ML20052D677

Safety Evaluation Re Induced Errors in Neutron Flux Detector Readings.No Changes Needed in Operating Parameters for Currently Operating Reactors

ML20052D677
Person / Time
Site:	Crystal River
Issue date:	04/22/1982
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20052D672	List: ML20052D671 ML20052D677
References
TAC-43263, NUDOCS 8205070081
Download: ML20052D677 (7)

Text

,,,..m..,

c.,

UN811 ;' ' I A t t -

/

NUCl E AR H E GUL AiOR Y L' W'.*:W' \\

I

?

~

a m..v.t % o c j e.

g-.....f

-g.

. (

r Inteod n; ion

• • '.r l '. ' n *. t

. t re, e ntly orrforu<4 5 d. O i

t' t

f1.o d. t r.

• c>r e r-a fi y a nd " is. t '-.'i.> '!

for 4,t'

<sonts than t'.v.c li,

. e f ollowt oj o>nveru t i os wi'M ~ M. !,

c-

- si

(ref. 2).

The utili t ir s

h ri;

~'

e 1

sud the roe.pon<> W. '

1 v.

.i 1.

ie nore.

In.rief the pr /1..

i. e (1 ) rr r t'ie A n m i r i s ;i o ii.. c s,

o: r 1...*

)

inerc.5.ing :he tearsi.nt f1 ax i.. c. i on

• a

( '11) * ; id the 77. a r..il ly i1 il. is.

l,1 r ej..tI n ernts :ie.,t.on :1.x

• ,..3. i i.

a* noi il. c o ' c ol r<.1...i ? t e rn ' w" s o f f " - t i v 1 ois -

t o. '..i. Je ( f o r a g h. n c oe e : or.i s ' l. -'1 ).

W a f. r t i

and potentially in an...unt !.r.ind tLit

.tr.all.

3.-

1 All of t he respon, ting ut ilitics.. o. t ild o

.o s ioll a r rc <.; un< o, b.ised on :4',.'

M

.1.* * :ns ni h e r.- in based on the calculations for the ';; e NS-W 1,4 ' i

'c.r<.

the problen concern.

Puie carried r o ". i r 4 1 f.s -

e<

th:refore, a seric**'a t di f ferent vici..

'qt.

U1 t

of po'ential flux crror irw er ises > ~ ' ai

• c ' <

8,'05070001 820422 PDR ADOCn 0300030; L--

PDR

/ -

j

't a

-s.

In *o t

t.;> era ting,tara < t c rs of t'.' f r r ".p t i f ve -

j the two preOntations 4111 c refer red to v

.e

~'.V. M % '

2'.

P.oth analyses were usrd Ir. f:, ming a.19 ; r,t for *ke r(vi.w.

t r'

T.

k J

l'v al u a.t i o.n.

Sned on the WPS$ st.ady Wl cc ecloi

'5 4t a 1 eitin).i.'.,

..r i

g-t (e

event, among the small stu. ifino ferr d, fr. :

,t.r. 1

'.rre

's, r)

e

' l un a turbine bypass with ta ak inlet (<qd f

.n

.: s. 21 d *** * - i

' < i 16*F.

Thry concluded that laeger stea,li..e

• f

'huilding preisure or variable low pr".sure trip.

t o

'e4 1

M i

stjeral overcoo'ing events.inclu fing tar'iine Ay; 's f ai1.i" h

r

..?aaline 'rnais.,-.11a

,6, f;..

and also dudied the largar

required, i

7

%mally BU has used 4 25 tr. int!'nt flux cre ar.

ru s, al 1

..;t' r

[~

assu.,cd t rrors and a trip set roint of 105.'a of fiell wr 1:.

a ti i p

.n s

5 n !.6d un A4191 calc al at I "s I frs. i u.

- s analyses eof 1171 a

'transla te, downco.wr t oo;$ era ture c honys to *. '11 t he 'i. i. ~., t,-

d. t

...i.

i c

taivera ture ret uction of.='iout 16*F corresi emis to 131 f. *fl..j N i ng trip l'ofnt of !?3%.

11oke exanined data fro' a na.'.fr of test *g

,e l

i,'_

l

^

relating tenperature and flux readings.

"4 sod un thc so t e s t s t i.e.,

.b. ', 1 1

I a relationship (linear with te.Teraturo) 50% n i nlet

'c

..,'o N.

tat a 951 confidence icvel).

It wo's141 ; revi in a 171 *4! at 10 C

4 S 'i of their analysis, however, they use d a l'

':I /l *F fai ter (l',

h, g

4 1

i.

6

~

i s

,- % r A

e b

~

,y;.

~

~

..I wi. t. '..

3 th. i r.g t Lo r 11r ui.it ed f.'JI. s i l e t fcr coc h r n *c t e r, a t i t s : ini~, i. -

.'e i :-

a relatfon'h'p betw.On reaster 1 r,

r stl t t-

• . * <. *.ir i

flon with (reor cod v2riable ir

,r.

.. e - cra *.1 t

r. J r ns i.e. ?. c t r d by the re c ',r F

. r t i. 9 i

in the Divis-Besse '.uSi.ittal).

T' ey ei a.:

.n ftig d m igr) pr.:er if strib.at icr:5.

The tr alts, b cf

's

.i1-of the in, r oved irJPR value.it the lo or i ilet '

i.

c tha t D'UR limit s (both 1. 30 and 1.43 whi.h i *... ;.

s a lu.?'

oi ri (

i ;

p.'nal ty) fall within the protec ted e. jit n ',r and Seyond,16*F overc oaling.

Ta, r.istri'ed'n, ali. slat.o

'o r

io..er condi tians

>re al so done to (.l. :k ;..yrt 4 e '. icos in !;.t -i' e' these limiting conditions.

Th~.e.^re also o

!., -.. c s t r. '

i, D'M and unter fuel i.ci t (O 't) li.aits.

• 'Ae, er for.ed plant spec t fic. uly...s

f. r

...b..

';"i

?-

. nt,

fri:1u'ing t:o turbine ! y:..'.s vr,t

( a l '.7 r i s,

. i

.l i !

a'.... ),. n.1 t h.* 1 a s.;r r s t r..I

a '. s.. t :.-

i 1 is ee :ir-4).

P 'y i...'d l '

':! /*f f o -*

.i o*

.I r

1 i fy

! <..15 ( g i < i..g a' o s t 11 ' f,91 f ar t '.o '... '

)

a t o nn :nf re o.. s cool in.J ici l

ilj,4 f.s r ' 1.. i ;J i.

Th 'y fea.4 t 4.1t D"il a'i'l Cl M 1 i i ? '. w. ce

"..t 1,

,+

reit a i f en i6ich would have S. en, revided ' y s 1,

!c

't r

o'-

o (ier.61. ) the sterlved t'JI - t,

, c r.i t..cc c, r ;r

.,m,

.+

+

.4 4

The review of all of the cu'. mitt sis has 1t id to t$a..mi r 'c :

4 nagnitude and extent of the e.ff ct and its <on

.s pe nces ' e-i ag o q' c o' interes have been suitably emined.

It.e l'.W c a f r ul a t i r a n d ' ' U.

• oca s. remants ctripicynt each ct'.cr $n tSe' ~4gni t #c of

I vs *

.s J

do the co::pleentary calculot:ons for t'.0 n/>gnito9 ef '

, a.a ao

. r. en to be considered during naxir e events. Using this info.

4 t ' nn t '.a : st'ction

.:?s en systm appears _ to be able to provide prot e. tion Sc'c,ro m " fi<ig 1 5

D'iB and CFM.., However, all f..tu re s aSai t t a t s s'. 't h r r 4.f i ca

' a l y '. i c, of j

overcooling events by B4W reac tors should in'clude in t'.e c.21y,is and g,r~,.sm-t nt s' tittals and tion an equivalent of the,infernation 'invol.ved in tF.c prt i

the use of the' penalties resulting frcci inf at r.ooling si;.iTar to t%e unless new values are justifled.

?-

.P J

The other event f avolving a pote tial indication error for the flux sby al, n

6 which in turn is involved in ten inating the.evant'hy a trip sig ial, is t he rod ejection act.ident.

In this case the error aris..s fra<a the (Lv p in F

• er distributi'on caused by the of. eted rod

'i.f ') t he e f fe: ti ve

<r r

, the.icar)1e us. 1 in I;. int level as seen by the flux detector di f fcnot fr

?

kinetics) analyses.

The vioS1m, as related tr trin, 6.ms11 only. <ist for small worth rods (neig%orhood of 0./;. '.k or l'ss) sinto 'he rite in flu Icvel is too'large to significantly af fott trip occ arr.'nce uid t 0:ing for lae ger reactivity insertions.

Since the MW "/oro po..<r"

.,v..nt anal

,as f

riornally involve high pressure trips rather thaa high fl.n trip,for smaller.

e rod worths, the probice is only relevant to the full power analyses siich are nonna11y analy, sed as tripping on high flux.

g*

e

'~T'_W

p,.

c i' 5-

,. i.

~,

T j ;,

,y...

The fttV su'mittalj. argued on the basis of engineering ju4,; nt tfat, if heat transfer out of' the feel pin during the transient woro 1 ic1 >vir 4 in t rie

.p ejection shalysts (5r -5 bas not ! cen the ( Ase in past su'di tt al s), t'.e ;.b.er l,l and eating inc# pates for the range of rractivity in'.ertion th4t i ht not 9

l J

cause flux trips hould not re'.olt in peak s nt al tfes ~.te64 tog lioit s mo cal /9m). Duke presented results of calculations of flux errors resulting 4

.J..

s14ti.,n of

[.

from a ninher of,rnd conf fsdrations, providing a basis for a corr

.. s a.

error with rod worth, and also presented ty;)ical power historios as a f,ntt ion 5

of rod worth. Free these it can be concluded that' there wN1d be a high I

f"lus trip ~ for a rod. worth above a'.out 0.1% I4 at a trip Ir.el of aheut RM

,(rather than the'gsvally assated 1121).

For rois ander this level thore might cot be a flux trip, however, power levt1s a id p'aif.g (ac tors wiciat"d 3.

i

.., 1 with these rod worths are suf ficiently low that tt e Ifmit far the v ot (PRO 3

cal /glil is not approached.

The initial transient is minar and the quisi The steady state is statilar to thit af the sing 1g ro}

ithdrai.11 c.*ent.

latter is described in the Midland SAR w'1ere It is indkated, in an in) lysis with heat transfer, that ?80 cal /gn is not a;;praad.nd (nar is 3'n c..<'r:d) a for tven f a ser rod worths than are involved h.'re (e.g., gre ater " san 0. M 9 ).

The review of the sub.nf ttals has lead to 'the conciusion t'.at the flux en ar auociated with the changed power distribution for rod ejection daes not

'significantly af fect the trip function for the laryr rei warth event s and that the consequences for the vaaller worth ev:'nts.src not of a raagnit ado to approach limits when considering the heat transfer that occurs.

(

f, '

'4

~' i s

.'s l

., -. 7 i

t

~ $.

Lnury agi dt,nclukipns,

' ;; r The ef fective neutron flux trip level in BAW rt,ic ters yay t e rai #d i

that nore. ally used in analysas 'w ange of increased flon 4*t<

u ti.m in t' n 1

docctner in cooldw. events and M iuse of pwor ( flux) StriNtiv

'- p s in the rod eje.ction event.

f' ' e ver, analysrs of as tre-a, h 9 dc n

.'s requiring high flux trip indicate that suf f !(.it nt i. irgin. aists in t he trip levels, as aupent ad by the f.:.provr, ant in ;P a:t pro <ided by ti.e cool ':tc, The that limits on,D'48 and CFM are not rzcreded in' operating r. actors.

reviE of this, analysis has re,eulted in agree.vnt with this conclusic'n for k

operating reactors. Howeyer, all future analyses of these econts for MW reactors should include in the ef fective trip levei for cooldo.:n events a suitable flu, error terrs of a n399f t.ede as discussed in this review, 0.n.

13t t.M1 for a 16*F cooldown, or as specifically d:ri.od for. the -ractor is has been done by Duke. For the rod ejection event the analysis of 'Se i screased error indicates that the only e.'cnts.which my he significontly af fectod are those with sn'111r r rod sorths for dich the e.nc., mnces are holow ifnits even.tithout a high flux trip.

The rnviow i is i t club.1

• bat no e eng s are nec& d in opos ating paraieters for cu. rently.. er.4f ing i < u tors because of thi.s errer.

8 2

ak' 6

9 4

• g 5 ;p s

7 u

a..

g

-s.

g x,

...'f:

.g i

v.-

~

a

,.se j,.

'- (8g -

i,

" f C M. g;)';;,

7 t.

a.

<. u, g p;.y, 4; 1

',, R icrences t

3 1.

t r tter fr.1 Ja.scs Taylor (Pf.'s') t o Vic t or r.t.;l'c,(W C),

r,'.

"Pc'. ult s of Pere nt Induc ed f box rrror inest ir:at ines. "9 4

P.

?'< r orandse froni t.. S. N!,< r:. t r in ('lDC) to T. " vak ( '.K ), "

+-

. "RW f n.ioced F lux Error. "

3.

let'<rs fri.1 the folin.f rig utili ties on t'.e i. fii

• ed A ' s'a*r NRC, 0;erating Peactor Branch 4 i

Toledo Edison, l'. arch 18, 1991 e Daic Pc ecr Co.. ?'. arch 19, 1981

• a.ru 'n'a "Jnicipal 'Jtil i ty Ci.tric t, "-i h 73, 101.

!', tro olit.in Edf sco Co., te,sti

+r 29,1 %1.

ilsrida P..or Corp.,*tarch 70, 1991 i

A.4 2n.a s roi.cr 8,L i gh t ro.,.Nb ry 30, l 'iRI.

Dated:

April 22,1982 I

k t

e 4

1 t

L Dp JA 5

%