ML19332F486

Assessment of Risk Significance of SEP Issues for Oyster Creek, Final Rept Phase II

ML19332F486
Person / Time
Site:	Oyster Creek
Issue date:	12/31/1984
From:	Davis P NEW JERSEY, STATE OF
To:
Shared Package
ML19332F481	List: ML19332F480 ML19332F483 ML19332F485 ML19332F486
References
NUDOCS 8912150069
Download: ML19332F486 (35)

Text

wQv p.

w m.;b3.-umauw,

s-1

_.m m, w mw%m:w.

s wm

- a

• < 1m..

m m m$!- ~

zu. [^

m.,_f-..

rer# wa' q[

wp : y $ N'.M <

.. TV5 5

• th ?

i2N' 9

y

% 1. m 4,u 1

?. i_o. <..?

^-'

+

a

1. (q'y.g, n

+e-r

<:a ;.

o 3

, f'.. '

m'

~ h f%y%, w%a s,'_;lW P?TR( ^Q."")

.g',._WC. 7 ? ?.- p}7kk-[-_bf"3 - W

" ~

M} %W[..

[,i f

Y'.

E.-'

.-21 E-z K. -

p 4

i eh,m.+ +n,?q ' 1lWv um q-e~ a~ w, -

,s 4

-. : - +

r e v):f ; m 4W

, ypgr W.-m t ; -s e_

Qv n,. y

~.f.

.'w jt

-y, _,_:

• ~

m g, '.'.QmQ. ~ '- ye m"' f,,, m p{lty c..-

• @f. _. 4:

6.

<h in -

p" ?

w;:

as s.

. J L.Jx y n i p

s

_y_

L. ', -

-b ~ha

$wl;<+;y' m e m)f w{ ~ l%-i r rW lW ',

,:,,,+

+ -,

,s U. 7 2.^- a,; Eg

' 4 g 7. '

g

~.. _ '

,c

=.,

• r

- m-5A

i f M%!

~ ^ '

• L

"-=.k fn k~? = - ~* D 6

-h?

';?

94 ' *

^ '.

r a

e r

7/YW'Q-My w'n..

.;; m.+

> _u,'

y. l 8

~

u u

s

' 'i? G J-8 r

+

g;f sn m

- w > n.

l Ys fwh l~hf,~ %W l}.f..l.

f.. p

~'~ ' '

.. we4 h<a - -- r... u [.M

' w:,

e 4

e g Tg y.v,M

--N,#

' I _' :

$id

,.sa; m

5 r aT m,

.;J s

a i

m v

' ~'

+ ';

\\

pq r 3 47 m-

, 44.-

.3 t ;p j

e i

~

N.,'l. 3,e,

' Mk.p% d:

t c d.r.^ffd.

Cd c. ~.

%g

< ' - t tmW n

~- -

e g

x, r

s

+

v<

'h$ & l}

~

$Q>

2-

.a m.

- + +

e 3,

c'

, w% u 4

n

..i t

a s __

s

.u'

= wr

)

,.f;.b i ' V m$y,s.4 u s v

t

?' 7_--

'f_>

" Q T',

~.

t g

V 4

,L

.k Y

v${[i,kk Wj, W.

r'

~

3-3 3.c ^*.

di g

m 3

nry_b m

L

gp i 3;

~ '

e a',-

1 J

sl b.ml.'% ;f 7 Y v.

y M

w. u'&

n',,

9

,1

...a ny e.

.m.-,.,e

+

4, m

t am w

?M' J

A.

r_3.

~E

+_.b.::.

n

);c

'4 I

.,f A ;-

~.l4 f,f k l,r -l FINAFREPORTT N

y w

+7a q

s

%sp.;g J'

m..,...y-AU.

m 1 T y

/,x. -- J.s A. -m. u

=

. n-pg ;il y..

y 4

.q

.3

PHAS" ELil gg ;,

a_..g

.s.

4 bu. 4e 4

&, e" AN ASSESSMENT OF?THEJRISKi m.

MMT,.., !

YSIGNIFICANCE OF SEP-lSSUES FOR:

ify$

~

OYSTER CREEK 7

?W

~ ;m%

v, y en v

' ~ -

m,@s%e, w s

/n cM-. M,F '

s d _On. -

i y

zo.

s

. k' ?

d Y

• y:

~ ', -

4

- w$g

% g jf.g

)

. p^[ '.

N e - i.

s

,C1 ~a

+

.A 7

gm C14

• 1 c, M.

~*h g

g

^

gJm7'6

.,r u,

s m

s,

~ >,

w w-:- %

m Mi OcK.

w i

2

, -v r

a 2,.cs c f

- x v

c gs

".v a,.

p.

t

2 m

r 8

tw-m.

',y i._ a p. n.

Op 7'-' r 4

Prepared'for; v

a..

+

+ u+ m

t. #

.~

g.

wt y

ETHE NfWlJERSE.Y, /BUREA.UiOF.5. [NATIONcPROTECTIONJ U b

(>qf',.

m._

i'

.p If

.w_ri e.y m'

w t

gnq-T.;

q,

• ," 5 ' sh',

n 1..

i

, m.y r %' 'y~%y., n!,,

3 m,

w-a t

-.fSE. h. Y

^

\\

y v-a 4

1

's

.' 1.a.s_ p,

'y 2 >

s

.s J

e.

By

~H 9,% JfM y-.

m.

~-

g' ~% :

4Y i !- <

< ~ -..

[ g c F.

7.

o

'P.

R.. DAVIS--

9-2 7.n

. P.~0.1 Box 1604 a.

Vg7ge Idaho' Falls, ID 83403-1604:

a 4

t j..,,

~-.t R.N.,,. e_.

>L s

m'r ' > _

f FM W

n r s 3 a

r.i -

L

?N-n,e

_.,,y.'

[.3 g

+< a,e ;

+

e vt (Rf [g

~;-' -

f,><

,m..

~.

~ y~c 2

m

,m

^an u

n 4..

4 j

Mp'! -

f,

g'

.-*.h l

43y@Ja""k y

.Deterber 1984 J

3 -

'8912150069 891211 M~.,# a w%Ml( ; $

$PDR1.~ ADOCK.05000219 u,+

-PDC m

s..

w. e i..-

ms.-

x c

w Y

-f 4,,

+fff f

_N

? ' f'

~(

y : M'?N J.c-., ;. " ' ' :p ;,

h n # c&,.M2 :

t

. 3

' ;?8

++g--

~#

m.

u; up:;, w A,::,42 y-LNG ce "r nm m Pt

. ^- -*~ i

iI J,

1. ,~

A-4: 4 -

n.c.

..w.

m

m.,cw,- ~r.

s

- e.,.,

v

x M**

. CONTENTS Section; Page 1

INTRODUCTION-1 11 RISK DOMINANT SEQUENCES FROM PRA STUDIES 2

A.

Millstone Point Unit 1 Dominant Accident Sequences S

B.

Browns Ferry Unit 1 Dominant Accident Sequences 8

C.

Peach Bottom Unit 2 Dominant Accident. Sequences 11 D.

Dominant Accident Sequences from the Precursor 11 Study Ill DESIGN COMPARISON FOR BWRs USED IN THE STUDY 14 IV RISK SIGNIFICANT ASSESSMENT OF SEP ISSUES 17 1/

CONCLUSIONS 18 REFERENCES 32 l

t I'

m y

+

LIST OF TABLES Number Page 11-1 Comparison of Core Melt and Major Release Probability 4

from PRA Results 11-2 Comparison of Consequences Associated with Release 6

Categories-used in the BWR PRAs Considered in This Study 11-3 Frequency of Core Melt Accidents for Each Release 6

Category from the Millstone Point Unit 1 PRA 11-4 Accident Sequences which Dominate Core Melt Proba-7 bility from Millstone PRA 11-5

-Major Causes of Failure for Risk Dominant Systems 9

f rom Millstone PRA Results 11-6 Accident Sequences which Dominate Core Melt Proba-10 bility from Browns Ferry PRA 11-7 Major Causes of. Failure for Risk Dominant Systems 10 from Browns Ferry PRA Results 11-8 Dominant Core Damage Accident Sequences for Millstone I 13 and Oyster Creek from Accident Precursor Study 111-1 Comparison of SWRs Used in Study 15

-111-2 Comparison of Millstone I and Oyster Creek Core 16 Melt Probability for Each Release Category

-IV-1 Assessment of 87 Issues Against Risk Significant 19 Systems IV Summary of Issue Evaluations 29 ii

+9(9+.4#

E O

\\g)ye/

%f[dig

////p k-@

IMAGE EVALUATION TEST TARGET (MT-3) 4 4

I.O lt M BEA

((0 EM l,1

[w EM IE I.25 1.4 1.6 4

150mm 4

6"

>,4k, %i, s<xg*k;;+

/A 4f 3, 7777 oh#i

<AFAu

- - _ _ ~ _ _ _

p+ 9@

l

4) &)

[////

bg,p IMAGE EVALUATION 1ss11aest mim jpp I.0 l;m m yljEM l,1 ['

Ele 48 1.25 1.4 1.6 4

150mm 4

6" tl;f,,f,,,f-

- [(,;;[k$

a

t. A.

s# @

3 o

k'h IMAGE EVALUATION M/

[pppg\\Tj g[,4@4, #

k///7ij.' h TEST TARGET (MT-3) 4

$'#/

'%hif 1.0 lf En IMa pl_@RM l,l

[8: E=23 1.8 1.25 1.4 1.6 l

l 4

150mm l

4 6"

4h *'%

1. f4*p xxxx

>>,,g;f,3ppppp

,4<4+

%f Lk=

4

++ @+

l

[bo M/

////p 4)y$v*:+/

IMAGE EVALUATION g.O 44e #4 TEST TARGET (MT-3)

%~', &

/4,,,,

l.0 lt M B La ti M M=a e

m l,l

• kN l.8 l1.25 1.4 1.6 ll 4

150mm 4

6"

+d *%

~/f4%%

fQo>W.&,,,,p

+g#+.(4 1y 3CA.

[qIid NnnwM [ h< J N J 2*

Lw ni ~

o'I

)

s yy s.

• g; d.t

,, =

%X,-

n.

gfV

g 4

$% S $gq n, '"' ;

3

, Q4 4 5'J i

s

,m'

~

b $ M, M N, I

\\

-m-

=

~

m w in E

1. ,"'t 7- -

w'w, u,'-

m.

7 ptp./ -

%)?

FANLASSESSMENTJ0FJTHE R1SK; SIGNIFICANCE R gm 0E SEPilSSUES FOR-0YSTER CREEKL

&,b

?; -

[.~.

C,.

- If _

'lNTR000CT10F ~

~

y j ;~

shi; report..presentsi theLresults of an effort < to examineithe risk s.ignificance!

[4 q'

w i

Jofjsome87 issue'sfidentifiedibytheNRCas'areasgof[pote'ntialnon-comp 1iancet rA(J f

.-NRCifrom some140 safety-related topicsTwhich were founditot apply'to the?

f Efor the-0yster. Creek nucleartpower plant.

Th'e87issueswerelderivedibyfthe

~

e o

~

@f(i 4

5

~

f 50yster! Crehk hlantiduring L the NRC!s LSystematic. Evaluation; Program (SEP)._

~

The s

iSEP; reviewiinvolvesiexaminingLolder plants to determine 'if: theyj are' deficient e

]

- fin' meeting; current NRC l regulations:which?may have been developed since thei 4

g plant l.wentiinto operation.

A, y

<n E

~

- gg Thefprocedure-used ins this1 study to determine the potential risk implication 1 t

l O %[ '

4 j4

-ofethei87;issuss was~!asifollows.

+

2 %u' -

Oli iTOee P'robabilistic Risk' Assessment-studies for plants similar to-

.h

~

gn

.L0yster? Creek were examined and7riskidominant-accidentjsequences-gE s.

Jidbn'tif ied.

@W

'4 4.s,.,

j

-4*

's 4 1 y

m' W

'&[

jh EA brief esubvey wasfun'dertaken to; i_dentify;designi differences zbetween? <

~

M @' ' >

Ethe; three' PRA plants and 0yster Creektto-determine' the Lextent 1to wh.ich

~

+

~

e 1

a

@k

,the riskudominanti sequencesfapply.

e + >,

o f

-i" W,

-l3).

JThe?.ihdividualL factors comprising lthe risdominant sequences were'

^

w 5

.'separatedito." identify _ risk ~ dominant initia t' ng events and.. system -

i

! f a ilu,res. --

&f.

% 3 F>,"

$4?

'Th'e major contribu. tors to failure for risk dominanttsystems were i_ den--

+

n tified.

4~y w ;

si c

( M ;M c5.1

?Each of-the.87-issues was examined to determine if they would influence 2

1 i

the Qwjor contributions to failure for the risk dominant systems.

]

tr

, j,

'i e

jb, '-

p

, i h t_ v j

4

j k

- q m w:

u

~

OE hh

_; h d

ager o

It should be noted_that a definitive conclusion regarding the risk significance of the 87_ issues could be reached in only a few cases.

This is because (1) many_ of the issues are associated with protection against events postulated to occur external to the plant (tornadoes, earthquakes, floods, etc).

None of the PRAs examined for this effort considered external events.

Thus, the risk sig-nificance of these issues could not be determined within the scope of this effort; (2) the issues were derived from topics which reflect current NRC regu-lations. These regulations were not developed on the basis of quantitative risk assessment studies, but rather on the conventional NRC approach of deter-ministic analysis coupled with engineering judgment.

Thus, a close coupling between the issues and risk assessment results would not be anticipated and did not, in fact, exist.

11.

RISK DOMINANT SEQUENCES FROM PRA STUDIES There have been six probabilistic risk assessment studies published and made available which consider risks from boiling water reactors.

Three of these studies were for reactors with containment designs similar to Oyster Creek (designated Mark I containment),.

These three were thus selected for use in this study. Of the three, the Millstone Unit 1 0) plant is closest in design details to Dyster Creek, and its use was emphasized in the study. ' The other two PRA plants, Browns Ferry Unit 1(2) and Peach Bottom Unit 2(3) are of later vintage design than Oyster Creek and have more substantial design differences.

The Oyster Creek and Millstone plants commenced commercial operation about one year apart (1969 vs 1970 for Millstone) while Browns Ferry began operation in 1974 as did Peach Bottom. A further discussion of the dif ferences among the plants is provided in Section III.

A Probabilistic Risk Assessment study for a nuclear power plant attempts to quantify public risk from the operation of the plant.

This is done by calcu-lating both the probability of serious accidents and their consequences.

The product of these two quantities is the accepted definition of risk.

Accident 2

rgyq%:g,,, y

+

,; gy 7,

y,3 g

y wWp

e, x%mq r

- 7

>2 x s

_w

g. _ gt;,

[",%

• 3 f'-

- b 3p 13 if 0

+

F; a s

A W" ~ 1 h ' A g,7 w m ' w

o

+

<ji ;

Jg' e p 3 kf /

i

~

.' ' W ur M

NW s' %s M

g

[V[gh{

dProbsbilitiesLarefcalculated by' ass'sssing the frequency of; accident" initiating.

i NT ^

Heventstand.then; calculating the probability of failur.e of thoseLsystems which -

N '-

? '. -

%g g ' fare designe'd to~ successful _ly-terminate theJconsequences of the~ event? A'com-?

\\

p%

fte:(dis' cuss' ion-ofzPRA methodology -can _ be =found in-Reference 4a

~

~

Q:

wC*

UBefore discussing [the risk dominant accident sequences from the_ three PRA'-

1 w

L studies examined 1for this: ef fort,.useful perspective can _be obtained by com-7

[',

.paringiavai_l Abl A PRA results. - Table.ll-1 provides this comparison, iThe ver-c

(

' 4 ltical scale l(logarithmic) shows-probability for"co.re melt-and major releaselas:

W listed at-the1 bottom of the: table.

Core melt probability is the? annual l prob-J

~

abilitythat:anlaccidentiresulting.infuel. melting;wascalculatedtooccun v

i.,[,

g lThiprobibility ofimajor release is the'annualEprobabilityEthatithecontain<

^

i ment wil.1; fail catastrophically following. a core melti ;Theisymbols;"B"Jand' a

s 3$ ~ "P" are used to ind_icate.these probabilities, respectively, for: boiling land/

i,

- pressurizdwatercreactors as calculated by the -.15 available PRALstudiesb.'

I>

- ' (Simbol.starou.nd the?B"1 indicate which results are from which-of the three o_

b'W_

!PRALstudies" utilized inf this studi. As shown.in the table, the Millstonel 3

N['

llP'oinltiUnitj core melt?probabilfi_ty result for' a. plant _ which most closelyLrepW y

m.

M' l resents 00yster Creektis towards.the upper part:of-the range of results'. lThe P{L

r

!freqdency offa ma'jor: release ~for Millstone -is the highest offa111BWRs Lyetl o

+

~

i 1significantly lower (a-factor of'10) t than the highest' PWRL results.: While the?

JTable IIL1L compariso' ns provide some' perspective, 'the results should; be? v'iewedt 4

? with considerable. skepticism? for; the fo110 wing' reasons:

(1)'the resultsiwere;

,3" Y?

jprodudsd at-different' times =(as much as 9 years apart).by different organie

A

~

m YJ (2ationsgemploying different methodologies, 'ari f2) large uncertainties;are -

x

/associsthd withlall results,.which tend toLdiminish the'significancenof difh

%7J

,ferences.

+,

gr i eturningito the. identification of dominant risk sequences from the PRA studies, R

2, N

'results from each study will be considered separately.

It is' important to recotj-nm snife~~ that none of-the three studies considered external events as accident initiators and only t_he Peach Bottom PRA actually calculated health consequences from the accidents.

Ruther than specific health consequences, the -Millstone and Crowns Ferry: studies assigned accident sequences to four release categories g

yhich were originally defir,ed for Peach Bottom.

Each release category has a L*'

[ spectfic se.t of characteristics (amount, timing, and elevation) which describe

^

,. ~

[ {.

h)

~.}

g

~

e %.p m

.n.

=

.v a,

+

a m

s

ct w-

. <~

M ;'M g M

Y.

9 -

1-.

pg 9;

n

.;.s.

r -

j m

9 sol..

.i-m i

,9 4

t-y s r.

J..

t e i

w a c-2' 1-

, p_

=.

p 5e-4.

.. q-e, b

. _7_

.,.m.

W 4

1-4 2.

g 1

=

epg, ' s 3

= -.. s.-

1: m:

4 m w.1111 a w

e mi

,e pr c

_.m.

e m.

...q__...+,..,.

e g

y. _..,

...COMMR{MN9F.--COM.-MER-AM)-MAEn n!

KAE-.pmAm-ty i - ;. n n; 15 g"g

- 7~

- ;J 5m; atT10M f RA~TEFUET57 %*

-u a e

z j

w am

?d

~~

+ ",+

i.

e.,~._. w.

~.+ -. =.

=.

u.n w.=. m.,

=.....v.

wr

- m..

m. ma..

, m...

~

' '. ;r;.;_. :.l., p E.,

Q.3 dI[ d.N[:'.f N3 3'

i r$

[

1N :- i.

{ *.M'E h.p ;.9 -* % -': 1

• ."*.i' d :U'* g _ Md[

h'*5M J

__"... F5_..:C;_=

2=3~

55_:T-5.. E..ii.. - ;i ~i i.if.e_l i. 3E..M. ii.d;. i. :..E i_:.d. d.. si,

._. i.=.. ~~B._ -.i d.. -u. M.1._9 F. fili

2

. _ ~

w,

. _ _. _... _. _.. _. _.._..-j

-. _....,..t.......

s

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

n.

. m z a.

_m..

+ -

~

~.

..a

~

n

. -.. ~

~-

4.~ +.

.~

.._.i.

,.9 j

4--

.. +

m.-.,-,

m.

q

.p.

q,a -

,e

.,..m

-. h_.4m.

..__-.4... -

e..- "... 1- ",

_- m. i, '

,9,.

4 r_

.r-r..+.m._

r-

,.a.

1 e...-

~

.r..,.'

m i--'

• L '

T

Ms'T'

'L1 't*

^

. [3 ",*T

*; "T- 'n~ lp' W' ' ' ~ ~i i r %~~T'~~~ i ' T' T' ' ~

~

T' i

~

a --

m m

.e x". M'-.; ~

1

~~

.o _

,. m

,w

.u.

i.a a

w.u 2

4,

9

. =

= ;. 7

=

._e 4

m r,

._ m

.=.~w=...

r._

.r :

.n-

. :2.

p.

c 2.

e. t.s.u.

=. i1.. =...,-

=.= = -..

u-V

=

=#tW iM

1. E=& #d=h

=.E=f=W M%

TF

=n

= G.E

+

4 ~

s s

m

.n,.,._

.;n e

• ~

4 1

1 s.

g' a pj')

1 4.

.7 4 :..

Y, j.'

.i *. ' $. "

9

'*?^

't.".

"4 L-4.

..r

.4

- +

s.__

m.

p; 3

- g m.i m a

rn-

1. . = g -:

i.W

.,r

v. '*L.' *.C.

_T4._

• .. ~_~~

..-.A_P. I ; 2.."*

".Z n _' 1

- ;;.;* ': '.M.1. 3..

_;.,;'.' _..~. _/;,_';. ;_t.

",, G O..r..;..

g

~.

.s.

- _.._. p _.... _.,., _ - _. _.

-....re.., _.._.-

.n

.t

.a..

~... -....

... _...m.

.o

-. u. 4

>i-

.J

, ;- p g

w

.m e._

7 -

m.

.r s

~,

o M..

' ','h r

.,. r _

.-w..

,-w..

, r.

,,, m j

3 g

?. -} y...e...,,a

..l.,,

5 l Q, a,

- t _

.e a

.o....

a m

m y

l. $ i..

--.,n

- g

.4 _3...: ;.

. :A..AJ

-.....n

",....l l r!

m.

"" i"f ;

1 " ' dh i-* f.

t'..

d, -

U.

... j

,. ',i TS -

1-1 -~ ~'

• "?; " 7 t"7 1' t."

1 T

T, l.-i i 17 t 7 "--'-t

-g _tp g,,,

-4;

+

=.,g.

.-p.

n.

-r

-..4-i 4 'N

.h,

$. + p.

1)

T 4

4 4

/,

_.4 7+

1 H..

-i

+ - -.

,.O -

.m - g

,s m L p....

=,...

. =.

.. a.

a...

=. un

=

g-n=4

m...

.o x,

u..

,p p

L=.. -.. -=

_. _.. r.

, =

r. - l..

.1

_=-

m--

r

a. : o.

.=.i..

a.

=.

.. = _. _= =.. =._. =..

u

.w,

L

=1. 'E.

.4..v}....

w

./{I.

j 3.-.

6

.,."r-.d.5

. 'i

• 2

• . f

.'."u+*r T

I 7

. f

.,,.,j _

p.

4 g

n.

. T~

+

y.c

~

. _.r-

.p 4-

= p.og 1.-

s. m y;-

n r

.qu 1

j.

M,r g

• 3 _-. L,

v. 4

.m

=.

3

.., 4.y %.

. i '.

%.m

.g n.

---

.Lc ps:

-r p ^ *.

h s.ur4 2 d. 4. ; t. e. Reu.!J 'to".r.. '

- "4

.-*^ vM

.4

..A,

.P

- P' ' =..-

"l.

s3vn

. P..re

.... --.... - _... a,-.

t-

)-

....=

g.2. o.e 1,

m.. e, _. t501 1 na. G..te.r. R

.t --,.ttFr_ - -

ex t.-

u

.l...dt

<-],

_... J.

,- m

.r(

,i..*. _.h, il, none U_

.. t.

,c

+

~

2_

m.,.,

~.,.

,~,

y.,.

=L

. Wrr-4, Browns.4e r r,y-U. W. t-4. -

__.i....._..

.w.

A

.....-t.~.n.

.my.,.

}

i m"

• r

.t_r... p.I..

3 Q F..

i

/4.=.._ Pea h Bottom Un' t ~

.r --- _

- I I

• '+*

3 n

I k.

k k

,9

^.

eI.

,t-,

. 4 j

1 o

-A J.

i j -

32 d:

Q

.-7 t

h.

l

'm l,.

d.

j-

[

4 g.~ p:

,' l L'.

. a._..

-. - ~

.,e

.-g..y

. - - +

.' i-o

.4i. ~

,,.e l

l t

-,i.4J-

,..-g

- ~. -

(.

+,. -.

.w-_,+.I

--4 I

p j

3 o

~.... ~. -

. d,,.,., _. l _

.4-yn i

i i.

[

l 3

j i.

P"*

• .4" 6=

-- i 4

i, i

i 4-f

..-..I.,.-,

9 m

w j 5

)

4 p,

-i e

.g l

{'

i'\\.U/.

i t

~

if

-ns-r g

I i

l

)

w i

i i

J

. p P' s

m.*

~,

s i

i L*

g

_A.. -

I

_Y.

%,

• hg &(

l'

^

c MQ nyy;

' MY ;/ wh '3" E

• f i

'+

%g ';'m n

q.

m

+

,?h& h?

w% l y

<m

.a k ' V, o

4 g

N

[gg#fh Y j

N

'1 categories and probabilities computed on the basis of the accident: Sequence -

p',^-

t p-

characteristics which'are impo'rtant to the release characteristics.'

Fori 1,

..(

example fin 1some ' accident sequences the containment i.s predicted to failj M

~

I4 b

7beforeithe'coremelts'.x-For=theselcases,;theaccidentsa're'assignedfto1af

) ',

Lhigh release: category since hold-up and1 retention ofl fission productsfint theD (e ;,

o

pre-failed. containment'would not be-as:significant'as oth'er sequences-1.n':

+

L.*g Lwhich;theycontainment is: intact-at the onset'of core melt'.s 1

h-c s

es R

' iForJeachl of: the three,PRA' studies,. four1 release categories were' found't'o' be '

3 m

adequate toYspan-the range' of radionuclide release characteristics for the :

l accident' sequences found to be important..In~the PRA which considered: Peach a

Bottom (3) ;:a. composite' reactor site was defined for the purposes of. calculating; j

u w' ",

accident 1 consequences.1: Based on-a more exp1,1 cit consideration-of'these ion-J

]

[w-

'sequencesk,Lthe Table 11-2 results were obtained. This table illustrates d

y 76 7 i

.the differentl average consequences in terms of early and latent fatalities scomp'uted forJeach release category for. the " composite" site..The' table.illus '

M tratss the rather significant differences in calculated : consequences; for each?

(e' Lrelease category.

For. Table ll-1, a " major release" was defined as ~a 4Cate-

/

'goryll, 2, or 3 ~ release.

However,-since the' containment was: preoicted toh

bc J

ifail for. accidentsfin all-four categories. a significant: release.wouldl occur rn m^

Q evengin Category 41(and fatalities are predicted to occurEas shown in Table Dl JII;2), the least exte'nsive release.

For these reasons, this study wi.llifocus c

Y

' on-core melt. probability-rather than. risk, buttit should be' recognized.that "

hw llsojuef core melt accidents impose more significant public consequences than others. A separate. assessment of: risks from'0ystir CreekO 6)fhas;been per -

formed"which considers the risk-from different' types of core melt accidents, y

y 39

[)

4c i,

-:A r; Millstone _ Point Unit _1 Domjnant Accjdent Sequences j

q eTable ll-3:shows -the frequency of core melt accidents for each 'of the four

]

g 4~.

Erelease categories discussed previously.

The total core melt probabil'ity is 9

7 J

dominated by accidents predicted to produce Category 3 and 4 releases.

)

.i Table 11-4 provides a list of the accident sequences wh'ich dominate the core celt probability for Millstone.

The 1isted sequences contribute 83F of the'

' total. No other sir.gle accident sequence contributes vore than 15 to the total.

j i

l,

u

,,yn r

Table 11-2 COMPARIS0N Of CONSEQUENCES ASSOCIATED WilH RELEASE-CATEGORIES USED IN THE BWR PRAs CONSIDERED IN THIS STUDY I

Averaoe Consequences l

Release Category ___. Ta7 eta faTitieFTatent Fa talities

[

l 4.3 870 l

i 2

0.05 533 3

0 290 4

0 39 i

i I

l 1

1 1

l i

Table 11-3

-l FREQUENCY OF CORE MELT ACCIDENTS FOR EACH RELEASE

[

CATEGORY FROM THE MILLSTONE POINT UNIT 1 PRA r.

[!-

F h

_ _ $59k

.I.[.$_$.

f [ p p fi U jaf] [ _[_((

t

[.i 1

1E' 6 k

2 8E-6 l

3 1E-4 4

2E-4 Total 3E-4 e.

t 4

1 l

F O

[.

I

y+r Table 11-4 ACCIDENT SEQUENCES WHICH DOMINATE CORE MELT PROBABillTY FROM MILLSTONE PRA Sequence l

Probability I~

5Contributiontoj Total CMP j

1.

T JCD 6.7E-5 l

22 l

4 l

f 2.

T JCEFG 4E-5 13 4

'3.

T KCEFG 3E-5 10 4

4.

T KCD 3E-5 l

10 4

{

10 l

' 5.

T LCD 3E-5 4

6.

TA 2E-5 l

7 2

7.

T JCMG 9E-6 i

'3 i

4 8.

T LCMG 9E-6 3

4 9.

T KCMG 8E-6 3

4

10. TM 6b6 2

5 Total 3E-4 l

83 i

Definition of Symbols:

T2 - Transient with loss of_ power conversion system T4 - Transient with loss of normal. ac power T5 - Transient with stuck open relief valve J - Safety' relief valves fail to close K - 1 solation condenser system failure-L - Isolation condenser makeup system faillire A - Reactor protection systems failure M - Shutdown cooling system failure n-C - Feedwater Cooling Injection System failure G - Containn.ent cooling system failure E - Low pressure coolant injection system f ailure F - Core. spray system failure D - Automatic depressurization system failure

g dhgEW % g y fo 7e

~

7 e " i; y.

=

~

hv l.

g ~

,h:

a p s, m,w m',W

• s 3

r s

~y'.

+ r

-*2; y@vQ%g MU-g,J e qm<

B0 1

4 4,

~

5 e

,m, e

gg 4

3 w &.m;ca m u >

4 s %g-y mt.

a' 1

m yy;

~

y" '.;

3',

%u%w;G.s k-[=s dg 4,,

m

=

r 2 '

1 7

6ThMsequences?are$designedj(first column)Lbylaisefies offlettersshichdrept dy 1

u bA - e ' f reseht anlinitiatingl event l(first letter) followed by a"sei ofi sy~ tem: fail-s s

ures-_(tubsequentiletters) which are predicted;to result in core Jmelt. i Thef e

s p

0,,,

probabilityfoffthe sequence is?the~ product-'of the1 initiating; event 1frequencyL

%~ '

landLttiesubsequentsysemfailures._ The letters areLdefined at: the bottom :

P w

g s/ cof[titableh W,

iq 3

3

,a t

r r

y

Thettable illustrates that the most risk significant initiating eventiis al gy itrans'ient=witholoss of normalJac; power-(T )-.- The frequency of,this transient;

~

4 4

z P

Lwas assessed 1(Tablei4;1? of Ref.il) to be_0;20/yr. l:iThis is the same value used!

^

un

[foreP'each Bottom _and Browns Ferry and represents a normal ye power lossLoncef uh every 5 yearsQTo; check; the< validity. of; this frequency.for _0yster Creek, LAppendix fiof 7the SEP(7 kwas examined. ~ Appendix F contains a review of L E

10ysterl. Creek operati_ngsdata and includes events Lwh'ich have occurred at' the:

J

^

plant.:p Although1the Appendix F. results-are somewhat confusing. it appears?

2

(T, ablen4(2);: that;twocnormal _ac ' power disturbances resultinglinsforced shut-1 downihave occupred atT0yster0 Creek in thet timeiperiod 1969-1981'. Thistisl p

g,.

icons is'tentiwi th a: 0l2/yr;:f requency1ofinormal acL power' loss.

n 1

9 y" -

In~orderbtotassessithe. significance ofHthei87.SEPfissues on-core melt proba '

!bility,Ritcis' necessary to examine not only the risk 1significant systems butD

' m, N,~

-laisoHthe major;causesicifusystem' fail _ure.

An >SEP -issue may -influence theMesign 3

Lorfoperation of aLrisk?dominantisystem, but if it doesn'tiaffect? the niajor-

?causes of-system failure:innthe risk dominant sequences,iit-wi11[notihave; Mk an; influence. on risk re'duLctioni r:e y.

J From lthel Mil'l stone PRA,-' Table. ll-5.was --comp.iledLwhich indicates t the domidanti 3

en

~

d MW 7 Scontributors Lto: failure' for each.of the risk 1 ominant Esystems identified in' af' iTableDII-4" r>s

. m

-@U l

n w H

B..

BrownsEferrL Unit 1 Dominant Accident Sequences Eg 1

a gw

. TableL il-6 lists-the. three dominant accident sequences from -the Browns ferry PRA study which collectively contribute about-90s to the total core melt prob-abilityLof 2E.4.

All other sew ences found in the study individually contribute s

a

)

• L 4-1d m,,

2 zy g

_,~

t

788W ERWMyNayy MMy %m t W jMg

- QMg W"NMiWt g@ ^~ m@/D ' =

• h m

, us

-9 4

n i "E

W rf: W ww i

~

• u "4

W ' l,\\

u V",('T '

i i kMM WM.. au g+C W 4

% E. +

s 4

4 o

mm p

u, W^W+rB: t@ w. - <

_g-3-

.cw--~

, np s

=,

t e

M qw& ^ ;w V. k,% :

• M N./

+

k lL:

-u Yt L

\\

s r

L m

1 w u.m g$ q uc'y w'~$ + m* I '!M.

sk l

y,

~

V?

MSW@m

' # k c

s&

V.

M

& &v b;: W % J ' '

fj;9 d f hfi [+ % f

_ t' s,

E JTable lb51 Q$

i h % @d g K >

a mun x:,

L MAdORI-ChVSESI0Fl.FAILUREFORR1SK00MI'NANT; SYSTEMS' 4

Myf "

f~ '

, iFROM MILLSTONE.-PRAfRESULTSl "plv%%n m;.,

.y y q-2:

r M

y

~

w A,

4

(

73 1

4 QQF

'ISystem e a

Major Failure Causes-

' F Comments-4~.-

N y ] Safety:rSliefh

' direct from: data.

[,nocausesidentifsd;

• 3 (W %

' va.lves7(J);

.c

.E -

>valvelfailurescontnibi (q

"a o

e l

^

- p, isolation: valves closed for?-

@MQ' n n.)lsolation Con?

iden' serf (K)"

4,,.

maintenance:

ute'abouti]00 f3

+

m w:,-

-t s

t g c [.

Tisolation~ Condenserir !failbre; oft makeup valve; 4h y4<,

1 Makeup <(L-);

X

5. $(% 7 U

w d 1 ;....

. common mode failure y

mechanicalifailure ofcrods?

i" E y

Reactor:Protec 1

(, V zritionyA)4 9

m my p '.

g' sm..

34 l Shu tdown iCoo'l i.ncj) y < breaker / transformers auto see-(G) below; Y

=

w,g -

L(M)'

i.

bus.. transfer" n

Q k

- ~.,

gl id h * '.}FeedwaterlCoblanti...

= gas turbine. break'eri listed iriLdecreasing fF pumpf reaker Lo'rder-of sig'ificancet E~

'~'"Jinjectionl(C){

b n

? LW O

gas turbine generator--

H M -(iserYice:. water system?

~.

9$,

• M

~

', y M

fl" tcohdensatettransfer;systemi d@@ y,

G}L -relafsL

M ~, o,.

pumpipermissiveJswitches.

q

.=

,a a

3.ii%

-[

r_

s

/

[g- [ContainmentiCool ' W breaker / transformer; isametfailuresiastforc

'g iingn(G)i ~ ' ~

auto bus > transfer-f(M)Vabove:

r i

s

%+ -

ALow? Pressure;.

e-v.

gasJturbine andLdiesel.

su > J '

4 7

D>M

1. Coo'lant. Injection.J

genera.t'or _ _

i#S a'

g(E)l 1

!gasSturbine breaker andi y 1

f N@

' diesel generator r

trip (reset logi~ 'for?

f c

wg y MM, 1 emergency power m<

t>~

e SW N.

- - - ~

Jl-W.s$ $ 9 ; Core 15 pray Injec-w,.m M

1 (same as above)-

sWy ctlo~nS(F)

I

[ 4[lT,

(Automatic Depres-l[fortherisksignificant:

human error

~

%g@ J g "

stirization (D).

~ ;sequencesi this; system will R.

S n.ot' automatically actuate-sur?rs.

'~

[

because drywell pressure

?

does not increase.

Thus,

.:q n_J,

, manual actuation is nec m 5l

-.s Sfh 7 i sary.

gdQ S

_ N

.)?

w

-csz 9

[ &.4 4 i. ww,.

n e--

w.-

e+--

<~~

, * - ' - + - - ~ - ~ - - ~ ~

4

'H w - -

,?0

)

g'pra:n m g '

s_.

J hihd h h h '

'~ '

p uTWQ ar m

$qq W

Rg#;W W R W x,m++m W MXN+eJ_lte V>

< - % As J

k

. w

syvy Wu %ymp[h$ N ngw g;

w e g w,d m.

m,, m'lT3 n

b

,-m

+>

w cm,

y s.

e,TTh,yy

_ - yg

^ ' -

M,v.

a i

a s

y/

-L

." f ?f d i..

.9%

5' Y

• 5 "". C/ A f h!Q H d-%.:tW I

/

dij H%w.

h 3

h f Ylll ES

~ k h "5I9 % W'wl

^

k b ?

> < ~.

5 S

W3~.

y-5 c

t 4

'., ',^

Qud_f.

f I

+

'- N c+"

~

4c A$a. E>

Wye MGp p 1"u'

' ' ' =

t s

M h+@wdyh e ~/+

%, T4, 5

M

-m z 4

W 4

l 4

W w-

,ma

,~ -

m n

2 y

m g,..

gq;(

up ggggc{m nu o u

wy m e u-1 i

w

?

e s

y.,_

'~j['t

e

-4 W:m. W. v/

e 4 --

a W'

. sr,

,c s

e M-*

4 r

<r._:

d.

~

A, a '

u

,,.a

-4.l:G._$@ % v s-2 c

f j

' Y Wnen $f;q $

y i

k #,Mfj a j t

p

,.. ;,9 ft p

i SIM;A -'h U

T

<x*

1Tableil1/61 hk!Nhb[m@Jgh [ M; C NCCidENTLSEQUENCESi hlCH i

pw' 9

e u n g a. $. p

/[;;f

' PROBABILITY:FROM BROWNS?FERRv;PRAS t

T E WPh

~

m.

s "

7 4

p up. _

<s.

..t,ps v(wag _,,

i t

n g_Mk. w y

's

-..i Sequence'

+

P robab._l i ty o F Contrfbution to TotalT AJ -

ym as i 6

+

h.DONd y

.~,.m y

19.7E-51.

4 1

,,W c r(T6R Rf

491 n

TK:% (x f, l

B y.

+

uw a

1. sl@,

i.

e2 6)TJB0 5.1E-5 4 M.

e.

-. ~. =

3, s

_ 126:

+,,

1

~m

-u,

,c dNK?d 5

imm L3 M T,R.B;Av.

22.8E-5' S14; O

RR 4

y a

.=

- 1

$EB1,

Total

-118E-4L 789 M,

., m, g.p gg M -

M

} [W m

p.

m s..

=""

J 4 Definition of Symbols':

4 n

. J',,

- a

, c

/

1. UW- { b

T p-LTransient

withlpower conversio~nLsystem unava11able:

i gffbf, w

~

i A

e.r

%' 2 6 s n e >

1 i

m4 A M..i, STp',':-Llossiof offsite? power transient em

,m

~..

m. nmm-4 s

n; x

^---l-

^

Y

$f IT \\-

d. j.. )

/

-T

.._ #m M i # R'B:

'u 3 Residu,al;.. heat: re_movalf(torus cooling) ='

s..

wr u.-

ww n"

g % ;"%

.BN-I loritrol rod driveisys1!em C

n w 2, :

+

w i~,

A a

]g ]n y> >, l ::RgRes'idualgheat removal.(shutdown 7 cool;irjg)_' ~

em

, b(m x y

v ( -p 34

< >m <

w u,

m, ~;t_fq

+

1.

_c

• mc

L q'

..+

r

, 3 a

m F

i.O Table' ll-71 4

<w a,

' _. -*i i.

rs

• Q.::

$.y.%.!N

";3~

tMAJOR.'CAUSES OF; FAILURE?FOR RI.SKt, DOMINANT:

4 NA SYSTEMS FROM:. BROWNS FERRY PRAERESVLTSu gu yp;,

t q

n W ' 'm' t

gjf;XA.

p pmy i,s:equence <S,y.s tem i Maj.or> Failure Causes m - ~Commentst 9M W,;. ~w v?

RHR-torus = cooling.

-o nu

.M A TT{R~Rf

.) Valve control' circuit R andLR

~

ea faults-Bothclailed

+

C 's ( '

- B L

(R )

are y;

i ethe valve 1 con-

. trol circuitsfault.

J

.[-ControlHroddrive(B) scram failure

- ~.

a *-

LTjBT

"#^

LT R R RHR-torus cooling- (R t diesel generator f aul ts BothRNandRA are mu B A -

B_

k;s$

y RHR-shutdown cooling Emergency equipment failed by each of' Jw;2w (R )

i cooling water these causes A

>m.gc y >.. ;%mMW ;h!

~

~~

^

ij r.%

Wllpo$ e n;i <>

w r_

rwa

WyETF4);"

u

~

~

pNN

• 4 1

S=

4-1 ey 1

% t

+

h=uk $m MCig r

y-c w;;

~~

%[m sf

y g

---y 3-lessithan Si toithe: total. J.The th'ree sequences include two-types of: transient _-

g{}t >

]nitiatingevent(:andthree"systemfailures,stwo:ofwhich(R Q

?, '

B andLR ) occur onlyg A

Lin ~ combination! Table 7 flists the systems and identifies the major failure w(

fcontributions.

@eachi ottom iJnit 2-Doniinant Accident Seguences; C.t B

s.

4

'Only* ;two dominantiaccident^ sequences contribute about 75'; -to theit g

g~;v[~

~

melt proba'b'ilityf based;onithe ReactorJSafetycStudyl results for. Peach Bottom (3).

?

s

_ No;other ; single [ sequence contributes -more than 3%.- fThe two:, sequences arei:T 1

y

• [

['and;T ich.arelt'ransient eventsifollowed by residual; core _hea't,(W) and_ scram C

failurek(C).- The1 residual' core heat removal failure is' dominated:bj, human error

1. ~.

2 i

<and=commonimode failures associated with test and maintenance.

SD'.; LDominantiAccide'nt-Sequences from the Precursor Study IntJuneil982,:the NRC published a stat'us report I describihg the results.toi idate.fromLtheir Precursor Study.

This study, conducted by:the_0ak'. Ridges

'g i7

N_ational Laboratory and.their contractor, Science. Applications, Inci, uses 1 ~

Li_censselEyent Reportsi(LERs) to evaluate the potential forlsevere accidenti gh,

[Precunor; events. :The program involves evaluating:LERs, which arefrequired!

ibyLthe1NRCitoL e prepared by pl' ant operating utilities for. various specified -

b 16ffnormalland;unusualevents,todeterminewhicheventshavethe: potential 2

2for.leadingito severeTcore damage accident sequences.

Event: trees:ars son '

istructedraroundethose:LERs which'are determined to;be'sig'nificant.- The; event.

,s

-treesLidentify.pathscwhich, given the LER event,Lwould' lead to: severe core =

damage; LThe ' paths represent combinations of system failures and,-in some -

t 7 nstances,Lin'itiating events.

Probabilities are assigned to these failures -

i

- Kandfeventssto; establish a::guantitative-estimath'of core. damage probability.

.Intsomejcases,fthe LER' event is not an accident-initiating eventibut.'rather-

sgy she7 discovery 3that:a system important to safety was in a failed or inoperable; g,. ' '. s

'istate.

In _theseicases, the probability of an initiating. event fo.r which the D ',

unavailable system was an important mitigating factor was estimated and'the' g

core damage probability computed.

~

LThe; Reference 12 status report covers LERs compiled during the 10-year period u

L1969 to 1979.

Some 19,500 LERS were screened.

It was ultimately determined j

a

that only~ 54 of these ~LER events were potentially important precursors to i

severe core damage accidents.

]

On '

1 II '

.}].

f h

$ W, I

m

~

The 54 precursor events with their associated event trees and computed core dasmage probabilities were examined as part of this study to determine if any Oyster Creek or Millstone Unit 1 precursors were found which could influence the risk significance of systems and components found, or omitted from, the Millstone PRA.

Of the 54 precursor events, seven involved the Millstone or Oyster Creek plants.

These seven events,along with the additional failures and/or initiating events required to produce severe core damage, are listed in Table 11-8.

The table (last column) also provides the core damage proba-bility as estimated from the precursor study.

These probability values are arranged in decreasing order.

In some cases, two different sequences involv-ing the LER event were found to be important core damage contributors.

in these cases, both sequences are listed separately as shown in the third column, and the probability value includes the total contribution from all sequences associated with the particular LER event.

As shown by Table 11-8, the highest probability sequence contributes about

-5 6x10 to the annual core damage probability at Oyster Creek, involving a loss of feedwater event.

This represents about 20% of the total CMP from the Millstone PRA (assumed to apply to Oyster Creek),

None of the remaining sequence probabilities in Table 11-8 are greater than 2% of the total Millstone CMP.

The particular loss of feedwater event at Oyster Creek and the associated system failures were not found to be a dominant sequence in the Millstone PRA. However, all of the systems involved in the' sequence (third column of Table 11-8, including feedwater coolant injection, isolation condenser, and automatic depressurization system) were found to be risk significant systems and were used in this study to evaluate the SEP issues.

Thus, the precursor study results do not indicate that any systems in addition to those identified in the PRAs used in this study are risk significant.

It should be noted that the precursor study report 02) has received some criticisms from various reviewers, most notably from the Institute of Nuclear Power Operetions(

)

The resul ts are, therefore, subject to controversy and should be :wd with caution.

Qc Qy=

1. MQ ~ c. 3qp:3myn :pqg;g;mqqx wmyymgbggwp, 3g~ m,ym.

~

gg g

[g m a( W > ~;

m.

was+

~ '

a v

+

a w

gm a

[m.w-

,Y

% (( ~, % ;gug; m;L3

" CX-YMN 9

TN;IM k)NN mr

[ ' Nhh.

h5kd q-w; M.

W,

~ %, v

., ~ 4.y

~

g.,

~

V

,a,4 I% J -. w=/,

' n =M ?g -,

3,,

,.y m: p e gg y.q:-y x. W ' y,,~, 3 ; f '

i9:? & s

~

g

,g 39 g,, - c g 3,5.a g; a;.

j.

.., e

n ra ta f

vi g -

G:: l _. M - WC,y W~

s m.

Q}'~

pp,

3_a

_)

, gy pg ggg

=,

c

..-m

.g 7,7 - g s

cm + m.-

r s.

r f

i.

~_

^,

e p-

"%y

,. - 3.-+

=

n og x

w m..c.,

,.1_.

s.,

a.

_n r

a.-

Y

.,h n-

- DOMINAriT rCORE DAMAGEE ACCIDENT' SEQUENCE $FOR HILL $"TONEf i ~:^ g% -

{ Table'll-8+

^

fi j

?

FROM ACCIDENT PRECURSOR STUDY-

~

i System Fa11ures.

i- -- Probability (per yr)"

Rant f Event l Feedwater, coolant injection;Lisolation c6ndenser,: and ?.

?[):

W :q 1.6E-5

" 4 Oyster Creek toss of Feedwater t automatic Ldepressurization system' ' '"

7 j

7 a,

'4E Millstone I i Loss of Offsite PoweE tif 1.r Diesel-generator and feedwater coolant injection 2,

Feedwater coolant. injection and automatic depresp l;

.surization or_. low pressure. n ection?

i j l';

g-

~

i 4

15E i Feedwater coolant ' injection and au'tomatic' depres-

'l E

Oyster Creek Valve Malfunctions 1;

surization system or low pressure coolant injection:

l:

~

j.

2.

Long term cooling.

SE-7 t

1

~-~

1. IDieseligeneratdr failures and isolation condenser I

'i-Oyster Creek

! Loss of Offsite' Power-T

'2. i Teedwater coolant iinjection and isolation condenser >

^

l

'and automatic depressurization system orjlow pressure

' C i

).

. coolant-injection f I

L-~

~

22E-8' i

f ite power and feedwater coolant injection andiisola-a

~

. f.

O fs

. ation condenser

. p Millstone I Diesel Generator Inoperable

~ '

L1E-8:

Millstone I Gas Turbine ~1noperable

~1. [0ffsite' power"and diesel generator and isolation con-IL i~

denserf I

p

2; L0f fsf te power and isolation condenser. and.automatica i

depressurization system or low. pressure coolant -inject. ion' f

1E-8a Millstone I

! Gas Turbine failure j hidffsite power.and diese1' generator andinolation condenser d automatic depres-'

2 2. L 0ffsite power and. isolation condenser ari Lsurization system or low pressure coolant injection -

g i

g

!g 1

+

p w

? [k 1

1 l

t y

1 e

\\

ix Z A

4 3,

[

,,,- h,

.e

s4 4

7.

sr3 s

W:

, S r.{ 1 " S:.

4 I

k

~

v

,4 2M.-- }

A i L Q, 1+

w

.s....

m..

%e_-__.z__,

_...n

~ [ [ N,

['

N.,

.?_

%Wf* ding q,q, 5Nc x_ _

4[ '

=

~

h}l1Mf%l' ? Y 1

~

h5 e

w _%,Q Y

=,4 m

y..jg W

}

g lhat'

$111'.1 ? DESIGN"fCOMPARISON1FORTBWRs US'ED' INLTHE STUDY.-

I w 3 w~

h

[AMief'compaEisonjohthe major d6 sign?featuresj of[ thei0yster Creek, Millstone (I) _

'fBrownsiferry$2),,andjPeacNiBotto[)?

O BWRs wasLunderiaken. - 'The~ purpose of the' 1

[l((

i,:comparisonwasito"obtain,someTpreliminarycindicationioffthejextent:thattthe;PRA3

$m-

plants 7com, pare w~ith10yster Creek.: --Of partic'ularjinterestfis the; comparison:

k[' &

betWeenL0ysteh Creek and Millstone Point UnitLl:-since lthe M.P.(1) PRA wasiused.

. extensively.by th'e NRCEin the'ir SE'P evaluationN) ?and$ al'so.wasi the - prima ryf em- -

~

iphasis:in'this studyc LThe_ comparison'_is11mportant tochelp establ_ishithe exten't; w

nto whiEh theEPRA results may apply-'to Oyster Creek, E~

G:

i N-E-, -Q '

qTiblejlH-lLillustrates1the;-differences among 'the plants with respect tocsome>

~

12 systems'important'to safety. The primary source for this comparisontis1 n

g

Reference 8f withisu'pplementaljinformation' from References 1, 2,: 3, 9,10 L R +

'andill = The tableiclearly shows that' the Millstone design' features are:much '

j;;f more similar to O'yster[ Creek;than:either Browns Ferry.or Peach Bottom. ; An yy 1 inconsistency was= found1inithe literature with, respect-to the existence of.

Karsafety-related lfeedwater coolant injection: system (FWCl) at Oyster _ Creek.t 4

o LReference 8? indicates,that no such system exists at Oyster Creek, and the mh FSAR fbrf thelplant(Ik does tnot include any mention of such a? system. 'How-:

sg UV lever,M oth References 9<and.10findicate:that the Oyster Creek design includes-

an:FWCILsystem whichE an supply core: cooling under high pressure accident ~

c conditions.

L m

In2che.cking further, it wasLdetermined that the Oyster Creek apparently does;

' not NaveiaLsafety gradeffWCI system (14,15)(. Accordingly, the: Millstone core "i-1meltiprobabilities were adjusted to ' account for the lack of an FWCl systdm at e

10yster LCreek.

Thisiwas1done'by examining each important core melt accident.

j ise'quence (sunmarized ;in Table 11-4 of this report from Table..-8.3 of Ref.1) h Tandideterminingiin which cases 'the FWCl failure had an important contribution g

totreducingLthe sequence probability.

The FWCl function occurs in a large;

number /of1the dominant'. sequences-for Millstone.

However, in many cases,

'J ifailure of the. system was caused by preceding failures of support systems.

ja-o-(predominately. electric power).

It was-eventually determined -that only three 3,

i caccident sequences would be appreciably affected by the absence of an FWCl 3!

1 i

system at Oyster Creek.

These three sequences are designated by T JCD, T KCD, j

4 4

Land T 0 in Tame IM. D e proba W m es of W se G ree s m e m s w m j

3 requantified by removing the FWCl f ailure probability rultiplier, determined

~

I q -,

'w &

c

H,- -

e

!%c-

_ sw.

r y[myy w~T :q?,:%+L gew m 3:e,(w%gq_qQ_:&g,;mul.e@c

+4 s a

g.

s lQWa, ' a%y yy=ag.g&@

u. ;:&m.-new-

-m'W,

- o,

' m-Q_. 9 Q.

LQ97 4g 7;"

g+,p~~

h,%m,.'6y w; a e x w wu,;;3dge.;,.wmyg h.

J;m; ;a.; -.,,

p'

~

w en n, g n

,n m.(,

cy

+ -, e.w, s, n,

en,.

gm

~

+n w, g<pg

^^ 4;:.o g _ w. g.. + -. ~.,.,, A ; s..' %g. :

ea

}.--

e r +.s n

y J A.~'r<,% y'T r a

Jg!6. }-Q).Q,:g%;lY ?? ts.

4*

~ m^

a.

@; < ?> &:

.m.__

~ t

- L u.

wr Cv, ap

.. p%d n

~.t Q

'M

.'a,,&g.

, ~.,.

.s

..u m.

~ >

p gy,

t w

m.

~ %.f -

~

e<

.r

- {; n q'Q.

~, _ _ '

y;,, Q.Q.

?

S)

N-'q; ;" ; M y ;L,v ] Qw gg~.}g., w'. ' " O; _, '$ &

+

. 7.y_ g '

jf, yg;g.a.4< yagga:.pgg.mpq.

n gr-g,

+

.y., -

1.y,. m g

2.m n 1s

+ -

.e y

q,_., g g y

k. '

.; ;,,,.. p'.

4

,, m w e..

3

^ p s y%% yp

'6

~

J a

r

-n

^

- > +.cf hf&&g&m i

=

y. '

A, c,e t, -m --

,e

','r' W,, ' h ',A

.r

.e

, ';l ?

je i <

w w. gg n.:,a.,-y c.

e m ~

m,.

.....^

Y

?c' gm t

f,,

't iM M >

m h

t s

o

..,..M

-y, s

-y 4-

.,m.,

e

-s-

~,

a

.. 97 g pp....;

,.p~g yg,

s.

.a

+

, m;m v i.

1

- 4

-g gg rn m p 4

s. -

~

' ' =.. -

m'G, y)3,; -

f a.

--[.

..W :;,

Sh. $

i t c..

.r,, l?;j g ML :a p

z t

_m

,,.,3-3)n.W F P.

n -

2 r vow n x ;Y un

gy-

u 9: 1.

=

HTabl e - III-1 s -#

.Cem u,

=me.

s r

y 44

~ COMPARISON OF: BWRs USED.IN STUDY-

. 'W 4..

n ater AUS-5 Press-Td #p Camercial' L

- t j'

.s Containment l _ U'Cl.

Shutdown LoopK

'. Pump ;

ure Relief?

unter M g

- Ernergency LPCIf.~.

Spray /Supp.! Select.

' Valves' InjeL RHR Cooling' Pool Coolv Sila

cperation Type [ContainmentiRCIC' 1 Condenser HPCI LPCS a.

r

.s..

s e;

s l

g.

a y ), f19 Motor:

1Xj;];,

7 {,

Oyster Creek 12/69 2

Mark.: ' '

s[

X:

'X X'

g

,~

.g.'

I Millstone Point (i}

12/70

_3 ~ ; ' Mark 1-

.X X

X

.XT lj-1X.

?x'

' Motor'L J. I 2

.X; E.

i

X -

'X Xa

Turbine, (X)

.g:

Peach Bottcm (2) 7/74 4' bJMarklT

.51 '-

Browns Ferry (1) 8/74

4 Mark 1 X

x:

.X:

, X

X

/

Turbine

Il

~

-k

.e m.%.

r

_ c,

+

c'l*%,

,n (1) See related text. Section'III,

.1

. 3 f

f

,g.,

k 1 '

i g

n-

.p' d

' 7 I,.

+

r_

A; }

v V

4 p.

g' 4

p

+

,3

+

t'

^*

..a4,'

' q.I' 3

,(

t

(

,..'b' M

a

.. i

,' i.

2

' ?^

,j b,. e e='

j+,

l.

P g

J.

,a..

-. Q ',, ' ',. ' :' Q'

^ '

,.s R-y,

...V'T '

f y

N a

Je

\\.

a.

,,";; ~

~ m

~

..w.s... a;

-,:-a g

2+

4 w

e

, ~~

[.

s i?

. ~

~

X.

, n

- -- cE

<.: %'s

\\

+-N

+

.',,s,,.

+

.v.

'h'A ip d

1-i H

'l@' + b' M-

r W

d: 6 -f _N

'M'Q:

Q., -

_.,x"

M, r-,.M.Q.--

t.

r..

+

c

,<,e

..u : g~

Q, r_-

. ~ _, _ _, _ _

r w.

5;ppaymy w

.w

_',M,<

t 4; '

h d

&[ bhke t,

N I

s

,,s mng r

s 1

e kj Q@$Q S' L

~

n 2

i p7 WL s

W' T

1

~

Ml.]

(byTsumming allicontribu'tionsito(FWCitf ailu'retas providEd in-Section;8"df -

TReferencefl? forfeach sequence. Thefcontribution to each release category of--

$[

M 4

~*%

^ftheselsequencesLwasureca'l.culatedusingTable(8.3?of;Re'ference11,andthe;

~

Ltotal core meltfprobab'ility' for. each release category wss' summed.

The: resul tf i

s

?is shown? iniTable- !!!-2 ;

g:

s (d

!Tabl elllI a,

?$e

COMPARISON l0F MILLSTONE'l' AND OYSTER CREEKi 7<

JCORE MELTLPROBABILITY.FOR EACH RELEASE CATEGORY ~

.ns i

j L4

, i

{~ <

--_el ea se_C.a_ted._6_pL_ob_aVi'lTtLF C-

. Total ~ Core 7

R r F 7

7-

____7 Mel t Pr_obabil_i ty._-.

p

~

d

. Mil'1 stone.1:

1E461

'~8E-6 lE-4

'2E-4

.~3E-4:

m e

. Oyiter1 Creek 1

.lE-6~

.8E-6L 1.2E-4' 3;2E-4 4.4E-4:

~

_ _ u_

-l Aside from thsifeEdwater? injection-system, Table' 111-1 indicates onlyLtwo -

S J

x M

t arbas?wtiere differences texist between 0yster Creekjand Millstone.

The first!

2

}differencelissthat' Millstone has -a low' pressure cool' ant injection.'(LPCI)f C9

s9 stem:while? 09 ster? Creek;does not.

However, 'this. difference is not expectedi w

f

tothavela;s'ignificant effect: on' theiapplication of the. Millstone-risk-dominant w '

1 sequences tos0ys.ter= Creek since. failure of LPCI for Millstone, sequences: always-R M

occu'rs[ concurrently with feedwater injettion-and;LPCS : failure. Thelfailure h

', icause forsthese!three systems is-loss-of onsiteTemergency' power rather than s

]it Jindependent-failures of the s' stems, Thus, 1t?isinot likelyithat:the proba '

y m;f Ebil'ity:of these sequences'would be.different if ihe~LPCI system were omittedh U

4 iThe final difference from Table III-l is the inclusion of -an LPCI loop seleca T

tion' feature.on Millstone and not Oyster Creek.

This, feature allows LPCI Qp

flowJto.be isolated to the recirculation loop which is postulated to rupture

{y

during a loss of coolant accident (LOCA).

This feature does not improve the relia-7.

Rbilitycof LPCI,- but? rather enhances the effectiveness ~ of it by not allowing water.to spill directly into the drywell.

Further, no LOCAs of any type were y

foundito be risk significant. accident initiators.

Thus, this dif ference is

}

[>,

not expected to compromise the _ application of risk dominant sequences from o

3 I.

'the Millstone.PRA to Oyster Creek, i

N

(

Eds '

I H

%ffff Why,

(

q

.4 W',

WW [iJh).^

s-

^

%2% ' }w' _.an:

i

,-WM, l[]

d m2,AlfurtherlmajorJdifferenc.e; between10yster' Creekiand Millstone was(found:

g.

-~w,..

$) [b ' ' L A whicn:Wnot[on LTab1'e 111'-1. : Mill stone Unit $ relies' oni one.laiesel. gener

~

a gEg, ;, f athr;'and oneiga'5f turbine; generator l for.onsitelemergency ac power, while) t li(

Oyster CreekL employsitwof diesel 79enerato'rs 'forLth'isEfunction. LHowever,<

@h

' J he: Mi11 stone' PRA:used the'same;f ailure. probability 1for bothi gas-turbinej t

and: diesel generators =.n ThUs, thisidifference-appears to.be of111ttle; con-

,,; {

• 1 sequence in..the context ofLthis study.'.

h (ItTshould/be= emphasized that this plant comparison: study was':far from :ex-t "hy4 '

c wM haustive, and that risk Esignificant differencesicodidLexist.'amo_ngLthe: plantsJ 3

bm

?further, it is possible that' errors andLomissions exist ::in-the PRA ' studies

^

Consequentlyp the risk significant which could invalidate their results.

{~

Qf ~,

eval.uationiof -the 87Jissues in7the following. section should1 bejconsidered-e.w W

l

>a1 preliminary scoping effort'to identify any significantaitems.

4

IN

RISK S'lGNIFICANT" ASSESSMENT OF' SEP?!SSUES Mb i Ech ofTthe 87-SEPeissues-as identifiedLin the SEP report was: evaluated' 4

E s y 1 against the riski dominant: systems, components, land failure causes as derivedt g;,

Lin Section II. LThe basic: purpose.of the evaluation'was to! determine if;thei cissue would ha'vevany -influence in terms of reducing l r,isks.1ThFbackfit' re--

p[.~

lquii'ement imposed byL the NRC for many :of the -issues was also, eiamined to 4

Ldetermineif;the/issueswithriskreduction'potentialLwereamongthose:with; w

backfitsrequirements.-

3

,7 E.:f tin reviewing the-871 issues, al1~ were found.to fall1 Lin'5. categories with respect?

ttoltheir; assessed potential fororisk reduction. These five: categories are as?

(follows.uwith definitions:

t

,e 1;

Undetermined - It could 'not 'be determined :if risk reduction potential #

Esisted for.the issue'because the area:being addressed was not' con-sidered inLthe PRA.

y

^2..

No --It could be determined with some confidence that the issue had.

9y no risk significance' based on PP.A results.

f

%I 3.

Probably~~Not - The potential _ for risk reduction appeared to be very Tow for th~e issue.

~

4.

P_ossibly - The issue appeared to have some potential for risk re-duction, but a definitive conclusion could not be reached for a j

Variety of.rCasons.

1 5.

Y,es_ - The; issue definitely addressed areas where risk reduction could be. achieved, j

wm v

i*:.

~

wms

\\

q i

Table IV-1 provides the results of the evaluation of the 87 issues. The first column provides each topic considered in the SEP, and the second_ column lists each of the issues (a total of 87, numbered sequentially) associated with the q

topic. The third column indicates whether NRC requires a backfit associated with the issue, and the fourth column provides a judgment of whether the issue appears risk significant.

Comments are provided in the last column.

Table IV-2 provides a summary of the results from Table IV-1.

As the table indicates, for over 40% of the issues, the risk, significance'was undetermined.

In all of these cases, the issue was associated with external events which could pose a threat to normal operation of the plant. The PRA studies did not consider these events.

For almost 50% of the issues, a determination of "no" or "probably not" regarding potential risk significance was made. Of the remaining 10 issues, only one was considered to have definite risk reduction potential, in 9 of the 10 issues, backfit requirements have been imposed.

The single case of a "possibly" risk significant issue where no backfit is required involves the reactivity control l

issue. This issue could influence the reliability of the reactor shutdown function which would have a risk significant impact based on the PRA studies.

V.

CONCLUSIONS This section provides conclusions from the risk significance review of the 87 SEP issues, as follows:

1.

The Oyster Creek core melt probability (4.4E-4 per Section llA) as estimated from revising the results of the Millstone Point Unit 1 PRA is relatively high, as is the frequency of major release prob-ability when compared with other PRA results.

These results should be viewed with caution, however, due to the inherent uncertainties in PRA results as well as variations in rigor and methodology em-ployed by the PRA studies and uncertainties in extrapolating the Millstone ! results to Oyster Creek.

An estimate of risk from Oyster Creek due to core melt accidents is provided in Ref. 16.

2.

Of the 87 SEP issues, some judgment of potential risk significance could be o de on only 52.

The rec ainder were related to external events not considered in the PRAs.

Only 10 of the 52 were found to have potential risk significance, and only one had definite risk sig-nificance.

Of the 10, t,ackf i t e e:p i o, ents at e tw ing m; mwd on 9.

-)P.

r M

_i j

-s I

Table IV-1 A55E55ENT OF 87 ISSUES AGAINST RISK SIGNIFICANT SYSTEMS Backfit Risk r.

Cs w nts Recuired Significant Issue Topic

)

Undetermined Evternal events not considered in No 1.

Condensate Transfer Pump Power

' PRA for Millstone or Browns Ferry -

11-3.8 - Flooding Peter.tial and Pro-

[4.l(1)]

tection Requirsents No Undetermined External events not censidered in 2.

Flooding Level Procedures [4.l(2)]

PU for Millstone or Browns fewy

+

II-3.B.1 - Capability cf Operating Plants to Cope with Design Basis Flooding Conditions Undetermined External events not considered in 3.

Canal Water Level Instrumentation Yes PRA for Millstone or Browns Ferry II-3.C - Safety Related ' ater Supply

{4.l(3)]

(UHS)

Undetermined, External events not considered in Yes

.4.

Isolation Ccndenser flooding P9A for Millstone or Browns Ferry

[4.l(4)]

5.

Iow Water Level Shutdown (4.l(5)]

No undetennined External events not considered in PRA for Millstone or Browns Ferry 3

i-Yes Undetermined External events not considered in 6.

Hurricane Flooding of rumps PRA for Millstone or Browns Ferry

[4.l(6)]

Undetermined f External events not considered in Yes 7.

Finoding Elevation [4.l(7)]

3

PPA for Millstone or Browns Ferry i

O}

Undetermined f External events not considered in 8.

GroundwaterElevation[4.l(8)]

No PRA for Millstone or Browns Ferry Yes Undetemined External events not considered in 9.

Roof Drains [4.l(9)]

PRA for Millstone or Browns Ferry j

' Undetermined I Some coeponents affected could III-l - Classificatien of Structures,

10. Classification of Structures, Yes Cnmponents and Systems {4.2]

l lhaverisksignificance,butin-Ce ponents and Systems (Seismic fluence on reliability of this topis is unknown.

and Quality)

'l

__ b i

• __m4.__,._

-h-'_w_,-_-

m sh-- ahed.59_wta.d_We

_Mp,_C

'mh

=e

_-h-'

_,,__,p4

~h__

_a. _,, _ _ _ _

,p.

[_

U t.

a

%4 lit s

Table IV 1 (Continued)

ASSESSMENT OF 87 ISSUE 5 AGAINST RISK SIGNIFICANT SYSTEMS Backfit Risk Con ents

? Ret;uired Significant Issue Topic Undetermined ' External events not considered in Yes

11. Reactor Building Steel Structure FRA for Millstone or Browns Ferry III Wind and Tornua Loading Above the Operating Floor [4.3.1]

Yes Undetermined

12. Ventilation Stack [4.3.2]

Y(s Undetermined 13.' Effects of Failure of Nonseisaic Category 1 Structures [4.3.3]

Undetermined Ne

14. Components Not Enclosed in Quali-fied Structures 34.3.4]

15. Exterior Masonry Walls [4.3.5]

no Undetermined Yes Undetermined ;

16.RoofDecks[4.3.6]

e t

c'

17. Intake Structure, Oil Tanks, and

. Yes Undetermined i 5

Diesel Generator Building [4.3.7]

6 18.LoadCombinations[4.3.8]

! NoII)

Undeterminedf No Undetermined

19. Soil and Foundation Capacities [4.3.9]

1 i

III-3. A - Ef fects of High Water Level on

20. Hydrostatic loads (Combination)

' No

, Undetermined f External events not evaluated in

! PRA for Millstone Pt. or Browns Fee

[4.4(1)]

UndeterminM{

Structures i

21. Hydrostatic loads (Short Duration)

~ Yes i

+

4

[4.4(2)]

I l

22. Below Grade Penetration flooding No Undetermined g

i

[4.4(3)]

i i

4 i

1

^ _

..c T

y-yw

- q:y-. ?%.

'p

.gr--

~

}g: gK. y.

b.%%Q'" r 1

~_.

~

-- y-:. g y.;

q Q,-w

• -[

Q '? _': _

yy.

Q 7

- - - - g

,y.-

' ' j-

^,4

m ; c f

-r o.

, m_

. _. _= ~ ~

...:. +, e.

z a,'~

~

_ ggww, zy_l C t.-u -

m m._

n

~

- i ugnj '

-w

^

~

5+3 '

2 c.

g t

,.pg;Q 1-

~ =-~

. D T

l _

?q b f 2+

7 4. s. ti

-Table IV 1 (Continued) 3 g

ASSESSMENT OF-87 ISSUES.'AGAINST RISK SIG'JIFICAVT SYSTEMS

. e.-

1 Backfit-Hisk-

~ ' Coments '

~

Required

• Significant Issue Topic-23.; Intake and Discharge. Canals'[4.5.1] ;j; No _ =

No 1

!!!-3.C - Inservice Inspection of Water i

i Yes Control Structures

24. Intake Structure Trash Racks and

. ] ' No..

. i.

- Intake Screens [4.5.2].

No(1} ' {

J 1

No--

t 25.-RoofDrains[4.5.3].

q f No

-26. Inspection Program [4.5.4]

Mo p

i

27. Emergency Diesel: Generators and.

f.Yes l. Undetermined J External events not eva III.4.A - Tornado Missiles

. ruel 011 Day Tank [4.6.1].

l

- t.. for Millstone Point or. Browns Ferry.

j

' 28'. Mechanical Equiprnent. Access Area

i. Yes

, ; Undeterminedj-g' 7

[4.6.2]'

i i

v g

g Undetermined 1 1

. :- 29[ Controi Room [ Reactor' Building.

No j

-and Turbine Building Heating, ing.

ll)

.l-7 i

Ventilating, and Air Condition.

j

, p Or/AC) Systems;[4.6.3]

4

.30. Condensate Storage Tank, Torus Water

.Yes l Undetemined'

Storage Tank,1and. Service Water and~ i

EmergencyServiceWaterPugs[4.6.4][

[

r.

111-4.B - Turbine Missiles-

'31. Turbine' Missiles.[4.7]-

Yes_ ]L Undetemined Jot considered in PRA.(Classified j

aslan external. event).;

q 1,

External events not. evaluated 1n Pfb:

^

. -:32; Truck ~ Explosions [4.8.1] L lp:. No :

dl7 Undetermined 4

ifor: Millstone Point or Browns Ferry ?

111-4.D a Site Proximity Missiles l

(including Aircraft):

1 - Undetemined L33...AircraftHazards'[4.8.2], -

I' ;Yes i

1

'*?

'n

,.g-

,,eu' J

4 1

h.E [,

,',9',g..

j

}

W

^

9

ET g y

"~

~ 4:

NG;;

M

~

3'"*

3 ~,7_.y ;%~L$t~q3 w

t, =* '

M

=

yy lll

~

j Q

>b

~ ' " ~

A c2..

~,

~

.c

~ "^

w

'y vgg,

c y_.

~

,m

[

fa 2s@!E.;

D

• -y

. ;s

?

a p.i.

d' Table IV-1 (Continued)

A55E55'*E*:T Cr 87 ISSUES AGAINST RISK SIGNIFICANT SYSTEMS

...-Risk'

.j--

t Backfit :

-1Recuired Significant

. Coments Issue Topic OI f No.

! Not abvious these' items were' definite'-

III-5.A - Effects of Pipe Break on-t

34. Cascading Pipe Breaks [4.9(1)]

I No Structures, Systems, and Com-i f

i considered in PRA studies; ponents inside Containment

'j l Not obvious these items were definite';

?? No

35. Jet Impingement Effects [4.9(2)]

No

).consideredinPRAstudies-l.

1.

l Not obvious'these. items were definite-I

~j'No

[

36. Drywell Penetration [4.9(3)]

no

1 considered in PRA studies i 1

i g

j ' Neither of these events were found -

111-5.8 - Pipe Break Outside Containment f.

37. LOCA Outside Containment '[4.10(1)]

]

No-j No:

to be. risk significant accident int

j...

t i

i tiators:in PRAs.

Yes

• No 38., Emergency Condenser. Isolation 4,

g q !

. [4.10(2)].

g..

~

1 t

1 l Unde'terminad i Seismic risks'not considered in:

l 39.-PipeSystems[4.11(.1)]

Yes Ill Seismic Design Considerations g

t 3 Millstone or Browns Ferry PRAs.

1 e

. l; Undetermined l See related discussion in. text.'..

40L Mechanical Equipment [4.11(2)]

'Yes I

f l.

41. Electrical.Eqbipment'[4.11(3)]

Yes

1. Undetermined t-e 2." Ability of Safety-Related Elec-

- No I Undetermined 4

+

I' I

.trical Equipment to Function

.}'

[4.11(4)]i

.g I

j

' 43. Qualification of Cable Trays {4.11(5)]' Yes Undetermined l-.

I 111-7.B - Design Coces. Design Criteria,

44. Design Codes, Design Criteria.

Yes

~- Undetennined -

toads.being considered are prin-:

Load Combinations, and Reactor

Load Combinations' and Reactor :

capally from external events not' Cavity Design Criteria Cavity Design' Criteria [4.12]._

considered in reference PRAs.' In-1

. fluence of codes and criteria not i addressed in PRAs.

E n.m.-m. _.

t

.a--. -

g.-

=,

^

~ ~

}

eq

,.p-

Q^

~

^

E'?

- Q, ';,

y

'x 3;

g-L m t.

~

g

.,r_u.

[

~: ' ]

~

+

.x

,... ~

.t

Table'IV-1 (Continued) s ASSESSMENT Or 87 ISSUES AGAINST RISK SIGNIFICANT SYSTEMS

~

Backfit Risk-Required Significanti

'Cw. ants Issue i

Topic

45. Loose-Parts Monitoring and Core

' No.

No i No accident initiating events ofi, j'

! risk significance would be pre III-8 A - Loose-Parts Monitoring and.

Barrel Vibration Monitoring 4

.j vented by this. provision based-Core Barrel Vibration Monitoring

~;

[4.13]

1

. on PRAs.

+t e

i Probably Not' Valves appear as contributors;toi i

46. Thermal Overload' Bypass [4.14(1)]

Yes

{

-j two risk.significant systems (Iso.-

111-10.A - Thermal-Overload Protec-jt - No 8.Probably Not, lation condenser makeup and resid-!

tion for Motors of Motor-

47. Magnetic 1 Trip Breakers [4.14(2)]

• ^ual heat removal). However, valve L

-e Operated valves j-g f ailure report (NUREG/CR-1363) -

j shows no failures' attributed to-

malfunction in thermal overload j-

-I' 4,

l protection,'although a large frac--

g

'I tion'(24%) of. unknown failures are i 5' -

. j:

listed.

i 1

IV Reactivity Control Systems including'

48. Peactivity Control Systems,'Includ- !

No

! Possibly

~5 cram failure'is a risk significant.

event. c Not' clear if the requirement ~

Functional Design and Protection ing functional Design and Protection i 1.t. ; involves effort that would reduce

AgainstSingleFailures[4.15]

scram failure' probability,'which Against Single Tailures l

i - was determined to be comon mode j

I-mechanical failure of control rods " -

1 i

i j'

I to insert for Millstone Point.

~

e tl.

! Pipe failure LOCAs not risk signi :

No V Reactor Coolant Pressure Boundary

49.LeakageDetectionSystems'[4.16.1] ' i Yes (RCPB) Leakage. Detection l

I ficant and' leakage detection systems.1 -

z not accounted for in PRAs':

7[ No

l 50 20perabilityl Requirements [4.16.2]

.Yes l

51.Lintersystem Leakage [4.16.3]1

~ No '

'No1 lIntersystem leakage not.rlsk signi-:

I ficant from PRAs-l No No' 52.' rector Coolant Inventory -

'alances [C 16.4].

- {^

j.

i

-- +.-._,.:.l -.

.. ~.

n

.-w

~

m, en / - ~., _,

a 5

~ m f

j 4

Table !Y-1 (Continued)

ASSESS"ENT OF 87 ISSUES AGAINST RISK SIGNIFICANT SYSTEMS Backfit.

Risk l

'. Required I Significant I Coments Topic Issue V.6 Reactor Vessel ;r.tegrity

53. Reactor Vessel Intgrity [4.17]'

I Yes No I Reactor vessel failure not risk l

[ significant from PRAs.

l 1

V-10.8 - Residual Heat Pmoval System

54. Residual Heat Removal System f

Yes l Yes l RifR is a risk significant system Reliability Reliability [4.18]

t i

and it appears from discussion on f

Page 4-26,27 that reliability.

j j

would be improved.

i 1

V-11. A - Requirements for Isolation of

55. Requirements for Isolation of j

Yes i No RWCU is only system affected, and High-anc Low-Pressure Systems High-and low-Pressure Systems l

i it is not a risk significant system i

failure based on PRAs.

[4.19]

l 2

a I

I

56. Water Purity of BWR Primary Yes No F-12.A - Water Purity of FhR Primary Coolant Coolant [4.20].

g 1

l 8

YI-l - Organic Materials and Post-

57. Organic Materials [4.21.1]

Yes i No accident Chemistry

58. Post-AccidentChemistry[4.21.2]

No No

59. Locked-Closed Valves [4.22.1]

Yes No See discussion of containment.

VI Containment isolation Systed 1

isolation in text.

60. Remote Manual Valves [4.22.2]

i Yes No I

61. Valve Location [4.22.3]

. No no i

i 62.. Instrument Lines [4.22.4]

No No i

63. Valve Location and Type [4.22.5]

No No

• ~

64. Administrative Controls [4.22.6]

j No No

'[

l i

s.

Q1 -

-jpr 7mg,.

7,,

,.g g.c gg m

n x

.x

~. ~

1 s ;

c,

.4.gc. g

- y

x g

w7...

vip y.

, p v _.y g.- -a

~, ;3

= < -

.n

.,.r (2y ti ;,y :. '

e}l.,

.4

'g.,

- N E iM

" c.gf y

~

~

iQ 1

di-

&O

,u-r

,~

Y 7-

@af q; 1 Table '1V-1 ? (Continued) 1' x:

ASSESSMENT.0F 87 ISSUES AGAINST RISK SIGNIFICANT 5YSTEM5;

+

Risk

. Backfit.

'l~

Required ^ t Significant '

~ Coments-Issue Topic

.I L

._ j.-

g a'

VI-7.A.3 - Emergency Core tooling

. - 65. Emergency Core Cooling System

'I

.:Yes

,1 No '

18 Emergency condenser failure not.

e l risk significant-(only condenser?

Actuation System [4.23]

t Actuation System.

-l.

- makeup)

~

.t-VI-7. A.4 - Core Spray Sozzle Ef fectiveness.

' 66. Core ' Spray Nozzle Ef fectiveness.

l No

. I No '

no low pressure accidents are risk i

-[4 24]i l.

significant. :Aisof core spray ef-1

-fectiveness not considered in PRA studies.

VI-7.C.1 - Appendix r -

ctrical

67. AC' Automatic Bus Transfers "Yes Possibly Leading of ernergency ac buses is" 7

'X Instrumentation and Control

[4.25(1)]

lrisksignificantbasedonMillstone;

-j: Point PRA; influence off automatic.

Re-Reviews

\\I 68..DC Automatic Bus Transfers I bus. transfers not evaluated.

~

~

-[4.25(2)]-

{L..

V1-10.A - Testing of Reactor Trip

69. Response-Time Testing [4.26.1]

'No-No.

Mechanical failure of control. rods,

dominates system failure:

System and Engineered Safety.

j-Features Ircluding Response-Time Testinr; 70.-Einstrumentation-for Reactor Trip Yes Possibly Scram failure is' risk significant.-

~ 5ystem:(RTS) Testing [4.26.2]-

The influence of. this issue on scrael reifability at Oyster Creek has not?

been evaluated.

l

'Wo Probably Not Based on evaluatOn on' 4-37 of SEPi

-l.71. Dual-Channel: Testing [4.26.3]~

1(not significant for Millstone. but-

. not -included in Table 4.1)J

+t' Q

4

'e'-=,

h e,.M.--_.-_._._._

_ m.a._ _,._a d i

-g77

, T _.

• Wy -,

y,

s.

3m gg y -

E7.m,

~

-~

_g

[

[

&'r ^

(('rg

~.;W._

~

p

. =.,

c.c w.3 hh,

J s.;-.e 1C

~ r

, zl;f k A

^

~

w 2.,d q'

W1 4

.L qq g.7 w g* u Table;lif-l '.(Continued)

ASSESSMENTOF87ISSUESAGAINSTRISKSIGNIFICANTSYSTEMS

-lp

^1A 1

' Risk Backfit l.- Significant" ;t Connents Recuired Topic

Issue

]

.)

t.

VII-1.A - Isolation of Reactor Protection

.. ' 72.1 Flux Monitoring Isolation' {4.27(1)] ',

Yes

[.Possibly i See Page 4-38 of SEP -

System f rom t.onsafety Systems. -- -

f

! ossibly'

').

Including Qualifications of Iso-4-

73. Reactor Protection System (RPS)

}

Yes 4

lation Devices

}-

ProtectiveTrip[4.27(2)]

l lP 3

1

~

~

. es 1 Possibly:

VII-1.B - Trip Uncertainty and Setpoint l.

74. Trip Uncertainty and'Setpoint Y

Analysis Review of Operating

' Analysis Review of Operating Data a'

~

Data Base Base.[4.28]

y.

~

j No{j) j.

System Control Logic and Design

{

. Control-LogicandDesign{4.29].

; Probably Not See #47 [4.14(1)], the only item -

VII Engineered Safety features.

1.

75. Engineered Safety Features 5ystem

^

}

Kfound by the SEP to be in non-'-

~

.-l compliance for 0yster Creek -

.;(

e g, - -

VII Systems Required for Safe jJ

'76. Systems Required for Safe. Shutdown

-Yes Possibly-

~

i-1

^[4.30]-

Shutdown VIII Onsite Emergency Power Systems

77. Onsite Emergency Power SystemsL Yes Probably Not Problems found at Oyster, Creek diesels.

(Diesel Generators) d

(Diesel Generator) [4.31]-

relative to this topic are not sig-j-

nificant' contributions to failure-1 (see related text).'

'i Vill-3.B - DC Power System Bus Voltage.

i78._ DC Power System Bus Voltage Moni-Yes Probably Not DC: bus failure not. found to be risk :

_n

!(-

' toring and Annunciation [4.32].-~

significant in Millstone Point and Monitoring and Acnunciation Browns Ferry PRAs.'JIssue does~not --

appear. to have a large. influence on :

+-

, DC availability at Oyster Creek.

(Ref. pagas 4-42 and 111, App. D.

y "of SEP report).

f

~ '

.l" D-s.

a e

.s tm.mmm-

--..s.._,w.m

~

~

m 7,

'~3

.4 E

Table IV-1 (Continued) y.f ASSESSMENT OF 87 ISSUES AGAINST RISK SIGNIFICANT SYSTEMS Backfiti Risk Topic Issue l Reavired 8 Significant Coments I

i l

i t

l No l Penetration failure not significant VIII Electrical Penetrations of

79. Electrical Penetrations of Reactor

? No Reactor Containment Containment [4.33]

to risk based on PRAs. See related j

discussion in text.

l 1x Ventilation Systems

80. Restoration of Ventilation [4.34(1)]

Yes Possibly could influence diesel generator'-

reliability j

i

81. Reactor Building Ventilation [4.34(2)]!' No l No j

~

82. Core Spray and Containment Spray Yes No These systems not important to risk Pump Ventilation [4.34(3)]

based on PRAs for Millstone and Browns Ferry. Does not appear that u

problem at Oyster Creek will decrease reliability to the extent that systems ~

5 become risk significant.

83. Battery, Motor-Generator and Switch Yes Possibly Affected equipment may influence Gear Ventilation [4.34(4)]

reliability of risk significant system.

XV Decrease in feedwater Tempera-

84. Decrease in Feedwater Temperature, No Probably Not These problems could influence the ture, increase in Feedwater Increase in Feedwater Flow, Increase frequency of transient initiating Flow, Increase in Steam Flow, in Steam Flow, and Inadvertent Open-events for risk significant sequences.

and Inadvertent Opening of a ing of a Steam Generator Relief or l

However, experience at Oyster Creek Steam Generator Relief or Safety Safety Valve [4.35]

(Appendix F, pg.' F-xii) shews apparent.

Valve ly very little influence from these problems on the frequency of tran-sients, which is not unusual compared to data for other plants.

I j

as

-qq~ %pc ge %.,@ ~ T:-gu e %%..,,

9LWF. 8W N,WMW9Msih yv -- g 2

>a--

y.

J M M M/J"~

f~

G.-

g-v w ce

-p; as,~

he

( ;; ' ' "

-Lf %,,Te @'

a _ > :rgtp.;

g,.

p, a

_g x

u.c.

=

+

?w'- ^ ~ n

.- _ { :-Q

. -m-~ ~' 7;. 4 %o n nmu m

,n a

52 ' ' ' *u-

, }..[?.c f.[;

,l?J, ^

e & ~:h?,

k '

n m

?

• l'

Nl5. l

%y P

' ~ ; T,_" :1~

~

p

.c, Dx

, gr t

y n

e

~

- u m

g se ggg.Qpm

, ~-

f'"

~

r 1_

.:. _L

^

. ' - M AS L 3

c~-

-:g

')y

,-(

.Q

.f yy ;, *r.*

"?"

~+

,,n.

y 4

-4.,, q%m. A g

, w;;,.w:

w e

" ~,,f ~ g.?.WQ:1.,Yl,;y fr

,Q}~ T g.

Q y.Qfg m.

~

,w kh7 3

.~ -

m,

, 96 g: ~4

py y c

.na

'gff.:

u ggy

' TablellV.11(Continued)

m. ;

&mM ' - f

.g %

y

' ASSE55 MENT'0F.-87155UES AGAINST RISK SIGNIFICANT SYSTEMS T

.~..

q,9.,

2,

p. 3...gff M

Topic Issue Required 8 ' Significant 1-

, Coments -

1

- ~V Backftt.j.

- Kisk.

- i

]'

g

- 8 1

XV Radiological Consequences of l:

,ES. Radiological Consequences of Failure i Yes l No -

. ffbt' found.to be risk significants E

' n :c.

Failure of Small Lines Carry--

?*

of Small Lines Carrying Primary l-

!in PRAs.-

1

, ~

Ing Primary Coolant Outside i

CoolantOutsideContainment[4.36])

y Containme_nt.

~ -

w-

- ~*'

l (t} _

- No '

LNot found to be' risk-significantb E

~

XV Radiol gical Consequences of..

1

86. Radiological Conssquences of a;

'.ho a Main Steam tir.e. Failure i-iMain' Steam Line Failure Outside-Ein PRAs.

N_

Outside Containnent

~

},

Containment.[4.37]

w; w L

s 6

XV Loss-of-Coolant Accidents Result-1 [

287'. L'oss-of-Coolant: Accident Resulting.

. Yes;

No '

Not found.to be risk significant'-

v L~,

4, ing From Spectrum of Postulated

' :from Spectrum of; Postulated Pipe.

win PRAs.

cf Pipe Breaks Within the Reactor:

s

~ Breaks Within the Reactor CoolantL

^

y _

~ Pressure. Boundary [4.38]:

Coolant Pressure Boundary;

g (1) According to the SEP, these issues are partially covered by other backfit' requirements (See Table 4.Iof SEP).

4 h

r e

c 7

,.~:..;...

,-n..

+ d a.

...q

.a i -.

.e

• ./

.4

.,,[

-I'

..n.,,.

~

~

I I'-

g. ;;,

fn M ' ;.$

y7 q.; : A :

7

-.j e

s.3E

~

' ? l-iIrw

,,.JQ.,

m

- y;L:*~'

,;-;T..z.' '

w l*'b'

~

"^

?%

r9 7r-s...

* ~

1'

>-f l',

1. ~

,3

. :n r

g

w-e Table IV-2

SUMMARY

Or ISSUE EVALUAT10NS l

No. of No. of issues i

iReguiringBackfitf Ri_sk 5,ignificant Issues Comments f

Undetermined 35 22 3 others partly covered i

under related backfitting No 35 14

,l under related backfitting 3 others partly covered probably Not 7

3

! 1 other partly covered under related backfitting Possibly 9

8

No backfit required for i

1 i reactivity control issue

(#48 f rom Table IV-1) i Yes i

1 1

Reliability of RHR (tS4 from Table IV-1)

W

ccwm

,v m K

+

)_h, ik i '

Jh 4s 8

t

-I

$i% ' '

i Th f l

=

- 3 m-y+.

lhe single' issue not covered by backfit requirements _is'" Reactivity-i p,

Control; Systems; Includ_ing functional' Design and~ protection Against.

N L$ ingle railsres"... The.:NRC positio'n on this issue as described in the j

,s

,3 13Ep wasl evaluated to. determine if conclusions could be established on j

u--

its risk significance.-

y,y r

N' On Pagel4-22 of the SEp. the NRC concludes that; on the basis of'a i

a

"" limited PRA" (which apparently refers to Appendix D of the~ 5EP), thei m

h i

g effects:of a" multiple-rod withdrawal on risk was determined to be.of!.

1 3

j h,f low; risk.: This conclusion. appears valid.. tioweyer,:the topic would:

..I.--

R seem to! cover also those situations with influence scram reliabilityt g

The' NRC staff furttier concludes (pg,' 4-22) that the Oyster Creek rod '

j s!

control system design based on a' review at Dresden Unit 2, meets-N O

current licensing criteria, and thus backfitting is not! recommended-j a

J n page A-34, it is-concluded for this' topic Lthat' "NRC has concluded 0

1$

ithat revisions to existing licenses are not warranted.

Staff effort-

.on this' issue will continue 'at 'a low level,"

O.

n P.

"r N

P

0n page 25 of Appendix D where Top _ic IV-2 is considered, under the

1

6

_sub-heading' "NRC evaluation", the following statement appears:

"There i

fexists(the potential for single failures at Oyster Creek which would O

E-

' allow multiple control. rod withdrawali,- which is~ considered nottac.

E

['_

~

ceptable," This: conclusion seems inconsistent with the evaluation on-j h'

'Page 4-22.

j q

c

'f

(

' The' assessmentiin Appendix D also includes' the.resultsl (no. details-id given)- of-at fault tree eval.uation. of the-Oyster Creek reactor :protec- -

~

tion' system, and a'" careful" evaluation'was made of the fault tree and:

uthe equivalent fault. tree-for Millstone, ;1t was concluded from this h 4., '

,. effort that "... failures in the rod withdrawal part-of the control

{,'

,roa system do not contribute to failure of, the RPS '.' This. is probablyL a valid assessment, HOWever, it addresses a narrow interpretation of-

' the. topic and does not consider scram reliability, further, the in-

{

creased probability of -uncontrolled rod withdrawal would increase the Mf f

fprobability of transient accident initiators (designated T in the Mill-2 g,,

(stonePR_A)' requiring. scram, ibis increase is probably not.significant, jfb; ',J Chowever, compared to the 6.6 event / year used in the PRA.

l.

hs.

N(

i

(,,

s S

Ip F

lt appears that the NRC position is somewhat vague and inconclusive.

As a result, this topic is considered to remain in the "possibly" category with respect to risk significance.

c An investigation was performed of a recent study (I2) to determine if 3.

identified accident precursors at either Millstone Unit 1 or Oyster Creek appeared significant in terms of identifying SEP issues of risk i

significance.

The study was undertaken by ORNL and involved examining actual events at nuclear power plants for the time period f rom 1969 to L

1979. The events were screened to identify those which had significant potential for leading to core melt accidents if additional system fail-

[

ures had occurred.

For those with significant potential, a core melt probability was calculated by identifying subsequent system failures required to produce core melt and assessing the probability of such failures given the state of the plant at the time of the event.

The precursor study identified 54 precursor events (from Licensee Event

[

Reports) with potential for adding to the probability of core melt.

The Oyster Creek and Millstone Unit i reactors were involved in seven of these precursor events.

However, the assessed core melt probability for these seven events did not in any case exceed 5'; of the core melt probability assessed for Millstone Unit 1.

Thus, it is concluded that the precursor study did not identify the potential for any core melt accident sequences for Oyster Creek or Millstone with a significant probability compared to the Millstone PRA results.

Further details miy be found in Section 11.D.

E

F s

REFERENCES 1.

Interim Reliability Evaluation Program:

Analysis of the Millstone Point

'Viii^tT1TuiTFa F Vowe~r~M a nt, NURt G/ CR-10'85', TdliWTp)Tii'a~tToni,~ l nc.,

~

~

Ja nua ry T98Y ~~^---

2.

Interim Reliability Evaluation Program:

Analysis of the Browns Ferry' UnifiltiuYfeaF~PiliriE,~ ffUkitI/CR-2802, 'iG&G 1dario,"JuTy~1982.

3.

Ryattor_ Saf ety_ Study WASH-1400, USNRC, October 1975.

4.

PRA Proc,edures Guide, fiUREG/CR-2300, USNRC, January 1983.

5.

I tts_i de_.NRC, J a nu a ry 24, 1983, McGraw-Hill.

6.

Overview of, the Reactor Safety Stu_d.y Consequerites _Model, NUREG 0340, USNRC, June 1977.

~

7.

Intearated Plant Safety Assessment - Systematic Evaluation Program.

OfstWCr eck jiu'cleaf Geidifatin.915tMion,' NuklG!OS22," January 1983T i

~

8.

Additional Information Required for NRC Staff Generic Rep ^ ort on Boiling Wa t}6r' i(Fac t'6rsl, ~ Nf DO-Y4 708, 'G6n6ra l E l eit ric' '06., ~ Kugist 19 79'.~

~

~

9.

Suneary Report on a Survey of Li9 t-Water-Reactor Safety._ Sy~ stems, h

NUREG'/Clf-20'69, f'. "X. Yiddlisoif,' ORNLi, Ipfli'1983.

'~"

10.

Design. Data,an,d Safetyy Ieatt.res _of__Corpercial_ Nuclear _ Power P1artt s.

ORNL-NSIC-$5, December 1973.

Central f etyy Analysis Jyport - Oyster Creet Nucle.ar Plant, Jersey Final Sa 11.

Power & Li9 t Co._

h 12.

Precursors to Fotential Severe Core Damage Accidents:

1969-1979 -

Stjiu[}@pdrt,' MfREG7Clf-~2'497,' 0ak Ridge Safio66'lTa665fory,"J6ne ~1982.

~

13.

Review of NRC Report:

Precursors to Potential Severe Core Damage Acci-

' '~ ' 69.-19'._79.,._ A _S_'ta t u s'.k. _epot t'., MTRL G./..CR '.249_7, IN'PCi ~82-025,' 5eritentic r

.l 9

~

~~

i

~

derit's':

14.

Personal Comunication with Mike Hober, New Jersey Bureau of Radiation Protection, June 28, 1984.

15.

Grourina of Light Water Reactors for Evaluation of Decay Heat Removal ~

Capatiilitv, NUR!G/Ck-3'713, 'DieoOialen ' National 'L'aisoVa tory, June ~1984.

~

16.

An Lstimate of Risk f rom Core Melt Accidents at the Oyster Creek Reactor (Appendix A), P. R. Davis, to be published.

I

)