ML20028H849

LER 90-012-00:on 901229,control Room Operators Received Annunciation of Reactor Trip.Caused by Rod Control Sys Failures.Failed Cards in Rod Control Sys replaced.W/910128 Ltr

ML20028H849
Person / Time
Site:	Prairie Island
Issue date:	01/28/1991
From:	Hunstad A, Parker T NORTHERN STATES POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-90-012-02, LER-90-12-2, NUDOCS 9101310104
Download: ML20028H849 (6)

Text

.

Northem States Power Company 414 Nicollet Mall Minneapolm, Minnesota $54011927 Telephone (612) 3N5500 January 28, 1991 10 CFR Part 50 Section 50.73 Ud Nuclear Regulatory Commission Attn: Document Control Desh Washingtsu, DC 20555 PRAIRIE ISLAND NUCLEAR CENERATING PLANT Docket Nos. 50 282 License Nos. DPR 42 50-306 DPR 60 Unit 2 Renctor Trio as a Result of Rod Control System Failures The Licensee Event Report for this occurrence is attached.

This event was reported via the Emergency Notification Syst<>m in accordanen with 10 CPR Part 50, Section 50.72, on December 29, 1990. Please contact us if you require additional information related to this event, hite v Thomas M Parker Manager Nuclear Support Services c: Regional Administrator - Region III, NRC NRR Project Manager, NRC Senior Resident Inspector, NRC MPCA Attn: Dr J W Ferman Attachment i

i a

m1 5

% . n 0.L U U O 4

ut ,,oca.. ...vu tea Co ....N gC,,,o n. . ,,,,,,,, ,,,,, .,,,,,,..

, ,,i . .. mm ,

f , 13tsuattD gyaCtN Pf a allpokst to Cowe(v wtw twig -

1 LICENSEE EVENT REPORT (LER) E*.'s'~'r's'Us*Ne'i 80c'sWs't'ila'0*tEi a s C" a ei a

1 emo a g gyasPCat,t wamaGluthi t,4 t e n C wwissicN Saancm it g331. uccs ;csts,4=a wasmihCtem NWCLE Am to tag pastawcas agevcitoN pacstCT 13I1041041 C 8 8 8C E cs uamactusme Ano SW0087.wasme=Ct0N,CC;Cleg, i 8 Atlutv mar *G 118 pocatehwuela us **66 4 pRATRTE IST.AND NUCl. EAR CENERATING Pl. ANT UNTT 2 o l5 l 0 ; o l 0 1 3 10 16 1 l0Fl 015 rit a ...

UNIT 2 REACTOR Trip AS A RESULT OF ROD CONTROL SYSTEM FATI.URES eveht oate m i tan =vuese m I s eroa t ents in i ofwis e aciutils i=volvio m wo=rNl oav l v Aa l vsaa l: J"Rl,g/6 T,'y,gluo=ti car jvtaa s awes . =u occast avussaisi oistolotoi ; i

~ ~

1l2 2!9 9 0 9l 0 0l1 l 2 0l0 o ll 2l8 9!1 oistoios o! :

ines atroat is suomittsD PvasvAsst to tut steytasuaNTS C810 CFa 4 sca re e.e es mwe es one seue==,s litt 0's a afine N 20.403161 20 aollel XX 60J3isit2Hwt 73J1tml Powla 20.404(alt 1Hil MJeleHit W.734sH2 Het 72J14sl Livsk rm e a ss,.,* .. .,

flei 11 0!O ro.aoei,iiiHel WJai.ia so.rai.H:H

- - - .o .. ..* - T e. ~n c se-OU N 4081emHiel W JJiaH2Hu W.731sH2HMal Jsdas

, ,9 ,'  : , , , ,

<+oe 20.dosisH1Hhl MJ3tsH2 Hall 60J11 alt 2ltwaH98 a':A i ,

, ' '-" = -

% 70 4081aH1Het tufatalcifial to 731sH2Hai UCtNiti Contact 80m TMiS 48a1129 gAug it LEPHONG NWM45m aata COct Arne A Hunstad Staff Engineer 611 !? 318 l R l ! l ! 1 I 211 COM*ttti Okt Uht PCa laCN Cous0NENT f AILus t OtlC#fst0 IN this 8 8Pont (13:

CAu$t SY$t tu C0wsoNggt W C. atpearsett df { CAWSt STrflW CO MPON E N t WaNuf aC-tUnta a t aca7a s t t-toneagg UT.o s

s.

n 4. .a[#

s .

s 'g:p s y

^ '

X A!A i ! I Wl 11210 Yes 'E M I I I l i I I ' '

Q g. r

, l l l l l l l Nh .

l l l l l l < s SVrettutNT AL st

• Cat t RetCit0 (143 WQNtM l day lYIAa sue wisssoM Yti ilt ver. temet - LR!tCTIO SVonstLSICN Ca til NO l l l sss,asc? u . uw ,, a. ,,,,. ., am . ,-~. a o tui  :

On December 29, 1990, Unit 2 was at 100% power. At 2234, Unit 2 tripped.

Control room annunciators, and the sequence of events generated by the plant process computer, indicated that the cause of the reactor trip was a negative flux rate trip from the reactor protection system.

An investigation into the cause of the negative flux rate trip revealed that ,

two circuit cards in the rod control system power cabinet 1BD had failed, The first failuro was of a transistor in the urgent failure alarm circuit. The second failure was the opening of a solder connection on the stationary gripper regulation card, causing the reference voltage to go to zero. In response to the reference voltage going to zero, the urgent failure alarm circuit should have generated an urgent failure alarm and a " hold" current that would be-applied to the stationary grippers for all the rods supplied by that power cabinet. This hold current would have prevented the rods from dropping into the reactor. But since the Urgent Failure alarm circuit had also failed, no current was supplied to the stationary grippers in control rod Bank D and they fell into the core, causing the negative flux rate trip.

The failed cards in the rod control system were replaced. Unit 2 was returned to service at 1330 on December 30, 1990.

.a C . n ,, . . w,

_ _;_ -c,- . _, _ ~ . _ _ _ . _ _ . _ _ _ _ ._ _._.._ _ - _ _ _ . _ _ __ _

.. . . ..~.....os............. , , , , , , , , , , , , , , , , ,

n ,..II .

i.1CENSEE EVENT REPORT (LER) ',"l".'o'!,8,e',2d',*el'!;Mii,,!f",l .T,".'",1 TEXT CONTINUATION *?,",",',",'o', *,18,$25'Qi*d!'/"','!/J v '"/ *$$fj

. a r t e.o o Tsom (3 Ifs c!

Cif asahAGlut h? .atNO 4WDGif,WalmiNGY04.DC 20Ma.

, ws. , , ... n , own aw=ev ui ua =W=et. isi * *GE mi t.. "* M t*  :

0'#3:

Prairie Island Nuc necr Plt Unit 2_ c16l0l0l0l3l016 Gen,maasnn 9 !O -

01112 -

o lo 012 M 0 I$

una,,,,., . ,

EVENT DESCRIPTION -

On December 29,1990, Unit 2 was at 100% power. At 2234, the Unit 2 control room operators received annunciation of a reactor trip. The operators verified the reactor trip using control room indication and stabilized the unit in the hot shutdown condition in accordance with plant procedures.

Except as noted below, plant systems and components responded to the reactor trip as designed. Control room annunciators (EIIS Component Identifier ANN),

and the sequence of events generated by the plant process computer indicated that the cause of the reactor trip was a negative flux rate trip fr.n the reactor protection system. A negative flux rate trip is consis, tent with a dropped rod or rods.

An' investigation into the cause of the negative flux rate trip was initiated

} following stabilization of the unit. The investigation initially centered on g possible blown fuses in the rod control system, llowever, no blown fuses were j found. Further investigation revealed that two circuit cards in the rod i

control system (EIIS System Identifier AA) power cabinet 1BD had failed.

The first failure was of a transistor in the urgent failure alarm circuit.

Because there are no alarms or annunciation associated with the failure of an urgent failure alarm circuit, the control room operators were unaware of the failure, j The second failure was the opening of a solder connection on the stationary gripper regulation card, causing the reference voltage to go to zero. In response to the reference voltage going to zero, the urgent failure clarm circuit should have generated an urgent failure alarm and a " hold" current that would be applied to the stationary grippers for all the rods supplied by that power cabinet. This hold current would have prevented the rods from dropping into the reactor. But since the urgent failure alarm circuit had also failed, no current was supplied to the stationary grippers in control rod Bank D and they fell into the core, causing the negative flux rate trip as they left their withdrawn position.

l

=ac e-ima cei

-- - ~ . _ ..-

j e '

• g eggan3ess W.S. .uCLsan a Gutaicar coumets.ow apreovto o 3 a o.31604104 S EPla tl. orsorgt 3

!- 1.1CENSEE EVENT REPORT (LER) l"e".'J#.'J'e80 U"ol'!Tvi's,I'8"J.Y %".?a'o i TEXT CONTINUATION - f ".'".'A'l'.'ffd ?J",*00'c'"!2 '"' 00iNf!

t u.

o,i",,'a."J.J.i""'M.2,',o".'!.".Z.i.n.

o voon. 1o . "J.J."c#

m u. 9??a

.......m oocan av=*ia m u na,v ea i.i ease un vi.a  !, H ,[, '6 (  ;

,y,"

i Prafrie island Nuc Gen Plt Unit 2 0 l5 l0 { 0 l 0 l 3l0 l6 9l 0 *-

nlll? -

0l0 0l3 M 0 l5 '

rut ar . n w. .-- aac w ma amn

?

I

.There were several secondary effects of the unit trip:

A control rod drive mechanism fan, a spent fuel pool cooling pump, and an auxiliary building normal ventilation supply fan tripped as a result of the automatic transfer of the power sources supplying the nonsafeguards 4160 volt buses. The equipment that tripped was restarted.

Some turbine drain valves indicated that they were not fully open. A work request was initiated to correct the malfunctions.

The nuclear instrumentation source range channels had to be manually re-

, energized because the intermediate range channels were undercompensated.

This undercompensation prevented the intermediate range detector output from dropping below the level where the source range detectors would be 1 automatically re energized. (This is an expected condition after a long period of operation at high power.) The intermediate range channels were recalibrated.

The failed cards in the rod control system were replaced. Unit 2 was returned to service at 1330 on December 30, 1990.

CAUSE OF THE EVENT A study of the rod position history performed after the unit restart, which utilized data from the plant process computer, indicated that the control rod a Bank D rods began to drop before the remainder of the rods. The results of this scudy confirmed the cause of this trip to be a negative flux rate trip caused by the unannunciated failure of the urgent failure alarm circuitry, followed by a failure on the stationary gripper regulation card, which allowed control rod Bank D rods to drop into the core.

ANAINSIS OF THE EVENT I

This event is reportable pursuant to 10CFR50.73(a)(2)(iv) since it involved an

-unplanned actuation-of the reactor protection system. The health and safety

- of the public were unaffected by this event. The reactor protection system 1 responded as designed to the negative flux rato caused by the dropped rods and l the plant was stabilized in the hot shutdown condition in accordance with i plant procedures following the reactor trip.

mac - ma iu.i

l IsA PM.I $$64 W.S. NUC4 tan 85 GULAt03 v COMui9510's g inPiett 4/30/92 I

LICENSEE EVENT REPORT (LER) ,',8,'n','#,o',."M ?t* o "'!@il,'*,#'i'.',' .77.'.'l,% -l TEXT CONTINUATlON *i,"'.",ll,",l"ff

.  !$'a life',", ','!M/ ONf!

W,'t!.e!!"rdfe",c*'Mie?"nWk Mia 08 esamAGluth? Amo eWOG87.wasnimorose.DC 20W3, escissiv asaast its DOCK 87muuss. m LI A muut4 A (68 9 A G t t3l iaa " R!.W."  : "';". .U prairie Island Nuc Gen Pit Unit 2 0 l8 l0 l0 l0 l 3l0 l6 910 -

0 ll l 2 -

0 lo 0 I4 0F 0 l5 text,, . c wane w sw mn i I CORRECTIVE ACTION '

An investigation into the cause of the negative flux rate trip was initiated 1

following stabilization of the unit. The investigation initially centered on possible blown fuses in the rod control system. Ilowever, no blown fuses were 3

found. Further investigation revealed that two circuit cards in the rod control system power cabinet 1BD had failed. The first failure was of a i transistor in the Urgent Failure alarm circuit. The second failure was the opening of a solder connection on the stationary gripper regulation card,

causing the reference _ voltage to go to zero. The two failed rod control system circuit cards were replaced. The two failed circuit cards were repaired. The cause of the transistor and solder connection failures will be addressed through the followup system assessment studies described below.

Increased rnaintenance has been done during recent outages ir. response to rod control system malfunctions that resulted in Unit 2 trips in December 1989.

Specifically, the Unit ? rod control system was inspected, repaired and tested by the vendor during the September 1990 refueling outage; the failed cards that caused this event were included in that-activity. Based on this event, the Westinghouse maintenance organization will be contacted and the adequacy and quality of their maintenance program will be evaluated.

The problems resulting from secondary effects of the trip were corrected before restart as described above.

Several other corrective actions have been taken or are under consideration.

Thece actions are designed to either minimize the potential for further rod control system component failures or to improve the ability to identify failures before they result in. operational problems. These corrective actions include:

l e A periodic walkdown inspection of both units' rod control systems has been instituted in an effort to determine the condition of rod control system circuit cards and fuses at power.

j Thermal monitoring of cornponents in the rod control cabinets is being done l to establish baseline information. This information will be used to periodically _ determine if_ components in the rod control system power cabinets are operating at elevated temperatures. Elevated component temperatures could be indicative of a component failure.

L n

o 1 NGC f enn W M$l '

~ _ ._ __ _. _ _ ._ _ _ ___. . _ _ _ . _

88 1 01M 3063 4A3. NUCLEAR $$GW4&f 06Y COMMalatON

. . . , . . . . m. c LICENSEE EVENT REPORT (LER) ',','s",','!,'o',,v" R"t u"o " '!;T',',, #"'.y ,"e',", .'.",,'j TEXT CONTINUATION R*".'.",'e'.*,f'",,d u"f u ll,'e!','!2 '"! "M*0!

,a,sa.o mg  ?"iO N 13 to i? O aC

• Os asamaGgoswt aug ouDGET,o A&MamCYom DC 20603.

,acsksiv Names m Docaat muussa L21 tla kVuetM fop ,act t3l sa. "Sm;'t'l < c'JJ,i:

Prairie Island Nuc Cen Plt Unit 2 01510 l0 t o 13!0 l 6l910 -

Ol1 l 2 -

0 10 0ls OF 0 Is ruin - . - .-.,...e w ,,nn

- Replacement of old or repaired alarm cards is being considered,

- The establishment of baseline test point measurements for an operating rod control system is being evaluated. This would provide baseline data for the rod control system cards. A deviatian from the baseline data would indicate a change in operating characteristics of the card, which would be indicative of a card component fcilure.

- Online monitoring of alarm card functionality by the plant process computer is also under consideration,

- A study has been initiated for the assessment of system lifetime and failure modes, and to ensure appropriate maintenance practices. Study results will be used in the determination of che need for system upgrades or replacement.

FAILED COMPONENT 1DENTIFICATION Failures occurred on an Urgent Failure alarm circuit card and a stationary gripper regulation card in a Westinghouse Full Length Rod Control System.

PREVIOUS SIMIlAR EVENTS Previous similar events were reported in Unit 2 LER 89 004. The transistor that failed during this event was not the same transistor that failed during the events described in Unit 2 LER 89 004.

81

.