolve questions regarding intent, !
tsj operators are expected to address alarms promptly, 4 p1 content and clarifications, 1 ta and that's routinely reinforced in the simulator, j pq They had a discussion about the intent of l ne the procedure guidance needed to be strengthened. !
, on the operating procedure,they deliberated and made pq Our procedures didn't specifically address prompt pzj the conclusion,an incorrect conclusion,that their pri thinking in responding to alarms, and Al-500 has pai OP covered their test.The authority when intent psi been revised to address that. Also, we revised the j i pq questions are asked rests with the interpretation pq annunciator response procedure to require immediate ;
psj contact,who was really functioning as Greg's pq response to a make-up tank alarm.
4 osj backup,as our assistant operations rnanager.They poi Procedural precautions regarding
- on could have consult the NCD-11,which is a nuclear pn adherence to operating parameters contained in
) pej operations directive.Those are the highest tiered per admmistrative curves,we disagree.We believe that pq procedures that we have in nuclear operations, poi OP402 provided adequate guidance regarding pq preparation of safety, regulatory, emironmental and pq adherence to the make-up tank curve. Remembering pq compliance reviews,and from that they would have an when you challenge the curve is when you add pri determined that a 50.59 cvaluation was required. pri hydrogen,and that step specifically calleu wt i pal They also could have simply exhibited a psi OP-103B, Curve 8.The make-up tank,I mentioned pq questioning attitude which was discussed in Al 500, pq that. Additional reference to the make-up tank )
i pq and AI-500 would have directed them to get some help ps) curve was added to the limit of precaution to OP-402
{
Page 196 Page 198 l go to make the decision.So that's why we disagree. pj for defense in depth.
p1 There were at least five opportunities that night to pi The responsibilities and limitations of 2
pl make the correct decision. p1 shift supervisors,we don't agree that that's a pi MR. BEARD: The details of that flow path pj weakness.The responsibilities,we feel,were Isi are in the back which we will give you with each 14 clearly defined in Al-500 at the time,and in your tej clement outlined so you can clearly see the flow gg package you've got those highlighted so you can look m Path.He was determined to go through every single m at that if you're interested. Again, though, we tai step,and we said no,we can make our point. tej have added additional guidance.1 talked about that
! ' pi MR. EBNETER: Were these operators and pi earlier,and it has to do with the CAPS philosophy.
om shift supervisors trained in all these procedures? pq l want to remark on the responsibilities pn MR.HICKLE: Yes,they were trained to pq and limitations of shift supervisors.After the two j pri use the procedures, the Al-500 and AI-400 pai tests were performed on the 4th and 5th and before
, pai procedures. pai our other crews knew about them,I specifically
- pq' MR. LIEBERMAN: Nevertheless, at least pq talked to,that I recall,at least two shift na two shift supervisors thought this would be okay to pq supervisors and presented a similar scenario to them
, nel do it that way. pej and asked them what their decision would be with
- on MR. BEARD
- I think we said as a root on respect to use of procedures.1 was assured in both ps) cause management was not successfulin reviewing the pej instances that that evolution would not be performed
, poi adherence to procedures and questioning attitude. pq without a test procedure,and they considered it to pq Our point is the procedures were adequate.We've pq be an unusual evolution.
pu already said that as a management oversight our pq Now, subsequently we've talked to other p21 ability to have them use those consistently was pri shifts,and other shifts have given us feedback that i gai inadequate.That's not an issue with us. ps) they would not have violated our operating curve,
! pq MR. HICKLE: The next concern,other pq but that's after the fact.That's after the l pq procedure guidance was lacking, such as guidelines ps) information about the evolution was disclosed.
i Page 195 - Page 198 (52) Miss-U-scripte BROWN REPORTING, INC, (404) 876-8979
~
6s PROCEEDhMiS BEh0RE STEWART EBNETER, CliAIRMAN '
PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 1
Page 199
)
pj Page 201 MR. GIBSON: Meaning they would have 19 MR.TANGUAY: As Pat indicated earlier, m conducted the test so long as it did not violate the 4
pi curve? m I'll be talking about the third and fourth apparent l
pj p) violations as well as the management issue as MR. HICKLE: That isn't what they said.
p) identified in the March 8 letter.
im MR. BEARD: I don't know if we asked them in The third apparent violation is three Iq that question.You just said in two cases where you to examples where we had inadequate corrective actions, m talked to shift supervisors they said they would not s pj have conducted the test. m specifically that corrective actions for engineering Ist calculations were not completed in a timely manner.
pi MR. HICKLE:I only talked to two pq personally. toi The first two examples related directly to the ny pq calculation of the make-up tank curve,and that will MR.GlBSON:I was only trying to pq be the focus of my discussion here this afternoon.
pai understand what you said.1 wasn't trying to expand 02) I'd like to dispose of the third example,
- 03) on it.
i ny pai if you would,and that was specific to corrective MR. BEARD: Are we ready for Paul to talk pq actions for safety-related tarks were not completed ps} about the next violation? I want to get back to one i
psi in a timely manner. I was personally involved in psi point as we're waiting for Mr.Tanguay.
' psi the decisions associated with delaymg the on There was management oversight.That's a pq term that we use ourselves often.The NRC uses that on corrective action associated with the tanks. At the pai time that that issue came up I made conscious pq term,and when we received your March 8 letter, ;
pq decisions based upon the other competing factors I poi although we had already been thinking of that from 1 um had for workload within my organization, recognizing I 99 the earlier inspection reports,1 guess this is a 99 the safety and significance of tuose issues,and pa case where the regulatory process works,contnry to paj what we had seen in the few t.aks that we had -
i paj uhat theTime Magazine article may say,that caused I ps; evaluated we had other higher priority,more safety pq us to really say what is management oversight, what pq significant issues,and I made a conscious decision pr does it consist of. ,
pq to push those out. !
Page 200 Page 202 pt We've determined that there were five l pj We did,however, list the tanks, "
m elements starting with the top,and that's m prioritize those tanks,and we have incorporated pi management expressing the standards and expectations pi that corrective action into the action plan that was p) through written guidance.That was very helpful to pi presented to NRC back in November as part of our set (q our understanding of what it was,and that's
[si point program, so it is ongoing on an integral part isj something that we're going to carry forward with us si of that.
m in the future.It really clarified our thinking on 4
m One piece of that that was completed in raj what elements were deficient and which weren't,and i rei the past when we got into that effort, we badn't si I think that's something that will be of pi identified there was an issue invoMng
- pq considerable benefit to us going forward.
pq instrumentation accuracy that had to be pq MR. GIBSON: That's good. I'd like to
- on incorporated,and we had to go in and make sure we pri share with you some of the thought that went into pri had correct, for lack of a better term, mechanical l pai that phrase in that lener,and that was that we i psi calculations which directly related to the pq felt you should have known about the problems that pq relationship of volume within the tank to the actual i pq were occurring on shift.The fact that your psi level.Once we had that physical level,we would go
- pej objective for meeting the chemistry standard was psi back and superimpose on it the instrumentation i on causing a real hardship on the crews and was Inn piece.The
~
mechanical-physical relationship has pai resulting in some cases hi a procedure being pe) been completed for all the tanks and now we're back poi violated and that this continued over some period of poi fitting in the instrumentation aspect of that.
pq time without your knowledge,that was one of the pq In hindsight,we should have done a more pn reasons that we felt that there was a management pu effective job in prioritizing getting on with those
] P2 oversight condition.
pri activities.However,at that time we were also 93; MR. BEARD:That stimulus to further our Raj looking to bring more work in-house.A lot of the j pq examination of what management oversight is and what pq AEs were downsizing.and we previously had relied on pq we did or didn't do was very helpful. 951 them quite heavily "Ihe other conscious part of the BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripte (53) Page 199 Page 202
PROCEEDINGS BEFORE STEWART EBNETER, CIIAIRMAN IN RE: l March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 203 Page 205 p1 decision was to keep that knowledge base and insight vi hindsight, going back and doing a review,what I've pi in. house and significantly enhance our working p1 determined is that the bulk of our venfications are l p) knowledge of the plant going forward. pi accomplished by a cursory review,and therefore what '
pj Root causes, there are several, as you pj I've concluded is that the level of rigor applied to la would imaginc.The first root cause,the engineers [si the verification process is not where it needs to to that were involved in developing the calculation did im be.
m not get the needed input from operations.They m in this particular case the verification 18: weren't licensed,they weren't trained in the pj was not commensurate with the safety and pi physical operation of the plant and were neglectful pl significance of a given calculation. Contributing pq in going out and getting the required information sq to this,there was ineffective communication and 4
on that they needed. Specific areas where they should no calculation between the design and system na have obtained some additionalinformation were in na engineers.Specifically,up front through the bulk na the area of validating assumptions right up front in na of the exchange and interaction with operations, say pq the calculation and making sure that the assumptions pq from theJune to September time frame,the bulk of pq they were using in the development of the curve were pq that interaction was between the system engineers
] pai in fact consistent with the procedures and the way pai and operations, design engineering wasn't that ;
, na the plant was operated.They did not do that. pa involved,and that was again exacerbated by the j
< pq They also did not go to operations and psi physical separation, the design engineer working on i pq ascertain the input that they would need to assure pq a calculation being located at St.Pete and the pq that they did in fact represent a curve that the pq system engineer being at the site and not really pq operators could physically use and that we could put pq being familiar with the design calculation itself. ,
tai out, as we already talked about, a design curve that pa A contributing factor on this third one 4
pa did not have an operating offset.Had they had that ga is the complexity of the make-up tank curve,which pq interaction they would have identified that as a pq was not well understood by anyone.There were a lot i psi need. psi of competing issues,as we've already talked about. ]
Page 204 Page 206 pj The second root cause in the area of p1 I don't want to dwell on this,but on the next slide m inadequate management is that we did not assure pj I just listed the issues just to show you that there pj operations input was in fact obtained.And pj were many, many competing issues here.Every one of p1 exacerbating that is that once we realized that that pi them ultimately forced us into the situation where rs was occurring we didn't step in and correct the e.1 we had this,and we had Appendix R and design Iq situation in a timely manner.I should have gotten (q limitations on the tank and the balance of the m more personally involved, recognized that wasn't m issues that unfolded unfortunately as we were tai occurring, and taken corrective actions to make sure tsj continuing to develop the calculations.
p) that that interaction was happening.In addition, im What did we do to address these issues?
pq we did not address at that point in time the human pq The corrective actions we put in place.first,we p q performance issues that were associated with the pq counseled the engineers,both the design engineer as pa} specific individuals associated with developing the na well as the individual performing the verification pa calculation. na process.
pq The third root cause was in the area of pq MR. BEARD: Paul, you pointed out that in es the calculation process.There were weaknesses pq this case many of these were directly as a result of ps) within the body or within the process that existed pq this violation,but some were broader and already in va at the time.Specifically, verification of pa place that de reflect corrective action to address og assumptions and design inputs was not effective.We pej the issue.
pq did not go and actually verify that the inputs that pq MR. EBNETER: Are these the engineers pq we were using for the calculation were appropriate. pq that Lieberman asked about before?
pq Secondly,the method of verification was pq MR.TANGUAY: These are design and pa inadequate.If you're familiar with the ANSI pa verification.He had asked about the system pa standard,it indicates that there's essentially tai engineers.We did counsel the design and pq three methods of verifying work,a review,an pq verification engineer,and the tourse of the pq alternate calculation, or by test. And in ps) counseling was done by the st.pervisor and their Page 203 - Page 206 (54) IWin-U-Scripte BROWN REPORTING, INC. (404) 876-8979
&ml% PROCLEDlh 131dORE h1LWAR1 E11NETER, CilAIRMAN
{
PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 207 Page 209 pl manager.The things covered within that counseling pj organization entirely to the site.That includes m session were the importance ofinterdepartmental p) myself.I was previously with the balance of the p) interaction and communication, making sure you've p) design organization located in St.Pete.We're now pi got a quality product, the importance of the pj physically at the site and have been sinceJune of is verification proces? and reiteration of an isi last year.The design engineering review board was iq expectation letter that I had put out months before im formed, and what this is is one of our senior staff m talking about the importance of verification and the m engineers leads a multi-disciplined technical review pi fact that the verification engineer was as mi of all calculations before they're issued.They
[9) responsible for the accuracy and adequacy of design pi challenge the design engineer on the issues within pq as was the design engineer.They shared that;it pq the body of the calculation,again trying to ensure ny wasn't something that was passed along- pq technical accuracy and adequacy.We also use p2j MR. LIEBERMAN: Paul,in your system what pri feedback into this, lessons learned, psi does counseling mean? ps; As we go through the process and collect ny MR,TANGUAY: It was a form of discipline pq precursor cards to identify issues, deficiencies or ps where the supervisor and manager and I also talked pq problems that we see within the engineering pa to them, sat down and talked through these issues- {
pej organization, we feed them back to the chair person ;
on There is a document within our departmental files, on of the design and engineering review board and say, om not in the performance, because this was the first pq 'These are problems or issues we've seen in the I usi step of a progressive discipline activity where you poi past,you need to add to your repertoire of pq talk to the individuals. Also,if you go back to pq questions as you go forward and begin the challenge t29 the two individuals'performanc- appraisal that was 99 as you look through the issues to make sure we don't {
p2j just completed you will see that reflected in there pri repeat them." -
ps) in the performance appraisal. pal As Pat indicated earlier, we've also now pq The calculation process now includes pq assigned important issues as management focused psi operations and system engineer sign off of design ps) items, which results in the identification of an Page 208 Page 210 p) engineering calculations.llack when the initial vi issue manager,and we did that in this specific case m calculations were done that wasn't there.it was pl once we got into the ove 211 isse and realized how p) left to the discretion of the individual design pi complex it is.We recognized Gary Becker,the pi engineer whether or not he went out and solicited pi former design engineer of the organization who l Isj input from the other organizations, and in direct , isi received his license and was now the issue manager )
io response to this specific issue we have incorporated is in this activity. Based upon his invo!vement,the m that as an in-line function. m scope and the complexity was further clarified,and ai When a calculation is being done,the 5) we made sure we had a lot of the issues identified m design engineer's required to sit down with the mi right up front and addressed from both perspectives, pq system engineer and a representative from operations pq and he was able to bring both perspectives to the pq and determine what the design inputs would be and ,09 table.
pri what the appmpriate assumptions are.Then there is p2) We also implemented a third party review ps) a physical signature within the body of the 03) of calculations.In this particular case,we had pej Calculations.Then,at the tait end.once the ny both Gilbert and MPR perform a review on those psi calculation is donc prior to being finally issued, ps) calculations. In both cases they did actual om again they sit down with the operations poi independent calculations.They didn't just take on representative and the system engineer and review on ours and review it.We gave them all the inputs and
~
om the results of the calculation,the product ofit, nei assumptions and the scenario. MPR did a manual not what it means,and the intent here is to make sure pe) calculation,as we did, Gilbert did a computer based pq that operations clearly understands what the bases pq calculation, which is also modeled via computer, so try of the calculation is and whether or not it has any pu the level of rigor was there,and in both cases the pri impact on anything in the operations organization, [22 results of their efforts pretty much paralleled ps) i.e. procedure revisions required, human factors, psi ours. In fact,I believe in both cases, from what 94 those types of things. po we can see,our curves were cons:rvative by psi We relocated the design engineering ps) approximately half a pound.When I say our curve, IIROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (55) Page 207 - Page 210
PROCEEDINGS BEFORE STEWART EBNETER, CIIAIRMAN IN RE:
March *27,1996 PREDECISIONAL ENFORCEMENT CONTERNECE Page 211
' Page 213 pl that's the curve we're currently operating to,the p) refueling number eight. It wasn't clear to me that m last curve.
m engineering had even looked at them.
pi We also have implemented the event-free p1 MR.TANGUAY: You said were the curves pl operation program to assist us in identifying issues p) reviewed?
rs) affecting our engineering performance in our area so tsj MR. GIBSON: The curves and calculations n we can track and trend them. Engineering management isj that support the curves.
m now attends operations turnover meetings on a daily m MR.TANGUAY: The calculations were
[sj basis.We implemented this a few months ago in an tsj reviewed,and that was clearly in the letter.
- a effort to improve interaction between the operations MR. GIBSON
- I knew the letter said it,
' to pq organization. It gives them an opportunity to pq but I just wanted to confirm that the engineer had pq identify issues to us so we can get rnore prompt 09 reviewed the calculation.
02: response and turnamund on those issues, and it's a
- 02) MR.TANGUAY: Yes.
03: rotating basis, one week at a time, and I will be on v3) I believe on the results,I've already pq it next week.It includes myself, my managers and pq covered several of these as pan of going over the
, psi all my supervisors on a rotating basis.
ps) corrective actions. One of the things we have seen
- osi MR. MERSCHOFF
- Paul, before you leave ps) is that the in-line reviews of the calculations have
, on that, engineering in this case failed to prevent on identified discrepancies prior to calculations being pai this from occurring.They had an opportunity in May ps) issued. As an ongoing process and based upon the va that wasn't acted on.lf this happened again today pe) precursor cards that we see looking at the design pq and you had that May of '94 PR from the cavitation pq review board, we are collecting data that pn incident, what would have stopped or resulted in pu demonstntes we are catching issues before the pri early resolution?
pri calculations are being issued,so that's a ceu '
sai MR TANGUAY: The fact that we physically psj indication to me that the processes appear to be pq have the design organization right on site working pq working. As I indicated, we're tracking and psi more closely with the system engineering psi trending that information as a part of the operation Page 212 Page 214 pl organization.When we receive Problem Report p1 program precursors and using that as input, and pi 94-149, that was predominantly reviewed by the m we'll continue to make changes as we continue to pi system engineer.That was their primary p) find them to be appropriate.
n interaction.The design engineer,although they had p) The fourth apparent violation,there were tsj a piece of that,it wasn't a significant piece.In tsj four examples of our failure to implement effective n today's environment,with everybody on site,I isj controls to assure that the design basis is m believe you would see closer interaction.In m correctly translated into operating procedures.The tai addition to that,at that point in time the syste's is; first three examples that are noted are directly rei engineering organization and design enginurinB tel related to the development of the make-up tank pq organization were split.The system engineers and pq curve,and again,that will be the focus of my pq design engineers did not report to the same person. 09 discussion here this afternoon.The last is the 02: Today they both report to me. 02) fire service tank volume which we have talked about usi MR. MERSCHOFF; So the September letter psi cartier. I'd rather not dwell on that at this pq that said the curve was conservative had no input ny point.
psi from design engineers? pst Root causes.These will be addressed,as psi MR.TANGUAY: They had some, but the bulk ps! I indicated,with respect to the make up tank issue, on was design engineering. on and even though they specific to the make-up ps) MR.GIBSON:Were the calculations pai tank,they do in fact has e a much broader poi reviewed by system engineering or design engineering pe) application. Essentially they apply to just about pq as a basis for the September letter? pq everything we do within our organization. At the pq MR.TANGUAY: Yes. pu time there was a generic lack of operation's gzi MR. GIBSON: Because, as I understand it, p23 involvement in the calculation development process.
pai the calculations had a statement in there that in pal As I indicated before,it was left up to the pq cffect they had expired and were no longer pq individual,the design engineer, whether or not she psi effective,that they were good only through psi or he solicited the appropriate input.In concert Page 211 - Page 214 (56) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
4.3 au : raws u.vuuo ou v4u. aiuuuu uu.u.4 uw un usuia..
PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 PIge 215 Page 217 01 with that,there was inadequate engineering pj MR.GIBSON: Well,my question was that pi involvement in operating procedure revisions. pi the September 2nd memorandum said in effect that the p1 At the point in time the operations pi curve in the calculations had been reviewed and was 81 organization would make a procedure revision,it vi correct.
te would go to the system engineer, and they would do gq MR. BEARD: Yes.
tai the review.The design engineers were left out of isi MR. GIBSON:And my question was were the m that process,they were not in line.One of the m calculations actually reviewed by an engineer as a tai things we do now is an in-line review for both ral basis for that response.
to) system and design.By doing this process where to: MR. BEARD: Because there's a difference pq operations is now involved with the calculation, pq between did engineering review the calculations and pi} engineers are now involved with procedures, and we og did engineering do a rigorous look at the bases for pri now look forward to having consistency between the pri it and all that sort of thing.Mr.Tanguay answered 031 way the plant is operated and the way it's designed, p3) yes.
041 alleviating that issue. v4j Now, what we thought your question was, psj There was ineffective communication pq and let me clarify it,is was the calculation
- 06) betWCCn engineers and opemtions.We did not paj reviewed in a broad sense.And the answer is yes, na communicate to them adequately that the curve in on we went back and looked at Problem Report 149 and psi fact was a design limit.The engineer, as we psi Mr.Ilinman's attachment to that, attachment two, poi indicated, knew that at that point in time. Also, psi where that specific corrective action was closed out pq we did not communicate to them that it was not a pq onJune 14th,1994.In that sense,yes,he reviewed 99 dynamic response and what we would expect to see in 99 it.Did design engineering and/or Mr.Hinman do a pri the tank itself,and the ineffective communication p2j rigorous review of the bases for the calculation, ,
pai was once again exacerbated by the physical p31 the assumptions and all of that?The answer is no.
p4) difference in location of the respective p41 That's an example of the weak- ss in our process ps) organizations. pq that we recognize,in the process for not only doing Page 216 Page 218 pj In addition. management did not assure ni the calculations initially,the detailed pi that the calculation and operating procedure gi calculations, but a review ofit.
pi processes included interdepartmental reviews.1 pj MR. GIBSON: Okay.
- 8) indicated there were deficiencies in both areas.We pi MR. BEARD: I don't want there to be any is were remiss in making sure that they did in fact isj misunderstanding about what we said.
(si include that and catch that. ist MR.TANGUAY: What did we do to address m (A recess was taken.) m the root causes that were just identified.Several tai MR. GlBSON: Pat,where are we? (s) of these I've already talked about as a result of rol MR. BEARD: Mr.Tanguay is about to - pi the previous violations so I won't go back and pq MR.TANGUAY: The corrective actions. pq reiterate those.But with respect to the specific pq MR. BEARD: We've gone through the root pu item that indicates both design and system pri cause for the fourth apparent violation and we're in p2) engineering review operating procedure revisions, vai corrective actions. 03; again,the intent there was to assure that the p41 l'd like to make sure on the record that 04) operators understood the design basis of the plant psi we're clear with regard to one of Mr. l.ieberman's psi in that the procedures did in fact reflect the p61125t Questions of Mr.Tanguay because we're worried pai calculation.
pq about that and we want to be entirely correct.1 on To enhance that understanding,
~
psi believe the question that we want to make clear was psi engineering personnel provided tramin ; to the poi that you asked Paul,and this is my synopsis of your poi operators as part of their last requal cycle,I poi question,if between May the 5th, SP-630 and pq believe.on the design basis of the plant.We pq September 4th did engineering review the pq physically developed a one and a half to two hour pri calculation.1 think that was your question. p21 session, took our design basis documents in and pai MR. GIBSON:I think I asked that pai explained to them what the content was, how to use p41 question. p41 them and what they meant, so we attempted at that psi MR. BEARD: Was that your question? psi point in time to enhance their understanding of the BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (57) Page 215 - Page 218
PROCEEDINGS BEFORE STEWART EBNETLR, CIIAIRMAN LN RE:
NarchI27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE ;
Page 219 Page 221 p) design bases of the plant. ni and design basis issues that we have in fact been p) We established an operations contact for gi identifying.
pi technical issues. Bruce had talked about that in p) MR. MERSCHOFF: While you're on sins of p; his presentation.That was Gary Hebb,a former p) the past,you told us earlier that you didn't have ist shift supervisor.He has been interacting with us is) an integrated plan to assure that design basis is on numerous technical issues and providing some tsj curves are not being used currently to operate the m excellent input so that we can get some consistency m plant.Thati a sin of the past.
Is) in the operations organization on issues that we tsj MR.TANGUAY: Correct.
tel need to address. rej MR. MERSCHOFF: When will you have such a pq Management structure and interfaces were pq plan,or is that part of the corrective action for vil strengthened,and I'd mentioned this before,that pij the violation?
pa now both design and system engineering organizations pa MR.TANGUAY: I don't have that pai are under myself and that we're all physically psj specifically identified here. Again, we were going pq located at the site,and those activit cs go 2
pq back to what Bruce had indicated where ops had gone psi directly to support and enhance improved support of pq back, reviewed what they had in the 10}B procedure, psi the opetuting organization. ps) and they did not have any where we appeared to be '
pn Results that we've achieved.The new on operating right up against there.What we're trying psi processes are in place and are working. Operations pej to do now is focus on the EOPs and set point ;
poi is directly involved in the calculation process and, vej programs,which are of a higherlevel ofimportance I pq as a result of this,there is an enhanced pq and safety significance, and then my intent is to {
pq understanding on their part of the design bases,and pi; develop a program to go back in a more structured mi the calculations that we issue are consistent with pa manner and look at the others.
pai the way the plant is operated.We have the design pa MR.EBNETER: And that was my I pq and the system engineers directly involved in the pq understanding of the previous comments,that all of pq operating procedure preparation,and I'd like to psi the curves are in 10}B and that you had reviewed Page 220 Page 222 pi clarify here under operating procedure. p1 that and were satisfied that there were no design gi We don't directly get involved in a pi basis curves being inappropriately used at this ni day-to-day operating procedure review.What we're p} point.
pi focusing on here,because of the significance ofit, pj MR. HICKLE: What 1 intended to say is I to are the EOPs and the APs.Those are the critical tg *vas satisfied that there were no occasions where we 16 ones where we've seen some issues.We have made tai were challenging curves.
m changes to the calculation and operating rev.cw m MR. EBNETER: Maybe that's what you did Isj process, and they have resulted in additional issues inj say.
ret and discrepancies being identified. rei MR. HICKLE:It didn't appear that the pq Any corrective action program that you pq curves were design basis,but we really needed p q take, particularly in this area, you're looking for p y engineering to tell us that for sure.
Da a couple of things.One is to preclude continuation pa MR. BEARD: I think you're well aware of pai of the problems that you've seen to date,and then va what we're into with our enhanced awareness of pq also there's a need to have a system that enables pq design basis translation into EOPs.And yes,Al, pq you to identify issues that exist as sins of the pq when I told you whenever it was that the EOP's okay, i psi past,if you Will. pq Well,I was wrong.We didn't do that properly and on The second item I have here is on we're doing that properly now.We've also got the ps! essentially a mechanism where we're going back and usi set points.We're doing everything we can right l pq identifying sins of the past as a result of the poi now.You cannot address every particular design pq implementation of the processes.And particularly pq basis issue at once, as we found out, and I think pq in those two areas,the set point and EOP review p9 you well know that on more than one occasion we've ga program, we are identifying issues as a result of pa bitten off more than we can chew in the time frame psi the intercommunication or interaction between the psi we told you.
- act operations and engineering organization,and those pq Paul said our process is to get around to ps) you're very well familiar with by the number of LERs psi doing all the curves to find out if they're design Page 219 - Page 222 (58) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
~
a su.: ravu.wuua 42uvist a 4 Au.us4 Lhat And, unutoaa PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 i Pcge 223 Pgge 225 pl basis. It's our plan to do that, not just his, and pi think that's the critical element here.We have m we can't do everything at once and v c haven't. m been working with the operations organization.
pj Right now we're focusing on set poinc., EOPs, tank p) There was significant management pj calculations,as well as a number of other things. p) involvement during the months prior to the September tsj We've assured ourselves that none of the other tsj 5th test, specifically in response to Problem Report Isj operating curves are even close to being challenged, pj 149 addressing the HPI flow test and operations a whether they're design basis or not,and they won't rn management.Specifically, Bruce Hickle,Greg Halnon pl be.When we can fit it into our resources and gj and other members of the engineering organization pj priorities,we willlook at all of those curves from pl got together on numerous occasions to assure that pq the part of the design curve and what is the basis pq there was interaction between the engineering and ny for the curve. pq operations organization on these issues.
pai MR.TANGUAY: Continuing on the results. pri Engineering promptly addressed the issues psi We're continuing to use,as I indicated earlier, vai and concerns that were identified by the problem pq performance indicators, particularly precursors, to og report, specifically the cavitation that was psi monitor our improvements and the effectiveness of ps; recognized or realized during the test.We were pq our corrective actions,and we'll continue to do pai able to prove at that point in time, based upon our pn so. As a result of the implementation of the na evaluation, prove and demonstnte that it was not psi focused item issue manager concept, the progress pai hydtugen entrainment,that it was in fact air poi that we see in addressing these issues are reviewed ps) entninment in the line because we had previously pq on a periodic basis and are statused monthly for pq performed maintenance and had not adequately vented pq Pat's senior staff. pu the piping out.
p2) As an additional item here,last week we p2; As soon as we recognized that we vented ,
psi had an inspection,and to show that we feel some of pai the system off, rerun the test and did not see a pq these activities are in fact taking hold, we were pq repeat, so we felt that we were very prompt in psj looking at the set point program as a follow-up psi addressing that at that point in time.There were l
Page 224 Page 226 pi irspection,and he gave us some excellent accolades pj also follow-up meetings to specifically address the !
m atxm. the quality of the product we're putting out m adequacy of the corrective action plan developed as l pi in that arena.When we came back the previous time pi a result of that specific pmblem report.
si he said he saw two typos in a 140 or 150 page pi Management involvement continued.There Isl calculation and this time he couldn't find typos, tsj were again interdepartmental meetings,and they had ist and we feel that's a direct result of the rigor l iq several objectives.one of which was to review a we're putting in these calculations and the process l m options to reduce operator burden.We were focused isj going forward. mi on trying to reduce the operator burden we've talked (9) All of these things we're talking about pi about here today.We were looking at and had pq today are playing a role in the development of pq implemented several changes for the operators to pq those We feel that if these activities and actions 'pu enhance their ability to maintain hydrogen pii were in fact in place at the time these other events !pri concentration where it needed to be.
ps) were unfolding we would have identified those issues psi Another objective here also was to obtain pq and we wouldn't be here today talking about these pq comments on the concerns regarding Curve 8.The ps; things. psi operators were specifically asked for any comments psi Next I'd like to talk about the fourth . psi relating to the September 2nd letter and the issues on item from your NRC March 8th letter,and this one lpn as they unfolded as a result of PR 149.
I l
pq specifically addressed the fact that management did nsi MRILANDIS: Paul, did you say you did I p91 not work effectively with the engineering and poi implement something to reduce the operator burden?
pq operations staff to resolve a longstanding operator pq MR.TANGUAY: Throughout this evolution, !
pq concern. As we've indicated before,we agree with pq as we realized there was a burden,we had taken some p21 that statement,and throughout the discussions here pri actions.We had gone in and we had raised the set pai today we've been presenting information that psi point on the hydmgen regulator.lf you give them a pq supports those.Within that statement I've taken pq higher pressure they can get the hydrogen in and !
psi the liberty to underline " effectively", because i psi maintain a higher pressure and make it easier.We BROWN REPORTING, INC. (404) 876-8979 Min U-Scripte (59) Page 223 - Page 226 ;
i
PROCEEDINGS BEFORE STEWART EBNETER, CIIAIRMAN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 227 Page 229 pj modified the valves within the hydrogen upstream of p1 cl.ne it out.That was from Gary.
m the regulator again trying to reduce the pressure n MR. LIEBERMAN: So when the September 2nd
- 31 drop within that line which enabled them to get pi letter was provided, what was the communication n hydrogen in quicker.We were trying to do what we pj associated with the September 2nd letter to the ist could on-line in a relatively short period of time in operators?
tai to address those issuce.We weren't insensitive to tai MR. HALNON: I had just promoted another m it,but there were some things that Bruce had talked m assistant to myself and gave him the memo,and he tai to earlier which were longer term and needed to be si wrote on it on the top of the memo " draft pi addressed at that scope. .a recommendation" and a note on it saying," Write pq MR. LIEBERMAN: You just mentioned the pq comments to G.Halnon",and he placed that into the 09 September 2nd letter.Maybe you should repeat what ny shift supervisor's in box.Then that crew came in 03: you said concerning the - ca that night that he did that and saw that.That's paj MR.TANGUAY: Management was looking to paj the only correspondence we can find.
pq find out what additional concerns the operators had, pq MR. GIBSON: He did not personally hand psi so they asked them,"What other concerns do you have sq it to that crew and did not have any personal pq regarding Curve 8." And my understanding is,in ps discussion with them) on response to that,that there was one item on MR. HALNON: To the best of our par identified,and that was a request to rerun SP430, ps; knowledge,no,because the crew was on midnights, on which is the full flow HPI flow test,and based upon poi and Carl was the one we're talking about.He didn't poi that we agreed we would in fact do that,and it was pq stay there until eleven o' clock when this crew came un a commitment to do it that we're in riJit now, and py on,I'm almost certain of that,and he doesn't '
pri it will be run this outage. pa recall talking to the crew. -
paj MR. BEARD: I'd like to read this into pai MR. BEARD:Where were you, Greg, between pm the record.This is documented evidence of pq September 2nd and September 5th? Were you in ps) management involvement.It's from Mr. Halnon to the psi training?
Page 228 Page 230 pl operators.1 think this is in the material. pi MR. HALNON:I had the IMPO E&A visit pi MR. HALNON: This is not in response to m going on during that time frame.
pi the September 2nd memo,which was your question. pj MR. GIBSON: Mr.Bergstrom did not urge in The September 2nd memo came out after what Pat's pl operators to take action on that memorandum) tsi about to read to you. Paul had got his timeline si MR. HALNON: The only thing we have right tai messed up a little bit. ter now is what he wrote on that letter, ano I think m MR. BEARD: This is in the materialI'm m that's included as part of one of the exhibits.
tal going to leave you.This is dated August 9,1994. pi MR.GIBSON: Did you ask him if he did?
pi There is an ongoing discussion on the issue of p3 MR. HALNON:I didn't personally ask pq hydrogen pressure in the make-up tank.I have pq him.
p4 talked with Mark van Sicklen and understand the pq MR. BEARD: I asked him again yesterday, na concerns.I discussed this with Bruce andJerry 02 Al.and he answered unequivocally no. I know in the val Campbell. and will be working on resolving the psi letter,the DO alert,and I'm sure again on the pq questions.it is important and Bruce has asked if pq enforcement with the shift supervisor you're going psi anyone has a concern to please write it down and psj to hear this: Management made us do it.That's psi send it to me. Even if it is not new I need to get pe) totally false.Mr. Bergstrom assured me that, Pat, na all perspectives of this issue so we can address the on I was new in the job,1 was getting up to speed on paj right areas.It appears what has been addressed has pai the issues,1 annotated this memo to be just what it poi not satisfied the concerns to date,and I need to be ps) was.1 did not in any way say, infer or otherwise I pq absolutely clear on the questions. pq say words to the effect that someone would lead you pq MR. LIEBERMAN: Was there any response to pq to believe here's your last chance, you better do j p21 that? pri something or it's going to be closed.I'd even use !
psi MR. BEARD: There was one. ps; a stronger word, but I'm not going to do that in pq MR. H ALNON: I got one response that said my this forum.That is totally false.
psi running SP-630 would be an inappropriate thing to psi MR. LIEBERMAN: Did he have any Page 227 - Page 230 (60) Miss-U-Scripte BROWN REPORTING, INC. (404) 876-8979
ggg g
- PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 '
Pcge 231 P ge 233 p) communication with him verbally? Based on your pj When would we have found this design m conversation did he - m basis curve? I could argue that we could have found I f p; MR. BEARD: No.1 don't know ifI asked pi it by September 30th if Problem Report 149 had been l pi him that specific question, but from what I did ask pi allowed to complete.1 mean,I could sit here and (q him it was clear to me that he did not. I did not (si make that statement,but it wouldn't be appropriate d
Iq ask him that question.I think that's also what you Iq for me to do so.
m just stated. m MR.TANGUAY: The fifth item fmm March im MR. HALNON: Yes. Again,I haven't asked tal 8th is that a series of engineering reviews of the l in him that specific question the way you worded it, ici adequacy of the make-up tank operating curve and )
pq but I have been under the impression that they sq other design basis parameters were in error, I l pu didn't have any words with them before the test. Du reflecting inadequate verification of design pa QUESTION FROM THE FLOOR:It was not just na parameters by management.1 was struggling with p3r a midnight shift,it was a Labor Day weekend,so ps) this, and in my addressing this this afternoon I pq there were even fewer people around. pq focused on the first four or five lines of that,
' pq MR. BEARD:Again, the memo itself has I
pq indicating that the reviews were in error.1 wasn't pq come back.It pointed out that corrective action vai sure how to deal with the last portion of that, on item, whatever it was, that there was an ongoing on where that verification wasn't done by management.
pq corrective action item ongoing.The memo stood for vel In my process management doesn't get directly i
, pe) itself.There was no close out. poi involved in the verification process of individual 1
I pq MR.TANGUAY: As you've indicated,there pq assumptions and inputs through calculations.
pq was management involvement.We were attempting to pq MR.GIBSON:I believe you previously
] pa address the issues, and unfortunately the events pa acknowledged that the verification pmcess was .
. pai that unfolded on September 5 ultimately preempted pai deficient and that more independent review was pq any other activities that were going on.We feel pq needed,and that's a managen. 't control that would psi that had that test not been run that the corrective psi be put in place without requiring personal Page 232 Page 234 p) actions that we were in fact pursuing would have pl management review.
m surfaced a lot of these issues and,once again,we m MR.TANGUAY: That's what I felt, because pl wouldn't be here today. pl management has the responsibility to assure that the pi The corrective actions that we've talked vi processes and pmcedures are in place to make sure to about throughout this presentation have been Is1 that the product is in fact correct,and that's (q implemented to address strengthening management rq where we had focused our energies on here today.We I m involvement,and we feel that through the precursors m agree with that statement,and the corrective
< Isl and the focus items program that management is more tai actions that we have put in place as a result of the tel in tune with the issues and will give them more to) discussions we've had on the other violations should l pq prompt attention consistent with their relative pq address those concerns going forward.
l ("I prionty, pq MR. BEARD: Al,at this point I can make
,' pri The last item, number five - na my conclusion or we can answer questions and then I 1 pa; MR. LIEBERMAN:If I could just ask one ps) can conclude.
pq more question. If it wasn't for the September Labor pq MR.GIBSON:I'm not sure what your pq Day evolution,you said the corrective actions were pq remarks are.You may want to hold your comments.
pq preempted.When do you think the company would have pq MR. BEARD: Then let's answer the on recognize the pmblem with the curve. On questions.
pai MR. BEARD: That's a very hypothetical paj MR. GIBSON: I would like, going back to poi question, and I don't think it's fair for us to try poi the violation on failure to follow procedure,I pq to answer that. I mean,I could say whatever I felt pq would like to explore that for a minute.What an I'm prejudiced to say and it would be difficult to py latitude or authority were operators given in pa refute it.I don't think that's a fair question. pa September of 1994 to change procedures for pai We feel that our pmcess which is in place today and pal depanmental procedures?Yourinternal pq which we'd embarked on during that time frame would pq investigation showed that two operators, as I pq have precluded something like that from happening. pq recall, thought the shift supervisor had the BROWN REPORTING, INC. (404) 876-8979 Min-U-scripts (61) Page 231 - Page 234
PROCEEDINGS BEFORE STEWART EBhETER, CIIAIRMAN IN RE:
Starch *27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 235 Page 237 pi authority to make on the spot changes.Is that p) require the review and approval of the DMPO,which ,
m correct? I'd like to understand what authority the gi is the plant manager, nuclear shift manager.or pi shift supervisor has.
p1 manager on call as defined in applicable plant gj MR. BEARD:This is with regard to the pi pmcedures.
, m second violation,the one we caught first? Processes concerning review and approval tsi tai MR.GIBSON:The failure to follow tot by these individuals remain the same.This is in procedures. m talking about those activities that would be j tai MR. BEARD: Do you understand the tal classified as unusual tests or evolutions. Finally, j tel question?
pi ensure that shift operations - and this is the l 0 01 MR. HICKLE: Not completely. pq duties of the shift supervisor- are conducted in ,
og _ MR. BEARD: Well, let's understand the pq accordance with plant pmcedures,the operating
- 02) question.
pri license and other requirements.This is the only psi MR. GIBSON:Ixt me try again.In vsj deviation authorization they have. Authorized en September of 1994 did shift supervisors have the j
09 deviation from license conditions or tech specs as ne tuthority to make on the spot changes to operating na allowed by 50.54X.lmplicit in this is an os) procedures. pej obligation to deviate from license deviations or on MR. HICKLE: Meaning did they have the on technical specifications only when this action is vai tuthority to deviate from instructions and operating pai immediately needed to pmtect the public health and poi procedures,or are you asking did they have the poisafety.
pa authority to initiate a pmcedure change process? pq Now,our operating procedures are 29 MR. EBNETER: More appropriately,I think pq conditions of our license.That is the oruy p2j you already put on the record or somebody did,and p2j autnorization to deviate that they have.
paj I'm not sure who it was, that you did not give the p3j MR. GlBSON: Let me ask just one pq operstors enough guidance in what to do on how to or pq follow-up question.IfI'm the shift supervisor and psj when to deviate from the procedures.Somebody said ps) the procedure doesn't provide the instructions or Page 236 Page 238 p) that. Al's asking you what authority did they have, oj direction that I need to accomplish what I want to ;
m so why don't you tell us that,that they could pi accomplish, can I make a pen and ink change to that pi deviate procedure without going to get some - pi procedure without independent review?
g MR. HALNON:I can answer that. 81 MR. HICKLE: You can't change the m MR. EBNETER:That's what we want to m pmcedure without going through the procedure change roi know. rs) process. l m MR. HICKLE: Let me show you two p1 MR. GIBSON: So some review other than my i ist viewgraphs that I think will address this isj own would be required for approving that pmcedure pi specifically,and if these don't answer it,this pichange?
pq will focus it a little more. {
pq MR. HlCKLE:In the instance that you pq lt's the duty of every member of the CR 3 pn just said,yes.Greg,do you have anything you want pai nuclear plant work force to strictly adhere to pri to add?
psi written policies and to comply with procedures pai MR. HALNON:I wanted to bring up,and pq written for the CR 3 nuclear plant.If a procedure pq this could be the root of where this question came psi directs them to take any action or perform steps ps) from, even though you may not know it, A1, that one pe) they know to be wmng or mav be wrong,then they pe) of the things I found that was in our AI-500 was on stop, restore and bring the di3 repancy to the on that there's a statement in there or step that says pai attention of the supervisor.This is an Al-500. pai if the procedure that you're doing and the operating pe) We're telling all the operations people they have to poi procedure you're doing has some steps that don't poi follow the procedures. pq apply,in other words,it may be the fact that the pq Now let's talk about deviations.There's pq procedure's written so that it's doing a release on 221 a few sections in our procedure entitled pri both tanks and you're only doing one tank, so pal responsibility of operations personnel, nuclear pai therefore you go through and when you get to the pq shift supervisor.This also came out of AI-500 at pq other tank you'd N/A it and write a note while psi the time.Certain operating decisions and actions ipsi you're N/A'ing it.
Page 235 - Page 238 (62) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
4.s su_: ravuwiaua uu van a ti.muu uun4 us, J.4nuiucu PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 239 Pige 241 pi AI-500 rnade the statement that if you vi reporting. Precursor reporting is what I had in ni came to one of those steps and it doesn't apply to m mind there, not deviations from procedures.
pi the task at hand and doesn't affect the overall p) MR. BEARD: You meant in a broad sense -
p1 evolution,N/A it and get a note from the pj MR. HICKLE: So that when they've got a is1 supervisor. Is] gripe or a problem complying with a management tai I found that some shift supervisors or tai standard of some kind there's a feedback mechanism.
m shifts kind of took that too far sometimes,and the m That's what I was trying to address.1 said that we tej notes that they made.some of them I think we found is; were somewhat deficient in that area because we had pi should have been procedure changes,and that was toi not developed our event-free operations program and pq back in the early '90s and the things that we have pq we did not have very wide use of our precursor pq found in the records.And we hit that head on.I pq p*~ ram,so some of the ways that we could have p2) put a paper out,and I can't remember when,I don't pri found out about low level complaints or problems or ps) know if it was before or after this task but it was psi workarounds,the mechanisms and processes werers't pq probably after,I would think,and it said that the 04j completely in place at that time back in '94.
psj only time you can N/A a step is,and I restated the ps) MR. BEARD: And then you said this.Use pq guidance in the Al-500, and you had to write that psi that one. I think this is driving at what we were on note. It caused a little furor, but they figured it on just talking about, pq out pretty quickly.That could have been some of ps) MR. HICKLE:And then management did not poi the misunderstanding where they thought they could poi provide adequate guidance on the use of routine pq N/A operating procedures. pq procedures for non-routine evolutions,and that's pq MR. EBNETER: My comment was based on pq where we said we disagree,that the procedures were p2j when you discussed management oversight.You said pai adequate for trained operators. ! didn't mean those .
ps) that was not adequate. Written guidance was okay psi two to conflict.
pq and communication was not okay.The process for pq MR. EBNETER:I just wanted to let you psi deviating from our standards was not okay,which is psi know I've been listening.
Page 240 Page 242 p) what we're talking about here.Your standard is pj MR. LIEBERMAN: There's a conclusion in p) that you will follow procedures,right? p1 your August review,Al.On page 12 is says Al-500 l
pj MR. BEARD: Right. pi and AI-400A do not provide the nuclear shift 1 pi MR. HICKLE: That's right. 81 supervisor a clear level of authority to perform tsi MR. EBNETER: And in this case they did (s) plant manipulations that are not directly covered by l to an evolution that was outside of the procedures.
to procedures such as raising or lowering the level m You said that was not okay.What was that based m within limits, cooling for temperature trending, et tai on? You said your guidance was not adequate or the tal cetera. j tq process for deviating from standards was not to) Does that mean you really disagree with pq adequate. pq that particular conclusion?
pq MR. BEARD: Let me go back - py MR. HALNON: I can address that. I've prj MR. MERSCHOFF: It was the fourth item of pri never read that particular statement but I know psi the first management concern. psi where it's coming from. Here the operators operated pq MR. BEARD: That's right.Give me a pq this make-up tank within the limits.On the 4th and psi chance to think. psi Sth they went over, but they started asking pq MR. HICKLE:I can clarify that. loq questions:Well,what is okay, can I operate between on MR. BEARD: Go back and show that slide nn this level and this level and not get an alarm, can pq again.This is where we went through the five ps) I move the pressurizer up and down to see if an pq elements of management oversight and where we nei instrument is tracking correctly.That was the pq thought we were adequate and where we thought we pq issue there that came out saying,"Well, give us pq were deficient,and I also want to get to another pn some guidance, what is a test,what isn't a test, p2l thing that we've said. pri what's troubleshooting and what's not."That was psi MR. HICKLE: One of the elements of ps) all included in our corrective action data.
pq management oversight is establishing processes to pq MR. BEARD: Let me go back,because I ps) identify deviations from standards, problem psi think I understaad your comment.Could you give me BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripte (63) Page 239 - Page 242
PROCEEDINGS BEFORE STEWART EENETER, CHAIRMAN IN RE:
Marcli 27,1996 PREDECESIONAL ENFORCEMENT CONFERNECE Page 243 Page 245 pj the page?
pj public health and safety? Hell,no.Was pi MR. LIEBERMAN: Page 12, Item C at the g3 concentration within limits? Yes.There was no pi top.
pi reason to exercise latitude.
gi MR. BEARD: My interpretation of that pi MR. GIBSON:I'd like to move on to the
- s when I read this and it's the , asis for our tsi next question.There's a document that I'll refer i im statement is that as these or
- :stors may say that (si to as Pat and G ary's expectations which was put out i m the current guidance allom us to do this,it did m in 1993.lf you read one of the bullets on those
- q not, and that's what this is saying.I mean, you
[aj expectations. it states it is okay to question any l
- si could take it another way,but that's what it meant, toi requirement,and requirement is underlined, )
am because that's one of the questions that we asked pq including regulations in company policy that don't pq them to look into.They did not provide them the 99 make sense. Rigid and blind obedience is not in the pri authority to do it. Clearly it was just an na best interest of nuclear safety.
pai objective to make the point - not that it was paj MR. BOLDT:What's your understanding of
)
pq confused.It's that they didn't have the authority ]
pq the entirety of that document? Do you have a copy psi to do it.That was the investigative team's pq of that document?
sq conclusion, that they didn't have authority to do nei MR. GIBSON: I have the list of bullets.
yn it,which coincided with ourview.
on MR. BOLDT: The reason I ask is in the saj Back in September when we had the ps) DOL letter which I understand you received a copy of 391 management review committee the investigation poi there was one page.It was a one page document.
- 2q confirmed that.That's what that means.I can see pq MR. GIBSON
- That's all I have.
- 29 how you might think it means otherm.,c, but it pq MR. BOLDT: That's the problem. It is
- 22: doesn't. p2) not a one page document.
pai CUESTION FROM THE FLCCR: Going back to pai MR. GIBSON: The reason I brought it up pq page 11,if you look under opinions, under A,it pq was to give you an opportunity to discuss it.
psi states that the operating crew needs to have some psi MR. BOLDT: That's why I asked.This is Page 244 Page 246 pj latitude for plant manipulations within operatmg pj the document.
- 2j limits performing evolutions not specifically pi MR. GiBSON:I brought it up,and 1 t me si required for megawatt production or compliance with p) just add a couple of other comments before you
- q regulatory requirements. pi discuss it.After discovery of the September 5th
- si MR. BEARD
- Yes. 1sj test, Greg wrote an e-mail to Mr. Fields commending '
- sj QUESTION FROM THE FLOOR:Doesn't that ist him for a questioning attitude, and I think. Bruce, n seem to indicate that they would have the m you had similar dialogue with Mr. Van Sicklen,I
- si authority? Isj believe, saying at least purportedly that he had a
- m MR. BEARD: No.That is a statement. pi questioning attitude and he should be commended for am that is an opinion.That's just a statement. Om pursuing something that he thought was a safety 39 Remember,the head of this was Dan Poole,who had pq issue.
- tri been a plant manager,and this is just a statement p2) Is it reasonable in your judgment for sai that in real life, when you're operating a plant, pai operators to interpret your expectation that if the {
34 ultimately you depend on the judgment and experience pq pmcedure doesn't make sense they don't have to i
- tsi and the training of the shift supervisors.They psi follow it?
I ysi have to have some latitude,and that's why the NRC pq MR. BEARD: Certainly not.That's l yn recognizes in 54F that if in your best judgment you nn another case where these two operators after the poi have to deviate, do so. Admiral Rickover told me pej fact are trying to use that as a rationalization, j poi many times,if you want to operate a nuclear plant nel which they obviously are, and that's BS. !
- 2m with trained monkeys, fine, write it down.Instead, pq Let me start from the beginning.As Gary l 24 we're going to train you and we've got this book of pu said,we spent a whole week with all of our 221 prucedures and you have to have some latitude. p2; supervisors working on leadership and expectations.
.2 31 In this case they didn't have that pal You have to go through the whole week to fully
- 29 Latitude, however.lf you g) back to 54F,it was in pq understand it,and you also have to understand all 2q our procedures.Was there an emergency here to the ps) of our expectations.What you have is part ofit.
Page 243 - Page 246 (64) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
AA C PMUU.MudW Du Utu. .H L Miu LDAL A ut, ULUMuig PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Pige 247 PJge 249 pj Throughout that course,and since we've worked very pj which people were involycd,but I'd used it so many m hard on saying whereas it's okay and we want you to m times that during his session of empowerment p) question and they did,we gave them an attaboy for p1 training he came up to me and he said,"I'm getting 91 that.On the other hand,there's pmcedures and p1 a little sensitive to you using this specific
[si processes we have in place and insist you go through (si example about communications."
isi particularly in a nuclear power plant to resolve is) So when Bruce told me that this test had m those questions.So on the one hand yes,but m occurred - again,I was in Rochester,NewYork,I'd tsj there's a way to do it. And these gentlemen were tal called back to the plant - he related to me what p) mature individuals.One had been a former navy pj was happening.My first reaction was the same as '
pq nuclear trained officer. pq Pat's.1 said it's a precursor,you realize that's pij MR. EBNETER:Is this document in the pq a precursor to the Chernobylincident.He related pri notebook? J 02: to me he was setting up a special committee.
paj MR. BEARD: No,but we'll add it.You paj And second to the fact that the event pq asked one specific question,and the answer is on occurred,which was obviously paramount in our ps) unequivocally no. psi concerns, my second most concern was having been pai MR. EBNETER:And you will provide that pe) through empowerment training, using that reactor prj document? 07) trip, having talked about communications, explaining pq MR. BEARD: Sure.We can send you the psi over and over again that's why we set up nuclear l ps) whole course,if that helps. pq shift managers,and my first question to Bruce was i pq MR. EBNETER: Don't get defensive, Pat. pq why didn't they use the shift manager.! was most pq MR. BOLDT: Let me make one comment about 99 disappointed in the crew when I found out which crew pri it, though.The empowerment course is a circle of p2] leader was on that crew that did not use a shift ,
pai activities,it's not just one.It starts with pa; manager. !
pq challenging the process. Enabling others to act is pq MR.EBNETER:The dinner' 's talking psj another piece, arid that is a key piecc.What psi about was probably an operator certificate Page 248 Page 250
- 0) happened in our view in this particular instance is vi presentation,and I don't recall that.That might j m the process was challenged,but the crew by m have been Ellis.For the record, that does happen p1 ineffective communication as well as some of our p1 in some of these certificate presentations where we gi process didn't enable the others to act that needed pl make our presentation on duties.
Isi to act. ! tsj MR. BOLDT:I think this is important ist On the very top of that second page under tsj also.These are the six steps that were in how I m enabling others to act, change was implemented and i m defined what a questioning attitude meant. Define pl we did not communicate at an increased rate.One tal the problem.Well.first of all,I believe saying ta other piece during this particular course, and p1 that there were two flags to when you really need to pq because I was involved in the 1991 reactor trips, pq question something.The first flag is having never 99 and this was another one,an example of how the pq done this before.That's a red flag.The second p23 reactor trip was a failure to communicate outside p21 red flag is I do this a lot.What does that mean?
pai the crew of what the problems were and how we should psj It means that I'm so comfortable I'm really likely pq have focused on it differently. pq to make mistakes.Those were the two flags for psj I have a little card that I generated, ps) entry.The first point was to define the pmblem pq and I talked to the folks.Stu,1 think you may psi clearly, accurately and thoroughly.We often try to pri have been at the same dinner,at one of the prj solve a pmblem that's not well defined.
~
paj operating license dinners.We talked about what a pal The second step was to consult the pq questioning attitude is, what it means, and I came psi resources,information and data,the rnan on call, pq up with a six step plan and we handed out the pq the shift operations technical adviser, system pq cards. Every time I would bring this up I would pq engineer, licensing, emergency plan,et cetera.The pri refer back to the reactor trip.That trip was p21 third step was to decide on the course of action,to pai covered in Mr. fields's shift.What we were trying pst get permission was what that step said,the man on p41 to get to is that open communications was crucial. f pq call,SRO,et cetera,if needed.The fourth step, psj l'd used it and I'd not mentioned which crew or !ast which was crucial,was before acting to consider the BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (65) Page 247 - Page 250
PROCEEDINGS BEFORE STEWART EBNETER, CliAIRMAN IN RE:
Starch'27,1996 PREDECISIONAL ENFORCEMENT CONTERNECE Page 251 Page 253 p) consequences of being wrong in step three.What if pl at those.We did that in the past in the electrical m I just made the wrong decision? And the fifth step a calculation program,the set point and the EOP p1 was to mitigate or eliminate the consequences of pi program.Again,you can only manage so much, pj being wrong. And step six was finally to act. 83 MR. BEARD:And the tank calculations.
Isj Those were issues that we had gone over and again (si That's all looking in the past.
isi largely which came out of an event which evolved - [q MR. BOLDT: The next point is to go back i m MR. EBNETER: So let me ask you m and look at the FSAR and ERs.
Isl something.Do you have one of those cards with MR. GIBSON:I think at this point I to '
sei you?
tot would like to ask for questions from ethers.What I pq MR. HICKLE: No. pq would propose we do is we give othec 'icre an pii MR. EBNETER: Do you have one of those pq opportunity to ask questions and then mayise we need pai cards?
pa a brief caucus among ourselves to make sure that pai MR. HALNON: No. na there are not any other issues that need to be pq MR. EBNETER: Do you have one of those pq addressed.
psj cards? j pq MR. BEARD: But I stdl need to make my psi MR. BEARD: Not on me, but I have one on pq conclusions.
On my desk in Crystal River. And that's also embodied pn MR. GIBSON: Do you want to do that now? j ps in our current event-free operations program. pq You can do that whenever you wish. j psi MR. BOLDT:I'll be the first to admit voi MR. LIEBERMAN: Wouldn't it be more '
pq that we're ineffective in following through with pq appropriate after all the questions?
pil that document. pq MR. IMBRO:I have no questions.
pri MR. EBNETER: And that's that point. pri MR. RICHARDS:I have a couple.I tnink ~
pa MR. BEARD: We were ineffective as of psi a couple of times today you mentioned that a curve pq September '94,and we've already said that.But pq is not a dynamic draw-down curve.It appears the
- 2si STAR is a key element of event-free ops,and I think psi operators thought it was. l Page 252 Page 254 p) that's emblazoned on my door at Crystal River.We si MR. GUTHERMAN
- That's correct.
m didn't invent STAR but we've certainly embraced it. m MR. RICHARDS:That was a fact that was ni MR. GIBSON: Moving along,I don't have p1 never intended to demonstrate how the tank would n many more here.One comment I did want to make, pl draw down in the design basis event for which it was 151 Paul.is that in your presentation you discussed is; drawn.
pi your corrective actions,and I noted that they're (q MR. BEARD: Yes. And that's another m focused on improving the process and improving m example of us not understanding and explaining to
!q future performance.Unless I just didn't hear it, ts) them what it was.We agree with your statement.
toi they did not focus on engineering evaluations and rol MR. RICHARDS:I've heard two versions of pq decisions that had been made in the past,and you pq that, too.You had two numbers today on the nu talked about things like a generic lack of 99 significance of the event. ! think the LER says ten pri operations involvement, too much reliance on cursory pri to the minus eleventh.You mentioned that earlier.
psi reviews,and a lack ofindependent reviews,which of psi Someone else said ten to the minus eighth.
9a course existed prior to your corrective actions. pq MR. BEARD: Let me answer that because I os) Now, mindful of Pat's comment that you psi asked the same question.The three times ten to the poi can't solve all the problems in the world at the psi minus eight is the frequency per year.The three on same time,I just wanted to make sure I didn't on times ten to-the minus eleventh in the LER reflects pai overlook something.My assessment of your pai the actual period of time that the curve was out of poi presentation was that you focused your corrective poi design.So the actual safety consequences of the pq action on future activities and have not made an pq time that the LER was three times ten to the minus 39 effort to go back to the past and correct errors 99 cleventh but per year it's three times ten to the p2) that might exist in previous engineering work. pa rninus eighth.
pai MR.TANGUAY: I had talked specifically gai MR. GUTHERMAN: I want to clarify your pq to the set point and EOP program.Those are pq first remark.
psi programs where we're going back and trying to look psi MR. BEARD: Why do you want to do that?
Page 251 - Page 254 (66) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
sa w cavu.waaw uu um. n c muu watita, unuium
. PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 PIge 255 Pige 257 pi We've settled the issue. pj My simpleminded understanding was that if plant m MR. GUTHERMAN: Okay. pl conditions were on the unacceptable side of the pi MR. URYC: I was curious, Mr.Halnon, p) curve and a LOCA occurred, calculations showed the pi about a cotnment you made earlier,that, quote Hebb si make-up pumps might cavitate. ;
ist does not want to operate the plant by computer. Ist MR. BEARD: Yes, sir.
ist What did you mean by that? p1 MR. HALNON: Pump, singular, m MR. HALNON: The computer is history from m MR. GIBSON: Whichever pump was aligned.
- si the standpoint of when you look at a computer point pj MR. GUTHERMAN: That's correct.
pi you're looking at it up to a minute old,and it's pi MR. EBNETER:Irt me tell you my view.
Om not operator friendly.It's not meant to be om If you're moving that curve,that's a dynamic curve o n operator friendly.You have a CRT above the control ny to me.1 don't give a damn how else you define it, j na board or you have it over here in a small CRT on the va but it's a dynamic situation because you're varying l
pa right-hand side of the control board,and Gary pai it at the point.Are you happy with the curve? '
v4) didn't think it was appropriate to ask the operators pq MR. RICHARDS:I can't answer that psi to take away their normal indication that they ps question.
psi monitored the tank with and have to monitor the data psi MR. EBNETER: Are you happy with the pn that was either old or out of place. On answer?
psi Our control board's set up in a mimic paj MR. RICHARDS:I understand the por standpoint,and on that mimic is the level and poi explanation.
pm pressure indicator,and if the operators are am MR. BEARD: You've made me do it.See pq affecting a change in pressure they'd have to go pq what you got us into? You're going m hear it again !
pa somewhere else to see how that pressure's changing, na fmm me. .
pa that would be a minute old,and he didn't feel that ps MR. GUTHERMAN: I'm sure I will.
pq was appropriate. pq MR. BEARD: We can spend the rest of the psi MR. HICKLE:And it wouldn't be practical psi day here.You know,I'm an engineer, too. In fact, !
Page 256 Page 258 pi to try to control the evolution with old data, ni I have - no,I won't say it.This curve is a locus pi especially when you're adding hydrogen to the pi of generated calculated points with respect to a pi approaching curve.You have to be able to tell that pl design basis LOCA.Whoever generated this curve in vi time,and the strip chart recorder was all they had 91 the past and in the future says if a IOCA occurs and (s) available. isi you're at this level when it occurs,what's the most tai MR. GIBSON: Jim.do you have anything at isi pressure you should have in there so you wouldn't m this time? m gas bind the pumps,or probably wouldn't, and you tai MR. LIEBERMAN: I'd be interested in pj calculate that.That's under a LOCA.That means si hearing Mr.Gutherman's clarification. pi water's pouring out somewhere in the plant.it pq MR. BEARD: Go ahead, Brian.That's all om means that this is coming down.this is coming down pq right.it's fine.You can say whatever you want. 99 all at the same time, and that's different from what na MR. GUTHERMAN: While we did not expect na happens when we normally operate the plant where psi that tank to draw down exactly on that curve,we do psi only this one's coming down.This isn't coming pq expect it will draw down within some reasonable band pq down,it's not water pouring out of some hole.
psi on either side of that curve based on the conditions psi So this curve is not a dynamic curve, poi present at the particular time with the fluctuation 5 psi it's a locus of calculated points.
On in temperature and instrument accuracy.That's all on MR. EBNETER: Curve is static, but the nel I wanted to say. pq conditions are dyramic.
poi MR. BEARD: Did that clarify? poi MR. BEARD: Yes, sir.The conditions pq MR. RICHARDS: That means it's not a pq which the curve is designed to prevent is a LOCA, pq perfect curve.It's representative of what a tank pq and that's dynamic.Now, my point is and what we pa ought to do within the accuracies of what you can pa were trying to say earlier is that however we na measure. ps) operate the plant, we have leakage in the plant and pq MR.HALNON:It's not a dynamic model. pq the level goes down or we may draw it down for some psi MR. GIBSON: I'm not sure I understand. ps) valid operation,as we do in SP 630, and the BROWN REPORTING, INC. (404) 876-8979 Min U-Scripts (67) Page 255 - Page 258
PROCEEDINGS BEFORE STEWART EBNETER, CIIAIRM AN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE l
Page 259 Page 261 pl operators noted that under that condition normal or p) speculate on what might have happened?
m assimilated HPI fluid test,that as the level was m MR. LANDIS: What I'm asking you is you '
p) going down pressure didn't seem to follow exactly pl were going to do a rigorous review of the n like this.it tended to come over some.There's pj calculation that was going to involve the l tsi nothing wrong with that.You would expect it Isl methodology evidently that you had back at that tai probably to do that because it did it.And you know is) time.You did not have the set point corrective m what? When we run 630 tomorrow we ce going to see )
m action program in place.You did not have the new tai the same thing, and there's nothing wrong with that rei set point methodology in place at that time. Is pi except that you want to give the operators an p1 that correct? )
pq operating curve that they stay below so that when a l pq MR.TANGUAY: That's correct.
pq system does that you don't challenge the design pq MR. LANDIS: That's all I have.
p21 curve. It's as simple as that. pri MR. BUTCHER: I have one clarification, pai MR. EBNETER:1 understand that we're paj and that's on Problem Report 94149.This test pq just arguing about a couple of terms. pq you're going to perform, as I understand it.it's psi MR. BEARD:We were sort of fuzzy,and I psj going to be a more severe test than what you 1 ps; think we've got ourselves clear on that. And when psi experienced as cavitation that the operators were on we run 630 tomorruw it ain't going to follow the on concerned about.
pq current calculations and it never will. paj MR. HALNON: That's correct.That's poi MR. EBNETER: Are there any other poj what's planned right now.
In questions on the curve? pq MR. BUTCHER: That's what I wanted to pq MR. LIEBERMAN:In all honesty,I thiw pq clarify.
gzi having that explanation just now was helpful. pri QUESTION FROM THE FLOOR:As of the day ps) MR. EBNETER:It was good, Pat,but it paj before they were going to run the spray pump,the pq was the same explanation that somebody else gave. pq decay heat pump and the make-up pump,whereas on the psi It's a good thing they were both the same.
psi previous they only had the decay pump and HPI pump.
Page 260 Page 262 p; MR. MERSCHOFF:I have no questions. p3 MR. BUTCHER: This will close ot.t the m MR. LANDIS:I have one question.The m test that was going to answer- well,it w.t; part p1 evaluation that was going to be done by engineering pi of the September 30th ra!culation,and the test this n to close out this issue by September 30th, was that p) outage was going to answer the questions on the ist going to be a rigorous review of the calculation? tsj concerns.
is) What was intended or what was planned? ist MR. BEARD:It will be the last action m MR. BEARD: Remember Stu's admonition. m step I think originally from 149. But I'll say isi If we don't know,we'll come back and tell you. {
Is) again what we're going to see in a dynamic system J pi MR. GUTHERMAN: That item was assigned to si response is it's going to be the exact shape of the og my group, design engineering,and it would have been om curve,and that's okay.
pq a rigorous review of the calculation.One of the )
09 MR. HALNON:It's more into the ;
v2i problems was that the calculation at that time was pri cavitation issue. l pai owned by an engineer in another discipline of psi MR. BEARD:It will address the og engineering,not in mechanical,so we would have had pq cavitation issue.I think that's really the real ps) to work together with them,but that was not a ps) issue.We think we addressed it back then and we're psi problem.We would have done so. pq going to prove it then.
on MR. LANDIS: Would that have included pn MR. LANDIS: Following up on Ross's pal review of all the assumptions and inputs that go pai question, are there any operator concerns now with poi into that calculation? ps) the test that is going to be performed,that it's pq MR. GUTHERMAN: Yes. pm not going to be complete,that it won't solve this pq MR. LANDIS: Would that have differed in pq issue?
pai any way from what you ultimately did do with the gzi MR.HICKLE:We've established a test pat make-up tank calculation.except that your pai director, Carl Bergstrom.and Carl has coordinated sq methodology today is probably more rigorous? pq all of the operations issues.He's also responsible
- 2sj MR. GUTHERMAN
- You're asking me to psi for the conduct of the test to make sure that it's Page 259 - Page 262 (68) Min-U-Scripte BROWN REPORTING, INC. (404) 876 8979
~~ ~ ~ _ - _. __ _ _ . _ _ _ _ _ _ _ ~ - - - .- - - - - - - -
- i. o u.i exotttute m*ont 3thmuu LBAh!ER, LIIAIRMAA PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 Page 363 Page 265 t 4
pj done with the right procedure and safety concerns, 19 no. And also when the alarm is in, unless you went pj so rm confident doing that that we're addressing m and took the computer point and did a plot like p1 the concerns that existed and that may exist today. pi we've shown on those graphs you have, we didn't know pi MR.TANGUAY: An additional thing we're pj how much over the curve you were. lf you were over 4
is doing, when we went to NRR a while back and is) the curve a certain amount,you'd have to go back te presented information regarding the calculation and Iq and take the data,and obviously we didn't do that.
m its development,we agreed at that point that when m We plotted it on a piece of paper fmm the strip pi we ran this test this outage we would go in and p) chart,and never more than one or so was shown under si we're going to add additional instrumentation to the si the curve,so there was no feeling to go back and
- pq key points to get as much information as we can pq check the computer data 99 about flows and pressures at various locations so py MR.GlBSON
- I'd like to follow up
- pai that we can take that and from it infer the data we prj because I have a different view.I believe that 1 psi need to get a better handle on the system,and the paj operators well understood that operating to the left
! 041 loss go coefficients as a means of verifying and pai of Curve 8 was operating in a region where Os validating what is there. So when they refer to a ps) engineering calculations showed the pump would
] ps) FT,that PT is intended to go get that additional vej cavitate in the event of a LOCA on data which is beyond the normal SP-630 we would on MR. HALNON:I don't disagree with you per run.
pej there.My point is that when the alarm came in, poi QUESTION FROM THE FLOOR: Brian,l'm poi unless you went back and realized the overshoot and pq still trying to understand this description of your pq how far the overshoot was,by checking the computer ny new and better Curve 8.Let me ask you one pu data,if you check just on your plot.on the graph, prj question.On your new replacement curve that's in pri and translated the point from the strip chart to the -
paj effect today,if you raised and lowered the level in pai piece of paper, that wouldn't show you that you were pq the make-up tank and drew a trace, would it more p41 a lot over the curve or even or . in most cases.
ps; closely mirror the slope of the current curve than psi So it was conflicting, and Bruce addressed that with Page 264 Page 266 ty it did back in 1994?
pj the poor questioning attitude they had as far as gi MR. GUTHERMAN: I would expect so because m rectifying those two concerns.
pi the methods have been more finely tuned. pj MR. GIBSON: Okay.
pi QUESTION FROM THE FLOOR:And I had one i QUESTION FRCM THE FLOOR:Following up rsi question for Greg.Back during,let's say, July rsj Mr.Gibson's question with regard to the latitude to through September of 1994, when there were a number (q that the operators have, very recently out at a m of cases when the Curve 8 alarm was in and at the m plant in which the operator knew the temperature isi time if you were outside Curve 8,according to your tsj cooldown issue, knowing that event, do you still pi calculation you could have cavitated one of the si believe that your procedure's very effective and pq make-up pumps,the one that was in service.When pq management expectations have been communicated to py the alarm was in, should the operators have declared ny the operators?
pri the make-up pump inoperable and entered a tech spec prj MR.HICKLE: Yes, we believe so.We na action statement? paj supplemented the LER, as a matter of fact,to pq MR. HALNON:In hindsight you could l pq clarify the decision-making and the human I psi probably make an argument for that,yes.That's why os performance of the shift supervisor and control em I went back to check if there were any in effect. pq board operator that were responsible for the conduct pn QUESTION FROM THE FLOOR:And why didn't l on of the,cooldown.But to make a long story short, ;
pai the operators do that during that time? psi the shift supervisor was involved with the procedure !
pq MR. HALNON: They didn't realize it made poi change.
pq the pump inoperable. pq MR. BEARD: Let me pause for a second.
py QUESTION FROM THE FLOOR:So no one in pq Stu,you were talking for a minute.Did you hear prj operations was aware,then, including yourself,that pri what Rag's asking and what we're talking about right p3) when you were outside Curve 8 it made the pump . pal now?
pq inoperable? Ipq MR psi MR. HALNON: Probably to that detail, hasj answe. EBNETER:I heard him.You're ring him.
BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (69) Page 263 - Page 266
PROCEEDINGS BEFORE STEWART EBNETER, CIIALRMAN IN RE:
"Marcli 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 267 Page 269 pi MR. BEARD: He's brought up the fact gy gave instructions to the nuclear control board m that - when was it,in February? pj operator.
p) MR. HALNON: January. p) He was questioned by one of our other i4j MR. BEARD: That's when we were cooling p1 SROs that was reviewing the data,and in questioning isi down and a condenser tube ruptured and we exceeded isj him we had to go back ourselves and go back to our isj the pressurtzer cooldown. [q supplier and get clarification to make sure we m MR. HALNON:It was the RCS cooldown m understand what tonperature to use for a !
sj rate, tai transition. And in doing so, we found out that he to MR. BEARD: And Rag, you said therefore p) took the wrong actions.Ilut it wasn't a case where pq because that happened do we still have confidence pq he felt he was stretching a limit in procedure or o n that our corrective action is effective. I think on taking latitude.It was strictly that he thought he pai thit's the essence of what you're asking. p21 knew what the answer was and he acted accordingly.
p1 MR. EBNETER: And what is the answer, yes psi So yes.1 think that the expectations nej or no?
- 04) that were put in place with our event-free program psi MR. HICKLE: Yes. !
pq were applied.I don't think that we can altogether !
pg MR. BEARD: Yes.Are we perfect? Are we Um climinate honest rnistakes.What we can do is when on ever going to be perfect? No. And I think we had a on those occur is to rnake sure we fully understand them pai good conversation on that issue with staff,with Mr. pm and that we're doing the things that we can to
)
l pej Jones personally talking. pq correct them. '
(2q MR. EBNETER: Rag,are you satisfied? pq QUESTION FROM THE FLOOR: Going back to
- 29 OUESTION FROM THE FLOOR:Yes. pq what you said earlier, Bruce,concerning some of the azi QUESTION FROM THE FLOOR:Two years have p2: issues on the barriers that were availaofe to the '
pai passed.This was in September of '94 when these paj operators on the 4th and the 5th. Relative to that pq things happened,and you say that since p41 time frarne, what was the conduct of operations as September '94 in fact the expectations program was psi guidance for a long response?
Page 268 Page 270 pi there in 1993.That's the time that you gave the gy MR. HICKLE: Do we have a copy of rt, (2) seminar,and three years from that day we still saw gj Greg?
pi some procedures not being followed under the pi MR. HALNON: In general or for that ni operators' actions which are outside the pl make-up time?
sl procedures.1 just wanted to know how does one tsj QUESTION FROM THE FLOOR:The AR-500 mi conclude that your program is effective in tq guidance alarm response procedures.
m communicating the expectations. m MR. HALNON: There were a couple of isi MR. HICKLE: The way that we conclude 181 steps.1 don't have it right here with me,but isi that is we have to go back and do a very thorough 191 there was no reference to promptness or time frame.
pq root cause evaluation on these kinds of errors that pq There was a step that talked about if the alarm came ny in the case of the cooldown rate that was exceeded, pq in due to the expected,you expected the alarm to pri the shift supervisor had made the decision,which by pai come in and there's parentheses,part of the vai the way is our best supervisor, he has no history of pa) cvolution's your doing or something to that effect, sq procedure compliance problems at all,and he was p41 it had some caveat in there about how to deal with pq involved with a procedure change where we clarified pq that.And I think that's what the operators kind of pm that procedure to talk about what temperature we use psi relied on,that step in itself.They were doing an on to monitor the reactor coolant vstem cooldown on evolution,the alarm came in,they knew why it was paj rate.Being involved with the procedure change that pai in, they knew it was going to clear when they poi was made several years ago, he felt he understood poi changed there actions, and that step kind of,1 pq the basis for the requirement and felt that that pq won't say permitted them to do that,but it kind of p9 requirement did not apply to the transition pq addressed how they were treating the alarm in that
- 221 temperature when the cooldown rate changes steps :221 case.
pai down.He never had cause to question any further. gai QUESTION FROM THE FLOOR:Was that just
- 24j lt was one of those cases where he was involved with p4) peculiar to Crystal River or is that somewhat psi the procedure change and thought he knew,and he pq standard in the industry?
Page 267 - Page 270 (70) Min-U-Scripts BROWN REPORTING, INC. (404) 876-8979
~
a ut: enoucunma uuunt sinmuu cuatica, unwona PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 I
Ptge 271 PJge 273 l pj MR. HALNON: I don't know ifit's pj MR. HICKLE:I think it did.
g standard for the industry.It was what was in our m QUESTION FROMTHE FLOOR:That was the pj procedures for many years. pj read out station.
pi MR. BEARD:I think a key point, Curt, p) QUESTION FROMTHE FLOOR:When you ts) and a contributing factor to this question of the [sj plotted it on the recorder it's over.
A alarms,we said procedures were weak with regard to isj MR. HALNON: At the worst point or some m timeliness and normal response.We fully m point going after.
Is) acknowledge that and they've been made better.Now, p1 MR. HICKLE: Maybe not initially.
pi in September of '94 was that the industry standard? pi MR. HALNON: The one on the 4th, using pq I don't know what the industry standard was,but it pq the same criteria that we used to talk about the 09 sure wasn't good enough and we increased our pq other nine,l'd say probably no,they wouldn't have pri standard. prj seen it.But towards the end when they were at the p33 QUESTION FROM THE FLOOR:There was also psi worst case they would have seen it,and on the 5th I j pq some discussion about the make-up tank level pq think that's the same thing that would have ps) pressure recorder versus the alarm indication or ps) happened.It was probably less than half a pound.
pe) computer indication,and you said the make-up tank psi l'd have to say that for sure.
pn recorder was the only instrument they had to make on MR. HICKLE: We said that the worst case pai the determination to whether an alarm was valid or poi was 55 incher over one and a half pounds,and we had pei not. pej that good a resolution on the strip chart recorder, pq MR. HICKLE: It provided real time pq During the evolution they would have been able to pq information. pq verify by the strip chart recorder tht they were in J pa QUESTION FROM THE FLOOR:My question is pri an unacceptable region of the curve without a doubt, . {
p31 that the expectation was to use that recorder to pai but probably not initially when they first got the pq validate the computer alarm. pq alarm.They would have had the same resolution pq MR. HICKLE: That's what they were doing. pq problem.
Page 272 Page 274 p) QUESTION FROMTHE FLOOR:Are you aware ni MR. HALNON: On the 5th they were .5 pi of the e-mail that came out from engineering that gi pounds over the curve and on the 4th they were a l
p; said use the computer point to add hydrogen and use pi little less than a pound,over a half. '
pi the make-up tank recorder to monitor operations, pi MR. MERSCHOFF: In the worst case on both (si that the alarm takes precedence? isi the 4th and the 5th would you have seen using the si MR. HICKLE: I think we discussed that. pi strip chart : corders that yeu were in the m That was the e-mail that went to Mark van Sickle. m unacceptable regiord is) MR. BEARD: What I heard and what I isj MR. HALNON. Yes. l wi believe is the case is the memo went from the system pi QUESTION FROM THE FLOOR:And on the 4th pq engineer to Gary Hinman and Mark van Sicklen. Gary om they were two pounds over as opposed to 1.7 on the !
09 Hebb didn't like the computer and that was the end 99 5th.
pri of it.so management wasn't aware of it at that time pri QUESTION FROM THE FLOOR: Going back to p3) until we were doing our investigation. usi what Brian clarified on the curve here. He said py MR. LIEBERMAN: So the system engineers pq that there's a band on either side of the curve.
ps) know that when they wanted to give guidance like psi MR. GUTHERMAN:lt's not an explicit psi that it should be going through operations? psi band,it's an implicit band.
pn MR. BEARD:Yes, sir.That's one of the on QUESTION FROM THE FLOOR:My question is psi symptoms we told you was weak and we've recognized pai that if the band is actually below the curve,if the pe) that.We've taken steps to address that. poi actual system response is below the curve,then pq QUESTION FROMTHE FLOOR:When you pq operating near or at the curve would put you outside pq pointed out on the 4th and the 5th and you said that pu the design basis,right?
pri they violated OP-402 by going over Curve 8 when they p2j MR. GUTHERMAN: The curve is offset for pai did the hydrogen addition,would that hydrogen psi all those uncertainties in the worst case pq addition,the amount over Curve 8,would that have py direction.
psi shown up on the strip chart recorder? psi MR. BEARD: It is now.
BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripte (71) Page 271 - Page 274
PROCEEDINGS llEFORE STEWART EBNETER, CliAIRMAN IN RE:
Marcli 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 275 Page 277 pj CUESTION FROM THE FLOOR:But what about to basis and therefore reportable,and I think that was m September 4th and 5th?
m exactly the intent of the other words.
p) MR.GUTHERMAN: Knowing what we knew at p1 MR.GIBSON: So it might have been more ni the time,that was the attempt. 81 accurate to say we may have been on, near or over, tsi MR. EBNETER: But it wasn't. (sj MR. KELLEY: Are you referring to the LER n MR. GUTHERMAN: But it was incorrect at tej statement under cause?
m the time. m MR. LIEBERMAN: Yes. ;
isi MR. EBNETER: The present curve has this MR. KELLEY:It says each time the (81 m margin built in,is that right? m operating point was on or near the curve as pq MR. GUTHERMAN: Yes. pq indicated by the main control board instrument, pq ;
MR. BEARD: When the system responds in a sq referring to the chart recorder,the make-up tank !
pa normal revolution and it doesn't match the '
va could have been operated outside the design basis.
psi calculated theoretical locus of points it's no big pq That appears to me to be in relation to the pq dez.l.It's not going to come close to the design p4) difference in the computer point and the control ;
psi basis. pq board recorder.
pa QUESTION FROM THE FLOOR:Thank you. l ps; MR. LIEBERMAN: Okay.
pn MR. LIEBERMAN:I hate to do this, but l on MR. RICHARDS:I want to make sure I !
psi just to go over the last question that Curt raised pq understand what you just said, though,and I think I psi that Brian has answered.Each time the operating vej understood before, but maybe I'm getting confused am point was on or near the curve could it have been pq because the curve was offset non-conservatively,it I pq operating outside the design basis c. it was pq was incorrectly drawn,and you had people operating ga cperating outside the design basis? pa in the acceptable region of the curve as it was pai MR. GUTHERMAN: Knowing that the curve ps) defined at that time for who knows what period pq was incorrect? pq because they were moving around in there and they psi MR. LIEBERMAN: This is November of '94. psi were in violation of the design basis because the Page 276 Page 278 pi MR. LANDIS: September, you mean. pj curve was incorrect.You haven't gone back and m MR. GUTHERMAN: My November memo is what m counted all those times, but now that you've figured p) you're referring to.The answer is it would have, p) out where the real curve should be I guess you can gj because we kriew the curve was non-conservative. p) append that there were lots of periods of time where tsi MR. MERSCHOFF: Near is an imprecise tsj they were operating in violation of the true curve N term.Some nears would have been okay and some ts) and not knowing it.
m nears would have not been okay. m MR. GUTHERMAN: That's correct.
- sj MR. GUTHERMAN:I didn't get to that tsj MR. RICHARDS: That's what I heard you m levelof analysis. Is) say before.
pq MR. LIEBERMAN: I'm talking about th.: pq MR. BEARD:I'm not sure what the point p q LER, not your memo, and these are your words, nN pq of the question is, Jim.ls there some inference pa mine,and so I'm asking the company,when they use va that we should have done something else? I'm not pai the phrase was on or near the curve could ha ve been sai sure what you're trying to get to, and that would be pq operating,what did you mean? What were you trying pq useful to us.I think the stateraent means just what pq to communicate to us? That's my question. psi Brian said, and I think that's what it meant to me pej MR.GUTHERMAN: What we were trying to ps) when I signed it.We were acknowledging right up on communicate,in my opinion,was the fact that the on front that with regard to operating by the main pai curve had not been redone by the time the LER was ps) board controlinstrument,which was the strip chart ps) issued so we didn't know where the correct curve was psi recorder,we could have been outside.We didn't pq going to end up. pq know. Are you inferring that we should have now pq MR. BEARD:I attended the i1-16 meeting pq done something else, gone searching for other times, ga where you explained why it was a design basis and in pa done more calculations,seen how many times and pai particular while we were on the evolution on pal wrote another LER or what? Our corrective action pq September the 5th we had done a sufficient p41 would have been the same no matter what.
psj calculation to prove we were outside the design psi MR. LIEBERMAN: In reading this,I would Page 275 - Page 278 (72) Dun-U-Scripte BROWN REPORTING, INC. (404) 876-8979
PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 PIge 279 PIge 281 j pl have expected this would have said each time the MR. BEARD: There was one before that and ni m operating point was on the curve it was operating m 1 said somebody note that.What was the first one,
. pl outside the design basis.Near the curve may have p) Larry?
pl been outside the design basis.it could suggest to pi MR. GUTHERMAN: You wanted to know when is) me that there's a possibility. But I thought at tsj specifically each shift was counseled regarding the tai this time there was a greater appreciation that if isi event.
m it was on the curve it was outside the basis and if m MR. MERSCHOFF: The first one I wrote isi it was near the curve it may have been outside the inj down was why didn't engineering review the LER for pi design basis. pi September 4th.
pm MR. GUTHERMAN: 1 don't understand. Is pq MR. EBNETER: Who kept track of these on ny there a question there? I'd like to answer it. 09 yourstaff?
pri MR. GlBSON: Let me try it. fic's pri MR. MERSCHOFF:I wmte down four of i psi questioning the accuracy of the LER.The question ps; them,yes.The first one that was asked was why j pq is you said if operated on the curve it could be pq didn't or did engineering formally review the LER pe outrMe the design basis.The fact of the matter is psi and note that the September 4th event was not vej at < m dme did ycu not know that operating on the nei mentioned.The second one was were Mr.llinman and j pn curve was outside the design basis? pn Salzman counseled for not bringing up the '94 4
pai MR. GUTHERMAN: We knew that the curve psi issue.The third one was when were the crews 2
poi was non conservative.Therefore operating on it poi trained or counseled on the 94 and 9-5 test.The pq would have been outside design basis. pq fourth was to pmvide a copy of Pat and Gary's j pq MR. GIBSON: Nu could have been but was pu expect::tions.
- pri outside.1 believe that ww t.9 question. paj MR. BEARD
- We're agreed.
i psi MR. BEARD: Yes, but I think somebody's psj MR. MERSCHOFF: Did I miss any more?
{ p4i hit the key point now that it comes back to me. At pq QUESTION FROM THE FLOOR:1 had one
] psi this point the information that we had was what? At psi more. Specific dates when the shift meetings were i
] Page 280 Page 282 p1 least the infortnation that we explored and also that pj conducted as close as you could come.
m we explored subsequently was the strip chart m MR. BEARD: That was his number four.
, pi recorder. And I think it's already been said here ni QUESTION FROMTHE FLOOR:I'm sorry.1
! p1 that in fact in our subsequent August evaluation the p1 guess I didn't hear him.
i tsi strip chart recorder indicated that they were okay, p1 MR. BEARD:It was actually number to So looking at the stnp chart recorder data - isn't si three.
m this right? Somebody's already pointed this out, m MR. EBNETER: Let's stop and go back over 1
(s) You couldn't tell on that because it was offset from tai those four.
pi the computer alarm.I'm not sure what's the intent pi hR. BEARD: Do we need to write those pq here. pq down?
py MR. LIEBERMAN: That's fine.1 think py MR. EBNETER:I just want to make sure.
pri what you've just said may have cleared it up, pel Go over the four.and whoever asked for them,do you )
psi meaning the main control board instrument. If this psi want it? If you want it, we'll get it in writing. l pq had said the computer,the digital,maybe my concern pq Go over the four. I psj would have been more valid. pq MR. MERSCHOFF: Number one is why didn't )
om MR. BEARD: I think that's correct. pq engineering or did engineering review the LER and l
pq MR. GIBSON: Could we have a review? pn notice that 9-4 wasn't mentioned. i
~
pal You've agreed to provide us a couple of things in psi MR EBNETER: Who needs that response? l poi writing. Maybe we should review that. pe) MR.LIEBERMAN:I would like that.
pm MR. GUTHERMAN: The items I have, Pat, pq MR. EBNETER: You'll give us a written py are we were to confirm whether Salzman and Hinman
)
pq response on that. ;
pri were counseled with regard to how they reviewed LERs pr MR. MERSCHOFF: Number two was were Mr.
ps) and specifically the description of the event. psi Hinman and Salzman counseled for not bringing up the pq MR. EBNETER:And you still want that.1 pq think you're the one that requested that. lpq psi 9-4MR. event.
LIEBERMAN: I would like that.
BROWN REPORTING, INC. (404) 876-8979 Min U-Scripte (73) Page 279 - Page 282 i
PROLEEDINGS BEFORE STEWARf EBNETER, CliAIRMAN IN RE: -
' March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 283 Page 285 gy MR. EBNETER: You need that written p) Do you have a deadline you need them by?
pi response. pi MR. EBNETER: When do you need them, Mr.
pi MR. MERSCHOFF: The third is the specific pi Lieberman?
gi dates when crews were trained on the 94 and 9-5 pi MR. LIEBERMAN: We'st not going to have tsj tests.
gs) this action out tomorrow.That's for sure.Say a im MR. LIEBERMAN: I might have wanted it. te week, m I would like that now. m MR. BEARD: One week.lf you can't do raj MR.HALNON: And we sr,id we'd do the best si those simple questions in one week -
pi we could with that. pj MR. GUTHERMAN: Give us until the 5th of pq MR. BEARD: We're going to write all pq April, the end of next week.
pu these to NRC. pq MR. EBNETER: That's fme.
p2; MR. MRRSCHOFF; And the fourth was the pai MR. BEARD: May I conclude?
psi expectations. paj MR. GiBSON: You may.But I'm going to pq MR. EBNETER: Are there any others that pq ask the NRC staff to step next door and make sure we psi you want a written response to now? pq have no otherissues to raise.
pq MR. LIEBERMAN:Will you have a copy of pq MR.LIEBERMAN: Five minutes.
pn the expectations that you could leave? pn (A recess was taken.)
ets MR. BEARD: Yes. pq MR. GIBSON:I think we can go back on pa MR. BOLDT: This has got notes on it. poi the record.The NRC staff has met to determine if pq MR. LIEBERMAN: Is it possible that you pq there were additional questions or issues that pu could fax us a copy? pn should be discussed during the meeting.We really par MR. BEARD: Sure. mai have no additional questions.We did note that
- psj MR.EBNETER:We need to get these psi you've agreed to supplement your last LER,and at pq responses in a very short period of time. pq so..ie point in the future we'd like to discuss the pq MR. LIEBERMAN: I'm only referring to a 99 schedule for doing that with you.
)
Page 284 Page 286 l 01 copy of the document.If you could fax it to us p; MR. BEARD:This is the LER for what?
pi tomorrow morning,I would appreciate it. gj Which LER?
pi MR. BEARD:We'll let you Xerox that, p1 MR. RICHARDS:I believe it was the l pi Jim. pi curve.
Is) MR. EBNETER: You can white it out if you isj QUESTION FROM THE FLOOR:Following the iq want,but I would advise you not to. is) September 5th event,the one in November.
m MR. KELLEY: h's just Conklin's notes he m MR. BEARD: Bruce, art you listening?
pj used in teaching the course. [sj Was that the LER iou were talking about we were si MR. EBNETER: So that's one of them pi going to supplement? I poi that's off the list,then. pq MR. RICHARDS: The LER says you're going I pq MR. MERSCHOFF: Yes, sir, number four. On to supplement the LER.
pr; MR. EBNETER: So you still owe us three. paj MR. KELLEY:Which one are you looking ps) And I asked a question and I didn't hear anybody. ps) at?
eq Are there any others that the staff wants a written pq MR. RICH ARDS: 9400901, March 1 st, '95.
psi respome to?There are no other questions we want a psi MR. KELLEY: It was originally written in vej written response to.The next question is when can ps) December and it was supplemented in March.
On you gwe us those responses?The staff will have to pn MR.GUTHERMAN: And we said again we're pai consic er them as part of their deliberation. ps) going to supplement.
pe, %R. BOLDT: I'll confer with these guys, poi MR. RICHARDS:In the supplement it says pq but I don't see any reason why they can't be faxed pq it will be supplemented, pq before the close of work tomorrow. pq MR. MERSCHOFF: The first page,the last p21 MR. BEARD: Don't overcommit.We've pai sentence,this report will be supplemented with psi learned that lesson.We've got an outage and 25 paj additionalinformation.Page one of nine.
pq million things going on. pq MR. BEARD:I want to make sure that we pq MR. BOLDT: When do you need them,Stu? psi understand what we're agreeing to do,and we'll Page 283 - Page 286 (74) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
4a tu.: ravu.wiava ou wu. m muu uuu us uomum PREDECISIONAL ENFORCEMENT CONFERNECE March 27,1996 )
Pige 287 Pege 289 1 (g agree to do whatever's right. pj ER. And the one person that we know for damn sure m MR. EBNETER:Well,let me put it this m knew about the September the 4th didn't say anything pi way.1 don't want to debate this this afternoon. p) in his review of the ER. ! want that on the l pi There's a need for another meeting with you on some p1 record.
(si of the engineering issues,I think that's clear, is) Conclusions.The first statement is not m that are outside this enforcement hearing.NRR and (s) a conclusion.1 think it's apparent to everybody.
$ m the region will get back with you on that meeting, m We accept responsibility for the violations.We ial and at that meeting we will discuss with you what is) couldn't do anything else,could we?There were m has to be done with that ER. p1 deficiencies in some elements of management 1
pq MR. BEARD: That clarifies it, because pq oversight in September of '94. And the reason I put pn I'm thinking what are we going to supplement it on that up there is because it relates to the pri with? What's the information? pri difficulty that we've had and I think you've had in i
- pa; MR. LIEBERMAN
- You may want to just psj having an enforcement in March of '96. In one way j v4) change it and say there's nothing more to 041 it would have been easier to have an enforcement in
~
psi supplement, but we just have something on the record psi December '94.This is March of '96.There werc psi we want to clear up. pej oversight deficiencies in September '94,we think j pn MR. BEARD: Do you understand, larry? on that we've addressed them, and of course that's what psi MR. EBNETER: That's fine. It's clear pm the next bullet says.We feel they have been poi that you said the statement and we need to get it usi addressed.They've been addressed,they were 4 pm covered,and we can do that in a subsequent sq addressed initially, parts ofit or parts of the pq meeting. pq corrective action of the violations,in our pri MR. GIBSON: Jim,did you want to say pri management review committee meetings. .
1 pai something? paj 1 went back and read it,and they were p4j MR. LIEBERMAN: My comment's just on the p4) pretty good with respect to what we knew about 4 psi end. psi September the 5th, the principles involved and the Page 288 Page 290 pj MR. GIBSON: Pat? p) human performance issues.They've been further pi MR. LIEBERMAN: I'll just - p1 addressed starting in February of '95 with our pi MR. BEARD:I thought I had the last p) management corrective action plan,which includes 8: word, Al. pj the precursor program, which includes all the things tsi MR. LIEBERMAN: This has been a long is) that we've talked about. It's also reflected,and tai conference.1 want to thank you for your patience is m and the information you've provided us.You've I a)previously this is something that's provided it to the NRC.not a recora nere, but I rei discussed a number of corrective actions that are isi Every year we go back and revise what we l im going to involve management of both engineering and tg call our long range plan for excellence and {
pq operations to pull off.The test of your success om opemions,a five year look ahead,and as a result l pq will be obviously in the results, and neither you pu of that we then develop our business plan for the i pri nor us can afford to have to have a similar visit pri coming year.When you read our current version of psi like this in the future. paj our long range plan and our 1996 plan,all of the p4) MR. BEARD: Totally agreed. 041 focus,the intent,the increased focus on safety and ,
psi May I conclude? ps) human performance that we've talked about that are l ps) MR. GIBSON: Yes. psi embodied starting with the management review pn MR. BEARD: Now I have a chance to put pn committee actions, carrying through the management psi one more thing on the record, and forgive me for pai corrective action plan and subsequent actions are nei doing this.1 want to talk about that ER again, pq cmbodied in those documents as a reficction of the um 009.The point I'd like to make is there's been pa first element of managewnt oversight which is to pq questions raised about if engineering review it,why 99 have wntten guidance on what our standards and p2i didn't the 4th come out. I'd like to point out that pri expectations are.
mai the one person who surely reviewed it and wrote a pal As another example of our corrective 941 memo of which we have a copy,and so do you,was Mr. p4) action I mentioned that we now have issue managers psi Fields,and that memo discusses at great length the , psi of key issues. IIere's the latest,and this is in
, BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (75) Page 287 - Page 290
VROCSEDINGS BEFORE STEWART EBNETER, CHAIRMMN IN RE:
March 27,1996 PREDECISIONAL ENFORCEMENT CONFERNECE Page 291 Page 293 pi this book that I will leave with Stu.These are the pl wouldn't have happened if our design hadn't been so pi latest issue manager updates for the management A complex.Other plants don't have dual systems.So pi corrective action plan,and it will be in your pl we could say," Hey,if our system hadn't been pi book.51 of 55 items have been completed.That's pi complex on this issue and we didn't have all these m 92 percent.Our make-up tank action plan is 82 tsi competing things it wouldn't have happened."It (si percent complete, and it's ver" cicar relating back tai wouldn't have happened if we had not decided let's m to the problem report referer e.This is where we m meet the chemistry guidelines.We could say that it Isj capture allof this. Isj wouldn't have happened if we'd given them an m I also would like to stress again that, m operating curve to start with instead of design pq and this again is embodied in our event pre-ops in om basis,and you would say rightly so.So what? So o n our '% business plan and our long range plan, and pq we're not going to take that approach.
l pri it's the back end of the oversight process,which is vai So why did it happen? It happened os selfessessment.There are three things I'd like to pal because,one, coordination and communication among na stress again. Senior management,that's me, Gary. 04j design engineering, system engineering and
! psi cnd the directors have embarked on a periodic take a pq operations was deficient.In addition,it happened l pe day or two sit-down and self-assess how are things nel because our process for managing complex changes was l pa going, get away from the forest and the trees and on deficient.it also happened because some elements l poi take a look.This is sornething Virginia Power does em of management oversight that we've talked about very pq very well.We've picked up that. poi clearly were deficient.Even with all that,it pq An a.ijunct to that is at the department l pq happened because the shift supervision of that crew pu manager level we do quarterly the me thing.We my exercised poor judgment and didn't meet their pai look at NRC inspection reports,the precursor p21 responsibility.That's why it happened. And our
- l ps; trends,IMPO, whatever it is, kick up our feet and psi corrective actions have addressed all of that.
l 941 take a day off.We do it off site. As part of p4) Why didn't we tell you sooner? Why psi that,and again, we picked this up fromVirginia psi didn't we know sooner? The reason that we didn't l
Page 292 Page 294 p1 Power,we take the elements of a good plant p) know sooner is that we didn't do an extensive root I m performer that's the NRC's own criteria and our QA A cause investigation in September beyond the pl department,and going down all those attributes p1 management review committee.But that's somewhat l gi gives us an assessment of how we're doing- si hypothetical and hindsight because the sub<luestion i tsi Basically we grade ourselves south level one,two or tsi is well, why didn't you do that, and I think we tai three. Im addressed that.We believed the operators,what j
l m We also have in place now a nuclear m they told us at the management review committee.
tal rssessment team headed by the former assistant plant te They wrote Probiern Report 267.Mr. Fields reviewed l
l m manager at North Hanna.It's staffed with a number R it, wrote a memo,didn't mention it.The technical l i om of top quality people,an organization that has pq issues were being addressed through Problem Report l py taken the lead on root cause analysis when needed. p q 149 and 267.We already knew that.There was no l l pri They operate our precursor and problem report pri reason to suspect otherwise.
psi program It's going to give us an added element pal We didn't know sooner because the three 04 that we've not had and that other top performing pq engineers who did know, system engineers, felt no l psi plants do have. pq reason - they didn't see that as an issue.Now,we pq Now,I said that I would come back and pq can say they should have,but they didn't until one pa answer the three key questions in our mind and I on of them read theJuly investigation report and poi think in yours as you've expressed,why did the nel noted, Gee,it's not in there,and thank goodness he pq make-up tank test happen,why didn't you know about poi came forth.We didn't know because the shift pq September the 4th sooner and therefore tell us, and pq supervisors,the shift supervision withheld pq why didn't you fix the curve sooner.I'll give you pn information,and the rest of the crew could have azi the answer.I thints we have collectively,but I p2 been more forthcoming.So that's why we didn't psi just want to make sure. ps) know. lf we had,we would have told you.We told 94 To answer the first question, why did it pq you as soon as we knew,we being management.
irs happen,1 could have started offlike this.It ps The third question is why didn't you fix Page 291 - Page 294 (76) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979
a n: PROCEEDINGS IIEFORE STEWART EllNETE t, CIIAIRMAN
{ 'PREDECISIONAL ENFORCEMENT CONFERNECE Maech 27,1996 Page 295 Page 297 pj it sooner, why didn't you ftx it between May and pj MR. EBNETER: We do appreciate your l m September,the curve,and I think we've already m coming in,and yourinformation I thought was well
! pj covered that.We had deficient coordination and pi presented,and we appreciate that.It does take a
! pj communication among design, system engineering and 8) lot of time. It took a lot of time by our staff.I a
tsi operations.The calculation development process was is) would expect your commitment is to get us that other tel deficient and management involvement was tai information by the 5th.
l m ineffective.That's why we didn't f~tx it.Our m MR. GUTHERMAN: Correct.
i tal corrective actions addressed that. tai MR. EBNETER: So we'll be looking forward
- rei You know,the hardest way to learn is m to that. And we 11 get back in touch with you with
} pq from self-experience.Now, we all talk about we've pq regard to the follow-up meeting on the engineering.
- on got to make use ofindustry experience and we try to 09 So with that,I'll just close the conference.
l pa do that and all that sort of thing,and na MR. BEARD:I do have one question.1 i
psi self experience is the hardest way We can verify 931 think the fifth item to your four is to whether we I v41 that.But you know, perhaps also it's the most 04j supplement LER 09 or not. Also, we'd like to ask l psi effective way from the standpoint oflasting.We're ps) about the tunscript.When willit be available and I
psi not going to forget this one.We've already paid a psj will we get a chance to review it for corrections?
pa high price for this self-experience.'lwo people on MR. LIEBERMAN: Our past policy and 081 have lost their jobs. Four others have had their paj practice has been that we do not disclose the vej careers adversely affected.There's been a pe) tanscript of a closed conference until after we um significant impact on our reputation and our pq make the decision on the enforcement action,and 99 morale.There have been significant management pu then we only disclose it if we get an FIA because it pa hours involved,and that's an understatement. rmj was a closed conference. ,
psi There's been negative press, we have four apparent paj MR. BEARD: But it wasn't closed to us.
p41 violations,and we're here.But we have learned. R*1 MR. LIEBERMAN: But if we give it to the psi Corrective actions are in place, and I will say on RSI licensee we give it to the world. If you'd like to Page 296 Page 298 pi the record that it won't happen again. m have it,we'd be happy to give it,but only under m The last thing I'll say is a saying that m those circumstances.
pi l've used with our people.They've heard me say it p1 MR. BEARD: Thank you.We do appreciate si before, but I think it applies.There's a saying 81 y ur forbearance in listening to us.
(s) that goes,In the moment of darkest defeat is the Isi MR. EBNETER:I should ask you one more tai greatest opportunity, and we'd like to think we've tal question before we close.Normally the rnaterial m taken advantage of that opportunity. . m y u're g ing to give us,is there anything that's is) Thank you. 181 private inf rmation or proprietary information that tej should not be disclosed?
tai MR. EBNETER: Thank you, Mr. Beard.With pq regard to experience and learning, experience is a "" # I "' * " **
pq difficult learning situation,that's for sure.It "" ## #' ' " * "**'#* I****"**
v2) it' na gives you the test first and then the lesson later, MR. MERSCHOFF: That book will go in the ps) and I think we all understand that.
41 PDR?
0 41 In closing,we need to remind you of tw psj MR. BEARD: Both can.
psi things.One,the apparent violations as wntien are psi MR. EBNETER: Thank you very much.We nei subject to change,further review,and particularly on appreciate it.
pn in light of the information you'll supply to us. pel MR. BEARD: Thank you.
ns: Secondly,the statements, views. expressions of the g,j (j(oceedings concluded at SMO p.m.)
poi staff or lack thereof are not intended to represent pq pq the final agency determination. As Mr.Lieberman pq pq has indicated,this is an agency process.I don't pa ga know what prediction - pai ps) MR. LIEBERMAN: liopefully sooner than p4j p4j later.That's about as good as I can say. ps; psi MR. BEARD:We're in favor of that.
BROWN REPORTING, INC. (404) 876-8979 Min-U-Scripts (77) Page 295 - Page 298
140( LDINGS bel ORE STEWART EENETER, CIIAIRMAN IN RE:
Marcli27,1996 PREDECESIONAL ENFO~ CEMENT CONFERNECE i
Page 299 til P1
)
p1 CERTIFEATE ,
1 isi I
tai STATE OF GEORGIA: '
m COUNTY OF FULTON:
a; I hereby certify that the foregoing toj proceedings were taken down,as stated in poi the caption,and reduced to typewriting under
- pil my direction,and that the foregoing pages 1 );
v2) throuph 298 represent a true, complete, I pai end correct transcript of said proceedings.
pq !
This, the 6th day ofApril 1996. j psj '
pel o7; Keith A.Wilkerson,CCR-B4381 My comrnission expires on the ps) 30th day of May,1999.
val l 2 01 Ril
)'
- ni 2 31 R41 PS) l l
l 1
i
]
l i
l l
i i
l 1 Page 299 - Page 299 (78) Min-U-Scripte BROWN REPORTING, INC. (404) 876-8979