ML19332D625

LER 89-055-00:on 891025,reactor Protection Sys 1A Shunt Trip Breaker Tripped on Undervoltage,Causing Loss of Power to Distribution Panel 1AY160 & ESF Isolation.Caused by Relief Valve Failures.Isolations reset.W/891127 Ltr

ML19332D625
Person / Time
Site:	Limerick
Issue date:	11/27/1989
From:	Endriss C, Mccormick M PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-89-055, LER-89-55, NUDOCS 8912050053
Download: ML19332D625 (8)

Text

m ,

-

!

,

' '

'

>

J[,c

,

10 CFR 50.73 '

.,

,

I

{l I t

,

.-

,

,

-

PHILADELPHIA ELECTRIC COMPANY LIMERICK GENEM ATINf8 ST ATION P. O. DOX A L S AN ATOG A, PENNSYLV ANI A 19464

!-

(216) 3271200 sat. 2000 l

'

u. a. .co nwic a. )... u. November 27, 1989

........'0'..".'.',L*."......, Docket No. 50-352 -

p License No. NPF-39 O U.S. Nuclear Regulatory Commission

.

A t.tn t Do!.:Went Control Desk

'

, ,

,

Washingten, DC 20555 1

e'UBJECT: Licensee Event Report ~

Limerick Generating ftation-- Unit 1 This LER reports automatic actuation of the Nuclear Steam Supply Shutoff System, an Engineercd Safety Feature (ESP), due to a Reactor Protection System /Uninterruptable Power supply breaker trip'from an unknown cause. This LER also reports two Steam Leak Detection ::ystem (an ESF) initiated isolationa of the Reactor Water Cleanup system following the breaker trip.

.

Reference:

Docket No. 50-352 Report Number: 1-89-055 Revision Number: 00 Event Date: October 25, 1989

• 4 Report Date: November 27 , 1989 Facility: Limerick Generating Station P.O. Box A, Sanatuga, PA 19464 This LER 16 being submitted pursuant to the requirements of 10 CFR 50.73(af(2)(iv). This LER-is being submitted late because of the complexity of the event. We regret any inconvenience this ,

may have caustd.

Very truly yours,

/ ..

.s k

&Q '

VAW:ch /

,

cc W. T. Russell, Administrator, Region I, USNRC P T. J.-Kenny, USNRC Senior Resident Inspector, LGS MP l

I 8912050053 891127 l

, PDR ADOCK 05000352 l, S PDC t l>

1 . _ _ . - . . . _ , _

... _ . _ _ _ . . _ _ _ _ . _ _ _

-

.

m.c p. Jns

• • • ' uA mucttaa (icwhatta7 c;mm.unica

.

• 4omio cus .o. i.e

.

, LICENSEE EVENT REPORT (LER) twis oru=

. .

81CikITT haasi (1) DOC 1 7 hw>tia m .a G 4 .J.

~

Lincrick Generating Station, Unit 1 oistoto 101315 12 1 lCFl0 l?

"'" Nuclear Steam Supply Shutoff System isolations due to static inverter trip and i Beactor Water Cleanuo Isolations due to relief valve failures.

a v e .,, o.n m

'

3 i u n au-.in m I ai* cat care m i er ia naciunis invetvio m l 7 hr- eay v tan vtaa a "'*.',0. 'j *".".01 Scat" '

c L 'i'a ' a ct u" h..is ccci at =v.it aisi o isio t oio s',  ;

~ ~

1l0 2l5 8l9 8l9 0l5 l 5 0 l0 1l1 2l7 g l9 ois sogo,o, , ,

,,,,,,,,,, rais . .oav is iv mis evenua=, to ran aiew;.t=nn c io cen pew ., . ,.. , nu l

-con m 1- nem, nam

_ w.nums.

L u.num,-i _

n.,uu 1

,,,,i, n..mnnu ummm ,m.,

u,v,p 1, 0, O ~~Ia.mmm g, nammai

] ummm ammai

--

] . .n mm ....

] ,,,,,g g ,,_ l

.m e a n .i mi.- m e uw W a:' ~ s .

n aosuomem H.numm u.ni.ams.. wen v ;' -

, a.mn n., u.nmmu. - '

w.n:.,m .. j uci~ sin e =v ct .ca t is un nu u.i '

n u e i.v.sia 48 d

• C;Ci C. R. Endriss, Regulatory Engineer, Lirerick Generating Station 21115 3 2: 7 -i 1: 2 0: O cow

• Life Chi teki sca sacM CC=PChikt f astumi Crt:Ria:0tmtmis an.cattisi

<

cavne overi , cc..e =t * ';** * *- "l','"l.'!:" [g cavi i.,i. co.. eses, e.gc y;;emy,jgi

'

-

<

r.a .e X CE Ri v Li2i6 (5 N g , i -

i ! i e i 'WU T

-

1:I s Q% qP 1 I I l l l I -

1 I i l- l l } W j

'

su uoi ra . .or n,ie i n.i

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

. U'J,il'd D vis o,,,.. ex*rerto suwwont carts T) mo *'n n 'i

, ,

. . . c1 ,s . . .o ,, ,, . . . . . -,.. n.,

On October 25, 1989, at 1700 hours0.0197 days <br />0.472 hours <br />0.00281 weeks <br />6.4685e-4 months <br />, during the transfer of the

'1A' Reactor Protection System (RPS)/Uninterruptable Power Supply static inverter to its preferred DC power source, the '1A' RPS

, shunt trip breaker tripped on undervoltage, causing a loss of

power to the 'lA' RPS distribution panel, 1AY160. As a result, l various automatic Nuclear Steam Supply Shutoff System (NS4)

! isolations, Engineered Safety Features, occurred. Reactor i l

Enclosure and Refuel Floor ventilation systems isolated and the

,

' Reactor Enclosure Recirculation System and the Standby Gas  !

! Treatment System started. The 'lC' Reactor Water Cleanup (RWCU) l pump failed to trip after the isol'ation. Except for the RWCU j System, all isolations were reset and nystems returned to service by 1735 hours0.0201 days <br />0.482 hours <br />0.00287 weeks <br />6.601675e-4 months <br />. The RWCU System isolated twice more during

- attempts to restore the system following the NS4 isolation. The cause of the RPS breaker trip and the failure of the RWCU pump to

.

! I I trip are unknown. The causes of the other 2 RWCU.isolations are j

a closed steam flooding damper (SFD) and 2 failed heat exchanger relief valves, respectively. RWCU was returned to service on October 28, 1989. One relief valve was replaced and the other blanked until a suitable replacement is available.

• A new model valve will be tested on the Unit 2 RWCU System to determine suitability. A re-evaluation of the failure mechanism is being performed.

l l

l g,. 2..

__,_. -

_

.M Ng j

. p ,

.

w. wa nwesen ess u e...e w ..we. ,

'

,

-

LICENSEE EVENT REPORT (LER) TEXT CONTINVATION **aovns owe no.a m -o e4 so gsw.i <

g pacasty na.s sip 886 8 8 7 *W** 8 a 48' gla hvoets see ,ast Lt l

Limarick Generating Station, Unit 1 "" N "h' W" b" - o -

818181014131512 8l9 -

01515 -

01 0 0l2 c7 O l */

r nr ., - . = . - c w =. ,, nn

, Unit 1 Conditions Prior to the Event:

Operating Condition: 1 (Power Operation)

Power Levels 100%

'

'

, .

Description of the Events

,

On' October 25, 1989, a non-licensed plant operator was

• transferring the '1A' Reactor Protection System ,

(RPS)/Uninterruptable Power Supply (UPS) static inverter (EIIS INVT) from its alternate supply to its preferred DC power  ;

supply _following the completion of corrective maintenance on

.certain capacitor connecting links of the '1A' RPS/UPS static inverter. During this transfer, at 1700 hours0.0197 days <br />0.472 hours <br />0.00281 weeks <br />6.4685e-4 months <br />, the '1A' RPS shunt trip breaker (EIIStBKR) tripped on undervoltage causing a  ;

loss of power to the 'lA' RPS power distribution panel (IAY160). +

,

This loss of power resulted in automatic Nuclear Steam Supply

. Shutoff System (NS4) (EIIS:JM) isolations of the following systems or subsystems (an Engineered Safety Feature (ESP) actuation), closing their inboard primary containment isolation valves:

o Drywell Chilled Water (DWCW) (EIIS:KM),

o Reactor Enclosure Cooling Water (RECW) (EIIS:CC),

'

o Primary Containment Instrument Gas (PCIG) (EIIS LK),

o Reactor Water Cleanup (RWCU) (EIIS CE), ,

o Residual Heat Removal (RER) heat exchanger vacuum breaker lines (EIIStBO),

l

'

o Primary containment exhaust to Reactor Enclosure Equipment

-Compartment Exhaust, and o Drywell Sump, Suppression Pool Cleanup (EIIS CG) and Traversing Incore Probe System (EIIS IG).

i l

g,.... -

y

_ _ _ _ ___ _ - - _ _ _ _ _ _

'

.

-

. 4 w h..

," w t avca.. .eavue... es es. ,

,

LICENSEE EVENT REPORT (LER) TEXT CONTINUATION maoves aos me sm- ::a i

,

ne a ss. ni...

sacmTV>+ 4tu 888"87 **'e s t m Lgm numeta is . ass ca l

Limerick Generating Station, Unit 1 "'

• l * "MMMW 818 t o l o 1413 I 5l 2 819 -

01515 -

0l 0 0l3 n, 017 rar .- e = ~ :. - ~~~- n n- sw , un >

The following systems received NS4 isolation signals but no valve repositionin'g occurred since the associated valves are norinally closed during power operations:

.

o Main Steam line drains (EIIS SB) and sample lines,

.

o Reactor Sample Line, -

o RHR shutdown cooling, sample, and drain lines, ard o Primary containment purge supply and exhaust (EIIS VA).

Additionally, the following ESP or partial ESP actuations occurred:

• o Reactor Enclosure and Refuel Floor ventilation systems (EIIStVA) isolated, ,

o Reactor Enclosure Recirculation System (EIIS*VA) initiated, o Standby Gas Treatment System (EIIS:BH) initiated, o An 'A' channel half scram occurred, and '

o A half Main Steam Isolation Valve isolation was generated.

<

Main Control Room (MCR) operators recognized the loss of power to

.the panel and restored PCIG, DWCW and RECW within approximately three minutes using the system isolation bypass switches in accordance with Event Procedure E-1AY160, " Loss of ?A RPS and UPS Power." Operators reclosed the shunt trip breaker, restoring power to the 1AY160 Panel from the alternate supply and reset the remaining isolations by 1735 hours0.0201 days <br />0.482 hours <br />0.00287 weeks <br />6.601675e-4 months <br /> in accordance with procedure GP-8, " Primary and Secondary Containment Isolation Verification and ~.Reset."

All isolations and all followup system actions occurred as designed except in the RWCU System. When the icolation valves for the RWCU System closed, the '1C' RWCU pump did not trip. MCR operators attempted to trip the pump from the MCR but were not successful. The pump was tripped by an operator opening the pump motor breaker.

l At 1810 hours0.0209 days <br />0.503 hours <br />0.00299 weeks <br />6.88705e-4 months <br />, operators were returning the RWCU System to service using the 'lB' RWCU pump when the Steam Leak Detection (SLD) systera (an ESP) (EIIS:IJ) caused an isolation of the outboard RWCU containment isolation valve on a high RWCU area temperature signal. The steam flooding damper in the ventilation duct leading to the room was found closed. Aftcr the damper was

. . ,

1 . . - .

- l

i e

e .

e meg aerm asse W L hWCklaa A43ptate., ggwee s.tige "8'

. .

tlCENSEE EVENT REPORT (tfR) TEXT CONTfNUATioN *aaovis owe meau-rea

-

s wa ss. sv.n sacMTV e6=s lia . escan awmaantai ,,g g g,  ;

L,,,,,,,,,,, g Limerick Generati.ng Station, Unit I na m "22W,'#.# ,

1510 l e le 13 l Si 2 819 --

015l5 -

01 0 Ol d 5 0l7 verta, - .. .

opened and water was removed from the damper's pressure sensing

'line, RWCU was placed in the gravity blowdown mode (reactor to main condenser bypassing the RWCU pumps rather than reactor to reactor). When the 'lB' pump outlet valve was opened at 2248 hours0.026 days <br />0.624 hours <br />0.00372 weeks <br />8.55364e-4 months <br />, the system isolated on a high RWCU heat exchanger room temperature from the SLD system. The relief valves (PSV-44-108 -

and 109)*on the RWCU regenerative heat exchangers were found to-be leaking steam to the room. On October 28, 1989, at 1620 hours0.0188 days <br />0.45 hours <br />0.00268 weeks <br />6.1641e-4 months <br />, the system was returned to service after replacement of

,

PSV-44-108 and blanking of PSV-44-109 under the administrative control of a temporary circuit alteration.

A four hour notification was made to the NRC at 2055 hours0.0238 days <br />0.571 hours <br />0.0034 weeks <br />7.819275e-4 months <br /> on ,

October 25, 1989 and at 0234 hours0.00271 days <br />0.065 hours <br />3.869048e-4 weeks <br />8.9037e-5 months <br /> on October 26, 1989, in accordance with 10 CFR 50.72 (b)(2)(ii),.since this event involved automatic actuations of ESPs. This written report is being submitted in accordance with the requirements of 10 CFR 50.73 (a)(2)(iv).

Consequences of the Event:

There was no release of radioactive materials to the environment as a result of this event. With the exception of the RWCU System, all systems responded as designed.

Except for the RWCU System, the isolations were bypassed or reset and the systems restored quickly.enough by operators in

'

accordance with plant procedures to prevent any adverse impact on plant systems. The time that RWCU was out of service caused a discernable increase in reactor water conductivity; however conductivity remained within the limits of Technical Specifications. Reactor water conductivity returned to pre-event levels within two days after the RWCU System was returned tom service.

l l

'

Immediate and follow up actions for this type of event (loss of power to a RPS panel) are provided in procedures E-1AY160 and GP-8. Additionally, licensed operators receive requalification training to review and practice responses to simulated plant transients of this type. This training reinforces immediate operator actions, minimizing the time that systems are isolated i and reducing the impact on the plant.

.

1 g, .. -

l

.

'

,

,

.

. . . us.=wcaansa: een : *c.

.

"

• UCENSEE EVENT REPORT (LER) TEXT CONTINUATION saaoves ove na s s -ca.

- . ..ns. sv ei oggMTY eamstu sccat? mwa.44a azi gla eveneta te 9.e 5 las l

Linerick Cenerating Station, Unit 1 "** l7 "E W I* W M e is t o t a le 1315[ 2 819 -

01515 -

0l 0 01 5 et 0l7 ren , , - . . . ..c , -, nn Cause of the Events

-

.

The cause of the actuation of the RPS undervoltage relay is unknown. Investigation by station engineers found that the output voltage of the static inverter was lower than normal due to a loose connector wire on a capacitor in the inverter. The loose connector was not one specifically worked on during the maintenance being performed. However, the connector could have been unintentionally jarred during the work on the other connectors. The lower than normal output voltage was above the undervoltage relay trip setpoint and the undervoltage relay setpoint was verified to be correct, however, the lower than normal output voltage may have contributed to this event. The relay we.s replaced on October 26, 1989. .The removed relay will be sent to its manufacturer for evaluation.

The cause of the failure of the '1C' RWCU pump to trip is believed to have been due to sticking of one of the pump trip relays. This, however, could not be reproduced during troubleshooting.

The second isolation of the RWCU System while the system was- ,

being initially returned to service is due to the closed steam flooding damper in the duct leading to the 'lB' RWCU pump room.

The closed steam flooding damper interrupted pump room ,

ventilation. The lack of ventilation allowed the pump room temperature to increase, resulting in the SLD system isolation of the RWCU System. The steam flooding damper had closed due to filling of the pressure sensing line with condensed steam from a previously identified steam leak from a valve in the area where the sensor is located. The isolation of the Reactor Enclosure Ventilation system due to the above NS4 isolation signal allowed this steam to condense in the ventilation ducting in this area.

The sensing line is located, as designed, at a relatively low point in the duct and acted as a collection point for the condensate.

The third RUCU isolation was caused by RWCU regenerative heat exchanger relief valve leakage increasing the room's temperature.

Prior to the event, both relief valves were leaking a small amount of steam. The combined transient of the NS4 isolation of the RWCU System and the subsequent failure of the '1C' RWCU pump to trip caused a pressure transient in the RWCU System. This pressure transient resulted in additional leakage. When the RWCU regenerative heat exchanger relief valves pass water, the water flashes to steam as it encounters the lower room pressure. The flashing water erodes the relief valve seats, causing them to leak more. When attempting to place the RWCU System into g, . -

I

^

,

.

.

..ee. ,... = . auc.sa. .inusareo ca.-*s.

-* * " "

UCENSEE EVENT REPORT (tER) TEXT CONTINUATION

• aovis ovm *- s*

• se :. a v :

pact 4tvma=4su d4 stet awmela 42' tan muwet a tea ,ast tp l

Linerid; Generating Stat. ion, Unit 1 "" N "R'W'MA*N c p f o lo l41314 2 819 -

0l 5l 5 -- 00 01 6 c50l7

~

se'$vTce*~~Ilie""$ $w T high energy water, re-established through the regenerative heat exchanger, leaked through the relief .

valves. The flashed steam vented to the room via an open funnel drain, increasing the room's temperature. The increased temperature was sensed by the SLD system triggering the RWCU -

System isolation.

. .

.

Corrective Actions:

Except for the RWCU system, the isolations were reset and the systems restored by 1735 hours0.0201 days <br />0.482 hours <br />0.00287 weeks <br />6.601675e-4 months <br />. The loose capacitor connector we.s tightened restoring the inverter output voltage to normal.

Ap a conservative measure, the RPS/UPS undervoltage trip relay was replaced at approximately 2000 hours0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> on October 26, 1989.

The 'lA' RPS/UPS static inverter was successfully transferred to its preferred DC power supply on October 26, 1989.

The RWCU System was returned to operation using the 'lB' RWCU pump on October 28, 1989, after one of the leaking safety relief j

!

valves was replaced and the other valve replaced with a blank.

The steam flooding damper was maintained in the open position  ;

under a temporary circuit alteration until the steam leak could  !

bs controlled. While the damper was maintained open, a second l damper in series with the first damper remained operable, i

i Actions Taken to Prevent Recurrence:

The other 'lA' RPS/UPS inverter capacitor banks were checked for proper capacitance to verify that there were no other loose connectors. Various measures to_ ensure that all the capacitor l connectors remain firmly seated are being evaluated to prevent reduced output voltages. These include " pull" testing during preventive maintenance or soldering the push on connectors in place. If a cause is found for the trip of the undervoltage relay, actions will be taken to ensure that similar relays do not have the same defect.

If a cause is found for the failure of the '1C' RWCU pump to trip, actions to prevent its recurrence will be developed; however, this is being considered an isolated instance. The use of the type of relay involved in the failure of 'lC' RWCU pump to l

trip is being evalua.ted. $

g, ..

_

.

h

. .

..e,- . n ~ a .. .. mu.. . o-u .,

"*'

-

,

tlCENSEE EVENT REPORT (LER) TEXT CONTINUATION *=o.nz ows ** * *-r.a

,

n ..n m n

....u n .. . n. = = n ~-n . m u......, ..n.

l Linerick Generating Station, Unit 1 .

"" W" O #C'S 8 Is lo la le 1315l 2 8l9 -

0 l 5l 5 -

0l 0 01 7 c5 017 a - . ~, . - e , .. n n The valve leaking steam which caused the steam flooding damper to close will be repaired during the next plant shutdown of t

sufficient duration.

As a result of this and other similar RWCU isolation events, a re-evaluation of the potential failure mechanisms of the relief valves is being performed. As part of-this evaluation, a -

different model relief valve will be installed on the Unit 2 RWCU System during the next Unit 2 outage scheduled to occur in December 1989. Other utilities are currently using this valve in the same application. The valve's operation will be monitored to determine whether it is suitable for permanent use in the Unit 1 and Unit 2 RWCU Systems. If the new model valve is not suitable, additional actions will be taken until a suitable valve and system configuration are obtained.

Previous Similar Occurrences:

LERs84-030, 84-040,85-007, 85-008,85-011, 85-024,85-026, 95-048,85-074, 87-021,87-027, 87-029,87-038 report NS4 isolations due to loss of power from a RPS/UPS static inverter.

A majority of these events are due to either a blown fuse or a defective voltage regulator card and, as such, actions developed to prevent their recurrence could not have prevented this event.

LERs 86-40,88-009, and 89-033 report isolations of RWCU due to leaking relief valves. Previous evaluations of the relief valve l failure mecnanism were inconclusive. In light of the new failures, the potential failure mechanisms are being re-evaluated for a more definitive cause and development of effective corrective actions.

L

'

Tracking Codes: -

X - unknown B2 - Failure due to Abnormal Wear l

l f pe*= a