ML19332F776

LER 89-013-00:on 891110,reactor Scram Occurred Following Main Turbine Trip,Causing Initiation Signals on HPCI & RCIC Sys Due to Spiking of Level Transformers.Caused by Design Error.Hpci Turbine secured.W/891211 Ltr

ML19332F776
Person / Time
Site:	Limerick
Issue date:	12/11/1989
From:	Endriss C, Mccormick M PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-89-013-02, LER-89-13-2, NUDOCS 8912190009
Download: ML19332F776 (8)

Text

<w 10 CFR 50.73 i

!i PHILADELPHIA ELECTRIC COMPANY LIMERICK GENER ATING STATION P. O. BOX A

• SAN ATOG A PENNSYLV ANI A 19464 (215) 3271200 ext. 2000 )

u. .a. u.co n uic x, s.., p.e. December 11, 1989 u - . . . 7 ". ', .".' ",' *! ". . . . .., Docket No. 50-353 License No. NPF-85 U.S.-Nuclear Regulatory Commission  ;

Attn: Document Control Desk l Washington, DC 20555 1

SUBJECT:

Licensee Event Report j Limerick Generating Station - Unit 2 J l

This LER reports a Unit 2 reactor SCRAM, as the result of a .

Reactor Protection. System (RPS) actuation. The RPS actuation l resulted from Turbine Control. Valve fast closure caused by an  !

actuation of a generator overall differential current relay. -

1

Reference:

Docket No. 50-353 Report Number: 2-89-013 Revision Number: 00

!- Event Date: November 10, 1989

!. Report Date: December 11, 1989 t l ,

Facility: Limerick Generating Station P.O. Box A, Sanatoga, PA 19464 o This LER is being submitted pursuant to the requirements of L' f -10 CFR 50.73(a)(2)(iv).

o Very truly yours, P

' I C

DCS:ch cc: W. T. Russell, Administrator, Region I, USNRC T. J. Kenny, USNRC Senior Resident Inspector, LGS fgf 8912190009 891211 ,

I PDR ADOCK 03000353 S PDC

gl .e , u 6 a 6. . ,

UCENSEE EVENT REPORT (LER) "a"**

IS$k Generating Station, Unit 2

" ' " , " , 'l" "l3S3 ' , "i ilo l7

"" Peactor SCPM frm high oower due to a generator trin on mqh ciiff erent.tal mrront frcrn an inmrrect setnoint, eveu, e.,e . te. es. . v mn .n . e...siu,se. .v.6ve. .

e DAY YSA. YSA. '

.Mb -

Min ecom,se .av VSA. eace6 sty ssaes cas, g g.g 81s lo l o i s i l l 1l 1 1l 0 8 9 8l9 0l1l3 0l0 1l 2 1l 1 8l9 ,,,,,,,,,, , ,

-.-.-,.--,....,.,m,

-. 1 ., ., ~.

_ 2 _

,, 0, 9, 8

. .m.

.-m _ _

y

. .m

.~.

_ _ _.m ,, ,

3 _

.m.

.m.

t b_ -

m

.m.,

t.cse e cos.v.c, ... . Lt. H.,

5 R. Endriss, Pegulatorv Engineer, Limer.ick Generating Station ,,, ,, "'*"'**"

21 i 5 32 ii;ii37 l 2O0i n u .. . u, . . , .

""" '""" ****'*"' "$Ya "tEo '

$$$hk *^u" "*"" ******' "iTe." "I'."I.if h[jfhkh!fkkff ,

e , , i i i i SW , , , , , , , gggg l' I I I I I l 6

~

m .. . . c .. H.

' 'I l l , ; j i ;

hkh,, ,,

eSE.os, v..-,,==, g anai c ,uw. em. m . __.;_= . _

, nei O November 10, 1989, a_ Unit 2 reactor SCRAM occurred following a Main Turbine trip. All control rods fully inserted as designed.

The turbine trip resulted from a Unit 2 generator trip caused by actuation of the A-phase overall differential current relay.

Following the SCRAM, reactor _ vessel pressure increased to 1125 psig and level decreased to approximately 0 inches instrument level. Operators restored normal operating vessel level and pressure. The High Pressure Coolant Injection (HPCI) and Reactor l Core Isolation Cooling (RCIC) systems received momentary initiation signals due to spiking of level transmitters. The RCIC system did not initiate and the HPCI system partially initiated due to the short duration of the initiation signals.

Both systems were operable and able to respond to a valid l initiation signal throughout the event. Both systems were l

secured and all initiation logics were reset. An " Unusual Event" emergency classification was declared and all appropriate notifications were made. The cause of the overall differential current relay actuation was a miscalculation of the trip setpoint performed during the original Unit 2 design. The calculation was corrected and the affected relays' cetpoints were adjusted. The miscalculation was verified to be an isolated event and no further actions to prevent recurrence are planned. Operators restarted Unit 2 on November 14, 1989.

m,ic ,

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

. ase <ama v. uctaaa u:vevoa, commes

. LICENSEE EVENT REPORT (LER) TEXT CONTINUATIDN amovio o=e no uso-me ,

ii.ian eme.

ene uvv naise ni ommitau== a

• u a umesa => .aes is Limerick Generating Ftation,' Unit 2 "*"" '"'*""

o ls lo jo j o l 3l5 l 3 8l9 -.

0 l1 l 3 --

O l0 0l2 or 0l7 verv u . sm mc a.- mna v nn Unit Conditions Prior to the Event:

Operating Condition: 1 (Power Operation)

Power Levels 98%

Description of the Event:

On November 10, 1989, at 1422 hours0.0165 days <br />0.395 hours <br />0.00235 weeks <br />5.41071e-4 months <br />, during the Startup Power Ascension Program, a Unit 2 reactor SCRAM occurred as a result of a Reactor Protection System (RPS) (EIIS:JD) actuation due to a Main Turbine Control Valve (TCV) (EIIS:PCV) fast closure, following a Main Turbine (EIIS:1RB) trip. All control rods (EIIS AA) fully inserted as designed. Following the turbine trip 6 l and reactor SCRAM, Reactor Pressure Vessel (RPV) (EIIS RPV) pressure increased to a maximum of 1125 psig a,nd RPV level decreased to approximately 0 inches instrument level (167 inches above the top of active fuel) as an expected result of the TCV fast closure. The turbine trip resulted from a Unit 2 Main Generator (EIIS GEN) trip caused by actuation of.the A phase overall differential current relay (587-6101A) (EIIS RLY). Main Control Room (MCR) operators restored RPV pressure-and level to their normal operating levels using the turbine bypass valves to control pressure and the Feedwater Level Control System (FLCS) n (EIIS:JK) to control level. An " Unusual Event" emergency ,

classification was declared due to the unusual shutdown of the

( reactor.

During the event, the End-Of-Cycle (EOC)-Recirculation Pump (EIIS:P) Trip (RPT) breakers (EIIStBKR) tripped, as designed, due to the turbine trip and resultant transfer of station auxiliary power from the main generator output to the offsite power source.

This resulted in a trip of both of the Unit 2 recirculation pumps. When the RPV exceeded 1093 psig, redundant control rod I insertion and RPT breaker trip signals were received as designed L from the Redundant Reactivity Control System (RRCS) Alternate Rod Insertion (ARI) and Anticipated Transient Without SCRAM (ATWS) -

RPT logics.

l Additionally, the High Pressure Coolant Injection (HPCI)

(EIIS:BJ) system and the Reactor Core Isolation Cooling (RCIC)

(EIIS:BN) system received spurious partial initiation signals due to momentary spiking of reactor vessel level instrumentation to below the initiation setpoint of -38 inches instrument level. .

The durations of the spikes were approximately 50 milliseconds.

The RCIC system received an initiation signal, however the spurious signal did not exist long enough for a full system initiation and injection to occur. The HPCI system received an

. geo.==

• • a oa u w UCENSEE EVENT REPORT (LER) TEXT CONTINUATION

....a..._

,.ciun nam o . ocean aviasa m un uuun . eau m  !

Limerick Generatinct Ftation, Unit 2

"'" " W a" i WS o l5 l01o lc l 3]5 l 3 8l9 --

0 l1 l 3 -.

0 l0 0l3 or 0l7 gen u . - we , mu w nn initiation signal long enough for partial initiation of the system however, sufficient initiation logic channels were not activated'to produce a complete HPCI system actuation and injection to the vessel did not occur.

As a result of the partial system actuation, the HPCI barometric icondenser vacuum pump (20P216) started, the HPCI' auxiliary oil pump (20P213) started and the turbine steam admission valve (HV-55-2F001) opened. Valve HV-56-2F059, which when open supplies '

cooling water to the HPCI lube oil cooler and the barometric condenser, did not open, and valves HV-55-2F006 and HV-55-2F105, which allow HPCI system injection through the Core Spray and the Feedwater system injection nozzles (respectively), did not open. 1 The HPCI turbine tripped on overspeed and restarted six separate times in a three minute period. A MCR operator observed the auxiliary oil pump cycling on and off, the tur.bine stop and control valves cycling open and closed, and the turbine speed erratically increasing and decreasing. After verifying that no true condition warranting HPCI system operation existed, the MCR -

operator secured the system by closing the turbine steam admission valve.

After recovery from the SCRAM and implementation of the necessary l

corrective actions to prevent recurrence, the MCR operators restarted Unit 2 at 0118 hours0.00137 days <br />0.0328 hours <br />1.951058e-4 weeks <br />4.4899e-5 months <br /> on November 14, 1989. The plant was. shutdown for three and one half days as a result of this transient.

A one hour notification of the Unusual Event was completed within 15 minutes of the event at 1435 hours0.0166 days <br />0.399 hours <br />0.00237 weeks <br />5.460175e-4 months <br /> on November 10,1989, in

, accordance with the requirements of 10CFR50.72(a)(3).

L Additionally, a four hour notification of the reactor SCRAM initiation was made.at 1715 hours0.0198 days <br />0.476 hours <br />0.00284 weeks <br />6.525575e-4 months <br /> on November 30, 1989, in accordance with the~ requirements of 10CFR50.72(b)(2)(ii) since a RPS automatic actuation occurred. Accordingly, this report is

_ being submitted in accordance with the requirements of 10CFR50.73(a)(2)(iv).

Consequences of the Event:

The consequences of this event were minimal. No release of .

radioactivity occurred as a result of this event. The TCV fast closure caused a Reactor SCRAM as designed and all control rods fully inserted.

. gnin=~ ~a-

l 1

U S nuCLain Et;uLatoa7 COhBaseMeese

get',ere M .

• • • - LICENSEE EVENT REPORT (LER) TEXT CONTINUATION manio om ao neo-oiw

$n91RES SQl44

. .u,, n. =u t aa= = = 6.. u . , .... . ,

I.imerick reneratina rtation, Unit 2

' ' ~ ~

., u . - , = -., n n l A turbine trip from high power is the most severe transient that

'the plant is anticipated to undergo, from an-instrument response  !

viewpoint. The level oscillations and resulting instrumentation spiking. experienced during this trip were expected based on previous testing on both Units 1 and 2 and all instrumentation responded as designed. All systems except for the HPCI and RCIC systems responded as-designed and the MCR operators successfully controlled the plant shutdown using the appropriate plant procedures. The spurious initiation signals received by HPCI and RCIC were a result of the' level instrumentation spikes. l l

Upon further investigation of the performance of the HPCI system during this event, station personnel and corporate nuclear l engineering division personnel concluded that the HPCI system l remained operable throughout the event. Had a valid initiation i signal occurred at any time during this event, the initiation signal would have existed long enough for the HPCI system to perform its' intended safety function. Additionally, the MCR operators were capable of manually placing the HPCI system in l service if needed. Prompt operator-diagnosis and action resulted I

in securing the HPCI system during the event, precluding any l I

damage that may have occurred if the cycling had continued.

Subsequent investigation of the HPCI system identified no damage or-deterioration of the system as a result of the cycling.

Therefore, the HPCI system remained operable throughout the event.

The investigation also concluded that the RCIC system remained operable throughout the event. Had a valid initiation signal occurred at any time during this event, the initiation signal would have' existed long enough to properly initiate the system.

Additionally, the MCR operators were capable of manually placing the RCIC system in. service if needed.

The following automatic RPS actuation signals also occurred that would have actuated a SCRAM if the TCV fast closure signal had failed, o Reactor water level < +12.5 inches o Reactor pressure > 1037 psig o Turbine stop valve closure ,

~

The RRCS ARI response provided a redundant method of control rod insertion and would have successfully shut down the reactor if required.

g.s. uu

. -l g'o mea . .

us.=ue m s :a:v m oav em anes LICENSEE EVENT REPORT (LER) TEXT CONTINUATION

• paarvso one no me eie. -

tapings el3146 pasekITV senast us occn81 = men sai gg, gg,,,,, ,,, ,,,, gy, .

Limerick Generatino Ftation, Unit ?

"" " # N o l5 lo jo lo l 3l5 ] 3 8l9 --

0 l1 l 3 _.

0 l0 0l5 or 0l7 vanw-. aac w an.wnn The RRCS ATWS recirculation pump trip-logic provided a redundant trip signal if the EOC-RPT logic failed to trip the recirculation ,

pump breakers.

I Cause of the Event:

The cause of this event was a utility engineering design error.in the calculation of the main generator high overall differential ,

current relay trip setpoint. This error occurred during the original design of Unit 2. This miscalculation resulted in the determination of a lower than necessary setpoint. This later resulted in premature operation of the overall' differential current relay.

The cause of the RRCS Alternate Rod Insertion and ATWS recirculation' pump trip' logics was an increase in reactor pressure to above the actuation trip setpoint of 1093 psig. The increase:in' reactor pressure resulted from the main turbine trip from high power.

The cause of the partial initiations of the HPCI and RCIC aystems was the momentary spiking of the RPV water level instrumentation.

This spiking of the level instrumentation was the result of the level and pressure perturbations that occurred immediately following the turbine trip and reactor SCRAM. The level spikes were too short-in duration to fully initiate the HPCI'and RCIC systems. The partial initiation of the HPCI system resulted in '

the cycling on and off of the system.

Corrective Actions:

Immediatedly following the SCRAM signal, the MCR operators '

verified that the SCRAM occurred properly and restored normal RPV level and pressure. In accordance with Transient Response Implementation Plan (TRIP) procedure T-101, "RPV Control," MCR operators controlled RPV pressure with the main turbine bypass '

valves and RPV level with the FLCS. After verifying that no true initiation signal existed for HPCI or RCIC system. initiations, .

the HPCI turbine was secured and both systems' initiation logics were reset by the MCR operators. Following the SCRAM, the Shift Technical Advisor (STA) performed General Plant (GP) procedure, GP-18, " SCRAM /ATWS Event Review," to verify all appropriate alarms were received and proper actuations occurred.

. gp.. uu

g' ' sies u. cuctsa ueuatoav comune

~ LICENSEE EVENT REPORT (LER) TEXT CONTINUATION . m oviooweno me-eio.

tupla88 SfJ146 encam massa m . occast avassa m ten muusen sei ... m I

,,.. . we =v.% . .... l Limerick Generatinq ftation, Unit 2 '"" ~~"

{

o p lo lo lo l 3l5 ) 3 8l9 --

0 l1 l 3 0 l0 0l6 or 0l7 1

. . --- -w~- wm

)

l Actions Taken to Prevent Recurrence: I I

Prior to restarting Unit 2, the calculations for all three Unit 2 '

generator ove'rall differential current relays (587-6101A, B, and C) were corrected and the trip setpoints were adjusted accordingly-by the Instrumentation and Controls (I&C) group.

Upon investigation, station personnel determined that the ~,

miscalculation affected. trip setpoints on only the Unit 2 overall differential current relays. The corresponding Unit 1 relay

, setpoints were verified to be calculated and set properly. This  !

event is considered to be an isolated occurrence and further i actions are planned to prevent recurrence.

Shortly after the shutdown, the HPCI system was inspected to y determine the cause of the cycling. Other than the need for a minor recalibration of the governor control circuitry, no i abnormalities were identified. After recovery from the SCRAM,

l. during power ascension for Unit 2, various tests at different L

L reactor. pressures were performed on the HPCI system. When the HPCI system was tested on minimum flow at.920 psig reactor pressure (duplicating the conditions of the original event), the overspeed trip condition did not recur. The HPCI system performed satisfactorily during all of these tests. Therefore, the cycling of the HPCI system is considered to be an isolated occurrence and is not expected to recur.

Further-investigation into the spiking observed on the RPV level instrumentation was performed during the Main Steam Isolation Valve-(EIIS:PCV) closure test performed on December 1, 1989.

Additional instrumentation was installed for this Startup Test Program test and the data verified the existence of the momentary

. spiking. This spiking was less severe than the spiking observed following the Unit 2 turbine trip and was again similar to the spiking observed during the Unit 1 Startup Test Program. Further evaluation is being performed to determine if additional actions should be implemented to reduce the effect of this spiking.

4

g' ,mu . ua muctisa co:uaronv em os l UCENSEE EVENT REP 3RT (LER) TEXT CCNTINUATION aono owe no um.co. 1

... ass em j

. .uvr ni occaer av=ea => ti . .u . .. .... .

Limerick Generating rtation, Unit ?

"'" "# U **U oislololoj315l3 8]9 -

0 l1 l 3 ._ 0 ;0 0;7 or 0) 7 sen u . s w ane w au .o nn l

l Previous Similar Occurrences: ,

Unit 1 LER 87-046 reported a reactor SCRAM from high power ,

resulting.from'TCV fast closure caused by high water level in the l moisture separator tank, and Unit 1 LER'87-048 reported a reactor l SCRAM from high power caused by high Electro-Hydraulic Control system pressure. No SCRAMS from high power have previously occurred on Unit 2. The-corrective actions for these previous l events would not have prevented this event since the previous SCRAMS were due to different causes.

I Tracking Codes: B8 - Calculation Error l

~

N$