ML20044B168

LER 90-002-01:on 900126,reactor Coolant Pump Current Monitor Inputs to Sfrcs ACH1 & Reactor Pressure Sys Channel 1 Experienced Reactor Trip.Probably Caused by Inadequate Test Switches & Isolation plug.W/900710 Ltr

ML20044B168
Person / Time
Site:	Davis Besse
Issue date:	07/10/1990
From:	Storz L, Stotz J TOLEDO EDISON CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-90-002, LER-90-2, NP33-90-002, NP33-90-2, NUDOCS 9007180022
Download: ML20044B168 (5)

Text

C: 1 I

n e

.x, ,.. . ,. -.

EI80N i ll July 10, 1990.. EDISON PLAZA -

- 300 MADISON AVENUE ..

TOLEDO. OHIO 43052-0001 Log No.: BB90-1971 i NP33-90-002, Rev.~l l

+1 Docket No. 50-346 License No. NPF-3 '

United States Nuclear Regulatory Commission

Document Control Desk Vashington,-D. C. 20555 I Gentlemen: j L

l LER 90-002, Revision 1 ,

Davis-Besse Nuclear Power Station, Unit No. 1 )

Date of Occurrence - January 26, 1990

i. 1 l.)

i.

.j Enclosed-plea:,e find-Revision 1 to Licensee Event Report 90-002. The changes.

are marked with a-revision bar in the left margin. Please destroy or mark-superseded any previous copies of this LER. .i 4 Yours truly, 1

h . $$

Louis F.. Storz  !

Plant Manager-

• Davis-Besse Nuclear Power Station LFS/plf ,

Enclosure ect Mr. A. Bert Davis Regional Administrator

.USNRC Region III ,

i Mr. Paul Byron DB-1 NRC Sr. Resident Inspector 9007180022 900710 PDR S

ADOCK;05000346 PDC

/ 9(h '

i , (

b l us,, seuoteam neoutAT0av coemssene0N I

[ was e.e. See .

M '* Appm0 Vee oms 800. 31es-4104

• V -

UCEK9EE EVENT REPORT (LER) 1 1

Occust asweteen m rases sa l enesLITV senast m Davis-Besse Unit No. 1 o l s g o g o l o g'l l 4 l6 1 [opl 0 ; 4  !

TITLE 44:

Reactor Trip from 73 Percent Due to Spurious RCP Monitor Circuit Signal eveetf DATstel Len --A lei RSPORY DATs m OTHeR F AceLittee R. .VeD lel ib 8 ACILITT esawes DOCKET NUMeente) es0 NTH DAY YGAR YEAR , 80,0, n < h80 NTH DAY YEAR c'l5l0l0l0; j' l 2l 6 91 9l0 0l 0 l 2 0 l1 0l7 1l0 9l0 o isio lo t oi l l 0l 1 ., s aww w nu Twee me 0m is euenmTTeo evneuANt 70 run necuinaisente s., is CrA o ren ., ,

,,,,,,,,,,, I T8Jfw "888

• seasem ,_

so.eesw n es = mHM E.

n Oi 7 "i 3 seanswmm

".sewm es esmweiam _ ,0Twegg.g

===wan= ==mai ==m*=<4: =  ;

e% 2; y _ _ _

essw an.m m g <,'sp(

eo ,s seasswnnm seass w m m se.Tammm m.Tswouw miswan.:

Lice 8deOS CONTACT POA YMee LOR not TeLOPMO8ee NUMesR [

esAase AmeA C00e '

Jan C. Stotz, Engineer - Maintenance Planning 4,! , 9 3,2,1; ,7,5 ,4l4 C00sPLOTe 0000 LING POR eACH COteroteestT p Altune De80Rese0 IN TMee RSP0 fit nel CAuse eveisW COMPONENT

" O' h ff '

CAUbe eYeTeM C044PONeNT "$ A8f' i ff l

X C, B , I, S , V V , 0 8,5 Y Nh'@If'  ! s , , , , , , , hhh X SI B l lRlV D l 2l 4 l3 Y E ' '

l l l l l l l supptensterTAL RepOAT ONPSCTeO nel he0 NTH DAY YEAR Yse II9 v.s. eenenen IXNCT90 SUshfAseION DMil NO l l l

.mTxACm.,. . ~. ~ . ~, , n t>  ;

on January 26, 1990, at 0846 hours0.00979 days <br />0.235 hours <br />0.0014 weeks <br />3.21903e-4 months <br />, during performance of Surveillance Test '

DB-HI-03205, RCP Current Monitor Inputs to SFRCS ACH1 and RPS Channel 1, the Station I experienced a reactor trip from 73 percent full power. Prior-to the trip, the l Station had been operating with three of the four RCPs running. Plant response to ,

the trip was normal with key parameters remaining in the normal post-trip band.

Steam generator header pressure was intentionally reduced to approximately 970 psig to aid in fully reseating two main steam safety valves solidly. After stabilization 7

of the plant, the Letdovn Isolation Valve, HU2B, vould not reopen. .The inability to restore the letdovn flov resulted in slover than normal plant cooldown and Mode 5 was achieved of January 27, 1990, at 0345 hours0.00399 days <br />0.0958 hours <br />5.704365e-4 weeks <br />1.312725e-4 months <br />.

l Immediate notification was made per 10CFR50.72(b)(2)(ii) on January 26, 1990,'at 1

1007 hours0.0117 days <br />0.28 hours <br />0.00167 weeks <br />3.831635e-4 months <br />. The reactor trip is reportable as an LER per 10CFR50.73(a)(2)(iv).

An action plan implemented to determine the cause of the RCP current monitor circuit transient concluded that there vere two major factors. They were the inadequacy of the test switches to provide current isolation and the use of a standard isolation plug which results in unintended forces on the knife svitch which carries the bypass current.

l The cause of HU2B failure vas determined to be thermal binding of the disc in the valve seat. The valve was replaced with a fully flexible vedge gate valve under HVO 2-89-0053-00. A similar failure occurred with RC-11, PORV Block Valve, during restart. It was replaced under HVO 1-90-1368-03.

L

-.e,.._______________________ , , u s m m. 9 ew e .

k yS esucklam pseutatomv Consessessoas GIIIC Pee aeta '

• • • **aovio owe s.o mo-oio4 =

UCENSEE EVENT h0 PORT (LER) TEXT CONTINUATION imme. swe FaWLITY .seast its [pocas, seuamen tal Len ,sumeen to: Past (3 ma " t' JJ. t'ra:  ;

Davis-Besse 11 nit No. I o is ;o j o 101314 l6 91 0 010l2 -

Oll 0 12 0F 0l'4 vertn- . .s s o asuw mi Description of Occurrence:

On January 26, 1990, at 0846 hours0.00979 days <br />0.235 hours <br />0.0014 weeks <br />3.21903e-4 months <br />, during performance of DB-MI-03205, Reactor Coolant Pump (RCP) Current Monitor to Sterm and Feedvater Rupture Control System l

(SFRCS-JB) Channel 1 and Reactor Protection System (RPS-JC) Chnnel 1-Surveillance Test, the Station experienced e reactor trip from 73 percent thermal power. Prior.to the trip, the Station had beer. operating with three of the four RCPs running. RCP 2-2 vas shut down January 22, 1990, as a m precautionary measure due to high indicated vibrations.

Plant response to the reactor trip was normal. Steam Generator pressure was reduced to approximately 970 psig to aid in fully: reseating two main steam safety valves (MSSVs). One of the valves that did not initially reseet fully was SP17A7, which also did net fully reseat after the January-18, 1989, event.

L The identification of the other MSW that did not fully resent is not known.

After-stabilization, the operators found that Letdown Isolation Valve, HU2B, vould not reopen. The inability to restore letdown flow resulted in the need to decrease RCP seal injection to slow the rate of pressurizer level increase. RCP L

2-2 (the shutdown RCP) seal return temperature' increased when seal injection '

flow was reduced. This resulted in seal return flow being isolated from RCP L 2-2.

The reactor trip (RPS actuation) is teportable under 10CFR50.73(a)(2)(iv).

Immediate notification was made to the NRC via the Emergency Notification System

-(ENS) at 1007 hours0.0117 days <br />0.28 hours <br />0.00167 weeks <br />3.831635e-4 months <br /> on January 26, 1990, per 10CFR50.72(b)(2)(ii).

Mode 5 was achieved on January 27, 1990, at 0345 hours0.00399 days <br />0.0958 hours <br />5.704365e-4 weeks <br />1.312725e-4 months <br /> to start the Sixth Refueling Outage.

l Apparent Cause of Occurrence:

The plant tripped when.RPS sensed that reactor power was above the trip limit ,

for operation with only one RCP running in each reactor coolant pump. RPS '

received a signal that indicated there was only one pump running in each loop because-RCP 1-2 test restoration induced a current transient in t: 1 RCP current monitor circuit making it look like RCP l-2 was off. Vith RCP 2-2 actually off ,

and RCP 1-2 appearing to be off, RPS reduced the high flux / number of RCPs trip setpoint to approximately 55 percent of full thermal power as designed. Since the plant was operating at 73 percent, RPS Channel 2, 3, and 4 tripped causing a

> reactor trip. RPS Channel 1 did not trip because it had been placed in manual-bypass per the test procedure. The trip occurred while restoring the test RCP setup that functionally checks the high and low current setpoints for RCP 1-2.

1-1 circuits had been similarly tested just minutes earlier without incident.

Subsequent testing and troubleshooting concluded that there are two major factors that can cause or contribute to a current transient. The first is the test switches used to bypass and then isolate the current monitor circuit under

.u s omm#.aua Es

]' "" 'a'

,. s ua auctian neoutavoav commeston edes ow.i amen

" ***aovio om ao .m-cos LICENSEE EVENT REPORT (LER) TEXT CONTINUATION anineswa occasi nune . m tennu . . -.... m enci6nvmaan m

m. '%t2. "' :.?:

Davis-Besse Unit No. 1 o g;oga;o 5; 4;6 9; O ,_, 0; Og2 ,_. 0;l 0; 3 o, O{4 von n -. nac e.-mnemm test. It does not provide sufficient, reliable means of maintaining. signal- I continuity to the non-bypassed circuits during testing. .The second contributor is the isolation plug which is inserted into the test switch during testing. It was noted that the plug could exert an unintended force on the knife switch which carries the bypass current. It was noted that with some repeatability the insertion / removal of the plug caused a current transient'of sufficient magnitude to trip-the current mon.itors in the other channels.

The cause of,the MU2B failure was thermal binding of the valve disc in the valve seat. When the valve was closed during the trip, the flov through the valve- l stopped. The valve cooled relative to its temperature with flow. The rigid j disc vedged into the seat as the valve cooled. When the operators later tried to open the valve, the stem pulled out of the disc.

Analysis of Occurrence:

There were no challenges to the Safety Features Actuation System (SFAS-JE) or I the Steam and Feedvater Rupture Control System (SFRCS-JB). Key parameters remained in the normal post-trip band. Minimum RCS pressure was 1790 psig and maximum was 2175 psig. Steam generator pressures ranged from a maximum of 1076 psig to a minimum of 970 psig on SG 1-1 and maximum of 1053 psig and minimum of 970 psig on SG 1-2. The transient from 73 percent vould be expected to be less L severe than a transient from 100 percent thermal power.

The MSSVs not' fully reseating did not significantly affect proper SG pressure l response as pressure was being controlled at approximately 1025 psig immediate l post-trip. The operators lowered turbine header pressure to approximately 970 psig to fully seat.the two MSSVs solidly.

Corrective Action to Prevent Recurrence:

1.

h Additional shorting devices that already exist upstream of the test switch vill be used to keep open circuit disturbances at the test switch from inducing transients _to the current sensed by the other (non-bypassed) monitor-circuits.

This vill be accomplished by a change to procedures DB-MI-03205 through 03208.

These changes vill be completed prior to their use for testing during three RCP

~

operation.

The standard isolation plug has been replaced with an isolation tool consisting of a single piece of insulating material. This vill minimize the disruptive forces exerted in this test switch during insertion / removal. The new. isolation tool has been successfully used during troubleshooting. Its use during this testing vill also be proceduralized by the change to DB-MI-03205 through 03208.

Previously scheduled inspections and maintenance were performed on the MSSVs this outage. l l

l

.....m...

. v . o,o i ....u. S l a an

Et

. . . .j<..

=

U $ 000CL.AA R.0ULA>OAV Cosen #ISBIDAs WRCP.sm M -

" **aovio omo si.o-e* -

LICENSEE EVENT REPORT (LER) TEXT CONTINUATION . .,3i, s

. .,, ... ooca v .. m t. . .,u . i., . .

L-viaa " b!?J'. .' 'l'J.l'.?

l]

O4

' Davis-Besse Unit No. 1- 0l5l0l0l0 3 46 9p _ Oj o 2 __

0 o, O'[4 l verv . . me s amm on j l L I' MU2B has been replaced under MVO 2-89-0053-00 vith a fully flexible wedge gate l H valve. The use of the fully flexible vedge should prevent further binding. A similar failure occurred with RC-11, the Pressurizer PORV Valve, during restart . l from the Sixth Refueling Outage. The valve was replaced under MVO 1-90-1368-03. I Other similar uses of the solid vedge gate valves will be evaluated for the need j to change to a design not affected by' thermal binding. j i Failure Datas l 1 This is the first LER (since all reactor trips became reportable in January j 1984) where RP5 reduced the high flux / number of pumps trip setpoint even though  : there were actually three of four pumps running. l l REPORT NO.: NP33-90-002 PCAO NO.: '90-0036 L' l I 1 l.

                                                                                                                                                                              .1 i
                                                                                                                                                                              ')

l l l' l i l l l seic .0mu 3444 e U S GPO 1988 O 624 638 456 C43I

                                        . _ . _ _ . _ _ _ _ _ _ _ _ _ _ . . _ _ .}}