ML20126F282

Forwards Description of Valves Affected by Stroke Testing & Effect of Valves Being Opened or Closed During Testing,W/ Respect to Encl LER 369/90-022-1 Re RHR Sys Being Inoperable

ML20126F282
Person / Time
Site:	McGuire
Issue date:	12/24/1992
From:	Rosenthal J NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:	Norberg J Office of Nuclear Reactor Regulation
References
NUDOCS 9212300209
Download: ML20126F282 (3)

Text

,*.

]

. , I 6m as g,,. ,

M *

1. *o

• UNITrD STATES 8' 7. NUCLEAR HEGULATORY COMMISSION 1; .t WALHING TON. D. C 20$55 DEC 2 41992

(**..*/

l MEMORANDUM FOR: James A. Norberg, Chief l Mechanical Engineering Branch Division of Engineering Technology Office of Nuclear Reactor Regulation 1

FROM: Jack E. Rosenthal, Chief Reactor Operations Analysis Branch Division of Safety Programs Office for Analysis and Evaluation i of Operational Data SUlUECT: RESIDUAL llEAT REMOVAL SYSTEM INOPERABLE The licensee event report (LER) 369/90-0221 (enclosed) discusses valve stroke testing affecting the emergency core cooling system (ECCS) at McGuire Unit 1. A description of these valves and the effect of their being opened or closed during testing are shown in Table 1. 1 The licensee has determined that testing (and consequent changing of vnive position) of these valves places parts of the ECCS in an inoperable condition. Therefore, the _

licensee has concluded that the valves will not be tested during plant conditions under which the systems are required to be operational. The immediate corrective action was to cease quarterly valve stroke testing and invoke the ASME Code,Section XI, IWV 3412 which states " valves that cannot be esercised during plant operation shall be specifically identified by the owner and shall be full stroke exercised during cold shutdown."

Thus, the valves will only be tested during plant shutdown conditions. This change will probably result in valve stroke testing every 12 to 18 months rather than the current

- quarterly testing in accordance with the ASME Code and Generie Letter 89-04. The licensee has assessed the safety significance of the ECCS inoperability due to the valve testing and it appears to be very low. - llowever, the licensee did not mention any associated increase in risk due to possible valve inoperability as a result of the longer test intervals.

While the licensee is to be commended for evaluating the risk due to these. valve tests, we question the prudence of not evaluating the risk associated with increased testing interval and consequent decrease in test frequency, p, p ,

s y

4ggpfy fi ,, .~ s W 4-9212300209 921224 f{'

PDR ADOCK 05000369 0 S- pog _

. ' James A. Norberg 2-It should be pointed out, that based on review of sketches included in the LER, and I statements made by the licensee in the LER, the valve testing would only result in i degradation of one train of ECCS. This is allowed by the technical specifications (TS). l The TS action statement states that with one train of ECCS inoperable, the inoperable train must be returned to operable status within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, or the unit must be shutdown to Mode 4, The testing lasts much less that 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. Thus, as we understand the LER, the TS would clearly allow the testing.

Also, we are concerned that other licensees may adopt this extended test interval. The result may be that a substantial number of valves will receive reduced testing with possible consequent lower reliability, Table 1 provides a list of the valves discued in the LER, and the effect of the valve testing on the ECCS system.

Please contact William Jones of my staff at (301) 492 4442 if you have additional questions.

OriEina' signed tirlack E RosenlNal Jack E. Rosenthal, Chief Reactor Operations Analysis 11 ranch Division of Safety Programs Olfice for Analysis and Evaluation '

of Operational Data

Enclosure:

As stated Distribution:

PDR Central Filei ROAD R/F llOrnstein WJones EBrown JRosenthal TNovak

- VBenaroya EJordan Dross GMillman, RES =.

MVagins, RES i y)lh'caf k ROAUg#f ROAH ROAB pA @ROAB-WJones v IlOrnstein EBrown -

JRosenthal 12/p/92 12/p/92 12/p/92 12/jr92 .

-J

l TA13LE 1. VALVES AFFECTED llY TESTING VAINE LIST t

NI13611 RilR llX 11 TO S1 Pump 11 l NS 3811 Rllit Pump 13 to containment spray nozzles containment isolation NS-43A RilR Pump A to containment spray nozzles containment isolution - .

NI 103A S1 Pump 1A Suction  ;

NI-11511 S1 Pump 1 A Mini Flow Line Isolation NI33411 51 Pump suction cross to CVCS isolation EFFECT OF TESTING If any of the first three valves (NI 13613, NS 3813, NS-43A) are tested and are open during a large break LOCA and could not be closed, the low pressure safety -

injection flow (via the RIIR system) could be degmded to less than the flow ,

assumed for the FSAR analysis. The licensee states that only one train is affected .

due to testing only one train at a time. The valves are normally closed. Cycling - i the valves open would reduce the flow injected to the RCS.  ;

if NI 103A (SI A pump suction valve) fails in the closed position (simultancotis testing closure, loss of power, and small break LOCA), SI Train 13 would be ,

inoperable.

If NI 11511 (recirculation line for low flow for pump 1A) fails in the closed position (closed during testing, loss of power and LOCA), Train A of SI would be inoperable. >

Failure of NI 334B (SI pump suction cross to CVCS)'in a closed position

. (simultaneous testing closure, loss of power, and large break LOCA) would result - >

in Train A of SI being inoperable during recirculation phase because the flow .

path from Train A of RilR is through valve NI 334B.

a u

L

~

{ -_ . _. _~ _-

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

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

,' e ,

. 4 1

LICENSEX RVENT REPORT (LER) ,

t <

• stttf k W (1)

• coarrr w m mam(2) rnorr n "h Guire fluclear St.ation, Unit 1 .

05000 369 1 ore 11i tate) Bot.h Tralnn Of The Residual Iloat Renwaval Systcan Woro Inoperablo During Quartotly Va.lvo roko TlJuo Tooting Docauno Of Isnpropor Schw1uling. 'Y

_ rvyn omtsi _ pn mocmarsi r m rt t e r7i erntra rAert.ttrr.n twvntwnras -

tu car stAm rEAA sinvEntA1. trvtstan ammi tar rum JActs.rty emr.s . rmer ma<nratai l

wmarm womura Hecuirn, (fnit 2 05000 370 i 7 16 90 90 022 01 09 17 90 05000 mATims 1 tura rmer is ,mtvrm evemAn so rwrmerra ce so ra tesrt me or enre nr tbe ra11~1*ritui

• (*) to.402(b) 20.405(e) 50.79(a)(2)(iv) 13.11(b) n 100 _ to.405(e)(1)(1) so.36(c)(1) 60.7)(a)(2)(v) 73.71(c) 11.(1 0) 20.4oS(a)(1)(11) So.36(c)(2) X So.73(a)(2)(vil) oruta

. (specifyac :in a,

h-> yyy[;gg. - so.40s(a)(1)(111) 2 so.73(a)(2)(1) so.73(a)(2)(viL1)( A)

@@y@i- $ to.405(a)(astiv) So.73(a)(2)(t!) 50.73(a)(2)(viii)(a)

/[ f h r * / Y '" ..o.40$(s)(1)(v) to.7)(a)(2)(111) So.13(a)(2)(a) 1.lentag norrAc7 M2H tuta tw[3a) a ver.mrms wisoren rry L. Podoroon, Supervisor, McGuiro Safety Roview Group AxxA caos 704 075-4407 n ewtAta (su !.lu ha< t.Anap m puxi F Alwn ti,.t.cy t hu) In tuta Prauvtt u) -

i? r MAN 17AcrVPIR RErORTABLE

(? %f t 1st Ors *n34 CUNPCFDf7 MANUrAC'IVPIR REPORTABLE . CAUSE ETSTD4 CCTGVNENT To kPics W* 19 kPPDS '5W

d a h$d

!dC Nh strrrtmrrrst. PrPoser rwerten(get EXPECTED W8m1 DAY YEAD SUDMISSION YES (If Dee. ocuplete 12PECTED SURM1&SION DA"E) X l 80' DAfr(15)

.TitACT (t.tsit to looo spaces, i.e. epproxianat.ely (af teen eingle-shace tytevritten lines (16)

On July 11,1990, Performanco (PitF) porponnel prepared a reicoue of proceduro pT/2/A/4206/02D, safety Injection (NI) Train B Valvo Stroko Timing Quarterly. On July 16, 1990. PRF Engincor A portormed the normal review of the proceduro ruincue. During the courno of the review, .PRF Engin a177 in onjunction with Operations pornonnol discovered thatwycling*valvoa ly,an p 2 NI-13 6ft,, acidual float RemovalHND)bHopt p g5xchangor 18 and ::D To -NIs Pump ylD,and D,gle at power could degrado ND oyotom operation in the event of a grggBreak-Inso!Of" Coolant *Accidentg(1DLOQA)/. 4 Do' gradation could occur when the valves wgro cyclod to the open position'. Thono valvoa have boon routinely otroke time testod in Hodon 1 (Power operation), 2 (Startup), 3 (flot Shutdown), and 4-(llot Standby) ainco plant otartup for each unit, respectively.

/Silbtroqu ntly, it was datormined that the name nituation was also true for valves 1 and and 1 and Qi{j < (D Pump B To Contaisunent Spray (NS) Nozzlos Containment Isolation, S-43, ND Pump A to NS Nozzloo containment Ioolation, during performanco of proceduren P't/1 a/nd 2/A/4200/02A and D, NS Trains A and D Valve Stroko Timing Quarterlies. This ,

incident in assigned a cause of Improper Scheduling of the valvo otroke timing for theco valves. Both units wer e in Hodo 1 a': 100 porcent power at the time of the event diocovery. Appropriato ochedulo changen have boon implomonted to ensure that the valvon will be stroke time tested only when the unito are shutdown.

~

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

3

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

3 . . i

n. ma us. ucums vuao.m m UCENSEE EVENT REPOHT (LER) TEXT CONTINUATION maono o. m is * '

. samtsimum .,

4. net 4 LAM 4 Of pucet,y ,evees,a pl gg, ,,yugg g ggy , A,, g,

"^a "Po!?.@,. S  %?.W . -

~

McGuire Nuclear Station, Unit 1 olsgojolol3l6l9 9l0 0l2l2 0l1 l2 or l' l1 ,

nw .. w .uo mcr w.mn ,

EVALUATION:

Baekground

• The primary function of the eResi'ddalMicat Removal Syst'eW(HD) ll!IIS DP) is t.o remove heat energy irom the Reactor [EIIS RCT) Core and the Reat: tor Coolant System (NC) [EIIStAD) during plant cooldown and refueling operations. The system is also ut.ilized as a part of the Emergency Cue' Cooling System (ECCS). It. has a secondary function of transferring refueling water between the storage tank [EIIS:TK) and the refueling cavity at the beginning and end of refueling operat. ions.

The ayatem has two parallel flow paths sharing a common inlet from the Containment Sump,'the Refueling Water Storage Tank, or Reactor Coolant Loop C (EIIStPSP). The return linea connect to the cold leg of each reactor coolant loop. Each flow path containn an ND pump [E1IS P), a heat exchanger (l!X)

[EIIS HX), associated piping, valves [EIIS:V), and instrumentation required for operational control. During nonnal operation, the system in not in service but is aligned in readiness for operation as a part of the ECCS.

Ref erence drawing page 6 of 6. -

Technical Specifications (TS) 3.5.2 and 3.5.3 specify the requirements for the operation of the ND Pumps with respect to their ECCS functicn. Both pumps are required to be operable in Modes 1, 2, and 3. During operation in Hodes 1, 2, and 3, when one train of ECCS becomes inoperable, the inoperabic train must be returned to operable status withta 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, or the unit must be shutdown to Mode 4. If both trains ,of ECCS are inoperable, TS 3.0.3 applies, and the unit must be shut down to Hode 5 (Cold Shutdown) unless at least one train is made operable within an hour. TS 3.0.3 states that. when-a Limiting Condit. ion of Operation and its associated action statenents are not met within one hour, action must be initiated to place the affeeted unit in a mode in which the specificatic'2 does not apply.

Description of Event On July 11, 1990, Performance (PRF) personnel prepared a reissue of procedure PT/2/A/4206/02B, Safety Injection System (NI) [EIIS:BQ) Train B Valve Stroke \ ,

Timing Quarterly. The reissue was done to incorporate timing requirements as specifiedbypenericJ,cgeg9,-34f On July 16, 1990, PRF Engineer A began performing the normal review of the Nl') .: sO) U reissued procedure. During the course of the review, PRF Engineer A (% j discovered the. procedure required valve ENhr.136BT ND l!X 2B To NI pump ,2B, be MW cycled to the open positior.. lie t. hen consulted witT0perations personnel about. possible consequences of cycling t.his valve. Upon investigation, i t.

was deteruined -tQa,tgygling 7.this valve tosthetopen posgion while.atepowerNI m coul d "d F sWd FtWDWys r tt; moi pj eAtto n rflowrinathe r gyqq typftsacLa r ggeB tea keLo s sdf g.; Coolant Accident, (LBLOCAbtTThis was also true for puit 1 procedure ,

PT/1/A/4206/02B, HI Train B Valve Stroke Timing Quarterly, and the corresponding Unit 1. valve.

-n

. -_ .- _ _ - . _ _ ., 'Y D. " -

3

m .-- _ ____ -

.m... _

, i l

us.esucuanco m on, -

ar n., -

LICENSEE EVENT HEPORT (LER) TEXT CONTINUATION maonoo ui 4m

- emus; amm txscat, samenta 60 og a ,etaset a (el pagg ga

scou t f 884884,498

• I t.aa l%. sag 6pL y ae McCuire Nuclear Station, Unit 1 o js lo lo lo l3 l 6l9 9l0 -

0l2 l2 -

0 l1 l3 or 1 lh w - = w .--me w amm on ,

PRF Engineer A deterinined that these valves have routinely been stroke time .

. tested during Hodes 1, 2, 3 and 4 since R1 ant.startupefor eachtutii t W I

• !respectively.- Ne then initiated Problem Investigstion Report (PIR) 0-H90-0180 to resolve the problem. He also initiated appropriate procedure changes to ensure t. hat these valyco would only be stroke time tested during cold shutdown conditions.

In addition, on July-16, 1990, Design Engineering (DE) personnel began an evaluation to determine past operability of the ND system. During the course of their evaluation, DE personnel discovered that the same situation was also true for valves 1 and-2NS-38B, ND Pump B Discharge to NS Nozzles containtoent Isolation, and valves 1 and 2NS-43A, ND Pump A Discharge To NS Nozzles Containment Isolation. They informed PRF Engineer A of this finding and he initiated appropriate changes to procedures PT/1 and 2/A/4208/02A and B.

This will ensure that tliese valves will only be stroke time tested during cold shutdown condit. ions.

Conclusica -

This event is assigned a root cause of Improper, Scheduling by PRF personnal of the valve stroke timing for valves 1 and 2NI-136B, I and 2NS-38B, and 1 and 2NS-43A. These valves have routinely been stroke time tested in Modes 1, 2, 3, and 4 since plant startup for each unit, respectively.

It has been determined that the potential exists for the ND-injection flow to be degraded if a LBLOCA occurred while these valves are open during-testing.

Since the valves have been tested quarterly per the Inservice Valve Testing (IW) Program, the ND system was technfcally inoperable when past testing was executed during Modes 1, 2, 3, and 4.

.PRF personnel involved in develol.ing the procedures in question had not previously considered that cycling these valves at power could possibly degrade ND system pperation during a LBLOCA. Therefore, the valves were scheduled for quarterly testing as required in accordance with IW program guidelines. As soon as PRF personnel became aware of the problem, l-j appropriate actions were implemented to ensure that these valves would be '

I= scheduled for testing only when the units are in cold shutdown.

A review of the Operating Experience Program data base for the past:

twenty-four months prior to.this event revealed no events involving TS

! '- violat. ions because of Improper Scheduling and specifically there were no events involving valve stroke.t.ime testing. Therefore, this event is not etensidered to be recurring.

.This event is not Nuclear Plant Reliability Data System (NPRDS) reportable.

There were no personnel injuries, radiation overexposures, or uncontrolled

l. releases of radioactive material as a result of this event.

L p +

1

.....m,i,............%

L ,, ,,,,

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

eine e W -

. . . -, .i . .  : a. . . u auct.s4a at ut.at , x UCENSEE EVENT REPORT (LER) TEXT CONTINUATION A=odo o .eu fJ'.i"%

, _ tencs4 niaues ',a ,,

7Acadt,feArt(se cioca 1 somta in tta avunasta tel , e..e gs u.. '

an,:,p,,*' Q:; ,p

~ . .a McGuire Nuclear Station, Unit 1, o l6 jo jo lo l3 l 6l9 9lo -

0l2l2 -

0 l1 l4 or 1 jl,'

b <<,,,,- e eo.< .m a m r-suwann CORRECTIVE ACTIONS:

Immediate None Subsequent: 1) PRF personnel made changes to remove valves'I and 2NI-136B from ptocedures PT/1 and 2/A/4206/02D, NI Train N D Valve Stroke Timing Quarterly, and a'dd them to f

k]hg.

procedures PT/1 and 2/A/4206/03B, NI Train H Valve Stroke Timing Shutdown, respectively. l z,

g\h) 402 DE personnel performed an evaluation to determine past S operability of the ND system, and discovered that valves 1 and 2NS-38 and 1 and 2NS-43 could have also degraded (b (ch operability of the ND system durir.g testing.

{ j

3) IW personnel made changes to remove valves 1 and 2NS 38 '

and 1 and 2NS-43 from procedures PT/1 and 2/A/4208/02A Q0 Y and B, NS Trains A and B Valve' Stroke Timing Quarterlies, and add them to procedures PT/1 and 2/A/4208/03A' and B,

_( g kk NS Trains A and B Valve Stroke Timing Shutdown, respectively.

b Planned: PRF personnel will initiate a Station Problem Report dN requesting e review to be perfonned by DE to identify -if other valves in the IW program have previously unidentified

$ limitations on allowed plant ' mode for_ valve stroke time-Y testing.

hSAFETYANALYSIS: .

w

_ b If a LBLOCA occurs while valve 1 or 2NI-136D,1 or 2NS-38, or 1 or 2NS-43 are V opeu during testing, the ND system injection ' flow could be degraded to less than the flow assumed for the Final Safety Analysis Report (FSAR) LOCA-Analysis. These valves have been-tested. Quarterly in Modes 1, 2, 3, and 4

. k since Start.up of each unit respectively. Therefore, the ND system was f technically inoperable when 'past. testing was executed each time any of- thesp R valves were in the open position..

4 The time period in question varied from test to test and valve to valve.

t However, this time would have, in each case, been very short. The actual stroke time of each valve is 5 10 acconds. The time to verify the valve position would have varied from test to test, and then the valves would have- g q had to be cycled closed. .The probability of one of the valves being tested - jr and a LBLOCA occurring together is extremely low. Also, since the-valves- .

- A t could be closed.in_an extremely short time (if the need arose) it can'be

~

p- concluded that no significaut- safety concerns resulted because-.of the low g

kprobabilityoftwosuchevents-occurringsimultaneously,

! In the event that _a LDLOCA did occur, Operations personnel would;have .been l/ - able to close these valves as necessary to mitigate any decrease in _ injection

.se ,,,.. . ,o m c o. i n m e.w e.ees =

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

e. 4 . us.ascteaau m , ,e

• • " , ,t.fCENSEE EVENT REPORT (LER) TEXT CONTINUATION nenonoco.m ,ui .

imacs enu

~

sac.4 sty esAme_tte Doctti eAmesta til tta sevnesta 44 cAosin staa "2rl.';l'." *j Mag licGuire Nuclear Station, Unit 1 o ls l o j o j o l 3 ] 6l 9 9l 0 -

0l 2l 2 -

0l11 l5 or 1l1 natua - s m m m a- sw. inn a

flow. Operations personnel are in cont.rol of the valves during testing and only one train is test.ed at one time. In addition, the other ECCS systems vould have been unaffected by the valves being open and should have been able to perfonn their safety function in addition to the flow that would have been provided by the ND system.

The health and safety of the public were not affected by this incident.

During the time period when both trains of ND vere technically inoperable no events occurred which required the actuation of ECCS.

O k

7,; , ,o, .'i.. . . cro, n e s m . m cm e

.  ;, e Tac co,m uu

"" ut x.uu uwm, a_...._

LICENSEE EVENT REPOllT (LEll) TEXT CONTINUATION Ae novgoo, . .

Exe ats; eauw .

Facatsif esAnst til NEII N# 8 III tia es* 88R 84J pagggg m- -

":a t't'  ;'J.a,y:

McGuire Nuclear 11tation, Unit 1 't t LRT de'ans., esees 6 ::C . 4 ese : ' ~- Amc Fewee JdtL4 hl(tF1 o l5 lo lc {o l 3l6 l 9l9l0 ,_

0 l 2l 2 - O g1 j6 or 1 j t-e GS O 90 f*  %$ U O O to el 80 f* *. O E

,, p 17 1'1 9' *

=

y

'y s .s -- -' - -

4 w' m .--.y fa Q p I

f+ #

1 Ij} } sb&l%# # {  ?~Cl!.

se dey- 'ge 44 .,,ae

_ rFNU . ..

g ,, g --  :

., t it1

-f t 2

'!.J v 9.kh

,ps +9 -0. .

y c..,9" .9 -  : 3.

,v g . _( . '~}'t ~ .- -

@4 "p g

ta-0 4 ,,

ex-0. ..f. . .; n,g i x83

+ ..; s: ~ '

4 tu+

[ '

h) ki" p

Z '8 ld' (f _

b 2 0 3 Y

I EX. 0

=.

- (-A

~y .P-

p'l m

~ f 3"-0

"M-<3

< 2 -

5S .

Ep a.

ys Ep .. '

EMT Sl=

~

5 T --

8 x T- @, D.

"' l- p'".-

=

N Bi a

l F59 s3ai n

21 .

m .

mR o"

.l et; a k B

-w r-

.. .. . . - . ~ . ~ . . - _ _ .  ;

.- i I.Iccussg EVENT REPORT (IER) vnly grnrrINUATION .,

pocErr resta(*) n er.m t,p raccm -

{

i w:ss.rvr see;m ,

' TEAA ESQUDf7131 kEvtsloot i truksen te.Marn i

05000 369 90 022 01 7 m il Guire f"uclear Station s -

+ .

Additional Information:

' On Ccptemi,er 20, 1990, Station Problem Report (SPR) number 3710 was issued in rosponse to the -

The SPR requested-the  ;

pican:d correctivo action yonorated in roviolon 0 of thin report.

[

performance of a doelgn atudy to identify valves wLthin the ECCS which have the potential to degrada safety related systems. The responsibilit/ of performing this utudy was assigned to thaisyctem Engineering Group.

• Ac c- result of the study, a memorandum was inauad on Auguot 25, 1992, which identified 3

- vciv00 in the EccS which have the potential to det rade the NI and chemieni and volume control

_(NV) cystomo (EIISsCD] when quarterly valvo_hter4ko time testing La porformed. ,

Tha vcivos involved ares 1 and 2NI-103A, Safety Injettion Pump 1A Suction.

f p 1) 2)

1 and 2NI-1158, Safety Injriction Pump 1A Mini-Flow Line Isolation.-

p 3) 1 and 2H1-3348, safety infection Pump Suction Croon To NV Isolation.- '

Cystem Engineering porponnot datortnined wh6n valvo NI-103A la cycled to the closed position, During the inoperablo.

c= loos of Trairl A power would rondor Train D of the NV ' system ~

Rocirculation Phase of a Safety Injectiors, tho llowpath of ND Train B to the NV pump muction is through valvo NI-103A.

~

Wh2n valvo .NI-135B in cycled to the closed poultion, a loss of Train B power would hender ~

Trcin A of tho-HI syntom inoperable. When a Safety _ Injection signal han occurred, and tho-NC cyctem pressure is above the shutoff head for the NI pump,'the_NI pump would not have a

--flowpath, since valva NI-115D is in the flowpath for the 'miniflow protection of the Train A, _

HI pump._ .

i

- Wien valvo'HI-334B in cycled to the closet puttion, a losa_of Trein h power would render 'I Trcin A of NI inoperable. During the Rocirculation Phana of a' Saf ety Injectioni- the flovpath cf Train A of HD to t'ho HI pump suction in through' valve Hi-334. (Reference p 10 of:10).

- Tha immodlato correctivo action taken to provent degradation of the involved systems was the -

Atnerican Socloty of caccction of the quarterly valvo stroke testing on the af fected valves.

Hochtnical Engineers (ASHE) code articio_IWV 3412 states, "valvos that cannot-be exercised

~during plant operation shall be epocifically identified by the owner and shall be full stroko exercised during cold' shutdown.e." On August 25,' 1992,- Syntom Engincoring personnel lesued

150,-

L c enemorandum to the af f acted station groups requesting valves 1.and 2NI-103A,1 and 2NI-1 cnd l' and 2NI-334B be removed from quarterly valvo stroke testing and advised v that' the valves

. i.

"

• w .aN l ~ ,

---

=-%.

-+*4 ey ,* * < +, --,e--w,e, +- e,-,-- u.- .wa

I

%~

re- e I.ICICNSEK EVKNT ItKPoltT (I.KR) Tit 1T CONTINUATION .

ACILITI SAME(1) . IUCMT MHREP(3) 11R WWGD(&) l'ACli( 3 )

rtAm asgutartAL nytstom muern mm .m JGui.ro Nucinar station 05000 369 90 022 01 g ca 11 would be moved to the cold shutdown List for valvo otroko timing. Syntom Engineering porconnot also updated the Hecuiro Pump and valvo Incorvico Testing (IWV) Hanual to reficct thic chango. Due to the similarit y of tha ECCSa, McGuiro Nuclear Station (HNS) Dystem Cngintering personnel chared the renulta of tho doelgn otudy with Catawba Nuclear Station j t

(CNS) Engineering porponnol.

In cddition to the NI and NV oyotemn experiencing potential degradod scenarios during valvo ctroke tooting, the potential also existed for the came ayatom degradation during the performance of inaintenance on thono valven. Valvon 1 and 2NI-147A, HI Pumpu Hini-Flow Noader To FW (refueling water storage tank), which f all into thin category, woro addod to the original 3 valvoa. When valve NI-147A in closed for maintenanco, a loss of Train A power would rondor Trainn A and D of the NI ayatom inoperablo. During the Injection Phaue of a safsty Injection, the miniflow path for the NI pumpa in through valvo NI-147A. As part of ths Eccs, it iu unlikely thin valve would be clonod for maintenanco during Hodo 1 through Hoda 4 an thle valvo la TS related and required to be opened with control power f rom the valve oporator romcyod.

On Augunt 27, 1992, Operationn Hanagement icoued Operationu Hanagemet.t Proceduro (OMP),

c:ction 2-1, Attechment 3, Operations special Order number 92-23 which explained the potential problem, and linted the affected valven. The Special Order requented that the volven were not to be reponitioned unloon directed by an approved emergoney or abnortaal proceduro (EP or AP) . If there were any questiono, unit ctaf f poroonal were to be contacted.

e Additionally, an extenalve work history revio.4 was conducted on Unita 1 and 2 by Safety hsview Group poroonnel to datormino if thoro was any past proventive maintenanco (PH) or correctivo maintenanco performod on the'affeued valvoo that would have caunod a potential d:gredation of the unfoty tolated cystems within the ECCS. The review covered all work ecqusato acoociated with tho valven from Auguot, 1900, to the proront. The reaulto of.the review revealed that the majority of work performed was dono during unit outagon. *fho work performed during non outago intervala revealed work performod on valves 1NI-103A and 2N1-103A which had the potential to degrado cafoty related nyotorno. On September 21,198C, under work requsat 60374, valvo 1NI-103A was cycled closed as part of the Motor Oporated Valve Analyclo cnd Tooting Syntom (HOVATS), and on October 3, 1988, under work requent 135461, valvo 2NI-103A wan cycled closed twice, an part of work performed on the 1.47 Bypana Panel.

A risk potential annosoment van performod on valves NI-103A, NI-ilta, and NI-3348, by the scfety Analysis Accident croup. A f ailure modo for valvo NI-103A 10 included in the McGuiro Probabilistic Rick Assosament (PRA) model, which corrooponds to a valvo unavailability of opproximately 1.3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> por year. During the year, the valvo is closed for < 0 minutos per ysar an a result of testing, which in 1000 than, the PRA accumed unavailability.

i

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

. LICENSKH EVENT REPORT (LER) TKIT CONTINUATION .

wittrr sA,agt) 5 toctrr mHauts) '

.tn wutents i Fact (s) tzAn sinutarrtAs, nyssion F1MArn M Mrem 05000 369 90 022 01 9 or 11 1 Cuira Nuclear Station Additionally, valvon 1 and 2NI-103A woro cycled closed for short periodo of timo during correctivo maintenanco, in 1900. However, the Gal valvo unavailability for 1998, io still I r e-than the yearly PRA accumed unavailability. The core molt cut not (the combination of dif ferent f ailure modbo 'of dif ferent piocco of equipment) annociated with thin f ailure modo .

i

.- s his a f requency not greator than 1.0 E-0 por year.

valvo HI-115n has boon qualitatively diomioned ao inconocquential in the McGuiro PRA because if the NC oyotom in ebove tiio shutof f head of the NI putop for Safety Injection coquences, it to assumed the NV pump in operating ouecoonfully. The added redundancy of the NI oyotum in not nococcary to avert coro damago. If the cocondary oldo host oink la loot, a blood path muet bo octablichod before the prDoary aido proncare rican above the NI pump chutoff head.

If thin to not done, the reouit would be a failure anyway. For the abovo roanone, the NI niniflow path la not critical to averting coro damage.

. Valve NI-334D la not included in the McGuiro PRA coro molt cut not, however, the convoon cauco failure of valvon 1 and 2 NI-332h and HI-333B (!1V and HI Pumpo Suction Croncovoro), dooo cppone in the model. The f alluro modo which directly correspondo to the unavailability of valva NI-334B, lo equivalent to 1.7 houro por year. During the year, the valvo 10 closed for

< 48 minutos por year au a result of tooting, which la loan than the PRA acousnSd unavailability. The coro molt cut not annociated with the f ailure modo hao a frequoney not greator than 1.0 E-0 por year.

'It can be conclu ed that the unavailability of'the subject valvoa due to tooting dooo not contributo significantly to coro dranago riek, and dooo not contributo significantly to ECCS unavailability.

It han aloo boon datormined by Operations, System Engineering, and Safety Review Group paraonnel that if in fact the event did prenont itoolf, titare are emergency operating procedurco, EP/1,2/A/5000/02, High Energy Lino areak Incido Containment, and EP/1,2/5000/06, Loss of Emergency coolant Recirculation, which direct OPS personnel of the stopo to take in order to raitigate the circumstancos. '

For futuro reference, Hochanical Engineering personnel will revice the McGuiro Deuign Bacio Document (DDD) to include information relating to this design study.

System Engineering personnot will also perform a dooign otudy on the Auxiliary Foodwater (CA)

[EIIStBA), Main Feedwater (CF) (EIIS SJ), Diosol Generator Engine Fuel Oil (FD) [EIIS DC),

Component Cooling (KC) (EIIS:CC), Dlosol Generator Engino Cooling Water (KD) [EIIS LB),

Diocol Conorator Engine Lubo Oil (LD) [EIIS LA], Nuclear Servico Water (RN)~(EIIS BI), Main Steam Supply to Auxiliary Equipnent (SA) (EIIS SA), Control Area Heating Vontilation and Air Conditioning (.VC ) [EIIS VI], Annulus Vontilation (VE) (EIIStVD), Containment Air.Roturn

~7" _- .

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

I

,,,_.___7- _ - _

LICENCE!! RVRNT I!EI' ORT (LIIH) TEXT CONTINUATION m L m m (al , ten wumata(s) sta vmen ) Fact m Ytu str,p.rraat m isnom wtMarm ettEntR ,

cuiro Nuclear station 05000 369 90 022 01 go w 11 Eachango and Ilydrogon skirnmor (VX) [EIIss!!D), and chilled Water (YC) (EIIS KHj oyotoma to '

dstcraino if valvon in thono oyotoma have the potential to degrado cafoty related syntomo whsn otroke timing and/or mainterisinco activition are perform <xi on the valves during modou they are required to be operabic.

ADDITIONA1. CORRECTIVE ACTIONSt fannodicto s systern Engincoring personnel issued a memo to af fected station personnel to immodlately coano quarterly otroking of the subject valvoo.

Subroquent: 1) system Engincoring poroonnel updated the Unit 1 IWV Hanual to reflect the chango in the valvo otroko timing tent frequency.

2) HNS Sydtom Engincoring personnel notified C!lS Engineering personnel of the resulta of the design otudy.

Plenned: 1) A donign otudy will be performed on the CA, CF, FD, KC, KD, LD, RN, '

SA, VC, VE, VX , and Yc systoma by System Engincoring peroonnel to datormino if valvoo in those oyotomo have the potential to degrado cafoty related eyotemo when otroke timing and/or maintenance activition are performed on the valvoa during rooden they are required to be operable.

2) For futuro reference, Hechanical Engineering personnel will rovion-the McGuiro Dt3D to include information related to the doelgn study.

3) Syntom Engincoring poroonnel will updato the Unit 2 IWV Hanual to reflect the change in the valve otroke timing tout frequoney, during the next .ubmittal to the NRC.

s

4) Safety Review Group Managomont poroonnel will broadcast the results of the donign study through the Nuclear Network System. g 3

pr\fd $ $ Y( b 9 ww

$ s$

pvRe day'1 &*

y ,,

frons) a ush -g Kp,r- NR- go h r nkp g s l M? g w n ',

g ,,J E 5, b r. } M}t3 4 &d

- (l% ~ (W A lfE

' [' } f JS1 -

fo [

~

[k) vf 6VP If [ b('U" m . t. m v% sal

~

~

t. ,. 3 .

~

IA CENS!!ft ICVENT IUll'OltT (E.Elt) 'TKKT CONTfttUATION v:ss.itr am( 3 .

- soc,rr insang a in m umop ru:(as

. stAn* saytrarr As.' twiston ertNarm M<nt.R Guira Cuclear Station 05000 3'>9 90 022 01 or tt ti, 1 1 1

1 :1 j i i , , .

r,

f" e ~ :ci

~

(({l} " rin! g-c rs a g .

u 2 ra g-_

e

.a

-- eb u a -

. aq orp fu

.e

. e, e .

= . ca .t.l

8.  :: g

_a R R ~: ._ m I[+

.3 . i c= ,

u w-

- m M.

a- r g' .n mg axe go .

_jl_R g

,k g -

TlU T m. ~

.c . y .. .., __

c (, . %. - 'k -.<~

e y, s u "5 o i u 6 r 4 - ir m a.i ,__ ,t4: t -."g-*g2 x .

O i D 0[, - 19 ..,

. .i 7-~ ,

~~

MEhrn"u.gi g.

.I e,, df.[ i 3

e, J

p -

3 en u m y

,_. ento ..: E .

n

.g.. m ,7 m -

g"r-g% agg{ *4 e' .,,

2= og og L ox!, _e

_3..n

. ge n a(g 2er E, 4..{ "

b fyc o

a-t d:

e e' *-

4W( 4Xl <

g' T O e ej 0?.6 ,

q U"rg

Ag e,4 Ag E ;B ;4 "' * - <. I g *o R -"

r c e' " '4 es 1; y g.

g W E et-q e >-f.,cje, Ej{si e (3N (h{fiFJ.%'gl lj 1

=

V

• {

Gl- _

% { R R '

.- . EE

, .h' .E f ,h

=.

E: @

EME-

, , , . , , .c sr i. . .,,

- - .- .

• 5. 5

• I f53 l

s 3 f ' @ h,, e[ e

$ I O

.g 3 3 Y Y Y Y l

t

,+

m a- .

w_a

~m - ,,

/ .. .

il l .

Duke Iker Company 1 C Sisutum bleCuire Nuclear Cencration Department nce behident 1!100llopers ferry Road (AIGOIA) (704)375 4800 Iluntersnile, NC28018 8985 (104)3TS-4809l'a

. DUKE POWER October 27, 1992 U.S. Nuclear Regulatory Commission Document Control Desk Washingt.on, D.C. 20555 Subj ec t. : McGuire Nucle..r Station Unit 1 Docket. No. 50-3618 Licensco Evont Report 369/90-22, Revision 03 Gentlemen:

Pursuant to 10 CFR 50.73 Sections (a) (1) and (d), att. ached in Lironsco Event Report 369/90-22, Revision 01, concerning the inoperability of both trains of the Residual !! cat Homoval System during quarterly valve stroke time testing. This report is being submitted in accordance with 10 CFR 50.73 (a) (2) (i) and (v11). This ev' ant is considered to be of no significance wit.h respect. to the health and safety of the public.

Very truly yours, Kllo <.

T.C. McMeck!n TLP/bcb Attachment xc: Mr. S.D. Ebneter INPO Records Conter' Administrator, Regicn 11 Suite 1500 U.S. Nuc1 car Regulatory Commission 1100 circic 75 Parkway 101 Marietta St.,.NW, Sutto 2900 Atlanta, GA 30339 Atlanta, GA 30323 Mr. Tim Reed Mr. P.K. Van Doorn U.S. Nuclear Regulatory Commission NRC Resident Inspector Offics of Nuclear Reactor Regulation McGuire Nuclear Station Washington, D.C. 40555 03004" 9211050170 900917 -

'PDR ADOCK 0D000369

\

S PDR l{/_/l f 8