ML20006E775

LER 88-001-05:on 880102,manual Reactor Protection Sys Trip Initiated Due to Decreasing Condenser Vacuum.Caused by Numerous Leaks on Main Turbine & Main Steam Reheat Interconnecting Piping.Seating Matl installed.W/900219 Ltr

ML20006E775
Person / Time
Site:	Brunswick
Issue date:	02/19/1990
From:	Harness J, Jones T CAROLINA POWER & LIGHT CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
BSEP-90-0130, BSEP-90-130, LER-88-001, LER-88-1, NUDOCS 9002260338
Download: ML20006E775 (10)

Text

hn -

R q., pn '

,y, ~. g,

, MC s Carol 6na Power & Light Company L.

[. ;- .

Brunswick Nuclear Project l 1 P. O. Box 10429 Southport, NC 28461-0429 February 19, 1990 FILE: B09-135100 10CFR50.73 SERIAL: BSEP/90-0130

,' O.SJ Nuclear Regulatory Commission l

ATrN: Document Control Desk Washington, DC 20555 BRUNSWICK STEAh! ELECTRIC PLANT UNIT 2 DOCKET NO. 50-324 LICENSE NO. DPR-62 SUPPLEt!ENT TO LICENSEE EVENT REPORT 2-88-001 Gentlemen:

! In accordance with Title 10 to the Code of Federal Regulations, the enclosed 1- Supplemental Licensee Event Report is submitted. The original report fulfilled' the requirement for a written report within thirty (30) days of.a reportable

- occurrence and was-submitted in accordance with the format set forth in NUREG-1022, September 1983.

l Very truly yours,

)

v w, I . L. Ilarness, General blanager Brunswick Nuclear Project

-Tt!J/mcg i.

Enclosure cc: tir. S. D. Ebneter 3~

tir. E. G. Tourigny BSEP NRC Resident Office c, M kI b

.gg229oggg?,%8$$33($

g ,

S

[

,y

I gcyoRM,ms u s. NuCLa:a t a tuu.T omv Commission y,, oy g , o,, ,o , ,,,,,,, .

L txPIIE$ it30!92 i

. EsTIMITED DUTDEN IE3 f.EEPONsf TO COMPLY P'TH THit I

[ ""

Eo'JMt#s'Uo* '

'dj'o'.'Est8E!!.h'TYTWit '

0"CES

. LICENSEE EVENT REPORT (LER) "

I

'rev"JTo"# Co'JJM "ot"W ^"='lis &' "WJ t" r D'MI'A%iM!Wo'e"afnL'!dnm.f!"

. Aciuf v AME m ooCASi Nuunk m '= m i

Brunswick Steam Electric Plant Unit 2 0 l51010 l 013 l214 1l0FlOl9 )

n"' Manual Reactor Scram Due to Decreasing Main Condenser Vacuum and Failure of Primary i Containment Group 2 Valves G16-F003, F004, F019, and F020 to Close on Isolation signn1 l

$VtNT DATl ($1 LtR NUM9th 166 REPORT DATE 47) OTMt h 9 ACILITill INVOLYLO 401 DOCkt i NUMDt Rist MONTH DAY vtAM YtAR "w" Mn'" 0*ffM MONTH DAv vtAR e Ac Litv NAMas 0 16101010 1 I l

- ~

l 0l1 0l2 8 8 8 l8 0 l 0l1 0 l5 ol 2 1l9 d0 0lDl0l0g0; l l l THit REPORT tt SutMITT80 PUR$UANT TO THE RLouthiMtNTS OF 10 C8R I (Cneca one er me,e er ,ne terremass (tu j ePt R ATINO MODS #1 } 20 402161 20 4061st

{ 60.731eH2Hevi 73.? t M 73.71ist R

to 406teH1Hil 60 36IeH1i

{ 00 73IeH2Hvi no, 0 i 5, 5 .0 0. m n.1 30 40steH1Hiid

.0 =isim 60 731eH2Hl) n .OneH,n...

to.736eH2HvdiH Al

_ ggy,;eg,g J66A1 u

90 4066eH1H6el X S0 t3leittHe) to.?3teHtHv4.Htt 30 406teH1Het D0.13ieH2Hud te 73teHtHal  !

LICIN$tt CONT ACT POR THIS Lth litt Ntyg itLEPHONE NUMetR Aria CODE T. M. Jones Regulatory Compliance Specialist e 911 l 9 4 l 517 l- l 21311 15 COMPLif t ONE LINI POR t ACH COMPONtNT F AILURE OttCRitt0 IN THl3 OttPORT l13)

R fa l CgUgg gyg7gg ggyppgggy M C. R TA t C1Ull Sv$ttM COMPONENT "$NAC o pn g pn g X JlM l1 SVI Gl 0 1 821 Y X JlM i II S I V GIO l 812 Y [

X JM g j Ig S;V Gi O ; 8, 2 Y X JlM l Il S I V G]0l 8l2 Y SUPPLtMINT AL RtPORT I kPECTED H43 MONTH DAY vlAR

$USMissa0N

• ^"

T'] vis iir Pe.. eeme= ruterro svesess,0N oA rts -] NO 0 16 1 15 910 AuT R ACT ,L-, M mo .ue. . e . w,e. .

e.,, ,,,.ea sye .ue rm,-,,,ea nnou n se >

b While performing a routine reactor shutdown in preparation for the Unit 2 1988 refueling / maintenance outage, a manual reactor protection system trip (scram) was initiated at 0017 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> on 1/2/88, due to a decreasing condenser vacuum.

Reactor power was approximately 55*, and vacuum had decreased to approximately

-22 inches mercury. During the expected vessel level shrink following the scram, vessel level decreased to approximately 153 inches, thus initiating primary containment isolation valve groups 2, 6, and 8 at low level 1

(> 162.5"). Operator verification of these valve closures determined that the group ? valves 2-G16-F003, -F004, -F019, and -F020 failed-to close. These are the inboard and outboard isolation valves for the drywell floor drain sump ,

(F003, F004) and the drywell equipment drain sump (F010. F020). The remaining safety systems operated as designed during this event.

Investigation of the decreasing vacuum condition determined it resulted froin numerous leaks on the main turbin9 and main steam reheat interconnecting piping to the main turbine, which were repaired during the unit outage. To date, the cause of the group 2 PCIVs failure to close has not been determined. By 6/15/90, a supplement to this report will be issued to update the root cause determination of the failure of the valves to close.

N~Cform364 (6 891

F-

> e ,. u a nuc66 2 mtrue.50.v com.iniO=

' ' ^""ov'Do****"*

• LICENSEE EVENT REPORT (LER) FAILUME CONTINUATION IKPiht5 6 31 et 8 ACILITY haul it) . Docal1asuMelA us 6la woutta ten PAGE(3 O'J4ft vIAa R1;;."

Brunswick Steam Electric Plant Unit 2 o [i,] 010 lo I 312 p 818 -

0101 1 -

d 5 012 0F 0l9 COMPLlit ONE Leht tom LACM CouPoktNT phitymt Ottcmisto 18s THIS htPORT (136 Caust svsflu COMPONt N t WANG g 8AC-R a,Tgit CAust Sv $tt M COMPONENT "")yE "f0 hPR's' L

X JMI I R IL ff GIO1812 Y l l l l l l l 1 1 I i i l 1 l l i I I I l i I I I I I I I I l l I I I I I I i f i 1 1 I i 1 I I I I I I I I I I l l I l l l l l I i i l I I I I I I l l I I I I I I I I I I l _

l l 1 l l 1 l l I I I I I I l 1 l l I l l l l l 1 1 I I I I I l l l l I l l I I I I l i l l I l l i I I l  ! I I I I I I l l I I I i l 1 1 1 1 I I I I I I i 1 1 I I I I I I i i i l 1 I I I I I I I I I I I I l i I I I I I I I I I I I I I I I I I I I I I I I I l I l l s

I i l l I I I I I I I I I i ,

I I I I I I I i i l i I I I (

l 1-1 I I I I 1 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I l l 1 1 I I I I I I I I I I I I I I I I I I I I I I I l 1 l l l i l l i I l l 1 I I I I i 1 l  ! I I I I I I l l 1 1 1 1 I I I i 1 I I I l 1 1

~,A,c,,. 0a . m.

L

, g, c.......... ,,,, ,,, ,,,,,,, ,,,,,,,,

..... . . = = .

UCENSEE EVENT REPORT RER) M'c'".f#c',7 Wino *'!Po*3'ill.**C.' ,"o'f.;'s  !

TEXT CONTINUAT60N P,5".'ie'.  ;!','"@f*f'.1105.','!2 Ui' Of5

• W,'

c. ., u .

.. #.'J.i".".a.

a Nf a l, 0.!.7,c*,'R..e.n..?'J,oo"c#

. .c it et v h.ani d.l pocai, e,uesih un it. e.Waeth #1 4 iSl vi.. " b!?J.*'  :'J1?: j Brunswick Steam Electric Plant Unit 2 M o l0 el3 2 p 8 l8 -

0l 0l1 -

015 0l3 0 19 vent v . = =o.- amm nn ,

Initial Conditions At approximately 2015 hours0.0233 days <br />0.56 hours <br />0.00333 weeks <br />7.667075e-4 months <br /> on 1/1/88, a power reduction was commenced from 69*. power to commence a scheduled 16-week refueling and maintenance outage.  ;

This initial condition represented the maximum ataainable reactor power due to ,

operating cycle fuel depletion. After approximately two hours, at a power 4 level of approximately 50*., problems were encountered maintaining condenser vacuum. At 0017 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> on 1/2/88, a manual reactor protection system (EIIS/JC)

- trip (scram) was initiated with main condenser (EIIS/SG) vacuum at f approximately 22 inches mercury (lig) and decreasing in anticipation of an automatic scram due to the main turbine (E11S/TA) trip on low vacuum at greater than 30*, power. At the initiation of the scram, plant emergency core cooling systems and other safety systems were operable.

Event Descriptfor.

• At 2015 hours0.0233 days <br />0.56 hours <br />0.00333 weeks <br />7.667075e-4 months <br /> on 1/1/88, reactor power reduction was commenced in preparat.fon i for the scheduled refueling / maintenance outage. As power was decreased, condenser off-gas flow slowly began to increase such that, at 2125 hours0.0246 days <br />0.59 hours <br />0.00351 weeks <br />8.085625e-4 months <br />, the

. augmented off gas system bypass valve (Ells /WE/TCV) automatically opened due l to high flow (setpoint; 150 scfm). This increase in off-gas flow was not  ;

unexpected in that it had been noted that off-gas flow had increased during  ;

previous power reductions. This off gas power relation is believed to be caused by minor steam leaks at high power 1cvels which become vacuum leaks at lower power levels. Although off-gas flow was increasing, condenser vacuum  ;

was showing slight improvement. Power was reduced to 48*. at 2215 hours0.0256 days <br />0.615 hours <br />0.00366 weeks <br />8.428075e-4 months <br /> and 2B  :

steam jet air ejector (SJAE) (EIIS/Sil/EJR) was secured with 2A SJAE remaining in half load per the plant shutdown procedure. After securing 2B SJAE, '

condenser vacuum began to decrease and the 2A SJAE was placed in full load at ,

2245 hours0.026 days <br />0.624 hours <br />0.00371 weeks <br />8.542225e-4 months <br />, and reactor power was increased to 51*.. These actions caused vacuum to reverse the downward trend and start improving. With vacuum improving, offorts were initiated to identify vacuum leaks for repair / isolation to allow the recommencement of the scheduled power reduction. L At 2345 hours0.0271 days <br />0.651 hours <br />0.00388 weeks <br />8.922725e-4 months <br />, vacuum again began to decrease with the decreasing trend beinj; at a higher rate than had been observed during the initial decrease following the securing of 2B SJAE. No evolution had taken place during the previous

, hour which would have caused this change. Attempts to place the 2B SJAE in i

? service were unsuccessful due to a low output from the startup permissive temperature instrument (E11S/JA/TC). Power was increased to approximately_55*. '

by increasing recirculation flow and control rod withdrawal in an effort to' terminate the decreasing vacuum trend. These efforts appeared to have no effect. With the vacuum continuing to decrease, the decision was made to ,

manually scram the reactor prior to receiving the automatic scram due to the +

turbine trip on low condenser vacuum at greater than 30*. power.

oe ,- . ,

L . - -

g. .

.. . u.... s. .

,,,,,,,,,,,,,,,, ,,,,,,,, j ucersEE EVENT REPORT (LEN W,,'",%VW,7,2,o 'M.,W ,%'ai ,

TEXT CONTINUATtON 27.' 4,'A',5,'ffM f'.M!,'#;,'M n' MS l Wla','s?.'MT@M.**.3's'r,?.'2% tMi l

- 0? anAa.Aosest rt Aa.n supott,wAentesGiose.DC 30003  ;

f acistiv tenant tu Doctti.su.stl. Os gga etWaWGA tSi PAtt LS {

"*a M!?U'. lJJ.R l Brunswick S' team Electric Plant Unit 2 ,

o l6 l0 l0 l0 l 3l2 p 8l 8 -

0] Ol 1 -

015 014 0F Ol9 l vartu e it .asimwancs momnn At 0017 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> on 1/2/88, a manual scram was initiated at a condenser vacuum of approximately -22 inches lig and decreasing. A normal scram recovery was ,

fuitiated using the emergency operating procedures. Immediately following the  ;

scram, reactor vessel level decreased to approximately 15'3 inches due to expected void collapse and returned to the normal operating level. As a result of decreasing below a reactor vessel level of 162.5 inches (low level 1),

an automatic scram signal was initiated along with an automatic primary  :

containment isolation system (pCIS) (EIIS/JM) signal for groups 2, 6, and 8.

A verification that these automatic functions, per the emergency operating procedures, occurred at 0020 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> determined that the PCIS group 2 valves (EIIS/JM/ISV), 2-G16-F003, -F004, -F019, and -F020, failed to close. These  !

valves are the inboard and outboard isolation valves (both located outside the  ;

primary containment) for the drywell floor drain sump (EIIS/IJ/SNK)  ;

(F003, F004) and the drywell equipment drain sump (EIIS/IF/SNR) (F019, F020). ,

Failure of these valves to shut represented a failure of both redundant safety  :

divisions as PCIS valves F004 and F020 are associated with one logic division,  ;

and PCIS valves F003 and F020 are associated with the other logic division. '

Following identification of the failure of these valves to close, each valve  !

was given a manual close signal from the Control Room reactor turbine gauge l board (EIIS/NA/CBD), at. which time the F003 and F004 valves went shut. No change in position status was noted for the F019 and F020 valves. At "

approximately 0023 hours2.662037e-4 days <br />0.00639 hours <br />3.80291e-5 weeks <br />8.7515e-6 months <br />, the F020 was observed to be in the closed position.

No evolution could be identified which would have caused the F020 to close during this three-minute time frame. Another manual close signal was given to the F019 valve at approximately 0023 hours2.662037e-4 days <br />0.00639 hours <br />3.80291e-5 weeks <br />8.7515e-6 months <br /> and again, no change in position .

noted'; however, the F019 valve was observed to be in the closed position at '

0025 hours2.893519e-4 days <br />0.00694 hours <br />4.133598e-5 weeks <br />9.5125e-6 months <br />. Again, no evolutfor. could be identified which would have closed  !

the F019 valve following manual operation. ,

Further review of the plant response to this scram indicated that the remaining plant safety systems operat ed per design. Five control rods ,

(EIIS/AA/ ROD) were identified to be at the 02 position following the scram and they were fully inserted by 0030 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br /> using plant procedures. -

Investigation Summary Decreasing Condenser Vacuum

• At approximately 2220 hours0.0257 days <br />0.617 hours <br />0.00367 weeks <br />8.4471e-4 months <br /> on 1/1/88,. vacuum began to decrease following the removal of the 2B SJAE from service per procedure. Following an increase in

^

power and the placing of 2A SJAE in full load, vacuum began an improving trend. One hour later at 2345 hours0.0271 days <br />0.651 hours <br />0.00388 weeks <br />8.922725e-4 months <br />, vacuum again began to decrease due to no apparent cause, as no plant evolutions had been undertaken within that hour 4,

a e , - -A . ..

o ,

,,,,,0,,,, .. .m .ut. . . ,,,,,,,,,,,,,,.., .

==,

, ......c.

LICENSEE EVENT REPORT (LER) f,'ll",,'l'&*,"WW,M8.*M/W"*4' ,y," 'Q L TEXT CONTINUAT60N f,7,'jfo'sP.'M 30f'J10eO,'Lif 'v"g' OMn >

1. f"i.','.T.TJi'7/=.a',Mn!"Mi'ei Of MAldAOttellif A880 $90Gil 5tASal'8010h.DC 38603  ;

I 9 ACILITV 85&441 tu DOCElfIfWhStA GI gga .g,ggg pgi . ,A.E (S) vtam f' $h"4  ; IMa i

Brun'swick Steam Electric Plant Unit 2 .

0F 0 b l1 -

0 15 01 5 0 19 term e ,= , ease wmac e.,u,ama otin i

which would have affected vacuum. Vacuum continued to decrease until the manual scram was initiated at 0017 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br />, at which time tacuum recovered quickly, decreased for a short period of time, and then gradually increased until leveling out at -28 inches of lig.

A review of the vacuum trend during this event and previous plant vacuum opereting history indicated that, although higher than normal air inicakage did contribute to the decrease in condenser vacuum, the high air inleakage flow rate would not alone cause the high SJAE discharge pressures observed during the event. It is unlikely that a large air inleakage source initiating at 2345 hours0.0271 days <br />0.651 hours <br />0.00388 weeks <br />8.922725e-4 months <br /> would have allowed the vacuum to recover and trend as noted~ '

following the scram. ,

As previously noted, air inleakage had been an identified problem prior to j this event during power reductions. Investigations were initiated on 9/25/87 i to identify and correct suspected air inloakage problems. The investigation consisted of a complete valvo lineup and helium leak testing. These

, investigations had identified several air inleakage sources during the latter part of 1987 and were still in progress at the time of the event. Repairs had +

- been made where system operation and safety would allow, with the remaining known air inleakage problems to be corrected during the outage. Additional air ,

inleakage sources were still being sought at the time of this event.

In addition to the known and suspected air inicakage, the decreasing trend in condenser vacuum is believed to have been caused by SJAE back pressure. It is believed that excessive moisture in the downstream piping (EIIS/Sil/PSX) of the '

A SJAEs was partially the cause for the decreasing vacuum. Indications of this probable cause included a high differential pressure alarm (possible cause; high moisture in the off-gas effluent) received for the off-gas system main ,

stack filter (EIIS/WF/FLT) which cleared when the standby filter-(EIIS/WF/FLT) j was placed in service as well as a noted decrease in the off gas system recombiner (EIIS/WF/RCB) temperature. Moisture carryover into the discharge l

• - line (EIIS/WF/PSX) may have occurred from the high level noted in the 2A SJAE j intercondenser (EIIS/SH/COND), which was operating at a level of 82 inches I.

(normal level is 55 inches).

During the Unit 2 1988 refueling / maintenance outage, extensive Icak testing l involving use of helium was performed on the main turbine and condenser and -

l l the_ interconnecting piping to the Condensate System (EIIS/SD) and the Main '

l: Steam Reheat System (EIIS/SB) as well as the main turbine structure to l determine the root cause of the incurred decreasing main condenser vacuum.

l On February 2, 1988, while the main condenser was intentionally flooded up, water was observed flowing from a discovered 3/4-inch hole in Miscellaneous '

i Vents and Drains (MVD) System line 2-MVD-267-4-E-3 (EIIS/SM/PSX). This 4" line '

l l.

l 1

\

l Nac f.an setA See , ,

4 g,e ... on ... . a , . ,,,,,,,,3 ,,,,,,, ,,,,,,,,

LICENSEE EVENT REPORT (LER) 32.',%',,wW.%"'.%"s, **"'.l7 ,7,.7 '.T!

E7.'t %%5'fW.,*,.0P

• d!,'#','! !S'l' *E3 TEXT CONTINUATION 1 W,W#,J e8 knA 604a.J '?.'m.a',r.a'?%L'"A e M?,li lett Aasp supGtt stAsmiesotoes DC 30403. i F ACtht?Y .s4 4 til D0cett) espaatth sti HA MA m P&D5 (Si

,... -.. g,o . m eio- .

Brunswick Steam Electric Plant Unit 2 0 l6 l0 l0 l0131214 81 8 0l0l 1 -

ols olA M nlo [

isxw . w. .as w ac a amm tm ,

• i

~

'is the shell pocket drain line from the main turbine moisture separator reheater (MSR) to the main condenser, which collects the MSR shell drains from four '

1 1/2" lines (EIIS/SM/PSX), each containing a 1/4" orifice strainer .

(EIIS/SM/PSX). 'dhou the main turbine is online, line pressure upstream of the strainers is that of the main turbine 7th stage extraction steam pressure, which varies with reactor power (80 psig at $6*. power to 170 psig at 2 00*. power. 4, '

Likewise, line pressure downstream of the strainers varies with reactor power, ,

I such that as power is decreased, the resulting reduced pressure in the line will cause greater main condenser air inleakage. The hole size was determined l to have allowed an air inleakage of approximately 96 standard cubic feet per '

minute (scfm) at a main condenser vacuum of 28 inches of mercury. Following .

discovery of thn subject hole, the affected line was replaced. In addition to. <

the hole in the MVD line, several other leaks were identified through helium 1 testing. Repairs to these components were performed during the unit outage. ,

Valve Failures While performing a scram recovery in accordance with plant procedures, it was

• determined that the group 2 PCIS valves 2-G16-F003, F004, F019, and F020 ,

failed to close on an automatic isolation signal (low level 1). Subsequent ,

operator action caused the valves to close as previously stated. The '

following actions were performed in an effort to determine the cause of the PCIS valves' failure to close.

January 2, 1988

1. Following the scram recovery, the four group 2 valves were successfully

• cycled during normal sump pumping operations with no problems noted with ,

valve operation. The Unit 1 (U/1) valves were also stroked to verify-  !

operability.

2. A visual inspection was performed on the wiring (EIIS/JM/CBL1) and relays (EIIS/JM/RLY) associated with the group 2 isolation logic on Unit 2 (U/2).

3. A maintenance history search was initiated to develop the operating history of the failed PCIS valves. This history review determined that the F003 valve had experienced three failures and the F004 valve had '

experienced one failure since the solenoids (EIIS/IJ/*) were replaced in the spring of 1986 as part of environmental qualification modifications.

The remaining two valves on U/2 and the four valves on U/1 did not have a failure history.

• EIIS component description unavailable.

21C Peen 30641640)

r pgm .....:....- . . ,,,,,,,, ,,,,,,, ,,,,,,,,

uCENSEE EVENT REPORT (LER) @ ",'i',*,duag ,;Q "'.5l'n ,P

• T T 4' 0 ,

i- TEXT CONTINUATION W,'ii'AT,lf,*153.T!.d!,'e'"',LT,".' O!!l.'1 M.

e

.t.t'.%74..!,'.T.i.?.#.v is .

a,.;

,eur, ii.ans m - .ocan oveen. ai te. .uusi. mi e..e is l

.... 'm un .

Brunswick Steam Electric Plant Unit 2

• O l6 l0 l0 lo l 3l2 9 8 l8 -

0l 0 l - l -

0l5 017 M 019 >

vert va as a ==sen s wmac a ,manvnn

't

4. Applicable sections of the logic system functional test procedures were '

performed, with no logic problems identified. r i

January 3, 1988  ;

1. The F019 solenoid valve (ASCO) (Ells /JM/PSV) was removed and l disassembled. A minor oil illm was identified on internal parts and minor debris was found in the solenoid valve. Ilowever, no cause for, or- <

evidence of, failure was identified.  ;

2. The group 2 PCIS valve monthly functional test was performed on U/2. No- l problems wore identified.

January 4, 1988 ,

1. A special test procedure was performed to verify the group 2 PCIS valve l system logic on U/1. No problems were noted.

I January 6, 1988 )

,.. 1. A special test was performed to simulate the conditions present during the U/2 scram. The operation of the valves was observed locally and strip chart recording of electrical circuit operation was obtained. No problems were identified with valve operation.

2. The U/1 valves were shut (normally open), only to be opened for testing and sump pumping operations pending resolution of the failure of the U/2 '

I valves.

i Q

January 7, 1988 l.. _ Removed and disassembled the solenoid on the F003 valve with no problems identified. '

(' 2. Performed an air cleanliness test of the air supply to the F003 valve and *

' ** a test of the pressure regulator supplying the F003 valvo. No problems were identified.

January 8, 1988 1

l 1. Removed the F020 valve solenoid for on-site vendor inspection and l

performed a pressure regulator test for its air supply. No problems were identified. l 4

W I l' 2 - _ - _ _ _ _ _ . _ _ . . _ _ _ _ _ _ _ _ . _ _ . _ _ , _ _ _ _ ___t.___ _ _ , _ -

gy .. ... .. .. , . ,,,,,,,,,,,,,,,,.,,,,,,,,

. ...a. us.. .  ;

uca stt aveNT RtPORT (LER)  %%'",0,otag e Ma, ***;i,"*,',,,,8?'l*,.? T,;' . !,','  !

i TaxT coNTINUAT60N R,74MS,2*S*g',*.1F,#f,'p';' .*iMN '

@','.'.'#Jo*.?'7&a.4,'r ,

os manacimwt a=o.voost. as e=oto=.ocseen 2tM. 4%

8aceLif y Isaast tu D0Cetti h6s.sht. L81 L h 9, Lamb 88 tel Pa.6 481 l

... t!te nr.?.R Brunswick Steam Electric Plant Unit 2 015 0F 0lo-0 l6 l0 l0 l0 l 3l2 % 8 18 -

0l0l1 -

0 l8 ,

rart ua e w .asenwme a asiawon January 9, 1988

1. Performed a visual inspection of the logic relays A71-K17 and A71-K18 (E11S/JM/RLY). An arc strike was identified on terminal 4 of the K18 l relay and visual evidence indicates that contact 3-4 had been welded  ;

closed as a result of the are strike.

2. Initiated activities to remove one valve actuator (Miller air motor) i (E11S/JM/84) for inspection, ,

a

3. The F020 valve solenoid, removed.on 1/8/88, was inspected on site by the -

vendor with no operability problems identified.

.- 3 January 13, 1988

1. Removed and inspected the in-line air filter (EIIS/JM/FLT) supplying the '

^

NO3 valve. No problems were identified (filter was very clean with  ;

approximately three years operating history). .

4 -

January 14, 1988

1. Replaced the A71-K17 and K18 relays, General Electric Part No.. ,

CR120A06002AA, on U/2, j

=* January 19 and 20, 1968 1.

Removed and inspected the valve actuator for the F020 valve. The r internals of the actuator had a liberal coating of grease in accordance ,

with vendor recommendations.

Corrective Actions -

i*

Based on results of testing / analysis performed to date, a definite root cause has~not been identified. Vendor inspection and analysis of the valves'  !

solenoid valves could not identify evidence which would explain the failure of +

the valves.to open. The inspection did reveal the presence of an oil base film  ;

in the solenoid valves' internals; however analysis of the oil film determined 1 it was oil used by the manufacturer for component assembly. Laboratory work at j

the Carolina Power & Light Company metallurgical and failure analysis facility f indicates the incurred failure of the subject solenoids may be the result of a higher than previously expected rate of degradation (oxidation) of the solenoids' valve seat material, othylene propylene diene monomer (EPDM). This is believed to occur when higher than anticipated ambient temperatures in the solenoid valve bodies, due to the solenoids' being normally energized, combine with the presence of copper in the solenoid valve bodies, which are made of brass. In order to further identify the root cause(s) and required correction actions relative to this problem, the services of a contractor have been retained.

w.c e , sana mi -

w am ** * ' """"** ... . e, "J.f*""9 4,

,. UCENSEE EVENT REPORT (LEm P,l,U,"4'f,*, M , % ;'. T 4 ,3 lf T.7 ,"o't. L ]

. TEXY CONTINUAT60N RR7,','.%',lf,*g;!@,".;1','#^'MT.' *Eff!  ;

U,.?MiT.'Jf" o manacewswta=o a 'B'o**i=oto3"#$fli tit.voosv'.E.4e seen  ;

nosom anani ni occasi =vu.ia m 6. um . . m Bruncwick Steam Electric Plant Unit 2 0 l6 j o l 0 l 0 l3 l 2 l4 8 l8 -

0 l 0l1 -

0 15 0l9 0F olg von v .sw c .ess.wasic s manwnn

' l On May 25, 1989, action was completed to install Viton-based seating material in the subject Unit I valves' solenoids to serve as an interim corrective measure until the subject failure mechanism is clearly understood and final corrective actions are 2mplemented. Completion of corresponding interim action' on Unit 2 is expected by December 1, 1989. Viton has been shown to withstand temperatures which would normally degrade EPDM-based material. In addition, the scope of this interim corrective measure has been expanded to appropriately ,

include other normally energized ASCO solenoids on both units utilizing EPDM as i a seating material (reference Engineering Evaluation Request 88-076 Rev. 1).

Until this action is accomplished, these solenoids will be cycled on a weekly basis to help ensure their operability. An update regarding the results of this effort will be reflected within a supplement to this report to be submittod

' .on or before June 15, 1990.

As a result of this event, the on-site Quality Assurance group performed a surveillance activity (QASR 88-007) on the failure of the valves to close.

Event Assessment ..

This event was assessed to determine 'if the event would have been more severc +

, under reasonable and credible alternative conditions as defined in NUREG 1022 (and supplements). This assessment determined that the first event (scram) would not have been more severe as this is an analyzed event in the safety analysis and the plant systems are designed for adequate mitigation. The second event (valve failures) would also not be more severe in that neither a reasonabic nor a credibic alternative condition could have provided a source term within the drywell. Without the source term, there is no increase in the quantity on material which would be released through these nonisolated penetrations.

A review of our records indicates that the failure of the valves to close is an isolated event for LER reporting criteria (redundant system failure);

however, a condenser vacuum problem did initiate a scram in 1985 on U/1 and was reported in LER 1-85-008.

[

l nac t sena em .

- -