ML20042F580

LER 90-012-00:on 900406,identified That Procedure for LOCA Would Not Ensure post-LOCA Core Flush Would Be Initiated in Time to Prevent Boron Precipitation.Caused by Personnel Error.Configuration Mgt Program strengthened.W/900504 Ltr

ML20042F580
Person / Time
Site:	Calvert Cliffs
Issue date:	05/04/1990
From:	Denton R, Volkoff J BALTIMORE GAS & ELECTRIC CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-90-012, LER-90-12, NUDOCS 9005090098
Download: ML20042F580 (7)

Text

.. .

. . . . . . . . - _ _ 4

-u

~f t "' .k*. BALTIMORE 1

p"

- GAS AND ELECTRIC CHARLES' CENTER e P 0, BOX 1475 e BALTIMORE, MARYLAND 21203 1475 r-FL E. DENTON .

MANA0ER- I

. CftVERT Ct6946 NUCL(pA i PC AER PLANT DEPARTMENT

j

.i

, + Mayl4, 1990.

U. 9. Nuclear Regulatory Commission Docket No. 50 317' f" Document. Control Desk License'No. DPR 53 .;

Washington, D. C. 20555 i

Dear Sirs:

d The. attached LEP/ 90 12,- Revision 0, ;is being /mnt. to y .es required under 10 CFR 50.73.

l e

Should you have any questions regarding thisi mport, we w e'... be pleased to g ,

discuss them with you. '

Va g truly yours, , ,,

'k . .;

J. ,

s Vr R.'E. Denton bx- 'l l

g r

, 1

& ~

J JV/ymb j

cc: Thomas.T. Martin ~

X Director, Office of Management Information l

i I; and Program Control A Messrs: G. C. Creel

1. - C. H. Cruse i

{ J. R. Lemons

.L. B. Russell 1

_g g R. P. Heibel 7

<<, , l 1 g

L .

i r . l p

a e

y .

4 T .

i f _

'+

9005090096 900E6%

[ PDR ADOCK 05005317 *:' g6f')fh 5- PDC 6 $[0 q j 1 r  :

+

+

$,6, e. ~

m. .

+ .

• t MPIRt $. 4/30/92 e

(

'..,. (SilMAf t0 SURDEN PER RESPONSt TO COMPLY WTH THis LICENSEE EVENT REPORT (LER) n'&"fNTs'Ul&o"NWfu'Ro"A EY?MAYt'TO T!it RICU"S AND REPORTS MANAGlutNT ORANCH (P4301, O $ NUCLE AR P PERWO K R T ON ROJ ('3 to 0 IC OP MANAGEMENT ANO fUDOET,WASHINGTCN.DC 20h03.

• ACILITv seAME 113 DOCKET Nyh00ER (23

'Calvert Cliffs Unit 1 e Pass ai TITLt 441 ol5lo10l0l3l1 l7 1loFlOl6 Procedure 6vsNT oAu m Change Creates Potential for Post-LOCA Boron Precipitation um NUMetR <ei  :

RePOR7oAftcri l

~

OTHER ACiutits iNvotyto iei MONTH oAv vtAR vtAR "=p,,* 6 gy,;;l MONTM DAY TEAR F ACMT V hAWH DOCKET NUMMRIgl Calvert Cliffs, Unit 2 o lslo g o g o; 3;1;8

- ~

0l4 0l 6 9 0 9l0 MIS 0 l 1l2 0l0 0l 5 0l4 9l 0 oistogotoi l l O* 3 RAfisse a 4 $ PORT IS SueMfTTto PVRSUANT TO TMt Rh0VIREMtNTS OF 10 CFR I JChere cae o# a'ere of 888 fo#e**aff lill 30 4 i., ,, 4 ,e , - ,

.,34eHan,,, ,3,,,,,

20 408teH1 Hit _l

_ to 36teHit 90.73teH2Hel F3.f t tel f 1106 ] l 80 405IeH1HIH 90.38telf2) 80.73teH2Hva0

_ OTMGR sA 20 405(eH1 Hull 90.73te H2 Mil

_ 30.73teH2Hv44H A) Js64) i 30 4054eH1H6vl to.73teH3Hel

_ 30.73teH2HvuiHet 30.4064eH1Hvl 30.734sH2Hi41 90.734eH2 Hd LICENS$$ CONTACT FOR TMi8 LI A Ital NAME TELEPHONE NUM0ER ARGA C004 John Volkoff, Compliance Engineer ,

30l 1 i 2i6to 1-13 16l4 19 I COMPLETC "JNE LINT POR.1ACM COMPONENT f Altunt 08SCRietD IN TMit RIPORT (131

'C AU S t' $YSTEM COMPONtNT M ' O' "I'g ,Tn A E g g "

CAUSE Sv8 TEM COMPONINT

[ R

,TqA E (

• I I I I I I I '

I i i i I.l- 1 j l l I I I I I I I I l' 1 I t j SUPettMENT AL REPORT EXPECTf D ties MONTM OAv ytAR .

-] v ts tu ,*, ee.,-* raPictio svewss,0N oA res $Utvi ION Ae: T R ACT w, M , co T] NO l  ; q

e. , e , e.,. e.,, ,,,,, a ,,. o ., l j

,,,.e., ,,ie,, ,,,e.,

On April 6, 1990, it was identified that based on the exining design information, the procedure for a Loss of Coolant Accident (LOCA) would . nct, ensure a post LOCA core flush would be initiated in time to prevent boron - precipitation. The Emergency Operating Procedure (EOP) had been revised on February 10, 1988 to increase the delay in the initiation of = the post LOCA core flush from 8 to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. However, this change was based upon calculations performed in 1975 and did not account for changes in licensed power level or boron concentrations in the Refueling Water Tank (RWT), Safety Injection Tanks (SIT), Boric Acid Storage Tank (BAST), and Reactor Coolant System.

Analysis using approved methods and conservative inputs indicates that boron j precipitation could occur as soon as 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br /> after a LOCA.

'i The February 10, 1988 E0P change had the potential to allow boron precipitation - to '

occur and interfere with long term cooling following a loss of coolant accident The cause was personnel error. The procedure change was made without adequate investigation of the design basis. Thirteen year .old design calculationsa.n were used without recognizing or accounting for changes in the plant which changed the initial conditions and assumptions in the design calculation.

Corrective actions include strengthening configuration management program, improved procedure review ' and approvals, ensuring safety evaluation of procedure changes are performed when required and correcting the E0P.

'NRC Persi 364 (6 496

?

POAtl 300A U.S. iUCLEM ltG LULATOAY COsmesNIOes .

exmise LICENSEE EVENT REPORT (LER) '""^" " ""'""'*"'""'"

ao Ae'ae"cMi TEXT CONTINUATlON E',fsg"#'U'"oin'o'S""?"no"i

.gAgug A,ng #1 s'un'Ao'lT,5 '

O!?u!&"T.UM,Io?s '

&%3" ' * '

L -,

S ACILITY NAAIL til DOCKET NURIBER (3) LER NUMet A (4)- PAGE (3)

YEAR > " $$',4 r "y'18,$ l 1

l Calvert-Cliffs, Unit 1 o l5 l0 l0 l0 l 3l1 l 7 9l0 0 l 1l 2 0l0- 0l2 or 0-[6 TEXT f# move ausse e moused, use asnessner anC Femi m/ (th I. DESCRIPTION OF THE EVENT (

D i

Boron precipitation is a concern during long term c^oling following a cold-leg l

break Loss of Coolant Accident. The low core flow under those conditions allows boiling to concentrate"the boric acid contained in the coolant to the solubility [

l limit where precipication would occur. ' Boron precipitation has the, potential to ~!

interrupt coolant flow and could cause additional core damage following a - LOCA.

The post-IhCA core flush is intended to prevent this precipitation. 7 On April 6, 1990,'it was identified that' based on existing design information, time initiation requirements for post LOCA core flush in the Emergency Operating 7 Procedure (EOP) for a - Loss of Coolant . Accident : (LDCA) had not been properly analyzed to ensure that boron precipitation would not occur. -The E0P had been revised on February 10, 1988 to increase the delay in the initiation of the post LOCA core flush from 8 to 24 ' hours. However, this ! change was based 'upon calculations performed ~in.1975, and did not account for changes in licensed power level or boron concentrations in the Refueling Water Tank 1RWT), Safety Injection i Tanks (SIT), and Boric Acid Storage Tank (BAST). The NRC was notified on April 6, 1990 in accordance with 10 CFR 50.72. Analysis using approved ' methods and conservative inputs indicates that boron precipitation could occur as soon as 18 ,

L hours after a LOCA.

At the: time the procedure problem was identified, Calvert Cliffs Unit I was - in I Mode 4 (Hot Shutdown) at a temperature of 270 degrees Fahrenheit and a - pressure j l of 350 psia. Unit 2 was defueled.

L t l

II. BACKGROUND l Following a LOCA, heat removal is accomplished by continuous boil-off of water in For a large . cold leg break, only .the : amount of injected the reactor vessel. 4 water required for decay heat removal is. delivered to the core; the remainder spills out of the break. 1 As water boils off to . cool the core, sclubles contained in the injected water l concentrate in the core. Water . from the RWT , SIT, and BAST contain large l concentrations of boric ' acid as- required by Technical Specifications.

Additionally, until August 1976, a temporary neutron shield consisting of bags containing boron was installed above the reactor vessel . head. These bagsqwould ,

be a very large contributor to the post-LOCA boric acid inventory in the l containment.

The rate at which boric acid could accumulate in the core following-a LOCA is a function of the volume of water in the containment, the total mass of baron in solution in that v. ce r (including the contribution of the tempora *y acutron shield), the reactor vessel mixing volume and' the amout.t of decay heat. . With these initial conditions, boron concentration in the core versus time following a large cold leg break LOCA was ,

i s

NRC Fonn SEGA (640) .

.. I

U.8,88UCLE AA L81ULAT0nY C0 emes 0SION t.en.t.>PORM i 306A mMD We NM606 -

- EX, LIES 4/30c2 LICENSEE EVENT REPORT (LER) ISs"^20~ 'u ME,'40'5Eit/ .of"I.VM"$

C

,, TEXT CONTINUATION ign'.",'o',"! .^3ctt"f'.M!,'e'#,'NU,".' "MNfi u

m, o u,WLi?Ta*uimia,NOToN, nAceusNT ANo evoort nAsM DC 20603."c3%myia l

1. Aciuty NAme m oocust Nousen m ~

un sevusen to PAos m v Aa .

• • gp .a,5yy,y;'

l Calvert Cliffs, Unit 1 0 l5 l0 l0 lo l 3l1 l 7 9 10 0l1l2 -

0 l0 0l3 OF 0 l6 taxT wm ==m w . sun-wmer assmm calculated by Combustion Engineering.-- The maximum allowable delay in the i initiation of the post IDCA flush is the time when the boron concontration i

. reaches the solubility limit. The ccncentration verses time curve and the supporting calculations are contained in a letter from BG6E to the : NRC dated April 16, 1975. This document is incorporated into the des 1 nF basis as reference (28) in the Safety Evaluation Report (LER) dated March 14, 1977,  !

\

On May 9, 1975, Combustion Engineering supplied a second ' graph with curves .;

showing boron concentration versus time both with snd without the boron from the  ;

temporary neutron shield. This letter did not restate the assumptions for the [

curve, but referred to the earlier related correspondence. The temporary neutron j shield was removed in August 1976. - The removal of the temporary neutron shield i removed approximately 48% of the boron used in the April 16, 1975 calculation.

Removing this- boric acid increased the time to reach the boron solubility limit ' {'

and is a conservative change. Therefore, the post-LOCA flush initiation time was i not changed, i

Conversely, core power level (and thus decay heat) was increased and maximum boron concentrations in the RWT, SIT, and BAST were increased during the period j between April 1975 and December 1987. The amount of boron added by these changes  !

and the increased rate of boiling from decay heat had' a smaller effect than the amount of boron removed with the temporary neutron shield. The effect of thesc  ;

changes on post-LOCA concentration of boron and the maximum time - permitted to L initiate the post LOCA core flush were not formally're-analyzed. .There was no . j formal documentation and incorporation into the design basis (either directly~or e by reference) that the initial conditions used to calculate the time to reach the l boron solubility limit had changed between 1975 and 1987'. [

l In October 1987, the Operations Section requested the Analytical Support Unit to determine if the requirement in E0P-5 to initiate - a post-LOCA- core flush could be extended beyond the existing 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> limit or possibly' eliminated. The removal =

of the neutron shield bags in 1976 was the suggested basis for removing this requirement.

On October 20, 1987, the Analytical Support Unit (ASU) supplied the Operations Section procedure writers with the May 9, 1975 Combustion Engineering curves-showing boron concentration versus time with and without the neutron shield. The ASU indicated that the curve justified extending initiation of post LOCA ' core flush from 8 to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. As stated earlier, this document did not include the q calculation or identify the initial conditions used to generate the curves. As a '!

result, changes in core power level and increases in boron concentration -in the RWT, SIT, and BAST since the curve was generated were not recognized by the Analytical Support Unit or communicated to the Operations Section procedure writers, e

EC Form 3ASA 1649) -

P0nte3054 . U.S. NUCLEA4 kt0ULATORY COceletOS40se N.R.C.r APPROYED OMS NO. 81604104 E XEES. 4/30/92

• T LICENSEE EVENT REPORT (LER) : 78lsi,','AgiDE.MM'5o*nT%eg,y,7,N,Twip , ,'

• • ' ^

4 TEXT CONTINUATION  ?,"*'E4'

%"M."#MWRn,'!# C s' "'CMf! '

1 P APE RWO RE U T ON J (3 SO4 04 0 IC '

OF MANAGEMENT AND BUDGET,WA$HINGTON,DC 20003.

F ACILITY NAA46 (H DOC 8Lif NUMBER (2) L4R NutdSSR 10) PAGE 13) vl&R a

QU'M' g b .

."y'1*M Calvert Cliffs, Unit 1 0 l6 l0 l0 l0 l 3l1 [ 7 9l 0 -

0 l1l2 -

0 l0 0 l4 OF;0 l6 TaxT rn ==. as = , ww. um i.,me p.,,,, assa im III, CAUSE OF THE EVENT P

The cause of . the event was personnel error, .The time requirement' for L the

'l

~

initiation of the post LOCA core flush was changed without an adequate review of -

~

the design basis. Thirteen year old design calculation'results were used'without -!

recognicing that the assumptions and initial conditions were no longer valid because of design and Technical Specification changes between 1975 and 1987, i Contributing ~ factors include: q

1. The FSAR does- not identify that a minimum and maximum tima limit exists for the initiation of . the. post-LOCA flut.h . The criteria needed'.to establish these . limits (equipment qualification, decay heat) boron inventory);are'also-not identified.

2. The Emergency Operating Procedure for a LOCA and-the specific provision for a post-LOCA flush are discussed in the FSAR. . However, . the l procedure . time _

limit for the initiation of the post-LOCA' flush in the. procedure was revised without performing a 10 CFR 50.59 Safety Evaluation.

3. The calculation Atermining the time - to initiate the post-LOCA flush (April. I 16, 1975 letter) was not. updated when the initia1' conditions and assumptions affecting this calculation changed (power level and boron' concentrations in. f

, RWT, SIT, and BAST). The-generic cause of the error was already' identified, l as part of the Calvert Cliffs " Performance Improvement Plan" J (PIP) .under ,

I root cause 7: Insufficient depth of. assessment and root cause-analysis.

.IV. ANALYSIS Post-LOCA boric acid concentration versus time was calculate'd based on current 1 operating parameters and conservative methodssconsistent with the April 16, 1975 .i submittal from BG6E to the NRC. The time.to reach the boron solubility limit was.

calculated to be .18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br />. Plant operating procedure corrections are based on l this calculation.

A subsequent analysis was performed using an updated method of decay heat  ;

calculation and crediting additional reactor vessel volume available for mixing, j This analysis determined that~ boric ecid precipitation would not occur for at I least 26 hours3.009259e-4 days <br />0.00722 hours <br />4.298942e-5 weeks <br />9.893e-6 months <br /> af ter a LOCA. The E0P iniciated core flush within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after i a LOCA. Therefore, this event did not constitute'a significant safety hazard, j l <

t

[

c a

_ 80RC Penn 30BA i&40)

L g -(

1 . . . . . . .

NRC Ponet te64 U.S. NUCLEA.1 Lt.ULATA.Y COMMtOStoes

' EMIRES' 4/30/22 . . ..

,',, LICENSEE EVENT REPORT (LER)- f#st'#lo'fMM'CM'5EliT' .#EY ,T."Wi"n's

.? TEXT CONTINUATION- N"*'",'o' nt M M^"$May@!"nUle'n,'W".'"5CfRA 1 PAPERWO Rt(% t'O N h 04 8 09 MANAQtMENT AND SUDGET, WASHINGTON,DC 20003.

F ACl417Y NAME On DOCKEY NUMBtm (2) LER NUMeth 101 PAGE (31 L jtna > " M ', -

"'1".W Cnivert Cliffs. Unit 1 0 l5 l0 l 0 l 013 l 1l 7 0l0 0 l5 0F itXT 17 auro spese 4 coeussef, use eeWroner NRC Fenn 3054 W {1h 9 l O{--J 1l2 -

O' [6 f!

i V. CORRECTIVE AriQQg .;

+1 h

1. The boric acid concentration in the core following a LOCA was re-analyzed i using current plant parameters. The re-analysis determined that boron- -l precipitation could occur 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br /> after a LOCA.

2. E0P-5, " Loss of Coolant Accident," was revised to require' the initiation of the post LOCA core flush between 8 and 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> after the LOCA. The change:  :

received a -10 CFR 50.59 Safety' Evaluation which was reviewed by the plant <j Operations and Safety Review Committee and approved by the Plant Manager.

3. Calvert Cliffs Instruction (CCI)' -101 " Implementing Procedure Development-and Control" is being revised to include a formal- screenin6 Process to identify procedure changes which require.a 10 CFR 50.59 Safety Evaluation. '

-y

4. The'Calvert Cliffs " Performance' Improvement-Program"'already identified the underlying causes of this event during-the PIP process. -The PIP providesfa-systematic program 'and schedule for improvements .and verifying that, the .{

actions taken are effective. The PIP includes the following completed 'j actions which are applicable to this event..

Formal training -in performing safety evaluations and. how.to identify when a' safety evaluation is required.- (PIP Section 4.5);- 1 Establishment of the Configuration Management Unit which is charged with accumulating and consolidating the . design' . basis to . reduce the effort required to verify that plant changes do'not adversely affect the design basis (PIP Section 5.3.4);-and StrengtheninS of the Plant Operations:and Safety. Review Committee including the esta;'.ishment of a subcommittee 7 which - provides safety'  !

assessments of proposed procedures and procedure changes - (PIP Section 4.7). 4 i

i i

i

9 s

. S s- ,

a s-i NRC FeMR 300A l$49) . '

, . . g

.o i y A

L . . . < . ..

NRC postas assA U.s. NUCLEAA ie0VL Atony cOamsestoss -j '

CASI - 2 APPROVCD oms NO.31904104 -

EXTlXE8 4/30E2 $

1 umD eU20eN eER Response TO COMPtv WTw ms UCENSEE EVENT REPORT (LER) INFORMATION COLLECTION r.EQUEST: 60 0 H;.8. PO ;W1hD . 1 I

0 ~

' ' a' ^" '"o .uRo N umAu TO Tua R CORD  ;

4 4

TEXT CONTINUATION *AN"D"RE"PO'RTS MANAGEMENT BRANCH (P430). U S.

REGULATORY COMMISSION. WASHINGTON. DC 20666. AND TO 1HE PAPERWORK REDUCTION PROJECT (31500104). OF F ICE OF MANAGEMENT AND OUDOET. WASHINGTON.DC 20603,

[ FAcatary NAMt til DOCKET NUMBER (2) Lf R NUMBER (G) PAOf (3)

YEAR 8%Q, "

uYaYn l

ol6 or o 16 calvert clirts, unit i oislololol3l1l7 9 l0 -

o l 1l 2 -

o lo

Text w a= esse = w = mMiMasmac re- essa on y i VI.; ADDITIONAL INFORMATION l' A. Previous Similar Event  !

[-

\

, No similar events have occurred.

i .

i B. Affected-Comnonent Identification '

s

IEEE 805. IEEE 803

Comnonent Svstem E . Component-4 Residual Heat Removal / BP ~N/A 3 j Low Pressure Safety Injection System - '

4 d

i  !

.1 l

a  %

I a i i

i r

e l 'f r

L i

j-  ?(

L:I i t p 3, t

k q

i 1

4

} /

.l

• r

.l 10 l

=< > i

i lly

'qq -. l' A' ll l >ll s ..

1 w+ =Reiwassaim .,,, .;.

t

.c 3

[}lj'; s{ 'su ['i: ~I. s. .. '" 'y < , f> j L /, . ..;, ljgi he

~

,.s 1, ' s .'.ql~ l' '.

..n > > t L. ,

u/ -