ML19289E852

Forwards Tech Spec Change Request 72,Addendum 1,adding Limiting Safety Sys Setting.Water Quality Rept & Certificate of Svc Encl

ML19289E852
Person / Time
Site:	Oyster Creek
Issue date:	05/23/1979
From:	Finfrock I JERSEY CENTRAL POWER & LIGHT CO.
To:	Office of Nuclear Reactor Regulation
Shared Package
ML19289E853	List: ML19289E852 ML19289E864
References
NUDOCS 7905290178
Download: ML19289E852 (27)

Text

b ".

ADDITIONAL CLARIFICATION ON OYSTER CREEK WATER LEVEL CALCULATIONS MAY 2,1979 EVENT 2047

.?99 7 9 D5J 9 0/ ? K x.

CLARIFICATION OF CRD FL0k' DURING OYSTER CREEK MAY 2, 1979 TRANSILNT A discussion of how the CRD hydraulic system will perform considering the conditions which existed during the transient of May 2, 1979 is as follows:

Docketed information in the Oyster Creek Technical Specifications Page 3.4-5 and FSAR amendment 18 question 1 (attached) describes the operation and capabilities of the CRD pumps and hydraulic system for a small (less than.002 ft.2) break. The modifications described in Amendment 18 have been made so the CRD pump will auto start on the Emergency Diesel. During the May 2,1979 Event one CRD pump was running continuously and the second CRD pump started at the time DG-2 auto started and closed in to power the I-D 4160V Bus.

Therefore two (2) CRD pumps were running continuously from before the recirculation loop discharge valves were shut until after the "A" recirculation pump was started and its associated loop returned to service.

Since there was a scram signal inserted during the entire event, of the major flow paths for CRD water is through the scram inlet onr.

valves to the CRD drives and then into the Reactor Vessel lower plenum.

Another flow path is through the CRD return line into the reactor vessel annu]us.

Although there are no exact numbers on the breakdown of flow between these two flow paths, it can conservatively be stated that with both CRD pumps running at least 60 gpm would be delivered to the core region through the CRD drives and c least another 50 gpm would be delivered to the vessel annulus through :he CRD return line.

2047 300

3.4-5 In order to alla for certain primary system maintenance, which will include control rod drive repair, LPRM removal / installation, reactor Icak test, etc., (all performed according to appnved procedures), Specification 3.4.A.8 requires the availability of an additional core spray pump in an independent loop, while this maintenance is being performed the likelihood of the core being uncovered is still considered to be very low, however, the requirement of a second core spray pump capable of full rated flow cnd the 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> operability demonstration of both core spray purps is specified.

Specification 3.4. A.10 allows the core spray system to be in-operable in the cold shutdown or refuel modes if the reactor cavity is flooded and the spent fuel pool gates are removed and a source of water supply to the reactor vessel is avail-able.

hater would then be available to keep the core flooded.

The relief valves of the automatic depressurization system enable the core spray system to provide protection against the small break in the event the feedwater system is not active.

The containmet spray system is provided to remove heat energy from the containment in the event of a loss-of-coolant accident.

The flow from one pump in either loop is more than ample to provide the required heat removal capability (2).

The emergency service water system provides cooling to the containment spray heat exchangers and, therefore, is required to provide the ultimate heat sink for the energy release in the event of a loss-of-coolant accident.

The emergency service water-pump-ing requirements are those which correspond to containment cooling heat exchanger performance implicit in the containment cooling description.

Since the loss-of-coolant accident while in the cold shutdown condition would not require contain-ment spray, the system may be deactivated to permit integrated leak rate testing of the primary. containment while the reactor is in the cold shutdown condition.

The control rod drive hydraulic system can provide high pressure coolant injection capability.

For break sizes up to 0.002 ft,

a single control rod drive pump with flow of 110 gpm is adequate formaintainingthewaterlevelnearlyfivefeetabovethy3 c,

thus alleviating the necessity for auto-relief actuation The core spray main pump compartments and containment spray pum doors. (4)p compartments were provided with water-tight Specification 3.4.E ensures that the doors are in place to perform their intended function.

Similarly, since a loss-of-coola:a accident when primary l

containment integrity is not being maintained would not result in pressure build-up in the drywell or torus, the system may be made inoperable under these conditions.

3 This prevents possible personnel injury associated with contact with chromated torus water.

l Amendment No. 21

e i

1-1 1.

Use of the Control Rod Drive Pump for Small Breaks For breaks less than 0.002 ft2 (about 120 gpm) the control rod drive pump has the capabil-ity of maintaining the water level nearly 5 feet above the core if it is properly utilized. In order 2

to do so it is necessary that the auto-relief valves not be actuated for breaks below 0.002 ft since sufficient mass would then leave the system to momentarily uncover the core thus resulting in the temperatures given in Figure IV-20 Amendment 10 to the Jersey Central docket.

Upon loss of A. C. power, assumed for design purposes, void collapse and the mass lost through the steam line until closure of the isolation valves is sufficient to trip the low-low water level signal set at 7' 4" above the top of the core. This, in conjunction with a high drywell pres-sure trip, would normally trip the autc-relief valves for any size break.

By selecting a level lower Gan the 7' 4" low-low level signal on which to trip the relief valves, and automatically plachig the control rod drive pump on the diesel upon loss of external A.C. power, it is possible to prevent the auto-relief valves from lifting for breaks less than 2

0. 002 ft,

Therefore, the auto-relief valve level trip signal has been lowered to 4' 9" above the top of the core. This makes it possible for the control rod drive pump to function as an inventory 2

makeup system for r ' ! quid breaks less than 0.002 ft without tripping the auto-relief low level signal.

2 An analysis for a 0.0015 ft break is shown in P!gure 1-1.

Upon loss of feedwater at time zero the mass drops sharply due to mass losses through the steam line to the condenser. A 10 sec isolation valve closure after the 7' 4" low level signal was used to maximize the mass lost. Upon closing the steam line isolation valves, pressure rises and after a 15 sec delay the isolation condenser trips. This causes pressure to slowly decrease, eventually actuating core 2

spray flow, nearly 1-1/2 hours later. In this example for the 0.0015 n break the level does not reach the auto-relief trip point. To just reach the auto-relief valve level trip would require 2

a 0.002 ft break under these same assumptions.

The control rod drive hydraulic pump's controls are to be modified so that if a standby diesel generator is started (on undervoltage from external power), and as soon as the diesel is ready to accept load, a control rod drive hydraulic pump is started, unless core spray has already been called for or is called for at the same instant. If low-low water level or high drywell pressure had already called for core spray or calls for core spray at the same instant as the call for control rod drive pumps then the start of the control rod drive pump is tempo-rarily blocked from the emergency bus (diesel). When the core spray booster pump discharge pressure is established the control rod drive pump is started. Even when a core spray signal blocks start of the control rod drive pump the delay will be only the 10 to 20 seconds required to start the core spray system. Thus even for this event the control rod drive pumps should be started in approximately 30 seconds.

It should be noted that the time factor to start the control rod drive pump is not critical.

The mass loss from the system occurs only through the break after isolation valve closure and the loss rate is not high relative to the mass of liquid above the auto-relief trip level.

2047 502

1-2 Even a minute delay in starting the control rod drive purnp would only cause the level to drcp less than 5% of the distance between the 7' 4" low-low level signal and the 4' 9" auto-relief low level trip. However, as pointed out above the control rod drive pump will be energized within approxi aately 30 seconds at the latest.

The temperatures across the break spectrum with the new auto-relief low level trip are shown in Figure 1-2.

These temperatures are based on the conservative design type calculations and no credit is taken for core cooling once it is half uncovered. Only three relief valves are assumed to function with a 120 second timer delay. Thus it is seen that the lower level trip 2

does not affect the relief valve function for breaks larger than 0.002 ft and indeed results in an improvement.

204/

503

D**D D

A c e/

c m

- 9?1 k

m

. t.a.n. -,

. o.

. r

;;t~.,

..2......i

. j..g.

q

. +

.... - +

.4-t+.M gigg.--.

i

+ - *. +. _.,

af.

,,,g

_.3. ] ;,,;

-- - - -q ;.

4 --

_.,,, 0-

.g*.

f.

- + -

+ " + -+-

• r

r-n:.:II:_.f.. :,

TN.,

2= :(:

'

1

--.. :;d. -z-.;;

..n::u 4.

+-

- - +

=z.,

.1r. 'wr ut1.

~

,: = lit"t.

*I'r I..1.j LO 8t M

n

".1 ; a: Yh:;n r:tL.:-

nt:

m- -

mn.. :-=.

.:u.

re t

3

=

y=

n,

--.;2.;;;_ ;u =:

; fr.

.3 q

his-.g mpn m.

.n. :23

f t+

msms.

.-=

--....=.a-r

."i. i. -

• E;,2e.

di!f:.l

.:c_'25,.

2 m-3:ti

!:~

-=

i!3 att :n:.::t 08"

_. r-$.y

~

~] "i51id HE;i i D lM @t-h lh $ li}iEiiiMii:-frt: :2515EM

'tiiE IE L--

i

~~ l N :: t w :-

.24. :-u =4.

i.

a:.......: :+

...n

t :.- : =, - -:.4i:=:::a na

i

,:r

.. e: n 3-nnlm-- ;" 9..::.

.5.

-u-f:11:.

s ar

r:

..'T o

i a

. 2..

tm.

..: i.-

. i...a.:-

m:l;p

.g-. dr.n:

~12. l.h

-=" ~.~.~

+-

.. -....F. t --._

m " -- m.p:.-

.jj 35

-r

= tn rn:

~

- -v+-+- -1

-w:

LL.a..:h!.ky

..,. 1. : -

. E.

_.t

.a..

.L..;

~

.... n.

r

.u. n.

.***),.,.t. a...

3 t.

m.

..3 u....

- -+.;.:-

r~

.p:

.w.

nia:fT'."~

huq =,::n4.n;}E::

.:-s --: : * -l,.

~

a.a.

a ::ki.

t. ;.

4 w ir-a.A.. -

= : -

nt === =

.,nu+n g...

-rr}-

...c.-

,rg:...:

., p :...._ :. ::-

9,,. :;.

_.,...J ::. :: _.. ! Q..,.._

~r,,.ly =.. g,._...~--.-+66..

. g.r..

m

~:.

-*-1

.u c

.J.

=r.= :

a= : : u.

"g

. I'.

t=

r.:

. 4

..r 2nu

_=_=.m.-+l=n n.p :

a v....:b _

.n ar =

-.h:i:l:4;t :r i

Ja::

NE,::-

du.: --i 4 nr:u

m

. 2:t

.:a

.4

& E= u imui Giq.MHEE6. :li:

-! "= Em-fWEE i.ii

J TM G=

==

h

=

L[

.u 4 il * -i%ilit fi-it =;iiE: :/

Ci-i M ai :::

11tii@H@cC hi iiifiir

'}W5 MNS:n-

-+4'i !M

-:; i.1 side:'I.?isINidiEEi -[ D:M:'

t, J-

. ".....i~i:

ur ; il @E:E -- h~ "-

udiiD

M g

%h0:iO.p.c. SITE. -' ~~ N!' I *iiNE}:n.i t: .. tr. .:l- - - y.

..s,.h. k:.Ig =.

un...

}...._:3n.n_.r p..

r.22 En.fi:_=.t.".. _ tr.-. n.----._ .n ~ -.,.r.. n. a_ rvi - - = - - .I) nt'r:2t' J.p n- ... j. _ _ . d t. Q. ".C...~. ~ . ~ '.: ='.:=E..a- 'l * :- :r C - ' l...

"3 '-i-E... : ':e.:-.". - /_,,

i = n t. r - 3,.. 9 his EMI:5! 55.E -isli * .lihi Nati?! 4.'5 55iii5fiihE:5! ;-- :i/ 'O.lilD5: "i r#- LL:44 ' '

4..-. h..

.M; E. .t. -l.fl.ar

n._I *:

. =..M. y

n ;.. J....

C:=.

.j. = 2:.. l u.... "33 l t .rur =:- 2 m._ ...._.2 t i-b.hiI i:h***n t= I hbhi$bbi: - -I hE ' - l_kN Y-E '__ hN

• 2*'

G. 1)

.T.i. A.a

:n:n;.. =. : ;:::.

. g;.. n tn.:t. 4..g.n nf r ;.... r.; .; lun u. 1 --

.. : n. r2 4

a ...l -*g

t.t + gr t.::"mr:: _;r: n:l: -

y::: = r: :/:! : ; :.:.= - u. "u :: = an :.:q r z.:n . = rt 2

• _ G. I --" - -

r m t--* N.;

.C

u.. Q n.t*. z,z--[....}=. ;+g--:n -=:.j:= r-

• =: n. s..: :..

-;.u_;.. _ _.. = MT- ' l D"n w.; ;;:----- - - tz-- -F }.

q = n.

N ..u _ r: =

t. tr. -

--+-t t=y - - + -.. ..u.:.a... a.. - a.. _M.;kah" a.:[n_.2-- n t-- .n#=g:. - ---+- rt...._f 5:_- =gt= 1 : _ _.... r r: r.== .._ n..'U. " - ur :...

• j l 2: t.--

--- y - N -~ ~~&--- - g nf.-..w:- 2"nt;:: =Elu.;;;+:-

14.

n ;.;;1.='l nA

.t- _. _,.. ~ -: --. u;.;,nn. : --- ~- - - .c; = _... RP. ns...:

2..

- =::t: r- -- m - :-: g U-M... _... i '.-.M _ : * )- . _...1.' .; a..u. s.....

}l.a.

-, _ I.;.. _ __.d-4-. _ :}.;. ;...... -. .u. g .4__ 4 i

nu-r- -- -----+----- L= ;;;

l :;.I;.n _:- ---

-- :----:. nt? -+-- _ =..==:- -P * - - - ~ ~

Z

...f*". r ;=. 4 .: tir --t_.-. : :: t-.--n:p . :; l.: -t": = _.. -^^:t I -= %) iM- ! ~#M ' 1 T 2 20' CE5 75 55k:f - i.l.5 ^iNiii-ld=- ~r:

----4

- V..._5.,". _n:i.n. 8 =... _ n1.: - - - ~ - g ::=, e., .p. . ~... %.-_g.- _...,.,.., _ r O 2:_:4: w - 3 v.; ..u ;==. e f_ .? -. ~ _ _ r. ._.- De E v e - - - + - - ..-.=:;. to J ... _.:_.. :=.;.:.._ Z gi_. :. , _ l-.n.=...=... _ T _. ].. 2._=.:. :. =. . n. ._;.un:;1 =.n..-..;_;.n-r_. ' -m-- -~--------l- .4__. a :,.. ' - :.n I~a...

• tr :t:- : :.:r.

,..n. ...;=. L

~.c e ".

-. - - - ---+. .n f;.;m --:--- t-t.... _. ..g._. .q__,_.p--- o. %,,, g ;. 2,_,, t ----- 4_ _ _. r c7 -g.__...__; ._-_.f_, .. =. _ _ = p..__ g; .. _.,--__.=1a n } t-: _j =__-.. _ _._._._4--- m,_.,, =..2 N. ,. = : : =. u_,.. .2.__. w -=.:=_=. =_.__.: =. _... _ 2.=._. --: _._. _ ;fr_. =: =_ u.-..._..__ _. E. =. =_.._u. E__ l, _ :.. _.=._:-- -- ,;; y;;n, .,.f m-

n--.-------,-r_--n-._

- - -:__-_---_y.,;;;;;.g;: ;x, . _ _,..;; : u_ ;

= -. -

i.

.. :== t--- -- - - =-*: :

._ urn nn : /=:..; fm-:_ --i=___.__..___ . ; h. :. _-H ----

3.n

n --- = r-- t._.__ . -.. {._=_ 7 .--_=m -:- ;..=. _4n_. .=r___ = = -. _ _ -

==: . - --iu as.. a =:un agj==T :: n;f - t------ r=- ; c:'} _;. =..:ru. . r.: -t:.

= :._. t= =

= --*:----__-: 4;u --- .t-- - :n. .. t : :./:: 1.;nfi.ir.i-S__ =. __ :..T.52'.%... ' 2 2.. _C U-. E2r 2~~ ~ ~ ~ ~ ~ ~ ~~ 2 7" d 'C' 3 - '"- U-f..ii.:.1.:d;.i_113, _~.... - 4- .'E.. =3..i.. -.. _.. "~ -_.1-a :.1: r -L :c._.;r . =. - ;,".,._"..".=._ __...;.. :--- t-- -: nn-t:----. [ t:- .1 =. .[..

=.3.---

..:;=-.=... =.: : m= .1 ;-- -- r -. -= u.;. g

,... _ _ g.; g..

- - - = .OC7,_4- .,,h ._._-*-b-_i_4 97. M*!-*.... : '.* - !. {

• • • • % a

..: g.,y _"".

_'_'_J 3_3.[2 5

-- - < ' -e .E nzs t --" f5 - +-n. .aC M.;.;;;. - =: 24'--V.._.4....a.-..~ ~ =1 ~. : t -r -.. :.., -ur.- l. . u.' y.. . _. - * : h 78-- --_d =h i l ~ ~I t .f_ . n.:.. _:.] =_. t....} .. =... - - - - . - _.--+--. 4 1 .---."-8-- 1- - ^" . : L:..

~.--- = : _-

-I ..a J... I. s .: L.- -- - ~ - .y;;L2rnst;. m. r- =m* ~.; _M N U;IMA; F.e 4 r.u.A f../ .a.. g.. s.. e -- 2.:. .....I.___ . g:u-t... ---"f :- .. ~. L.- :._ := __ u - -1 n.....-t:- t t. r g.p _ _._.4;p:;2t,n**;='.M: nef.. WN P % + -* at:fx:2"==:.7 ar" %.;br;=t'&- -- t:--" - - - - -" tn =-.cf. 204/ 504

e . f,' r1-' : 2: ..j. .::M tt..

• • ~

t:U . L-+ dOYr MN

• • "d E ~%-d Ni 'd. idi.--if

. i P.' - " = ' . TT. - l, i e fp_, tidr:.: , !.! y" - - i

n. _

j gj..g g .p. <. a ,h l E '. l J.q. ty,,

5&

PF $ D P* "* ". -27:.:3 7 m u m , 4; -t ,a a .rH ,,--J rui - 3.s y d as s.,. n. .g ,g a 6 ..g ~i + 4. g jp y rg. ,m ar. -r .4 s.- W m.

r..'.y

x r-2 i.

i a ..g r --b :6 u 4!t f12.:* * ' I. MT [ '~* 4 II*t !. I .-SM ENui 51 N^.I.* " yg

f5 4j 'fp5

') dW -f O N

1. ~ NN N5- ' Eb*-I-5

'#33 U }! 5 < T" " - h. '. -.9 4.Y.N 5 3 '-~ H Js W j. -tr I:'.' ...:.:. ;"t;;:

..::- 7.<

r.cu N

- M r 5. 2. ?. E

Q ;N %

4 :%

^^

.- 2:~

*.

"*1=.: 2:

.n ~1 ~~ ~' ..t: e. P, E. ..4.,.. 3., .....,t,:b.,h.. x.. - - ..... a. l..!.. 3.,

7. t

..,.a.. e .3 .w o. 4 3 ..a ,4, n.t *,. .v W...g .e .p.

g.

-4 L. ..M.. .a.. . :,3 ..J

• .4

..a a. ..v ..a

• *1

.e.. m. .g 6- -.y c i j I;_ r

• ! g

.Z J'- .7 =- 3.. L w.. n.. ' Je _. ":._.f..J.', * * *-23 :.; y... ?. " * '.i :*~

• .J.

M( ~

• t

' 4 "J* - *-. T.'._! W. 9' 2.

1

2.,.' - -

. ;1 g

-.;."'.:I.' ..... d' *:-) ~* . ~. - l g'

'%{

,*. ;_'***3 S_.

m. u, M"' ~ 4..r,i ... s. 1, 4,... w 4 .H .' 1-** .4.- ,, 9 7.. . i g 4 4 Z q....,.a 4.. 4 o 44 g,,, .g, l.,. c. 4.i 4 + ..l

n. es.4

= q. .. W, a J.. V J.. a .,4. ~. t...- T,. ..,e i t- ..+ 4 .~. +. . m.,9,4+... ,w ..% a .7 9. 4

• ".r,t.*.".~"t.*-].,

-' -t,*-"i a ", '.,..' *,.t,,r'...!.*. }, e. .i ..J.. . t-- Q_ l y.* . a..t..g' j -i ---:::.: .{. 'i = .h V.Q.. 7 -'3 p.1 4-g .i J f .. WI.! E: in-- + e -1 .i inn >:Z -i Rd 4 %J '.Q : n n ..a = i m v. .a .e- - - - '-4.. =_ . = _ - .;--.n. :- 3 . _. m --s 7 -. (P \\. n. w.n 4 .i NW: m a . u:. e. , u. n v .m v,

r.,

o, m, t 3 M. -t-Q -D _ y u l O .~, s = .c _-,. = - i. i w ..q. 1.. -..g.y y i .~ . h. .hU L i -a '<: n.- = g e-st [: g .- p : 4 ,l. g v' m )M -l. F 6 N i V, '- l. ( l f.t: j ja .l. .. h yg ... o 4-4-- N ! 3, s .l. 4.. i .4. a v .g .e r -- ..~.. g a..y. t.. O.,, 4l

7.

=-. ~ ' m 9., .: i 4

5

~ Ei .a.- =

'*. #r 5

. i, ~ (. ! ,= i r - --5E

d. 7

= ~~

n

~_: -: I j w a .. ) v. ..2

n..,

_s, 4 -_4-- a.. i .n ... _.., i g -.J J.. h-i: -f -4 ~ = i.1 3L ~l -J.. \\-. "-4 - f. ad :c

l 1

a ..s_- c.. _c. .- ;:.;.. - :- -.:.2 n. i.r -;c

v. i !.~ ~ '.

ie .r y s y -l . [.c.. . :r t-t m s t 0 ~1 u a \\ .w m, w .6 U T l7 7") h j_, .+ r, rg r.; n.= v T-- I 6 l - -~ u a-N . s..y n i 1 .y }. '_ l ~ .y 13 %.q W

i. :-,,j

%y. m E::

• Qi 50 -.l fT T l

~ ~ .: m- =- n.

m:

o. l " ! l

{. "1 -. .-r n.31:r- -r

-- n

^r._3 -r m. q l

-n ;; :. e ~ l ' ~~~ a l " " ~ ~ ' nd. p* n

b:,.- t-c

.- Ru 2,-

.; ;;n

.m ? '-g- .,h an c.

r m.

r ~~ 8 ".'3 @ ~ f7 k T. ; +

• ~

"_J:

r..+ '. + -*. ' *.

~,,'.r.T'_a: -.. '.. ;*i

i.-

J*L U"+1.. ; .*J : l. 'J.,1. a.1 y 3.- e s

-,t

.*, r. 4., m,.- 4. .m.. .u 4 f : [* r !1:1J,.:..5 24 i e . q. y $t y-

a,-{,'

1*" q ;t - ga.... yt! titte j 1T. +d _,, 'e""a,, J.2 . L~ ;d y' 1 a. ni i a,,.. a,,s t. t

a...,.~..

. o

g..

g ,,e y r w , i NN IJ k.,ih:h. 5.n .l 'I.h.. h N C .c-.p-.k

1. e..

..t lh 1,1 ,,i 4 c 3 >g..}] n

.a,'

,} 3 r i,s .~ r s ;;d h w.~ m i. = ~ c 1

r..r,s.rp. :,r'

,,m.;;

• n;.r 4, q c g

-1

'.J.

p m. :

2

.;.n. :.n..A. - F* f i-q s_ a.. s 4

• y'l'v'- % * - a r*-,y.

~ ~ F,, ' ' + ' '- ',. 7- '*..::p' t, 7*: T+h'-' J'

r'u ' -*-

'4' A 6 g i 2 ~

7.. O t.

e ,4 t.

y. n.;,,i H

3 a.'n h,.,,.o,i,n ,a. i a 1 .., a. u. . an ,a, . 3a..,3.-,.n viit -n a, m m ,, n a g u'.

GENERAL ELECTRIC ADDITIONAL CLARIFICATION ON OYSTER CREEK WATER LEVEL CALCULATIONS I. VOID FRACTION ASSUMPTIONS A. Collapsed Level Calculation The conservative calculation of total collapsed level inside the shroud obviates the necessity for accurate void fraction calculations. This calculation is further clarified below. The calculation of initial mass inside the shroud at 172 seconds does invc]ve the calculation of void fractions. However, once the mass is initialized, changes in mass are independent of void fraction. At any time, the " residual mass" in the upper plenum as a result of collapsing all voids inside the shroud is given by: hj a lj - ) [c,g aOY 7p O e gypA53 (1) A 4 fP y y MJr~b]SYi = where M = Total li'ttuid mass inside shroud. T That is, the total amount of liouid in the shroud is calculated and used to fill the shroud from the lower plenum upwards. Whatever liquid mass is left over after filling the core and bypass regions is then the " residual liquid mass" in the upper p'enum.* This is obviously the lowest amount of mass possible in the upper plenum as the voids filling some of the space is the core and bypass regions have been neglected. Once the mass is initialized, the total liquid mass can be tracked by integrating the continuity equation for the liquid, and the " residual mass" in the upper plenum calculated as the amount Icft over after filling the core and bypass regions. However, since a similar numerical integration for M (total mass) had already been pe-formed to calculate M the upper plenum mass, u, thesenumberswereutilizedinthecalculaEionsforcomputational convenience. As shown on the following page, this results in a " residual mass" as defined above and which is independent of the assumed void fractions during the transient.

• This resulted in an initial " residual mass" in the upper plenum of 22700 lbs.

as compared to the actua1 mass of 31600 lbs. 2047 506 In Appendix 1 of the May 12, 1979 submittal, the mass in the upper plenum, was obtained as: f-$ core -N - f owe K f i i 3 gyppSS L (2) up ' i Pt.couof This can be arranged in the form Af ~~ U CORE coke uf ~N fk p y(I;sj f pf Y ~ ~ 81fRSS In the information on Clarification of Responses to Questions (1) and (7), the residual mass in the upper plenum as a result of collapsing all void inside the shroud was calculated as: ( ~k ~ Uf BTfASS g 6yfgs] coK6 core Substituting for M fr m Equation (3), up if,=(Y+Mg V (s) fz i This is the quantity desired in Equation 1 except for the inclusion of the vapor mass in the upper plenum, which is less than 1% of the " residual mass". The same result could have been obtained by integrating: et kk .+ ) ( k -- c3 f (6) T T furrr'L 2047 507 N where is the flashing and evaporation rate, and then applying Equati@ 1. Equation (6) is independent of void fractions during the transient. Further, even if the initial void fraction in the core were off by a factor of 2 (i.e., 0.40 instead of 0.20), the initial residual cass (and minimum residual mass) in the upper plenum would be de-creased by 6700 lbs., but the top of active fuel would still be covered throughout the transient. This shows-the insensitivity of the conclusion to the initial void fraction. B. Void Fraction Values Even though the extreme insensitivity of the collapsed Icvel calcu-lation to void fraction assumptians has been demonstrated, further justification for the values used is provided below. The void fractions were calculated with a correlation which accurately predicts void fraction data at very low flow rates. The relevant data were taken in the ATLAS loop at San Jose. Average bundle void fractions were measured by using differential pressure transducers to measure the elevation pressure difference Thebundle mass flux was limited to values below 0.10 X 106 lb./hr.-ft.' where the friction contribution was negligible. Axial pressure dr p was-measured over 12 different lengths of a 36 inch long adiabatic section ' at the top of a 49-rod test section with a 12 ft. heated length, and an average was obtained for each run. The average uncertainty in' the data is 0.011 and the standard deviation is 0.00467. For the 80 data points, the average error in void fraction prediction was 0.0001 and the standard deviation 0.0044. The range of test parameters was: Geometry: 7x7 bundle Pressure: 800 - 1400 psi Mass Flux: 20,000 - 100,000 lb./hr.-ft.2 Void Fraction: 0.06 - 0.72 The core conditions calculated at 3 minutes were: 4 Pressure: 920 psi Max Flux: 19,000 lb./hr.-ft.2 Void Fraction: 0.21 (average) 2047 508 4 '\\ t Thus the data covers the conditions encountered. The measurements were made at the end of the heated length to obtain a constant void fraction along the length to improve measurement accuracy. No change in void fraction is expected in the unheated region as the geometry does not c:iange. At.these low flows and void fractions, a bubbly flow regime would exist in the core and bypass regions. As the recirculation valves must have been very nearly shut to produce the low-low-low alarm (see original letter) no significant changes in power, void, flow or flow regime occur beyond this time. Thus, the changes in void fraction should also be accurately predicted. (For the collapsed level or " residua? mass" calculations, the void history is of r.o consequence). II. JUSTIFICATION TilAT STRATIFIED FLOW IN LOWER PLENIN IS CONSERVATIVE 'Inc assumption of stratified flow in the lower plenum is conservative with respect to the calculation of the mass in the upper plenum because: 1. Lower plenum temperature is minimi:ed as initial enthalpy is re-moved and replaced by cooler liquid. (None of the cooler liquid is lost). This maximizes the density and mass in the lower plenum. Thus for a given total mass inside the shroud, a minimum mass in the upper plenum is calculated. 2. The assumption of stratification minimi:es the subcooling of the inlet flow to the core and thus the reduction of vapor generation in the core. This will tend +o maximize inventory loss in the shroud due to vaporization. The above conclusions are valid if no significant vapor generation and condensation occur in the lower plenum. Ccmparison of the lower plenum mixed mean temperature with the saturation temperat.:re shows that the plenum was sufficiently subcooled throughout the :ransient to inhibit any flashing. 204/ 509

EXXON Responses to Questions Regardinq Oyster Creek Water Lovej Calculations t Sovide fhrther, qualitative description of the calculations 1. Perfomed to dotamine voide present in the core, owe bypase,! Justify conserv and standpipes at t=Z72 eaa. upper pteman, (actuct inuentory ?_ calculated inventory at that time). Void fractions were used only in a relative manner in the boil off and in the mass balance analyses. Only one flow regime existed Use of the void fraction throughout the duration of the analysis.in this relative manner with the flo minimizes the uncertainty in mixture level as a result of un- ! certainties in the void fraction model. l The flow regime remained constant throughout the analysis pert'od. At the time of low-inw-low level alann, flow through the recircula-tion valves had ceased. Thus, core flow thrcughout the analysis The steaming rate was primarily period was maintained at a low rate. governed by the decay power which remained in the 4 to 27 rangeIn throughout the transient. changed very little (<.07) throughout the analysis period as shown in Table 2.2 of XN-NF-79-47. The major uncertainty in void fraction is associated with the parameter which effect core voids during the period of interest; the decay power, The rate of de-core pressure, core inlet subcooling and core flow. crease in decay power is well known and the pressure used in the analysis was that recorded during the transient (PSMS output). A change in core void fraction of 0.02 was calculated for the maximum uncertainty in core inlet subcooling (subcooling changed from zero to that given by saturated temperature less recorded reci-culatio temperature ). was calculated due to the difference in core steaming flow created by Statistically, combining asstsning no flasning in the lower plenum. these uncertainties yields a void fraction uncertainty of less than This corresponds to a 5" uncertainty in the two-phase mixture The core is calculated to remain covered .04. level in the upper plenum. when incorporating the uncertainty into the calculation. I a d S L Alni, 2047 510 6 1

.2 2. Justify your statement (pg.12 App. 1) that stored energy release from the core is <5% of decay heat after the first 3 minutes. Describe how vessel vaLZ heat, fuel stored energy, vessel intervata stored energy, and coolant "superheat" are removed as Tsat decreases due to depreneurination in the time period of internet la to 20 min. after scram). Row was this accounted for in your caleutationt ~ Values of these heat sources compared to decay heat should be shown at a feu times of interest (3 to 20 mir.. after scram). The boil-off analysis considered stored energy sources in addition to decay heat. From zero to 172 seconds nucleate boiling would remove all the initial stored energy in the fuel. The only stored energy to be considered is that caused by depressurization; this includes the stored energy in the fuel, vessel, vessel internals, and coolant. These energy sources were included in the flashing rate (f4 )f as calculated by equation (3) of XN-NF-79-47. Table I shows tile magnitude of each of the stored energy terms from 3 to 20 minutes. As can be seen in Table I, the stored energy in the core, vessel and vessel internals represent a small fraction of the total energy release. The liquid flashing portion of the stored energy was significant because it was conservatively assumed that the portion of the initial lower plenum inventory (at 172 secondsJ re-maining in the lower plenum during a time step was flashing. 3. Describe how you accounted for mixing in the annutus. What uae the resulting recirculation flow temperature vs. tinc7.Tuatify con-servatiam of-the recirculation flow tementure vs. time (reautta in towest minimum core water tenet). The annulus and downcomer temperatures were assumed to be at saturation conditions corresponding to the pressure indicated by the PSMS computer output. This assumption is conservative since it minimizes the density of liquid in the annulus, hence minimizing the hydrostatic head available for driving the coolant through the recirculation isolation valve bypass lines. There is no means to heat the downcomer above this temperature as the downcomer liquid would flash untti its in-ternal energy corresponded to that of saturation conditions. Therefore, no less head than calculated would be available to drive the liquid through the bypass lines. The liquid recirculation flow temperature was taken as the average of the pump suction temperatures measured and logged by the PSMS system. _2047 }11

un Justif){ that Describe the type of trtizing asswned in the Lover piemon. 4. the assumed type of mixing results in the toucet minimum core water Zevet. The specific volume of water is almost linear in the pressure range of Tnterest. Therefore, the total volume of cold and hoc liquids will be the same whether mixed or stratified. The calculation assumed stratification of the cooler liquid injected at the measured temperature below the warmer liquid initially in the lower plenum. The wamer liquid was conservatively assumed to be at saturation conditions and was conservatively flashed-off to reduce core-side mass inventory. Analysis of the P5MS output data showed that the lowerpienum was sub-

3) alam.

At that time, the measured cooled at time of low-low-low ( pressure was 935 psia and corresponding saturation temperature 536'F. 3 alarm Pcior to the trip the measured temperature was 526*F and at Therefore, the liquid in the lower plenum was no less than 505'F. The assumption of 10F* subcooled, and probably 31F* subcooled. saturation conditions in the lower plentra minimizes the eventual 3 alarin, no voids liquid level and, since it' is clearly subcooled at can exist in the lower plenum for level swell t, the liquid level in the shroud and consequent cori and shroud mass inventory available for Further calculations based upon boil-off is, therefore. minimized. the PSMS-output indicate that the lower plenum remained subcooled throughout the event. is capable of condensing the maxirmen .ivotify that the iso-condene steaming rata it la ascumed to condenee in your cateutatione (including 5. courece from question H2). Examination of the measured pressure during the event shows that the isolation condensers removed system internal heat generation andDe-stored energy in accordance with their heat exchange capacity. creasing pressure indicates greater heat removal capacity than internal generation, and tho depressurization rate is then limited by the heat exchange capacity to remove internal stored energy. significant extra heat exchange capacity is demonstrated by a single isolation condenser from s70 seconds to 4.5 minutes where the pressureSince the flashing decreased from approximately 1000 psia to 850 psia. rate in the calculations corresponded to the measured pressure, a match between capacity and heat removal is maintained in the level calculations. 2047 U2

Table I Sum of Energy Sources During Transient (3-20 min) Fraction of Stored Enerf_ VpC (BTU /*Ex10) 4 (BTU x 10+6 Decay Energy Source Decay Energy 45 Stored Energy Core 2.2 1.2 .03 Vessel

• 3.1

.4 .01 Vessel Internals 3.4 1.5 .03 9.8 .21 Liquid **

• A transient conduction analy' sis showed a maximum of 30% of vessel energy could be released over a 17-minute period.

Only the portion of the vessel surrounding lower plenum was considered.

• • Included assumption that lower plenum flashes.

204/ 51 3 es

ADDITIONAL CLARIFICATION ON OYSTER CREEK BOUNDING ANALYSIS LOSS OF l'.'7?NTORY TRANSIENT ANALYSIS 2047 314

RESPONSES TO STAFF QUESTIONS ON MAY 19, 1979 SUBMI' ITAL, LOSS OF INVENTORY TRANSIENT ANALYSIS RESPONSE TO QUESTION NO. 1 At the time of MSIV closure the level inside the shroud is being held high by the continuing recirculation pump coastdown. With the combined action of significant flow and rapidly decaying power, low quality water is spilling over the separators into the downcomers (approximately IS.5 feet above the top of the active fuel). The core spray valves are opened at 285 psig. The collapsed water level in the shroud at that t'ime is estimated to be 6'-7" above the top of the active fuel. This estimate is based upon the total liquid mass of the system at that pressure corrected for the specific volume at saturation conditions. (See response to Question No. 4) Since the 285 psig condition would yield the minimum liquid mass to f.11 the system, this would also be the point of minimum collapsed water levtl. No credit is taken for the mass added by the control rod drive flow in d veloping this estimate. RESPONSE TO QUESTION NO. 4 The extrapolation of minimum collapsed water level inside the shroud beyond the PTSBWR2 simulation period is quite similar to the techniques utilized to determine minimum water 1cvel during the May 2nd scram transient. The total mass in the system at the time of system isolation is determined from the PTSBWR2 analysis. During the depressurization, following the 125 second transient simulation, the liquid mass and specific volume may be determined from t!;.: saturation characteristics at any pressure of interest. The resultant collapsed water level is derived by distributing the available liquid volume through the system volumes from the lower plenum upward. The minimum mass of liquic would occur at the lowest pressure prior to the core spray initiation which terminates the event. As noted in the response to Question No. 1, the minimum collapsed level occurs at 285 psig and is estimated to be 6'-7" above the active core. 2047 515

RESPONSE TO QUESTION NO. 11 There are actually three time delays associated with the automatic operation of the isolation condenser. The first is the delay from actuation signal (low-low-water or high reactor pressure) to discharge valve opening initiation. This delay time is associated with " seal-in" time for isolation condenser. initiation and may be set by the plant operators. The second delay time is the opening time fixed by the electrical and mechanical characteristics of the motor operated discharge valve and is approximately 20 seconds. The third delay time is the high flow isolation signal that is concerned with the high recirculation flow signal which causes isolation of the condensers. This latter delay is currently set at 35 5 seconds, and was addressed in the responses to Questions Nos. 8 and 9. The bounding loss-of-feedwater analysis presented in our May 19, 1979 submittal explicitly considered the first two delays discussed above. For purposes of that analysis, the first delay, from actuation to valve opening, was taken to be 10 seconds with no reset upon clearance of low-low water level. The results of the analysis show that brief water level surges, due to action of the pressure regulating control system attempting to maintain system pressure, could be sufficient to recover low-low water 1cvel and reset the delay timer. The analysis shows that time period from initial low-low level indication to the brief recovery is 3.4 seconds. Therefore, to insure proper " seal-in" time allowance, the technical specifications will require that the delay be set at less than three seconds. The change in initiation delay time will have minimal impact upon the transient results presented in our May 19, 1979 submittal. The isolation condensers will initiate earlier, but the time of MSIV closure and, consequently, systda mass inventory will be unchanged. Therefore, the minimum collapsed water level will remain the same as reported (6'-7" above the core). The only difference as a result of the shorter delay time would be a milder pressure rise following FGIV closure and a slightly earlier depressurization. 2047 516

CLARIFICATION ON THE BOUNDING ANALYSIS FOR OYSTER CREEK Response to Question No. 2 -- CR0 hydraulic flow was assumed to contribute zero inventory during the entire transient. This is conservative since at least one CRD pump would actually be providing CRD flow to the reactor vessel during the entire transient. Response to Question No. 6 -- No credit was taken for the mass of water stored in the isolation condenser in the Bounding Analysis. Response to Question Nos. 8 & 9 -- In order to assure that isolation condensers would operate properly even considering four (4) loop operation, pump coastdown curves and the condenser isolation logic was checked. A copy of the pump coastdown curve is attached. Note that it takes approximately twenty-five (25) seconds for a pump to be in essentially a natural circulation mode. A check of the isolation logic for the isolation condenser and Oyster Creek Procedure 609 3 002," Isolation Conde see Test and Calibration",show that should a high flow isolation signal exist, it must persist for 35t 5 seconds. Even assuming that no other time delays existed, this would ensure that isolation condensers would function properly if they had initiated automatically. Plant experience, as verified by alarm panel data sheets, show that on December 13, 1978 and February 2,1979, automatic isolation condenser initiations (time delay s I sec) occurred simultaneously with an automatic five (5) recirculation pump trip. These events did not resuit in any high flow isolations of the isolation condensers. These actual events, combined with the knowledge of pump coastdown behavior and high flow isolation logic, confirms that the isolation condensers will function as designed when automatically initiated for five (5) or four (4) 1 op operation. $0bl bhl

nanuat noot RECIRC PUf1P COASTDOUN RATE I I I I I I I I I I I I I I I l i 0.9 2 F R O.8 A C T I 0.7 0 N ~ ~ O F R 0.5 A T E D 0.4 S P 0.3 E E ~~ D ~~~~' rs-C) 0.1 s N I I I I I I I I I I I I I I I I 0.0 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 TIME (SECS) .}}