ML20115H522

Test Procedure Pr 94-267 TP-1 FPC Crystal River Unit 3 Makeup Tank Pressure Limit Curve Verification, Rev 0

ML20115H522
Person / Time
Site:	Crystal River
Issue date:	11/16/1994
From:	FLORIDA POWER CORP.
To:
Shared Package
ML20115H481	List: ML20115H511 ML20115H514 ML20115H522
References
PROC-941116, NUDOCS 9607230175
Download: ML20115H522 (10)

Text

._. . . . _ _ .__

Rev. 0 Effective Date __

TEST PROCEDURE I

PR 4 94.E47 71 l

1 FLORIDA POWER CORPORATION '

CRYSTAL RIVER UNIT 3 i . l

! l MAKEUP TANK PRES 5URE LIMIT CURVE VERIFICAT10N THIS PROCEDURE anneretre SAFETY REIATED COMPONElffs APPROVED BY: INTERPRETATION CONTACT DATE:

INTERMtETATION CONTACT: Manager, Nuclear Plant Operations 9607230175 960710 PDR ADOCK0500g2 G

,M 4

i 4

h, i

iABLT OF CONIEN15

, O J

i E l 1.e a

putentr.......................................................'. ... I i 1 2.8

) Rf5FRFar rt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......

2.1 . . . . . . .1 . . . .\

1.2- .. f 2 lis. . a

............................. 3..

m E . . 2 (M1 u ...............................

1 j 1.s --- 1 ma a 4

1 3.3 i g ;iiJii ...........................'..............

g............. I 3.2 j 3.4 l ly g ... ................................

i 1

,2 a

== 1 88 <

TH................. ........................ 2 j 4.0 i,

imi--iinne..................................................... p3 5.D FDt t GW. UP ACT nur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. . . . . .

j 5.3 raar'tacrart n ..........................................

d 3

3 i

e t

,i n

e^

1 e

1 ,

• 1 b

PR M 267 TP.1- Rev. O Page i O

_ _ _ , _ . _ _ ._ . _ - . _ - - ~ - - - - - -

1.0 puRpost The purpose of this test is to verify the accuracy of OP.1038 Curve 8.

• Maximum Makeup Tank overpressure". This procedure will provide "after the fact" documentation, review, and approval of the testing previously performed, which served as the bt.sla of

.probles Report M 267.

2.a arrrararre ~

. 2.1 IfrLINENTINE REFIREMIS 2.1.1 Op.402, Makeup and Purification System 2.1.2 Op-1038. Plant Operating Curves 2.2 DEVELSPflENTAL REFERDEES 2.2.1 CALC # 190-0024 Rev. 5 2.2.2 Probles Report # 94-149 i

j 2.2.3 Probles Report 8 94-267 l 3.0 l HR~ i faaaTRiMafiou 3.1 DESCR1FT10N l 3.1.1

! Maximum Makeup Tank overpressure must be limited to prevent i hydrogen entrainment in the suction of the Makeup Pumps during a LOCA.

i The martaus overpressure allowed at a MUT level of 55' is derived in Cales1ation 190-0024 Rev. 5. The methodology used a incluees calculating the fari press ring worst case LOCA flow i conditions which would result in eluun of water in the MUT outlet Itne. This safe press 2 i I

compressed to a level of $5' in the fulf uti w itne is then i

Law. Lhe idea) Gas i OP 103bCurve 8 is generated by expanding and compressing j this safe 55' pressure over a 0 to 100 inch range utiltzing the Ideal Gas Law. .,

Y, .

i I

l PR 94 267 TP.1 Rev. O 1

~

. Page 1 s

j 1 . - - .

i i.

i 1

} .

3.1.2

Data. collected during $P 630 and documented in PRf 84-149 suggests 1 '

that OP 1038 Curve 8 is noncenservative. If the faff gas space i

behavier can be described by the ideal Gas Law, uhtch is an v

assumption of the calculattens, then the leff pressure response te a level decrease should parallel the limit curve. The actual pressure responsg curve converges on the 16stt curve when pressure is in the acceptable regian and level is decreased.

1 1 .

3.1.3 This test will tattially fill the 1817 to a.ievel of 45*.

4

. 187T i .

pressure will then be increased to the curve 8 limit. The system

~

will be alleued to stablitae for apprestestely 30 minutes. The ,..

ftfi level will then be blod to a level of 85".

be refilled to norms) operating level, and falf The fulf will then necessary to remain within the Carve 8 Itait. vented if j og the test,

. the Austilary Building operator shall be stattened to the AB d

prepared te vent the Inft if pressors is above the limit and a LOCA occurs. Data will be collected uttitring'the RED 45 system.

j 3.2 LIIIITS Alg) PRECMT1tMS 1

3.2.1 If during the performance of this test,181T pressere is above the limit et OP.1038 Cerve 8 and indications of increased RCS leakage occur, immediately terminate the test and went the Itif to reture MUT pressure to within the Itaits of OP-3038 Curve 8.

3.2.2 For me-k located in Radiation Centrolled Areas, due consideretten

)'

aust be given to the ALARA program. Thts may reselt in a determinatten that special preparattens and/or precautions are necessary.

1

{ ,

3.3 KCEPTANCE CRITERIA l

' 3.3.1 ~

Actual MUT pressure is less than er egual to the pressere Itait of 1 07 1038 Curve 8 at all levels utthin the test range.

1 3.4 PREBEIRil5ITES

~

3.4.1 The Al eserator shall be stationed in the Aust11ary Butiding prepared to vent the ftlT if nettfied by the Centrol Room.

PA 94 267 TP.1 Rev. O Page 2 9

t ij. I m m.i10NS 4.1 Il at any time during the performance of this test indicattens of increased RCS leakage are present, IlER immediately terminate the test and went the MUT if necessary to return WT pressure to within the limits of OP-1038 Curve 3.

4.2 Raise IRIT level to 86* per OP-402 Section 4.5.

4.3 Raise W T pressere to the limit of OP 1038 Curve's per OP-402 Section 4.1 .

4.4 Allow the system to stabilize for apprestaately 30 minutes.

4.5 Lower ItJT 1evel to 55" per OP-402 Section 4.4.

1 4.6 Raise MUT level to normal operating level per'0P-402 Section 4.5.

4.7 3 pressute is above the limit of OP 1038 Curve 8, vent.the WT per OP-402 Section 4.19. i l

4.8 Collect RED 45 data en computer points I-35g (MUT Level) and K-401 (IRIT Pressure) on a one minute frequency for the period of the test.

4.g Utilizing Excel, prepare an "X Y" plot of the collected data to evaluate acceptability.

i 5.e FetteW-UP ACTIONE 5.1 CONTillGEllCIES l

1 5.1 jI the Acceptance Criteria of this procedure is not met, i

.M a Probles Report should be generated. t l

l PR 94-267 TP. I Rev. O Page 3

ll l

i I

\

} To: DTIELDS

! Frost: GEALMON i

Subject:

Journal 8/18 i Date: 09 18-94 Tine: 4:21p l

1 To: . JOURNAL Bruce will be coming to discuss the events over the last week in reference to the part pressure curve. Ever of.what happened, so to dispell any rumors,yone knows somedetails.

details here are some First, A- shift is not "in-trouble". We, as a department, were I

i brought into the line-light by an ' action to raise attention of a l potential safety issue. 'me methods to raise this attention brought l up a philisophical issue we must all be clear of the expectations.  !

When I say "all", I as "m == of all of CE-3 management and I workers, not just operations. The philosophy is in how we deal

{ with operating curves and limits. First, when given an operating f curve, we will comply with it at all times. Second, we must give i

j to operations curves we CAN comply with and when complied with asstre the safety of the plant. Without both of these, we, as a 3 plant, let ourselves down in attempting to comply with our Code of i Ethics. The safety of the plant is utmost, ur. fortunately drastic

! measures are felt required to bring safety subjects to a head.

l Also unfortunately, in bringing the IEIT issue to a head, we passed i

l into an unknown region of compliance, into clearly unacceptable i

regions of a curve which was alread/ thought to be non-conservative. The questioning attitude we ask you to exhibit was excellent. The thought processes and safety culture for this issue' i were excellent. The controls put on the evolution were excellent. c l

i The . philosophy of not accepting an != m iste, unsatisfactory answer was excellent. The data obtained was excellent. The one i place we felt we could have done better was to have a pre-approved j procedure, one that without a shadow of a doubt, showed no t unreviewed safety questions for crossing over the curve. I feel

! our management team could have done better if I more agressively pursued the completion of the issue. Whether it be my background

- that I thought I fully understood the issue, I do not know. I do know the three days 'I spent with the subject, almost non-stop, tausta me a lot about the issue I did not.know and should have. In closinc,. Bruce will discuss. with each of us his expectations on operating conservatively. Blesse do not take offense by surmising that he e M *= you are not. By clearly stating expectations, it will open up discussions where, once again, we will learn more about nuclear power. The end did not justify the means, even though in the big picture, it will bring the issue to closure. I encourage you to talk to Dave or Rob about the issue. 'msy both handled the qu', scions extremely well and professional. Bruce commented on this and felt their attitude contributed a lot to our learning process and brought the issue to the right people the quickest.

. ~. a >- 4.r..~. - <= . . ,

f ,

Plorlds INTEROFFICE CORRESPONDENCE Power wpn ga 9 8 "

• 888 j

sus as usam l

4 l sus.nen CrystalRherUnk3 PR M 4 Mf,M UT4 Easel /3rtugasPremme v 1

CAP Ema M438%3, Duipt Beds Emma Dsesemissdss ,

l

PBm erNeff s

tak E. E. Yest ama IGovember M,1994 l -

uran 4mp

! The papass of his soc h a dessuna es Asign huis kuus mekades tr es utgest j pahism. 'as mahades esasidued het es 3g33p etes esimbs msw and apansism se es i IstatAumuhmus is poemius CF 135 Oms Aashans D. 'ns ashese sens is CD-l 15s was desubsped to passide amr pamus4smi Emin wthh amme Ep1 pump hopig l

dudag es susr needens pbsen af a Less amak tmCA (IXOCA) anddma. CBS has j huquandy spumed as es sans and, is ens immasa, opused is es massepshis asian eres i sens.

i Paliniasy malysm hsw sous em es amen eune h ase.esannahe due to homment ammapenas k the ensubstan which developed to sens. The emma supe wE he lansed leise es ehdag amma. ommedy, opemens k spumens to shes as er base as

adminisesshe sans whisk is psmEst as sad 1.5 75N ian des es @lWB suns. At tis tims them hmasomshis mammes esttis mspensingBakwlRemuelapsesass sans wbss the odoubliams aus esepises. 'as basis er deskung that opendse en er a es he of the 05135 sans emmenses synelas suedde the daign basis is as tBows 4

! 'ns want as LEEACA assidst smalysed Aur CBS tess a esse emellas and asumimment 4 lameday sandpointis a emid les husak. Opusies la es unsesspubis agles of to @l5B eaw as the sont of a IXOCA would assak in desmes to to MR pumps ese o hpdessus l

i essaismaat tus the malump mak. Epils act nedsDad as se ensamist symma ist amo aoebag i k these IEDCA analyms (see Aashuset 3. Esmover, ens unique IDCA desenhg af

! genias disandon is a pondstad husk is a een Seed Ens. Elass LPI kisms is sin momer l vessel thnagh the esas $ sed lissa, any assung wear tes L7t in the emis emmishg en tsuk weeld not assit es vousl. Cassidsdag a singis tihes :its power asses is es eher usin, as LF1 would be enthhis tr esse soaBeg sad Ep! weeld he used to midges this samt is es

short tsa (sms Antshasst S . emnerym tout RAW.15154, A4 thuds Anshuir 4(Owe Messes IAss BruskArB&W's 230s.afW, Ausmak Vier thhs MenW. .

. Abhough to een Sood Ems asudes 'line loses which Esdt the hemk she as 0.44 SP .

(esmidend as haussenes hasak she, as hisadoes sus er tes 14CA la mais ensuch to l psumpt systems to aspeed as tbsy would is a ImacA. Thesstus, susdes piphs hand lemus would be esapsahis e toss far a ~sismis IXOCA. The samlas piping head hans am a added puunsour la dessnaising es hidal semelons deemed by the surve. This, la um, i

I

)

1 i

l 1

i b

3 . .. l

)

i unims es Op !M4 sene a daden Emdt der his esem hemmen k asemos puusten of epipment asuded O adtpass to auss, i

In seashales, speedse sa er e es ist af die 51033 asus a es samt af a LELDCA er i

sus Assiliss 1DCA weeld tsus mated in EFIpump damags. TMs k asaddeed armedes i sumids as endes hemis at to pint per 10Crase,72(b)(1)00(& Fasess asema as a ass-l 44e0 ryou aquis e -

I

. i i

i s

// i

\

4 Vf __ -

i lamier assiseming nedys.-

\ -

i am 3.R.Musada P.L ttmessy L 1 cessa C. E i LE

! J. W. Caughou

! J. E I. ass F. E h L L Belmr G. E. Batman L L WEssa

• l P. V. Plumlag

autumsgar j Pas j nomsnis vessem e nt

~

i I

l .

I I

l l

l O

9

yyy

- ._ __.. _ (.

mst-*m J ef I .

i MAXIMUM MAKEUPTANK OVERPRES8URE

e .

.i .

o m

e

/

3m i

/

i .

1 .

A m

l?

, i ,

3 -

1 . -

- MAex*re trun.nmer m m -

! 1 - -

., a

/ y/

tm l ,

f f

• / /

\ . . /

.. /

/

I 10 r" s / .

WEMAR. EMMN p - '

! n=*

oisoissammewmommanmmmass insomnwamerinehmed Iridicatt Walk M %.)

i 1

4 l

oPlas nn. p., n

)

i a

i l to get soro consiCtent an how wa 1chel equipment. Input from the ch:p3 j will be necessary but we will take the lead in ctandardizing guidanos cnd i - looks for plant labels. It is to our advantage to ensure labels in the t

plant (including the control room) are the best we can get them. The i remotor building offers a unique challenge with the radiant heat. The

! waterfront and harm have weather to contend with. All of these concerns i need to be pulled together to make a good program. please get back to me

! if you are interested in owning this program.

.............................................e**.........................c i . .........

1 The NUT issue has blossened into an arena of enferoament. I will be glad

! to speak to any questions about this issue. suffice it to say that Dave, i Rob, Jim, and Jack responded to questions from the NRC Office of j Investigation in a very professional and osmpista manner. I sat in on all

! of the interviews and was very pleased with their. responses and the'NaC

questions. The overriding issue remains violating the. operating ourve and

! the raamaam W ad it. Imts all ensure we understand the operating i envelopes we are +==kad to operated this plant in and bring in the proper 4 resources to help us de our jobs if any shallenges oosur.

...............................................................ee..... . . ec I will be going to Atlanta next week to talk to the Kerry Iandis, my counterpart in the region NRC office. No is very interested in CR-3 and hopefully we will have a good dinoussion on the event s of the past year.

.; ..............................ee...... eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeet Good Luck and Thanks for your hard work. Nope your holiday plans go smoothly and safely.

Greg d

= _ - _ - - . -..

b '

)

i - _

l I

j i

I

< ~

FLORIDA i POWER

, .g 4

Ils!ill" 4 IIr. Steuert theater -

RegtenalAdelaistrater.Restes!!

U.a. Nusteer begolatory Comisstes 101 Marietta Street N.W., Seite 8900 Atlants. Georgia $3333

Reference:

A. W to FPC 1etter. 3R1194 01. dated November 4. 1996

8. PPC to set letser. 3F1896.se, secas December 8, tes6

$shject: Iggyantved f tan ca.fr.01. Nakaan Tank nament1an bear ler. Ihmeter:

This letter auselements ser letter of Deember 1.1998 (Referenes 81 tr prortdta'g addittamal tefomsttee free ser sentiestag revies of enresolved ites 96 33 81, estous tant (IER) operetten. This addittees) isformatten includes FPC's serresttee actless to resolve the teses and farther disciplinary action taken egetast the roepenstle Shift Supervleer and Assistent atft Superviser.

As you kasu this metter has been the sehject of es investitettet hF the NAC's Office of Investigattens (0!). The fosas of that levestigattes, as we useerstand it, is ahether the acttens of en FPC eserettag shift et Crystal River unit 3. ta esaducting as taaetherited evoletten en the IGN es 8estamber 5.1994.'sometitutes a utilft1 violette of precedural reestreassts. FFC tates say IEC concern la this repaN sorteusly. We have perefore rettened the events of Septemaer 5. 1994, the perspective of shether a villfts) violettee esserted.

As eslained in more detail holes. FFC in no say tendense the actless of the Shift Seerviser and Ass 1 stent Shtft Superwtear. Fpt has taken approprtsta disciplinary attica with resaN to the reopensitte taetviensis, amat me heltese to he asseusta in 1T ever:11 performones reeeres.ght Basedof all seentheour stremastenses retten, hemover taetustag their me e est helleve tant the todtvidents' estions reos to the level e/ eeltherete stoconduct. They were estivated by a destre to attain esta to support a legitteste tscheital censors with the voitetty of the sun operettag serve.

Moreover, they did met understand at the ties that the curve reflected easten hasts lletts.

amanumassemereensaar temutMMstasseem.aWEEansLAesemus n . m a i m n . w w u u ,,. m . m .

- mmefd Qr

kl. S. liesleer RegelItery Caetasten 3F9895 13 Pope !

Under these stremetenses. FM done est helleve that ser purpose maand be served tr K esfemsenet esties agatest the tatteteesla. We e m alas consemed with ths acesses that seat se esties might osed. FR esmagemaet has earned hard to testill a esset attttede seeng the earteferes at Crystal River. Further settes egetest 1 invenes seald have the estatended effect of disseuregteg others perestag.1egittante sensores.

, addittamat Carreetion settaen la addttles to the stees dessetted to ser Desenter S.19M 1etter. FPC has tahee the followtes correettve esttees:

1. The Shift Superviser and the Aeststemt Shtft Onporviser were removed free lleensed dettes and reasstened etthis the Operettees Osserteest. Their correst esties tess1ve preenderes deoelopment and other operettesal seppert areas edges their SM escartages is helpfel. FFC has es latest to place the tedividuals hoek en shift at tkts ties. FFC will esseelt wits the K before reterslag either testettual to ehtft dettes.

2. to a' FPC has established comprehensive fashion, a detailed fellestestatsal the semples se esttee ples assestate isones le resolved with esteep tank eserattes and the seter eterage teekfreester betidtag soap level. The esttee pies tes edge further volteattee of 8p 1888; Cerve 8. to eenere the correst operettag regtes and alare values.

If ear FPC 11seemed perseses) had st11ft11; violated operettag preenderes.

FPC emeld est heettate to tehe owes stremper settee, inslottog disoberge of the perseas resposettle. Oer forther rettes of this setter has conftrend eer conviettee that although the Shift's esttees la esapestieg as eseethertsed evolettee were emesseptable and unrrested stress disstelles. the todtvideals did est sageps to deliberste etssendest.

Casteluntant Remarotea haltherste N4teasemet The facts servenedtog the seentherited evolettee se the IGN en soonertsed to l j eer Dessator 2.1994.1stter med the E's leepesttes. Aseert (asforsees A). I j he provies here a smeery of the fasters that reflest the state of sted of l the tedividuals tese)ved. If the W la suore of eRF evidense of stenpfel 3

intest, pisase settfy FPC se that we see take appropriate artise.

l

• 1e hrtef. en fler it.1994, dortag the refueltog estege at CR 3. the Shttt t Supervtser and Aseistemt Sktft Superetser more to sharge of the sporeting crew that eerfereed Servet11 esse Procedere (SP) 638, a fell flee test for the Nigh Pressere lesesties peeps and sheek v41ves. tertie perfereleg SP-430 the operatoes sheerves a settseable secrease to the M lovet, tedtesties a peasthis stest. meteep velve alone with sevitettee of I 14fP !C. This sonettles ~ues to probles Aspert (PR) 94 8849,

dated May it.1994. PR M-014e ested that. based as sosporteses er

' Operattees sorseasel of the esteal drop te M 1evel with the eastge leR .

overeressere serve to SP 1gst (Cerve 8), a serve plotted with the esteel  !

data potets troeded tenere the seesceptable restes of Cane 8. Pt M.

0149 further ested that this esserted even theagh the tettial IRR ever-

' . pressors level dertes Sp 430 was holes the easteus alleensla pressere ser

) Cerve 8.

c4 -

.m: mers - ru m:ve es. as ..u i

1 l

l l

. .- u. s. mai.er asesi.Yor, camini.

87808 18 l

%8 l

i e The earrective esttens for pt M.Olet included evolesttag the IIR 1, vel drop that sneerred te essentas unether Cerve 8 mes esseMahle.* The avoisetten peNermed g11sesSystes tapiesertag [Nesleer Plant Tashn Soport]sessiedades The decreastag stense hensees the pletted serve and OP 1938, terve 8 espears to be estely ese to the feet that both serves are answergtes se aere seig. esses se this avoimattaa, it dans not appear the piatted serve useld have enterer (sic) the menestehte resten of Cerve 8. la edettia seere is seatsrverfee hofft sese Cerve 8 to essere that teatremst error, for essepts. emeld not treste se essessive everpressere essetttes. [septestsaddes]

e These constestems more ales deseensted la a asserendes free Nesleer Most feehnical Seppert, dated Septemer t.1994. A seer of the senerendes uns prwrided to the Shift Seperrteer med Aseteteet Shtft Superviser, ftr the perpsee of determining stather they had eer additteest sensores er mesttees before the tsese uns sleses.

e The Shift Supervleer and Assisteet Stift Superviser have indicated that they did est have ensegh tefersatin to hees whether they still steeld he semesmed with the M everpressere/lesel servs. asseretegly, se Sentenhor S. App 6. tkesetssted the evelatige (deserteed to ret's letter of Osasseer t.1996) to gather addttteest data by sensertog the systes ressenes as IIff level mes lesered. to sorrytag est this evolettes, the Rift Segerviser med Assistaat SMft Seeretser have indicated that they felleged the agoraties precedure for tes fIl estes (OP 408) set tettially Dieged the erstem se the operstieg serve Itatt (i.e. act is violattes of the serve). 1her emmesly destees est to tehe the Itif level holes tte sleerly stated les level 1tatt of $$ tastes.

As the ftff level ses redeses, they toch dets se the systes respasse.

Sean after the evolettee hopes, the overpressere detfted tote the unacceptable operettee regte with respost to fEff level. The sees fatted to terstaste the evolutten at that potet er otherwise take secrepriate acttaa. The evoletten did, benever. lead to englatertag eenftstag that the operettag serve . ems teesserete and mamammaarvettve. . its eMft reported thte discrepancy to their management and tettisted a Prehles Report (Pt M 0 ESP).

The IEE's Enfereesset Peitty.18 C.F.R. part S. Appendix C. Sestim VI!!.

stktes that enforcement settens involvtag tedividuals are *stgetftseet paraennel esttens uhtch will he elessly teatre11ed and juditteesly applies."

Aseerding to sectise VI!! of the teforceanst pelley esforeement esttees aestest en instvisual are reserved for *(ejere eartees violettens. tasteding these involving the tetagrtty of as todtofdeel (e.g.. lytag to the IK).

Appiteatten of this peltsy,.is reflected la the fellestag cases teveletag enesteerteed esttens by opersters, shore the ige has teessed teateidual enforcement seesttensa 33 envid fame map. IA 94 08 (19ps) (IE prakthtted 500 free eagestag in 11esanes estivtttes for three years fellestag seitherste

' *88 . DEI "aEF1 - afW tr1J Martve es. Go 48

O i

l .*

11. 5. huslaar hope 1%ry Castasten )

3F0595 13 i Page 4 l

- ~ n t. ni r - _ ,,ga n .

esser se of sisposttlened " costrel red

1. seinem ca- testdest)Cittes gsad Nesleer Poser Stetten.

31 (1993) ans

• 1 i

(lee teseed(orders sospensing lleense gef feet l 3

laitt handling 1). EASW 90and 33 (1991)1sg endif> the 11ceses for Good Cttles meere the )

tadtvidualenpased in eastes1sttees to sehe op for en error se the plaesseet L asI unian, ease.~ (Qyster Creet aselaer of a feel asseshly)p GenerellegStatten), . Sep; qas* * *8E

~' 1*-- EA W=83s i (1980) (IK isease vtalattens te both licensee een sentret fees aperater for 1

seerster's es1therste aestrusttee of elem tape desensating safety 1tett l vietattes): Mm *- (Feesh Betten Ateets reser Statten) p B.300 (1998) i (MC saettenes lleenees that fatore deltterste yleisstems of presseures l goveretne entry tete high rettetten areas possit te esfereenset attlee agatest Esth Itcensee and tedtvideals tegel .

j Nessver, in cases est lesslying deltterste totest te stelate regulattens or i presseures, the NAC hee feud fereaft1 diestplisery setten by the itseases to i he sufficient. and has est perseed enforessent esties apsiset tes 2  : V- " m=1an, -- tase. (Versest Vashee Butlear' poser Stetten),l dettes the seater centro tA roomtettviduals ter an (licenses removed from tavolved.

19eemoed jN -85a ches a tp

tesheteten esosed vtatatt a of reestring semessary sentaineest eartog moveemet of irrestated feel te esent feel pool)

i 1teessee L teeke ca- (N.B. coorster pay a Heensed Rettesesand plant):

ehtft(A M.13 (1984) foresen she fat t(ed to felles presseures for wertinattshrestattenarte).

l tztaamatfee feetars i

Althmesh the Shift's acttens te sendesties the ovelettee withest proper i i J were clearl j authoritfirst,

.e w . t= .v.f toeppropetate to. s toothe en. for esseamattes por,ese of g fastere hertog das steeld he i' 4

i to deterates whether a testettal esagers with Cerve 8 setstet. Tkts uns met I

a saae of a fallere to felles precedores estivated by teoreper er wrongful intest. Seeens. the Stift Segerviser an6 sessisteet Shift Superviser ett set j

realtas that the escrattag cerve was a deetp hosts itett. They believed ,

that the serve effectively estehlished aestatetrettve itetts, as had been i

{

J indicated to the Septemer t.19M. esserenses to shtch testeserieg. had l 1

centledad that Curve 8 ses 'asserete and reseenably sensarvettve.' 18 fact. l l emir after further design bests svelsatten did FPC castlede that operetten te the esasseptable regten of the serve seestttsted operetten setside the desten beats of the system (see Attachneet 4 ef ear Desseher 1,19M letter),

J

~

i i This mes aise est a taas ehere se operator knasteely fatted to felles se j

eelicit proseders) step or other reestresset, or disregarded the advise of i

ether centrol fees screennel that ble astion uns insenststeet ette l

cretadores. The Shift Segerviser and Aestettet Shift Superviser referred to DP 40s and estatatoed the left-1evel etthin the earnal eserettag Itetts of 85 l

to N inches. la asettles, they hose elee teditated that they more est eere ,

i j

that Curve referenced in 8 the(of OP 105) porttee o f OP 408 goverstag vesttog and hydrogen 1

(seetten 4.N). These fasters de not aussee the operstars' fatters te att GA 1

a tleely ammer ease the systes reopease ertfled tete the sneesentab)e reage i of Cerve 8. Mesover, this case steeld be 61stingetshed free one there sperators tsok afftrustive acties that vesisted as sep11 tit precedural

'egetrosent, j .m ' inert - arw wu usertoo as, se .su

. s*d j

i I

s k

1 .

} .'.

i i

- w w W. 8. Neslear Regulatory Csestssien i .

! 3F0886 13 Page 6 l

amma lie rettersta that FM in no ear sendonesSuas the actises of the sh es astten is centesttap as unsetherteed evoletten en the Ing.

tenem ettele with good operettag priestales and annapsment's e n ectatless.

llevertheless. FPC reestas esseerned wtm the asesttee tesect of forther a enforsommet saastions egetast the tedividus) Rift Superviser and Assistant htft Superviser in the aksease of deliberste etscensust and ta vise of FPC

{

discipitaarp estisms . Based.epen tie fasts of this case. thereftre.

@does act believe esforcement aatten acetast tim ladtvidual esersters ris j unrrested le C.F.R. g St.(s.a.I)for

. a violettee of the K's deltterste ate r Steserely.

l j P. M. esert Jr.

Senter flee President llusleer Operettens plerff met Desument Centeel Desk 1 Office of lavesti teos en !!

Chief. Breach 3, Inspector Senior Resident W Project Manager i

f i

O I

d O $ .

4

! i i

i 1

l l May 23, 1995 1 Mr. stewart Ebneter -

1 magional Administrator, Region II

) c. s. Nuclear Regulatory Commission i 101 Marietta street, N.w., suite 2900- -

j Atlanta, GA 30323

~

Reference:

A. FPC to MRC letter, 3F0995-13, dated May 5, 1995

, R. Manager, operations ' Journal' to operations j May 19, 1994 subjects 13nresolved Item 94-22-01, Makeup Tank operation near Mr. Enneter: s l The purpose of this letter is to provide my perspective of events i

j and proposals presented to you in FpC to MRC letter, dated May 5, 1995 (Reference A). My name is David A. Fields and I was the-i i

Nuclear shift supervisor on duty september 5, 1994 when an evolution was performed and data taken on the askaup tank

! operating characteristics. This information was presented in i Problem Report 94-0267, Mtrf Pressure Limit Curve Inadequate. I believe my input into the discussion is important. These are my comments,. they abould have no negative effect upon the individuals who worked for me on shift. ,

I

! I strongly disagree with the tone and content of FPC's letter of l j May 5. I do, however, believe that Dr. F. M. Beard is sinoere in l l his desire to limit or avoid NRC individual enforcement actions  ;

i against the operators who actually performed the evolution. Ist  !

! ne provida you with a very short suanary of why I felt authorised l l to partora the evolution and why I continue to believe that the j evolution was not a ' test'. This is not an effort to say that  ;

j ' mistakes' were not ande. I personally made judgnantal errors l which have made this situation nere damaging ts FPC and to the people who worked for me and are now under scrutiny. I will discuss these errors and I will take is11 responsibility for i them.

Nydrogen over pressure in the MtFF was first identified as a i

sanoern in IE Nation 88-23. FFC's response was slow and not without controversy due to the iact that calculations were being performed by partly,'

Gilbert Associates and comments beak and forth between site, St. Petersburg engineering, and S&I in Reading, PA.

were tima consuming and there was no clear pro $est annager. At that time, the MIFF was administreively limited, by 0F1033, Curve i 8, to 12psig. In April 1993 a nos curve 8 was introduced, a j variable curve based upon level in the MIFF. The inoraamad over pressure was desired to maximize dissolved hydrogen in the

) reactor coolant system. The nudear operator on my shift had j concerns at this time about tac ...dlity to amargener berate and I

( .

?

1 i

Jx.

]Pf 3 m v -sv w

l i ,

! l a

4 1 l l  !

! i i  !

! systen response in the event of a fire. The concerns were not >

l adequately addressed to his satistaction. During Refual 9, my l 1 shift performed SP-630, MPI Full Flow Test, and serious problems  !

l were encountered. The primary cause was determined to be  !

incomplete venting.of the pump's suction piping. A secondary concern identified and included in the Problem Report (PkS4 149) was the MUT pressure response during drawdown and the fact that it appeared'to be converving and would cross curve 8. Nearly all of the issues and resolutions aaaamening PR94-14e were routed

• , ths @ sy nuclear operator. Me had many, conversations, letters, and meetings over the various issues with engineeriig and annagement. Ne kept me up to date and intarmed, but I did not play an active role in the issue and resolution. At this name time, following restart from RF 9, there was increased attention to Res hydrogen concentration. Management demanded that operating shitta operate high in pressure and on the curve (curves). On september 2, Engineering sent a letter to the Director, Plant operations saying that PR94-149 issues had been resolved and curve 8 was conservative.. The' draft of this letter was brought to me my operations management and I was asked if there was anything'my shift wanted to de er respond to h======

the issue was going to be closed.

Our response was to perform the MUT drawdown evolution on septa =ber 5, 1994. We had calculation 190-0024 nov 5, which we could see had several incorrect assumptions. The calculation

-assumed that the hydrogen over pressure was an ' ideal' gas. The calculation,used pump suction pressures based upon an assumption that MUT to BMST swapover would essur at am RB level instead of a Bust level. We sould aise see that the antamiation an=a1=ead that with two MPI pumps operating off of a single sustion handme, that a column of water 2.27 feet above the pump suction weald result if a worst case Iack break size oomarred. We as a shift asked ourralves if we could legally perfarm the evolution. We reviewed 0P-402 and concluded that we had approved guidamos for lowering the MUT from 86" to 558 One member of the shift questioned whether we should drain below SS* and the ANSS said no,,that to go outside the procedure would ha a ' test'. No procedure Limits or Precautions would he violated. We reviewed the Annunciator Alars'i.s.w.e.h e'and determined that at the conclusion of the evolution, we would take actions to resters the tank and clear the alarm. Since we understood that the surve was a concern in the event of a large break Iack and with an equipment malfunction which resulted in two MPI pumps operating off of a single suction handae, we deciGad te. station the Auxiliary Building Operators in a position to be able to immediately vent the pressure from the Mgr. All of this was discussed at a pre-job hriefing. We fully expected to get the MUT Pressure High Alarm. We fully empeoted to drew a synton response tt.n was above curve 8. A third licensed oper*,:er was stationed in the control room to plot Nrf lovel and pressure on curve a so we would know exactly where we were et all ti:.es and could abort the evolution 12 something unexpected happer.sd.

1 e

l i

i 4 .

d .

k We did not violate Op-1035, Carve 8, is a system response curve.

this curve, we placed the MITF at 84" with hydrogen over pressure exactly on the curve. Using an approved, routine procedure we lowered the level to 55". Actual system response, the data we took, draw the real and correst Curve s. ue did not violate Curve e, we. proved that the engineering assumptions used to derive the published curve were incorrect. The smooth curve that we included.in Problem Report 94-0267 was in fact the real Curve

s. The published serve was inoerrect because it was based upon poor engineering assumptions.

The data we collected showed an error of 1.7 psig at 55". This j number is significant in that the 2.27 foot solumn assured in the

calculation was inadequate, and hydrogen entrainment of MPI pumps

! was truly a concern. Lost in this entire issue is the fast that j we were right. We had a safety ooneern which was not addressed j hy Management or Engineering and we proved that it was valid. On September 9 operation was administrative 1y restricted to 2.0 psig below Curve 8 and this was increased to 2.5 peig on September 14.

! Engineering reluctantly agreed that OP-1035, Curve 8 was act j corr.ct.

! For 35 minutes.on September 5, my shift operated in a region that later turned out to be outside of the Ep1 System Design Basis.

That is 35 minutes with extra people stationed and briefed for i I

actions to take in case of an emergency. 'All' other times that the MUT was operated 'ea' Curve 8, CR-3 was. metside of the same design basis. The plant operated outside of design basis for months in Cycle 9 and following restart free Refuel 9. Our efforts have had a direct positive affect upon the safe operation of the plant. Engineering is conducting a semplete evaluation of this complex issus. Two additional one hour design basis reports have been ande and a twenty six item MUT Action plan has been developed to validate system assumptions and calculations and ,

IOP iwC-e guidannen.

We initiated FR94-267 en September 7. Within days, the subject of ' design basis' and 'unauthorised test' was raised. Weather I nor anyone en my crew know that Curve 8 was a design basis limit.

No one in Management kasw it was a design basis or they would not have insisted, demanded that we operate on the curve. thiy.would Engineering provies us with a surve and an annunciator alarm that indicated a one hour NRC report was due, if understood that it was a design basis surve? In fast, Engineer osm1A not definitively decide Wt a design basis had been violated until November 16 when the ea hbar report was ands. It is aise not totally clear, in my opinion, whether a ' test' was estan11y partarmed. There is no clear policy or g.i'elines in CR=3 Administrative procedures of what consti- cs a test. I use as a

' rule of thumb' that 12 an ovelaties is cotered by an approved procedure, then a test precedure is not rsquired. If I theaght that the evolution that we performed on the MUT was a ' test', I

.ould not have authorised it's performa..--

1

(

• l I

i

• Dr. Board's.1sttar of May 5 presents Management's view that my shift performed an unauthorized test and violated the MPI system

{

design basis. He is asking you, the NRC, to not take individual

enforcement action against me and my assistant for our astiions.

I agree with his conclusion that neither of us willfully or knowingly violated any NRC requirement. I do not agree With his assessment that management has fully understood this issue and has taken appropriate disciplinary action against us. Their

! actions appear to as to be a reaction to what they 'think' the i NRC will do or what they 'think' the NRC expects management to do. Fellowing the Management Review Committee investigation of i ny shift's actions concerning the MUT evolution, my superviser, i the Manager Nuclear Plant operations, presented his view of the i

MUT evolution and lessons learned in his e-mail Journal to all i operations Department personnel (Beferanos 5, attaohed). This l was the most balanced and insightful discussian of the issue that i I bcVe seen. It is clear to as that management expectations had not been established prior to this event, but were being stated l

j in his journal.

Additional disciplinary actions were not taken against me er my assistant, until after the office of Investigation annonnood its

{ arrival on site to perform an investigation. I was working the

mid-night shift and was called at home and told not to some into

! work, that I was 'off shift'. The Assistaat Shift Superviser was i already off shift because he had been asked, prior to September l 5, to come off shift and be the Emergency operating Procedure

! (E07) Coordinator. Dr. Beard's letter mise states that the company has no intention of restoring either of us to Licensed duties. Es has told me that I will never go back on shift. Mis letter was the first time, however, that my assistant has ever been notified that he was administrative 1y removed from raannamd

, activities. After the Management Review Committee completed its investigation, there were no restriotions, and ne diMiyM y actions taken against us. Dr. Beard assured both of us personally that FFC had'iooked at the issue and felt that appropriate action had been taken and the issue was closed. Why has FPC changed their position? .No now information has surfaced.

Region II and o&I interest and. involvement seems to be the only reason these additional actions were taken.

e

_ . ~ . . _ . _ _ _ _ . _ _ _ _ _ _ _ _ . _ . _ _ _ . _ _ _ _ _ _

e i

1 1

! I do not understand the purpose of Dr. Seard8s letter. The i

• politics' between yPC and the NRC is completely foreign to me.

fvery licensed person on ry shift answered all O&I questions

• l i truthfully. I believe that their report will contain no 1 i

iniornation or allegation of any willful wrong doing. Any '

J actions which the office of Enforcement might take against as i

seem to pale in comparison to the actions which my company has l already taken. My career and reputation have been ruined over l this.Assue. It now appears that my assistant's career is also I being harmed by these unilateral actions. '

After Dr. Beard showed me his letter and I expressed my disappointaant, he assused me of not being willing to sooopt any responsibility. I as responsihte for overy thing that happens on ,

i j my shift. I autherised the NOT evolution and take all i responsibility for any errors which were made. No individuals '

reporting to me, including my assistant, should he considered for any individual enforcement. I was their supervisor and I made i the decisions... period.

The following discussion is my perspective of areas where.I could have made better decisions.

1. The Nuclear operator on my shift took the lead on the ISPF and the concerns the operations department had with operating with elevated pressures. yellowing SP-630, he worked tirelessly with engineering to help resolve the problems identified. I kasu that the O&I report will show his dedication and frustration with the issue. I should have gotten more gWay involved and helped his with his concerns and got his questions raised to the appropriate levels. I let his down. If I had been a more aggressive, effective superviser, this entire issue could have been avoided.

2. I as the one person on shift who should have recognised that curve 8 was more than an operating surve. Ne had the calculation, and we could see that the surve provided was the assumed system response to a NOT drewdown. This information should have alerted me that- no conservatism was included. Z .am a degreed engineer, I spent five years as an officer in the 05 Navy nuclear program, I worked as an engineer for Duke Power far three years, and I worked for FPC.as an engineer for seven years prior to hecoming a shift supervisor eight years age. I as the only person on shift who could have been expected to roooghise curve 8 as a design basis. If I had made that distinsties, them I would not have performed the evolution.

3. I did r = notify or consult with the 8th/ Shift Manager prior to tak:.r.; data on the NUT. That was a serious mistake in judgement. I mistakenly did not place the proper importance se

- the evoluticr. .e were going to perform. I de not think the STA would have t " ad any objections; however, that is an opinien and does not an . a my action.

- - =' ^^

e f

I l We were not a group of frustrated operators bent on proving a 1 point and taking inappropriate action to embarrass engineering or i annagement. I honestly believed'that I was taking appropriate j actions as a shift supervisor to address the canamens of W

operators within my department. The data was taken on the aid.

l night shift only because we were on the mid-night shift. "Our intent was only to take data we thought would keep a safety concern open.

I hope this letter has given you a different perspective en the MDF issue than the one provided'by Dr. Beard. I'helieve the O&I

report will come to the conclusion that there was no willful

! misconduct associated with any aspect of the event. I en disappointed that FPC has chosen to say our actions were taken because we had ' technical' concerns with Curve 8. I believe this wording trivialises a valid safety oonearn. I an disappointed that FPC has taken punitive actions inst as and my assistant

' over an issue as important and sensi ve as this. My assistant is the finest Assistant shift superviser that I have over werked with. Eis ability, integrity, and moral principles ce outstanding. It is wrong and inappropriate that he should auffer 4 because of my mistakes or the mistakes of our management. If

} nothing else comes of this discussion, I would like'to see tia j removed from any threat of individual enforcement.

! My position and FPC's position on this issue are clearly not in j agreement. I would like to be included in any further discussion j of this unresolved item. I welcome the opportunity to attend any j Enforcement Conference. I consider my discussion of this event j as a protected activity detined and governed by 10 CFR 50.7.

I f sincerely,

( k.

l David A. Fields Nuclear Shift Supervisor Nucinar Operations cc P. M. Board office of Investigation Chief, Branch 2, Region II Senior Resident Inspector NRR Project Manager v attachment l 1

l To: DFIELDS attachment, Reference 5 l

From: GIIAIJ1001 subject: Journal s/is Date: 09-18-94 Time: 4:21p j To: *JOURIEAL

e.....ee.........................................................

i

truce will be coming to discuss the events over the last week in j reference to the ' NUT pressure surve.' Everyone knows some details i.

of what happened, so to dispell any rumors, here are some details.

- First, A- shift is not "in-trouble". We, as a department l were brought into the line-light by an action to raise attention of a potential safety issue. The methods to raise this attention brought

up a philisophical issue we must all be clear of the expectations.

l When I say "all", I as talking of all of CR-3 management and workers, not just operations. The philosophy is in how we deal j

with operating curves and limits. First, when given an operating curve, we will comply with it at all times. Second, we must give to operations curves we CAN comply with and when complied with assure the safety of the plant. Without both of these, we, as a plant, let ourselves down in attempting to comply with our code of Ethics. The safety of the plant is utsest, unfortunately drastic measures are felt required to bring safety subjects to s head.

Also unfortunately, in bringing the NUT issue to a head, we passed into an unknown region of compliance, into clearly uneeseptable regions of a curve which was already thought to be non-conservative. The questioning attitude we.ask you to exhibit was excellent. The thought processes and safety culture for this issue were excellent. The controls put on the evolution were esse 11ent.

The philosophy of not mosepting an incomplete, unsatisfactory answer was excellent. The data obtained was anos11ent. ghe one place we felt we could have done better was to have a y.. -.._. J procedure, one that without a shadow of a doubt, showed no unreviewed safety questions for crossing over the curve. I feel our management team could have done better if I more agressively pursued the completion of the issue. Whether it be my background or that I thought I fully understood the issue, I do not know. I de know the three days I spent with the subject, alaest non-stop, taught as a lot about the issue I did not know and should have. In closing, truse will discuss with each of us his expectations on operating conservatively. Please do not take offense by seraising that he thinks you are not. By clearly stating expectations, it will open up discussions where, once again, we will learn more about nuclear power. The and did not justify the means, even though in the big picture, it will bring the issue to cleours. I encourage you to talk to Dave or Rob about the issue. They both handled the questions extremely well and professional. Bruce commented on this and felt their attitude contributed a lot to our learning proossa and brought the issue to tr.& right people the quickest.

e ..e e e e e e. .. .. . . ... ... . . . .. . . e e e e e e.... .e e e e e *

• e ... ......e.....*

' ' ' ' ~~ ' * ~ ~ .-. - -_ .___ - _ __ _ ____ -

i i

/m A,,

unne smes NUCLSAR REGULATORY CORBASSON Aenen a I ist MAmeTv4 syngr.m.W mWFsM i 1, ATLANTA, omnWA M l

l

\.h. July 7.1995

EA 95-125 i nr. P. M. Beard Jr.

j 5 enter Vice President. Nuclear Operations

ATTN

Manager. Nuclear Operations Licensing. NA11 Florida Pouer Corporation .

i

.15780 West Pouer Line Street -

l Crystal River. FL 34438-5708

• 1 i SMJECT: NRC 0FFICE OF INVESTISATIONE REPORT 2-94-035 l E IN5pECTION REPORT NO. 50-302/95-13 l

Dear Mr. Board:

l This refers to an investigation by the Nuclear Regulatory Commission (MC)

Office of Investigations (0!) cespleted on May 24. 1995, and inspections conducted by Mr. Itoss Butcher of this effice hetmeen September 5. 1994 and

' July 5, 1995 and documented in M C Inspectica Report No 50-302/95 13. This  !

special instaction reprt also susmarizes related findings discussed in NRC l Inspection leperts 50-102/94-12. H-01, 95-07, 95-08 and 95-08. Durtag these l revious, the MtC est:aned the facts and circumstances surrounding a september 5.19m event involving pressure control of the reactor coolant ,

system nahaue tank: and revisued the adequacy of destga centrol and corrective i actions that affected operability of emergency core cooling system pe gs. The subject inspection report and the synopsis of the O! investigatten are enclosed. At the conclusion of the inspection, the findings were discussed ,

with those members of your staff identified in the enciesed report.

Based on the results of our inspections and the 0! investigation, four apparent violations have been identified and are beine censidered for .

escalated enforcement action in accordance with.the " General Statement of  !

Policy and Precedure for NRC Enforcement Actions" (Enfc.w Policy).10 CFR Part 2. Appendix C. la addition, enforcement action is V'.ng censidered against the licensed operaters involved in the september 5, 1994 evest. 1 In regard to the first apparent violaties, on September 5.*1994, licensed operators planned and conducted an evoluttaa that allound the makeup tank pressure to exceed the acceptable operating region of OP-1038. Curve 8 for approxtmately 35 minutes. In addition, the operators delayed their response to the annunciator for the makamp taak overpressure condition uhtle they continued te drain the makeup tank, causing the tank overpressure te diverge further into the unacceptable ration of curve 8. These apparest intentional acts resulted in a violation of Technical Specificatten 5.5.1.1 which regoires tuplementation of precedures Al-500. Conduct of Operations: OP-402. Makeup and Purification System: OP-1038. Plast Operating Curves and AR-403. PSA N Annunciator Response. Had an Engineered Safeguards actuation occurred while in this condition, cavitation and subsequent inoperability of one of the high pressure injection pumps could have resulted.

0

+ - - - , . - ,

i

)

• i i

)

1 FM 2 l The 0! investigatten concluded that the shift swerviser, assistant shift i superviser, and two chief operators deliberately violated Crystal River i Nuclear Plant precedures. 10 CFR 50.5 (a). Deliberste Misconduct in part, i ershibits licensee seleynes from engaging in deliberata misconduct that

{ causes a Itcenses to he in violation of a condition of ray license issued by i the Commission. ~ Deliberate misconduct is defined by 10 CFR 50.5(c) as an j intentional act er emissten that the person kness constitutes a violatten of a regstrument, preca3ers instructies er policy of a licenses.

4 The record in this case reflects that licensed operators planned and conducted i an evoluties that they fully egosted would result in exceeding the makeup j tank overpressure limits specified in precedures The record aise reflects j that the Itceased operaters intentionally delayed tuplementatica of the reeuirements of the annunciator response procedure in order to gather j additional data en the overpressure condities. Apparently, the licensed operators involved were aware of the precedural requirements and intentionally violated the precedures. The NRC censiders these apparent intentions) acts te constitute an apparent violatten of 10 CFR 50.5(a). We are aise concerned that appropriate management eversight and centrol was not asertised to preclude inteattenal violatten of plant precedures. -

j The remaining apparent violattens involved failures to meet the requirements i of 10 CFR 50. Appendix 8. Criterion !!!, Design Castrel in that the destga

! hasis.was not carrectly translated inte drawings, precedures, and instructions, for: (1) operation of the makeup taak; (t) operaties of the manual suas over of the ECCS pumps' sucties from the berated water storage 1 tank to the reacter heilding sue s'and (3) maintaining adeguate investery in j the reacter hetiding sump to provide adequate not positive section head to one j

~

low pressure injectica pump with the high pressure injection (WI) pump ,

suction crosstie valve open and supplying tus operattag Mp1 pues. Tuo of l i these violattens aise involved apparent violattens of-10 CFR 50. Appendix 8.

! Criterien XVI. Carrective Action, in that once the design deficiencies were j identified they were not adequately corrected in a timely menner.

These apparent violations indicate significant womenme in ammagement control of the reviou and resolutten of significant candittens adverse to quality. Operater concerns ahest gas entratament in the high pressure

. injectten pues, ' identified in prehlem reports and cerravandamea h

! eperations and engineering, were not adequately resolved over a significant i period of time. Subsequent to identification of the design deficiencias in

! makeup tank overpressure lietts, magineering revious of the design assumptions l for the pressure / level operating curve of the askeup tank were not thorough.

The curve issued by engineering contained errors and was neo-conservative.

The revised curves issued by engineering aise costatmed errors and were non-conservettve. The curves permitted the plant to be operated outside the design basis. Indicattens of deficiencies in the design assuottens for

vartees tank levels in other safety related tanks alse vere not aggressively

pursued.

1 i .

No Notice of Violatten is presently being issued for these inspecties j findings. The musher and charactertratten of the apparest violattens j described in the enciesed inspection report any change as a rees1t of further j MC reytow.

I t

1 J l I I

7__._______

1 -

i i

! FM 3 i A closed predecisional enforcement conference to discuss these apparent

violations has been scheduled for July 25, 1995 at 10

00 a.m. in the E 's i Region 11 office in Atlanta, Georgia. In addition, as discussed with you en i July 5.1995, we are aise scheduling predecisional enforcement conferences j with each of the licensed operaters involved in the September 5.1994 event.

j The decisten to held conferences with you and the licensed operaters deles not

mean that the K has deterstned that the vtalattens have occurred or that enforcenant action will.be taken. The purposes of these conferences are to discuss the apparent violattens, their causes and safety significances to provide you the esportsatty to point out any errors ta ser inspectten report;

, and to provide an appertanity for you to present year corrective actions. la'

your discusstem, you should specifically address the concerns deserthed above

! with regard to management oversteht and contre) of licensed activities and i include any attigattog considerations not previously identified. In additten.

this is an opportunity for you to provide any infomatten concerning your

! perseectives en 1) the severity of the violattens. 2) the applicatten of the i facters that the NRC considers when it determines the sneemt of a civil l penalty that any be assessed in accordance with sect 1Ja VI.S.! ef the l Enforcement Policy, and 3)..any other application of the Enforcement Policy ta j this case, including the exercise of distretten in accordance with sectten .

vit.

l Please note that the NRC Enforcement Policy was revised and became effective with its publication in the Federal Aepfster (60 FR 34381. June 30,1995)

(Enclosure 3). Because the apparent violattens in this case were under review

! before the effective date of the revised Policy, the K will utilize 4

whichever versten of the Policy accrues to the benefit of the licensee.

j During the conference. you will be provided En opportantty to address any

applicatten of the revised Enforcement Policy to this case. You will ba j advised by separate centsponeence of the results of our deliberations en this

matter. No response regarding kne apparent vielsttens is required at this

ttee.

I l Pursuant to 10 CFR 2.790 of the NRC's " Rules of Practice", a copy of this J

1etter and its enclosures will be placed in the IIIC Public Decament Reem.

Should you have' any questions concerning this letter, please contact us.

i l .51 ly b  ;

i lis W. Mars tracter '

Division of Projects l

Decket No. 50 302 .'

License No. DPR-72 EA 95-126 i Encissures: 1. Synopsis of NRC Office Of

Investigattens Report 1-94-036 j 1. IRC Inspection Report 50-302/95-13

3. Revised Enforcement Policy .

1  ;

{ cc w/encis: (See next page)

! I i 1

1 h

! M 4 1

cc w/encis

{ Gary L. leidt, Vice President Bill Passetti j

4 Nuclear Production (SAtc) FPC Office of Radiation Contral i 15750 West Power Line Street Department of Health and

Crystal River, FL 34423-5708 Rehabilitative Servicas '

1317 Wisemoed Boulevard

8. J. Nickle, Director Ta11ah====a. FL 32399 0700 1 Nuclear Plant Operations (NAtc) '

j Florida Pomer Corporation Joe ftyers. Director 1 15750 West Power Line Street Division of Emergency Properadmeas

! Crystal River, FL 34428-4708 Departmast of Community Affairs j 1740 Canterview Drive L. C. Kelley, Director (NAt!) Tallahasses. FL 32399-1100

{ Nuclear Operations Site support. FPC 15760 West Power Line straat Chairman i Crystal River. FL 34428 5708 Beard of County Coastssioners

! Citrus County

{ Gerald A. Williams 110 N. Apopta Avenue i corporate Counsel laverness. FL 35250 f

' Florida Power Corporation -

MAC - A5A Robert B. Bersum

P. O. Box 14042 Bau Nuclear Technologies j St. Petersburg, FL 33733 1700 Rockville Pike. Suite 525 j Rackville,le 20052-1431 j Attorney General

Department of Legal Affairs The capitol

! Tallahassee, FL 32304 i

l 1

e q@

l 1

i i

j SYNDPS!$

I i

On Noveeper 29. 1994, the U.S. Nuclear-Regulatory C:umnission. Region !!.

] Office of Invest 1gattons initiated this investigation to estermine it'cartain 1 reactor noerators at Florton Power Corporation's Crystal River Nuclear Plant j (CRhP) :si1berately vto) stas CRNP proceeures my consucting an unauthdrited j evolution involving in the masaus' tant. the relationship netween the water inves versus pressure The investigation aisclosed that on' September 5.1994, the Operations mid

', sntft oeliberately 411ouse the makeup tant water levei to' decrease, within j allousele lietts, without se3nsting the make j entering a scenititte area of overpressure. usThis tantpronibited overpressure area to wasprevent eescribee by a CRNP proceeural decament which displayee a slot (curve) of paretsstBle tant leve) versus pressure response. The DurDese for the conduct l

! of this evolution by the operators was to attain actual tant level verses i cressure ressonse cata for comparison to the proceeurai curve. This curve sescrioen tne permisstole operating region.

j When tne overpressure entereo into the unaccessatie operating regten.

annunctators activates. and the operators knowingly continues to attain data

! without taking any action to alleviate the overpressure and a11eues the i

unacceptahle everpressure conettien to exist for 15 minutes. The data

! gatheren by the operators confinued that the precoeural curve differes from

! the actual curve.

i i Basse upon the evidence developed in this investigation. tt is caecluded that i the snift suoervisor assistant shift supervisor, and two chief operaters i caliberately violatte CRNP procedures by exceeding the allownele makeup tank 1 overaressure, overpressure. and delaying taking appropriate action to reeuce makeup tant l

Case No. 2-94-036 1 Enclosure 1 i

1. ""% uurten sutse

/  %- NUCLEAR REGULATORY COMRASSION

- '. namou a i

• usagTTA sTaggr, Nw,, gyrfg333

{ ATLANTA. GEonsa m S

g.

so..e

! Report No.: 50-302/95-13 Licensee: Florida Power Corporation

• 3201 34th Street. South - '

St. Peterstnery, FL 33733 f ,

. Decket No.: 50-302 Lt.conse No.: Opt-72 f ,

Facility Nome: Crystal River 3 l

i Inspection Conducted: September 5,19M through July 5,1995 Inspector: ((' d R. htcher, senior Res# dent inspect.or Date 61gnes Accompanying Inspectors:

T. Cooper, Resident Inspector L. Mellen, Reacter Inspector. RI!

R. Schin - Project Engineer, R11 Approved by: ff

1. u ndts, section Ghtef Date 4ignes Division of Reactor Projects.

SINOWUtY Scope:

This special inspection report documents inspections conducted by the NRC between September 5,1994 and July 5,1315 and susmarizes the related findings ,

of NRC Inspection Reports 50-302/94-12, 95-02, 95-07, 95-08 and 95 09. These inspections included reviews of: ~

The unauthorized evolution by licensed operators regarding the operation of the makeup tank outside of precedural operating ,imits,

- Operation of the makeup taak per approved operating instructions that resulted in operation outside the design basis of the makeup and purification system,

-h Mh- (

i i

j -

Emergency operating procedures that directed the manual swapover' of the Emergency Care Cooling System (ECCS) puuss suction from the Sorated l Water storage Tank (BW5T) to the reactor building same at a EMET water j level that could have resulted in the loss of the ECCS pumps, and l

Emergency Pressure Injection operating LPI) pump (precedures thatoperating to supply two directedHigh thePressure alignment of one j Injection pues that could have resulted in the loss of the only

{ operable LPI pump.

l Results: ,

i .

j Four apparent violations were identified:

i

Apparent violation 50-302/95-13-01: Deliberate appration of makeup tank outside the acceptable operating region. (paragrarA 2) l

! Apparent' violation 50 301/95-13-02. Examples 1. I and 3: Operating i curves for makeup tank outside design basis and failures to take j adeguate corrective actions for significant canditions adverse to e quality. (paragraphs 3 and 4) 1 Apparent violation 50 302/95-13-03 Examples 1.1 und 3: Inadeguata t

j design asseptions for borated water storage task swesever level:

l failure to take adequate corrective actions for stenificant conditions

adverse to quality; and failure to adeguately trrasiate design basis

requirements for available stored fire protection water inte procedures.

1 (paragraphs 5 and 6) 1 Apparent vietation 50-301/95-13-04: Inadequata not positive suction

. head to an Engineered 1sfeguards pump during accident conditions.

(paragraph 7) l l These issues were previously being followed up as URI 50-301/94-12-01

! Makeup tank operation outside the acceptable operating region, and i URI 50-302/95-08-04 Discrepancies in the tuplementation of the fire i service water tank level versus volume calculations. These unresolved j ' items are closed.

l 1 i l

e

REPORT DETAILS

1. Persons Contacted Licensee Employees

• P. Seard. Senior Vice President Nuclear Operations

• G. Becker. Manager-in-Training. 5tte Nuclear Engineering Services

• R. Davis. Manager. Nuclear Plant Maintenance

• P. Fleming. Senter Nuclear Licensing Engineer

• S. Gutheman. Nuclear Engineering Supervisor

• S. Mainen. Manager. Nuclear Plant Operations -

.- *t. Hickle. Director. Nuclear Plant Operations

• M. Jacobs. Corporate Ceemanications

• L. Kelly. Director. Nuclear operations Site Support

• J. Masada. Manager. Design Engineering

• P. McKae. Director. Qualtty Programs

• P. Tanguay. Director. Nuclear Engineering and Projects

't. Wil' ians. Legal Council

• K. Wilson. Manager Nuclear Licensing Other Itcensee ensloyees contacted included office, operations, engineering, maintenance, chemistry / radiation, and corporata personnel.

Nuclear Regulatory Cameission

• R. Butcher. Senior Resident Inspector

• T. Cooper. Resident Inspector

1. • C. Evans. Regional Council. RII

• • K. Landis. Chief. Reactor Projects Branch 2. Region II (RII) I

1. • E. Merschoff. Director. DRP, RII

1. • T. Peebles. Chief. Operator Licmasing Branch. RII

1. • L. Raghaven. Licensing Project Manager. Office of Nuclear Reactor Regulation (NRR) t*R. Schin. Project Engineer. RII

1. • L. Watson. Senior Enforcement specialist. RII

1. • G. West. Engineering Psychologist. Human Factors Branch NRR

• Attended exit interview

1. Participated in exit interview via telephone ,

Acronyms and initialisms used throughout this report am itsted in the last paragraph.

2. Fellowup of Apparent Operator Misconduct During September 5.1994 Event (92901)

On September 7.19M. a PR was issued regarding the adeemacy of the l MUT hydrogen everpressure curve in OP-1038, Plant Operating Curves.

PR 94-0257. MUT Pressure Limit curve Technical Basis Inadesmate. listad operator concerns regarding the engineering calcelation (190-0024.

Revision 5) that the operating limits curve was derived from. The need to maintain H overpressure as high as possible was to address RC5

i l

i i

4 4

2 i

! chemistry control. The industry guideline for Act He concentration is

- all ec/ES. At CR 3. the operators annually manipulate the M level to j attain the maximus N overpressure by lowering MUT 1evel, increasing He pressure to the maximum allound value per the curve and thansressing the M 1evel to increase H pressure. e RCS H, concentration will stabilize at at 25 cc/Kg. Operators were instracted by endat'to maintata hydrogen overpressure as high as possible due to hydrogen concastratten concerns. The N55/AN51 shift relief checklist dated July 11.1994. 'under Comments /Special . Instructions, directed the operators to keep 881T pressure as high as possible. Op-1038. Plant Operating curves. Curve 8. Maxtaus Makeup Task ~ Overpressure, pists felf

, alleuable evertressure (psig) versus Itif indicated water level (in.).

l The purpose of Op-1038 is to provide operational information for plant j startup, shutdeun, and other plant operations and evolutions. The

< operators are to use this curve to determine that the falT is in en t

acceptable operating region.

i Recently, the operators expressed concern that the engineering

! calculations regarding the acceptable He overpressure on the IEff were

{ non conservative. Operators had observed the felf pressure versus level j variance from the curve during-operation and were convinced that curve 8 was neither accurate nar conservative. This observation had been l j documented in PR 94-0149. IRN-80 Stuck Open, which resulted in an 8

unexpected drop in the MUT 1evel. On September 5. 1994 in order to verify actual MUT pressure versus level differed from that shoun en 3 curve 8. the operators adjusted Stri pressure to fall on the curve at a i IllT level of 88 inches (ftlT high level setpoint). The system was

j. allound to stabiliza and them the IRIT 1evel was bled doun to the les I l level setpoint of 55 inches. Operating procedure 0P-402. Makeep and i Purification System. paragraph 4.1.15 directs operators to entstata the PRIT level between 55 and 88 inches. As noted earlier, curve 8 in OP-1038 is then used to matatain the MUT allowable overpressure (in psig) versus indicated water level (in inches). When the operators

! allowed the MUT to bleed doun. the MUT pressure entered the unacceptable j l region and the difference between the curve and actual pressure -

increased throughout the entire level decrease. At a fElf lower level of

! 55 inches. ftfT pressure was approximately 1.7 psig above the curve.

j PR 94-0287 stated that the 1.7 psig equates to approxiestely 3.9 feet of j water. Calculation 190-0024 Revision 5. only ensures a column of water i

in the IRIT line 1.27 feet high and therefers the errer in curve 8 is i larger than the margin provided by the calculation.

j- TS 5.5.1.1 requires procedures be established, implemented, and i asintained covering activities as ..- L 1 in Regulatory Guide 1.33 I

Rev. 2. Appendix A. February 1978. Regulatory Guide 1.33. Appendix A recommends precedures for startup, operation, and shutdown of the i reacter coolant system. Procedure Al-500. Conduct of Operations.

l peregraph 4.3.1, Procedural compliance, states it is the duty of every i member of the Cr)stal River Plant work force to comply with precedores.

! Procedure OP-401. Makeup and Purification System, steps 4.19.8 and 4.19.9 required operators to refer to curve 8 of Op-1038 for anstaanfalT l ,

k l

l 2 __

I l

I 3

l%

overpressure. Procedure OP-1038 Plant operating curves. Curve 4.

! Maximum Makeup Tank Overpressure, defines the allouable maneus tank j pressure versus level operating region during operation. AR-403, j PSA H Annunciator Response, annunciator M PRESS HIGH/LOH. reenires operators to take action to reduce M pressure to within the limits of OP-1035. curve 8 when a valid alars is received.

However, on September 5,1994. operators allound the makeup tank pressure versus level to exceed the acceptable' operating region of OP-1888, curve 8. The operators then delayed actions to comply with AR-403 when the makeup tank high pressure alars annunciated. Esconding the acceptable retten of curve 8 of OP-1035 iad delaying the annunciater

. response is an apparent violation of the requirements of Technical j 5pecification 5.5.1.1 which requires implementation of plant procedures i AI-500. Conduct of Operationst OP-401. Makeup and Purification Systemt j OP-1038, Plant Operating Curvest and. AR-403, PSA H Annunciator

! Response.

An investigation of the apparent deliberate failure to follow plant i

precedures was conducted by 01. The investigation was completed on i May 24.'1995. The 01 investigation concluded that the shift superviser, i

assistant shift superviser, and two chief operators deliberately i violated Crystal River Nuclear Plant pescadures by exceeding the j

allouable makeup tank overpressure, and delaying taking appropriate action to reduca makeup tank overpressure. 10 CFR 50.5, (Deliberate misconduct), paragraph (a), in part, prohibits licensee employees from engagtag in deliberate misconduct that causes a licenses to be in violation of a condition of any license issued by the Ceemission.

Deliberate atscanduct is defined by 10 CFR 50.5(c) as an intentional act

or omission that the person kness' constitutes a violation of a f

reentrement, precedure. instruction or policy of a 1teenses. The j licensed operators involved apparently were aware of the precedural reeutrements and intentionally violated the precedures. The NRC considers these intentional acts to constitute an apparent violation of j

10 CFR 50.5(a). Management oversight and control of centrol reen g

j operations is discussed in paragraph 8. The apparent violation of plant procedures and 10 CFR 50.5 is identified as apparent violation 50-302/95-13-01. .

l 1

3. Reviou of Design Basis of M Operating Limits (g2903)

FSAR 5ection 6.1. ECCS. states in part that upon a valid actuation I signal, the Makeup and Purification System is automatically switched frem its norssi operating mode to the emergency operating made (High Pressure in,1ection) to deliver water from the SW5T into the reacter vessel. Unstated in the FERR is the design feature which receiros the hydrogen overpressure in the M be limited to prevent the M from being emptied which could allow hydrogen gas to enter the suction of the HPI pumps (which aise function as the makeup pumps in the Makeup and

Purification System) and result in damage to the pumps.

) -

1 i

i i

I i

l 1

l

j. 4-

% i j Based on the continuing operator concerns, licensee management initiated a comprehensive review of the igli H overpressure issue. An engineering evaluation was completed on November 16, 1994 that concluded that '

operation on er to the left of the OP-1035 curve at the onset of a'

LBLOCA er core flood line LOCA would have resulted in HP! pues ' damage.

1 This means that operation on curve 8 of OP-1038 resulted in operation i j outside the design basis of the plant.

i Subssemently, the licensee recognized that they normally operate with l .-

the two trains of HPI iselsted from each other en the section' side of ,

j the fRIPs. One train is aligned with its suction from the ftff (and also '

i the SW5T after an Es signal), while the other train section is nomally  !

j iselsted from the first train and is aligned to the SWIT after an i i Es signal through a separata ptpa from the OMET. In this case, escass '

j hydrogen pressure in the ftff during the enset of a LOCA could cause gas

! binding in one of the tus E5 selected HP! pumes. The other E5 selected l

HPI pumo would not be affected since its suction is aligned directly to

the SWST.

4 l However, one LOCA scenario, a postulated break in a core flood line in j conjunction with the potential gas binding of high pressure injection 1 pump due to a high overpressure in the makaup tank, could lead to the

{ 1ess of the safety function as esplained belos. In this scenarie LPI i cooling water enters ints the reactor vessel through the core flood i lines. Any cooling water from LPI in the train cantaining the break would not reach the vessel. A single failure in the other train, i.e.,

loss of the B emergency diesel germter, would result in less of the other train of LPI and loss of one train of HPI. The remaining train of HPI cooling water would then be reestred to mitigate this event. The com flood line nozzles have inserts which limit the break size to 0.44 senare feet which is considered an intermediate break size. The blowdown rate for this LOCA is raoid enough to prompt systems to respond i as they would in a large breas LOCA. Therefore, the ftff pressure limit i curve constitutes a design basis limit for this event because a high

! overpressure in the makeup tank could result in emptying the makeup tank

! prior to switchover to the BUST resulting in gas binding of the remaining HPI pump and loss of the safety function. Conseguently, a pipe break in the A core flood line concurrent with a LOOP, and a start  ;

failure of the B emergency diesel generator could result in a reacter i coolant system bloudoun and unavailability of both trains of LPI and one ,

train of HPI, as well as loss of the second train of HPI due to hydrogen i gas binding. It should be noted that with operator action the A LPI l pues could be manually aligned to the OM5T and used for injection of cooling water.

• i

! 10 CFR 50, Appendix 8. Criterion III Design Control, reentres that -

1 measures be established to assure that applicable regu_latory l reemirements and the Design Basis, as defined in 10 cm 50.1, j Definitions, and as specified in the license applicatten, are correctly translated inta specifications, drawings, procedures, and instrections.

3 The failure to translate the design basis to ensure proper operation of

)

i i

i 1

i i

l I

l

i j 5 I

the Makeup and Purification System such that the system is autsmatically l

] switched from its nomal operating mode to the emergency operating made l I (Mi Pressure In,1ection) and is capable of delivering water from the '

l into the reactor vessel is an apparent violation of 10 CFR 50 Appendix 8. Criterion III. This is identified as example 1 of apparent

, violation 50-302/95-13-02.

1

4. Review of Revised Operating Curves for fRIT (92903)

~ On September 9,1994, short term instruction (ITI)'94-019 was issued reesiring operators to matatain Inff pressure approximately 2 esig' below

the limit shown en OP-1038. cane 8. On September 14.1994. STI'96-021 was issued stating that engineertag had identified that the calculated j i error was greater than 2 psig and therefore, operators were directed to

) maintain OP-1038. 187T pressure curve 8. appreminately The semanent 2.5 psibbelow revision, iston 13 tothe limit shoun OP-1038, was en i issued omiJanuary 30, 1995, and contained two new curves to replace the existing curve 8 titled. Maximum Makeup Tank Overpressure.

!' On January 31. 1995 the licensee reported the operation outside their design basis to the intC. The licensee determined that the short ters.

j instructions issued on September 9 and September 14. 1994, and the new

pressure versus level operating curves for the Inff (curves 84 and SB in OP-1038. Plant Operating Cunes), issued on January 30. 1995, more non-conservative. The STIs and the new curves were based on design

assumotions that did not carrespond to current EDP reestrements.

i specifically. E0P-os. LOCA Ceoideun, step 3.35 and 3.53. and EDP-07

. Inedequate Core Coeling, step 3.9. require realigning ECCS pump section to the R8 sump and aligning for piggy back operation of the fRIPs when the BW5T reaches an indicated level .of s 5 feet. The calculatten used to support the STIs and generate the new OP-1038 curves assumed the swap I over to the R5 sump and piggy back alignment was completed prior to l reaching 5 feet in the BWST. The STIs issued on September 9 and Septemeer 14. 1994 and the curves issued on January 30,19g5 did not provide adequate margin to ensure that hydrogen entratament .in the high )

pressure .eakeup pumps was prevented during design basis events when the makeup tank was operated within the specified pressure and level limits; and, therefore. the interim curves allowed operation,of the makeup tank outside of the design basis 'of the plant.

The 11ceasee's immediate action was to issue a short term instruction with the following guidance te operations:

(1) Due to inconsistencies between the design assumptions used to generate OP-1038. Rev.13. Curves SA and 88, and top-8 LOCA Caeldown, maintain fRff pressure a minimum of 7 and a anzimum of 11 psig less than the limit given in OP-1038. Rev.13 Curves SA and 88.

(2) When transferring LPI section from the SW5T to the RB sump and establishing HPI suction from LPI (EOP-8. steps 3.35 and 3.53 and 4

EDP-7. Inadequate core Ceeltag, steps 3.9 and 3.10) valve

$ 1 1

y 6 l t

alignments need to be performed in a timely manner. While i performing the numbered dotati steps in oreer, both A and I train

! valves listed within the step should be stroked simultaneously.

j (3) Due to the location of the transmitter, indicated SWST 1evel will j not decrease below 2.33 feet. i

!- (4) The STI was not to be altered or rescinded without DNPO app mval.

l .-

1 10 CFR 50, Appendix 8. Criteries III. Desies Centrol, requires that measures be established to assure that applicable regulatory reestrements and the Design Basis, as defined in 10 CFR 50.1, ,

Definitions, and as specified ta the license application, are correctly  !

tresslated inte specifications, drawings, precedures, and instructions. I I F5AR Section 6.1, F.CCS, states in part that the upon a valid actuation signal, the Makeup and Purification System is automatically switched i j from its normal operating mode to the emergency operating mode (High Pressure Injection) to deliver water from the berated water storage tank i j into the reactor vessel. l 1

4

! OP 1038, Plant Operating Curves, Curve 8 Muzimum Makeup Tank Overpressure, defined operating limits for centrol of the reacter coolant system makeup tank pressure versus level. Operators were instructed by management to maintain the makeup tank pressure versus level close to the limit defined by Curve 8 to maximize hydrogen i overpressure.

l l

10 CFR 50, Appendix B, Criterion XVI Corrective Action, states, in i part, that measures shall be established to assure that conditions

! adverse to quality , such as nonconfomances, are promptly identiftet l and corrected. In the case of significant conditions adverse to i quality, measures shall assure that the cause of the condition is

! determiced and corrective action taken to preclude repetition.

1 ~

I .As discussed above, the design basis for the ECCS was not correctly translated into drawings, procedures, and instructions for the emergency l operating mode (High Pressure Injection) of.the Makeup and Purification j system. The STis issued on September 9 and September 14, 1994 and

Revision 13 to OP-1035. Plant Operating Curves, uhich replaced Curve 8, i Maximme Makeup Tank Overpressure, with new Curves SA and St. Maximum

Makeup Tank Operating Pressure Versus Level on January 30, 1995 all

! allound operation outside the plant design basts.. An Engineered l Safspeards actuation while operettag on the new curves, could have i resu ted in cavitation and. subsequent inoperahtlity of at least one of the high pressure injection pumps and, for a given scenario as described i

in paragraph 3, without operater intervention, could have regelted in 4 the loss of all MPI pumps. Therefers, the corrective actions for the previously identified problem with the curves were inadequate to prevent operation outside of the design basis. Failure to meet the requirements of 10 CFR 50 Appendix 5. Criterien III and XYZ for the interim curves

]

I 1

1

i J

~

, 7 and the curves issued in Revision 13 to 0P-1038 are identified as asamples 2 and 3 of apparent violation 50 302/95-13-02.

s

5. Review of Design Assumptions for Sarated Water storage Tank Suepower j Level (92903) j On Febreary 2.1995, the residents euestioned the licensee regarding j ' . their design assumettens for the SW5T 1svel for suas over from the SW5T a

to the RB sump. The inspectore identified the following additional

). conce ms:

The indicated BWST in the centrol room reeds from 0 to 50 feet.

i The accuracy of the SWST 1evel instrumentation contains more than j 0.5 feet of uncertainty.

j -

The top of the 14 inch line from the IW57 to the E5 pumps is

located at approximately the tuo feet level in the SW57.

3 j -

Vertaxing of the berated water in the SW5T was not accounted for.

The SWIT contains a vertex breaker, homover, the Itcensee has been unable to locate any of the calculattens for this device. The- -

, licensee's preliminary calculattens indicate that even with the i

vertex breaker, vertaxing would be likely to occur between 3.5 and 4 feet of SW5T 1evel.

i -

Operaters could perform the SWIT to RB sump transfer function in accertance with procedures at anyttee the SW5T 1evel became less then 5 feet.

At the maximme flev ratas, the drew doun of the SW57 approachss one feet per minute.

On February 1, 1995 the licensee made a report to the NRC regarding their finding that manual susp over of the Es pups from the BWIT to the RB sump any not occur in time to prevent vertexing in the BW5T.

Preliminary calculations indicated that a minimum of 4 feet is reentred

, in the SW57 to prevent vertexteg and therefore ensure adequate NPSN.

Engineering calculatten M95-0005 dated February 6,1W concluded that vertexing in the BW5T has the potential to begin at 5' l' in the tank.

Taking level instrument error and calibrettaa tolerances inte consideration increases the level required by l' 2* and the suspever to the RB sump suction should be complete before as indicated level of 7 feet is reached in the SW5T.

The licensee's analysts indicated that after disposittening these considerations, it was acceptable to ratse the BWET same ever to the 15 foot level. The E0Ps have been revised to reflect that the suspner should occur starting at 15 feet and be completed by 7 feet.

i

! i

' l l

\

S ,

The swapover from the BWST to the R8 sump is a manual operation and has I j beu designated over the years as follows: l (1) 6/75 to 5/79 2.5 feet in the BWIT (2) 5/79 to 6/83 3 feet 9 inches in the SWIT l (3) 6/83 to 6/90 2.5 feet in the SW5T ,

i l (4) 5/90 to 4/93 1.1 feet in the RB f (5) 4/93 to 2/95 s 5' feet in the BWIT  !

l The licensee is conducting a more rigorous analysis of the swepover level. There are vertex and NPSH considerations for the pups taking suction from the reactor building sums. Additionally, there are

Trisodium phosphate baskets for pH control in the RB lower basement i 1 areas. The calculations for their placement assume a certais velisme of

! water in the sump and a certata flew rate. Both the volume and flow f rate of SW5T water into the sump area will change if the suas over level in the SW5T changes. These changes will result in a different rate of .

l TMsodium Phosphate dissolution.

l i

i 10 CFR 50 Appendix B, CHteHon III Desipa Centrol, roomtres that l measures be established to assure that app :icable latory

reestrements and the design basis, as defined in 10 FR 50.2, l Definitions, and as specified in the license application, are correctly

translated into specifications, drawings, procedures, and tastractions.

l i FSAR Section 6.1.2.1.2. LPI, states that when the IW5T 1evel reaches an j elevation of 5 feet, the operator will take action to open the LPI System suction valves from the RB emerTency sume, permitting recirculation of the spilled -sactor coolant and injected water from the

! RB sump.

1 l E0P E0P-08, LOCA Cooldown, revision 1. steps 3.35 and 3.63, and E0P-07, l Inadeguate core Coeling, step 3.9, revision 1, require realigning the

LPI pump suction from the BW5T to'the R5 sump and aligning for piggyback oporttien of the Make up Pumps'(ItaPs) when the W5T reaches an indicated level of less than or equal to five feet.

As discussed above, the design basis for the ECCS was not correctly translated into drawings, procedures, and instructions for operatise of the manual swap over of the ECCS pumps section from the IW5T to the RB samp in that on February b 1995 an engineering evaluatten identified that initiation of suas over of ECCS pump section from de IMIT to the RB suas should be comp)eted prior to an indicated level of seven feet to prevent vertering and resultant disabling of the ECC5 pas. State 1975 1 (escept for the time period of June 1990 through ApH11993) plant

} precedures have required the manual swepover frem the SW5T to the R5 i sump at a level of five feet or less in the SW5T whte is insufficient j to assure that all of the ECCS pumps would not be damaged by vertering.

i i

. , , . - - - , -- - _ , , ~ ,m, e,

~ - . _ . _ - . - - _ _ - . - - - - - - - - - . - _ . - _ _ _ _ _ _ _ _ -

I t

This is an apparent violation of 10 CFR 50, Appendix B, Criterion III.

Design Contral and is identified as example 1 of apparent violation 50-301/95 13-03.

6. Follow-up of Licensee Event Report 92-003.. Personnel Error and Lack of ,

Techr. cal Review in Past Proceeurs Revision Process Leads to incorrect '

Proceeures Resulting in Violation of Technical Specification and Design Basis (92903)

On August 1,1991, the licensee identified a potential.. problem ,

cancerning a calculation to support the basis for asssNas EDS fuel' oil storage volumes were maintained as regstred by T5 and design basis. On ADH1 15,1991, the licensee deteretned that two procedures for documenting the volume of fuel stared in the EDS fuel oil storage tank had been erroneously revised, resulting in one occurrence of failure ta meet the statsum volume of fuel assumed in the design basis and 14 occurrences of failure to meet T5 reestrements for minimum fuel volume. The cause of the probles was attributed to a failure to recognize that due to suction locations, same of the volume in the storage tanks would be unusable and could not be taken credit for.

As part of the corrective actions, the licensee stated that the relationship of-suction point to tank level for other tanks havtag a TS required minimum volume would be veHfied. A corrective c: tion plan.

which prieNtized the vaNous tanks, was developed. DH O nally, the various tanks were scheduled to be completed by December ISg4 vith the highest priority tanks being scheduled for completion by Decer 1993.

The priority 1 tanks included the CST, the BW5T the EGDG fuel til day

, tanks, the SASTs, and the CFTs.

On September 19,19H. the corrective action plan was revised, as none of the steps for recalculating the volumes had been completed. The new completion dates called for the project to be completed by ApH11997, and the priority 1 tanks were scheduled to be completed by March 1995. i This delay has an impact on the concerns on the SW5T discussed in i previous paragraphs. The calculation of the SW5T volume concerns has a direct impact on the BWST issues. This issue was a previous opportunity  !

for the licensee to identify and correct the problems with BMIT section.  !

A recent lutC review of the fire water storage tanks FST-1A and FIT-1B tank calculations, Mp3-0028, revealed a discrepant condition betmeen the '

FPP and the EDED reentraments. The FPP required that 345,000 gallons of water be contained in each fire water storage tank. The EDID reestred a minimum capacity of 300,000 pallons of water be available from each tank to the fire pumps. However; the capacity of the FITS is less then 345,000 gallons of Egghlg water in either tank. When full the tanks each contain apprestaately 318.000 gallons of MAggg water. The requirement in the FPP for each tank to contain 34a,000 callons does not appear to correspond to the design basis requirement of 100.000 gallons.

The licensee's volume calculations of the tanks concluded that when

1 i

i 10 1

{ 345,000 gallons of water are contained in the tanks, only approximately i

197,000 gallons of usahlt water are in the tanks. .

c$

l

, The surveillance procedure, $P-300, Operating Daily Surveillance Leg, requires that each tank be verified to contain greater than 35 feet of water by level indicators FS-1-LI and FS-t-LI. These indicators are

read on a 0 to 37 foot scale in the control room. The 37 foot level corresponds to apprestaately 315,500 sallons of 314A11 water. The 35 fact requirement specified in $P-360 correspenas to approriestely 295,000 gallens of gask13 water. The preesteral requirement does not appear to account for instrument tolerances, which betmeen the i

tremanitter and the indicator, are nearly 14 inches, This worse case candities, considering the instrument tolerance and a level of 35 foot i

in the tank, would enhy ensure a value of approximately 283,000 gallons 2 of ugablg water in the tank. This value is estside of the design basis.

The teensee has stated that they norus11y operate above the 35 foot

{ 1evel, since there is an alare eight inches above this level. The alam is calibrated to assure that approximately 301.000 gallonsofgggk}g i water are available in the tant. Housver, the leve) switch has as  !

allowable tolerance of four inches, meantag that the alare, set within the tolerance, could correspond to only approximately 199,000 gallons of l water, which is outside of the design basis.

h When made aware of the inspectors concerns, the operators inemased F5T levels to the maximum the tank can held, to assure that there was enough water available to guarantee that the design basis calculated i

minimusi requirements were met. ThisplacedtheFITgasklgwaterabove i the levels where concerns exist. Operators were infomeo of this issue i

by a note in the shift supervisors' log.

The Crystal River Facility Operating License No. DPR-72, paragraph

{ t.C.(9), Fire Protection, requires that fire protection measures be tuplemented. FSAR 5ection 9.8 states that the fire protectica program i has been formulated in accordance with specific fire protection j governing documents listed in F5AR Table .918. Table 9-18 includes the FPP. The FPP required that 345,000 gallons of water be sa ggingg in i

nach fire water storage tank. To imelement this requirement, the EDE i

required a minimum capacity of 300,000 gallonsofwaterbeartilabla from each tank to the fire pumps.

l l 10 CFR 50, Appendix 8. Criterion XVI Corrective Action, states, in j part, that measures shall be established to assure that conditions adverse ts quality , such as nonconfernances, are promptly identified and corrected. In the case of significant conditions adverse to quality, measures shall assure that the cause of the condities is determined and corrective acties taken to preclude repetitica. Failure

to toplement timely corrective acties to review potential significant i

conditions adverse te quality involving safety related tanks, including i the BW57 and FIT, is a violation of the requirements of 10 CFR 50, 1 Appendix 5 Criterion XVI and is identified as example 2 of apparent j violatica 50-302/95-13-03. -

i . .

3 1

1

) . .

i 1 4

l l 11 4

j 10 CFR 50. Appendix 8. Criterion 111. Design control, reestres that

measures be established to assure that applicable regulatory reoutrements and the design basis, as defined in 10 CFR 50.2  !

l' Definitions, and as specified in the license applicatten, are correctly l l translated inta specifications, drawings, procedures, and instructions. j Failure to translate the design basis requirements of the FIT into -

operating procedures is a violation of 10 CFR 50. Appendix 5. Criterion J i III and.is identified as example 3 of apparent violation of  :

50-301/95-13-03. This issue was previously follound up under

• URI'50-302/95-08-04. This unresolved item is now' closed. l

.7. Operation With One LP! Pump and Two MPI Pumps While in the Piggy Back l Mode (92901)  !

i 1

On March 22,1995, at 5:07 p.m. the licensee made a 10 CPR l 50.7t(b)(1)(11)(3) report regarding the finding of inadequate past LOCA j RB water inventory to support the current E0P receirament to align one

operating LPI pump with two operating HPI peps. During a follosus j engineering investinatten of previously identified prob ens involving i operation of the far outside of the desten basis (See LER 94-009.

i Personnel Errors in Determining MLIT Level / Hydrogen Pressors. SW5T .

Vertexing and RB Suas Level Parameters Result in Potential for Operation

Outside Design Basis, and URI 50-301/94-12-01 addressed in

{ Its 50-301/94ett. 95-01. and 95-07) the licensee identified a condition j specified in E0P-os. LOCA Coeldown, revision 2. steps 3.39 and 3.57, and E0P-07. Inadeouate Core Cooling, revision 1. step 3.10 that directs the i , operator to accomplish the fol lowing:

l 11 only 1 LPI pump is operating, lhtlLensure IRIP section cross tie j valves are opent i

!

• 9tlV-62 i

• ftlV-69 l This created a system alignment where one LPI pump could be subjected to  !

i supplying 2200 l

two HP! pumps,and gpa100 nominal to the RV 540 gpm recirculation SpaAto f1eu. the suction recently revised of each flow of calculation (M90-0011) shows that insufficient water inventory would exist in the RS to provide adequata NPSN to the single LPI pues at the i noted flow rates. This lineup could result in the Loss of the only j l operable LPI pump.

As immediate corrective action. the HPI puuss' section cross tie valves were caution tagged to the SSED. STI 95-0022 was issued on

s. March 22. 1995, to provide operators with addittenal guidance. The STI 1 is required reading for a111perators and it advised operators of the i reason for the revised calculation and to alert them to the tagging

order on the HPI suction cross tie valves. The STI alas alerted the i

operators that E0P-07 and E0P-08 were affected. The STI had a i 10 CFR 50.59 safety evaluatten. an independent review, a PRC review, and

] ONPO approval. The contre) copy of the affected E0Ps to the control j reon were marked with a red pen at the appropriate steps to restad the I

4

l 1

I l

1 12 operators that a STI affecting that step had been issued. The femal change to the EDPs is scheduled to be accomplished within ten days. The inspectors verified that the E0Ps had been marked as specified and that the STI was in the control room.  !

l 10 CFR 50. Appendix 8. Criterien !!! Design Control, requires that i measures be established to assure that applicable regulatory  !

reestrements and the design basis, as defined in 10 CFR 50.2.

.. Definitions. and as specified in the license application.' are correctly transisted in to specifications, drawings, precedures, and instrections.  ;

1 FIAR Section 6.1.1. which describes the design basis of the ECCS states, in part, that the ECC5 has been designed to perform its functions if a single active failure occurs and that one of the design functions of the ECC1 is to provide long tem cooling by recirculation of injection water from the reactor building sump to the core through LPI.

E0P-08, steps 3.39 and 3.57, and E0P-07.' step 3.10. direct the eserstors to open the HP! pumps' cross tie valves in the event only one LPI pump is available to supply suction when in the pigr'-back made of operation.

These procedural directions have existed since gril 8.1993. .

Neuever. enLineeHng calculation M90-00tl. revision 5. dated March 12. 1333 deterunned that during post LOCA operation there was inadequate inventory in the RB sump to provide adequate NPSM to a LPI pump, with the HPI pup suction crosstie valve open, supplying two operating HPI pumps. This. lineup could result in the loss of the only operable LPI pump. This is an apparent violation of 10 CFR 50 Appendix 8. CriteHon III. Design Control, and is identified as apparent violation 50-302/95-13-04.

B. Review of Management Oversight and Control of Licensed Activities In regard to the actions of the 11 censed operators on September 5.1994, the NRC is concerned that appropriate management oversight and contre) was not exercised to preclude the apparent intentional violatten of plant precedures. In the licensee's letter to the NRC dated May 5,1995, the licensee characterites the operator's concerns about ,

the makeup tank operating curve'as a legitimate technical concere, yet i the record reflects that engineeHag found the curve to be adequate and  !

proposed that the issue be closed. It is not clear that management was  !

properly involved in resolution of these diffeHng technical opinions. i In addition, operators did not seek approval of. management in conducting '

the evolution to evaluate response of the system. This raises questions as to the adecuacy of ceumunications between management and the licensed operating staff including whether management has clearly conveyed its expectations in regard to procedural adherence and the need to use established review mechanisms for plassed activities that are entside restine operation.

, As stated in the cover letter, these apparent violattens also indicate significant weaknesses in the management contre) of the revice and 1

j l

e l

l .

l l 1 -

l I3 l

I resolution of significant conditions adverse to quality. These j weaknesses include (1) the failure to adequately review operator

! concerns en gas entrainment in the high pressure injection pumps which I

had been identified in several problem reports and correspondence <

l between operations and engineering; (2) inadeonate engineering reviews )

of the design assumptions for the pressure / level operating curve of the i j makeup tank; and (3) failure to implement timely corrective actions for indications of deficiencias in the design assumptions for various safety-related tank levels. The root cause of these issues appears to

!. be a lack of management oversight of the review process. The IK has

proviessly expressed concerns with management eversight and comettment 1 to program implementation in meetings with licensee management en i j November 12. 1994 and March 1. 1995. j

, 9. Exit Interview I I

The inspection scope and findings were summarized on July 5,1995, with l

! these persons indicated in paragraph 1. The inspectors described the j areas inspected and discussed in detail the inspection results listed l j below. " Proprietary infomation is not contained in this report.

1 f IgggItanNumber h Descrintion and Refennce j 4

i

! EE! 95-13-01 open Deliberate operation of makaus tank l speration outside of acceptable operating j region. (paragraph 1)

EEI 95-13-02 Open Operating cunes for makeup task outside design basis and failure to take adeguate carrective actions. Three esamples.

(paragraphs 3 and 4)

, EE! 95-13-03 open Inadeguate design assumptions for berated j water storage tank swapover level.

inadeguate corrective actions, and i inadeguate stored fire protection water.

(paragraphs 5 and 6)

EE! 95-13-04 open -

Inadeguate Net positive section Head to an i Engineered safeguards pump during accident l conditions. (paragraph 7) '

URI 94-12-01 Closed Maksep tank operation outside the i acceptable operating region.

- (paragraphs 1, 3, and 4) l i URI 95-08-04 Closed Discrepancies in the implementation of the 1 fire service water tank level versus ,

j volume calculations. (paragraph 4) i 1

i 4

l

f

14 j IIRR 115.thdttr, Status Descrintion and Reference i LER 92-003 Open Personnel Error and Lack of Technical i Review in Past Precedure Revisten Process Leads to Incorrect Procedures Resulting in i Violation of Technical Spectficaften and Design Basis. (paragraph 6)

]

1

10. Acronyms and Abbreviattens -

1 AI - AMaistrative lastraction i AIS Auxiliary fluclear Operater i ANES - Assistant stuelear Shift Superviser l BSP - Retiding Spray Pump i B&W - Babcock & Wilces j Bil57 - Berated Water Storage Tank j CCHE - Centrol Complex Habitability Envelope-

- Cubic Feet per Minute

, CFM

! C0C - Certificata of Compliance

! CP - Compliance Procedure '

CROM - Centrol Red Drive Mechanism -

CREV5 - Centrol Ross Emergency Ventilatten System CVT - Constant Voltage Tressferner DCP - Decay Heat Closed Cycle Coeling Pump

DCV - Decay Heat Closed Cycle Cooling Valve Olr - Decay Heat Pues OHV - Decay Heat Valve DIIPO - Director lleclear Plant Operations ECCS - Energency Core Coeling System (s)

EDSFI - Electrical Distribution System Functional Inspection

! EEDG - Emergency Diesel Generater

! E0P - Emewy Operating Procedure

! E5F - Engineered Safeguards Feature

! ESAS - Engineered Safeguards Actuation System ,

3 F - Fahrenheit i FCil - Field Change listica .

FLUR - First Level Underveltage Relay FPC - Florida Power Corporation . l FSAR - Final Safety Analysis Report HEPA - High-efficiency Particulate Air HPI - High Pressure Injection IR - laspection Report i LCO - Limiting conditten for operation j LER - Licensee Event Report 4

LOCA - Less of Caelant Accident I LPI - Law Pressure Injection i MAR - Modification Approval Record

! MP - Maintenance Precedure

pelP - natsup Pump

ftff - Makeup Tank j ftpv - Makeup Valve J IICV - llen-cited Violatten i

1 f

a

}

}

l

I i

4 l 15 j NOTIS - Nuclear operations Tracking & Information System 110V - Notice of Violation l

itP5H - Not Positive section Head i

NPTS - Nuclear Plant Technical Support 185 5 - Nuclear Shift Supervisor

IRIREE - NRC technical report designation j 01

- IEIC Office of Investigations OP - Operating Procedure s -

g PM - Preventive Maintenance -

I pee ,

s

- parts per million j '

PR - Problem Report i PNC - Plant Review Camatttee i psi - pounds per square inch I

psig - pounds per sguare inch gauge QC - Quality Centroi

0A - Quality Assurance i RS - Reactor tu11 ding j RSS - Reactor teilding Spray RCA - Radiation Control Ares RCP - Reacter Caelant Pump RC5 - Reactor Ceolant System

• RPS - Reactor Protection System i RV - Reactor Vessel IRI - Raw Water l

RMP - Raw Water Pump RWV - Raw Water valve SCRA - Self contained Breathing Apparatus suit - Second Level undervoltage Relay SP - Surveillance Procedure '

SR - Surveillance Requirement SRP Staneard lleview Plan 55t2 - Shift Supervisor on Duty STI - Short Ters instruction SWP - Service Water Pump TDP - Training Department Procedure i i

TIS - Training Information System T5 - Technical Specification T5I - Technical Specification Interpretation .

URI - Unresolved item -

VIO - Violation WR - Nork Request 4

e i

l unessmes j

[ muci,sasnaamatoswootenesssoas l

1

! f * "aa"m"m",'e"manassiesse ""

i

\ ..... /

EA 95 126 w KML

,, , - .a l

p. 4/m nr. P. M. Seard Jr.

4 Senter Vice President. Neclear dperettens ACTX3N gggTg:

DY:

M M** ** f * ^ - l

! ATTN: Roma i

n.rida es.ger, iteclear tim Operettees Licensteg.

- r c.r.or.

1874p West Power Line Streat slfillII.y 1111:I111111

Crystal atver. FL sedes-s7es g g )

SUBJECT:

uRC INSMCTION Rtp0tf 10. 86 30t/95 22 l

Dear Mr. Board:

Ig I 3$$$33 $ll ggg3g This refers to a special taas inspection conducted on December 1115.1995, at the Crystal River factitty. The perpees of the taspecties was to deterstes

' unether operetten of Crystal River llett 3 ans safe and"in accordance with MC reeutromants. At the conclusten of the taspecties, the findlags sere discussee with these members of peer staff teestified to the enciesed report.

Areas esastand during the taspecties are identified la the report. Illthis these areas the inspecties tecladed selective asesinettens of precedures and records, and intervleus with perseasel cancerates the facts and strometances pertataise to centrol of pressere and level la the reacter costaat syntes maksen tank.

3 Based on the results of this laspection, and the rossits of the inspostles four apparent violattens sore i

j descrlhed in Inspecties identified and are Report betag considered 84 308/95 88, lated enforcement a for esca 3

accordance with the 'seneral statsenet of potter and preceders for lut Enforcement Actions * (taforceanst pelley),1R5341600. In addities.

enforcement action is being considered amatast the Itcensed operaten taselved la the apparent violattens en SeptemberTend 5. Isos. =$1ch are doesrthed holes. A synopsis of an IK levestigatten of activities related to the

Septesher iluly 5.1994, apparent violettens uns provided to see hr ser letter of

7. 1995.

An investigettee of activities as September 4,1994, is la progress and a synopsis of that tesestigattee will be provided to see eben the j investigattaa is cesplete.

l .

i With regard to the first apparest.vtelatten described to the enciesed report, j

nine examples of operators exceedtog the ansteen everpressure curve for the make up tank whlie adelag hydrogen more identified. Additlema11y examplas rettewed by the llAC. tae actise reestred Iqy the alam res. ponse in the nine procedure was not carried out for perteds of tlas ranging fres 30 stantes to 190 stastas.

The saceed apparest violatles sensores toe evolettees of Septenhor d and 5.

!' 1994. These evoluttans men sles apparent violations of the anstam everpressere curve. Sless these evolettees sore met regetres by p)ess -

cemettisms at the ties, het som testead intttated by operators for the

,4 l

l I

i khv a. I s0 f

4 l 3  !

l

$ l l

m I l i i l porpose of gathering data relattve to the maneus tank curve, they are i

3 considered to save been tests or esperiments, perfavnence of tests or esperiments vitanut a written safety evaluatten which provides the bases for i the determinatten that an unreviewed safety question is not involved is a 5

i violatten of'It CFR 50.5p.

.The third apparest vtalattoa. with three examples. tavolved fattures to aset i le CFit, part St. Appsedtz 5. Critertes IV1. Carrective Actten la that once l- destge defitteettes unre identified by peer staff, they were not adegestely j corrected la a timely menner.

t The fourth apparent violaties, with four esemples. tavelved fatteres te aset i '

the requirements of 1g CFR part 50. Appeeds: 8. Criterten !!!. Destge l Centrol in that the destga basis was set correctly translated late drewtags, i

precedures or instructions, for: (1) operation of the ashaup tasks (2) operatten of the mesual suas ever of the Emergency Care Casting syntes

' maps suction from the herated unter storage task to the reacter betidtag suse: (3) antatatning adecasta levantary la the reacter emildtag seep te provide adessata net positive secties head to see les pressure taisettes pump

! with the high pressure injecttaa (llp!)lasection pumps: and (4) estatat supplytes two operattag litch Pressere adequate water questity la the fire uster storage tank.

.j No Notice of Violatten is presently being tsesed for these taspectlen l findtags. In additten, please be advised that the musher and charactertrattes I

of the apparent vtelations deserthed la the enclosed inspection report any l change as a result of further NRC review. A predecistema) enforcement l conference to discuss these apparent violations mitt be scheduled at a later date. us response to these apparest violattens is regetred at this ttee.

In accordance with 10 CFR 1.790 af the IEE's " Rules of practice', a copy of i

~

this letter and its enclosures will be placed la the NRC public Oscuenet Rose.

'Should you have any questions concerning this letter, please contact us.

St 'i

[

) 1.b j

p 1. '

t F. 8theon. Dires l

Divisten of teacter Safety Decket No.: 50 302

[ License No.; Opt 72 .,

li

Enclosure:

NRC Inspection Aspert 50 302/g5 22 i

i (cc w/ enc 1: (lee page 3) l i

l.

i i

4 I

3 FPL cc w/ enc):

Gary L. Beldt. Vice President Nuclear Production (SAIC)

Florida Power Corporatica Crystal River Emery Comples 18760 West Power itse Street '

Crystal River. FL 34438 6708 .

8. J. Nickle. Director Nuclear Plant Operettees (NAtC)

Florida Power Carperetten Crystal River Esergy Comples 15750 tiest Pouer Line Street Crystal River. FL 34428 6708 .

L. C. Kelley. Strector (SAIA)

Nuclear Operations 5tte Support Florida Power Corperstles Crystal River Energy Comples 15760 West Power Line Street Crystal River. FL 14428 6708 Careerste Consel Florida Pouer Corporation NAC ASA P. O. Ses letti l St. Petercharg. FL 33733 Attorney Genere)

Department of Legal Affairs The Capital Tallahassee. FL 32364

. (cc w/enci cent'd . See page 4)

O i

o

i a

j 4

i 4

l FEL l (cceMac1 cont'd) st119essettt Offtee of Radiatten Centre)

Desertenet of Health and i hhaht11tative servless

.- ISU mesmead. Boulevard

!. TaBehessee. FL 32396-075 .

l Joe IWers. Director Dietsten of toergency Preparedness Dessrtaaet of Ceemmetty Affetts

278 Centervios Drive

! Tallahassee, FL 32399 1100 l

Claimen l Board of Cemety Cemeissteners Citres Cemety l

IN E. Apopka Avenue laserness. FL 34450 4245 Adert 8. Sersum SW heslear Technologies 13I0 ser.tytile Pike. Sette 5t5 l Reckville,it 10812-1631

1

! i i l l

1

?.

lt 0

i .

l i

1 .

i e

i e

j usesevans f NUCLEARWe R80lAA10RY" p

e awestrasner.a.w.,enssus 3 anAfHa,eEEE4M ss Report Ns.

50-301/95-21 Licensee: Florida Peuer tem 3301 34th Street St. petersburg, F133733 Decket No.: 50 302 License No.: Opt 71 Factitty Name: Crystal River leuclear Plant Unit 3 Ins et r / N Me r ma ' ':_ 2=r game sigest Accompanying Inspectors: C. Rapp, teacter lespector R. schts, Reacter laspector Approved by: h-napels o. r.artstensen, satef I 33k/p Date 5tgeed l Maintenance tranch Divisten of teacter safety SL89Wtf Scope:

An inspecties was conducted te detemias -if samagement provided adopate guidance, tratalag, precedures, and other support necessary te east management's espectattens for operetten of the nake-up Tant systaa. The inspectica reviewed Make up. Tant data to deterates if the systen uns operated in accordance with approved precedures and if appropriate acttees unre taken for est of specificatten condittens.

Results:

Four apparent violations were identifted:

Apparent vietatten 50 302/g51101: Nine examples of operetten of the askeup tank evt.stda the acceptable operating regten uhtle adding hydregos.-

Apparent violetten 50-302/95 11 0t: Tae esseples of conducting se .

unesthertaed test er experiment without a urttten safety evaluetten centatatap the bases for the deteristaatten that sa enrettesed safety questies did est esist.

% vutsUZ$T T4 ff . _ _ - _ . _ _

i i

i I

4 1

! 2 l Apparent. violation $0 302/99 22 03: Three examples of inadesunta corrective action concerning leadequate revistems to Curve 8. Esimus kke up fank Overpressers; inadequate review of Probles Report g414g; and inadequate j corrective actten for required task volumes.

I

! Apparent violatten 50 302/95 12 04: Fear examples of landensate desty

! control centerntas the incorrect design informatten costaland in Carve 8 which 1

allowed the plant to be operated setstde of its design basis, inserrect i

swapever point for the berated water starege task, inadeemste not positive j section head for the les pressere safety tajectten pumps dortag sannover to i

the teatter tutiding sump, and statena volume reesirement for the fire water j storage tank not being set.

l Weaknesses were identified in the human facters aspects of informattaa

! provided to operators for control of askeup tank level and pressure, lack of j tracking out of calibratten data, the failure to place the askeup tank

' level / pressure alarm in a restine calibretten program, ineffective communications between operattens and engineering and within both departaasts, i

vague guidance provided to operators in precedures for uhan precedures are j adequate for evolutions being performed, for alars response times, and for j determining when evolettoes constitute a test er superiaant.

I i

! l l

i l

l 1

i f

1 l

1 fi I

l l

l REPORT DETAILS

1. Persons Centacted i**

Licensee Personeel 5 ' i

• K. Baker. Manager Nuclear ceaf tgeratica Centrol '

G. Becker. Nuclear Plant Operattens Evaluator

• s. Belt, vice ernstesat Nuclear Production

- *R. Bright heleer Prtactple Licesstag Eastaser J. CassheII. Nenager Nuclear feuer Technical Support i

• t. Davts. Metatenseca Manager i

• D. deMonfort, Nuclear Operattens Instructor j

• M. Donovas. sepeniser Nuclear Power Technical Support l

• R. Enfinger. Senter Licensing tagineer P.. Floatas. Senter Nuclear Liceastag Engineer

• A. Frised, Nuclear Principle Liceastag Eastseer

• B. Guthermen. Nuclear Licenstag Nasager a l

• G. Halmen. Manager Nuclear Plant Operattent i

V. Hermances. Senter Nuclear Employee Centern specialist l *l. Minkle. Director. Nuclear Pouer Operattens i *L. Kelly. Director. Reclear Operations site Seppert i *W. Kisner.Jr.. Senter Nuclear Schedule Caerdtaater Manager

• K. Lancaster. Nuclear ions Trataing l

• J. Llad. Nasager Nuclear '

• t. Longhouser, Manager Not ear Security j 'J. Maseda. Manager Nuclear Eastaaering Design

• R. McLaughlte. Nuclear Regelstery Specialist j B. Moore. Manager Nork Centrols

( g *S. Robinsea. Nasager Nealear guality Assurance J. Smith. 5eperviser Operater Training

• 0. Stenger. Attorney

• P. Tanguay. Otretter Nuclear Engineertag and Projects 1

- *S. Weinberg. Attorney R. Widell. Neelear Operattens Training i

Other Itcoasse eseleyees contacted included engtesers technicians, operators and effice sereannel.

NRC Personnel ,

l *R. Butcher, senter Resident laspector i *T. Ceeser. Resident laspector

• P.Region Kellogg!.

I Senter Pro, lect Manager. Divistes of Reactor

'K. Landis. Branch Chief. Olvisten of Reacter Projects. Region !!

• C. Race, Reacter Inspector

• t. Schin. Reacter inspector

• Attended (mit latentes.

i l

l l

1 i

i f

! 1. Operetten of the Makeup Tank i The taspection documented la NRC laspecties Report (!R) 50 302/95 13 l reviewed the circumstances sorroundlag the Septeater 5.1994 event i involvtag pressure centrol of the reactor coolant system ankeep tant.

5 Dotatis of the review of the September 5.1994 event are decamented la j paragraph 1 of IR 50 30t/9513. Apparent violatten After forther80 34t/9513-01 vertes, the IRCwas teentified as a result en of tusthat rettes. se Septestier e and 8.1996. est escastems l,

has onlyconcluded did operators that,fatt ta felles precedores, tut'te CFR 50.89.uns alse

v$ elated to that the mentpelattees more est reestred by pleet condituns

and as approved precedere esisted for the conduct of the tests. This

! apparent violatten is discussed la detail is Section 1.1 of this report

! and identifled as apparest violettaa 50 80t/95-21-02. Apparest violation 50 303/95-13-P1 is adstatstratively closed and the failure to follow procedures en Ses%eser 5.1994 tientified ta 58 302/95 13-01 is incorporated late asampia 1 ef apparent violatten 56 302/95-23-81.

Paragraph 3 ef IA 50 302/95413 documents the NRC rettew of the destga hasts of ten operattag lletts. Dertag this rettes it uns detersteed -

that the IRK pressure iteit curve constitutes a design basis itett.

Apparest violattea $6 302/95-8101 descrthed la detall holes therefore

' represents examples not asir of a fe11ere to felles precedores. but also examples of the failure to operate withis the design hasts Italt.

1.1 Itakaup Tank Data a! The taspectors revieued ashe-up tant data for the time perted June 1.1994, threegk Septesher 7.1994. to deterstes the number of tiens curve 8 Itaits sa sake-up tank pressure and lesel had been saceeded. The inspectors aise observed the aske up task pressure and level instruentatten in the costrel ross, restened drastats and calibretten records for that tastrueestatten, reviewed the design ca)celatten that seppersed Curve 8. reviewed operator legs, and disenssed the lastrumentattes destge and operatten with engineerl led tettt%9fs.

lastrumentaties la the centrol rees for aske-up teak level and pressure included a high eressure alars, cemester potets, and a

- chart recorder. Daring the tsaa perted June 1.1994, through Septandier 7.1994 the alars was driven by the cesputer such that whenever the cesputer vales for make-up tank pressure escaeded the curve 8 Itsit for the esisting make ep tank level, the alarm would be activated. ifhen the pressure was equal ta er below Carre 8 the alarm wou)d be de activated (there was as progressed dead  ;

band). operaters seeld dispiky the eseputer tetets for sake.us  !

tank level and pressere en vides screens aheve the mala centre) l heard er en the right side of the control heard. The cessuter data was saved by the plant tempster every stemte and uns available for the laspector to retten. The chart recorder was located en the vertical sectten of the anta centret hearW and I

]

i s

l l

4 3

displayed aske up tant level and pressure en ene strip chart. The thart records were saved by the licensee and captes were available

,4 for the inspector to rettes.

Plant cesputer records of aske up tank pressure end level l tedicated that terve a Itatts had been exceeded, and the related 1 centrol reen alars had been validly activated, se numeress

! different occastems durtes the ties persed in questica. The tespectors selected the eleven anat sigetftcaat escastems for i forther review. When the alare had been is centtaunusly for more.-

i 4

than 30 stentas and terve s had been esteeded by aere than 0.5 psig. These occastens were:

l Mastaus pressere i Duratten

' D111 Ilm IR1REMI.), ever Cervg e inste)

1) 7/23/94 12:13 14:14 .

!!! 1.08 10:27.-11:14 48 0.68

!) 7/21/94 la

3) 7/17/M 14:44-16:01 78 9.65 i

4) 7/28/94 14:26 17:29 184 1.10 l

4 7/10/94 09:38 11:38 ISO 8.73 5)

6) 8/6/94 09:55-11:15 141 0.82 i

! 7) 8/8/94 10:08 11:14 of 1.54 1 0.51

8) 8/24/94 13:34 14:50 37 1

9) 9/4/94 04:34-05:06 43 1.38

10) 9/4/94, 15:21 16:46 65 1.87 l

1

11) 9/5/94 04:45-05:21 37 1.T1

. the coopster data indicated that l

i Os eseraters the first occasies reduced aske ep task (7/23/94) level to sheet is taches (te El about inch 11 les psts level 11stt)4 to about 1 pstg (ateve the Curve 8 Itatt), thes,

. increased level to ahest St.5 inches. As level was increased.

• make up tank pressure increased to seest'tf pstg. The ceapetar data indicated that the Curve 8 alars would have been activated i during the hydrogen additten. ressined in through the level tecrease, and than stayed la for about another 95 stoutes as pressure gradually decreased t,e below the alars potet. Chart recorder data tadtsated apprestantely 1.0 tsue hscher level (54 j

taches and 81.5 lashes) and asprestastely 0.5 to 1.8 pate neuer j - pressure (11.5 psig and 27.5 estg) than the computer esta.

+

tattheatles and assuracy of these instruments are addressed later i

4 i

i i

4 la this report. When plotted on Curve 8. chart recorder data indicated that asas-up tant pressure escended Cerve 8 dering the hydreven additten. crossed to the acceptable reglen of Cerve 8 during the level increase, and then roastaed to the acceptable region. Operator legs incleted hieedtag 585 gallons free the SCs

.te the 'C' RCET then feeding 335 gallons from the 'C' RCN to the EE. There were as other leg entrtes cancerning this evolettee.

.As assessment of operater legkeepteg is addressed later to this report.

The inspectors assessed whether operater settena dertag the 7/I3/94 evoletten were la asserdance with operat preesderes.

precedure Op 408 Maksen and portftcetten System. .75, step j

4.4.3. directed operaters to step redactag make.ep taak leset

' den gaff decreases to les level alors'. State the precedere

- directed operators to step reductag level after getting the 55-tech les level alors and the data tedicated that the sehe ep tant l 1evel was tacreased, within apprestastely 11 steetes, to aksee H l i

inches, the taspectors.cencledad that operaters did not violate the procedure when they decreased sehe ep tank level helse N i inches. Neuever, the inspectors ested that engtasertag i calcolatten 130 8024 free ehtch Cerve 8 uns dortved, asemed that the este ep tank emeld sleeps he operated at en tedicated lesel of l

greater then er equal to 55 inches. The inspectors concledad that 4 the operettag precedere and les level alors setpetat Let M 4 taches) were weak ta supporttag the eastseering calculetten. The  !

taspectors aise concluded that operators violated precedures dertas subsagnost parts of this evolettes. Operaters did est felles Op.401 instractions for adding hydrogen. The appitceble i i

' step ta Op-402 specifically stated " refer to Cerve 8 of W.NE ,

l for assimum 181T overpressere". Neuever, opersters essesded Carve 8 daring hydrepea additten. Operators aise esteeded the l ~

adetatstrative listts of Op 1035. Curve S. Mexism Naksep Task Overpressure, for se estended perted of time (abset att staates). l Engineertas calcolattaa 13D 8084 assened that the make ep task was1d always he operated at an ladicated pressure of less that er

aqual to Cerve 8. In additten, precedere AR.483. Aansestater Response, regstred .that, for a salid alars, operators take acties l to reduce pressere. to witkta the Itstts of 0p.1938. Carve 4.

neuever, during this eveletten, operatore did met take tienly acties ta response to the 'INEMp TAE pMIS MIW' namesciater.

Operators stated that vesttag the make-ep taak to rednes pressure l

would take appremiantely tes etnotes hacesse of' the ragstred auxillary operator actions. The austitary operator actions included selectine a waste gas decay task, deastag natt.

contaminatten tiething and entertag a contamtmated area. clestag one asemal valve and spesieg another, entttee the costantestad j

area, and starttag a waste gas cegresser. Dartas this eseletten, testaid of reductag pressere ta response.te the alors, aperstars j - inces tsad level (and pressere) to the ankow tank and then left 2

i - the alars is for apprestastely as addittamal 35 elastes eithest taktas acties to reduca pressure. tattle the campeter/alers and

)

i 1

3 i

l 1

N l

l i 1 5

1 chart recorder data differed, all tastruments were operattenal and ]

i tadicattsas free all should have been consteered valid. Operetten i with a valid pressure indicatten outside of values alleued by I l

precedures, as ladicated by the alarm, constiteted a vietattee of the precedores. Te rely en the chart recorder tedicatthh and j disregard the alarm and camputer data, withest havlag the alars/competer tedicatten checked by estatenasse personnel and l

preven to 'A tavalid, useld.be esasidered eencesservattee and j unacceptable. As discussed later, ta this testance.the chart recorder tedicatten was apparently tenecerate and esacesservative.

i 1

The failure of operators to fellow precedures by violattag the lietts of OP 401 while settag hydrogen en 7/13/94 is identified as an example of apparent violatten 54 38t/g5 21-41.

l On the second occasten (7/25/94h. computer data tedicated that

l. sperators reduced aske.us tank level to apprestattely 5s.5 taches j

(pressere went to shout 15 ests).' then tecreased level to apprestaately 83 taches chile pressere lacreased to apprestostely Ig psig. The data tadicated the Curve 8 alare useld have been i activated when level was decreased to $8.5 taches. roastead is j threega most of the level tacrease, and them danctivated as level

lacreased above 73 inches. Chart recorder data ladicated that pressure went free a law of apprestattely 15.5 psig to a high of

! appresientely 28 pstg uhtle level west free a 1ew of sheet 5g I

j taches to a high of about M taches. Ilham pletted es Cerve 8, the chart recorder data tadteated that aske.up taak pressers was se er below Curve 4 throughest this evolutten. Operater tecluded i gallons j bleedtag 400 sallens to the 'C' RCET, them feed from the 'C' RCST and 100 galleas free the T to ratee j

hydrogen pressere to 27 psig. There were no other leg estries for j

this evolutten. The taspectors concluded that operators vtelated procedures dertes this oveletten. Operators escoeded the administrative lletts of OP-10SS. Cerve 8. Mastaun Makeup Tank j

j Dverpressure, for an estended perted of time (about 48 steates).

la eddttien. seersters die not take timely actien ia response to j

the 'MAKtUP TAlet PRESS MIST

• annectater. lastead of Pedettag pressure ta response to the alars as required by the essenciater response precedure. eparators tecreased level (and pressors) la the sakeup tank. 7his is an asespie of apparest violettaa 50 303/96 12 01.

On the third occasten (7/tf/94), cessuter data indicated that operators started at a mahe up tank 1evel of apprestostely Tt inches and a pressure of apprestaately 19.5 pstg, added hydrogen to increase pressere to about 11.5 pate, then tacrossed level to atest 77.5 laches (pressure went to about Il psts). The data .

tedicated that the Cerve 8 alare woeld have activated dertap the hydrogen additten, reestand te durtag the level tecrease. then stayed la for about amether 78 stoutes as pressure gradshllF decreased to belos the alare potet. Chart reaseder data tedicated

- that tattial make up taak level was about 75 taches aedstettial pressure was abest lg patg. pressers was increased to abast 21

_ . _ _ .___.___.m_ ___..~..___ _ _ _ _ _ _ __ _ _ __ ___ _ . _. __

t

) S psig, then level was increased to anest 78 taches (pressure west l to aheet 24 psig). then pletted en Cerve 8. the chart recorder i data indiented that aske.ep tank pressers was holes Curve s i

throughest this evoluttaa. Operater legs inciated feettag 150 gallens from the 'C' RCET and 30 gallens of domineraltaed unter to the make.ep tank. There were no other leg estries for this event.

The inspectors coesladed that operators vis1sted precedores during 1

this evolutten. Operators did met felles Op M tastreettees for adelag hydrogen by seneally typasstas the 15 ps 2

regulater. Tee aspitsable step ta Op-402 spect 3818. Add

• Deterates assissa IRff everpressere ustep Curve 8 ef destrud assuet of hydrogen while ensuring 181T pressure Itatt is j not ascended.' lieuever, operators esteeded Cerve 8 Wortog i hydrogen addities. Operators also escoeded the adstetstrative i lietts of Op 1838. Carve 8. for an estended perted of time (thest t 78siegtes). In addities. ductag tais evetetten, operators did not take tteely action in response to the *WEUP TM ptts$ NIEN*

1 annunctater. lastead of reductag pressure in response to the

! alars as required. eperaters tacreased level land pressere) ta the i eekeup tant and then left the alars in for abset as Ints addttlesal is an 70 minutes utthest takteg action to reduce presssrs.

computer ladicated that On f 1 (7 operators started with a aske.up task level of about 73.5 taches to tecrosse

! and a pressere of atest 18.5 estg. added i pressure to abest to pets. then tacreased eve to about 88 inche

! I (pressere uset to abest 38.5 psig). The data tedicated that the

! Carve 4 alats sesid have activated when the hydrogen use added, l remained la through the level tacrease, then stayed to for sheet 100 stantes 1 as pressure gradually decreased te holes the alars pelat, recorder data indicated that the letttal aske-f se tank level us abset 74 taches and pressure was aheet 18 pets.

4 pressers was increased to aheet 23 pstg thes ,

! se Curve 8, the chart recorder data tedicated that sehe.ep task 1 pressure was se er holes Curve 8 throughout this ovelettee.

l Operator legs tecluded addtag hydrogen to the enks.ep task to the l anslaus for the curve, feeding 47 gallons free the 'A' RCET and l

! 253 gallons free the 'B' SCET. than adding 60 gallens ofThe j desineralized unter to the make up tank.The taspecters canclud t entries for this evoluttom.

l l

eparators violated precedures dertas this evo bypasstag theil este hydrogen regulater. The appitcable step la Op-402 spostftcally regetred *easurtag M pressure limit is est exceeded.' lieuever, operators escoeded the Carve 8IRff pressere Operators aise eassaded the 1tatt ductag hydroges addities.adetetstrative lietts of Op!

la additten, during this asslettee.

of time (about 184 eleutes)ly action la response to the "IBEIM eserators did est take Stee Instead of reductag pressere is TAlet Ptt35 Illte annunctater.

i l

I I

f response to the alars as required. eperators tacreased level (and  ;

! pressure) ta the makeus tank and then left the alars to for asest i an addittenal 180 minutes withest takteg actten to reduce.g

• l pressure. This is an esample of apparent vietattes i

i ,s l 50 301/95 12 01. )

j De the fifth occastes (7/30/94), competer data tadicated that

! eparators started at a aske ep task level of sheet 74.5 leches and i a pressere of about 19A psts. added hydrogen to leersene pressere l

to abast 23.5 este less' the alare estivated). then'ehset 199 j atastes 1 ster tacreased level to abest St.5 taches (pressere seat i to about 28 psig). The data tedicated that the Cerve 8 alare i useld have estivated uhen hydrogen uns added and roastead la estil j it cleared when levelthat uns subse

data todicated the tattla)quently ratsed.. Chart reco i inches and pressure was about 18.5 pster pressere uns tecreased to l aheet 13 psig, thes level uns tecreased to abset 84 taches j (pressers went to aheet 25.5 pstg). When plotted en Curve 8. the i chart recorder data indicated that make up tank pressers reesteed i helow Cerve 8 threechest this evolution. Operator legs tecluded l

feedias 45 gallens from the T RCET and 255 gallens free the 't' RCST to the make-en task.1here were no other. leg astries for thts evoleties. The taspectors concluded that opersters vtelated precedores dettag this evolettaa. Operators did met felles W 402 tastructtens for eddies by sensally typasstag the 15 pstg i hydrogen regulater. The appi le step in OP 4et specifiestly i

regetred 'essertag left 1tatt is met succeded.* Maurer.

4

[ operators escoeded the 8 Inff pressere Iteit eartep todrages  ;

l additten. Operators ales seemedad the administrative 1 atts of Op-1038. Curve 8. for se estended perted of time (ehest 19e l stantes). la addittaa, dertag this evoletteer, operators did met l l take timely action la response to the 'fguEUP TALE PRtII BlW" i annunctater. Instand of redscles presserit ta response to the  :

j alarm as reestres, operators teeE as acties for ehest ist stastas.

i then increased level (and pressure) la the makeup task to clear 4 the alare. This is an esemple of apparent vtalattes  ;

l 50-302/95-12-01.

/

De the sixth occasion (4/E/94), campeter data indicated that operaters tacreased enke-up task level to abest R laches

{ pressere (pressure went totoaheet 17.5data abast It patel pois. The,that tadicated them the Curve added 8 h alarm weeld have activated een the hydrogen was added, thes 1

stayed la for aheet 141.atastes as pressure gradually decreased to

' helow the alarm.peist. Chart recorder data tedicated that the maximum pressers was about 25.5 pste at ahoet 81 taches. thes

! pletted on Curve 4. the chart recorder data indicated that make ep tank preasure renalmed hales Cerve 8 throughest this suelettaa.

i Operator 1ees tectosed addleg 188 gallees from thaJCt48t to the aske-up tank. Thereante as other les entries far,this eselettes.

i i The taspectors concluded that operaters vielstedjroundses dortog

this eveletten. Operatore did not follow OP-40t teatreuttees for

i i

l f

4

)

i

i i

• 8

! adding hydngen by manually hypassing the 15 pste hydropes i regulater. The appitcahis stap to Op set spectitcally required Neuever, aparators 3

• easuring Itif pressure Italt is act ascended.-

l essended the Curve 8 Inff pressere Instt during hydrogen seditten.

l Operators also exceeded the adelaistrative limits of Op-las, Is '

4 Curve 4. for an estended period of time (ehest 141 stautes).

f additten, durtag this evolettes, aparaters did met take Steely

! action la response to the "IUEEW 1st 9Rt35 NtW ammusctater. )

1. .

lastead of redactag pressure ta respeese to ths.alars as vegstres,

! operators left the alors ta for abast 341 atestes. This ts se j

esemple of apparent violatten IG.38t/95 tt 81.

I On the seventh occastes (4/4/94), seeputer data indicated that

! eperaters started with a este-up task level of thest 79 tashes and

! a pressure of thout te pstg, then added hydrogen to tacreses l pressun to shout 27.5 psig. The data tedicated that wasthe Cerve added, 8 thee d

l alarm would have been activated when the 11y decreased to stayed in for abset 67 stoutes as pressert below the alars point.. Chart recorder data tadicated that the I

' eastaun pressure was abest 26 si at a level of sheet 80 lashes.

When platted as Carve 8, the recorder data indicated that make-up tank sure remaiend en er below Curve 8 threaghast this evolutten. were as operstar les estries for this suelettaa.

l The taspectors concluded that operators visistad procedures during this evoluttaa. Operaters did set follow Op.40t testrucktens for adding hydrogen by menen11y typeastes the 15 pale regulater. The applicable step in OP-est spectftca Iy reestred l enaria wr prenere itets is ut mes. sed.- non = r. epe nters d '"-

SYttsf,.'tt.PCtt.LIF.E'un.lMt.f="1"1.'E a'" fff.'!'la l Carve 4. for an estanded perted of ttee (abest of stautes).

additten, during this evelettes, ters did not tabs timely 4

.P Q5 & at . m a n sisistr .  !

tasiead'ef redome*H l i eperators left the alaru la for ehest 67 stestes. This is an

! esaap)e of apparent violattaa 50 30t/95-t! 81.

t'

' operators started at a este up test leves or eas6 '#s = deb 6 . ,

pressure of ahest la pale, teoreased level to shout 31.5 taches Smar rint 145,D st%!b=_td8t 88r8t'P'.**.'"a*"  !

alars would have been activated ines hydrogen was added. thes stayed in for abost 87 minates as pressere gradually decreased to

below the alors point. Chart recorder data indicated that the j annimum pressure was abast EF psig at about 88.5 taches. lous pletted sa Curve 8, the chart recorder data tadicated that make-up j- tank pressure renalmed holes Curve 8 throughest this ovelation.

There were as operstar. leg estries for this evolettes.. The taspectore caecinded that operators violated presadures dortag this evolution. eparators did met felles Op 442.tastreettees for adstes hydrogen by asses 11y typeestas the 15 psis hydrepes 1

i l

i i I e

I

i t i i

g regolater. The app 1teable step la OP.482 specifically reestred

• essertog stif pressors Iteit is est esteeded.' Noeever, operators caceeded the Cerve 3 leff pressure limit derleg IWdrogen ad$ttlen.

Operators aise esteeded the adstetstratta itetts of Op.nces. la Corn 8. for an estended perted of time (about 87 stestes).

addities, darteg Ots evolutten. eperaters did est tehe tteely acties is response to the *1gEEUP TAIK Patss MleN' assenstater.

lashel of vedestag eressere in response to tie alars as restred.

i emerators left the alarm ta for about er stoutes. This is en example of apparaat vtalattes 50 30t/96.II.01.

On the sloth sceastee (g/4/94) computer data tedicated that operators started et thest 80 taches and D pstg in the enke up tank, tecreased level to about 85 inches (pressere meet to shout 26 pste), added hydroges to tecrease pressere to aheet 31 pste

! (when ha altre activated), reessed lowl to about St.5 taches j (pressere east to about 14.5 estg), tesreased level to about 79

' inches-(pressere west to thest te psig), then vented the enke-ep task to ehest to pstg. ' The data tedicated that the SS tech atetese level ens exceeded during the owlettes. Also, the Cerve i 8 alars useld have been activated from the time that lydrogen was edded at the besteetne of the evoletten estil the mate.ap task was l vested at the and of the evoluttaa. The data aise tedicated that.

as leal ens redoned from shout 85 inches to atest 51.5 taches, l pressere esseeded Curve 8 by as tecreastep aseset. Chart recetter f ,

data tedicated that pressure meet free a aten of sheet 25.5 pstg

! at a level of shout 85 taches, to a les of sheet 14.5 at a i level of sheet 52 tsches, thee to a high of abset 25 at a i level of abest 80 tsches. 18tes plotted se Cerve 8, j recorder esta tedicated that make.ep task pressere uset shoes s Cerve 8 as level was being decreased below aheet te techos, l

remained aheve Curve 8 durtag the level decrease, then returned below ".arve 8 dertog the level increase. Operator legs testeded feedtog 125 gallens from the "t* RC8ft adding regens bleedtes 924 gallons to ths 'B' RCST: and feedles its ens free the 'A' i RCST. 380 gallons from the 'C'. RCET. ael 380 allens from the *C' i

RCET. There sore es other leg entries for th : evoletten.1he taspectors concluded that operators vtelated precedores dertog this enletten. Operators did est felles ap.401 testreettees for i

addine hydmma hr manne11y kreetstes the 15 eme regolater. The appilcable step la OP ost spec'ica y regetred

• easering self pressure 11stt is not esseeded.' fleesvar, aparators

. escoeded the terve 8 IElf pressure llett dering hydrogen addities.

i Operators aise exceeded the adotatstrative Itaits of Op-1838

' terve 8. for an estanded parted of time (abest 43 stoutes). la addities, dortog this eveletten. eparators did est take Steely action la response to the 'lltKDF TNK.pAESS Mlgl' anseestater.

lastead of reductag pressere to clear the sters as reesteed.

4

• ehersters redeced.Sevel (and pressure) intick caused theMurve E 1 Itatts te he exceeded by an increestes assent. Opersters left the l -

alare la for about 43 mimetas before vestles the sehe-ey task to i

i i

1

.i I

E._____._ _ _ _ _ _ . _ . _ _ _

1 l

1 i

le clear the alars. Evetutions, met reestred by plant condittees.

- conducted to tellect data constitute a test er esperteest.

10 CFp 50.5p requires to part that tests er eenertaaets that are conducted must contata a written safety evaluatten that provides the bases for thy as sereviewed safety questten does set esist.

The evolutien conducted en 9/4/94 without a arttten safety y

evaluaties is identified as esemple 1 of apparent violettes i

56 303/95 12 01.

i On the teeth escasten (3/4/94), campster data tedicated that operators started with a anko-up tank level of atest 31.5 taches and pressure about 30.5 pets, added hydrogen to tacrenes pressure to thout 30 este, then atest 95 stastas later a differset shift of

emerators vested the aske-se tank to reduce pressure to about 13 pstg. The data ladicated that the Curve 8 alare usuid have been

' activated from the tlas that hydrogen was added sett1 the Staa that the enke.up tant was vented. Chart recorder dets tedicated a maalaus pressere of about 25.5 psig at a level of abset e tsches.

j When pletted en Curve 8, the chart recorder data teficated that aske.up. task pressure ramalmed en er helow Curve 8 dortag this i

evolution. Operater legs taciudad venting the este g task at the t and of this evolutten. There more no ettaer les estries for tkts evolutten. The inspectors concluded that operators vielsted procedures dortap this evolution. Operators did est feller Op.402 instructions for addles tydrogen by manually bypasstag the 15 pstg q

hydrogen regulater. The applicable step ta op-est spectfleally required "enseries fIK pressors 11att is est amemaded? Nuesvar.

j .

eparators exceedet the Curve 8 palf pressere 1tatt dortag tpdragea

. 9 j addtttes. Operators also escoeded the adotaistrative Itsets er Op-1038. Curve S. for an estended perted of ttee (about 85 j sinutes). la addities, dortag this evolutteer, sperstors did met take timely acties is response to the "fWWEMP TALK pMIE WSt*

annunctstar. Operators left the alars 1a for thout 85 enestes hefers they vented the enke-up taak to reduce pressure. Ihts is an example of apparent violatten 50 303/95-21 01.

On the eleventh occaste's (t/5/94), computer data tedicated that operaten started with 4 aske up tant level of abast 84 teshes and a pressere of shout to psts, added hydrogen to tecrease pressure to aheet 30 psts (des the alars activated), tacreased least to about 96.5 lashoe 'When the alars cleared at a pressere of about at pstg). decre level to shout 53.5 inches (pressere seat to stemt 14 pstg), then tecreased level to about 31 teshes (pressure went to atest 17 psig). The data indicated that the Cerve 8 alarm would have been activatM for shout 25 minutes from the time hperegen was added until the level was tacreased to abant 35.5 inches and the alare clea'ad. Then the alars useid hose base activated for abast 37 alastes free the ties the lonel dearesse was hogun estil the llam tiaered dartag the subsegeset lesel increase. The data alas ladicated that, as level ses deersoned following the alars pressura easseded Curse 8 by as teoreestas amenet. Chart reeneder data indicated a saataan pressere of shout 1

l

^

, )

b l s

I f.

3 v l 11

! 30 pstg at a level of sheet 86 taches and a statsum pressure of

! about 14 pstg at a level of about 54 taches. When plotted on Curve 8. the chart recorder data indicated that make upstaak

, pressure meat above Curve 8 shout stduey during the level decrease

! and returned to en er below Curve a during the subseedeht level 1

increase. Operater less tacinded feeding 185 gallens from the *3*

RCET: addlag t bleedios SEE es11ons to the 'B' RCST: then feedta lens from the 'A' W 300_ga11ess. free the 'C' RCETand g1 gallens from the "B" W. There were as other leg i estries for this evoluttaa. The tespecsers concluded that eseraters vielsted erecedores eartag this evoletten. Operstars

{ did not fo11em Op 408 tastrustless for addtag hydrogen by messa11y 1 bypessing the 15 pstg hydrogen regulater. The applicable step ia j Op 442 specifically regelred 'essartag fEff pressure itsit is act j enceeded.' Neuever, operators ausseded the curve 8 ptff pressure

iteit dortag hydrogen addities. Operators aise exceeded the l administrative 1tatts of Op.1038. Cane 8. for an estanded perted i of time (shest 37 atnotes during the level decrease and subsessent j level increase). In addittaa dertog this evolutten emerators j did not take steely acties la ressesse to the *INEtuP TAISE PRESS HIGH" annunciator. lastaad of redscles pressers' to clear the alare as required. operators cleared the flest altru by teoreasing level (and pressure). Men the alare came is uhtle operators vers  !

reducing level '

j uhtch caused the(and Cervepressure), aparators 8 Itatts to be escoededcoattneed to reduce level by en tacreasing assent. Operators subsegnantly cleared the alare after 37 stestas  !

by tecreasing level Land pressure) in the make-up tank. l Evolutions. not rege9 rod by plant casettiens. conducted to collect data constitute a test er escorteset. 10 CFR 58.59 ragstres la part that tests er esperteents that are conducted must contata a

writtaa safety evalsattaa that pegrides the bases for uly an unreviewed safety seestles does not esist. This evoluttes was ,

conducted without a writtaa safety evaluation and is ideettfted as i example 2 of apparent violattes 50-30t/9512-82.

'Of the 11 occantens rettooed. the taspectors noted that tuo of them, the statt and eleventh occastems, differed in sees esterial j respects from the other ates. On stas of the occastens. eparaters a were taking actions to tecrease aske.g task hydrogen pressere (as receemanded for SC5 cheststry centrol) den enneeded Carve 8 lietts. Neuever. en tus of the escastens (9/4 a.a. and 9/5/94

operators escoeded Carve e itetts while reductag make-ep i

a.m.)

tank level (and presserv) ans uhtle performing an evoletten tha I was not required te support RCS chemistry contre), Da mine of the i accastena operators teek delayed actten (or as actlan) fellewlag i the alars to reduce the ameset by thtch sake up tank pressure .

i exceeded the allowable ropten of Carve 8. Neuever, en tuo of the i operaters teek-

! eccastems noncenservative (9/4/94 a.m. and actions fellestag the a9/5/94 a.a.)lars to fwther the amount by which aske up tank pressere aseneded,the allausbie i reglen of Curve 8. Also, the Itcenses stated that eMme of the i escastens (9/4/94 a.m. and g/E/94 a.s.) operators plammed the i

1 I

4 k

1 i

i 1

i 1

}

l l 11 evoluttees to challenge the accuracy of Cerve 8 and teak data during the evolutions. In additica, sa 9/5/94 a.a. eperators stattened an auxiliary operater by the make up tank vest to take 3

action if a plant event occurred while the make up tank m 2 escoeded Curve 8. The inspectors ceae)eded that uhllt waters j vietated precedures, and the destga basis Itatts estabilshed by i the 18# pressure 1tatt curve, en each of the 11 escastees

retteund, available tafersetten indicated that the latest of the j sperators durtag the tus escastans (9/4/94 and $/5/94) ettysred from the ethor stas escastens. Does enantherland aestpu14ttens

! of the plant en 9/4/94 and 9/5/95 to additten te violettag the precedores Op 402. Op 1838. and AR.403 uhtle editag hydrogen, aise violated le CFR 50.59 in that the manipulattens more not vegetred by plant eenettiens and no approved procedure entsted for the conduct of the tests. As described previously, these tue unauthertred tests are identified as esamples 1 and I of apparent vtelatten 50 302/95 11 02.

Elle reviewing the computer and chart recorder data the inspectors noted that many substaattal changes ta make up tant level had been made where Curve 8 was not escoeded. The licensee's analysis of this issue found that, during the ttas perted .1mme 1 tareagh September 30 1994, there more 480 manipulattens of make-up task level er presseret tecluding GIS level reductions or level tacreases. 49 hydrogen addittees, and le vents of the tank. He Itcoasse further stated that II of these sentpulattens (ealy 3.15) rossited in the cageter. generated annunciator for make-up task pressure betag ta the altre o conditten. This indicated that, en most aske.up taak level or pressure annipolattens operators did not escoed the pressore Itatts of Curve 8.

la summary, the inspectors concluded that operato'rs emeseded Curve 8 Itaits and violated operating precedures en at least 11 occastens during the time perted June 1.1994, threegh Septester

7. 1994. Two er these occastens were unauthorized tests er esperiments not required by plaat condittens.

2.2 16enas facters and Centre) Rees lastrumentatten The tospector assessed the human fasters aspects of the control rees tastrweents for aske-up tank pressure and level. The computer data for pressure and level were clearly dispityed digitally. Pressure was displayed to the nearest one-hundredth of a ests and level was displayed to the nearest one.hendredth of an inch. The chart recorder was auch more difficult to read.

Pressere could be read to apprezinately the nearest ene half psig and level cos1d be read to apprestastely the nearest task. Stoos the Cerve 4 calculatten (190-0034) provided asly tuo feet of estar (about 0.8 psig) as protection for the 191 peopv the toepeater -

considered that the readability of the chart reeerder uns est suffittent to support operating the aske-g 4ank en er meer

__.____ _______ ~._ __ _ _ ____ _ __ __ _ __

l 1

i

}

I

) 13 4

i Curve 8. Operettee en er near Curve 8 was alleued by precedures j and was encourated by Itcensee manatament to mestatte the i concentratten in the reacter coolant system. Also, therei ne i display in the centre) resa for praatetty to Curve 8. Tephe a

seteentnatten of prennetty to the 1tatt of Curve 8. esetaters j would have to seeselly plot make up tant pressere and level en a

copy of Curve 8. The lack of such a display, uhtet could been been provided by the , made it very diffleult.fsr emeraters to tecrease task pressure class to Curse 4 i without esseedtog Cerve 8. taasectors castledad that the i huana facters aspects of the IEff leformatten displayed to the

] centrol rees were soak to sepperting operaties user Curve 4. ,

2.3 Calibretten and Accuracy of Makeup Tank Instrussets l

I The inspector assessed the calibraties and accorsty of the lastrussats. Make up tank pressure tapets to the cesputer and chart recorder were from the same pressure trasemitter and electrical buffer. Malatenance precedure W 15.A calthrotten records fres October 17,1994, tadicated that the chart reenrder l pressure string read abset 8.7 pate 1ew (at Il psig). N enke-i op tank pressure ses eerustly controlled between apprermately 15 t and 30 pstg.

! withta the 1).11 psig seriana tastrement error asemed byThe taspe j Calculaties 196 4024 dated Augest 28,1998, frem thich Cerve 4 i ,

was derived. The tespacters rasted that the chart recorder feed i Iow (by about 1.0 to 4.8 pstel and est of tolereses, at pressures 4

! above 25 psig (i.e.. at 54, Ts, and 100 peigh en tuo conseestive i j 1992 and Octsher 27,1994. Seseek l i calibrations, occasten, it hadnovember been reca 27,libreted and left reading accarately.

However the Itcassee had me trending peggres to identify (and ,

l tattiate corrective actlan) when aa < astrument, that ses '

i calibrated by a asistesaeca pescedure end possibly tapertant to i safety, was repeatedly fesed to tedicate teaccurately estside of i the allowed toleresca. From the as found catthraties data, the

insposter seeld nos detamine whether the chart recorder pressure

would have read withis the 1.12 psig a*==ad anstems testrumet 1 error, at a mate.up task pressure of 29 er 30 pets, dertog the l period July three$ leptember,1994. them inferend of these chart

, recorder errors, the Itteesee prowtly checked the calibraties of ,

the chart recorder for aske-up tank pressure and faced it to be i within the alloweble talerance. The taspectors identified the i lack of tastrument set of calibratten tracking (and corrective i action) as a weakness la the licensen's maintenance program.

j ..

The inspectors noted that the computer make se task pressere and 1

the Curve 8 alars sere not checked by calibraties procedures. By

compartsen of the competer data with the chart recorder data, the j taspectors fesed that the computar generaHystedtestedelper -

~

pretture att ne diert retortler by abast 9.5 la lit esta'about 4

29 psig 'and by thest 0.1 psig at about 17 psig.- The inspectors

! found that actten takes ta completa MAR 93 06-4641A. dated ,

i

14 i

July 12, 1993, uhtch had installed the cesputer pressere pelat and terve 8 alars, had failed to revise precedures to assure that the

'.' lastruments weeld be calibrated in the future. h flat fenettesal test had tested the turve 8 alare setpoints in July INS. The tespectors noted that calibretten precedures useld have reestred the tastruments to be calibrated every to years, and that less

than tus years had transetted stace the testruments were 1

lastal)ed. Therefers, the safety significance of the elseed.

'- precedure rettstens was steer. The taspectors.tdenttfled this fatture of the toplamentattee of mR she8-et eta to appropriately revise procedures as a unebeess in the Itcensee's modtFtcattaa a' process.

Level tapets to the campster and chart recorder were free tuo Through a different level transettters and electrical huffers.

selector switch on the vertical section of the meta centrol heard, centre) rosa operators seald select either tranesitter to the

, chart recorder. The ese selected treassitter useld then sepply the tempster. There were se records of the settch pesttles, there was no tadicated preferred euttok positten, and operators stated that they occastenally had changed the switch positten.

Calibrattaa of level transmitter LTI en December 8,138, per servet11ance procedure Sp ISOE, Enclosure 1, found that the chart ledicated ehest 1.0 taches high at et and recorder 90 incheslevel string (The enke ep tank level was normally of level.

controlled hetuosa about 55 and se taches.) The cesposer petet as not checked at that ties. Calibratten of level tremesetter

.I LT1 en March 18 1994, per Sp 180E, lectosure I, feued that the

chart recorder Iovel stries tedicated about 1.5 taches high et 80 inches of level and thest 1.0 taches high at St taches of level. That ca11bretten aise checked the cesputer potet and fened that it indicated ahest 1.0 faches high at 60 and 90 taches of

.: level. On hath dates, the as feed condittees were generally within a11euskle tolerances, and the instruments more recalthrsted and left utth apprestantaly more error. Sy reries of the aske ep tant level calibretten data, strip charts. and campster data, the taspectors determined that chart recorder normally rund higher than the cesputer by atest 8.8 to 1.0 inches. The taspectors concluded that the 1.0 and 1.5 inch errors found dortog

-calibratten were within the 1.7 inches mentous tastrummet errer assemed by Calculatten 19D 8024, uhtch derived Curve 8.

ah=

In summary, the tospectore identified tus Itcensee related to make-ep Lank testruneet calibration: one weakness ta the Itcensee's maintenesse propres in that est of calibrettee tracking and corrective actten was lacktag for tastrussets that 4

were relied spea to seppert a safety.related calculattee, and one weakness la the itseasee's endificattes process in that a MR that tastalled instruments did est rettse precedures to regette l

4 4

d

i i

i i

d 2 Il

~

pertedic calibratten of these instruments. The taspectors concluded that everall, the make up tank testrementatten (alars.

canester, sed chart recorder) uns suffitteet to scopert operatten 4

of the aska.ep tank utthin the authertsed regtes of Cerve 8.

I 4 2.4 operators Legs

.- la reviertag operater legs, the inspectors fened that the

! 1eformaties in the legs std met thoroughly or sessistently l the toapectors teocluded that desseest the ovelettees. ItsueverInte 11eenses regatraments for the state of the lees did est vie legheeples, as the licensee's regattements are wages. The laspectors aise noted that osereters se shift dortog all er part

of the eleven ovelettens tac)eded 3g differest 1ttessed operators.

apprestostely 1005 of all lleensed operators se shtft at Crystal 3

tiver 3. (The oeersters were se a sta. shift rotatten, and each j shift facinded five licensed operators, toe $80s and three A0s.)

Also, the inspectors noted that there were shift turnovers, during three of the eleven occastens, when the seks-up taet high pressure

slare useld have been in. The shift managers were seppesed te j

attend the shift turnovers. and than ==ald have had the espertonity to beceae emere that the seks up task high pressers alars as in.

The taspectors concluded that most of the itsensed operators et i Crystal R$wer 3 were involved to the nine esseples of apparent l vie)stian 50 301/gl-II 01. Also,11eenses senegement had emple l l espertonity to identify the apparent violetten.

i, 2.5 asportaktlity

• The inspectors rettewed the timeliness of the liceasse's reporttag

! of the operatten of the make-up task outside the destga hasts, 1

with emesssive hydrogen overpressure. The operater evolettens

that challenged the accuracy of Curve a escarred on 4 and 5, 1g94. Probles Report Pt 94 8t57. IEft Pressere Technical Basis Inadessate, was dated Septester f 1g94. I,tcensee

decessets and statements tedicated that sagtesering deterstand en hovember 35. 1994 that Cerve The Itcenses 8 ens further nascenservettve deterstand and outside that sporetten es th destga hasts.

er shove Curve 8 weeld poteettally sites sake.g test bedregen to enter and damage a make.up (1r1) pump dartes a sortate esset, a core flood 1tne break coinc4 dent with a fatters of one emergency

- diesel generater. The Itcensen made the required one.hser le CFR Sg.72 report at 17:55 en lievesher 16,1934, and submitted the regetred LER en December 1g legt. The taspector concluded that the licensee made the see-heer report preeptly afser diesevering

, the entside of design basis canditten. The Ltt uns sadisttted j

late, and was prevtessly identified as fl4 58 30t/94-87.st.

l i

j i

i

s i

i i

d 16

3. Engineeriog Soport l

The taspectors assessed engineering scopert to operattens. They

! revtemd pretes reports and engtaurtag deartment eerrespedence i related to the Curve 8 prehlems reviewed the engteserief calculattee free which Curve 8 as derived and discussed the issues with systes

'

• engineers, design engtasers nad operators to assess the adaguetr of

' engineertag support for this tsees.

3.1 Accuracy of Carve 8. Mastens Makeup Task everpressers Caree The aperators stated that the purpose of the evolettees as September 4.1994 and on September 5.1994. uns to casfies their perception that Curve 4 was lacerrect. The tospectors reviamed September 1990s englasertap support to operattens dertag June The spectfical y, the issue regardlag the escuracy of turve 8.

inspectors reviewed prehlen Asperts M 93-0018 Peteattally inadeemste Nake up Pump NPSN. dated Jansory te.1993: M M 8149.

• M-60 stuck Open dated May 10. 1994 and R M 4267. M Pressure Cerve Technical Basis landequate, dated Septesher 7, 1994. The taspectors aise rettamed related REA M4747, sensera air helag draus late the sehe up peps eartag the perforunnen of SP 836 088 Fell Flow feettas). dated June y.1994: and W este up pasp

( M1305. caneeratag operater burden la l operehtitty daring accidest scenaries, dated 1. ISM. la i

addittaa, the laneetters verteund enstasertag Calcelattes 19-l .

0024 dated August 25. 1988. from e lck Curve 8 was deriveds W 93-06-06 01A. letpoint Cheapes to M Dverpressere. Les MT

l Level, and 9R15-4913 and a related safety evalsatten dated July 15.

i 1993. The inspectors ales reviamed captes of vartees related laternal tacertog senersada and campster atti sessages and discussed issue regardlag the accuracy of Carve 8 uttk systes l engtasers and destga sagtesers.

f hased on this rettes, the taspectors soucleded that the operators'

! sencarn with the assarnar of Curve 8 uts Westified as e i centriheting facter in Probles Aspert M 94 0149, dated Nur it.

i 1994, and was aise identified more clearly to R M 80ST (after

! the Septesher 4 and 5 eyelettens). A done 14. 19 M. urtttes

' enstasering evalsatten of a stated concers regardtar the saareer of Curve 8. 4tch had hoes reported ta R 964149. concluded that the data sepp 1ted did set ledicate an error in Carve 8. The i engineering evelsatten identified me errors te calestatten 190-(

0024. that derived Carve s. sehseevent Itcensea review of Curve 8. after Septesher 5,1994, identified several errors related to the calculetten of Cerve 8. These errors tecluded the emissten of the effects of gas ahoorption. temperatere etmapes.

' and the partial pressere of water v . 1 sternal assereeds.

electrente anssages, and statements eastseers ladiested that M

~ unre discussed seueral M 4149 corrective acties (and Cerve ttmas betwasa engtesers and operaters,

)teclaitag at asettage en j

July it.1994, and en August 5,1994. La additten, sagtesers sure i

}

l t

i i

_ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . . _ ~ ~ _ _ _ _ _ _ _ _ _

3 l

]

17 aware that at least one operater was not satisfied with the

. ressenses from engineering and made attempts to get the operster's

! concerns more clearly documented. Engineertog essagers (ans emerattens managers) did not get proactively tavolved ta4he

{ .

prehless with resolving operator's concerns utth the atteracy of j Curve 8. The inspectors concluded that the failure of 'the j

engineerine evaluatten to reopense to pt 94 8149 to identify 1

errors ta Calculatten 196 0084 with respect to the slope of Curve 8 constiteted landegaste corrective acties. his landecente corrective actian is teosttfted as esemple 1 ef esperant vtelattes 56 30t/95.tt 93. Addettenal asemples'of fatture to take adegeste corrective acttens unre identified as apparent violattens la 1

It 56 30t/9513. As deserthed in Seetten 7 of this report, these i examples are i late esemples I and 3 of apparest violatten 50- 22 05.

Internal Itcenses he aise indicated that engineers and

' operators discussed the eethed utt1taeA by operators to metatata adequate hydrogen overpressure in the MIT to achieve Il sc/kg of

! diseelved hydrogen la the ES. Alse. a memorandes free a systes engineer en August 8,1994, to ens SRO and one M (apparently is response to questlans from edvised the operators to see the cesputer potets fer enha.up pressure and level uhes maktag addittens steen they usund give apre accurate laticattee chart recorder. The asme further stated that the alors teek precedence over the recorder indicattees. Yhts tafernst response to a queetten uns set disseminated to other operators by l

engtasertap or by operettees. The laspectors asted that the l

response a se uns met commistent with a cesservattve operettog 1 prestice that nar valid tatteattene (i.e. alars, eesputer. and recorder) he kept out of the unacceptable operattag regtems. This was emether tadtcattee of a snakness to teamentcattens betmeen eperattens and engtesering ard aise within operettens.

l

! Records available to the inspectors tadicated that the 1 calibrattem/ accuracy of the make-up taak pressure and level instruments (alors. camputer, chart recorder) unre met shalleeped by operators or sapineers. Helatenance records revealed that the make-up tant pressure ladicaters had not been calibrated between November 1992 and October 1994 and then ta October 1994 the chart recorder pressore ladttettaa uns found to be erseassus and noncenservattys. (The tastruussts were scheduled for feettat j calthratten every tus years). No corrective actten was taksi to preclude recurrents of this problem. -

The inspectors eencieded that engineering responded to p%)eas or questless that unre clearly stated te Pts and SEAS. %suever, the quality and tiaallanes of these responses varies.~ mtis the aheve roepenses were taaseplete and usttualy, others appeared te he

- therseek. La ens asegle, eastaaers identifted. spa cases of make-es pump cavitatten identtfled ta pt 94-4149 to he'atrapsskuta left i efter matetenance la certata sections of piping qthey thea i

1

\

i

18 lettisted sodtittations and precedere changes to assure that the sections of pistne were vaated after metatemen:e. In asetter example, RfA 941308. engineers performed a detailed Espeer.fregee sectsten asklag analysis of various potential corrective acttens to reeuce er ellatsate operater burden in ensurtag make.up pop operett11ty darias en accident. An esaapie of amether issue where I

sperators were dissatisfied with serrective acties tavelved the emettency feeduater centrol systs, where operaters had to restteely plass the EFW pumps ta mesual centre) and reduce the f)eurstes dortog tremstests. Ints actten was reestred to the 10Ps and soccessfelly praettend on the steelster, but uma considered by operators te he en ennecessary baresa.1 hts operator harden was se the Ilceasee's Itst of *eperator workarounds' for uttch corrective actiens were hetag appropetstely perseed by plant annagensat.

Elle reviewing Calculatten 190 0024. dated August 28, 1982. free l uhich Curve 8 was derived. the inspectors noted a discrepancy between a calculatten asampsten and licensee procedures. The calselattee assumed that, dertog a LOCA. the switchever of ECC5 pumps' suction frem the 85T to the reacter hetidtag sup useld actor by precedere at sa 5 seep level that useld egnate to a SWST l 1evel of sheet 14 ft. The calculattaa assesyttee stated that the I value for gust settchever level seeld be valid through refuel 8 l ubich had occurred ta 1993. Nessver. Iteensee emergency operating precedures in effect dertae Joe - September 1994 regetred esersters to begin the entlebevor st a fuBT 1evel of 5 ft. The tospectors fesed that a IWR to reise the level of certate

• l testruments in the reacter betidtog (for flood plane conceres) ked heen installed la 1993. The fulR tedicated that the emergesty <

eserating

t a were to be changed but did not ledicate that l talculatten 190 0084 was affected. Fellertog the IWE the ,

emergency operattee precedures had bee 6 changed to regetre the settchever at a tutT 1evel of 5 ft. The Itceasse showed the inspectors an internal engineerleg eensraade dated Hersh 24, l 1955. with an evaluatten sheetag that.the i ft settchever leve uns appropriate. thtle the 14 ft level uns based on tuo runstag  ;

HPI (enke-up) pumps ta one trata, the 5 ft level uns based se only  !

ese runing HPI pene por train and thus less fles and less pressure less free the INST to the F1 pumps. The taspector  ;

retteund the evaluatten la the assereedse and canciedad that it '

was a logical and reasonable esteesten of Calesisties 19D-00t4 with respect to ensartag the make-ep task hydrogen was kept est of the HP1 pumps. However. it everlooked potential verteslag in the Busi and tatredactles of air late all ECCE pumps. Alas, it 4 overleeked the fact that the EDP started switchever at 5 ft in the IMIT and that there assid be ties (and gusi level decrease) lese)ved durtes the settchever. Further, both the same end calculattes 19D 00t4.everleshed the fact that *with sehe g tank level allemed to se belee sort, aparators usukhese as tedicottee of adesmate aske-up pue W5t and as a result could.deside to step the affected peup. The inspector vertfted that the emergesty

l 19 eserating precedures included a reestrement for operators to essere that no aere than one HPI peep per trata was reamlag then SWET 1evel decreased below 25 ft. The internal engineertag ammerandum was signed by a senter neclear engineer and's sueleer engtnoering supervtser. Neuever ne fores) change sqs aeds to the effletal calculatten. The formal change to the calculatten, a osality record. steht have identified the discrepancies with the calculattaa and useld have regelred ledspendent vertftcetten by a sected qualtfted taeor ta addtttaa to the supervtaars 1.

The inspectors that prehles report pt 94-0149, dated 13.

1994. had identifled the need for a feveal caltslattaa to agpert the 5 ft $ WIT suspever potet. The omstaaertog schedule for completten of that analysts uns Septesher 30, 1994. Atee, probles report pt 94 4257 dated September 7.1994, identified the esed to updata calcolatten 190 e0t4.

The inspectors concluded that the Itcensee's failure to ate a new offtetal calculattaa er rettse the calculatten 1 la March 1993 was an apparent vielstles of 10 CFR 58. Appendis 5.

Criterien !!! regetrements for destga sentrol inclettag independent vertricatten of destga calculattens. Yhts landequate desten centrol stil be included la esewie a of apparaat weetatten 50 302/95.tt M. Apparent violattes 95 13 03, example 1, paragraph 5 of IR 50 30t/9513 deserthed an apperust vtaletten of 10 CFR 50. Appendia 8. Critertes !!!. Design Centes), for inadequate destga assumpttens for hereted water storage task seapever level. This apparest violattes is adotatstretteelF

! closed and incorporated late apparent violatten 50308/96Et-M.

esemple t.

Additional examples of inadagnate destga control sers identified as apparent vtalattens la IR 56 308/95-13. As descrthed ta 5ectica T ef this report. these examples are ines late exagles 1,3 and 4 of apparent violettaa 50- 11 85.

.3.1. Review of Operability Cancers Resoluttes Evaluatten Report

~

As part of the review of the Itcensees' new operahtitty evaleotten process, the taspectors revissed the operahtitty evalenties concerntag an unsecured secties of RB semp grottag testalled over the ECC5 pep secties pit. mile tastalled mately for peressnel protection, t&ts grattag was credited to the FSAR for pregnettag objects larger then 1.5 taches from entering the (CCS pas auctten pit.

To determine if"the unsecured B same grating could affect the eserahtitty of the ECC5 pumps, the Itcenses calcolated the pressure en the grattag created by the verst casa.fleersta est of the sucties pit and into the 5 semp and camperedAheHe the

- force necessary to lift the grottes, tened en this talediatten, the licensee conciadad the weret case fleurate taased tenurftetent force to Itft the grattag. The Itcessoa ales considered the forma

i 3

le en the grattag resulting free speatne the DH erop itse with the reactor at pressure. This action ucs directed by precedere to erstect the ON poses from fat 1ere la the event the M peeps hai i been eserated at law flow for as estended perted of ties. The j licensee dismissed this as a peteattal cause using *engineertag i judgement" hased on the calculated flew rate necessary to esert a force suffletest to Itft the grattag. Ustag the some methodelegy for esiculattag the force from secttaa pit outfles the licensee calculated that greater thes 3.000,000 spa useld be restred to ilft the grattag. Dieuever, SW calesiattens by the taspectors, sstes the Itsensee's methodology, fesed that only 138,000 gpa was reestred to lift the grattag. gotasse of the sigafficant difference hetuosa the tuo values, the leseesters enestioned how the vales of 3,000.000 sps was cales1sted. The Itsensee stated the value was calculated by stretterni enstaesring j

and not by mechnetcal engtseertag. Structural engineertag fatted j

? to recogatae that flev varted as a function of the senere rest of

} the differmattal pressere and had performed an inappropriots i

extrapolatten. Furthermore, structural sagtseertas did met j tenmunicate this value to mechanical engineertag for review. The Itceases said the value uns questioned by mechaalcal emetenertag

i after the operaktitty reselsttaa uns tasued, but an ese6entes entsted to amend the vales.

[

The taspectors geestioned the effect of steaWuster taptagement sa the grattag due to spostag the M drap Itse with the esector at i

pressors. The taspectors retteued the procedural goldesee provided kr the Itseases. Prior to directtee to epse the 5 drop ,

l lies, actless more takes to redoes E5 temperature to less thaa 200*F. Furthermore, the W drop line was reestred to be speasd only if M pap operattaa at les fleurate useld sessed ten hours.

Based en the preceder:1 actions to reduce RCS toeparatore below ,

200*F and the estended time perted before this actten useld be roeutred. the taspectors concluded there was as potential for staanVvater taptagemast on the grottag.

The teseectors concluded that, dile there was no tenedtate

. operahtitty concern the operett11ty concern reseletten process j

lacked adesunta revtems to ensure that cesservative operehtlity determinattens were sede. Furthermore, the fact that plant

managensat had accepted the resolutism tedicates that leesfftetest 1

management atteattaa uns given.te this particular reselstles and a willtapness by senagement to accept, withest review, engtesering evaluattens that resulted in coattaved plant operetten. The fallere of the strectural magissertag group to commetcate the value to the mechaattal emptaaering groep for review is emother example of usak ceassaicatsens between plant argentaattees.

la suunary, the taspectors identified tus apparent vtalattees and tus

- weakeasses in the area of engineertag seppert to opersttees. The fat 1ere of the anstasertag evalsatten in roepense to PR 34-0014 to identify errors to Caltslattaa 190-0034 with respect to the accuracy of d

I 1 .

f 1

L

1 1 1

i

(

)

i -

i i i l

} 21 1

j Curve 8 constituted inadequate corrective actten. The failure to

generate a meu er revised efftetal calesistion for turve B ta north 4 1993. Le support the rettsed operstlag procedure lowertag SIET

switchover 1evel to 5 ft. constituted inadessate sesign control j lacluding a lack of independent vertftcatten. The infomal I communicattens between englasertag and operattens regarding operater ese j of aske-up taak lastruments and lack of such commsatcattens within i

operatleas'tedicated a weakness ta letra and interdepartamatal communicattees. Also,.the tacorrect calculattaa and lank of j

cesemaicattens withis the eagleseriet departmaat, sith respect to as evaluatten of en unsecured section of 5 ause grating, indicated a weakness in the Itceasee's operability evaluatten process.

j 4. Procedures j

i l The taspectors reviewed the precedures that were la effect dertog the i j period June 1 to September 5. 1994. Particular emphasis was placed en  ;

the adeouacy of guidance available to operators for the conduct of plut 1 i evolutions. rescendtag te plant alarus. and detemistag when as i evolutten was covered by existis$ plant precedersa.

i Al-800, Conduct of Operattees, revisten fl. provided general goldence en

sperater respensthtittles and procedure esape. Al 500 secties 3.3.1.15 4

stated that operators vers respoestble for tattne steely and ,

j actions to essere safe operaties of the plant. Furthereere. -  !

J eostion 4.3.2.2.4 stated that annuncteter reopense precedoms shall be i used to diagnose alams not directly related to lateattenal sentpelatten of plant centrols and for aar alars the operators were est epitettly l fastilar with. No further guidance en timeliness la respeedtog to alarus er toplementing alars response operator actions uns provided ta Al 800. AI 800 sectica 4.1.35.1 stated that it was the duty of every member of the trystal River Unit 3 morkforce to strictly amore to written policies and to comply with precedores written for Crystal River Ustt 3. Neuever, Al 500 sostion 4.3.3.3.1.a stated that when the adeevacy of asistlag pmcedures was guesttened, shift supervisten seuld make the deterstaatten as to which procedural regstrements unre j applicable, a

i NOO-11. Implementatten of Technical sputftcatten, revisten 3 provided a

! mechaatstic process to determine when precedures more regMred for conducting activttles covered by Technical Spectftchtten A.S.t.1. This process did not provida guidance for detemtalog when precedures were poteattally landequate er when en activity was. net covered by entsttag precedures. Precedure N0012 was an opper tier precedere and was implemented by Al-4004.. Description and General Adotatstrattaa of Plant Procedures. Precedure Al 4004. rettsten 4. gave direcites se the generation process for nem procedures tectedlag any revleus that were reentred. However, as guidance sa when new precedures should be peserated was present.

e

4 i-tt b- .

i Procedure AI 400E. Performance and Transetttal of Procedures, provided a checklist to assist in deteretning if a precedere constituted a test er j

infreteent evoletten. Nomever, this checklist only would be esas if a new precedure had been generated and not to gotermine if an evolettaa

] helag conducted estag esisting procedures was a test er an infrequent j evoletten. He goldance was present to assist la determining if as eveletten betag connected oss a test er tafregeest evoletten.

Furthermore, the Itceasee did met define what activities constilsted a test'er esporteest. .

The inspectors discussed eith the Itceasse the tssess conceratog eben nos procedures would he developed. The licenses said there was no i

' specific gettance and it useld have been the shift sopervisers' judgensat to deteretas if nos or editttonal precedural psidence ens regeired. The iaspectors asked if the shtft sepervtears reestved any trotning that would assist them la asking this deterstaatten. The lleensee stated they retted et the shift supervisers' exportance and licensed operater tratalag.

. prerlded Precedure direction forA!.48tl. Procedere the content Writing and forest (Eacept for E0P/AP/VP)Specificall of plant .

the seidance for sparattas precedures stated parameters that ear jeopardtae egetpoest safety if amesaded steeld be tactaded in tas Instts i

j and precautions section of the precedure. precedure Op 3M Corte 8 ans a limit established for the purpose of protecting the high pressere

^

injectten puses free hydrepen gas letrosten. Centrary to the goldense ef Al-40tt, ne reference to Cerve 8 ans present ta the Itstts med precasttees sectlen of precedure W 408. Habasp and Purificatten System.

revisten Fl. AI 40tB did not provide goldence sa ese of plant serves referenced within precedores.

j Procedure Op.401 rettsten Fl. testractions for moraal aparetten of the makeup and portftsatten tecludtag 184 hydrogen gas addetten and water level changes. Yhts precedure uns presented to a sectiemal femat to alles for operater floatht11ty by teplementtog only the applicable section er secttees. nouever, a precastina er Insttetten given in one section did met apply to other sections of the preceders.

The only precautions and 1tettations that were appitcable thessghest the entire precedure were .these stven la the precastions and llettettees section. The precauttaa er liettatten for etR pressure uns presented only la the section for Itydrogen gas addtttes. Re further golduce for Ian pressere Itatt was given ta procedure Op-est incledtog louertog and raistag 98K water level.

l Precedure AR 403. PSA N Annunciater Response, revisten 11. Provided .

guidance for operater response ta a laK high pressere stars. The ala m was driven by a tempartsen of Ian level and pressure to a cespeter a19ertthe that aperatinated the IIK pressure limit (terve 8). 1he geldance for a valid alaen uns to essers IElp-141 and ISAF-148 ene sleead j and ta reduce pressere eithts the IRR presseru Itstt (Cerve 834. Be i

~

eeldasca for tteeltess: to lettista operater actises er the selbst for reductag IER pressere uns gives.

j:

1 l

1 1

4

i I

i i.

e 23 procedure Op 1038. Plant Operating Curves, provided the adotaistrative operating limits for aermal plant operations. This precedure did est ~

3 provide guidance en eermal plant operattens or en the appitcahtitty of

administrative estrating Itatts. la fact. precedure Op.1938 waresed i j enly to provide revisten/ control when adotatstrative operating j " itetts 1

were revised.

procedural guidance was met provided to assist the shift superviser in detersteing when esisttag pmcedures unre edeemste er if new precedures ~

were necessary. Forthermers, the tempacters concluded that l

l' Op withinsetthe did11mits not provide suffistaat of precedure sophasts Op 1835 Carveen 8. esistatalesAdditionally,presse es l 1

spectfic guidance for timeliness of respease to plant alares uns l l present. l l

The inspectors concluded that operating precedores, as written. were l

3 adeounte for operattens within the authertsed region of Curve 8. All nine of the examples of apparest violatten g5 11 01 involved hydrogen j

addition, for which precedures were clear la tavettag Cerve 8 Instts.

However there were weaknesses te written directions to operators regarding applicability and use of precedures.

j

! 5. Training j

The inspectors reytened the Itceased operator taittal and requalification training programs ta deterates if adequate trateleg i entsted.

The Itcensed operator regenlificatles progres uns presented over a tue-year perted. Mauever, est all precedures er systems were covered withts that tus. year perted. The inspectors determined that the AI-4N sortes precedures wers last covered la 1991 and the Makeup and portftcetten operating precedure was last covered in 1990. Procedure At e uns covered duttag the licensed operater requalificatten cycle prior to September 5. 1994.

The Itcessee supplemented the itsensed operator roguallftsatten program l with the Itcensed operator ragstred reedtog program. This program uns l vsed to keep operators suere of precedural changes. Estdance for the licensed eserator reestred reading progres coasisted of a process for developtog the reestred reedtog list and preesssing of sign off sheets.

However, ne formal directlen en the latest of the regstred readiet program was present. Furthermore. precedures that the operators used l reettnely were not always included as required reading. , l Tratatng on plant curves and edeteistrative limits was presented during initial license training. Reettaa plant operations, such as IIR -

hydrogen additten er water level chasses, were not specifically tecluded dering licensed operater regesilficatten trataing. TheIteesses asolained that covertag reattaa plant operatises was not seasidered as effective use of tratat ressurses lificaties because the operatore ausre feelltar traistag instead suphestand estas with such operattees.

1 1

J

' 14 the ahneraal and energency plant procedures to respond to plaat 4

transients. Ivalsattens of operater performance ears condected dortag observed real tles staulater tratatag. The licensee was using WEEE 1921. Deerster Licensing Esamlaer Standards, as the evalestive standard for acceptable operstar performance. M REE left was used by NRC enastners la deterstes if as tedividual demonstrated sufficient j kneeledge and abtitties to receive er estatate as operator's Itcense.

WREE 1021 evolusted operator performance la a vertety of areas includtag precedere use and amorence, alars response, and coepitance with adotatstrative itetts.

The inspectors caecinded the licensed tratalag program was adequate.

i Laployee Centere program 6.

i The inspector rettewed the ECp to determine the adammary of the program.

The inspecter rettesed IRB Massal procedure 155 35. Asvtsten 5. dated an. The

.  !!/11/94 to determine the centent and perpees of the i purpose of the progree is te call attenties to the reestruments j which prahthit discristnetten by FPC tts centracters, or seb.

! eentracters aestest as employee for espagtag la certain protected j activities. Tkts includes providing tas IAC tafermettaa atest possible j vtalattens of reestrements taposed ty the Atests Act er the a process to

! taarty Raurgentsatten Act. It esseres that esplepass

! supress caneerns or asks soggesttens withest fear of retrthetten er l discristantion. The progres establishes a prosess for decamenttee.

tavastigattag, and resolving the concerts. Additieselly, the progres asseres that exiting employees have as opportanity to ideettfy conceras.

The program contains a confidentiality provisten to protect the identity ef the tedividual peerldtag the coneers. This protecties is est l akseluta in that there are seus circumstances under eticA the identity

heard j

could he revealed. These taclude as order of the caert.

or statlar legal entity. Also, as ladividual is censidered to F if he/she acts ta a meaner l watved reaseaablytheexpected right to confidentiality to disclose his /her identity.

l The program requires notificaties of the tadtvides) of the reseletten to of his/her eencern. The program also allens the tadtvidual the attens appeal the resolution of the concern to the Strector. Neclear Sits Support for rettee of the caseers and reeslettee to deterstas if j further investigative acties is seesssary or desired.

The inseector discussed the program with the ECR and reviewed the records of the progren inclustag several conseras that were completed and others that mere st.111 active. This revies le 1995, tedicatedthe mester that la th past year, the prograireceived 30 to 40 concerns.has droppe this decrease la activity with, plant ansagement and the Ett and caeciaded that the reason for tais decrease ses the tepleenstettee of the procerser card system. This system uns tepleaseted to elles the Itcassee to identify concams prior to their rtstes to the level store a N - -. iv _ ____ __ _ _ . _ _ - _ _ _ = _ _ _ _ . _ _

_m__ _ _._ _ . _ . . _ .__.___ . _ _ _ _ _ _ _ . _ . _ _ . _ . _ _ _ . . ._

~

25 probles report was reestred to be generated. The precurser cards are reviewed at the daily plea of the day meeting sad are assigned to individuals for action. The cards are then tracked to closure, noth the 1tteesee and the inspector conciadad that this program was ani enhancement of their corrective acties program and that its wt ad teplementatten should have Pseused the muster of espleyse . The inspector concluded that the closed concerns had received proper atteattee and that the open sencores sore att11 active with long ters corrective actions ta progress.

The taspector revissed audits of'the ECp conducted in December INB and July 19M . The audit reselts indicated that tratatag of persemel en the ECP was effective. Seventy five percent of the individeals surveyed indicated the espected degree of fastlierity with the program. Eighty percent of the individeels surveyed indicated an adegaste onderstanding of management's espectattes as to when an ECp fers shoold be embattted.

In general, the servey tedicated that employees' generally were pleased with the propres and management's teamitaset to the progras. The laspector randesly latervleued personnel while eastta and arrived at essentially the sans conclestees as the audit reports.

7. Otspositten of Apparest violattens identified to NRC laspecttaa esport 50 302/95 13 The inspecties desmested to NBC IR 50-302/95-13 tdenttfled apperust vtalations involving the adegency of design control and corrective actt. ens.

Apparent violattes 50302/9513-02. example 1. described to 3 of 12 50 302/95 13 tavolved the fat)ere to translate the dost basis to ensure proper operattee of the Makeup and puriftsattes Systes, such that the systen 13 aatematica11r switched from its mersal operattag made to the emergency operattag made gh pressere la,1ectise and is capable of delivering water from the ate the reacter vesse , as as apparent violatten of 10 CFR 50. Appendts 8. Critertoe !!!. This apparent violattaa is adstatstratively closed and tacorporated lets .

apparent violattaa 54-308/95 21 04 esemple 1. l Apparent violatten 95-13 01. examples I and 3. deserthed la paragraph 4 of it 50 301/9511 tavolved the fat 1ere to meet the regelreensts of 10 CFR 50. Appendia 8. Critertes !!! and IV! for the laterts serves and the curves issued la Asvisten 13 to Op 1038. This apparest violation is adstatstratively closed and incorporated into apparent violatten 50302/9512-03, azaeple 2.

Apparent vtalatten 95 13-81 example 1. deserthed in paragraph 5 of it 50 301/95 13 tavelved the sanzal saepover from the tuST to the RB susp at a level of five feet er less la the BW5T. which is teseffletant to assure that all ef the ICC5 pumps useid not he damaged by vertestag.

This is an apparent stelattes of le CFR 58. Appendix B. Cetterten III.

Design Centre) for inadegoate design essenettens for hereted ester 4 .

J ea 9

~-- . .- - . . - . - . - .- - . . - - - - . -. --. .-. -

}..-

l 1

l J

26 storage tank swapever level. This apparent violatten is j l

e aanimistratively closed and tacarporated late apparent vtalation  ;

$0401/9512 04, example 1.

j 1

i Apparent vietatten 95 13 03. example 1. described in paragraph 6 of 12 50 302/95 13. tavelved the fatlers to toplement timely corrective l acties to review poteattal significant condittens adverse to esality l tavelving safety related tanks tacledtes.the OMET and FIT. which is a j violatten of the reestrements of 10 CFR St. Appendix 8. Criterica XV!'.

4 j

- This apparent violatten is adotatstratively closed and incorporated late i

apparent vietattaa 50 308/95-12 88, esemple 3.

Apparent vtalatten 95 13-03. sample 3. described la paragraph 6 of 1

1R 50 302/95 13 levolved the failure to translate the destga basis reestrements of the FST inte operattet precedores uhtch is a violation l

of le CFR 50. Appendis 8. Criterien !!!. This apparent violatten is administratively closed and incorporated late apparent violattaa

! 50 302/95 12 04, example 4.

't j Apparent violaties. 95-13 04, described in paraproph 7 of !R 50 302/95-i 13. tavolved inadequate destga centre) ta ensure adequate teventary to the RB sump to provide adeguate upSN to a LP! puup, with tAs IFI peup l secties crosstle valve opea, supplytes tus operettag NPI pups. This 1 lineup could result in the less of the only operable LP! peop. This was identified as an apparent violatten of 10 CFR 50. Appendts S.

Criterten !!! Design Centrol. This apparent violattee is administratively closed and incorporated tate apparent violatten

{ 50 308/95 11 04, asemple 3.

lt .

8. Exit laterview l

1 i

The taspection scope and findtags were sumartand on December 15. 1995 with these persons indicated in paragraph 1.1he taspectors described l the areas inspected and discessed in detall the taspection restits ,

i itsted beleu. proprietary taformatten is met contateed la this report. l l

133g ItenNumber1131g1 Descristian and Reference EEI 95 22 01 Open ' Nine esemples of operaties of the enheep task outside of acceptah)e operattag retten.(paragraph 1.1)

EE1 95 12 02 open Two examples of anaethertzed tests /

espertaants during which the plaat was

- aperated in a neeceaservative mammer outside the acceptable operattag region without a safety eva)matten. (paragraph t.1)

1 i

k i

27 j eel 95 t! 03 Open Three esemples of inademate corrective  !

i actten concerning inadewate revisten to '

j curve 8. Inadecoata rettews. inadegoate J

tank volumes. (paragraph 3.1) j EE! 95 22 N Osen t Four esamples of tsadequate desten control conceratag Cerve 8, varteus setpoints, and j ,

tank velenes., (peragraph 3.1)

! E11 M 13 81 Closed j Item closed by inserporating the example-inte 95 82 82.

t j

EE! 95 13 42 Closed Itas closed by incorporating the esemples l late 95 12 g3 and M.

l EEI. 95 13 83 Closed Itse closed by lacorporating the examples inte 95 2,2 03 and N.

j EE! 95 13 04 , Closed Itas closed by incorporating the asample 1 -

tote 95.!! 04.

j 8. Acronyms and lattialtsas

? 4 ONET terated vatar Storage Tank, ON Decay Neat e, 4 ECCI Emergency Core Coeling System ECP leployee Concem Program E3 Employee concern Representative EFW Emergency Feedwater top Energency Operating Precedere FpC Florida power Corporatten FSAR Final Safety Analysis Report NPI Nigh Pressure Injection LOCA Less Of Coelant Acetdent MAR Modificatten Action Roguest IllP Make up Pune

, Mlf Make up Tank NGO Muclear Opers"ons Department

'NP5H Not pe="' , w. tion Head PR Prob 1L ,' w;t R8 Reacter Wilding RC81 Reacter Coelant 81eed Tank RC5 Reacter Coolant System RfA Reevest for Engineering Assistance R0 Reactor Operator 3R0 5entor Reactor Operator e

Q Q 9 Page 4 of.4 C15tVE SA, 881.8C OPDATI0llAL CRITRIA De 8E515 LIMIT curve is based en Engineering Calculatten 2 6-0053. Aav. 2.

e m .. uR,1 curve ,s based e. i tr t er e, ,e, - a.d iewei . read from W-14-LIRI.

o h e M ACCEPTABLE RER10ll is es er above the OEIIM LMIT curve. Operstian in the IIIACCEPTABLE RERIM shall require an evaluaties is accordenen with CP-111

'lattiation and Processing of Precurser Cards and Probles Reports'.

e h AIBRBEIA19R AURI is a vaHable alam based as a Pressure /Velume relatieaship for the, makeup thak and lattiates at 3 psi helen the design limit curve at a level of 5 taches.

e De COIFUf5 ALARR is a variable alars based en a Presesre/ Volume relattenskie for the makeup tank and initiates at 4 psi helow the design iteit curve at a leve) of E taches. ,

e The ACCEPTABE RE*?'il is below the AIOREICIATOR ALAM above 'O' psig and amea=Pa= the PREFERRB RER1tII. It is bounded by the Les and Nigh Level alarus (55'&100")respectively. The C0lpUTER ALAM is provided to alert the operator that the makeup tank pressure is approaching the annunciator alarm. Escoedtog the computer alars does not require any ismediate acties.

Exceeding the AISEV'ATOR ALARIt will require operater actica in accordance with the appropMata Annunc ...or Respesse Precedure. This area is esasidered a M51RICTW RESIM.

w The PREFERRE RER30Il is an area below and to the right of the CIEWWIg aLasu er,g equal to er greater than 15 psig. This region should provide adeguate lodrepes concentration (1E5 cc/kg) in the RCE if askeup tank pressures are ceasistently maintained in this area.

o operation outside the PitEFERRB RER!tBI is acceptable duHag plant maneuveHagt however. un$er steady state conditions, efforts should be made to operate in the PREFERRE RES!011.

o Makeup Tank Pressure should be maintained above 0' psig during aermal operating conditions, and during venting evolutions. This will prevent damage to the IRIT during a drew-dema free a.destga basis accident.

o Exceeding the WT R!st and Lalt LEVEL alare setpoints (200*, 58') will reeuire action in accordance with the appropriate Altra Raeramem pesendure(s). These areas are REETRICTD RER10118.

N G

Sm OP-1038 Rev. 15 Pepe 34

7-i i

1 j .

sm=e numens emmenes aus i  ::: :: : * : :: :: :: : : :: : :: :'::: 1: 3 0 3 : ::: S si l 6 \ # .. i i i i i i

1 ' 4 A . _ g e 1 T  : y ,

e

\ \ _

. e i I

• I g

k 4 t _

V i T 11 ,

m. g t

. X \ l 6 .

'k l ei ,

I m

1 i i

\s i  ;

i \ t i .

I

, i 1 , \ h a 1 J ,1 ag  ;

e \ -

8 t \ h i ,

et

.# T 1 Ie I s" L 1

\t .. l. g

1. ing

• E I 3 ,

I k Wik l '

e i Mi .

l a

4 ke a L t

\ __ E \ '

I a,

k EI 't

• 1 ====

, g' ,

I '

4 \ E fl L' "I

e > I 4  ; e L ===i i g l

\ 4 o ,,,

g t , ,l i i  ! !

1.

I I i -

1 I 't \ . '

me i I 6 . t \ q

• I i ii 1, 1

.\ .

"I g I ,t & I

) f, i a '

0 g 4 e: i ei i:

/\ u ri *

• 1

\'

,. g 5E m as l it

'l I 3

Q iI i

6

( \ L t'

I I I N O a

~

jff i% e { *\ .g

- * ' i i

- 8 o

l. l j 2

i , e i e

, 1 ,

g o i i mum m

.i

' I ' ,,,.,u, i

• I \ __ g

. i ! .

, i s i I Ii i 1 3 .

• I git i g , a g g q

e. i g g I i 4 4 i jg 8

g i o i

i le , ii . L p (,

8 l.. i; == i y g j,  ! 'i l # 1 1 a

.  :  : 6 s - t et t e il 1, I

I

, Q8

• 1 I !

Ie I \

e ,

h , 1 .

E g I ;

• 6  ;  ; g g t ,

g I;

'

• l l uman g E .

t M

l ano m ammummann g 8 3 R R R S S = *

, x -

is <,ii ti 11 : a l

'  ?

l l'L

( k ' I t  % V .

x ,

s .

h. .

s 3 ;.. I E

1 \ .

( n

• i d\ <

T . I I f*

. r  : g i. g y 4 .

I

, I  % \ >

g r g ( ' 4 -

1

(  !

s i g

.a .I '

i 1 '.- e j q '. g s, -- 3' I ' ') l r ,

1 1 <.

')

i ( <

I ' t.

• L i

.g i =q __ 4. T 6 11 ,

a > .. \

H = T O

e

-l g '61 1 .

l ,..g~

a 't \

, . ,i, '. , i 3 . ,

] .i

- i is 1 . ,

. 1 1 1 Vs g i e ,

I ll a,l

g. , gl g .g ,

i i i ,- s a .-

.i , ., ,

l ,

i 4it i i i t # 1 i

- i i i . , a s g ,,

i. i i ,, e i e s v -

t t iii g ,.a /.i T t-a

, g j g  ! ili ,I l A: -

Y T

-a E l

M t \ ' *

. l

., a .

.i, i l g ,

, t. 1, n .

iii,i e4 ,

3 4 i,4 g 1 <

i t "

1 t .

,t4 i l

,r, i- is il a to l I tl ,'i s  ; I  :# g' 1 n

, , j i===. i is ii t t .

\. '

'I t  : ,  ! i  ! L i

h. ' I 16 1: , 't I i Illi 4 l s t i n

, e -

1 f \ \ '.

I il- t 1 1

,.i, ... . .. . ,

'i

,6 ii,

i. t- 1[ It

\ g l

i' 3

7 l  ; . 1 i l

• 8

_.