ML19325E900

Forwards Text of Info Communicated to All Util W/ Westinghouse NSSS Design on Pressurizer Safety Valve Set Pressure Deviation Info Request Response

ML19325E900
Person / Time
Site:	Crystal River, Comanche Peak, 05000000
Issue date:	10/19/1989
From:	Loftus P WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:	Rossi C Office of Nuclear Reactor Regulation
Shared Package
ML19325E899	List: ML19325E898 ML19325E900
References
NUDOCS 8911130006
Download: ML19325E900 (13)

Text

.. _ _

l P.2 E-Oct 19 '991680s LEC-Cast asSA

\\

l October 19,1989 Tot (rnia Rossi 492 1163 i

Subject:

Pressuriser Safety Valve Set Pritsure Davistlun !nfundlien Request Response t

The following pages provide the text of the infomation consnicated to all nuclear utilities j

i with Westinghouse NS$$ designs on the above subject. This inforestion is being provided in i

response to your discussion with Bill Johnson en October 18,194 and our subsequent telecon.

We plan to follow this infomal.,

comunication up shortly with a foms1 letter.

{

please contact me or Carl Hirst if you need additional infomation.

i i

k 4

i P. A. Loftus. Nanager ProductLicenisql Westinghouse t e ear Safety j

(412)-3746282 4

t I

f i

?

8911130006 891101 PDR ORQ NRRB i

PNV 6

l p,3 o:719 '8916:09 WCC-t#51 ey i

l l

4 i

October 18,1g8g j

pranauptmar Rafatv Vaiva tat ernaaura haviattan i

This letter is to provide you with inforestion M1sted to a potentis) Code i

deviatten of the ressortaer safety valve set pressuro free the ASMEhle inferinstj j

l end the plant t niet) specificetten requirements. by to Westinghoupe l

estion following rev$se

}

beine provided yese Based en an evaivation of the avattable Safety Reviou Ceanitt that this issue does set constitute a l

infomaties, the SAC eens substantial safety hasted and, es such is not reportable by Westinghouse to the INlt under,0CFA part II. De 41lC then eveLutted the significanceFra this i

of this istut utilising the criteria of 10CFRle.59.the SRC tencluded i

i utt11 ties for their evaluation.

[

l SYl10pstl ASNE lection !!! defines set pressure and !revides an opening pnssure f

tolerance that is specified in percent of he set eressere for pressures the set pressure for the pressuriser safety l

above 1000 pet.

TypicallyIn plant Technical Specifiestions. Recent plant 15 valves is tall psig +/ d test data indicate that the opening pressure experience anmore than one percent free the erising) set eressure when the operati i

valv4 is natalled at temperature conditions lifferent Free these use chanets 1

i during the set pressure test. percent can occur. "his potentselly pisce Technical specificattens, ASME Code Sections ll! I reestrements, and sis criteria.

l thus, outside the bounds of the plant Itcensing I

IDENT!FICATI0li 0F 1880t P ific ts: and Electric (Pt&I) South Carolina Electric and 8as In'1989 two utt11 ties

).

The valve testing houseWeslernSoviceCenter(

GM6' forged body (V.C. Summer and cast hdy (Diable at the Westi signs. The set pressure tes a were serfereed using a loop included Cros eesi configuratten. Test condittens included the conFro) of ambient air Canyon) valve temperature to simulate as. installed plant condittens.ed setting the l

valves to t4ll pale its using apprestostely 800 'F het water in the lees. The los> 6 ell was subsequently drained and the set pressure checke with stees. Tw valve set pressure dropped appnainstely 45 to 05.

l it i d that set Based on the testing performed at the Witpressurechan l

Plants setting their valves on steam and installing them en het er cold water loop seals have a 1

resultant set pressure higher than 2448 pote +15.

Since the trend is for the set pressure difference to increase as the temperatu I

will resvIt in the largest set pressure increase.

t l

m

p.4 OCT 89 'e916tas IEC-CAST 40%

' o-

.i I

The FSAR Itc sing buts aulysw were en1 sated since pressuriser ufety valve set points above the nominal 2500 psia 15 value could have e potential adverse'tapact en the FSAR 11const basis criteria, d ere spec fica 11y the Less of credit is taken for saflety valve relief,d heter ud RCCA (Joction analyses Les#furtine Trip, Feeditas treak, Locke were enasined.

Typicallyhn each of these analyses, the pressuriser safety valves (PSVs) i l

tre actua and erevide sufficient relief caescity whidi 11sita the peak i

pressure in the RC8 to an acceptable value. Should the ptV set pressure i

the eartta to the anximum allowed pressure for each of these to increased,be ponentially reduced to a point wiere the licensing basis events would criteria would no ' enger be antisfied, Westinshouse has performed sensitivity studies on the is ect of increased PSV set eressures for each of the four potentially affested transients.

The results of these analyses are contained in Attachment I.

81stlarly, the effect of a lost lovp seal during nomal plant operation and pressuriser Safety relief transient condittens have seen reviewed for the case in which a Pressuriser Safety Valve has been set and is installed in a loop seal configuration.

If the 1eep seal is lost as a result of a trusient Ilftine the PSV, the PSV is exposed to steam at the valve seat and a reduction In set pressure due to the increase in temperature is experienced. The reduction of the valve's set pressure from the nesimal value of 2500 psis to the p0RV set pressure and actuating at that point, does not affect the licensing basis I

criteria since'no crolit is taken for the p0RVs in the licensing basis analysis. A further set f.ressure reduction to the anxious 85 below 2500 criteria, however,

,sia t..t gt.d t. V in:t. the u c 5 5 lysis or evaluation.

confi matten wou d require p ant specific an

)

If the loop seal is leet during normal plant operation, the PtV is exposed to steam et the valve oest and a reduction in set pressure due to the increase in temperature is experienced. The redsstion of the valve's set pressure from the nominal value of 1500 pois to a level which opens during nemal plant operation is bounded for one PSV as defined by the current analysis of an inadvertent opening of a psV.

l SAfffY I480t The eressuriser safety valve is classified as a Safety Class 1 component and is required te prevent the pressure in the reacter coolant systes from arceeding its desien condition, typically 110 percent of t4ll ps1g (t600 psia). The deviation of the set pressure varies from 4 to 8 percent as seen under various lese seal conditions. This set pressure deviation is l

entside the bounds of ASME Code Section !!! and XI requirements and should i

l be reviewed by each utility in con.iunction with their Technical

$lecification. AIME Code Section !!! is not met since the set pressure of j

tte pressuriser safety valves is outside the opening tolerance specified.

i l

Likewise, ASME Code Section XI for intervice testing reevires valves not exceed the started set pressure criteria by more then 35.

I I

-me--+

I f

007 19 'te 16811 E C-CAST M 4 p,g l

)

i l

l i

C a tuston As a result of the tests tenddeted at the Westinghene Western Service J

Center, it has been detemined that the pressuriter safe valve spt l

pressure wilh very based en the methodelegy used in setti the valves.

l The varian:o curs when the valve is set at conditions r than

'as.instelhed. That is.when either the test media er ambient i

temperatures differ free,the operatine media and ambient tesperatures a i

i set pressure shift is'sessible. Crosty eftl design valves set with het t

water and ambient air temperatures of apprestaately 300'F and 130'F, test respect 4 vel, superience61 a set pressure shift of 45 to es when the media was to saturated stone. Thus setting a valve et plant i

ambient air vi h stees as a media and installine it en a leep ses1 filled with 300'F water can result in a set pressure 45 to 85 hipar than l

saticipated.

16ete that similar data does not exist for other safety valve sfass, designs, er other plant specific temperatures.

l l

Crosby Valve and tape Co. agrees that the valve set pressure should be

,establishedattemperaturesrepresentingas.installedmediaandambient t'

temperatures.

I RECOMIEl47!0ll8 I

i Utilities shuld review the entsting methodelegies that are currently in prettice at their plants relative to setting and testing of pressuriser t

safety valves, their current FSAR analyses and the licensing bases for the s1pnt to detofeine their complian;e with safety valve set pressure tolerances as specified in their "echnical Specification.

i

[

t i

I m.

..e 4 i

CCT 19 'eW 16811 nKC-CAST 195A p,g

)

~

)

l ATTACmpff I

)

l Nst!FICAT!W FOR CONTlWED Opt 2ATI0li j

j The FSAR licensing basis analyses wem evalusted since pressuriser safety valve set points above the nasin618500 psia 15 value opuld have a i

potential adverse en the F8Ah licenst basis criteria, where credit is taken for sa ty valve relief, spot fica 11y the Less of i

wem enemined. p, Feed ine treak, Lethed noter and EC4 tjoction analyses j

Lead /"urtineTri 1

in each of these analyses, the pressuriser safet valves (PSVs Typicallyked and provide sufficient relief capacity which ifnits the posk) are actus

~

pressure in the RC8 to en atteptable value, should the PtV set pressure be increased, the margin to the easieue allowed pressure for each of these r

events would be potentially reduced to a point where the licensing basis criteria would no longer be satisfied.

l i

s I

Westinghouse has performed sensitivity studies pm the insect of increased PSV set ressures for each of the four potentially effested transients.

The fell sensitivity studies were performed on the 1spect of increased set pressures for each of the four potentially affected i

transients:

Less of Lead / Turbine Trip For the less of lead / turbine trip analysis, sensitivity studies show that with no credit taken for any relief capacity free either the l

licensing basis IIsit for this Conditten 1 event). However,gn (the PSVs or the p0RVs the peak hCS pressure nacoeds 1135 of desi the f

pressure remains below 1805 of design and thus, the peak RC8 pressure does not cause stresses to sateed the faulteil conditlen stress i

limits. This enelysis is based upon the ans'ysis documented in the F8AR, and all of the conservative bounding assumptions are applied.

l 1

Feedline treak For the foodline break event Westinghouse has performed analyses l

which demonstrate that with a 105 increase in the P8V set pressure, t

from 2800 psia to 8780 psia, the manieue K8 pressure rentins below I

1805 of desten. In addition the core remains covered throughout the transient an4 no everpressurInstion of the secondary side escurs.

This analysis does not take credit for best estimate operation er p0RVs, and retains the tenservative assumptions which are presented l

in the FSAR. Thus the peak RCS pressure does not cause stresses to exceed the faulted, condition stress limits.

3 s.

00T 49 'C916812 6CC-Oest de5A py i

i Locked Reter Wostinghouse has perfomed locked roter anglyses for a typical i 1 pop j

phant, which bounds 3 and 4 loop plants. This analysis ashes siellar j

senservative assuuttions to those found in the FSAR analysis. No l

credit was taken for any retted capacity free the PSVs or PORVs. The the peak maximum K8 pressure rensins below 1995 of design. Thys,d condition l

I RC8 pretsure does not sause stresses to esteed the faulte stress limits. There is no adverse 1spect upon the ro ls.in 05 or l

the peak sled tesperature analyses documented in the FsAR.

l RCCAtjection l

Westinghouse has perfomed a bounding everpressurisation analysis for i

the RCCA ejection event which is dpcumented in WCAp 7688. This analysis is perfomed under 90L HFp conditions and makes estressly conservative a)sumptions togerding ejected red worth. The PSVs are assumed pperable with a set pressure of flog psia. The peak pressure is calculated to be less than 8000 psia. A 105 intresse (tle psi) in the PSV set pressure would intresse the peak pressere by ne asre than t

RSO psi, resulting in a peak RCS pressure of 3660 psia. This is I

as dis essed in erotte hen 1995 et desten eressure. Howeverlive ana s, usin l

psia. g 30

, a more dotatiel, but still conserva i

i SCAp4 I

methode egy saltuinted a peak RC$ pressure less than i Thus even with the additional 150 pet bias due to the 105 setpoint I

below l

shiff,fdesign.

l thepressurewillessainlessthan2850psiawhichify In addition this analysis used se estreme l

1805 o I

suservativeesestede.dworth.stims.dasatoav.es,reaterthan the conservative values typically presented in the F8AR. Lower, but still conservative, ejected red worths would yield lower RCS l

pressures. Thusl even under conservative assumptions, the peak RC t

i pressure will ne esteed that which wov1d cause stresses to esteed the faulted tenditten stress Itatts, j

Based on the results of these sensitivity studies, the cattulated pressure spikes for these transients de not challenge the pressure integrity of the l

primary systes cesponents.

f f

limilarly, the effect of a lost loop seal during normal slant operetten and pressuriser Safety relief transient conditions have been reviewed for i

the case in which a Pressuriser $sfety Valve has been est and is fasta11ed, in a leep seal configuration.

i

[

i

~

-~-,w---+,w--e v---.,

m ee,

--we.---

~~------n--

2 l

cci e,9 'ee asias W.c-r.ast assa p.e

,*c i

a If the 1eop seal is 1eet as a result of a transient lifting the PSV, the f

esure PSV is exposed to steam at the valve seat and a reduction in set p The reduction the due to the increase in tesserature is experienced.

set l

valve's set pressure free he nominal value of 3500 pois to the i

pressure end actuatine at that point, does not affect the Itsensing basis

'i sriteria since no trolit is taken for the PORVs in the 11 sensing 64 sis enslysis. A further set scossure reduction to the staisus 85 M1ew 2000 psia is not expopted te vlelate the licensing criteria, however, confirmation would require plant specifts an& lysis er esalvation.

l I

If the 1eep son) is 1est during permal slant operation, the PSV is exposed j

to stees at the valve seat and a reductlen in set oressure due to the increase in temperature is enterienced. The redvetten of the valve's set i

pressure free the nominal val'oe of 1500 psia to a level which opens during l

I normal plant operetten is bounded for one PSV as defined by the current entlysis of an inadvertent opening of a PSV.

l 1

i i

i l

i l

l r

l l

l 1

I I

e.

4-Et4 CLOSURE 3 N 71. 1999 E6ItB MMPT Ch LDETDM Fra i m sq Ag((ggy RilO9 M 04W1Ebl (L#tED TIPE IN POlMd) trTIM75 PCNM.Y ACTIVITIES LDSTEPM Fn i N #9 LIMITEPM A f N89 UNrMPM PILLDAPB ACTI N M M D CTIN CDfN

{ Nim!ZATICN Pf.AU WWIN & W.

3 TO 6 T, 1 N NW 3 E.

N 1999 TT3G 1M m

1 0

0 0

0 g2 1

0 0

0 0

PO3 5 0

0 0

0 1

WID 1

0 0

0 0

ean a 1

0 0

O_

0 m

1 0

1 0

1 TEMAL S

0 1

0 I

a M wt ass & Fed fCr ICneterm follcase sisy how tuon assigwd to nultiple trw*two and a single TM PCfit may how taan mos1 Fwd to similar wts at multiple plants.

i n

W i

, ' Pete b.

1 11/01/09 PERFDRWttINDICA10R$$16NIFICANTtytN1; o

PC.i NAMt (VENT EVthiLESCAlPl!DN DTR ll6NIFICANtt l

Hit c

CRYSTAL RIVER 10/N/99 HPl LINE FLON INSTRUNENTA110N DittRMINED NOT 10 Pt 0 !Mt1V RELAlt$ (DulPMENT SUFFICl[NTLY ACCURAff FOR OPERA 10RS TO CORRICTLY 64 LANCE LOOP 1NJEC110N Fl.0NS IN ACC0FNNCE NITH THE PLANI (MtF6tNCY OPERA 11N6 PROCEDU4t$.

i 4

0

r-g m.-

i..

l

(

i N AC1R SCRM IU N RY Etttelut 10/29/09 l

1. PLW1 MtlFit M1A i

Mit Blt!

W!! Peurt llIRAL Cault 80rLI-mm m

l Cat 10NI WM KLW TOTAL

!$1 1H f

I i

10/27/M llWLO CWrm 2 AM N EOU1 M NT W

4 6

4 l

i f

I l

i l

i ItKTOR ICRM SuwRY i

Ett telut 10/22/09 i

1. PLANT M CIFIC NTA i

HTE 817t Wii P0urt lighAL C4 ult Cartl-mm m

I CAflIul AINE ILLOW 79f4L 191 198 t

10/10/99 Nlut NILE POINT 2

1A Pttl0NL M

4 1

3 10/10/09FARLif 2 160 A E80lPftNT W

4 6

4 10/19/M NILLif0NE 1

70A EOUIPIENT W

2 1

3 10/19/09 l#1M POINT 3 100 R PERSONEL W

2 10/19/99 FARLtf 2

2A PERSONEL M

4 1

l j

t e

e G

P f

l 1

l~

t

--+,-,,

.-_.--e..-__...,,,,,---,,n,.

--..,_-,n..-.-

--,.-.,_.__,.n.

.-,----,---,,.e,,

1

!!. COMPARISON OF WEEKLY STOT!STICS WITH INDUSTRY CVERAL28 3

SCRANS FOR WEEK ENDING j

10/29/89 BCRAM CAUSE.

POWER NUMBER 1989 1989 1987 1986 1985 i

OF WEEKLY WEEKLY WEEKLY WEEKLY WEEKLY SCRAMS (5)

AVERASE AVERASE AVERASE AVERASE AVERASE YTD (3)(4)

(B)(9) l l

88 POWER >15X EQUIP. RELATED

>15X 1

3.0 3.1 3.9 4.3 5.4 PERS. RELATED(6) >15X 0

1.1 1.0 1.3 1.5 2.0 QTHER(7)

>15%

0 0.1 0.5 1.2 0.4 0.6 I

88 Subtota! 88 1

4.2 4.6 6.4 6.5 3.0 88 POWER <15X EQUIP. RELATED (15%

0 0.4 0.5 1.2 1.4 1.3 PERS. RELATED (15%

0 0.3 0.3 0.6 0.' S 0.9 OTHER

<15X 0

0.0 0.1 0.3 0.2 0.2 I

88 Subtota! 88 O

0.7 0.9 2.1 2.4 2.4 888 Tota! ***

1 4.9 5.5 S.5 S.9 10.4 MANUAL VS AUTO SCRAMS TYPE NUMBER 1909 1988 1987 1986 1985 l

OF WEEKLY WEEKLY WEEKLY WEEKLY WEEKLY SCRAMS AVERASE AVERASE AVERASE AVERAGE AVERASE

)

YTD l

l t

MANUAL SCRAMS 1

0.9 1.0 1.4 1.0 1.0 AUTDMATIC SCRAMS 0

3.9 4.5 7.0 7.V 9.4 J

l j

l i

i i

g pw,

,,e e

~~

---r,

,s

-w--,---

w

,~--~--v,

--w-,-

,,ss-v

---

e----

~ - - - - - - - - - - -

(,,

!!. CDMPARISDN DF WEEKLY STCTISTICS WITH INDUSTRY CVERASES f

SCRAMS FOR WEEK ENDING 10/22/89 SCRAM CAUSE FOWER NUNDER 1989 1985 1987 1986 1985 DF WEEKLY WEEKLY WEEKLY WEEKLY WEEKLY I

SCRAMS (5)

AVERASE AVERASE AVERASE AVERASE AVERASE YTD (3)(4)

(S)(9) l Et POWER >15%

l EQUIP. RELATED

>15%

2 3.1 3.1 3.9 4.3 5.4 PERS. RELATED(6) >15%

1 1.1 1.0 1.3 1.8 2.0 OTHER(7)

>15%

0 0.1 0.5 1.2 0.4 0.6 88 Subtotal se 3

4.3 4.6 6.4 6.5 3.0 f

48 POWER <15%

EDU!P. RELATED

<15%

0 0.a 0.5 1.2 1.4 1.3

[

PERS. RELATED

<15%

2 0.3 0.3 -

0.6 0.8 0.9 g

OTHER

<15%

0 0.0 0.1 0.3 0.2 0.2 88 Subtotal as 2

0.7 0.9 2.1 2.4 2.4

• • • Total 888 5

5.0 5.5 B.5 S.9 10.4 l

i i

MANUAL VS AUTO SCRAMS i

TYPE NUMBER 1989 1985 1987 1986 1985 DF WEEKLY WEEKLY WEEKLY WEEKLY WEEKLY SCRAMS AVERASE AVERASE AVERASE AVERASE AVERAGE I

YTD 5

i MANUAL SCRAMS 1

0.9 1.0 1.4 1.0 1.0 AUTOMATIC SCRAMS 4

4.0 4.5 7.0 7.9 9.4 i

t l

a i

.. ~, _,...

I..

l L..

\\

\\.*

!$3.h 1.

FLANT SPECIFIC DATA BASED ON INITIAL REVIEW OF $0.72 REPORTS FOR THE WEEK OF INTEREST. FERIOD IS MIDNIGHT SUNDAY THROUGH MIDNIGHT SUNDAY. SCRAMS ARE DEFINED AS REACTOR PROTECTIVE ACTUATIONS WHICH RESULT IN R0D MOTION. AND EXCLUDE PLANNEE, TESTS OR SCRAMS AS PART OF PLANNED SHUTDOWN IN ACCORDANCE WITH A FLANT PROCEEURE. THERE ARE 111 REACTORS HOLDING AN OPERATING LICEkSE.

l 2.

COMPLICATIONS: RECOVERY COMPLICATE) BY EQUIPMENT FAILURES OR i

FERSONNEL ERRORS UNRELATED To CAU5: OF SCRAM.

~

3.

PERSONNEL RELATED PROBLEMS INCLUDE HUMAN ERROR. PROCEDURAL

)

OEFICIENCIES AND MANUAL STEAM GENERATOR LEVEL CONTROL PROBLEMS.

]

4

'OTHER" INCLUDES AUTOMATIC SCRAMS ATTRIBUTED TO ENVIRONMENTAL CAUSES (LIGHTNING), SYSTEM DESIGN, OR UNKNOWh CAUSE.

s k

6

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _, _