ML20212F736
| ML20212F736 | |
| Person / Time | |
|---|---|
| Site: | 05200003 |
| Issue date: | 10/31/1997 |
| From: | Loftus M, Spencer D, Woodcock J WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | |
| Shared Package | |
| ML20212F716 | List: |
| References | |
| WCAP-14989, WCAP-14989-R01, WCAP-14989-R1, NUDOCS 9711050077 | |
| Download: ML20212F736 (264) | |
Text
{{#Wiki_filter:. , w p ,' w - , . , ,. ,.. , .. . . ... . . . ., . . 4 ,;,,,;. . . . . , , .4 . . , ,
'.,p ;p'. . ',idg a"7
< * . 3 .,; ". . TVy,q. ', gl' ' .6 - <
,. 4. ,' . # . .e 'i , .-(;' .r., ., y .,",'g .I,, . - ' ' . ,;, ./.9 ., ;~ , : ' n .
. ,". - . .. ",.,og.,#',
.r .- .*9 s*
e' o
, ,f,p..."g' ita'J*'- ec,
*,e , ': . .4 * , ? al 4 1.1 ',.g '%
.( * ' ,. *.Tc P 6 l' ' ;. F Y
., ' ' . J' .' . .
._ ,f. .y *
/ ,9.'j,'... ', '.g .; : [* p' ' ..l, y . '* P, . , .b o .~ '
"...',s ,..'.,.,4
. l U. .J. ' .'jo -..q.. , . .e. .,7 y.% 'y' . .
.' ~ 6 .,..5.
f . .
*. i. . k,,J. d 7 .
. a.v t n,.'a p'.
.n . ..'1..,.;~..
(-
., .,/ .? $? .. a . < - (* e, r g f - c fA
. . '., . .g .s t. ,
V = ' -
. ,' .,..t. . , . l$ ", h . ' [. . ' '. , -I) f . , i' '
l ' 1l
,.",. ,; . ', '. ..4.(:'c . "'. ' ' :f.'j 37h; g
ay.. ,.S - 4 , ). ,. p #;.",'.
.jf ,,
'. , h ' .' ' .
+
,.',[f, d. ,:.ff($.l'. _.
."1.; -
s'i *
. i. D, og , J. ; :. tlg g
. ' '. . :.--.g,; - ..-.....,;,...: .. ' - ..,. , *-;
.".,. ..- ., ;, 4+'. * ;7.#4 , , e. , ' '* f. .
' . j?.. p .r.
kje ,,s.
..... ; '. '; . e.......,yp.r....,.g. .e e =
.._,c: .
* .j .
. .g.. .
. .t. s,(* ..
+
,,,,f 3 , . . _ .. . ;-.
'n'
. t. ' ,. ,. ' ". ,.:'- __, ', ., _.y., . . ;'._ . , :. M ': l; :> . . , ;, _ Qh- f. " W
.- - ;: '.'*. .1- . t l*' . g* ;p
- _ . . , .. a.-a :1.j
~ e' w a'<. .
.4> ' -
.', ' . . , e ;. . '. . ;: .
., . ;.. . ,. ,1. g., .: .n.. , .r .:..;.f l,(, ,y;,4, ',, . . ;. .p.
- , , ;;._ y y , _. . iQ. m;p ,, ,g;, 9,,
. ' '...g i. , , . ' d . .,' ,' . , .-g {;.. , ,l+::gl g ; _ = t . l y (J '_.. , g 9
, ' j {' ,-{ !,.' .
- . :, .- 'y ' : , ' , . ' .: J'.: 3 9 . .Q: R*'g *: '.q'j.y ;. q . y, : ,' .f l - ~
.- . -_ .-,; { ;> 4.'.. :;.. 't: -: ~-
b 4:.ys' .- e
.- ' : w .
- "( ; .'; :.( c3. : .'-?
.d . ,'., , ; h' '_Q, ,.,%s d*.f'(:
c .n 7 :g_:.h:;<5: J{ <,'.
- ' " . . . l ' . ,A . ,~.....;,e j [ ' _, '., " ". .:}.4 ,
., 6/ g .; ;.,y-{ ,
f
' _ . .'g. . n. . Y- . , : ._ :;:: n
; .b '
',' .: : . 2; . .. > . . ';l.y .%,. '._ -
- ' '. ,1l5Nl" ): ' Y
; y:, m' c(_ y ~ .Q y"'t
. '.f;
- w. . -.' , . . *.c y v **
- tJ..t
' , ). ',...v. . ', '.y:'YW$,;',. .n . (.;'f .. (: :A. ,'sf,..,.>.*,'y' , ' ...
... ,, g-.1,. . . , .'..%.* , \'. . 0
+ q y ' ;. .; ',
, . e. ,W }p .\ '
, . S. .'r .
}. ;'* . ~~ .), ., ' . . , ,
'Y .
-
- g":' : ; ,, ', , ' .. ' :. , ' .l, .. ^
.l ' l ' ..\ . ' ' . ' . . ' ' < ^ ' '"_', . Y. o
-s id
. .' . *; ". , ..e ,** . m ...p...
.,..?< 4,p.c, f '. j, ..,.,.8.'
2
* ',.e,8.._
. ' . . .. * .s;*
-[ . .. * , . '.,e *
.q
- 5
- 4. .
,'g*
- , . i
; . . - ;. ' , , 'E j,' .. .
[4)"%, ;.:
- . , l- .. ,
[f : i ,. .. . .[ ,-l !. ,...".la,, ,' ',7, .: ,';i " . : : ., -
. . . .: . . . .' . . . ' .. .. , .,i . ' .4,c.'.'.1/ h.: : . .; .T .4.l, t, ;.- J. ; t?:W.W y s,
- . '<p.'.
7 . Sf 6. . .S : " ^
. '. . . . _..- . . -.'.i , ,y : ,:., . :, !'..,:. :. 7 ...f . p+f'pb"' . , 7
-:' ,;r .;.. ;;:'.7., ., ?.
. a. . : ty,4 s .. E:~;;.p - .. .
. .g" . : .-. ;:' ..'
~ . *, . , . . . .. . < . . . .
. ;.- , ;; ;, ;' ,, _._. : g. .. ,
.;f , ; . ,r - g. . .,,.",,.g._.,,.f.b.,*.
8 '.- -
; . . - i
*.i - -
.c
;s.. *. .:.,...'..,.x,.. . : :. ' . ' . JL. , p g %.. u.,.:,'.'g'_
. . .s , :. - :. - , : '., 4 .. *
- .= , . . . . . . .
, g
? :*.
=
. . . . -.'.,4 e.a. : .
.. ,-..,uf, .... -;....y.. ,, . . ,s
...,.,...e,.. ,
. , . - ,'. . . , . t .l,
* [ '- . .f. ', .,- r .' l .
- 4 . .
.n. ., ,, . ..- 'r.. .. ? ' y,, . . ;.,<', ,.e . . .' .\, : , :>. -' . s .. .. ..% :,s:, ;, e,s
. .n
',~L . :
. .' .s ' :. . . 3,
' . .' * .'cy ... . -
;,:.,..._'.* , s w_ . - t. ;'; . _{ *
.- .,Q..
;, .* n. _. % :;,.*, ; -:.. _. ',_'i.-! , . . l kl .y.
.o.
.. . y . . .
t
". ' ' ,'}; ._.,m,,
1"s Jilk
, . . ..s. , ;.
.? . .
b,,_.,_
".:. . . ; ' p, s
... ' ., a'. t . . ., , . - ?..' . ~; . m V e ;; ', J _ * ' ;
, .; Q' .'r,.;: . , ;.;. :::.f_
..y.:_i'. :: .:
..~ ,_
'..*', '_,_...e. _
. .; :....*., :) ' -; ,
.,_..'f,.- .:: ); :, .:.,, ._}.f,y,.: "4 , y ' .,-
...,e.
,:.,.'*- 's.* , y '
...."...,',..A:
. . i , : .:. .i.,f! 2
- p. - -
- 9: - ;
.-.. .. ..3 , _ . ;_. . q , , ..!
/
..s ,. ,... .
..v.
- i. .V.,'"..'.n..
- -,. . . ;y; .(; -
; ..' .- _ . . - . . . : ,.,5; ;..; ;,. ;g:.y; ; ..,.,,_e . '-Q ,, }.,: . .
,4 . . ,
s
' :- l1.- ..p. '; 4 El:. . .? f.. #.4' i . .; s ,
(- ; ,,' N. ' ?. .
' :. 'ol.,'. ,;
_'y.'7_....;,"'.;._,'..,
.% .'.:'-
- g'::, {
', . .M , :.....'?'.
,. e
.,.., f ,.-l;.J
- .* ;;. . i
! -(* S '. :.
- i.-l.
. g*, .: ;' ,.,.,., lg
(, .
~
p ::,,- ;d - '
- '. .' ,..g . -,-.' ' ., . ......_';.
e
... f
.- .. .... . . ,n*.........V;'gr."..',,... , . , - . ' *,;..
S'.6. n
- . .~ _. f d_,';,:. K'_. .~, )' , .. . ' - %,
.;s, , .. . . g: . : >,,... - ' .% : i ,.... ' '
' e : . i,. * :: .
3
.. .._..';...,. * ...p: . . .*,::-
3' .t. .j; ::., . .*? u .'
\
- i; . . . .,.... .':_'.m, x.-.Y. ,,.,..;
.:.s
.. s
. . e.;*.,.'y.
. . . . ~.
..: .o
, . . ,. .- s =. -
'. ; '.p.;
;- 4
' '. . . ' h' ; } ~.: .
-[,.,.,. '
.. - ;y Q:. . 'l l},,',%,r
- ;. .l '.-. .ij;,;_2 P_ - -
. 4. ;
. . . . -, __ . s- .** * =. . .
. .; s'
.- '....,p',.
. . . , .;-, R.'.
- ; ',,(- . . . . S .
, . ' .', . , . ..; j._ ',,"l ; ', ', ) l , ,..' ._ >f.
. - \. . .,f.o '. ,.. . , '. : , l '. . . , ' , . .
~ _
_',, j f' j_I . '
; :~
=......,....g'. ' .' '} : : ' ; ,'_,j, . ?l.} *;,L t_ f _ ._'.N.'w.. l, b
. . s .s . :-
.' -;- r.
f; n o .' . ' .. 6 . .. , *: o ; a
- ' . . . . ' _. . ,-- Jce
- a.b.'. "...
. . ' . , . . . ' . - ...:,' ' .(
.. '4. ... {7'. f, y'.,. i. /.
.Q . :- ,l..' ; . '."- (9; d ',. - ' ,; . ,,
. Q ts' . . .: .,..
- - ; , /,' s p ,
.c. -
. A. *.. ,.. - . - . ' . , ...; .,..T : ..
g
.'3-l'..; .'. . a.:. i., y ;. .; ,' . . ',- -;.1l
;- ..4.._,...,
, ' . .n
-r Y'.1.-:....,._,,... : %i,,.? . , . , . j ). ; ;,, .
- ,. .p t , t . .
) n,,
e g d -
? ..f.,; :. .t- a ,, . . . . . . ,..,...g.. r.
. :. c,,; y.y l ,. y l;; p .. .:. ,6 , . :. ,.f.
.. p ,, ,~.: R- , .. ,. %- ..,...,.: ; : , ' _ : s :3,. . .- :. ; - .
c., ,. ;.: . t e,e.,.'3
} .% ; . _ . __
- - .- . ; . i, .. s..,
e
. ;y; .. . . . -p( s.w . , q ;, ,.-;
e. y
. . .. .:. . . .; .', ,::;./
_ , _9, . : . . ,, j - ,j $. .J.,y -[. . ;_, Q . ..}. ._:?Q[? 4., '.:.,. . .l;- .;.[_; ; ;.
-n . .,
g;).o.[v _.:.) y; ; _;.~
* ,t ...
;p.. s. .. . - .:.. . : s. .. . ..-,;~l ... ~-:. 't'). . . . : ,..r ..:... ,.p;r..
a yg :p,- ;g...L. ..
.- : n,;. ' . _ - -,
..n,.. . \
. . . , . . i ,;. .i t.; . -.
_ ; :. . -. , . '_: , _ . , _ ( , :::; y, . ._.; _s[p. g, :36. . , v:n, i .. g .: .... .. _ ; .;. .._ G... . .. .
.'c :- .-
.. g.
' 3,-
v q.. l.,7.;
. . . - . ... i,k., ' .' ":. ; y: *::.g
. .~. . . ;t: . , '. ..).-.,5.- '. , , , ' , :
t.'. :. ... . . , . y '. ' ,l. _ ... ~
.. .. .y . . .'. g .E*. yj> , -
'.. J.
. , . . .I .
...'4,'
* ' ' . . . - ' '. ; .'.} [. '.,4 { . ' ;' . -
,; ,'.g[;g' ;': h ? : . yl ' , . ' :q., , , '
.,8 6.
y: .. U-{:*., ... . .'.
.'q# ..; * [
~~.-N.. 'E ,f . 1*.[- 'f'; '
'?
' .. . ' hf * ;: '.' ., f 4.. - _.
f'llll. +%'; ? .f'*'.'.;'. _ O - k ' _'.'f :byq '.. : '. .c,
.e .:
_,...,..'.l.' . , ' ' . ' . ' ... ) :: ; 9 . .!a,',. .M ,
- g . , a ;Q ;l[ . G. Q. Q;' l,.
, , . ,- . p ;? m ' .- n. .m.. .sc ' ..s . . :;, *;w. _a, ,. . ; . ',h..
%., s.- * . . . %. . .l*.. . . . . 4,.
c . .g . a
,.t -_ , g,i %l:; b..... % ,.. ';'i _ w?< . ' O . ; -l
'- ..' . ..'.,s,~... .:..< .
.;.:ls.f t .v,A'."r j ,;9 ..p ,? bi*
.. .' . !: ld ':.. .ti'-;['Q.y . ,. h,.'..'e.. 4: . . ' '
y' Q.,,3 ,' . .- ::
- n
\'.- .- -
g p' l 4. /. = ,'; ,g
- 5- .
y
;lg3 . 4:. . . . 'w,.',1g! t .
,..i....,--g
- l+;. -
. . .. i p . r. ..
J.. .a .y%j.Q, ,.xys-:,, .Y ;,Q
- . .. e. -Q
;.) i % f gg;gs s' 4.,
-:.'. j, i '. ' y.\
' .). .l.)p...-;[y . ._ Q '_.') . gg,Y..xQ'!l%. ;.
, .. g - ._y....;...-,. ; .s..,
. ; . r. - : v. eN}.;ll
- .. . s
.s' D.
.t ?... .e
.y :. y ,, . .**t',_' .*
. '. ' :.) ':. . -.',u' p,y;.b
- ga 2<
> c.y.;1%.8 s'k.)' c s.%.y:y h:W ;.f*. g'f,~y.
* .gj . l4.< Q, *4.).4ya y..pl-:!Q 1. ,
e.
.e- , e & - .' .-
G ',
; .t g.
? %;f.f.z.:M y lsQf,ap QCQ Q ' bQ'* $'h.g
;L%. .&pf gl?'2l&il.t:~
.?
. . _If a
'.' l.' . ,- -
.a.y k..
n
.;..n'...a_ .p bd. :QtQ
.h' y.;. .
yL y
. .._:. %p;: . ,:. ;,c .e .9s,.i%>a.- 5 y.. .. h. 4 y e,.g .v
- 3. . . -
. , . - - '- , . : , .. ;. l2.
. 4. . - q._ ; q, s . ;; -4 ;9;e,ygyp:,.;. t:a qgQ,.,gg.. .m}p? t.p:?
,;;;..mn y 3 . age m,;.g;.p\ .. . ,, ; . y 3
*;' O
. , . . .. ., . .n 6 4 s ;. b.i -0 .: :. h.{ ,. ,
3':Q..msy E
. [.
[. . . . . . . . .; ... . . . ,' c;. ;}. -}.m,; .1
): 3h (,, ** . {,.
- l. hW
. p,
- %,..f Q .? p.: , y mM y
?
'..l .;z.. .. .l:. (.1 ,.. . ; ' .; 3.;.;",:',,. ,f,9..a.'.c.g
. .a mk 3 ' Q1 ,, @
' 9 . , . . . . g. 3 *g '
- 3 j ;:' ;' . ?.. l -l .) f.; .f.. ' _
. :; l 'kU{,Ef.'f ,
f j 7 .g
; f -
- f{ -f.I ;
F, $p.6
- g.f.
- J .. 5 plwc J. '.1n .- .
a Ql':::m.cg.:g 4 ,
. .. . .. . . . .,,;.rvw - . ,
% 3 . %.m
. %. .* { (, hp , hki ';
Westinghouse Non Proprietary Class 3 WC AP-14989 $$$$$$$$ Revision 1 N Accident Specification . - and Phenomena Evaluation for AP600 Passive ' Containment- '
~
Cooling System -
\
Westinghouse Energy Systems s P E; c 20 LO3 A PDP
k WESTINGHOUSE NON-PROPRIETARY CLASS 3 WCAP-14989, Revision 1 J Accident Specification and Phenomena Evaluation for AP600 Passive Containment Cooling System Mike Loftus Dan Spencer Joel Woodcock October 1997 Westinghouse Electric Corporation Energy System Business Unit P.O. Box 355 Pittsburgh, PA 15230-0355 C 1997 Westinghouse Electric Corporatir>n All Rights Resene.i o:\3692non.wpf.lt>.101497
.)
i TABLE OF CONTENTS LIST OF A CRONYMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi SUMMA RY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
1.0 INTRODUCTION
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 1-1 1.1 OBJECTIVE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1 1.2 REPORT ORGANIZATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3
1.3 DESCRIPTION
OF AP600 PCS DESIGN AND OPERATION . . . . . . . . . . . . . . . . 1-3 1.4 OVERVIEW OF PHENOMENA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-7 2.0 PP.OCESS FOR PHENOMENA IDENTIFICATION AND RANKING . . . . . . . . . . . . . . . 2-1 2.1 PROCESS DESCRIPTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1 2.2 TESTING PROG RAMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3 2.2.1 2.2.2 Heated Flat Plate Test ..................................... 2-7 Wind Tunnel Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7 2.2.3 Condensation Tests .......................................2-7 2.2.4 Air Flow Path Flow Resistance Tests .......................... 2-8 2.2.5 Water Distribution Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . ........ 2-8 2.2.6 Small-Scale PCS Integral Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9 2.2.7 Large-Scale PCS Integral Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-9 2.3 SCALING ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-11 2.4 RANKING OF PHENOMENA ..................................... 2-12 3.0 ACCIDENT SPECIFICATION . . . . . ....................................... 3-1 3.1 ISSUE AND SUCCESS CRITERIA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1 3.L1 Design Cri teria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1 3.2 CONTAINMENT SYSTEMS AND STRUCTURES . . . . . . . . . . . . . . . . . . . . . . . . 3-2 3.2.1 Inside Containment ....................................... 3-5 3.2.3 Outside Containment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-6
3.3 DESCRIPTION
OF TRANSIENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9 34 EVENT SCENARIO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10 3.4.1 Initial and Boundary Conditions . . . . . . , . . . . . . . . . . . . . . . . . . . . . 3-10 3.4.2 Loss-of-Coolant Accident . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-16 3.4.2.1 Description of LOCA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-16 3.4.2.2 Temporal Partitioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-20 3.4.3 Main Steamline Break . . . . . . . . . . .......................... 3-24 3.4.3.1 Description of MSLB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-24 3.4.3.2 Temporal Partitioning (MSLB) ........................ 3-25 4.0 PHENOMENA IDENTIFICATION AND RANKING . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1 4.1 PHENOMENA OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1 4.2 THE AP600 CONTAINMENT PHENOMENA IDENTIFICATION AND RANKING 4.3 TABLE........................................................4-8 TEST AND SCALING RESULTS USED IN PHENOMENA RANKING . . . . . . . 4-15 4.3.1 Test Results Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-15 4.3.2 Scaling Analysis Results Sununary . . . . . . . .................. 4-18 4.4 RANKING OF PHENOMENA LISTED IN PIRT . . . . . . . . . . . . . . . . . . . . . . . 4-20 o:\3692non.wpf:lb-101497 Kevision 1 October 1997
il 4.4.1 Brea h So urce . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-22 4.4.1A Mass and Energy Release of Break Source . . . . . . . . . . . . . . . . . . , , . 4-22 4.4.1 B Break Source Direction and Elevation . . . . . . . . . . . . . . . . . . . . . . . . . 4-25 4.4.1C Break Source Momentum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-28 4.4.1D Break Source Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-31 4.4.1E Droplet / Liquid Flashing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-33 4.4.2 Containment Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-36 4.4.'2A Circulation / Stratification in the Containment Volume ........ .... 4-36 4.4.2B Intercompartment Flow in Containment Volume . . . . . . . . . . . . . . . . 4-39 4.4.2C Containment Volume Gas Compliance . . . . . . . . . . . . . . . . . . . . . . . . 4-43 4.4.2D Fog in the Containment Volume . ...........................4-46 4.4.2E Hydrogen Release .......................................4-50 4.4.3 Containment Solid Heat Sinks .............................. 4-54 4.4.3A Liquid Film Energy Transport on Containment Heat Sinks . . . . . . . . . 4-54 4.4.3B Vertical Film Conduction on Containment Heat Sinks ............ 4-56 4.4.3C Horizontal Film Conduction on Contai unent Heat Sinks . . . . . . . . . . 4-58 4.4.3D Internal Heat Sink Conduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-60 4.4.3E Heat Capacity of Containment Heat Sinks . . . . . . . . . . . . . . . . . . . . . 4-63 4.4.3F Condensation on Containment Heat Sinks . . . . . . . . . . . . . . . . . . . . . 4-65 4.4.3G Convection From Containment Volume . . . . . . . . . . . . . . . , . . . . . . . 4-68 4.4.3H Radiation From Containment Volume to Containment Heat Sinks . . . 4-70 4.4.4 Initial Conditions Within Containment . . . .................... 4-72 4.4.4A Initial Temperature in Containment . . . . . . . . . . . . . . . . . . . . . . . . . . 4-72 4.4.4B Initial H umidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-75 4.4.4C Initial Pressure . . . . . . . . . . . . . . . . . . . . . ....... ............ 4-77 4.4.5 B reak Pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-79 4.4.5A Break Pool Circulation / Stratification ......................... 4-79 4.4.5B Break Pool Condensation / Evaporation . . . . . , . . . . . . . . . . . . . . . . . . 4-81 4.4.5C Break Pool Convection Heat Transfer within Containment Volume ... 4-83 4.4.5D Break Pool Radiation Heat Transfer within Containment Volume . . . . 4-85 4.4.5E Conduction in Break Pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-87 4.4.5F Compartment Filling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-89 4.4.6 IRWST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-91 4.4.6A,B,C,D,E, F . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-91 4.4.7 Steel Shell . . . . . . . . . . ................................... 4-93 4.4.7A Convection Heat Transfer From Containment Volume . .......... 4-93 4.4.7B Radiation Heat Transfer from Containment Volume to Steel Shell . . . 4-96 4.4.7C Condensation on Inside Containment Shell . . . . . . . . . . , . . . . . . . . . 4-98 4.4.7D Film Conduction on Inside of Steel Shell . . . . . . . . . . . . . . . . . . . . . 4-100 4.4.7E Internal Film Energy Transport on Steel Shell . . . . . . . . . . . . . . . . . 4-102 4.4.7F Conduction Through Shell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-104 4.4.7G Heat Capacity of Shell . . . . . . . . . ......................... 4-107 4.4.7H Convection to Riser Annulus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-109 4.4.7I Radiation to the Baffle . . . . . ........ .................... 4-111 4.4.7T Radiation to the Clumney . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-113 4.4.7K Radiation to the Fog / Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-115 4.4.7L Outside Film Conduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-117 4.4.7M Outside Film Energy Transport ........ . . . . . . . . . . . . . . . . . . . 4-1 19 4.4.7N Evaporation to Riser Annulus . . . . . . . . . . . . . . .............. 4-121 4.4.8 PCS Cooling Water . . . . ....... ......... ..... ......... 4-124 o;\3692non.wpt.it>.101497 Kevtsion 1
- October 1997 =
y* % e h, d ;, '
111 4.4.8A PCCWST Wa ter Flow Ra te . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-124 4.4.8B Water Temperature ... ................................. 4-128 4.4.cC Water Film Stability and Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . 4-130 4.4.8D Film Stripping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 134 4.4.8E Film Dra g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-136 4.49 Riser Annulus and Chimney Volume . . . . . . . . . . . . . . . . . . . . . . . 4 138 4.4.9A PCS Natural Circulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-138 4.4.9B Vapor Accelera tion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 141 4.4.9C Fog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-143 4.4.9D Flow Stability . . . . . . . . . . . . . . . . . ................... 4-145 4.4.10 Ba ffle . . . . . . . . . . . . . . . . . . . . . . . . . . ......... . . . . . . . . . 4-147 4.4.10A Convection to Riser Annulus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . 4-147 4.4.10B Convection to Downcomer Annulus . . . . . . . . . . . . . . . . . . . . . . . . . 4-149 4.4.10C Radiation to Shield Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-151 4.4.10D Conduction Through Baffle ..................... ......... 4-153 4.4.10E Condensation on the Baffle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-155 4.4.10F Heat Capacity of the Baffle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-157 4.4.10G Leaks Through Baffle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-159 4.4.11 Ba ffle Supports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-161 4.4.11A Convection to Riser Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-161 4.4.11B Radiation from Shell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-163 4.4.11C Conduction from Shell into Baffle Supports . . . . . . . . . . . . . . . . . . . 4-165 4.4.11D Heat Capacity of Baffle Supports . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-166 4.4.12 Chimney Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-168 4.4.13 Downcomer Annulus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-170 4.4.13A PCS Natural Circula tion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-170 4.4.13B Downcomer Annulus Air Flow Stability . . . . . . . . . . . . . . . . . . . . . . 4-172 4.4.14 Shield Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 173 4.4.14A Convection to the Downcomer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-173 4.4.14B Conduction Through the Shield Building . . . . . . . . . . . . . . . . . . . . . 4-175 4.4.14C Convection to the Environment . . . . . . . . . . . . . . . . . . . . . , . . . . . . 4-177 l 4.4.14D Radiation to the Environment . . . . . . . . . . . . . . . . . . . , , . , . . . . . . . 4-179 4.4.15 External Atmosphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-181 4.4.15A Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-181 4.4.15B Humidi ty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 183 4.4.15C Recircula tion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-185 4.4.15D Pressure Fluctuations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-187
5.0 CONCLUSION
S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
6.0 REFERENCES
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1 APPENDIX A - SYNOPSIS OF PIRT EXPERT REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1 c:\3692non.wpf-lb-101497 Kevision 1 October 1997
IV LIST OF TABLES Table 2-1 _ Containment Analysis Processes Used to Initially Define Test Program . . . . . . . . 2-6 Table 31 PCS Compartment Gas and Heat Sink Volumes and Areas Considered in Evalua tion Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-8 Table 3-2 Comparison of Key Containment Analysis Resulta . . . . . . . . . . , . . . . . . . . . . 3-10 Table 3-3 Initial Conditions for AP600 Containtnent Pressure Calculations . . . . . . . . . . . . 3-11 Table 3-4 Sequence of Events Leading to the Development of the PCS Cooling Film . . . . . 3-22 Table 3-5 Large-Break LOCA Sequence of Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-23 Table 4-1 Phenomena Identification and Ram.ing According to Effect on Containment Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . 4-9 Table 4-2 Summary of Conservative Estimates of Postulated Sources of Hydrogen During a Containment Pressure DBA LOCA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-51 - b o:\3692non.wpt:lb-101497 Key 3sion 1 - October 1997
_ _._ . . _. ._ _. - _ , . ..._ ___ . _ . _ . . . _ . _ _. _- __ _ .._ . . ~ . _ . v i
- LIST OF FIGURFS -
4 1 , _ Figure 1-1 Relationship Between AP600 PCS PIRT, Testing < Analysis, and Evaluation Model .12 ! Figure 1-2 Passive Containment Cooling System Arrangement . . . . . . . . . . . . . . . ....... . ! 1-4 , Figure 1-3 Schematic of Passive Containment Cooling System Showing Major Flow Areas and . Heights . . . . . . . . . . . . . .........._.......... . . . . . . . . . . . . . . . . . . . . . 1 -5
- Figure 1-4 Genera.lized Schenatic Representation of Containment Pressum Relationshi l
Transfer through Momentum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-9 . . .
.p to Heat
- Figure 2-1 PCS Phenomenon Identification and Ranking Confirmation Process . . . . ..... 2-2 Figure 2-2 Section View of AP600 Large-Scale PCS Test Phase Two Configuration . . . . . . . 2-10 Figure 3-1 AP600 Containment Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1
- Figure 3-2 Simplified AP600 Containment Compartments . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
- - Figure 3-3 Transient Mass Release Rate in AP600 During a DECLG LOCA . . . . . . . . . . . . 3-12 4
Figure 3-4 Transient Energy Relcase Rate in AP600 During a DECLG LOCA . . . . . . . . . . . 3-13
- Figure 3-S Transient Mass Release Rate in AP600 During an MSLB . . . . . . . . . . . . . . . . . . 3-14 Figure 3-6 Transient Energy Release Rate in AF500 During an MSLB . . . . . . . . . . . . . . . . . 3-15 Figure 3-7 PCS Delivered and Applied Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ; 3-17 i Figure 3-8 DECLG Containment Pressure vs. Time ..............................3-19 i Figure 3-9 Four Time Phases for DECLG Event , . . . . . . . . . . , , . . . . . . . . . . . . . . . . . . . . 3-21 i Figure 3-10 MSLB Containment Pressure vs. Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-26 o:\3692non.wpf:1b-101497 - Kevision 1 October 1997 r -4+,- . - - -,e-.r- -. -
-n -,,- . - - - ----m., - -
vi ' LIST OF ACRONYMS CMT - Core Makeup Tank DBA - Design Basis Accidents DECLG - Double-Ended Cold Leg Guillotine ECCS - Emergency Core Cooling System IRWS7 - In-Containment Refueling Water Storage Tank LOCA - Loss-of Coolant Accident LST - Large-Scale Test MSLB - Main Steamline Break PCCWST - Passive Containment Cooling Water Storage Tank PCS - Passive Containment Cooling System PIRT - Phenomena Identification and Ranking Table PRHR - Passive Residual Heat Removal PWR - Pressurized Water Reactor PXS - Passive Core Cooling System ' RCS - Reactor Coolant System STC - Science and Technology Center T&AP - Testing and Analysis Plan o \3692non.wpf:Ib-101497 Revision 1 October 1997
vil
SUMMARY
l The purpose of this report is to specify the limiting design basis accident (DBA) scenarios and l relevant passive containment cooling system (PCS) design parameters, identify the phenomena associated with the AP600 containment pressure response to a DBA, and document the basis (test, scaling, sensitivities, expert review) for phenomena ranking. This report also describes the l treatment of each phenomenon in the AP600 evaluation model (EM). The most limiting DBAs are the double-ended cold leg guillotine loss-of-coolant accident (DECLG-LOCA) and main steamline break (MSLB), due to the large mass and energy releases to the containment and the resultant containment pressure increase. The timing and progression of these two evenis provides different challenges to the containment: the LCCA is a relatively long transient compared to the MSLB. Characteristics of the mass and energy release such as superheat and momentum are also different for these two DBAs. The AP600 PCS was designed to maintain the AP600 containment below a pressure of 45 psig and reduce pressure over the long term. This function is accomplished primarily by absorption of energy by the volume and stmetures inside containment and by the evaporation of water that is applied via gravity directly to the outer containment shell surface. s The PCS consists of a 550,000-gallon water storage tank located above the containment shell, a set of weirs located on the containment shell to uniformly distribute the gravity-fed water from the storage tank onto the shell, and an air flow path to transfer energy from the shell to the environment. A systematic process involving results from separate effects and integral effects tests, containment scaling analyses, sensitivity studies, expert review, and engineering judgement was followed to identify and rank the respective containment cooling phenomena. Where useful, detailed phenomenological studies have been performed to address specific items (e.g., circulation and stratification, water coverage in Reference 1, Sections 9 and 7, respectively). This evaluation found the following: The mass and energy release is important since it is the source for containment pressurization Gas compliance limits the rate of pressure change Condensation mass transfer in the presence of noncondensibles inside containment is the most important phenomenon for pressure reduction inside containment Circulation / stratification and their effects on noncondensible distribution are important phenomena that affect the condensation mass transfer
SUMMARY
Revision 1 oA3692non.wpf:1b-101497 October 1997 ______-__a
vill Evaporation mass transfer from the containment shell is the most important phenomenon for energy transfer outside containment Natural circulation through the PCS air flow path is an'important phenomenon that affects the evaporation mass transfer Film coverage and stability on the external surface are important for heat removal by evaporation Convection and radiation heat transfer are much less important than condensation and evaporation For each PIRT phenomenon, a summary of the basis for the PIRT ranking and the treatment of the phenomenon in the EM is provided in Section 4.4, as follows: PIRT Ranking Basis for PIRT Ranking Test Results Scaling Results Sensitivity Studies Expert Review e How Phenomenon is Implemented in Evaluation Model Justification Of Evaluation Model Treatment Of Phenomenon Test Experience Modeling Guidance Sensitivity Studies Evaluation Model Treatment of Uncertainty, Distortions Appendix A provides a summary of the expert review process followed for evaluating the ranking of the containment DBA phenomena. The information documented in this report provides in a single location, the containment DBA approach to address phenomena models and uncertainties. Specific reference to the PCS scaling evaluations (Reference 2) which support the use of experimental data and a roadmap describing the methodology used to address the phenomena that significantly influence the containment DBA pressure transient are included. 4
SUMMARY
Revision 1 0:\3692non.wpf:lb-101497 October 1997
1-1
1.0 INTRODUCTION
1.1 OBJECTIVE The purpose of this report is to identify the phenomena associated with the AP600 containment pressure response to a design basis accident (DBA) and document the basis for phenomena ranking. The ranking is used to focus attention on the most significant phenomena and to identify aspects of the AP600 containment evaluation model which should be bounded or othenvise properly addressed to provide confidence in the predictions of the DBA. This report specifies the limiting design basis accident (DBA) scenarios and relevant passive containment cooling syst.em (PCS) design parameters, identifies the phenomena associated with the AP600 containment pressure response to a DBA, and documents the basis (test, scaling, sensitivities, expert review) for phenomena ranking. This report also describes the treatment of each phenomenon in the AP600 evaluation model (EM). This report supersedes Reference 3. The AP600 containmt.nt is a large, closed volume that undergo <.s pressurization during a postulated DBA. The containment shell serves as the boundary between the inside, pressurized region and the outside, atmospheric region. The passive containment cooling system (PCS) is used to transfer energy from the containment shell to the erwironment during a postulated DBA. In addition to the PCS which transfers energy from the containment to the environment, the AP600 containment has several other pressure mitigation features. These include a large volume, a large amount of steel and concrete heat sinks, and non-safety grade containment fan coolers, that absorb the energy from breaks in the primary or secondary side, and mitigate containment pressurization. This report is the starting point for the evaluation of the AP600 containment pressurization process and the PCS design. As shown in Figure 1-1, the Phenomena Identification and Ranking Table (PIRT) supports the other key containment evaluation activities and reports, since it identifies the containment phenomena and provides the basis for importance of the phenomena to containment pressure. This report focuses only on DBAs that may lead to containment over-pressurization. It excludes beyond-design-basis events, severe accidents, and under-pressure events. The limiting design basis accidents considered in this evaluation are described in subsection 3.3 and 3.4 and were selected as those that pose the greatest challenge to containment design pressure. INTRODUCrlON Revision 1 o:\3692non.wpf:1b-101497 October 1997
.- - . . . . . . = . - . _. . . .. -_- -_-
12-
)
l l l PIRT n EST WORTS
/
WGOTHIC WGOTHIC
^L N APPLICATION VALIDATION TEST A ALYSIS REP R REPORT REPORT REPORTS v
EVALUATION MODEL Figure 1-1 Relationship Between AP600 PCS PIRT, Testing, Analysis, and Evaluation Model INTRODUCTION Revision 1 o:\3692non.wpf:1b-101497 October 1997
13 1.2 REPORT ORGANIZATION The rest of the introduction provides a description of the AP600 PCS design and operation and a brief overview of the most significant phenomena, to introduce the reader to the physical arrangement and important processes. Section 2 describes the process of developing and confirming the PCS DBA PIRT, summarizes results from testing and scaling, and defines criteria for ranking. Section 3 specifies the events to be evaluated and shows success criteria. Section 4 contains the PIRT and a summary of the bases for ranking. Results of testing and test analyses, scaling analyses, sensitivity studies, and expert review have been included in the bases where applicable. Section 4 also includes a roadmap for each phenomenon that cpecifies how each phenoJnenon is addressed in the evaluation model for the containment pressure DBA calculation. Appendix A describes the expert review process and summarizes the results of the review, including supporting information for the PIRT ranku'gs.
1.3 DESCRIPTION
OF AP600 PCS DESIGN AND OPERATION The PCS makes use of the steel containment vessel and the concrete shield building surrounding the containment. The major components of the PCS are: the passive containment cooling water storage tank (PCCWST), that is incorporated into the shield building above the containment; an air baffle located between the steel containment vessel and the concrete shield building that defines the cooling air flowpath; air inlets and air exhaust, also incorporated into the shield building structure; and a water distribution system, mounted on the outside surface of the steel containment vessel, that functions to distribute water flow on the containment. The PCS arrangement is shown in Figure 1-2. A water recirculation path is provided to control the PCS storage tank water chemistry and to provide heating for freeze protection, as shown in Figure 1-3. The major flow areas and heights are also shown in Figure 1-3. The internal containment geometry is shown in Figure 3-2. Containmem systems and structures and spatial organization are described in subsection 3.2. Operation of the PCS is initiated upon receipt of two out of four Hi-2 containment pressure signals. System actuation consists of opening the PCS water storage tank isolation valves. This allows the PCCWST water to be delivered to the distribution bucket above the center of the containment dome. A weir-type water distribution system is used on the dome surface to maximize the wetted coverage of the dorr.e and vertical sides of the containment shell. A corrosion-resistant paint on the containment shell enhances surface wettability and film formation. INTRODUCTION Revision 1 o:\3692non.wpf:1b.101497 ( October 1997
1-4 l l 1 l l 1 l l PCS Water : i. M; e#L Storage Tank GElli gySS 4;nbi*$ W ;!"A-p[ggw.;t'+m]ffj$,f0IhN. 7gjy Air f g gy: I y:
- Air inlet h, ,y -
%l 7i inlet i
l l fi w pjaoe : il6%& Steel - ' h ~' dsd -. ?;
- Concrete Containment ,Jyl Shield Vessel (Ek. +
%q~idh JF%' GIE@!ggy)t,7*;q+
- py un
- ~ =q - Building 9
\ \ . l,h}b $?
Air Flow /' a Baffle @
^E Ea BMrsuQt:** - - -
wem#wdR, EUL m iRN '
; if$ $ igja g ~
1 ~% g r AP600 Ultimate - Heat Sink 7 ry i Figure 1-2 Passive Containment Cooling System Arrangement INTRODUCTION Revision 1 c:\M92nonwptib-101497 October 1997
15 l l i I!!
!'l
, h k
# h a
i=i v h m y l m gg L-
< 1 jfj'l{'
_ si
>-C-}$!
<9p 8 a sees
. 1 .
i I
\' -
r-- -1 , , ' * - i, E ~ EI1 M' ,II -- - -- - -- - - G
2) s -
l gj gjl L. _ . ,.J ",,,- n
. ..f
/
/ _
a . i n w k I+ -g . 0 -
~ ,
s oNN C C ' g :/ - w* b i,,,... -
.. : a.: u ...-::.:,.;.
. n
' ll n!(;
L 1 5 i I _ t' w g b y
=
I Figure 1-3 Schematic of Passive Containment Cooling System Showing Major Flow Areas and Heights INTRODUCTION Revision 1 o:\3692non.wptib.101497 October 1997
1-6 The PCCWST (550,000 gallon volume) provides water for containment wetting for an appropriate period of time following system actuation. The flow of water is initially established at approximately 440 gpm for short term containment cooling. The flow rate is , reduced over time to approximately 63 gpm at 72 hours as the decay heat decreases. The flow rate decreases as the water level in the storage tank decreases. A combination of different height standpipes in the storage tank is used along with orifices to adjust the delivered flow rate over time. The cooling water not evaporated from the vessel wall flows down to the bottom of the inner annulus into floor drains. Floor drains with 100 percent redundancy route excess water to storm drains. The drain lines are always open (without isolation valves) and each is sized to accept maximum external PCS water flow. The interface with the storm drain is an open connection such that any blockage in the storm drains would result in the ann'lus drains overflowing the connection, draining the annuin independently of the storm drain system. The path for the natural circulaMon of air upward along the outside surface of the containment shell is always open. The natural circulation air flow path begins at the shield building inlet where atmospheric air enters through openings in the concrete structure. Air flows past a set of fixed louvers and is forced to tum 90 degrees into the downcomer. After flowing down the downcomer, curved vanes aid in turning the flow upward 180 degrees into the riser. The riser and downcomer are separated by the thin-walled baffle. Air flows up the riser to the top of the containment vessel and exhausts through the shicld building chimney. The important dimensions (References 4,5, and 6) associateci with the PCS air flow path are: Containment inside diameter - 130 ft. Containment shel! thickness - 1.625 in. Riser gap - 12 in. (approximate) Baffle steel thickness - 0.12 in. Shield building inside diameter - 139 ft. Downcomer gap - 42 in. (approximate) 5hield building concrete thickness - 3 ft. The inlets for the PCS air flow path are placed circumferentially at the top outside of the shield building. Wind tunnel tests show that this provides a symmetrical t . let to a plenum in which pressure is circumferentially equalized, minimizing the effect of wind speed and direction, and limiting the effects of terrain and nearby building interference and turbulence. There are 15 air inlets, each 16-ft. wide by 3-ft. high. Screens are provided at the air flow inlets and discharges to prevent debris from entering the annulus. A special process trace heating system provides for heating of the a inlet and chimney structures that may be sensitive to snow or ice buildup that could cause blockage of the air flow path. The chimney th ough which the air / water vapor exhausts is located above the air inlet to provide additional buoyancy and to minimize the potential for exhaust air being drawm into the air inlet (i.e., recirculation). INTRODUCTION Revision 1 o:\3692non.wptib-101497 October 1997
17-The PCS system design is relatively simple with few active components, and all components i are accessible for inspection, maintenance, and repair. The PCS cooling water conditions and the valve positioning on the water storage tank are governed by the tecimical specifications and are monitored. The PCS water storage tank also has connections to the spent fuel system and the fire 3 protection system. The spent fuel system connection is for long tenn makeup to the spent fuel pool in full core offload accident scenarios. The fire protection system connection is to a segregated suction of the tank which provides an SSE protected water supply for fire protection. 1.4 OVERVIEW OF PHENOMENA As the containment atmosphere heats and pressurizes in response ta a postulated accident such as a high-energy line break, energy is removed from the containment atmosphere via several interrelated processes both inside and outside containment: c Circulation of the steam, water, and noncondensible mixture within the containment atmosphere Mass transfer via condensation on the initially " cool" heat sinks and inner shell surface inside containment Heat transfer via convection and radiation to initially " cool" heat sinks and inner shell surface inside containment Heat transfer via conduction through the steel containment shell Sensible heating of the relatively cool PCS water Mass transfer via evaporation on the outside shell Heat transfer via radiation and convection to the baffle, condensation or evaporation on the baffle, and radiation and convection to the shield building and environment Buoyancy-driven circulation in the external air flow path There are many parameters that can affect the above processes including the initial and boundary conditions, break size and location, thermal resisteces, and the actual performance of the PCS such as the amount of water coverage on the outside of the contakunent shell or the air flow rate over the shell. An overview of these energy transfer processes is prmided in subsection 4.1.
\
j INTRODUCTION Revision 1 o:\3692non.wptib-101497 October 1997
1-8 l The containment rate of pressure change is linked by mass and energy transfer resistances to ! the intemal heat sinks and through the containment shell to the PCS air flow path as shown i schematically in Figure 1-4. The energy and mass transfer resistances depend on values of the local velocity, air / steam concentration, and temperature which are govemed by the momentum inside containment and in the PCS air cooling path. The contauunent shell serves as both a heat sink and as a conductor of energy to the environment. As criergy is transferred from the shell to the air in the riser, the air becomes less dense than the air in the downcomer. This density difference causes an increase in the natural circulation of air flow through the downcomer, up the riser, and through the chimney to the exit at the top of the shield building. The relative importance of these processes changes as the transient proceeds, h transient nature of these phenomena is described in subsection 4.4, " Ranking of Phenomenn Listed in PIRT." INTRODUCTION Revisica i o:\3692non.wpf:Ib 101497 Octokr 1997
1-9
.. ..........,s.., '
..- *lnagy8 1,tane } -
- rmn ...
p
~ '.
', convectm ..
J i ' , A=4 em
... J .g i
.. ~.
i * -
. , . . . . . Ugwd . . .Rrn'i Delt Hast Rustd Pressure Mate s= a.m. %
.~
of Change '
; s
, .. Extemet i ,
- Doumtemer.
'. floor Crumney
) glemal Heat Smks g T,', . Pbe Pati a - e . concrete k -- - -
!! (
Steel. }: ', SteeLJacketed Concrete f{ # 4 Internal Extemal Momentum Momontum Figure 1-4 Generalized Schematic Representation of Containment Pressure Relationship to Heat Transfer through Momentum INTRODUCTION Revision 1 c:\3692non.wpf:1b-101497 October 1997
-21
- 2.0
- PROCESS FOR PHENOMENA IDENTIFICATION AND RANKING This section describes the process for identifying and ranking the phenomena in the AP600 containment for a DBA event. The process that was followed provided for structure, independence, and completeness; and was also iterative.
2.1 PROCESS DESCRIP' HON The process for identifying and ranking the importance of the phenomena in the AP600 containment following a postulated accident was initiated in the late 1980s. This process started witli the identification of accident scenarios that previously posed the greatest challenge to the containment and its respective cooling systems, i.e., large-break LOCA and MSLB. These transients have traditionally provided the largest mass and energy release to the containment (see subsection 3.3 and 3A for discussion of transients considered). However, it was recognized that the AP600 containment design differs from those in existing pressurized water reactors. Since the AP600 containment DBA relies on energy transfer to the emironment rather than to any active heat removal systems (containment fan coolers) it was necessary to extend the phenomena identification process outside the containment to include the environment. The identification of phenomena also helped to define what new tests needed tc be performed. The phenomena identification and ranking process involved the key steps showm in Figure 2-1. Tne phenomena that occurred during the most limiting scenarios were subsequently identified and documented by the Westinghouse engineers most familiar with the containment thermal hydraulic response. These personnel have extensive experience in analyzing fluid flow and heat transfer mechanisms, and evaluating the pressurization of nuclear power plant containment buildings. The identification and ranking of the containment phenomena was then subjected to test comparisons, scaling analyses, and sensitivity studies. These checks (described in subsections 2.2 and 2.3) verified that all containment phenomena were properly identified and ranked in importance. PROCESC FOR PHENOMENA IDENUFICATION AND RANKLNG Revision 1 c:\%92non.wpt:1b.101497 October 1997 l
22 Define scenarios and success criteria ir Review phenomena and existing tests 1r Document phenomena identification and ranking if . SETS and LST data analysis 1r S<:aling 1r AP600-specific quantitative uring phenomena reports Enp8'*'"' I 4 U Y v henomena lis No complete? Yes u Document PIRT and its bases
-6 Figure 2-1 PCS Phenomenon Identification and Ranking Confirmation Process PROCESS FOR PHENOMENA IDENTIFICATION AND RANKING Revidon 1 o:\3692non.wpf:1b-101497 Octobt r 1997
2-3 2.2 TESTTNG PROGRAh1S l The need for tests to support the AP600 containment design was identified in the late 1980s. At that time, containment phenomena with characteristics that were unique to AP600 were identified and an initial testing program was defined. The following provides an overview of the test program structured to address 10CFR52A7(b)(2)(i)(A), subsections (1) and (3). The design basis events calculated for the SSAR are used as a starting point to examine the types of data needed to support containment DBA. Each transient type is examined to see J how the uniqueness of the AP600 design imposes additional model verification requirer ents on the safety analysis code, EGOTHIC, and supporting models. The differences between the existing PWRs and the AP600 are also considered, since the basis for many of the safety analysis criteria and methods is rooted in the bases for the current generation of PWRe.. The phenomena were compared to the available containment test database (Table 2-1) to indicate the need for additional tests for validating analytical models. The important thermal hydraulic phenomena were identified, and the existing verification for the safety analysis codes was assessed against the current verification of the code, as well as the applicability of the data verification for the AP600 design. The assessment indicated which models required additional verification for the AP600 specific geometry or conditions. The assessment also gave an initial indication of which phenomena are of most importance for representing the passive features of the AP600 safety systems. Although many of the phenomena occur in current PWRs, there are no safety grade active containment cooling systems credited for the containment DBA in the AP600. Such active systems in current PWRs lead to somewhat different thermal hydraulic conditions in AP600, so that AP600 specific verification was needed. Test programs were established to address: evaporation and condensation mass transfer, including the effects of hydrogen, extemal air cooling of the steel shell, internal circulation and stratification, a external liquid film distribution (stability and coverage),
- effects of wind and turbulence, integral tests Scused on long term heat and mass transfer data (the LST).
Separate effects tests were performed at various facilities, as discussed below. Because the initial blowdown peMod for AP600 is not significantly different from that of current operating plants, integral effects tests, focused on the long term cooling for AP600, were identified to examine the integrated heat and mass transfer behavior of the PCS. Subsequent to the design of the PCS integral test facilities, scaling methods developed in the 1990s have been applied as described in NUREG/CR-5809. Scaling has been used to confirm the PIRT ranking (Reference 2, Section II) and to specify the applicable data from the PCS
~
PROCESS FOR PHENOMENA IDENTIFICATION AND RANKING Revision 1 o \3692nonwpfib-101497 October 1997
2-4 Large Scale Test (Reference 2, Section III) for separate effects correlation validation and b' GOTHIC code validation. ( The following paragraphs provide a brief overview of the test facilities developed for the AP600 program, and where applicable, supplemental test data available in the open literature. Summary conclusions from each of the Westinghouse sponsored tests is given in section 4.3.1 to support the formal process of identification and ranking of phenomena. Heat and Mass Transfer Mass transfer data in the literature are not based on surfaces with highly wettable coatings. Therefore, test fecilities were designed to provide evaporation and condensatic.n data for surfaces covered with the highly wettable inorganic-zine coating used for the AP600 containment shell for comparison to data from the literature. Evaporation separate effects tests were conducted at Westinghouse Science and Technology Center (STC) on the Heated Flat Plate test facility (Reference 7). Condensation tests were performed at the University of Wisconsin (Reference 8). It has been found that mass flux rate correlations available in the literature adequately represent data from the inorganic-zine coated surface, and that the primary effect of the coating is to improve film stability and coverage (see below), in both evaporating and condensing modes. Evaluations of AP600 containment heat and mass transfer phenomena were supplemented with test data available in the open literatce. These included the Hugat heated, parallel, vertical, isothermal plate tests; the Eckert and Diaguila heated vertical tub- test:,, the Siegel and Norris heated, parallel, vertical flat plate casts; and the Gilliland and Sherwood evaporation tests. These tests as described in Reference 9 provided additional data to validate models for convective heat and mass transfer in AF600. Extemal Downcomer/ Riser Flow Path Pressure Drop The integral Large Scale Test matrix included a representation o' the external riser air flow path with various magnitudes of resistance at the inlet. The Air Flow Path Pressure Drop test facility at STC (Reference 10), representing both the downcomer and riser, provided the basis for estimating the unrecoverable loss coefficient in the AP600 external flow path. Internal Circulation and Stratification Data available in the literature to address internal circulation and stratification included separate effects tests of circulation within cavities and integral effects in larger scale facilities (LST, CVTR, BFMC, HDR, NUPEC). The integral Large Scale Test (LST) facility included a representation of the AP600 internals. Although the LST intemals did not represent inter-compartment flow paths, data from the LST have been considered in addressing PROCESS FOR PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA3692non.wpf id-to1497 October 1997 1
25 stratification since the LST test matri < addressed a range of imposed boundary conditions representative of a significant portion o: the AP600 transient. Extemal Liquid Film Distribution - Stability and Coverage The literature available for film stability and water coverage is primarily focused on uncoated surfaces (for e' xample, polished steel or polished copper). The surfaces studied in the literature have rather high wetting angles (about 60 to 70 degrees), while the coated surface used in AP000 has a low effective average wetting angle near 10 degrees. AP600 tests used to determine liquid film stability included the effects of surface preparation and coating material. Data from heated and unheated surfaces and at various scales have been obtained from separate effects plate geometry, full scale unheated sector of the AP600 dome (with partial vertical wall), and the heated LST. The full scale sector Water Distribution Test included the effects of surface alig; ment irregularities and allowable tolerances. Effects of Winci and Turbulence The effectiveness of the AP600 wind-positive PCS flow path design has been tested in wind tunnels at several scales under the direction of the University of Western Ontario. Tests addressed prototypic plant sites to determine the effects of buildings and topology on the air flow into containment. PCS Integral Heat and Mass Transfer Tests The integral Small Scale Test (SST) was designed to examine extemal heat and mass transfer and water film behavior. The PCS Large Scale Test (IST) was designed to examine both extemal evaporation and internal condensation in an integral setting. A scaling evaluation of the LST has been performed te determine how well the LST characterizes the thermal hydraulic phenomena expected in the AP600 during a design basis transient. (Reference 2, Section III). The following sections provide a brief description of the test facilities sponsored by Westinghouse. PROCESS FOR PHENOMENA IDENTIFICATION AND RANKLNG Revision 1 0:\3692non.wpf:1b-101497 October 1997
2-6 Table 21 Containment Analysis Processes Used to initially Dehne Test Program AP600 Containment Uniqueness Validation AP600-Specific wrt H Does it vanidation Containment Process Plants Exist? Needed? Tests identified Evaporative film Yes No Yes PCS tests,1/8-scale tests, cooling heated plate terts Condensation, with No Yes Yes CVTR, U. of Wisconsin noncondensables Not AP600 l specific Air cooling of steel Yes No Yes LST to simulate air passage shell Internal circulation No Yes Yes LST to supplement the and stratification Not AP600 eusting test database (IIDR, patterns in specific BrMC, NUPEC, CVTR) containment Effect of hydrogen No Yes Yes LST to simulate containment on containment heat Not AP600 transfer specific Liquid film Yes No Yes Film flow experiments to distribution on investigate the water containment (film distribution-stability and
- heated pic.<e tests coverage)
- large-scale heated film flow tests (LST)
- full scale unheated water distribution test Effects of buildings Yes No Yes Wind tunnel tests with and wind velocity on building effects and site effects air flow over steel shell PROCESS FOR PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wptit 101497 October 1997
27 2.2.1 Heated Flat Plate Test The Heated Flat Plate Tests were performed to generate heat and mass transfer data for evaporative cooling with parameters that bound the range of expected conditions on the AP600 containment shell and in the riser (e.g., heat fluxes, ligeld film flow rates, cooling air velocities). A secondary purpose was to observe the film. hydrodynamics including possible - formation of dry patches due to surface tension instabilities. The test section was a vertical,6-ft. long,2-ft, wide,1-in. thick flat steel plate that was coated with the highly wettable inorganic-7inc coating used for the AP600 containment shell. A clear acrylic. cover provided a channel for the forced air flow a..d allowed obsenation of the applied liguld film. Preheated water was supplied at a metered rate to a simple distributor located at the upper end of the plate. To simulate heating of the containment wall, the test plate was heated from the back side using a high temperature fluid flowing through copper tubes welded into grooves in the back of the plate. Tests were performed in two orientations: vertical to represent the containment sidewall and - 15 degrees from horizontal to represent the upper portion of the dome. Tests were performed with no water on the plate and for a range of water film flow rates. Two of the tests were performed with very low film flow rates (as low as 15 lbm/hr ft) at relatively high 2 heat flux (as high as 6000 BTU /hr-ft ) to force the film to dry out before reaching the end of the test section. See Reference 7 for more information on the Heated Flat Plate Tests. 2.2.2 Wind Tunnel Tests The Wind Tunnel Tests were performed to test the aerodynamic response of air flow past the AP600 containment building and through the PCS air flow path under a variety of conditions. Three scale models (1:30,1:100, and 1:800) of the AP600 structures were used to simulate the shield building air inlet and exhaust, as well as the surrounding buildings and upwind terrain. Tests were performed to assess the response of the cooling air flow path to large extemal pressure fluctuations. See Reference 1, Section 6 for more information on analyses of the wind tunnel tests. 2.2.3 Condensation Tests A series of experiments to examine condensation vf air / steam mixtures flowing over cold surfaces were performed. The test section was 6.25-ft. long with a 2.75-ft. entrance length and a 3.5-ft. long condensing surface. The channel-cross section was square with a flow area of 0.25 ft.2. The 3.5-ft. condensing surface length was coated with an inorganic zinc paint used on the AP600 containment shell to promote surface wetting and condensation. The condensing surface was held at a near-constant temperature of approximately 86*F by cooling plates located on the back side, while the steam flow rate and inlet temperaturr were PROCE.55 FOR PHENOMENA IDENTIFICATION AND RANTING Revision 1 o u692non.wpf:1b-101497 October 1997 1
[ 28 - varied. The effects of a noncondensible gas (helium) and the orientation of the surface (for simulation of inner shell surface orientation angle) were also examined. See Refertace 11 for ' more information on these tests. 2.2.4 Air Flow Path Flow Resistance Tests The Air Flow Path Tests were performed to measure the hydraulic resistance in the PCS air flow path using a 1/6th4cale test (24-degree section). The test used a fan to force air through the flow path to characterize the pressure drop and flow resistance at approximately prototypic Reynolds numbers. The tests resulted in design changes to streamline the air flow pt t reduce the pressure loss coefficient. See Reference 10 for more information on these tew. 2.2.5 Water Distribution Tests The Water Film Formation Tests (Reference 12) were performed to show the wettability of the se.ected inorganic zine coating for the A?600 containment shell and to characterize general requirements for forming a water film over a large surface area. An unheated,8 ft. long,4 ft. wide steel plate, painted with the selected inorganic zinc coating, was placed on a pivoting frame to simulate the various angles on the containmem dome and sidewall. A stream of water was applied to the center top edge of the plate to see how it would spread to cover the surface. With a flow rate of 1 gpm from e 0.5-in. diameter tube pointed perpendicular to the surface, the water spread to form a 1 foot wide stripe of film down the S ft, length. These same results were obtained at plate angles of 90 and 11 degrees from horizontal. The film thi.kness was not uniform near the point of application;it was thinnest just below the application point and thicker on both sides. The film stripe continued to spread (more slowly as the surface became more vertical) and a very thin, wet region was created at the edges as the film traveled downward. Wtous film spreading mechanisms were also investigated. A dam and weir system was foed to be the most effective in distributing the water to create a wavy laminar film over the entire width of the plate. The Water Distribution Tests were used to deter 4e the water cwerage as a function of flow rate on the containment outside surfm _ to determine the time to establish steady-state coverage on the AP600. A full +cale tes ction representing a 1/8th-sector of the containment dome and a portion of the vert 3idewall was built, and the performance of various weir distribution systems was tes' Ihe tests were perfonned at ambient conditions and included flow rates of 55 - :20 gpm equivalent flow on the AP600 containment. See Reference 13 for more information on the Water Distribution Tests. PROCESS FOR PHENOMENA IDENTIFICATION AND RANKING Revision 1 0:\3692non.wpf:1t>101497 October 1997
29 2.2.6 Small Scale PCS Integral Tests The small scale tests were designed to provide heat and mass transfer data for both the inside and outside of the test vessel. The test apparatus consisted of a 3-ft diameter,24-ft. high, steel pressure vessel that was intemally heated by steam. The vessel was surrounded by a clear, plexiglass shleM that formed a 15 in. wide annulus for either forced or natural circulation air flow. The tests were performed with varying steam flow rates, water film flow rates and temperatures, and inlet air flow rates, temperatures, and humidity. Instrumentation was provided to measure internal steam concentrations, extemal water evaporation rates, exit film temperatures, air velocity and temperature, and humidity. See Reference 14 for more information on small-scale tests. 2.2.7 Large-Scale PCS Integral Tests The large-scale PCS test (LST) facility was built to provide long term heat and mass transfer test data for a geometrically similar model of the AP600 containment vessel. The tests provided experimental data for evaluating phenc,mena inside contairunent, and for determining the relative importance of various parameters that affect heat and mass transfer on both the inside and outside containment surfaces. The LST consisted of a 15 ft. diameter,20 ft high pressure vessel
- hat approximated the AP600 containment vessel at approximately 1/8th linear scale. A plexiglass cylinder was installed around the vessel to form the air cooling annulus (also called the nser in this report). Air flows upward through the annulus via natural circulation to cool the vessel. A fan was located at the top of the annular shell to provide the capability to induce higher air velocities than can be achieved during natural circulation alone, so that riser Reynolds numbers in the range of AP600 could be simulated. A liquid film was applied to the outside of the test vessel to provide evaporative cooling. Two rings of Ftubes provided the capability to apply water in a manner similar to the water coverage observed in the water distribution tests. See Figure 2 2 for an overview of this test facility.
Test coaJitions were selected to provide steady-state heat and mass transfer data over a range of conditions representative of a DBA. These conditions included pressure, steam flowrate, cooling air flowrate, and water coverage. The LST was designed to sufficiently encompass the conditions expected during long term cooling for the most limiting AP600 transients such as a large cold leg LOCA and MSLB. The LST did not simulate the blowdown phase of the LOCA and MSLB transients. Rather, tests were performed over a range of the initial and boundary conditions to assess the impact on heat and mass transfer rates. PROCESS FOR PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:u692nortwpf:1b-101497 October 1997
2 10
- [xhaust ran Outtet T/C's Located by - 4 * -Detto P Cett Equal Circuaterentist Areas
-rised Velocity Meter Vater bstributch System F.e Dffuser
\
Portable Anemometre g taffte Traperatures Internet Velocity Heters Heat rius Meters til 1"Si ffff'fff f f Insee/Dutside Volt Plesivass Baffle Temperatwes Traversing T/C T Portable Anemometer
- s' for Traversing g .
,/--itund Tem,>eratures J L/
\ Traversing T/C's
.,eam me,a,or -et
, Transparent Baf fle Steam Release Partable Anenometer m Flos Area for Traversing g Loop Compartment ,
/
Dperating Dec% V y A S rimed Inlet T/C
'Open' Volume openg Gutter s - *=.
, ve,e-te . -
Steam Injection Line steam rios Meter Helsum lajection Figure 2 2 Section View of AP600 Large-Scale PCS Test Phase Two Configuration PROCESS FOR PHENOMENA IDENTIFICATION AND RANKING Revision 1 0:\3692nonwpf.lb 101497 October 1997
f 2 11 For most tests, steam was injected through a diffuser located under a simulated steam generator compartment below the operating deck (which simulated a LOCA). The steam rose as a plume, and air was entrained in the rising plume resulting in a natural circulation flow pattern within the simulated containment. Thermocouples located on the inner and outer surfaces of the vessel were used to determine the temperature and heat flux distributions. Tests were also performed with an elevated steam source to simulate an htSLI3, with parametric variations to examine the effect of source direction and momentum. See References 15 and 16 for more information on these tests. 2.3 SCALING ANALYSES The scaling analysis results have been used to support quantification of the importance of various phenomena in the containment cooling process (Reference 17). The scaling analysis performed for the AP600 containment was submitted for review and revised to incorporate NRC comments (References 2,18). In Reference 2. Section it control volume equations were developed to describe the rate of change of the containment gas energy and pressure. These equations were coupled by conductances to energy equations for internal heat sinks and to the external PCS through the shell. Scaling groups (PI groups) were developed by normalizing and nondimensionalizing the conser vation equations, using initial and boundary conditions, in a form that shows the important dimensionless pa;ameters in each group. Values were calculated for the P1 groups during each time phase to quantify the relative importance of the transport processes and components. The evaluation of the PI groups assumed that the containment steam / air atmosphere was well mixed. Nondin.ensional parameters and relevant test data were defined for assessing stratification and internal flow field stability. The P1 groups were evaluated for containment energy and pressurization, conductances to heat sinks and the shell, momentum in the air flow path, and momentum within the containment. The conclusions from the scaling analysis, which support the importance of the phenomena identified in the PIRT, are discussed in subsection 4.3.2. In Reference 2, Sec' ion 111, top-down scaling is used to determine the most important system level phenomena during blowdown and long term phases of a Large LOCA transient and to show how well that phenoraena are preserved behveen the LST and AP600 plant. The results of this analysis are used to determine to what extent global containment data (i.e. containment pressure) can be used fmm the Large Scale Test (LST) for WGothic code validation. PROCESS FOR PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA3692nonwgtt loi497 October 1997 l
.. _ l
2 12 ' 2.4 RANKING OF PHENOMENA The purpose of ranking the phenomena is to identify the important phenomena that are to be ; bounded or othenvise properly addressed in the containment evaluation model. The criteria ' for ranking the phenomena for the AP600 containment cooling process was based upon a combination of test results, scaling analyses, sensitivity studies, expert review, and engineering judgment. The application of these items in the confirmation of the PIRT rankings is discussed in subsection 4.4. The final PIRT rankings are provided in Table 41. s I i I L i PROCESS FOR PHENOMENA IDEN11FICATION AND RANKING Revision 1 as3692non. wet.ib.101497 October 1997
31 3.0 ACCIDENT SPECIFICATION The accident specification for the AP600 containment cooling process consists of a statement of the issue and success criteria, description of the contahunent systems and structures, identification of the transients considered, and description of the event scenarios. 3.1 ISSUE 'AND SUCCESS CRITERIA ~ Most commercial nuclear reactor designs include containments to limit the release of radionuclides to the environment during a postulated breach in either the primary reactor coolant system (RCS), or the portions of the secondary cooling system inside containment. 1he containment is a pressure vessel designed to the ASME Boiler and Pressure Vessel Code requirements for a specific design pressure. Vessel penetrations for entryways, piping, and instrumentation are scaled to limit leakage of the vessel atmosphere to the environment. A group of design basis accidents (DBAs), known as high-energy line breaks, has the potential to release significant quantities of high temperature, high-pressure steam / water inside containment, and may increase the internal pressure to values that challenge the design pressure. Both the primary and secondary coolant sy:tems have large values of stored energy as a consequence of the large volume, high temperature, and high pressure of their steam / water coolant and the heat capacity and high temperature of the cooling system boundaries that include the reactor vessel, steam generators, turbines, pumps, and piping. 3.1.1 Design Criteria The AP600 PCS has been designed to maintain pressure below the containment vessel design pressure during a DDA with no credit for active containment heat removal systems and witn no operator action. The only active part of the system is a one time automatic valve opening to initiate PCS cooling water flow, based on a safety grade containment over-pressure signal. For the PCS DBA, the PCS design will be judged to be successful if, for any postulated DBA, the PCS can: Maintain the peak pressure difference across the containment shell below 45 psig Reduce the pressure difference at 24 hours to less than half of the design value Provide containment heat removel for a sufficient period with no operator actions Cantainment design criteria, contained in General Design Criteria for water-cooled nuclear power plants (10 CFR Part 50, Appendix A and 10 CFR 52.47), are addressed in the SSAR (Reference 19). The following paragraphs provide a summary of the applicable regulations. Criteria addressed by the PCS DBA PIRT: The containment structure should be able to accommodate the calculated pressure and temperature conditions resulting from any 1.OCA. This is to be accomplished without ACCIDENT SPECIFICATION Revision 1 oc\3692non.wpf:1t>101497 October 1997
3-2 exceeding a design leakage rate and with sufficient margin. The margin should reflect consideration of (1) potential energy sources such as energy in steam generators, l limited metal water reaction that might result from degradation but not failure of the ECCS, (2) limitations on the emount of information available on accident phenomena, and (3) conservatism in the calculations. (Criterion 50)
- Documentation in an application for design certification should include, evidence that the performance of each safety feature of the design has been demonstrated through either analysis, appropriate test programs, experience, or a combination thereof (10CFR52.47(b)(2)(1)(A)(1))
evidence that sufficient data exist on the safety features of the design to assess the analytical tools used for safety analyses over a sufficient range of normal operating conditions, transient conditions, and specified accident sequences (10CFR52.47(b)(2)(i)(A)(3)) Criteria addressed elsewlere in the SSAR: Containment will establish an essentially leak tight barrier against the uncontrolled release of radioactive material (Criterion 16) Systems are required to be available to remove heat from the containment to negate pressure buildup that would otherwise result (Criteria 38 through 40) A system is required to be provided to remove fission products from the containment s atmosphere to reduce the consequences of ongoing leakage (Criteria 41 through 43) An evaluation of the AP600 against the Standard Review Plan is provided in Reference 20 (10CFR50.34(g)) Criteria for severe accidents addressed in the Probabilistic Risk Assessment: Severe accidents, detined by 10 CFR 50.34(f) for near-tenn operating licenses, 10 CFR 52.47 for standard design certification, and 10 CFR 50.44 for combustible gas control are addressed in the Probabilistic Risk Assessment, and are excluded from the PCS DBA PIRT. 3.2 CONTAINMENT SYSTEMS AND STRUCTURES Temporal and spatial partitioning were used to organize the PIRT. Subsection 3.2.1 describes the spatial partitioning. Subsections 3.4.2.2 and 3.4.3.2 describe the temporal partitioning, which is event-specific. ACCIDENT SPECIFICATION Revision 1 au6nnonnt.18101497 October 1997 i,
33 The inside of containment is a large, closed volume (1.7x10 6 ft3) which undergoes ! l pressurization during the accident. The containment shell serves as the boundary between i the pressurized region inside, and the ambient region outside. Therefore, the containment i cooling system components were segregated into three global regions: intemal, external, and shell. In order to more conveniently present the components in the PIRT, spatial partitioning of volumes inside and outside containment was used. Each global region was segregated into the major volumes shown in Figure 31. Phenomena evaluation address specific components where appropriate. For further convenience, the initial and boundary conditions were considered as " components" that can affect the containment pressure response. Thus, l the PIRT can be used to assess the relative importance of factors that affect the AP600 i pressure transient as well as the calculation of the DBA pressure response. 3 l t i l lE l i i i 1 i ACCIDENT SPECIFICATION Revision 1 o:\3692nortwpf.lb 101497 October 1997
l 3-4 i ST N MY l
/
/
i d s
\
1 air ir.let
/
/
iWh /
^
annulus annulus environ-riser / downcomer rnent
/
/
/
/
/
/ y' /
wi e.
/%#
- / _l m
< shell
- - . Nshield building IRWST& W baffle = = i --- \ - -
Break Pool Figure 31 AP600 Contahment Structures ACCIDENT SPECIFICATION Revision 1 c:\3692non wpttb 101497 October 1997 1 i
35 l 3.2.1 Inside Containment t The intemal containment includes the following key components:
- Break source
- Containment volume ,
;
- Containment solid heat sinks !
- Initial conditions j
- Break pool ,
j
- In-containment refueling water storage tank (IRWST) a-l The break source, which acts as the internal boundary condition, includes the steam, water, j and/or water drops depending on the mass and energy release rates. The water drops i
suspended in the steam inidally flash a small fraction of their mass to steam to reach thermal i equilibrium within the containment atmosphere. After flashing, the large surface area of i these many tiny vrater drops maintains the atmosphere at or near saturation for up to l thousands of seconds, l The inner containment atmosphere includes the mixture of steam, water, and air contained 6 3 whhin approximately 1.7x10 ft of volume. The subcompartments below deck are large open volumes with relatively larSe interconnections that promote ndxing throughout the i below-deck volume. All compartments bclow-deck are provided with top openings to , minimize the potential for a dead pocket of noncondensible concentration. \ The distribution of gas volumes and intemal heat sinks corresponding to each intemal
- compartment are listed in Table 3-1. Figure 3-2 shows a cross-section of the containment i with typical compartments. Additional information on the containment compartments can be l obtained from Reference 1, Section 4.
The containment solid heat sinks include the steel hardware and concrete structures within
- the containment. The containment shell is listed separately from the internal heat sinks since the shell has a Sigmficantly different boundary condition due to the extemal evaporating
- film.
4 The initial conditions include the temperature, humidity, and pressure of the containment volume as well as the temperature of the solid and liquid heat sinks inside containment. Liquid from the break that is not dispersed as drops with the steam accumulates in the
- bot +om of the steam generator and reactor cavities to form the break pool. The break liquid
- is assumed to leave the break at the containment saturation pressure. Liquid from drop j fallout and condensation on intemal structures below the operating deck also drains into the ; break pool.
i ACCIDENT SPECIFICATION Revision 1 on3692non.wyt:1b 101497 October 1997
---_1,y.--.,---s y+ ..79, ,,%_. _
..,_,,T-wer-=-r-----=-v-'=-'* TC'-- **y v"-'vm'-*t'ry'vW- r-T-ww w tv-m*'w--t=--Mwwy-w--wrw w pw w-~ yy-"+vy y- r-
. - _ _ _ _ _ - . _ - - _ ~ _ _ - - _ _ _ ___ __ _ . _ . _ _ _ _
3-6 The IRWST collects the condensate that fonns on the shell above the operating deck via gutters. After primary system depressurization, the IRWST provides a gravity flow of borated water into the reactor. Widle condensate flows to the IRWST, the water may heat but only to the temperature corresponding to the steam partial pressure of the atmosphere at the operating deck, Consequently, it can not become a vapor source by heating from the atmosphere, either while above deck or after draining into the tank (IRWST water is assumed to be at an initial temperature of 120*F). 3.2.2 Containment Shell The shell, which is 1.625 in. thick steel, is an important component because it stores energy and provides the path to transfer energy to the ultimate heat sink-the environment. An inorganic zine coating is applied to both the inside and outside containment shell surfaces to promote wettability (and therefore, more efficient heat removal) and to provide corrosion resistance. The PCS water is applied directly to the top outside surface of the containment shell. The water is distributed across the surface by means of two sets of welts. The water absorbs energy from the shell, heats up, and evaporates into the air flowing up the annulus riser and out the chimney. 3.2.3 Outside Containment The components outside the containment make up the PCS air flow path, where the evaporative, radiative, and convective energy transport processes occur that transfer the thermal energy from containtnent to the environment. The physical components of the PCS air flow path are the thin-walled steel baffle and diffuser, U-shaped baffle supports, chimney structure, and thick walled concrete shield building. These outside components define the downcomer annulus, riser annulus, and the chimney volumes that make up the PCS air flow path. The PCS cooling air flows from the environment, through the inlet screen and downcomer, up the riser, through the diffuser, and out the chimney to the environment. l l ACCIDENT SPECIFICA'llON Revision 1 o:\3692ncn*T 01b-101497 October 1997
3-7 2% (h 1W SG REFUELING ~ ~ CANAL \ / 140' [" i _ '
" CMT CMT IRWST 120' v
~"
]L J L. -
cvs Ace f - gg 1 RPV [ STAIRWELL M 4 REACTOR CAVITY J W Figure 3-2 Simplified AP600 Containment Compartments ACCIDENT SI'ECIFICATION Revision 1 o:\3692non.wpf;1b-101497 October 1997 {
9 y> n
-i
- g. Y m
$O s' kO p
w Table 3-1 PCS Compartment Gas and Heat Sink Volumes and Areas Considered in Evaluation Model A 4g 1 C u) Gas Concrete Jacketed Concrete - Steel 5m Control Eb
$C Volume Volume Area Area Area Area Volume Area 0 2! F to 3 (ft ; Vol Fract (ft )2 Fract (ft2) Fract (ft') Vol Fract (ft2) Area Fract h Above Deck 1393900 0.81 8200 0.42 8200 0.13 1600 0.13 71FJO 0.33 h $h (minus sh II) 55 Above Deck - - -
0.00 - 0.00 7300 0.58 53600 0.25 hk
-n ~
(shell) IRWST 11000 0.01 - 0.000 11100 0.18 148 0.01 7100 0.03 O E Circulating g Compartments a SG East 29200 0.02 0.00 4500 0.07 110 0.009 6200 0.03
$ SG West 79100 0.02 -
0 03 4900 0.08 120 0.01 6200 0.03 E CMT 157200 0.09 8400 0.43 13500 022 2200 0.17 47700 0.22 Ln, Refueling 44300 0.03 - 0.00 7600 0.12 78 0.006 3300 0.02
% Stairwell 16000 0.01 -
G.00 2200 0.03 29 0.002 2800 0.01 a e Dead-Ended l Compartments Accum NE 13300 0.008 0.00 3600 0.06 140 0.01 5000 0.02
$ Accum SE 10200 0.006 -
0.00 2800 0.04 140 0.01 3200 0.01 CVCS 15700 0.010 2800 0.15 700 0.01 560 0.04 6600 0.03 g Reactor 5300 0.0tC - 0.00 3600 0.06 17'1 0.01 1000 0.005 g Cavity 9 TOTAL 1730600 1.00 19400 1.00 62700 1.00 12595 1.00 213600 1.00 a y Note- These values are consistent with Reference 1. The scaling analysis model described in Reference 2 does not 9 3 ecteunt for a number of heat sinks listed in this table. The rationale for the heat sinks used in the scaling gx I analysis is provided in Reference 2. XR
, a 3
5' 8U
~~
i I
i 3-9 As shown in Figure 3-1, the outside contaltunent hardware is broken down into the following key components:
- Riser annulus and chimney volume
- Baffle
- Baffle supports
- Chimney structure
- Downcomer annulus
- Shield building
- External atmosphere
- Initial conditions of structures (grouped under " Ins!de Containment")
The riser annulus (approximately 12 in wide) is formed by the baffle and the outer surface of the containment shell. The chimney volume is located at the top of the shield building where the air and vapor mixture exit the riser annulus. The baffle is the thin walled steel plate that divides the annulus between the containment shell and the shield building into the riser and downcomer. The baffle supports are U-shaped brackets that position the baffle at approximately 12 in, from the containment shell. The chimney structure consists of the concrete and steel structures at the top of the shield building. The downcomer annulus (approxim/ :ly 3.5 ft, wide) is formed by the inner surface of the shield building and the baffle. The shield building is the 3-ft. thick concrete structure that surrounds the steel contam' ment shell. The external atmosphere acts as the AP600 plant ultimate heat sink, i.e., the external boundary condition. The atmospheric conditions which may affect the containment energy transfer process are the temperature, humidity, and wind conditions. The PCS air flow path interacts with the external adnosphere only at the inlet and outlet since the shield building concrete is so thick (3 ft.) that any thermal interaction through the concrete with the emironment is insignificant.
3.3 DESCRIPTION
OF TRANSIENTS The results for the two limiting containment overpressure DBAs provided in the SSAR (Reference 19) have been used for phenomena evaluations. The transients are selected for use in identifying the phenomena which should be considered in PCS design basis analyses. Minor variations in transient progression were not expected to lead to additional DBA
' ACCIDENT SPECIFICATION Revision 1 oA3692non.wphitr101497 October 1997
3 10 phenomena. The peak containment pressure, the containment pressure at 24 hours (LOCA only), and the peak containment temperature for the two limiting events are provided in Table 3 2. Table 52 Comparison of Key Containment Analysis Results Peak Containment Containment Peak Pressure Pressure at 24 hours Containment Transient ipsig) (psig) Temperature PF) Double-er.ded cold leg guillntine 44.0 18,9 280.3 MSLB at 30% power 44.8 N/A 353.6
'Ihe results of these limiting transients are sufficiently typical of AP600 PCS performance to allow identification and ranking of the physics of the containment cooling process, at an appropriate level of detail for the PIRT.
Based upon peak containment pressure, the two most limiting transients are: the DECLG LOCA, and an MSLB at 30-percent power with delayed main steamline isolation valve closure. The LOCA cold leg break analysis includes the long term contribution to containment pressure from stored energy sources, such as steam generators. These two transients, which are described in subsections 3.4.2 and 3.4.3, provided the highest energy release and contairunent pressure. 3.4 EVENT SCENARIO This section provides a description of the two limiting transients, ar.d includes the initial conditions, boundary conditions, and key assumptions in the modelJng. 3.4.1 Initial and Boundary Conditions Both of the limiting transient events (DECLG-LOCA and MSLB) are assumed to start from the same initial containment conditions. The initial conditions assumed for the safety analysis are consistent with the Technical Specification limits, and are presented below in Table 3-3. Additional discussion of the effects of these values is presented in subsections 4.4.4 and 4.4.15. ACCIDENT SPECIFICATION Reva$ i o:\3692non.wpf;1b 101497 October ie ! l l
3 11 l 1 l Table 3 3 liittial Conditions for AP600 Containment Pressure Calculations Parameter Reference Values Environmental temperature 115'F Environmental pressure 14.7 psh Environmental humidity 22%
- Shield building, baffle, and chimney temperature 115'F PCS Cooling water temperature 120'F Containment air temperature 120'F Containment air pressure 15.7 psia Conte.inment air humidity 0%
Shell and heat sink temperature 120'F
'llased on 80*F wet bulb temperature Three transien' boundary conditions are provided for each transient,1) break mass release rates of liquid and vapor,2) break energy release rates of liquid and vapor, and 3) PCS cooling water flow rate to the shell. The mass and energy release rates for the LOCA are presented, respectively, in Figures 3-3 and 3-4. The mass and energy release rates for the MSLB are presented, respectively, in Figures 3-5 and 3-6.
The following mass and energy sources are accounted for in the long-term LOCA mass and energy release calculation:
- Core power, temperature, and pressure increased to account for unceitainty and instrument dead band
- Decay heat (1979 ANS plus 2 sigma)
- Cole stored energy (+ 15 percent)
- RCS fluid and metal energy (+ 3 percent RCS liquid volume)
- Steam generator fluid and metal energy (+ 10 percent fluid mass)
- Accumulators, core make-up tanks (CMTs), and IRWST
- Zirconium water reaction (conservatively considers energy due to 1 percent of fuel cladding)
The ener5y release rates are calculated in the mass and energy model so that the energy is released quickly which results in a conservative containment pressure calculation. Other characteristics of the LOCA and MSLB are discussed in subsections 3.4.2 and 3.4.3 respectively. ~ ACCIDENT SPECIFICATION Revision 1 on92nen.wptn*101497 October 1997 f
3 12 . 100000: :10000000
~
Uquid Ma 10000 Uguid Flow g1000000 ? 5 g Gas Flowy\ Gas Mass . l
- , g S
E 1000! 5100000 f
~ ~
h 5 5 1005
/
] .
10000 10 ...iii.. . . . i ii . . .... . ... ..l 1000 1 10 100 1000 10000 100000 Time after Break (seconds) Figure 3-3 Transient Mass Release Rate in AP600 During a DECLG LOCA CCIDENT SPECIFICATION Revision 1 o:\%92nonspf.1b 101497 Octob-r 1997
3 13 ! . 100: :100000 10 Gas Rate 10000 f . 4- - 5 1: Wud Gas EnewA s1000 g
- Rate 3
s , @
= : p 0.1 i
Wud Erww 100 9
- : tD b f ! 31 Q - -
m j 0.01; ;10 f-1 1 0.001 . ,...o 1 1 10 100 1000 10000 100000 Time after Break (seconds) Figure 3-4 Transient Energy Release Rate in AP600 During a DECLG LOCA ACCIDENT SPECIFICATION Revision 1 c:'.3692nortwpf.th101497 October 1997
3 14 l l l l l l
...............ee.. ............e.................................... '
M
/
. .......................... .............................. r........ .
s.* , l 1 000. ...~..........
~. .". . ~..... ~. ~. ~. . . ~..~. ~. ~..~.
- e. . .d.*/j.d. . . . . . . . >
. . ~ . ......
~ . . ~ . ~ . ....
~ . ~ . ~ . ~ . . ~ . . ~ . ~ . ~ . ~ .-1
................................y...................................
.g g .
......................,./................. .......................... .
...................s....................... .......................
/ 9
. ................/................................. .. ..............
. ...........c.............................................. .......... -
.t 100 . . . ..- . . . . . . . . ... . 10000 1 10 100 1000 Time after Break (seconds)
Figure 3 5 Transient Mass Release Rate in AP600 During an MSLB ACCIDENT SPECIFICATION Revision 1 o:\369hmwpf:1b 101497 October 1997
l 3 15 1 00
.. . . . . . . . . . . 1. .0000
. . . . . .=. . . . . . . . . . . . . . . . . . . .
g l _. 10.,..................................................................
..... ... ................................................................. 1M
........ ......................................................... . u,
............... m .......
- ................................. /.... ..,.
............................ ....................... y .............
th Q A .'.... > 4..
..............................,..c.s....,
.............................t..
1%
........................... 3g
....................f1 .............. ...... ...................... . Q
. ................. s#............ ............... ...................
. ....... y'........s.......................................... .......
/
'/
0.1 . i... . . . ii.i. . . . \ .. 10 1 10 100 1000 Time after Break (seconds) i Figure 3-6 Transient Energy Release Rate in AP600 During an MSLB ACCIDENT SPECIFICATION Revision 1 c:\3692non.wpfh.101497 October 1997
3-16 The third transient boundary condition is the PCS cooling water flow rate to the shell and a typical flow profile is shown in Figure 3 7. It is calculated assuming a single failure of one of two valves (located in parallel) to open. The delay in the flow to the containment shell is attributed to the time for filling the pipe, the distribution bucket, and the first weir. The largest flow resistance is in orifices in the standpipe so that the single valve failure assumption affects the gravity-driven flow by less than 2 percent. The reductions in the flow rate at about 9,000 seconds and 80,000 seconds occur when the level falls below the first and second standpipes, respectively. More information on PCS water flow is given in Reference 2, Section 7.6.6 and Reference 1, Section 7. The enviroriment is also a boundary condition and may change over the course of the transient. The effect of environment induced disturbances is addressed in subsection 4.4.15. It is assumed in the safety analysis that the environmental conditions are constant at their initial (conservative) values. 3.4.2 Lc.ss-of-Coolant Accident 3.4.2.1 Description of LOCA With the reactor core at full power, a high-energy primary coolant line is postulated to break, releasing a combination of steam and water to the containment. It is assumed that the nonsafety grade containment fan coolers do not c,perate. The break is a DECLG rupture of the RCS piping in a steam generator compartment. The 22 in. inside diameter cold leg pipe is the second largest diameter pipe in the primary system, but produces a higher second pressure peak than a hot leg (31 in. inside diameter) break. The flow resistance from the reacty to the break is less for the hot leg break, so less of the reflood coolant goes through the steam generator. Consequently, for a hot leg break, the steam generator stored energy is released more slowly, over a period of hours, by convective heat transfer to the containment atmosphere. For a cold leg break, more of the steam generator stored energy is transferred eailier by means of passive reactor cooling system water, which leads to a higher peak containment pressure. The DECLG blowdown releases approximately 6,000 ft.3 of water into containment. Fellowing initial reactor system depressurization, the accumulators deliver their remaining inventory of wate into the primary system followed by the CMT delivering water to the primary system. 'ihe IRWST then provides a gravity flow of water into the core from its initial 70,850 ft.3 inventory. Approximately 40,000 ft.3 from these water sources will fill the reactor cavity and lower portions of the steam generator compartments and flood the reactor hot and cold leg piping elevations, effectively flooding the core. ACCIDENT SPECIFICATION Revision 1 l o;\3692norwpf lb-101497 October 1997
3-17 80 64 - -- -- -- -- f~~ m
~
l 48 - -~ W --- -- -- -- - Outnow f 32 - Second Ww Outnow . _ I 16 --
-4 -
l 1 0 ' 1E0 1E1 1E2 1E3 1E4 1ES 1E6 Time (seconds) Figure 3-7 PCS Delivered and Applied Flow ACCIDENT SPECIFICATION Revision 1 o:\3692non.wpf;1b-101497 October 1997
3-18 Water from the break that is not entrained into the atmosphere as drops drains into the reactor cavity and, as the break flow continues, the level rises to the bottom of the steam generator compartment. At approximately 15,000 seconds the break flow level rises to the level of the floor of the CMT room, the maximum flooding height. Continued steam generation frorr, the reactor into the atmosphere condenses on the extemally cooled containment shell, and the condensate flows back into the IRWST through a system of gutters. The operating deck is sloped to return all other above-deck condensate and rain-out to the IRWST. The continued release of both core decay heat and heat from structures with long thermal time constants, such as the reactor vessel, combined with water from the passive core cooling system (PXS), produces steam that causes the containment pressure to continue to increase following the blowdown, although at a slower rate. Initially, the steel and concrete heat sinks inside containment remove significant quantitles of mass and energy from the containment atmosphere, but these reach their maximum effectiveness after several minutes. However, within a few minutes after the initiation of the accidant, the PCS external cooling water and air flow become fully effective in removing energy from containment, and thereby limit the peak containment pressure. The AP600 blowdown containment response to a LOCA is similar to that of existing two loop plant designs. The blowdown phase of a large break lasts on the order of 30 seconds, after which the energy release rate remains below approximately 1 percent of the peak blowdown energy release rate. The release rates are determine:! by the size of the pipe break, with the larger the pipe, the more rapid the blowdown. A range of LOCA break sizes have been examined to select the limiting case of a DECLG. Following the blowdown, the containment pressure drops as the steam flow rate rapidly decreases and the reactor lowa plenum refills, while the containment shell and internal heat sinks absorb some of the energy released during the blowdown. As the internal heat sinks saturate, the continued lower release rate during the peak pressure phase causes the containment pressure to increase to a second peak until the source energy release rate reduces to values below the capacity of the heat removal systems and the containment begins a long term depressurization. A representative pressure history for the AP600 DECLG LOCA is shown in Figure 3-8. i l ACCIDENT SPECIFICATION Revision 1 o:\3692nortwpt:1b.101497 October 1997 (
1 I 3 19 I D. &afuced Subtoo%e /
.1 Pt*S WaterflowChenen i i t i I 10 100 luuv 1M04 1H06 D(N Figure 3-8 DF.CLG Containment Pressure vs. Time ACCIDENT SPECIFICATION Revision 1 a\3692nortwpElb.101497 October 1997
3 30 3.4.2.2 Temporal Partitioning (LOCA) Temporal partitioning is used to help evaluate the transient ey recognizing that the most important processes early in time may not remain important for all time. By separating the transient into time phases, the number of phenomena that are important at any one time phase is reduced, thereby reducing the complexity of each time phase. The DECLG LOCA transient is partitioned into the following four phases as showm in Figure 3-9:
* " Blowdown," which lasts approximately 30 seconds
* " Refill," which lasts from 30 to 90 seconds
* " Peak pressure," which lasts from 90 seconds to the time of pressure peak (about 1200 seconds) -
* "Long term depressurization," which lasts from the pressure peak and beyond The time phases are chosen to be useful for explaining the physics of the containment pressure transient and for scaling the various pressure phases. A useful partitioning for scaling is to choose inflection points to segregate the pressure curve. Such an approach is used for PCS analysis, as shown in Figure 3-9. The naming convention is to use the most prominent aspect occurring during each phase as the descriptor. Where the mass and energy release drives the shape of the curve, the phenomenon driving the mass and energy release is used. The mass and energy blowdown drives the initial pressure rise, so the first phase is named " blowdown." During the refill phase, while the reactor lower plenum is being filled, there are no releases, leading to a brief period of depressurization until releases begin again.
The refill period lasts about 60 seconds until the inflection point is reached, and is simply called " refill." During the third phase, after the refill period, the containment repressurizes up to the next inflection point at peak pressure, so the third phase is called " peak pressure." Continuing from the time of peak pressure, the containment depressurizes, other than for perturbations resulting from the degree of subcooling in the PXS and external PCS water flow rate changec. Therefore, the final phase, from about 1200 seconds and beyond, is named
*long term depressurization," or simply "long term." Although the temporal partitioning is based on the containment pressure transient characteristics, the external cooling water film flow is the major means of heat removal with its own time sequence. The time sequence of events leading to the development of the external film is shown in Table 3-4. The film coverage begins after only 37 seconds, although the time to reach a quasi-steady state of coverage on the side walls is estimated to be 337 seconds based upon a flowrate of 440 gpm; the timing of events would vary for other at amed initial flow rates.
ACCIDENT SPECIF] CATION Revision 1 c:\3e92non.wpEltw1C1497 October 1997
3-21 l Pea k Blowdown 8thll Pretsstre Lone Thrert
/\
I Reduced Subeoeha / J PCS WaterflowGanon i I I I i 10 100 1000 le+04 le+05 D(880) Figure 3-9 Four Time Phases for DECLG Event ACCIDENT SPECIFICATION Revision 1 o:\3692nortwptib101497 October 1997
3-22 Table 3-4 Sequence of Events Leading to the Development of the PCS Cooling Film Time External Shell Surface Activity (seconds) Teinperaturem (.9 lireak triggers containment pressure setpoint 0 120 Valve opens (solenoid actuated, air-operated) 20 120 Pipe and bu:ket fill 37 121 First weir fuls 112 134 Second weir filt 187 144 i Steady coverage P established 337 174 Note: (1) Temperatures listed are calculated using the containment evaluation model which takes no credit for water on the containment shell until steady coverage is established. The extemal wetting processes, coupled with the slowly increasing external shell temperature, results in an increase over time in the wetted external shell surface area. As shown in Table 3-5, the flow of external cooling water onto the dome begins shortly after the start of refill, so during blowdown, the wetted areas are assumed to be zero and the entire shell surface area is dry. Furthermore, the shell time constant of approximately five minutes means the external shell surface temperature is too low to evaporate, radiate, or convectively transfer significant energy until after refill. Ad '.itional information on the containment shell temperature profile is provided in Reference 1, Section 7. Significant evaporation correspond; roughly to the time that steady water coverrge on the shell is achieved. As a result, heat transfer and evaporation from the outside of the shell are not significant during the blowdown and refill time phases. ACCIDENT SPECIFICATION Revision 1 0:\3692tum.wpf.It*101497 October 1997 l
3-23 Table 3 5 Large Break LOCA Sequence of Events
'Ilme (sec) Event 0 Break occurs Blowdown begins 20 PCS valve opens 30 Blowdown ends Refill begins 37 Water coverage begms on dome 90 Refill ends, peak press ne phase begins 337 '
Steady shell water coverage assumed to begin (440 gpm) (60.5 lb/sec) 1,500 Long-term containment depressurization begins 12,937 PCS water decreases to 122 gpm (16.9 lb/sec) 110,137 PCS water decreases to 72 gpm (9.9 lb/sec) 259,537 PCS water decreases to 63 gpm (8.6 lb/sec) A discussion of the break source and external cooling water characteristics during the four LOCA time phases follows. Blowdown Phase (0 - 30 seconds): The containment pressure increases from 1 atmosphere to approximately 3.5 atmospheres during the 30-second blowdown period. The mass released during blowdown is
- approximately 40 percent steam ar.d 60 percent liquid, and the break flow is choked, or nearly choked, throughout blowdown. The extemal water is not on the shell during blowdown.
Refill Phase (30 90 seconds): ( Following blowdown, the accumulators refill the lower plenum of the reactor with a high flow rate of cold water so that releases from the break cease for about 60 seconds. As the reactor water level rises through the core, typically termed reflood, water is turned to steam. The resulting steam and water flow rates from the break are very low and increase with time. The mass and energy release rates are two orders of magnitude less than the blowdown rates, and can be approximated as zero during the refill phase. With a negligible steam source rate and high condensation rate the containment pressure drops by a few psi from itr peak at the end of blowdown to the end of containment pressure refill phase at 90 seconds. The external water film is not considered for heat removal during this phase because the film is not well dc eloped until after refill and the extemal shell temperature is too low for effective heat removal. ACCIDENT SPECIFICATION Revisica 1 o:\3692ron.wpf:ltwl01497 October 1997 l
3-24 Peak Pressure Phase (90 - 1200 secondsh The post-refill and peak pressure steam source velocity is low encugh that a negligible amount of the break water will be entrained and dispersed as water drops. The external wetted coverage increases from near zero at the beginning of the peak pressure phase to a maximum at 337 seconds. The peak pressure phase ends when the PCS heat removal begins to exceed the' hest source and the pressure transient turns around. Long Term Depressurization (>1200 secondsh The long-term steam source velocity is low enough that no break water is entrained and dispersed as drops. The extemal wetted coverage remains at the coverage consistent with the source flow rate, liquid film stability, and the evaporation rate (evaluated more fully in Reference 1, Section 7). 3.4.3 Main Steamline Break 3.4.3.1 Description of MSLB Steamline ruptures occurring inside containment may result in significant releases of high-energy fluid to the containment environment, resulting in increased containment temperatures ar d pressures. The quantitative nature of the releases following a steamline rupture is dependent upon the configuration of the plant steam supply and protection systems, the :ontahunent design, the plant operatirig conditions. and the size of the rupture. The main steamline starts from she top of the steam generator, makes two 90-degree turns, and passes vertically down through the operating deck into the CMT compartment. The line makes another 90-degree tum and runs through the CMT compartment to the shell, where it passes outside containment. Part of the main steamline is shown in Figure 3-2. The break is assumed to occur at the top of the steam generator compartment, where the resulting flow pattern muumizes interactions with the bci ow-deck heat sinks in the CMT compartment. For this evaluation, a double-ended guillotine rupture of a main steamline is postulated to occur with the reactor at 30 percent steady-state power. The MSLB analyses that define this limiting case are presented in Table 6.2.1.1-1 of Reference 21. Since steam generator mass decreases with increasing power level, breaks occurring at lower power generally result in greater total mass release to the plant containment, and thus the greatest containment pressure increase. The main steamline isolation valve closure is assumed to be delayed permitting both loops to blowdown until closure (10 seconds after break). Offsite power is assumed to be available, since this maximizes the mass and energy released from the break (due to continued operatiet of RCS pumps and feedwater pumps). The availability of ac power in conjunction with the passive safeguards system, CMT and pecsive residual heat removal (PRHR) ACCIDENT SPECIFICATION Revision 1 l o-\3e92non.wpf:1b-101497 October 1997
3-25 maximizes the mass and energy releases via the break since this maximizes the reactor cooldown rate. When the PRHR is in operation, the core-generated heat is dissipated to the IRWST. The hELB blowdown flow is assumed to be superheated steam throughout the transient. Beyond the end of blowdown, the source energy release for an MSLB remains a; zero. The peak pressure during an MSLB is determined by the containment volume, steam / air circulation to the intemal containment heat sinks, and time response of the heat sinks. Although the extemal PCS water flow is credited in the h5LB SSAR analysis, its effect on containment response is not as pronounced as in the LOCA because the hGLB transient is much shorter. The pressure history for a representative AP600 hGLB transient is shown in Figure 340, 3.4.3.2 Temporal Partitioning (MSLB) The blowdown for the MSLB transient lasts approximately 400 seconds, after which there is no additional flow. With no mass and energy source, the containment pressure decreases rapidly as the internal heat sinks absorb energy and extemal cooling is provided by the PCS. Consequently, the MSLB evaluation does not extend beyond blowdown. ACCIDENT SPECIFICATION Revision 1 o:u692non.wpf:lt>101497 October 1997 __J
3-26 - I 1 L ~ I I I
') 1 2 3 4 X1E Tune (asc)
Figure 3-10 MSLB Containment Pressure vs. Tune ACCIDENT SPECIFICATION Revision 1 o:\3692non.wptib-101497 October 1997
l 41 4.0 PHENOMENA IDENTIFICA110N AND RANKING This section prm' ides the basis for the identification and ranking of the phenomena associated with the AP600 containment cooling process. An overview of the phenomena is described in subsection 4.1 Subsection 4.2 shows the PIRT and discusses how phenomena were grouped. Subsection 4.3 sumnurizes test and scaling results used to support the rankings. Subsection 4.4 provides the basis for the ranking of each phenomenon, based upon a combination of test results, scaling analyses, sensitivity studies, expert review, and engineering judgment. In addition, a roadmap of how each phenomenon is addressed in the evaluation model for the containment pressure DBA calculation is provided to document the methodology in accordance with 10CFR52.47(b)(2)(i)(A), subparagraphs (1) and (3). 4.1 PHENOMENA OVERVIEW An overview of the key phenomena involved in the containment cooling process is shown in Figure 4-1. This figure illustrates the energy transfer processes starting at the break source inside containment and leading to the environment. This figure represents the key hardware (shell, baffle, shield building), volumes (containment), and flow areas (downcomer, riser) involved in the respective heat and mass transfer mechanisms. The phenomena identification process was based upon identifying the p Tenomena or parameters that may affect the containment cooling process. A summary description of the significant phenomena involved in the containment cooling process is presented below, starting with the break source and progressing to the environment. Detailed phenomenon descriptions are provided in section 4.4 Inside Containment
- Flashing of high-pressure water (for LOCA) or release of superheated steam (for MSLB):
The break source steam disrupts the initial containment atmosphere with a forced jet that may later transition to a buoyant plume. The pressure inside containment will increase as long as the source rate of mass and energy into the gas atmosphere exceeds the absorption rate of mass and energy from the atmosphere by liquid and solid heat sinks and the containment steel shell. The characteristics of the break source mixture such as direction, momentum, and density are parameters that can affect circulation and stratification within containment. The density of the break source for the MSLB is lower than for the LOCA. PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:u692non wpf:1b-101497 October 1997
4-2 PCCWST cNmney g,,, (air exit) AliTW. rc Condatons l
- g. .
y film > conductance and capacitance air fkw / i l l tacn h evaporabon
~
& baffle Y
rog fam flow / ememal s'r i ikw condensanon drops
\ j annulus
/
annulus fim neer / downcom conductance & / especitance p
+ D$ radiation
,/
radiation radiation gas n#
/ -> convecnon m / convecten " '
/
condenaaten /
#// **
~"
/
mn/ a shelf 7%ucten
/.
sir *s /m/ w
\~ shield
=g-_building\
s stratircation natural convecnon Figure 4-1 Overview of Containment Building Phenomena PHENOMENA IDEN11r1 CATION AND RANKING Revision 1 o:\3692natwpelb-101497 October 1997
4-3 Saturated drops are formed during the blowdown process for a LOCA by entrainment into the high velocity steam and are transported with the steam into the containment, resulting in thermal effects as follows. The drops flash initially to thermal equilibrium with the containment atmosphere, and later evaporate as the containment pressure decreases and, therefore, may be another source of energy release to the containment environment. (See the paragraph below titled ' Fog inside containment during LOCA' for a description of the effuts of drops during the LOCA post-blowdown period.)
- Natural circulation and mixing of the steam / water / alt mixture inside containment:
The break source jet or plume cauas convective trmsport of steam, water, and warm air inside containment. The subsequent heat transfer and condensation from the warm, steam-rich gas to the initially cold heat sinks produce additional body forces and wall plumes that induce further changes in the convective flow field. Natural circulation is composed of the overall convective flow patterns that occur on a compartment scale and also on a large, containment-wide scale. The compartment scale circulation is due to wall layers, jets, plumes, and entrained flow. The large-scale circulation is flow between compartments induced by pressure, density, elevation, ar,d momentum differences and results in intercompartment flow. The pressure transient is mainly affected by parameters that influence mass transfer. Mass transfer has as its primary parameters steam concentration and velocity, the latter only for the case of forced convection. Largeccale circulation and entrainment into jets or plumes can drive mixing and can affect local values of steam concentration and velocity near the heat transfer surfaces. Jet and plume entrainment within compartments or the above-deck region can also result in stratification, or the existence of a vertical steam concentration gradient. The circulation ar.d stratification that occur in the AP600 have the potential to locally reduce heat and mass transfer rates by transporting and concentrating noncondensibles, or conversely, to improve heat and mass transfer by concentrating steam. Segregation is the separation of steam and air into different compartments due to convective and boundary layer transport processes. For example, condensation in the dead-ended below-deck compartments has the potential to segregate air and steam, creating an air-rich atmosphere within the dead-ended compartment, slightly enriching steam in the remaining atmosphere. Such a state will increase mass transfer to the shell and heat sinks above deck and decrease mass transfer to the heat sinks in dead-ended compartments.
- Fog inside containment during LOCA:
Fog is delivered inside containment by the break source steam / water interaction during blowdown only. The near-sonic velocities within the primary system pipe during blowdown are expected to entrain a large fraction of the liquid water. Post-blowdowm, PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA3692non.wpElb-10?497 October 1997 1
4-4 the primary system pipe break steam velocity is small enough that the entrainment rate may be assumed to be negligible. The vapor volume fraction is about 0.998, so drops coalescence is not expected. The fog is removed from containment primarily by
" gravitational settling" and by phase change, Neither settling nor phase change will eliminate the fog immediately, so a range of fog content is considered relative to thermal performance and circulation. The fog has a strong effect on the effective density of the gas mixture, and hence, on buoyancy-induced phenomena. Fog also increases the effective heat capacity of the gas mixture. The releases from an MSLB are superheated and thus contain no fog.
The effec,t of fog absorption on radiation can also be significant; however, inside containment fog will effectively enhance the absorption of radiant heat by the opaque gas, . which is already a significant absorber.
- Radiation heat transfer from high temperature mixture to the shell and internal heat sinks:
Energy is transferred by radiation between any two components-solid, liquid, or gas-that differ in temperature. The greater the temperature and the temperature difference, the greater the energy transfer. Radiation is also enhanced by high emissivity surfaces. The liquid film surfaces have emissivity of 0.95 to 0.96 (Reference 22, pg. 216), the inorganic zine paint emissivity is 0.90 to 0.95 (Reference 23), and the concrete emissivity is approximately 0.90 (Reference 24). Radiation between gases, and from gas to solid or gas to liquid, can be significant when the product of the steam partial pressure and radiation beam length are of the order of I ft-atmosphere (Reference 22, Section 5-10). This is the case inside containment, where steam partial pressures may be as high as 3 atmospheres and beam lengths are frecuently greater than 10 ft.
- Convective heat transfer to intemal heat sinks and containment shell:
Convective heat transfer is a boundary layer conduction process that is driven by a temperature gradient in the presence of a flowing bulk fluid. The greater the temperature difference, the greater the heat transfer. The bulk fluid motion may be due to a state of torced convection, free convec+ ion, or a combination of both, which results from convective flow driven by entrainment or density differences.
- Condensation of steam on inside of containment shell and internal heat sinks:
The condensation of steam is the convective transfer of mass to a liquid film on a heat sink. Condensation is a boundary layer diffusion process that is driven by a steam partial density gradient. Condensation removes the gas enthalpy (hs) fr m the atmosphere, transfers the heat of formation of the gas (hfs) to the heat sink, and leaves behind the liquid enthalpy (h f) with the condensate. PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA3692non.wpf-1b 101497 October 1997 l
4-5 Liquid film thermal transport (heat capacity): Liquid films form due to condensation inside containment and flow under the influence of gravity and shear forces due to extemal gas flows. The intemal liquid film carries away the liquid enthalpy (h f) that accounts for approximately 15 percent of the enthalpy of the condensed steam (hs ). The ratio of the energy retained by the film to the energy removed from the gas is represented by the ratio of PI values PI,g/(PI,jg + PI,jg) presented in Reference 2, Table 8-4. Values calculated from Table 8-4 range up to 14% The liquid film that forms on the intemal steel shell surface is collected at the crr.ne rail, the stiffener ring, and at the deck elevation, and then drained into the IRWST. Compartment filling: The condensation of steam from the intemal containment solid heat sinks (other than the shell that drains into the IRWST), in addition to the release of water from the RCS and fog "settimg," leads to filling of some of the compartments below the operating deck. Filling may restrict the natural circulation flow pattems through the lower compartments or submerge solid heat sink surfaces, and thus affect the energy transfer process.
- Liquid films conductance-vertical and horizontal:
The existence of the liquid film (due to condensation) on the internal shell and heat sink surfaces offers a thermal resistance to the transfer of energy from the containment gas to the shell or heat sink surface. The University of Wisconsin tests (Reference 25) showed that downward-facing surfaces with inclinations greater than l' from horizontal drained by flowing films, while inclinations less than l' formed drops that rained off. Rain occurred with a heat transfer coefficient greater than 1000 B/hr-ft 2 .F. Less than 0.09% of the containment shell has a surface slope of I degree or less, so rain from the shell surface is of little significance. The flowing film Nusselt number correlation (Reference 9, Section 3.10) shows a minimum at the transition from wavy laminar flow to turbulent flow, from which the minimum film heat transfer coefficient inside containment is approximately 600 B/hr-ft 2 'F. Condensate on upward facing horizontal surfaces can develop rather thick films (greater than 0.05 in.) that may be more limiting to heat transfer than the heat sink internal resistance or the mass transfer coefficient.
- Concluction heat transfer through intemal heat sinks The intemal resistance of the containment solid heat sinks must be considered in the transfer of energy from the containment atmosphere. The intemal resistance can be scaled s relative to the surface heat transfer using the Biot . number, ht/k. Heat sinks with Biot numbers less than 0.1, such as steel with thickness less than approximately 0.5 in., may be PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692nortwpf-lb-101497 October 1997
l 4-6 simply modeled as lumped masses. Heat sinks with Biot numbers greater than 1, such as concrete and water pools, limit energy absorption by intemal resistance rather than by l surface trensfer coefficients. Containment Shell
- Conduction heat t ansfer through steel shell:
The internal resistance of the shell must be considered in the transfer of energy from the containment atmosphere. The internal resistance can be scaled relative to the surface heat transfer using the Biot number, ht/k. Heat sinks with Biot numbers less than 0.1, such as steel with thickness less than approximately 0.5 in., may be simply modeled as lumped masses. Heat sinks with Biot numbers greater than 1, such as concrete and water pools, limit energy absorption by intemal resistance rather than by surface transfer coefficients. [ Biot number for shell?]
- Stored energy in containment shell and internal heat sinks:
The containment volume, shell, and other sinks (concrete, steel, and water pools) provide a large capability to store energy from the break. Outside Containrrent
- Radiation heat transfer in PCS air flow path:
In the riser, downcomer, and chimney, steam partial pressures are on the order of 0.1 atmosphere and beam lengths are on the order of a few feet, so radiation to or from gases in the PCS air flow path are relatively small. Drops or fog in the riser annulus may capture radiation. Radiant heat transfer from the shell to the baffle may keep the baffle at a temperature greater than the air in the annulus.
- Convective heat transfer in PCS air flow path:
The convective heat transfer phenomenon in the external annulus is similar to the intemal containment convective heat transfer, but is driven by other parameters. These parameters include the environmental conditions, such as temperature and wind velocity.
- Liquid film thermal transport:
The extemal liquid film from the PCCWST is supplied at a temperature maintained at 40*F to 120 F per Technical Specifications. The energy absorbed by the sensible temperature increase of the film, before it begins to evaporate significantly, accounts for up to 8 percent of the energy transferred by evaporation (Reference 2, Table 8-4, PI,,). The energy absorbed by temperature increase is sometimes referred to as the subcooled PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:u692non.wpf:n-101497 october 1997
47 l heat capacity of the external film. Most of the external film is expected to evaporate, but i the portion that does not evaporate runs down a drain and away from the PCS flow path. Other characteristics of the liquid film such as stability, stripping, and drag can affect the energy transport process, as described in subsection 4.4.8.
- Evaporation of water on the outside of the containment shell:
The evaporation of water is a convective mass transfer from a wetted surface. Evaporation mass transfer is driven by a concentration gradient and will take place if the partial pressure of steam at the wet surface is greater than the partial pressure of steam in the adjacent environment. The evaporation rate is the product of evapolation mass flux times the wetted area. The evaporation mass flux is a function of annulus conditions and the film surface temperature. The wetted area is a function of the shell heat flux and film temperature and flow rate, all of which are interdependent and vary with time and position. The heat flux to the shell is dependent on internal containment conditions and shell conduction. The wetted area and the film flow rate are dependent on the applied PCS water flow rate that decreases with time, and are also affected by the film stability.
- Fog generation in the riser annulus:
The external air flow path bulk gas is superheated at design basis conditions, so fog absorption of radiation i.s not expected for the AP600. When fog does exist, as in the large-scale test riser, its effect on radiation through the riser air is not significant. Radiant heat transfer from the shell is a second order effect relative to evaporation. Since fog can potentially impact the buoyancy of the annulus, it is evaluated for that effect. Fog present in the annulus will absorb a fraction of the radiant heat leaving the shell, which will cause a fuction of the fog to evaporate. Evaporation of fog adds low molecular weight steam mass to the riser annulus, increasing the buoyancy in the annulus riser.
- Natural circulation in the PCS air flow path:
The shell heating and the cooling film evaporation on the - Ae of the shell induce a natural circulation air flow in the PCS air flow path (downcmer-riser-chimney, as shown in Figure 44). The resulting air flow rate affects the heat and mass transfer coefficients from all the bounding structures (shield, baffle, shell, and chimney) to the moving air. The characteristics of the air flow on the liquid film may have an effect en the momentum and energy transport processes, as described in subsection 4.4.8. PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf-lb-10149' October 1997
4-8 4.2 TIIE AP600 CONTAINMENT PilENOMENA IDENTIFICATION AND RANKING TABLE The PIRT rankings for the four LOCA and one MSLB time phases are presented in Table 4-1. This table contains the component or volume in the first (left-hand) column and the phenomena or parameters most closely associated with that component or volume in the second column. The table is arranged to show the energy transfer processes starting from inside containment with the break source and ending with the ultimate heat sink, the environment. For both simplicity and convenience in this PIRT, each energy transfer process has been assigned to the component or volume '.o which the process is most closely associated, either containment hardware such as the steel containment shell, or a gas mixture such as the containment volume, or a flow area such as the annulus riser, in particular, the heat transfer fluxes have been associated with the respective hardware surfaces and generally, with the energy source. For example, radiation heat transfer between the shell and the baffle was assigned to the shell. The PIRT includes parameters such as the initial and boundary conditions, which are important to the analysis of the containment cooling process. Results from testing and scaling used to support the PIRT ranking are summarized in subsection 4.3. A description of how each phenomenon was ranked can be found in subsection 4.4. The numbers and letters for the components and phenomena in Table 4-1 refer to the specific paragraphs in subsection 4.4.X, where X represents the component or volume number in Table 4-1. For example, the basis for ranking break source mass and energy in the containment can be found in subsection 4.4.1A. mm PHENOMENA IDENTIFICATION AND RANKING Revision 1 l o:\3692non.wpt:1b-101497 October 1997
4-9 Table 41 Phenomena identification and Ranking According to Effect on Containment Pressure Component or Phenomenon or Parameter LOCA MSLB Volume Blow Ref;ll Feak l Long Term Blowdown (30-90 (>1200 sec) (0-400 sec)
@ 30 sec) (90-1200 sec) ,,7)
Inside Containment: 1.) Break Source - A.) Mass and Energy H N'A H H H B.) Direction and Elevation H N/A H L H C.) Momentum H N/A H L H D.) Density H N/A H L H E.) Droplet / liquid flashing L L L L N/A (thermal) 2.) Cor.:ainment A.) CirculatiorvStratification H H H H H Volume B.) Intercompartment Flow L H H H H C.) Gas Compliance H H H H H D.) Fog (circulation) L H H H N/A E.) Hydrogen Release L L L L N/A 3.) Containment A.) Liquid Film Energy L L L L L Solid Heat Transport Sinks g B.) Vertical Film Conduction L L L L L
" ' 'I C.) Horizontal Film L H H H H Conduction D.) Intemal Heat Sink M H H M H Conduction E.) Heat Capacity M H H M H F.) Condensation M H H M H G.) Convection from L M L L L containment H.) Radiation from L M L L L containment PHENOMENA IDENrlFICATION AND RANKING Revision 1 oA3692non.wpf:1t>.101497 October 1997
4-10 Table 41 Phenomena identification and Ranking According to Effect on Containment Pressure ! Component or Phenomenon or Parameter LOCA MSLB Volume Blow Refill Peak Long Term Blowdown (30-90 (>1200 sec) (0-400 sec)
@ 30 sec) (90-1200 sec) ,,c) 4.) Initial A.) Initial temperature M M M M M Conditions B.) Initial humidity M M M M M C.) Initial pressure M M M M M 5.) Break Pool A.) Circulation /Stratificati;,a L L L M L in the pool B.) Condensation / L L L M L evaporation C.) Convection within L L L L L containment volume D.) Radiation within L L L L L containment volume E.) Conduction in pool L L L M L F.) Corapartment filling L L L L L 3
Inside Containment: 6.) (RWST A.) Mixing / Stratification (gas L L L L L
& water)
B.) Condensation L L L L L C.) Convection L L L L L D.) Radiation L L L L L E.) Conduction in liquid L L L L L F.) Liquid level changea, L L L L L PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf:1b 101497 October 1997
4-11 l Table 41 Phenomena identification and Ranking According to Effect on Containment Pressure Component or Phenomenon or Parameter LOCA MSLB Volume Blow Refill Peak Long term Blowdown (30-90 (>1200 sec) (0 400 sec)
@ 30 sec) (90 1200 sec) ,,,)
Containment Shell: 7.) Steel Shell A.) Convection from L L L L L containment B.) Radiation from L L L L L containment 4
..) Condensation H H H H H D.) Inside film conduction L L L L L E.) Inside film energy L L L M L transport F.) Conduction through H H H H H shell G.) Heat capacity of shell H H H L H H.? Convection to riser L L L M L annulus I.) Radiation to baffle L L L M L J.) Radiation to chimney L L L L L K.) Radiation to fog / air L L L L L mixture L.) Outside film conduction N/A N/A L L L M.) Outside film energy N/A N/A M M L trensport N.) Evaporation to neer N/A N'A H H M annulus PHENOMENA IDEN11HCATION AND RANKING Revision 1 o:\3692non.wpf:1b-101497 October 1997
442 4 Table 4-1 Phenomena identification and Ranidng According to Effect on Containment Pressure Component or Phenomenon or Parameter LOCA MSLB Volume Blow Refill Peak Long Term Blowdown (30 90 (>1200 sec) (0-400 sec)
@ 30 sec) (90-1200 sec) ,ne) 8.) PCS Cooling A.) PCCWST flow rate N/A N'A H H L Water B.) PCCWST water N/A WA M M L temperature C.) Water film stability and N/A N'A H H L coverage D.) Film stripping N/A N/A L L L E.) Film drag N/A N/A L L L Outside Containment:
9.) Riser Annulus A.) PCS Natural Circulation L L M M M
& Chimney Volume B.) Vapor acceleration N/A N/A L L L C.) Fog N'A N/A L L N/A D.) Flow stability L L L L L 10.) Baffle A.) Convection to riser N/A N/A L M N/A annulus B.) Convection to N/A N/A L M N/A downcomer C.) Radiation to shield N/A N/A L L N/A building D.) Conduction through N/A N/A L M N/A baffle E.) Condensation N/A N/A L L N/A F.) Heat capacity MA N/A L L N/A G.) Leaks through baffle N/A N/A M M N/A PHENOMENA IDENiu 1 CATION AND RANKING Revision I o:\3692non.wpf'1b-101497 October 1997
4-13 Table 4 t Phenomena Ideratification and Ranking According to Effect on Containident Pressure Component or Phenomenon or Parameter LOCA MSLB Volume Blow Fwfill Peak Long Term Blowdown (30 90 (>1200 sec) (0-400 sec)
@ 30 sac) (90-1200 sec) ,,c) 11.) Baffle A.) Convection to riser air L L L L L Supports t
B.) Radiation from shell L L L L L C.) Conduction from shell L L L L L D.) Heat capacity L L L L L 12.) Chimney A.) Conduction through L L L L L Structure chimney B.) Convection from L L L L L chimney air C.) Heat capacity of L L L L L structure D.) Condensation on L L L L L chim.ney 13.) Downcomer A.) PCS Natural L L M M M Annulus Circulation B.) Air flow stability L L L L L 14.) Shield A.) Convection to N/A N/A L L L Building downcomer B.) Conduction through N/A N/A L L N/A shield building C.) Convection to N/A N/A L L N/A environment D.) Radiation to N/A N/A L L N/A environment PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf:1b-101497 October 1997 l
4-14 Table 41 Phenomena identification and Ranking According to Effect on Containment Pressure Component or Phenomenon or Parameter LOCA MSLB Volume Bl:w Refill Peak Long Term Clowdown (30-90 (>1200 sec) (0-400 sec)
@ 30 sec) (90-1200 sec) ,g) 15.) Extemal A.) Temperature PCA N/A L L L Atmosphere B.) Humidity ffA t#A L L L C.) Recirculation tCA N/A L L L D.) Pressure Fluctuations PCA N/A L L L
+4 4 PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\36 mon.wpf:1t>.101497 October 1997
4-15 4.3 TEST AND SCALING RESUL'Is USED IN PHENOMENA RANKING This section provides a summary of the results from tests (subsection 4.3.1) and scaling analyses (subsection 4.3.2) used in ranking the phenomena, organized by report. The specifi<. ranking bases for each of the phenomena is provided in subsection 4.4, including specific test information and scaling PI groups. 4.3.1 Test Results Summary This section provides an overview of results from the containment tests, organized by test program, which in conjunction with the scaling analysis results, engineering judgment, and sensitivity studies were used to provide the final importance ranking of the phenomena. Test results are also referenced as they apply to specific phenomena in subsection 4.4. Heated Flat Plate Test (Reference 26) Results The following are the key observations and conclusions from the Heated Flat Plate Test analysis: A stable, wavy laminar water film formed on the hot, coated, steel surface in both orientations-vertical and 15 degrees from horizontal. As the water flow rate was reduced, the waves in the film became smaller and eventually disappeared. The water film was able to wet and rewet (after dryout) the hot, dry surface (at a surface temperature of 240*F). The two low-flow, high-heat flux tests showed that evaporation up to the point of dry-out of the water films on the heated surface produced stable film evaporation. The water film was not adversely affected by the countercurrent cooling air flow up to the maximum air velocity of the test range (5.9 to 38.7 ft/sec.), i.e., there was no water-film stripping. Nominal heat transfer from the dry surfaces to the air (no water film) agreed well with the Colburn heat transfer correlation (Reference 9, Section 4.1). Water film evaporation and resultant heat removal agreed with mass transfer correlation predictions. Radiation to the air baffle wall and subsequent heat transfer to the cooling air occurred and accounted for heat transfer in addition to the convective heat transfer. PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:u692nmwpf:1b-101497 October 1997 ?
4-16
~.-
Wind Tunnel Test (Reference 27,28,29,30,31) Results l Based upon evaluation of the wind tunnel test data, it was found that wind induces more containment heat and mass transfer than a quiescent atmosphere since wind drives more flow through the AP600 annulus (wind positive). Also, a literatute review provided data to show that the effects of recirculation due to thermalinversions or strong winds have a negligible impact on the PCS heat removal (Reference 1, Section 6.3). Condensation Test (Reference 8,25,32) Results The following are the key observations and conclusions from the Condensation Tests: A slightly higher average heat transfer coefficient was observed on the horizontal condensing plate than on the vertical condensing plate. The presence of helium affected the heat transfer coefficients the same as other noncondensibles at the same molar concentration. Increased heat transfer coefficients were observed with the steam jet impinging directly on the horizontal plate (simulating steamline break). Water Distribution Test (Reference 33,34,35) Rcsults The water coverage for an equivalent AP600 flow rate of 220 gpm was estirrated to be 25 percent from the top of the dome down to the first weir. About 70 percent of the surface was wet between the first and second weirs and about 100 percent wet below the second weir. The coverage decreased from 100 percent as the applied water flow rate decreased. At a flow rate equivalent to 120 gpm on the plant, water began to spill over the first weir at about 2.5 minutes and over the second weir at about 5 rainutes. The time to completely fill the weirs to their steady-state level and to establish steady-state coverage of the dome and sidewall was conservatively estimated to be about 10 minutes. These tests reflected the plant design flow rate of 220 gpm at the time the tests were performed. Recent plant evaluations have been performed with a design flow rate of 440 gpm. Small-Scale PCS Integ_ral Test (Reference 36) Results The following observations and conclusions with respect to the water film were drawn from evaluation of these tests (Reference 14). A stable, uniform, wavy laminar film was formed on the inorganic zinc-coated steel surface using simple weirs. PHENOMENA IDENTIFICATION AND RANKD4G Revision 1 o:u692nonwpt:1t>.101497 October 1997
4 17 The film remained stable and uniform on the vs .ical sidewall of the vessel at a terage, evaporating heat fluxes in the range of those expected on the AP600. l
- The local heat removal rate at the top of the vessel where cool water was first applied was significantly higher iban the vessel average heat removal rate.
The overall heat removal capability with a wetted surface and a well-mixed air / steam mixture inside the vessel agreed with analytical predictions. Large-Scale PCS Integral Test (Reference 12, 37) Results The following important observations with respect to film behavior were made during the tests: As the pressure and temperature increased inside the pressure vessel, dry spots first began to form in the wet, but low tiow regions on the dome and sidewall. The dry spots grew vertically, separating the original continuous film into several wavy laminar flow stripes. At higher heat fluxes, dry spots also formed just below and in line with the J-tube location. The central wavy laminar flow region of the individual film stripes was surrounded by a region of laminar flow with no visible waves. The thickness of the laminar flow region appeared to continually decrease out to the edge (or bottom) of the film stripe. The widths of both the wavy laminar and lammar flow regions of the stripe were observed to decrease with increasing heat flux. The film stripes remained stable (i.e., they did not split or bunch up to form thick, narrow rivulets) as they evaporated on the vertical sidewall. An evaluation (Reference 38) of the LST data provided important conclusions on both water coverage and heat re;noval: Evaporation was the primary mode of heat removal from the outside of the vessel (approximately 75 percent) followed by the sensible heating of the subcooled film (approximately 17 percent). The remainder of the energy (8 percent) was transferred by convection and radiation. Striped film coverage provided better heat removal than forced quadrant coverage for the same wetted perimeter. PHENOMENA IDENiu"lCATION AND RANKING Revision 1 c:\3692non.wptn401497 October 1997
4 18 The heat removal rate appeared to be more affected by ambiant air temperature than I by liquid film temperature. The heat removal rate had a relatively weak dependence on annulus air velocity. The highest heat flux occurred near the top of the dome at the elevation where the external film was applied except for the hsritontal, high-velocity steam jet injection test case. Injection of a high-velocity steam jet (simulating a MSLB) resulted in a well-mixed vessel and thus, a relatively uniform wall tempecature and heat flux over the evaporating surface. The heat removal rate increased as the steam con:entration near the containment shell increased (by raising the injection location). 4.3.2 Scaling Analysis Results Summary A scaling analysis showed that the range of AP600 operation was adequately covered by the separate effects test data (Reference 2, Section 10.1). Reference 2, Section 10.2 describes how the LST scales to AP600. Reference 2, Section 11 describes important scaling distortions in the LST, and how the distortions are addressed. The scaling analysis identihed the relative importance of the phenomena under study, such as condensation mass transfer and evaporation mass transfer via the appropriate non-dimensional parameters or PI groups. Specific PI groups from the scaling analysis relevant to each phenomenon are given in subsection 4.4. The PI group values specified in the ' Scaling Results' subheading under the Basis for PIRT Ranking heading in subsection 4.4 reflect the scaling calculations documented in revision 0 of the Scaling Report (WCAP-14845). A number of the PI group values to be documented in Reference 2 (to be issued 6/97) will be updated from those reported herein to correct a calculational error. However, the change in PI group values is insignificant, and does not impact the conclusions documented in this report. The evaluation of the PI group values provides the following conclusions. Inside containment: The break source steam mass flow rate is important since it drives the pressurization. The gas volume is important since it relates pressure to stored mass and energy (volumetric compliance or capacitance). The liauid condensate is important since it carries away part of the energy of the condensed steam. For mass transfer to heat sinks: PHENOMENA IDENTIFICATION AND RANKING Revision 1 1 o:\3692non.wphlb 101497 October 1997
I' 4 19 The internal steel, concrete, and steel-jacketed concrete heat sinks are important since they absorb energy and condense steam, thereby reducing pressure. Intercompartment circulation affects the distribution of noncondensibles and velocity, which are important parameters for mass transfer. Stratification is important since it can increase the concentration of dense non-condensibles and locally limit the utilization of heat sinks and conductors within a compartment or conversely, improve mass transfer by concentrating steam. The, horizontal liquid films are important since they can produce low conductares that insulate upward-facing horizontal surfaces. Conta;nment shell: The shell is important since it is a major heat sink and the only energy transfer path out of containment to the riser. The internal conductance of the shell is important since it limits energy absorption and transfer rates. The condensatien and evaporation mass transfer conductances are important since most of the energy transfer to and from the shell is by mass transfer. The liquid film stability is important because it can limit the area for evaporation and evaporation is the dominant process for energy transfer from the shell. Outside containment The buoyancy and flow resistance in the PCS air flow path are important and have a strong effect on the evaporation rate. The downcomer is not a significant contributor to air flow path energy or momentum. 4 PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpt:1b-101497 October 1997
.. ___ a
-- .- -.. - - - - - _ - - . - - - - _ .~. - - ..-.- ... - -. ...-
.4-20 4.4 RANKING OF PHENOMENA LISTED IN PIRT This section discusses the purpose of ranking phenomena and provides the bases for the rankings. A roadmap of how each phenomenon is addressed for the containment pressure DBA calculation is provided, and includes the following information:
- PIRT Ranking Basis for PIRT Ranking Test Results Scaling Results -
Sensitivity Studies Expert Review How Phenomenon Is Implemented In Evaluation Model Justification Of Evaluation Model Treatment Of Phenomenon Test Experience Modeling Guidance Sensitivity Studies Evaluation Model Treatment of Uncertainty, Distortions Bases for Ranking The phenomena relevant to containment DBA are shown in the PIRT (Table 4-1). The PIRT has been structured into high level groupings consistent with the discussion in subsection 4.2. Phenomena were ranked based upon their effect on the energy transfer process and containment pressure reduction. The ranking of the importance of the containment phenomena was based on a combination of test results, scaling analysis results, sensitisity studies, expert review, and engineering judgment. Sources of information used to rank each phenomenon are provided in each of the following subsections. The phenomena were ranked either:
- H- High importance to the energy transfer process and containment pressure reduction
- M- Medium importance to the energy transfer process and containment pressure reduction
- L- Low importance to the c.nergy transfer process and containment pressure reduction N/A - Not applicable to the energy transfer process PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf:1b-101497 October 1997
4-21 Purpose of PIRT Ranking The ranking is used to establish the most important processes for containment pressure mitigation and to provide guidance on the appropriate level of detail in assessing uncertainties or developing bounding models, as follows: Phenomena with a High or Medium ranking during any LOCA time phase or during a MSLB need to be considered in the evaluation model. Consideration may include showing that neglecting a phenomenon is conservative for pressure predictions. The effect of important parameters is assessed in developing uncertainties or bounding models. Phenomena with a Low ranking during all LOCA time phases and during a MSLB are those that have a small effect on pressure. As a result, it is acceptable to use an available best-estimate or realistic model in the evaluation model. In some cases, the low ranked phenomena may be neglected if their effect on containment pressure is small or if it would be conservative with respect to containment pressure to neglect them. The bases for ranking each of the phenomena or parameters identified in the PIRT and the justification for how the phenomena are addressed in the evaluation model are provided in the subsections below. The last digit of the section designator refers to the PIRT high-level grouping number and the letter corresponds to the specific phenomenon entry in Table 4-1. LST Distortions Related to Evaluation Model Representation The AP600 containment evaluation model requires both system level validation and a mere detailed level, or component level validation. The LST integral system test response is compared to the AP600 system response in Reference 2, Section III and is 'used to identify the system-level distortions in the LST. The component level phenomena that must be addressed are those identified in the PIRT In the following subsections, the test basis that supports the t PIRT ranking and the modeling basis for treating each phenomena in the evaluation model are discussed. PHENOMENA IDENTIFICATION AND RANKING RevisioU o:\3692non.wpf:1b-101497 October 1997 1
4-22 4.4.1 Break Source I 4.4.1A Mass and Energy Release of Break Source PIRT RANKING HIGH for all LOCA phases, except N/A for refill. HIGH for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) ,
- Test Results The LST covered a range of steam flows (0.11 - 1.54 lbm/sec) that demonstrated that steam flow was a dominant parameter for vessel pressure (Reference 15, Figure 4.3-1)). The LST results support a ranking of High for this phenomenon.
- Scaling Results Scaling shows that the mass and energy releases from the break source are the driving forces for the containment response for both the LOCA and MSLB events, except during the refill portion of the LOCA when there is no breah source. Pressure scaling PI values (PI p, prk work) equal 1.0 for all LOCA phases and MSLB (except for LOCA refill when there are no releases). (Reference 2, Section 8.5)
- Sentitivity Studies A WGOTHIC peak pressure analysis sensitivity case using nominal M&E releases instead of DBA assumptions showed a shift of the peak pressure to the blowdown phase and about a 4 psi decrease in peak pressure (Reference 1, Fection 10). The sensitivity of containment pressure to mass and energy releases supports a nnking of High for all LOCA time phases and MSLB (except for LOCA refill when there are no releases).
Expert Review The expert review, summarized in Appendix A, provided a ranking of High for all LOCA time phases and MSLB (except for LOCA refill when there are no releases). EIIENOMENA IDENTIFICATION AND RANKING Revision 1 on3692non wpf:153-1014s7 October 1997
- 4 23 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter)
I ja,e PJSTIFICAilON OF EVALUATION MODEL TREATMENT OF PHENOMENON (Test E:.perience, Modeling Guidance, Sensitivity Studies) l ja,e PHENOMENA IDENTIRCATION AND RANKING Revision 1 o:\3692non.wpf:1b101497 October 1997
4-34 . I ja,e EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I Ja,c PHENOMENA IDENDFICARON AND RANKING Revision 1
" W h *Phlb101497 October 1997
4 25 4.4.1B Break Source Direction and Elevation - PIRT RANKING HIGH for LOCA phases - blowdown, peak pressure LOW for LOCA phase -long term N/A for LOCA phase - refill HIGH for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) , Test Results The LST steam concentration measurement results (Reference 33, Section 2.4) for tests 222.2B and 222.4B shawed that direction and elevation can be important for containment pressure due to their 2ffect on circulation and stratification within containment. Sensitivity Studies Sensitivities were performed to assess the effect of circulation and stratification on calculated pressure for LOCA and MSLB. The LOCA releases are at the elevation of primary piping in the steam generator compartment. For the LOCA, sensitivities to 8 var aus break release locations below deck (Reference 1, Section 9.3.2.5) show that the intemal solid heat sinks reach near maximum relative effectiveness well before the time of peak pressure and the muimum calculated change in pressure is less than 0.5 psi for the range of directions and locations studied. For the MSLB, jet direction is not a significaat parameter since the jet kinetic energy is sufficient to drive circulation through the containment (Reference 1, Section 9.2.2) leading to relatively uniform steam concentrations. The MSLB elevations are the steamline above the operating deck (Reference 1, Section 9,4.1.1 and 9.4.3) and the steamline in the CMT room (Reference 1, Section 9.4.1.2 and 9.4.3). The steamline break in the CMT room was shown to yield more effective containment solid heat sink utilization and provided a peak pressure benefit of approximately 1.7 psi. Expert Review The expert review summarized in Appendix A provided a ranking of High for all time periods except for the refill period when there were no M&E releases and during the long term period when the M&E releases were very small. s PHENOMENA IDENIIFICATION AND RANKING Revision 1 on3692non.wpt:1b-toi497 october 1997
446 liOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) ( , t ja,c JUSTIFICATION OF EVALUATION MODEL 'IREATMEffr OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) Ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 0:\3692non.wpf 1b 101497 October 1997 1
" 4 27 I
l ja.c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja.c PHENOMENA IDENTIFICATION AND IMNKING Revision 1 0:\3692non wpf Ib-101497 October 1997
4-28 4.4.1C Break Source biomentum PIRT RANKING lilGH for I.OCA phases - biowdown, peak pressure N/A for LOCA phase - refi'.1 LOW for LOCA phase - long term lilGli for hiSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) ,
- Test Results The LST results (Reference 38, Section 2.4) for tests 201.1,202.1, and 203.1 compared to tests 201.2,202.2, and 203.2 showed that the velocity (momentum) can be important on containment pressure due its effect on circulation and stratification within the containment.
Sensithity Studies See subsection 4.4.1B for summary of relevant sensitivities. Expert Review The expert review summarized in Appendix A providtd a ranking of High for all time periods, except for refill period when there were no M&E releases, and during the long term period when the M&E releases were very small. HOW PHENOhiENON IS IhiPLEhiENTED IN EVALUATION hiODEL (e.g., boundary condition, torrelation, code option, noding, input parameter) I ja,e PHENOMENA IDENTIFICATION AND RANKING ' Revision 1 o:\3692nortwphib 101497 October 1997
4 29 JUSTIFICATION OF EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) ( Ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS l ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692non.wpf;1b 101497 October 1997
4 30 I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 0:\3692non.wptib 101497 October 1997
4 31 4.4.1D Break Source Density ! g PIRT RANKING ! HIGil for LOCA phases - blowdown, peak pressure l N/A for LOCA phase refill , 2 LOW for LOCA phase -long term HIGli for MSLB i BASIS FOR PIRT RANKING (Test Results, Scaling Results, "ensitivity Studies, Expert
- Review) ,
) j-
- Expert Review The expert review summarized in Appendix A supports the ranking of this phenomenon as High for all time periods, except for refill period when there were no M&E releases and during the long term LOCA period when the M&E releases were very small.
HOW Fi!ENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I jax JUSTIFICATION OF EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I jax , PHENOMENA IDENT1FICATION AND RANKING - Revision 1 c:\3692non.wpf tb 101497 October 1997
4-32 l ja c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I j..e 1 PHENOMENA IDENTIFICATION AND RANKING Revision 1 a\369:non.wpf:1b-101497 October 1997
4 33 4.4.1E Droplet / Liquid Flashing The blowdown liquid and entrained droplets enter the atmosphere saturated at the containment total pressure where they are exposed to the containment gas mixture of air and steam at the steam partial pressure. Since the liquid and drops are initially superheated, they evaporate quickly to reach thermal equilibrium with the gas mixture. The MSLB releases superheated steam, so droplet / liquid flasinng is ranked N/A. PIRT RANKING LOW for all LOCA phases N/A for MSLB UASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Rcview)
- Scaling Results Droplet / liquid flashing can increase peak containment pressure by adding steam mass. However, the scaling results (Reference 2, Section 8.5) showed that the thermal effect of drops were not very important with pressure scaling PI values (P1 p.A d )
less than 0.05 for all time periods. The scaling results support a Low ranking for this phenomenon. Sensitivity Studies The assurned fraction of liquid turned into drops during blowdown was studied over the range of 0 to 100 percent, and the effect on thermal performance (Reference 1, Section 5.8) and circulation (Reference 1, Section 9.3.2.6) have been examined, it is reasonable to assume that the fraction of liquid turned into drops during blowdown in AP600 is significantly greater than 5%. The sensitivity results showed that with greater than 5% of the liquid delivered as drops, peak pressure and long term pressure were only weakly affected, although an effect was seen on the blowdown pressure prediction. These sensitivity studies support a Low ranking for tlus phenomenon. Expert Review The expert review provided in Appendix A provided a ranking of High for blowdown and refill and Not Applicable for other periods. The final ranking of Low for this phenomenon was based on the scaling and sensitivity study results described above. PHENOMENA IDENTIFICATION AND RANKING Revision 1 a:\3692non.wpf:1b.101497 October 1997
4-34 i HOW PiiENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary l condition, correlation, code option, noding, input parameter) ja,e JUSTIFICATION OF EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,C PHENOMENA IDENTIFICATION AND I'ANKING Revision 1 o:\36 mon.wrf 1b101497 October 1997
4 35 i Ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja.c PHENOMENA IDENIIMCATION AND RANKING Revision 1 a%3692nortwpf.1b.101497 October 1997
4-36 ' 4.4.2 Containment Volume 4.4.2A Circulation / Stratification in the Containment Volume Circulation and stratification are phenomena used to describe convective flow inside containment. Segregation is sornetimes used to describe one of the possible effects of circulation and stratification on steam distribution. Predicting these phenomena requires the modeling of specific convection processes. Convection is the Dow of Hulds due to acceleration, pressure, shear and body forces acting locally on each microscopic fluid element. On a macroscopic scale, the collective motion of Guid elements organ'2es into recognizable fluid structures, such as jets, plumes, and wall plumes that interact with the bulk fluid by entrained flow. On a compartment scale, the combination of jets, plumes, wall plumes and entrained Guld produces circulation pattems. The strength of the compartment-scale circulation is respnsible for the steepness of vertical density gradients, or stratification, within containment gas volumes. Dead-ended compartments may stratify in a stable manner and not experience circulation. Intercompartment convective transport produces a pattern of containment scale circulation inside containment. The intercompartment flow can occur due to one or more of the following four processes:
- Pressurization of one compartment by the steam source, as during LOCA blowdown
- Momentum induced convective flow Condensation in a compartment creating a lower pressure Net buoyant force not counteracted by pressure or mcmentum Condensation dominates convection and radiation by more than an order of magnitude based upon the results from the scaling analyses. Circulation / stratification affects condensation via steam density and velocity fields.
PIRT RANKING HIGH for all LOCA phases HIGH for MSLB l BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert l Resiew)
- Test Results l
The LST results (Reference 38, Section 2.4) for tests 222.2B and 222.4B showed that l circulation and stratification within the containment can strongly influence mass transfer rates. LST tests have been evaluated (Reference 1, Section 9.2) to support a specific evaluation of the effect of stratification on containment pressure for various PHENOMENA IDENTIFICATION AND RANKING Revision 1 l o:u692nongf:1b-101497 October 1997
4-37 postulated scenarios. While circulation and stratification can strongly influence steam concentration distributions, sensitivity analyses (see subsection 4.4.1B) show that the internal heat sinks reach near maximum relative effectiveness well before peak pressure. The steam concentration dist.Butions continue to affect the PCS heat removal throughout the transient.
- Expert Review The expert review summarized in Appendix A provided a ranking of High for all time periods, except for refill when there were no M&E releases and during the LOCA long term phase when the releases were very small.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, conclation, code option, noding, input parameter) I ja,e JUST!FICATION OF EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,e 1 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja.c PHENOMENA IDENrlFICATION AND RANKING Revision 1 a.sx9 non.wpr:151oi497 October 1997
4 38 l l ja,c PHENO? iENA IDENTIFICATION AND RANKING Revision 1 o:\3692nm.wyttb.101497 &toWr 1997
4 39 4.4.28 Intercompartment Flow in Containment Volume Flows resultir ' from the driving forces discussed in subsection 4.4.2A are affected by the l geometry and loss characteristics of flow paths between compartments. Flow through l compartments and the above deck region produces a pattern of containment scale circulation inside containment. Intercompartment flow can tend to reduce stratification gradients, and affects the steam concentration near heat transfer surfaces, which affects mass transfer rates. Since intercompartment flow affects parameters, such as steam concentration and velocity, that are important to mass transfer, the ranking of intercompartment flow is the same as that for condensation on the steel shell (subsection 4.4.7C) inside containment. PIRT RANNING LOW for LOCA phase - blowdown HIGH for LOCA phases refill, peak pressure, long term HIGH for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Test Results The bases for the effects of intercompartment flow are developed using HDR, BFMC, and NUPEC tests (Reference 1, Appendix 9C).
- baling Results in Reference 2, Section 8.5, the work PI group values for steel and jacketed concrete show that mass transfer is a dominant intemal containment phenomenon. Since most of the steel and concrete are located below deck, and since steam can access these heat sinks only by intercompartment flow, it is concluded thtt intercompartment flow is an important parameter for mass transfer.
Sensitivity Studies Sensitivity stud + tow that the details of the flow path (location, elevation, loss coefficient) are not very important by comparing the blowdown response for one-node and multi node models (Reference 1, Section 8). Note that intercompartment flow paths must be modeled to avoid artificially pressurizing the break compartnwnt even though the details of flow paths do not dominate the pressure response during blowdowm. Evaluations and sensitivities (Reference 1, Table 91) were performed for various assumed circulation patterns which lead to a range of steam concentration transients PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692nmwp1M01497 October 1997
4-40 in the various compartments during a LOCA. Since internal heat sinhs saturate well before the time of peak pressure, the effects of varying the rate of heat removal by internal heat sinks did not sigrcticantly affect peak pressure or long term pressure for the LOCA (Reference 1, Section 9.3.2.3 and 9.3.2.4). > Since the MSLB transient is essentially over before the PCS becomes highly effective, internal heat sink condensation rates are more important for MSLB than for LOCA.
- Expert Review The internal and external experts provided differing opinions on the ranking for this phenomenon as described in Appendix A. One expert stated that intercompartmental flow was of liigh importance for blowdown (LOCA and MSLB) since it establishes the " initial conditions" for the remainder of the transient, but of Low importance for the remaining time periods of LOCA. Other experts did not feel that this initial condition was important. All agreed that details of flow paths were not important during blowdown since volume compliance or energy stored in containment volume was dominant and the time constant for heat sinks is long compared to the 30 second LOCA blowdown. It was agreed that gas content (steam or air in subsection 4.4.2A) was important, but the rate at which it moves is not as important relative to the time to reach peak pressure.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja,c 1 PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692nortwptib101497 October 1997
/
4 41 JUSTIFICATION OF EVALUATION hiODEL TREATMENT OF PIIENOMENON (Test Experience, hiodeling Guidance, Sensitivity Studies) ( 4 t ja,e s PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692nortwpf4101497 October 1997
4-42 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o;\3692rmwpf.1b 101497 October 1997
' 4-43 4.4.2C Containment Volume Gas Compliance 4
4 The mass and energy that enter the internal containment volume are stored within the gas , volume and cause the gas pressure, temperature, and density to increase. Storage within the j gas volume is reduced by condensation and energy transfer to the heat sinks and to the i containment shell. As internal gas pressure increases, the energy stored within the gas ! increases. The storage of a portion of the delb ered work and energy within the gas increases l its intemal energy, and the energy change due to a pressure increase can be called " gas compliance " The term " gas compliance" is used to distinguish the con:ept from the well known gas compressibility discuesed in most thermodynamic texts. Gas compliance is a capacitance term, and as such, does not increase or decrease pressure, but only reduces the i rate of enange of pressure in response to the addition of energy to the volume, i PIRT RANKING ' HIGH for all LOCA phases . HIGH for MSLB 1 l BASIS FOR PIRT RANKING (Test Results, Scaling Resu%, Sent;dvity Studies, Expert Review) i i
- Test Results t
All tests that include a pressurized vessel also include gas compliance The volume l used in test comparisons directly affects the resulting predicted peak pressure.
- Scaling Results i
i During the LOCA blowdown period, pressure scaling shows that the energy stored by j the gas accounts for a significant fraction of the break source. Although rates of l pressure change, other than during LOCA blowdown, are much slower, the pressure i PI group from scaling is relatively constant. The pressure PI group results show that j the ratio of gas compliance to the product of the source work and time constant l remains relatively constant for all accident phar,es. Therefore, the importance of gas p compliance relative to the source end system time constant remains high for all accident phases, and is ranked High. The scaling results (Reference 2, Section 8.5) l provided pressure scaling PI values (PI pg ) of approximately 0.76 for all time i periods, l F i e Sensitivity Studies
- Although specific sensitivity cases with variation of initial volume have not been performed, pressure it directly proportional to volume as can be seen from the ideal l PHENOMENA IDENTIFICATION AND RANKING Revition 1 l o:\3692non.wpl16101497 October 1997 l
444 gas law, Sensitivity cases show initial pressure to be approximately a direct adder to tlw value of the predicted pressure (Reference 1, Section 5.4).
- Expert Review .
The expert review sununarized in Appendix A provided a ranking of High for all time periods HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I j.e JUSTIFICATION OF EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,e PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\%mmettb.101597 October 1997
4-45 s EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS ( Ja,c I ja,e PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:u692non wpt.1b-101597 October 1997
4-46 4.4.2D Fog in the containment Volume l
'he steam flow velocity in the co:d leg pipe during the blowdown phase of a DECLG LOCA {
is high enough to entrain a large fraction of the available liquid and disperse it as fog, or ! very small drops. The velocity following 1.OCA blowdown is not high enough to create and entrain much liquid, so a negligible amount of the liquid will be converted to drops after blowdown. The drops are expected to persist for as long as hours, so all LOCA time phases are affected, although drops delivered during blowdown are expected to settle out before 24 hours have elapsed. An evaluation has been performed for drop flashing during blowdown
- and evaporation of drops due to post blowdown containment temperature chenges (Reference 2, Section 7.1). Results showed that the maximum evaporation leads to only about a 5% reducMon in drop diamc'er, fer a drop being reduced from its temperature to the minimum system temperature of 120'F, based on the assumption of 50% of the break liquid mass being small drops. Ther efore, drops, if delivered, will exist until settling or wall
- i deentrainment reduces the drop content.
During blowdown, the maximum mass flow rate for drop production is estimated to be approximately 0.5 times the stram mass flow rate, so the steam / drop rnixture density can be approximated as 1.5 times tae steam density, or r = 1.5 PM/RT = 1.5 x P x 18/(10.73 x T). The resulting mixture density is thui 1.5 x 18/29 = 0.93 times as dense as pure air at the same total pressure and temperature. Such a foggy source mixture may be higher in density than the mixture already in contain nent. Therefore, the blowdown source jet may not be as buoyant if it contains fog.14 0wev.r, LOCA blowdown flow rates are high enough to pressurize a com},artmeat, so the fic,w distribution during blowdown will be govemed by loss coefficients though openings exiting the break compartment, not the source densitv. Therefore, the break buoyery does not drive circulation during blowdown. The heat capacity of the foggy source mixture is significantly increaseJ over that of steam alone. PIRT RANKING LOW for LOCA phase blowd'av n HIGli for LOCA phases - refill, peak pressure, long term N/A for MSLB BASIS FOR PlRT RANKING (Test Results, Scaling Results, Sensithity SNdies, Expert Rcview)
- Sensi'.vity Studies The effect of a range of drop diameters on thermal performance relative to contairunent pressure has been studied (Reference 1, Section 5.8 ) and the limiting assum} tion of small drop diameter was chosen based on its impact on calculated PHENOMENA IDENTIFICATION AND RANKING Revision 1
, o:\36 mon *rf Ib.101497 October 1997 . _ ~ _-
. _ _ .. _ _ m. . _ . _ _ _ _ _ _ _ _ _ . _ _ . ___ _ . _ ._ _ ..__ _
4 ' 4-47 containment peak pressure. Since well accepted phenomenological models are not available to predict the mass and size of drops created during blowdown, sensithities have been performed for a range of liquid assumed to be converted to drops during blowdown. Results have been used to assess the thermal effect of flashing drops during blowdown (Reference 1 Section 5.8), as well as the effect of drops on ! circulation in the evaluation model (Reference 1 Section 9.3.2.6). The fraction of break flow liquid which forms drops during blowdown in AP600 is likely to be much greater than 5 y treent. For that fraction or greater, the amount of drops does not significantly affect the pressure transient, consistent with a Low ranking for thermal effects. Expert Review The intemal experts agreed that fog in the containment has a Low important i ranking for all LOCA time periods. The extemal experts provided a ranking of High for the LOCA blowdown phase as summarized below, and in Appendix A. The extemal experts' basis for a High ranking during LOCA blowdown is that since fluid issuing from the DBA bre ik is postulated to act as the fog source (broad drop size spectmm) and since fog thermodynamic and thermal properties are significantly different from steam or a steam / air mixture, a High ranking for blowdown should be assig,ned since these properties could significantly lower the containment pressure history. (Note that this issue is addressed with the High rank during LOCA blowdown for droplet / liquid flashing, as described in subsection 4.4.1E.) F HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ ja,e PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\s692nortwpl:lb-101497 October 1997
a 448 JUSTIFICATION OF EVALUATION MODEL TREATME OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) { s I l I r 1 ja,c i PHENOMENA IDENTIFICATION AND RANKING Revision 1
.. o:\3692nmwpt.1b-101497 October 1997
4-49 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692nortwpf:1b101497 October 1997
4 50 4.4.2E Hydrogen Release Following a LOCA, hydrogen may be postulated to be released to the reactor containment atmosphere by: Reaction of the zirconium fuel cladding with water Corrosion of materials of construction Release of the hydrogen contained in the reactor coolant system
- Radiolysis of water The telease,of hydrogen has two effects on containment pressure. The partial pressure of hydrogen gives a direct increase in total pressure due to the addition of mass to the gas, and an indirect increase in pressure due to the noncondensible degradation of condensation rates.
Hydrogen production from zirconium water production during a design basis LOCA or MSLB is not expected. Design basis analysis criteria for LOCA and MSLB preclude fuel failure, and limit worst case fuel cladding temperatures below that for which zirconium water reactions would occur. For conservatism, LOCA M&E releases include the energy associated with zirconium water reaction of 1 percent of the cladding material in the core; however, hydrogen is not assumed to be released into the containment gas volume. Production of hydrogen from corrosion of aluminum and zine occurs in the environment inside containment following a postulated LOCA, ar.d is a function of the pH of the water contacting the metals. Corrosion is a relatively slow process of hydrogen gent.ation at the marginally acid pH of the blowdo.vn and ufety injection water, requiring a period of days to generate signincant quantities. Following a LOCA, the production of hydrogen from corrosion has been conservatively estimated to be 2990 standard cubic feet (scf) over the first 20 minutes, and 5380 scf over the firct 24 hours. Dissolved hydrogen is contained in the primary coolant and in the vapor space above the pressurlier liquid due to the hydrogen overpressure maintained by the chemical and volume control system. Dissolved hydrogen is conservatively estimated to be 1170 scf. Most, but not all, of the pre-accident liquid inventory is released during the LOCA blcwdown. A conservative evaluation assumes that all 1170 scf of the hydrogen dissolved in the primary coolant and contained in the pressurizer vapor space is released to containment during blowdown. Hydrogen from radiolysis of water is considered to be generated due to the radiation field in the core. After the blowdown, the passive core cooling system (PXS) refills the reactor vessel and maintains the water level in the core. Radiolytic hydrogen generation continues in the core due only to gamma radiation from radioactive decay. A scry conservative evaluation gives a hydrogen production of 280 scf over the first 20 minutes and 5400 sci during the first 24 hours. PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692nenwpf:1b 10H97 October 1997
4 51 The above evaluations consider only conservatively high sources, and ignore significant sinks, which would reduce the amount of hydrogen actually delivered to the containment atmosphere, due to: Solubuity of radiolysis products in water in the break pool, liquid films, and drops Recombination or reaction of the chemically reactive hydrogen gas with the primary system piping and other containment materials and surfaces
- Recombiner operation The above conservative estimates of postulated sources of hydrogen during a DBA LOCA are summarized in the following tr.ble. Comparison is provided to the pressure increase represented by the mass addition, as well as the relative increase in number of moles of noncondensibles to assess the potential impact on condensation rates.
Table 4 2 Summary of Conservative Estimates of Postulated Sources of Hydrogen During a Containment Pressure DEA LOCA Integrated Total I!ydrogen Release from Source Up to indicated Time (scf) Source 20 minutes (peak pressure) 24 hours Corrosion 2900 5380 Initial solution 1170 1170 Radiolysis 280 5400 Totals 4440 11950 Assessment of Effect on Containment Pressure Predictions Time of Evaluation 20 minutes 24 hours Partial pressure increase due to mass 0.05 psi 0.14 psi addition at containment conditions increase in number of moles of 0.3 % 0.9 % noncondensibles PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA3692nortwphib 101497 - October 1997 j I I
6 53 The hydrogen sinks significantly reduce the amount of hydrogen actually delivered to the containment atmosphere. Because hydrogen sources are small and are further reduced by the sinks,it is concluded that hydrogen is not significant during any phase of a containment pressure design basis LOCA. Therefore, hydrogen is neglected in the mass releases, both for the mass effect on pressure and the noncondensible effect on condensation rates during a design basis containment pressure calculation. It should be noted that the LOCA evaluation model energy released to containment includes, as a conservatism, an amount of energy equal to that resulting from reaction of 1 percent of the zirconium in the active fuel region, but no hydrogen mass is assumed to be added to containment from the break. There is also no significant source of hydrogen from the secondary side during a htSLB. Hydrogen from sources on the primary side has no significant path into the secondary side, so they are not considered significant sources for an htSLB. Since the secondary side does not utilize borated water, corrosion rates are less than for the LOCA, and the htSLB is over by 500 seconds, limiting the total production of hydrogen. PIRT RANKING LOW for all LOCA phases N/A for hiSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert , Review)
- Test Results The LST results (Reference 38, Section 2.5) for tests 217.1,218.1, and 219.1 showed that the addition of helium into the simulated containment vessel well beyond DBA concentrations (up to 20% by volume) affected the heat removal rate as predicted by the non-condensable partial pressure in the mass transfer correlation. This information supports a ranklag of Low importance for all time periods of the LOCA trarcient and Not Applicable for hiSLB.
- Expert Review The experts agreed that the hydrogen release was of Low importance for the LOCA transient and Not Applicable for the htSLB, as summarized in Appendix A.
PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wptib 101497 October 1997 i
4-53 POW PIIENOMENON IS IMPLEMENTED IN EV.iLUATION MODEL (e.g., boundary ondition, correlation, code option, noding, input parametert I j.e JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON nest Experience, Modeling Guldr.nce, Sensitivity Studies) I jax EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I l PHENOMENA IDENTIFICATION AND RANKING Revision 1 a:\3692newpt:1b.101497 - October 1997
I 4-54 4.4.3 Containment Solid Heat Sinks 4.4.3A Liquid 1 N.n Energy Transport on Containment Heat Sinks Liquid films form due to the condensation on concrete and steel heat sinks inside containment. . Liquid films flow under the influence of gravity and shear forces due to extemal gas flows. The internal hquid film carries away the liquid enthalpy (hr) that accounts for approximately 15 percent of the enthalpy of the condensed steam (h g) based on the convention of zero htemal energy at the triple point of water. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results Th Oe : results s (Reference 2, Section 8.4) confirmed that the liquid film energy treer or *.he containment heat sinks was of Low importance for all time periods with tn2 sum of energy scaling PI .alues (PI ,, g,,, + PI ,, g, ce + PI ,, g, p ) between 0.0 and 0.08 for di phases of the transient. Expert Review Expert' agreed that film energy was an element of the condensation process and shoulo be combined into the discussion on condensation (items 3A,3B should be included in 3F to be consistent with traditional definition of condensation which includes the film resistance along with mass transfer resistance). HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) { ja,c l PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf:1b.101497 October 1997
4 55 > JUSTIFICATION OF EVALUATION MODEL TREATMENT OF FHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) [- S,C EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS ( ja,e i t i PHENOMENA IDEN1RICATION AND RANKING Revision 1 c:\3692nortwpf-1b 101497 October 1997
4-56 4.4.3B Vertical Film Conduction on Containment Heat Sinks Liquid films on concrete and steel surfaces inclined more than a few degrees flow fast enough to limit the film thickness to approximately 0.005 to 0.10 in. This results in a relatively high heat transfer coefficient so the rates of heat transfer to structures are limited by either the structure intemal resistance or the mass transfer coefficient, but not by the film. The relative magnitude of energy transfer resistance on the shell are summarized in the conductance scaling analysis (Reference 2, Section 8.2). PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR FIRT RANKING (Test Results, Scaling Results, Sensitisity Studies, Expert Resiew)
- Test Results The University of Wisconsin condensation tests (Reference 32, Section 3) and Chun and Seban data indicate that the effect of a condensate film on the heat transfer rates was negligible over the range of parameters tested due to the very small film thickness. This information supports a ranking of Low for all time periods.
Expert Review The experts all agreed that the conduction through vertical films on the solid heat sinks was of Low importance for all time periods as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e g., boundary condition, correlation, code option, noding, input parameter) I ja,c PHENOMENA IDENflFICATION AND RANKING Revision 1 au6 mon.wpe.n>.101497 October 1997
L 4-57 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) [ ja.c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I
~
ja.c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692nonwpf;1b-101497 October 1997
4-58 4.4.3C Horizontal Film Conduction on Containment Heat Sinks Liquid films on surfaces facing up that are horizontal, or nearly so, can develop rather thick films (greater than 0.05 in.) that may be more limiting to heat transfer than the heat sink intemal resistance or the mass transfer coefficient. Horizontal surfaces facing down, with the inorganic zine coating, experience film flow for slopes greater than 1 degree from horizontal. Slopes less than 1 degree drip with water and have heat transfer coefficients greater than 1000 BTU /hr-ft2 ,.F. The condensation rate during blowdown will develop a film thickness of less than 0.01 in. Combined with the short time for drops settling, the film thickness will remain less than 0.05 in. PIRT RANKING LOW for LOCA phase - blowdown HIGH for LOCA phases - refill, peak pressure, and long term HIGH for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sersitisity Studies, Expert Review) Scaling Results At 24 hours there is no condensation, so the film will either drain away or evaporate, in which case the film is nonexistent. During post-blowdown LOCA, and MSLB, condensation and/or rainout of drops may cause the film to reach a thickness sufficient to degrade heat transfer into the surface. Therefore, this phenomenon is ranked High for these phases. Expert Review The internal experts concluded that conduction through the horizontal film on the solid heat sinks was of Low importance for all time periods since the film would reach the saturation temperature very quickly. However, the external experts concluded that this phenomenon was of High importance for all time periods except for the short LOCA blowdown period on the basis that the horizontal film can become thick enough to retard heat transfer into horizontal surfaces. Based upon this information, this phenomenon was ranked Low for the LOCA blowdowm period and High for other LOCA phases and MSLB. PHENOMENA IDENTIFICATION AND RANKING Revision 1 o;\3692nonwpf4101497 October 1997 l
4 59 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja.c JUSTIFICATION OF EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) l 3a,c EVALUATION MODEL TREATMENT OF UNCEPTAINTY, DISTORTIONS I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpt:1b.101497 October 1997
4-60 4.4.3D Internal lleat Sink Conduction During the LOCA blowdown, most of the mass and energy release is stored in the containment gas volume. As containment continues to pressurize prior to the initiation of significant cooling by external evaporation, pressure mitigation is primarily by heat transfer to intemal heat sinks. From the point of view of containment gas, heat sink effects are small relative to the source and could effectively be neglected during LOCA blowdown. From the point of view of the heat sink, the heat sink temperature increases during blowdown (about 15 percent of their temperature increase at peak pressure). Thus, the heat absorbed by heat sinks during blowdown must be tracked to provide an appropriate heat sink temperature during later phases. Once evaporative cooling begins to be effective, the PCS heat removal becomes dominant and results in the pressure turning around at the time of peak pressure. In the long term, internal heat sinks saturate and are less effective than the PCS. Internal heat sinks are one of the sources for removing energy from containment by condensation, and their intemal conduction limits the amount of condensation that can occur on their surfaces. PIRT RANKING MEDIUM for LOCA phases - blowdown, long term HIGH for LOCA phases - refill, peak pressure HIGH for MSLB BASIS FOR FIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Rt iew) Scaling Results The scaling results in Reference 2, Section 8.S were used to calculate the sum of the PI values applicable for this phenomenon (PI , pwork, st + PI ,pwork. cc + PI ,pwo,g, jc ). The calculated values of 0.08,1.95,0.94, and 0.29, for the four LOCA phases, and 0.64 for MSLB support the rankings assigned. Sensitivity Studies A sensitivity to the steel jacket to concrete air gap is shown in Reference 1, Section 10, which indicates that the effect of small changes in intemal heat sink energy conduction rate does not significantly affect predicted peak pressure because net PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA3692non.wpf-1b-101497 l October 1997
4-61 internal heat sinks energy removal rate is second order well before the time of peak l pressure (Reference 1, Section 9.3.2.3).
- Expert Review The experts provided differing opinions on the ranking for th's phenomenon as summarized in Appendix A except for the LOCA refill period and the MSLB blowdown period which were both ranked as High. Expert 8 stated that initially the heat sinks play an important role (in blowdown and refill) but they decrease in importance as the heat sinks become saturated (peak pressure and long term). Expert 8 felt that heat sinks are the only mitigating feature (besides containment volume), so it must be ranked High for blowdown. Experts 5-7 felt that the heat sinks were of Medium importance during blowdown. The final ranking is based on engineering judgment applied to the scaling, sensitivity, and expert review.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ jae JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) {. Ja,c PHENOMENA IDEN1u1 CATION AND RANKING Revision 1 oA3692non.wpf:lb-101497 October 1997
4-62 - I ja,c EVALUATIDN MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wptib-101497 October 197
4-63 4.4.3E Heat Capacity of Containment Heat Sinks The heat capacity of solid heat sinks within containment is a function of the material specific heat capacity, the mass of heat sinks, and the temperature increase in the source. The initial heat sink temperature affects the amount of total heat that can be removed by a heat sink for a given increase in containment temperature. Internal heat sinks are one of the sources for removing energy from containment by condensation, and their heat capacity affects their temperature (and thus the amount of condensation on their surfaces). PIRT RANKING MEDIUM for LOCA phases - blowdown, long term HIGH for LOCA phases - refill, peak pressure HIGH for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Sensitivity Studies A sensitivity to the steel jacket to concrete air gap is shown in Reference 1, Section 10, which indicates that the effect of small changes in internal heat sink energy conduction rate does not significantly affect predicted peak pressure because net internal heat sinks energy removal rate is second order well before the time of peak pressure (Reference 1, Section 9.3.2.3).
- Expert Review All experts agreed that the heat capacity of internal heat sinks should be rated the same as conduction (subsection 4.4.3D).
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ ga,c PHENOMENA IDENiuICATION AND RANKING Revision 1 c:\3692non.wpf:1b-101497 October 1997
444 i
}a,e JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies)
{ ja,c EVALUATION MODEL TREATMENT OF UNCERTANTY, DISTORTIONS' I ja,c PHENOMENA IDENiw1 CATION AND RANKING Revision 1 c:\36 mon.wpf:1b-101497 October 1997
4 ; 4.4.3F Condensation on Containment Heat Sinks L Condensation is a boundary layer diffusion process that is driven by a steam partial density gradient. Condensation removes the gas enthalpy (hg) from the atmosphere, transfers the heat of formation from the gas to the liquid (hg- h) fto the heat sink, and leaves behind the-liquid enthalpy (h)f with the condensate. PIRT RANKING MEDIUM for LOCA phases - blowdown, long term HIGH for LOCA phases - refill, peak pressure HIGH for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Test Results The vclidation of the condensation mass transfer correlation used on inner shell surfaces included an evaluation of LST data as described in Reference 9, Section 3.9. Using measured values of total pressure, wall heat flux, wall surface and internal fluid temperatures, and air partial pressures, heat and mass transfer coefficients were derived and the results for condensation were compared to the correlation. Results showed that condensation accounted for about 80 to 95 percent of the wall heat flux. The test range covered internal conditions representative of AP600 post-blowdown quasi-steady heat and mass transfer conditions. Since internal heat sinks have a significant surface area, condensation on them can affect containment pressure.
- Scaling Results The scaling results in Reference 2, Section 8.5 were used to calculate the sum of the PI values applicable to this phenomenon (PIp, worg ,, + PI p, worg cc + PIp, wo,g ;c ). The ;
calculated values of 0.08,1.95,0.94, and 029, for the four LOCA phases, and 0.64 for -! MSLB support the assigned rankmgs. Sensitivity Studies Reference 1, Section 9.3.2.1 shows that even removing all heat sinks during blowdowm only affects containment pressure by 3.5 psi. The sensitivity study eliminated not - PHENOMENA IDEN11HCAT10N AND RANKING - Revision 1 oM692nonwpf:1b-101497 October 1997 l l
446 only the dominant process of condensation heat transfer, but also the lesser processes cf convection and radiant heat transfer.
- Expert Review All experts agreed that the heat capacity of internal heat sinks should be rated same as conduction (subsection 4.4.3D).
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ Ja,e s JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,c PHENOMENA IDriN1w1 CATION AND RANKING Revision 1 c:\3692non.wptib 101497 October 1997
4-67 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS - [ Ja,e PHENOMENA IDEN 11HCATION AND RANKING Revision 1 o:\%92non.wpf:1b101497 October 1997
4-68 4.4.3G Convection From Containment Volume Convective heat transfer is a boundary layer conduction process that is driven by a temperature gradient in the presence of a flowing bulk fluid. The bulk fluid motion may be due to a state of forced convection, free convection (wall layer ), or a combination of both. PIRT RANKING LOW for LOCA phases - blowdown, peak pressure, long term MEDIUM for LOCA phase - refill LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results The scaling results in Reference 2, Section 8.5 were used to calculate the sum of PI values applicable to this phenomenon (PIp , q, g + PI p, q, cc + PI p q y ). The calculated values of 0.01,0.33,0.12, and 0.02 for the four LOCA p'hases, and 0.15 for MSLB support the assigned rankings. These PI vanes, which include both convection and radiation heat transfer, are roughly 50% attributed to convection.
- Expert Review The external and internal experts had differing opinions on the ranking for this phenomenon for the refill, peak pressure, and long term LOCA periods as summartzed in Appendix A. However, the experts agreed on a Low ranking for the LOCA blowdown phase and MSLB. All experts agreed that convection to heat sinks was less important than condensation. The final ranking was assigned based on consideration of the scaling results, the expert reviewer comments, and engineering judgment.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf:1b-101497 October 1997
4-69 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test - Experience, Modeling Guidance, Sensitivity Studies) I ja c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,e k PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692rmwpt:1b-101497 October 1997
4-70 4.4.3H Radiation From Containment Volume to Containment Heat Sinks Radiation from the containment air / steam mixture to the solid heat sinks can be significant when the product of the steam partial pressure and radiation beam length are of the order of 1 ft. atmosphere. This is the case inside containment where steam partial pressures nay be as high as 3 yanospheres and beam lengths are frequently greater than 10 ft. Radiation is also enhanced by high emissivity surfaces. The liquid film surfaces have emissivity of 0.95 to 0.96 (Reference 22, pg. 216), the inorganic zine paint emissivity is 0.90 to 0.95 (Reference 23), and the concrete emissivity is approximately 0.90 (Reference 24). PIRT RANKING LOW for LOCA phases - blowdown, peak pressure, long term MEDIUM for LOCA phase - refill LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scalmg Results, Sensitivi+y Studies, Expert Review) Scaling Results The scaling results in Reference 2, Section 8.5 were used to calculate the sum of the PI values applicable to this phenomenon (PIp , q, + PI ,pq, ce + PI p, qge ). The calculated values of 0.01,0.33,0.12, and 0.02 for the four LOCA phases, and 0.15 for MSLB support the assigned rankings. These PI values, which include both convection and radiation heat transfer, are roughly 50% attributed to radiation. Based on the scaling analysis results, it was determined that radiation to the internal heat sinks has a much smaller effect on containment pressure reduction than condensation. However, it is not negligible for the LOCA refili phase. Therefore, it is ranked Low a for the blowdown, peak pressure, and long term phases of LOCA and Low for MSLB, and ranked Medium for the LOCA refill phase. Expert Review The external and internal experts had differing opinions on the ranking for this phenomenon for the refill, peak pressure, and long term LOCA periods as summarized in Appendix A. However, the experts agreed on ranking for the blowdown periods for both LOCA and MSLB as being of Low importance. All experts agreed that radiation to heat sinks was less important than condensation. The final ranking is based on the xaling results described above. PHENOMENA IDENUFICATION AND RANKING Revision 1 l o:\3692nortwpf:1b-101497 October 1997 l
4 71 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter). I ja,e JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) [ ja,e EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS [ ja,c I PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692non.wpt.1b-101497 October 1997
4-72 4.4.4 Initial Conditions Within ContaMment i The initial conditions within the containment may have an important effect on the containment pressure response. These conditions, which include temperature, humidity, and pressure, affect the capability of the solid, liquid, and gaseous heat sinks in containment to absorb energy from the break. Sensitivity studies showed that these three initial conditions had an efket on the peak and long term containment pressure for a LOCA event, although a smaller effect on blowdown and refill phases. The MSLE is mitigated primarily by condensation on intemal heat sinks, as well as the containmen; shell inner surface. Therefore, initial conditions for MSLB would have similu importance as for the LOCA peak pressure. 4.4.4A Initial Temperature in Containment PIRT RANKING MEDIUM for all LOCA phases MEDIUM for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensithity Studies, Expert Review) Sensitivity Studies The sensitivity results (Reference 1, Section 5.5) showed that a decrease in the initial temperature of both 6.ternal volumes and structures from 120*F to 50*F subsequently decreased the LOCA peak pressure by approximately 1.4 psi, and the worst case MSLB peak pressure by 0.2 psi. This information supports a rankmg of Medium for all time periods since the initial temperature affects both the initial time period as well as the subsequent time periods. Expert Review
'1he experts had differing opinions on the rankmg for this phenomenon for the blowdown, peak pressure, and long term LOCA phases as summarized in Appendix A. However, the experts agreed on the ranking for LOCA refill period as Medium and for MSLB blowdown as High.
PHENOMENA IDENTIFICATION AND RANKING Revision 1 l o:\3692non wpElb-101497 October 1997
4 73 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL -(e.g., boundary S condition, correlation, code option, noding, input parameter) a,C
- JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experienet , Modeling Guidance, Sensitivity Studies)
{ . Ja,c PHENOMENA IDEN11HCATION AND RANKING Revnion '. oA3692non.wpf:1b-101497 October 1997
. 4 .
EVALUATION MODEC TREATMENT OF UNCERTAINTY,- DISTORTIONS I ja.c ( a PHENOMENA IDEN1wlCATION AND RANKING Revision 1
- c:\3692nmwpt.1b.101497 October 1997
4 75 L 4.4.4B Initial Humidity in Containment PIRT RANKING - } MEDIUM for all LOCA phases MEDIUM for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Sensitivity Studies The sensitivity results (Reference 1, Section 5.3) showed that an increase in the initial humidity from 0% to 100% subsequently decreased the LOCA peak pressure by approximately 1.3 psi, and the worst case MSLB peak pressure by 1.1 psi. This information supports a Medium ranking for all time pericds since the initial humidity affects both the initial time period as well as the subsequent time periods.
- Expert Review The external and internal experts had differing opinions on the ranking for this phenomenon for the blowdown, peak pressure, and long term LOCA periods as summarized in Appendix A. However, the experts agreed on the ranking for LOCA refill period as Medium and for MSLB blowdown as High. The final ranking is based
- on the results of the sensitivity studies described above.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja.c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) { PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692non.wpf:1b-101497 October 1997
4-76 ja,c t EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c C PHENOMENA IDENIwlCATION AND RANKING Revision 1 o:\3692non.wpf:1b101497 October 1997
4-77 4.4.4C Initial Pressure in Containment PIRT RANKING' ' MEDIUM for all LOCA phases MEDIUM for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Sensitivity Studies The sensitivity results (Reference 1, Section 5.4) showed that a decrease in the initial . pressure from 15.7 psia to 14.5 psia subsequently decreased the LOCA peak pressure by approximately 1.7 psi, and the worst case MSLB peak pressure by 1.8 psi. 'lhls information supports a Medium ranking for all time periods since the initial pressure affects both the initial time period as well as the subsequent periods.
- Expert Review The external and internal experts had differing opinions on the ranking for this phenomenon for the blowdown. peak pressure, and long term LOCA periods as summarized in Appendix A. However, the experts agreed on the ranking for LOCA refill period as Medium and for MSLB blowdown as High. The final ranking is based ..
on the results of the sensitivity stud!es described above. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ l*# JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test _ Experience, Modeling Guidance, Sensitivity Studies) [ jax PHENOMENA IDENTIFICATION AND RANKING Revision 1
- o:u6 mon.wpalb.101497 October 1997
4-78 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ju PHENOMENA IDEN11MCATION AND RANKING Revisionl o.\36 mon.wpf:1b-in1497 October 1997
4-79 4.4.5 Break Pool 4.4.5A Break Pool Circulation / Stratification For the LOCA event, the containment pressure interactions with the break pool do not become significant until the beginning of the long term phase. For the MSLB event, most of the released mass condensate drains into the IRWST. The break source is largely superheated, so the break liquid to the pool is not considered significant. PIRT RANKING LOW for LOCA phases - blowdown, refill, peak pressure MEDIUM for LOCA phase - long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sencitivity Studies, Expert Review) Scaling Results The scaling results (Reference 2, Section 8.5) showed that the pool is not a significant source ever. though the PI group is biased to maximize the surface temperature by assuming it tc be stratified. Also, there are no significant driving forces for mixing in the pool. These conclusions are based on the assumption of a thermally stratified fluid which is physically reasonable and maximizes the pool as an energy source for containment. Expert Review The experts all agreed that the mixing in the break pool was of Low importance for all time periods except for the LOCA long term which was ranked as Medium as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, a,rrelation, code option, noding, input parameter) [ ja.c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf:1t>.101497 October 1997
4-80 I ja,c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) { ja,e EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS [ ja.c PHENOMENA IDENTIFICATION AND RANKING Revision 1 l o:\3692non.wpMb-101497 October 1997
4-81 4.4.5B Break Pool Condensation / Evaporation The break pool which contains saturated water may function as either a heat sink that condenses some of the steam in containment, or as a heat and mass source that provides additional steam flow into the containment. Based upon the results from the scaling analyses, the break pool is a minor heat and mass source. The evaporation from the break pool provides a small increase in the containment pressure for most of the LOCA transient. PIRT RANKING LOW for LOCA phases - blowdown, refill, and peak pressure MEDIUM for LOCA phase -long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) 1 Scaling Results , The scaling results'(Reference 2, Section 8.S) showed that the PI values p (PI , weg p) were less than 0.07 for the time periods of interest assuming the pool to be stratified. Since the pool represents approximately a 7 percent increase in the containment - pressure rate during the peak pressure period, it was ranked as Medium for the LOCA long term phase and ranked Low for the blowdown, refill, and peak pressure LOCA phases, and Low for the MSLB event.
- Expert Review The experts had differing opinions on the ranking of this phenomenon as summarized in Appendix A, although all agreed that the condensation on the break pool was of
- Low importance for the LOCA blowdown and refill periods and MSLB. The final '
ranking is based on the results of the scaling analyses described above.. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) i I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf lb-101497 October 1997
4-82 JUSTIFICATION FOR EVALUATION h10 DEL TREAThiENT OF PilENOhtENON (Test Experience, hiodeling Guidance, Sensitivity Studies) l K J k l' ' EVALUATION hiODEL TREAT iENT OF UNCERTAINTY, DISTORTIONS u L I Rt ir 1 l*# s
~ ~ ,^ PHENOMENA IDENTIFICATION AND RANKING Revision 1 o'\3692non.wpf:1h101497 October 1997
4-83 4.4.5C Break Pool Convection IIcat Transfer within Containment Volume The convective heat transfer between the containnwn' steam / alt mixture and the break pool is an order of magnitude less than mass transfer, and is thus insignificant for all time phases of the LOCA event as well as for the hiSLB event. PIRT RANKING LOW for all LOCA phases LOW for MSLB UASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Scaling Results The scaling results (Reference 2, Section 8.5) provided P1 values (Plp, q, p) of 0.0 for the time periods of interest. This information supports a ranking of Low for all time periods for the convection and radiation between the containment and the break pool.
- Expert Review The experts all agreed that convecti;n heat transfer between containment and break pool was of Low importance for all time periods as summarized in Appendix A.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, ccdc option, noding, input parameter) l l [ 3a.c l l l PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf-lb 101497 October 1997
4 84 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I jax l EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS ( ja,e
- i 1
- 1 1
J PHENOMENA IDENTIFICATION AND RANKING RevWon 1 o:\3692non.wpf:1b.101497 October 1997 yy?*WT t- P41' ty 'W"'**W'W'Nt-'t='+T@-matr--' Af PD~*r' *"'w** '
- e 'y w p-w" - ' % ~ --~Wgw==rWe
i l 4-85 4.4.5D Break Pool Radiation lleat Transfer within Containment Volume i Radiation heat transfer between the containment steam / air mixture and the break pool is insignificant for all time phases of the LOCA event as well as for the htSLB event. PIRT RANKING LOW for all LOCA phases LOW for htSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Scaling Results The scaling results (Reference 2, Section 8.5) provided PI values p ( P1 , q, p ) of 0.0 for the time periods of interest. This information supports a ranidng of Low for all time periods for the convection and radiation between the containment and the break pool.
Expert Review The experts all agreed that radiation heat transfer between containment and break pool was of Low importance for all time periods as summarized in Appendix A. HOW PHENOhiENON IS IhiPLEhiEhTTED IN EVALUATION hiODEL (e.g., boundary condition, correlation, code option, noding, input parameter) {
}"
JUSTIFICATION FOR EVALUATION hiODEL TREATh1ENT OF PHENOhiENON (Test Experience, hlodeling Guidance, Sensitivity Studies) I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf.lb 101497 October 1997
.- - - - _ . - _ . . . _ . __ _ _ _ . - _ _ _ . _ - . . _ . - _ , _ - . _ _ ~ . _ . . . . . _ . . _ . . - _ . _ - _ _ . . . _ , _ - -
4 86 ' L"/ALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS l 1 1" I l PHENOMENA IDENTIFICATION AND RANKING Revision 1 a s3692nortwptib-101497 October 1997 d
1 4-87 4.4.5E Conduction in Break Pool The internal thermal resistance is a complicated function of ie stratification and circulation of incoming fluid to the pool and of the interactions between the pool and its boundaries that are also heat sinks Thus, conduction in the break pool was given the same ranking as pool circulation / stratification (see subsection 4.4.5A). PIRT RANKING LOW for LOCA pna.,es Mowdown, refill, and peak pressure MEDIUM for LOCA phase - long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review
'Ihe experts agreed that the conduction in the break pool was of Low importance for ,
all time periods as summarized in Appendix A except for the LOCA long term phase where the external experts ranked it as Medium. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, aoding, input parameter) I ja e JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) ' I jae PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non wptib 101497 October 1997
4-88 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DibTORTIONS l l l Jae l l i A IDEGFICATION AND RANKING Revision 1 October 1997
4-89 4.4.5F Compartment Filling The break pool fills with break liquid and below-deck condensate. The break pool starts from the sump at the lowest region of the containment, and as the break continues, it floods more volume in additional compartments. Enough water is collected by 15,000 seconds to close the major flow path into the steam generator compartments. Closing this major opening while a smaller opening remains open will reduce the large-scale circulation induced by the break source in the below-deck compartments. By this time however, the net effect is second order for below deck heat sinks so gas circulation below-deck may not be very effective at condensing steam. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that compartment filling was of Low importance for all time periods as summarized in Appendix A, except for the LOCA long term phase where the external experts ranked it as Medium because of the potential to block flow paths. But since the heat sink area in the filled compartments is only 11% (see Table 3-1 for dead-ended jacketed concrete areas and stairwells) of the total heat sink area, the effect of filling on heat sink utilization is small, and during the long term phase, the net effect of heat sinks are second order. Flow path blockage does affect circulation and is evaluated in Reference 1, Section 9 as it relates to subsection 4.4.2B - Intercompartment Flow. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692rmwpfib-101497 October 1997 _ _ __- J
4-90 3 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) ' I , - P t l'* i
- EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS
[ i jax PHENOMENA IDENTIFICAT10N AND RANKING Revision 1 o:\3692non.wpf.1b 101497 October 1997 v.--,- . , - . - - . . . . . . , _ _ _ , , _ _ _ _ _ , , , _ _ , _ _ ,
4-91 4.4.6 IRWST 4.4.6 A,B,C,D,E,F The operating deck floor is sloped to drain into the IRWST. The above-deck condensate drains mostly into the IRWST, while that below deck drains mostly into the break pool. The IRWST water is initially assumed to be at a temperature of 120'F. Vents are provided at the top of the IRWST, which provide a path to vent steam released by the spargers. The vents open on small pressure differentials. The condensate that drains into the IRWST comes from the shell and from condensate on the above-deck heat sinks, at a temperature and steam partial pressure lower than that of the above-deck gas but hotter than the tank water. Consequently, the added liquid is stable, and the temperature of the gas at the IRWST liquid surface will be cooler and more dense than the gas above the tank structure. Any air / steam that enters from above will condense some of the steam on the water surface, leaving behind a more stable atmosphere that resists flow interactions with the gas above the tank structure. Late in the LOCA transient, th( IRWST drains into the RCS to provide additional core cooling and therefore the liquid level decreases with time. The effects of IRWST level and subcooling of IRWST liquid on break source are addressed conservatively with the mass and energy release model. Consequently, the direct interactions between the containment volume and the IRWST volume by conduction, convection, or radiation are expected to be negligible. P RT RANKING LOW for all LOCA phases LOW for MSLB ' BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts all agreed that the IRWST phenomena were of Low importance for all time periods as summarized in Appendix A since the IRWST is somewhat isolated from the containment atmosphere.
' PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\mxnwpf:1b.101m October 1997
4-92 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary I condition, correlation, code option, noding, input parameter) t jax JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test l' Experience,'Modeling Guidance, Sensitivity Studies) I jae EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I j ax
.P i-i PHENOMENA IDEN1w1 CATION AND RANKING - Revision 1 o:\3692non.wpf 1b-101497 October 1997 -
y,wy,-..g----_.sg-s y p.,,,,., y..w----yeogr ,-yagw,g,
4 93 , .! 4.4.7 Steel Shell i , 4.4.7A Convection Heat Transfer From Containment Volume Convective heat transfer is a boundary layer conduction process that is driven by a j~ temperature gradient in the presence of a flowing bulk fluid. The bulk fluid motion may be due to a state of forced convection, free convection (wall layers), or a combination of both. l PIRT RANKING y . 1 ] LOW for all LOCA phases
- LOW for MSLB l
i , BASIS FOR PIRT RANKING (Test Results, Sca!!ng Revilts, Sensitivity Studies, Expert - l Review) 1
- Scaling Results l The scaling results in Reference 2, Section 8.5 were used to calculate the sum of the PI
{' p + PI p, g d.). The l values applicable to this phenomenon interest (PI , q,,, support the + PIcalculated re assigned - l rankings. These PI values included both convection and radiation heat transfer. The scaling analysis shows that convection from the containment steamA.ir mixture to the j steel shell has a smaller effect on containment pressure reduction than condensation. j' In combination with radiation, convection had only a 5 percent effect en containment pressure for the later time phases of the LOCA event (refill through long term). l l
- Expert Review i
l l The experts agreed that this phenomenon was ranked as Low for the LOCA and ! MSLB blowdown periods, but the intemal and extemal experts had differing opinions !- (Low vs. Medium) on the other LOCA phases, as summarized in Appendh A. The [ internal experts felt that convection and radiation to the inside of the steel shell were i: much less important than condensation on the shell and should be ranked Low for all
- time periods.
i G i j PHENOMENA IDEN~klCATION AND RANKING Revision 1 a:u692non wptib.101497 October 1997
~
1 I 4-94 4
- HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter)
I i t k
- i 3
}'<c l
)
i l l JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test ; j Experience, Modeling Guidance, SensitMty Studies) - ' [ ! l. 1 1 ja,e PHENOMENA IDENTIFICA110N AND RANKING Revision 1 oA3692nonwpf:1b-101497 . October 1997
4 95 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I jax d PHENOMENA IDENTIFICATION AND RANKING Fevision 1 a\%92rmwpf.1b 101497 October 1997 l
l 4 96 - 4.4.7B Radiation lleat Transfer from Containment Volume to Steel Shell l The scaling analysis shows that radiation from the conttirunent steam / air mixture to the steel shell had a smaller effect on containment pressure reduction than condensation. However, in combination with convection, radiation had greater than a 5 percent effect on pressure for the later time phases of the LOCA event (refill through long term). PIRT RANKING LOW for all LOCA phases LOW for MSLB BAST.S FOR PIRT RANKING (Test Results, Scaling Results, Smsitivity Studies, Expert Review) Scaling Results The scaling results in Reference 2, Sectioni 8.5 were used to calculate the sum of the Pl values applicable to this phenomenon (PIp , q, n + P1p, q,,, + P1, p q, dd h calculated values of 0.08 or less for the time periods of interest support the assigned rankings. These PI values included both convection and radiation heat transfer. Expert Review The experts agreed that this phenomenon was ranked as Low for the LOCA and MSLB blowdown periods, but the intemal and external experts had differing opinions (Low vs Medium) on the other LOCA phases, as summarized in Appendix A. The final ranking was based on the scaling results described above. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja,e PHENOMENA IDEN11FICATION AND RANKING Revision 1 c:u692nortwpt:11 101497 October 1997
4 ~)7 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PliENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I jax EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I j.e
~
PHENOMENA IDENTIFICATION AND RANKING Revision 1 'o:\3692nmwpf;1b 101497 October 1997
=
m
4-98
- 4.4.7C Condensation on Inside Containment Shell PIRT RANKLNG HIGli H all LOCA phases HIGli for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
Scaling Results The scaling results in Reference 2, Section 8.S) were used to calculate the sum of the PI values applicable to this phenomenon (PI p , wo,u + PI ,p wo,x,,, + PI , p work, ds)- The calculated values of 0.02,0.61,0.47, and 1.04 for the four LOCA phases, and 0.37 for MSLB support a High ranking for all phases except LOCA blowdown. Sensitivity Studies Sensitivity studies show that the penalties on heat and mass transfer in the evaluation model have a significant effect on pressure (Reference 1, Section 10). Expert Review The experts agreed this phenomenon should be ranked as High for the LOCA refill, peak pressure, and long team perioc'.s and for the MSLB blowdown period as summarized in Appendix A. There was one differing opinion on the rank for the LOCA blowdown period (High vs. Low). For the final ranking, a High rank was assigned for the LOCA blowdown phase based on engineering judgment. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) ( ja.c PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:We2non.wpt:1b.101497 October 1997 '
=
- -* -y-m m. w m ie ygr3-,- w g-me.ey-m,e-+m-,.e- ..--r . -- , . ;%, r
4 99 I l , i l JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, h#deling Guidance, Sensitivity Studies) ( , jax EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS l Ja,e PHENOMENA IDENTlHCATION AND RANKING - Revision 1 a:\3692newpt15101497 - October 1997
4 100 ! 4.4.7D Film Conduction on inside of Steel Shell ' Tests at the University of Wisconsin show that surfaces with slopes less than 1 degree may ; experience drops fallag from thr liquid film condensation, which has a higher heat transfer
~
coefficient than film wise condensation without drops. Since only 0.09 percent of the shell i heat transfer surface area has a slope less than 1 degree, horizontal films are not significant for the shell for any time phase. On the majority of the inner shell surface, a flowing liquid film exists with a heat transfer ! coefficient greater than 600 BTU /hr ft2 .F on the inorganic zine coated surfaces. 'this compares td a heat transfer coefficient of 50 to 100 BTU /hr ft2 ,.F for condensation onto the film. Scaling of heat transfer conductances showed that the temperature drop through the j film is small relative to the other series temperature drops, therefore, film conduction was ranked as low importance for both the MSLB and LOCA events. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Scaling Results The scaling analyses (Reference 2, Section 7A and Table 8 2) for the heat transfer in series show that the film conductances are much greater (or resistances are much smaller) than for the other conductances. 'Ihis information supports a ranking of low for all time periods for the inside film conduction.
- Expert Review '
The experts all agreed that the film conduction on the inside of the containment shell was of Low importance for all time periods as summarized in Appendix A. l PHENOMENA IDENTIFICATION AND RANKING Revision 1 oT3692nortwptib101497 October 1997
4 101 liOW PliENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correhtion, code option, noding, input parameter) I jax - JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,e EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,e PHENOMENA IDENTIFICATION AND RANKING Revision 1 oc\3692ncetwpf-1b101497.
- October 1997
I 4 103 4.4.7E Internal Filan Energy Transport on Steel Shell ! 1 Liquid films form on the inside of the containment shell due to condensation. The liquid film that forms on the internal shell is collected at the crane rail, the stiffener ring, and at the deck elevation and drained into the IRWST. The intemal liquid film carrier away the liquid enthalpy (19 ) that accounts for approximately 15 percent of the enthalpy of the condensed steam (h,) based on the convention of zero intemal energy of the triple point of water. FIRT RANKING LON for LOCA phases blowdown, refill, peak pressure MEDIUM for I OCA phase -long term LOW for MSI B BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Exyrt Review)
- Scaling Results The scaling results (Reference 2, Section 8.4) provided PI values (PI ,, g,,, + PI ,, g,,, +
PI e, f.d,) of between 0.0 and 0.11 for the four LOCA phases and MSLB, Expert Review The experts agreed ' hat the film energy transport on the inside of tle containment shell was of Medium importance for all time periods as summarized in Appendix A. Since the film energy transport on the solid heat sinks (subsection 4.4.3A) was ranked Low, this information supports the assigned rankings. liOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) Ja,e PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:u692nmwpt:tb 101m October 1997
4 103 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guldance, Sensitivity Studies) l ja,e EVALUATION MODEL TREATMENT OF UNCERTANTY, DISTORTIONS ( . }"# PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692nonw).f:1b 101497 October 1997
4 104 4.4.7F Conduction *!hrough Shell The time constant for the steel shell is approximately 5 minutes. For the LOCA blowdown and refill phases, although the shell remains nearly at its initial temperature, conduction ' through the shell affects the later phases, and therefore was ranked as High. Later in the LOCA transient (peak pressure and long term phases), the conduction heat transfer through the shell accounts for approximately 1/5 of the containment to riser energy transfer resistance based upon scaling analyses. Therefore, the shell conduction was ranked High in importance. During an MSLB, the shell acts as a steel heat sink, so conduction is ranked H.igh. PIRT RANKING HIGH for all LOCA phases HIGH for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Scaling Results With a normalized conductance of 1.0, the shell conductance is significant in comparison to the inside and outside gas-shell conductance for all but the dry shell outside (Reference 2, Table 8-2).
- Expert Review The experts had differing opinions on the ranking of this phenomenon for LOCA blowdown and refill as summarized in Appendix A, although all agreed that shell conductance was of High importance for the LOCA peak pressure and long term periods as well as for MSLB blowdown period. For the final ranking, conduction through the shell was ranked the same as condensation (subsection 4.4.7C), since conduction is the means by which condensation energy is transferred out of containment.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\%92non wpf:1b-101497 October 1997
4 105 ! l 1 f-1 i ja.c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) - I jax PHENOMENA IDflfrIFICAT10N AND RANKING Revision 1 o:\3692non.wpt:tSto 4,7 October 1997 -
410$ I j.x EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS l Ja.c PHENOMENA IDENI1FICATION AND RANKING Revision 1 o:\3692non wptib 101497 October 1997
4 107 4.4.7G lleat Capacity of Shell l The heat capacity of the coated steel shell is a function of the specific material heat capacity, the mass of the shell, and the shell temperature rise. The initial shell temperature affects the amount of total heat that can be removed by heating the steel shell during the initial pressurization. Over the longer term, the steel shell heat capacity provides a significant heat storage capability, acting to dampen perturbations of short duration. The steel shell is the path for heat transfer between the internal condensing surface and the extetnal evaporating surface. The heat capacity of the steel shell is important when the difference between the energy in and energy out are significant. PIRT RANIGNG HIGH for LOCA phases blowdown, refill, and peak pressure LOW for LOCA phase -long term HIGH for MSLB BASIS FOR PIRT RANWNG (Test Results, Scaling Results. Sensitivity Studies, Expert Review) Scaling Results The scaling results in Reference 2, Section 8.4 are used to calculate the net energy through the shell. The sum of the PI values representing energy on the inside of the shell (PI,,q, , PI,jg , , PI,,q,,, , Pl j g
, , Pl ,q.d, e , PleJgd,)less the sum of the PI values representing energy on the outside of the shell (PI,,qux , PI,,q,,x , P1,jge,x ,
PI,,qa,x) results in values of 0.02 for the LOCA blowdown period,0.62 for the LOCA refill period,0.44 for the LOCA peak pressure period,0.02 for the LOCA long term period, and 0.38 for the MSLB blowdown period. The high scaling value at the start of the refill period shows that the heat capacity at the end of blowdown is much more important than the calculated value of 0.02 would indicate. Expert Review
'Ihe experts had differing opinions on the ranking of this phenomenon as summarized in Appendix A, although all agreed that it should be ranked as Low for the LOCA long term period and High for the MSLB blowdown period. The final ranking was based on the scaling results described above.
PHENOMENA IDENTIFICATION AND RANKING Revision 1 0:\3692non.wpf:1b-101497 October 1997
4 108 i HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., ' nundary - condition, correlation, code option, noding, input parameter) I ja,e JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I jax EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,e PHENOMENA IDEN11rlCATION AND RANKING Revision 1 c:\3692nortwpf.lb-101497 October 1997
..-. - -- ,.- _..-. . n :..:._-...-.
4 109 4.4.71R Convection to Riser Annulus PIRT RANKING LOW for LOCA phases - blowdown, refill, and peak pressure MEDIUM for LOCA phase - long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Senaltivity Studies, Expert Review)
- Scaling Results The scaling results in Reference 2, Section 8.4 were used to calculate the sum of the PI values applicable to this phenomenon (Pl ,c q,,,x+ PI , eq. dsx). The calculated values of htween 0.0 and 0.M for all time periods support a Low ranking for all LOCA phases and MSLB. These PI values included both convection and radiation heat transfer.
- Expert Review The internal experts agreed that the ranking of this phenomenon should be Medium for the LOCA long term phase and Low during all oSer time periods as summarized in Appendix A. They agreed that convection to the riser and radiation to the baffle during all periods should be of Low importance except for the long term which was ranked Medium due to the highei dry fraction and, hence higher temperature of tl e shell.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I Ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 a:\%92non wpub.101497 October 1997
. _ . . . . . . . . . = . . . . .
4-110 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test / Experience, Modeling Guidance, Sensitivity Studies) I 3 Ja.c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I j a,e PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf;1b-101497 October 1997 l __ ____ _ ________.m -_. --- -
4 111 4.4.71 Radiation to the Baffle Radiation from the steel shell to the baffle cools the shell and heats the baffle above the temperature of the adjacent riser air. PlRT RANKING LOW for LOCA phases - blowdown, refill, peak pressure MEDIUM for LOCA phase - long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results The scaling results in Reference 2, Secticn 8.4 were used to calculate the sum of the PI values applicable to this phenomenon (PI ,, q, m,+ PI e . dsx + PI ,, q, i,f ). The calculated values of between 0.0 and 0.06 for all time pe.!ods support a Low ranking for all time phases. These PI values included both convection and radiation heat transfer. Expert Review The internal experts agreed that the rankmg of this phenomenon should be Medium for the long term period of tue LOCA (external experts agreed it should be Medium for peak pressure) and Low during all other time periods as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, neding, input parameter) I ja,e i PHENOMENA IDENTIFICATION AND PNG Revision 1 a.\3692nonwpf:1b 101497 October 1997
4-112- , 1
. JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies).
I 1 i t i n l j ja,c ! EVALUATION MODEL TREATMENT OF UNCERTrJNTY, DISTORTIONS i ( 1 1 - ja,e ) 1 . d i
- PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692rmwpf:1b-101497 October 1997
4 113 4.4.7J Radiation to the Chimney The dry upper dome surfaces of the steel shell can radiate to the missile shield which defines the bottom of the chimney volume. The low temperature of the liquid due to subcooled film a reduces radiation from the wet upper dome surface to the missile shield.- The radiant energy absorbed by the missile shield would be transferred to the dumney air volume by convection, adding to chimney buoyancy. This is judged to be a small effect and is thus I ranked Low for all accident phases. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANK 1NG (Test Results, Scaling Results, Sensitivity Studies, Expert
- Review)
Scaling Results The scaling results in Reference 2, Section 8.4 were used to calculate the sum of the PI values applicable to this phenomenon (PI ,, q,,,x+ PI e , q, dsx + PI ,, q, g ). The calculated values of between 0.0 and 0.06 for all time periods supports a ranking of Low for all time periods. These PI values included both convection and radiation heat transfer.
- Expert Review The experts agreed that the ranking of this phenomenon should Le Low during all periods of the LOCA, and Low for the MSLB blowdown as summarized in Appendix A.
. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter)
( )- ja,e PHENOMENA IDEN11HCATION AND RANKING Revision 1 o:u692non.wpt:1b 101497 October 1997
4-114 l JUSTIFICATION FOR EVALUATION MODEL TREATMENTT OF PHENOMENON (Test l Experience, Modeling Guidance, Sensitivity Studies) Ja c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ]ae PHENOMENA IDENI&lCATION AND RANKING Revision 1 o:\3692non.wpf Ib-101497 October 1997
4-115 4.4.7K Radiation to the Fog / Air i If drops are postulated to exist, they will capture radiation emitted by the shell that would l othenvise be deposited in the baffle. Capture of the radiant energy will raise the temperature l ' of the drops and riser gas flow, with the result the the saturation temperature increases and some of the drop vaporizes. The calculation performed to assess the effect of condensation from a supersaturated vapor (see subsection 4.4.9C) showed drop formation caused the net riser mixture density to decrease, so it might be expected that vaporization of drops leads to a density increase. That is not the case, however, since the energy source for drop vaporization is not the riser gas mixture, but rather is the shell radiation. The drops only vaporize as-required to balance the absorption of the radiant energy. Consequently, the gas mixture temperature increases, causing the mixture density to decrease, the drop mass decreases, causing the mixture density to decrease, and the gas molecular weight decreases due to the increased gas fraction, causing the mixture density to decrease. All of these increase the buoyancy and the PCS natural circulation air flow rate. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results The scaling results in Reference 2, Section 8.4 were used to calculate the sum of the PI values applicable to this phenomenon (PI ,, q,,3x + PT e. q, dsx ). The calculated values were less than 0.% for radiation from the shell, and thus will be less for radiation to the fog / air nuxture. These PI values included both convection and radiation heat transfer. This information supports a ranking of Low for all time periods. Expert Review The experts agreed that the ranking of this phenomenon should be Low during all periods of the LOCA and for the MSLB blowdown as summarized in Appendix A. PHENOMENA IDENTIFICATION AND RANKING P.evision 1 o:\%92non.wpf-Ib.101497 October 1997
l- 4 116 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I jae JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS l jax i e PHENOMENA IDENTIFICATION AND RANKING Revision 1 l OA3692non.wpf:1b101497 October 1997 j
4-117 4.4.7L Outside Film Conduction A flowing liquid film exists with a heat transfer coefficient greater than 600 BTU /hr-ft2 *F on the inorganic zine coated surfaces. This compares to a heat transfer coefficient of up to 113 2 BTU /hr-ft 'F for evaporation from the film. Scaling of heat transfer conductances showed that the temp,erature drop through the liquid film is small relative to the other series temperature drops, therefore, film conduction was ranked of Low importance for LOCA peak pressure and long term phases and MSLB. The rankmg is N/A for LOCA blowdowa and refill phases since no external liquid film is present. PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results As discussed in subsection 4.4.7D, the film conductance was much less important than shell conductance, evaporation, and condensation based upon the scaling analyses. This information supports a Low rankmg for all time periods. Expert Review The experts agreed that the ranking of this phenomenon should be Low during the peak pressure and long term periods of the LOCA and Low for the MSLB blowdown as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., bounda.y - condition, correlation, code option, noding, input parameter) I ja,e PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non wpf:1b-101497 October 1997 1
4-118-JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test
- Experience, Modeling Guidance, Sensitivity Studies)
[ Ja,c EVALUATION MODEL TREATMENT OF UNCERTAINIY, DISTORTIONS ( ja,c i PHENOMENA IDEN11HCATION AND RANKING Revision 1 o:\3ennon.wpt:1b-101497 October 1997 l
4-119 4.4.7M Outside Film Energy Transport The external liquid film is supplied at a temperature that is below that for which significant evaporation would occur (120'F assumed for safety analyses). The energy absorbed by the film and the resulting temperature increase is referred to as the subcooled heat capacity of the external film. Most of the external film is expected to evaporate, but the portion that does not evaporate runs down a drain, away from the containment shell, carrying a small portion of containment energy. PIRT RANKING N/A for LOCA phases - blowdov a and refill MEDIUM for LOCA phases - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results The scaling analysis shows that energy absorbed by sensible temperature increase of the film, accounts for up to 8 percent of the energy transferred by evaporation. The scaling analysis PI group for subcooled heat capacity, PI ,,px , has a maximum value of 0.08 at the beguunng of the long term phase and less during other time periods (Reference 2, Table 8-4).
- Expert Review The experts agreed that the ranking of this phenomenon should be Medium during the peak pressure and long term periods of the LOCA, and Low for the MSLB blowdown as summarized in Appendix A.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) { ja.c PHENOMENA IDENTIFICATION AND RANMG Revision 1 o \3692ncnwpf:1b-101497 October 1997 l
4-120 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS { ja,c c el PHENOMENA IDENTIFICATION AND RANKING Revision 1 i o:\36 mon.wpf.lb-101497 October 1997
r I 4421
- 4.4.7N Evaporation to Riser Annulus Based upon the scaling analyses and test results, evaporation is the dominant process for energy transfer from the shell and hence has a major effet on the rate of pressure change inside containment. Evaporation on the AP600 dome begir, within a few minutes of PCS drain valve actuation, as water begins to spill out of the bucket. However, evaporation is not assumed in the evaluation model to take place until later in the LOCA transient after the shell heats up and full water coverage is achieved.
PIRT RANKING N/A for LOCA raases - blowdown and refill HIGH for LCv A phases - peak pressure and long term MEDIUM for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Test Results The LST results (Reference 38, Section 2.1.2) for tests 202.2,207.3, and 208.1 showed that evaporation was much more important than convection and radiation. These tests confirm an importance rankmg of High for the time periods when the shell is wet.
Scaling Results The scaling results (Reference 2, Section 8.4) provided PI values (PIe . fg. esx )f 0.02 for the peak pressure period and 0.81 for the long term period of the LOCA transient. The high scaling value at the start of the long term period (0.81) shows that evaporation at the end of the peak pressure period is much more important than the value of 0.02 calculated at the begnuung of the peak pressure phase would indicate. Expert Review The experts agreed that the ranking of this phenomenon should be High during the peak pressure and long term periods of the LOCA, and Medium for the MSLB blowdown as summarized in Appendix A. The early LOCA time phases were judged to be Not Applicable. PHENOMENA IDENiu1 CATION AND RANKING Revision 1 oA36monwpf 1b.10u97 ' October 1997 _ _ _ _ _ _ _ _ _ _ _ _ _ . l
4-122 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary
.ondition, correlation, code option, noding, input parameter)
~[-
ja,e JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,e PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692non.wpf:1b-101497 October 1997
wW
'g -
4 ppCATION AND WC Revision 1 ggpon.wpf:1b-101497
4 124 ' 4.4.8 PCS Cooling Water 4.4.8A PCCWST Water Flow Rate
- Water flow rate from the PCCWST affects the amount of heat that can be removed due to l subcooling (see subsection 4,4.8B). Water flow rate is also directly related to the film flow rate that influences film stability (see subsection 4.4.8C).
PIRT RANKING N/A for LOCA phases - blowdown and refill HIGH for LOCA phases - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Resiew)
- Test Results The LST results (Reference 38, Section 2.1.2) for tests 202.2, 207.3,208.1,216.1A, 216.1B,202.1,202.2,207.2, and 207.4 showed that water flow rate, which directly affects the evaporation rate, is important. These tests confirm a ranking of High for the time periods when the shell is wet.
- Sensitivity Studies
] The sensitivity studies (Reference 1, Figure 7-9) showed that a 65% increase in the nominal film flow rate decreased the peak containment pressure by approximately 1 psi. The increased nominal film flow was in excess of that which is evaporated. The coverage area affected the long term pressure (Reference 1, Figure 7-10) showing a
- relatively strong influence when reduced coverage areas are considered.
- Expert Review The experts agreed that the ranking of this phenomenon should be High during the peak pmssure and long term periods of the LOCA, and Low for the MSLB blowdown as sur marized in Appendix A.
PHENOMENA IDFNIIFICATION AND RANKING Revision 1 0:\3692non.wpf.1b 101497 October 1997
4 125
- HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter)
I ja,e JUSTIFICATION FOR EVALUA' lion MODEL TREANENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) { l ( I PHENOMENA IDENnFICATION AND RANKING Revision 1 cx\3692non.wpf:1t>101497 October 1997
4-126 I l ja,e EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c . PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692nonwptib-101497 October 1997
l-4 127
}a,e PHENOMENA IDENTIFICATION AND RANKING Revision 1 mW2non.wpf.lb-101497 October 1997
4-128 4.4.8B: Water Temperature Water temperature applied from the PCCWST affects the amount of heat removed due to subcooling. Energy scaling shows that the amount of break energy removed from the vessel by subcooling is less than 8 percent. Since the energy removed by the subcooled film is proportional to its temperature, the energy was minimized by assuming the external water is at the maximum temperature. Water temperature is ranked the same as outside film energy transport (see subsection 4.4.7M). PIRT RANKING N/A for LOCA phases - blowdown and refill
, - MEDIUM for LOCA phases - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Test Results The LST results (Reference 38, Section 2.2) for tests 203.2 and 210.1 showed that the water film temperature had a smaller effect than the air temperature. These tests would suggest an importance ranking of Low for the time periods when the shell is wet.
l Scaling Results
- ~
The scaling results (Reference 2, Section 8.4) provided energy scaling PI values (PI ,, q,3,x) of 0.01 for peak pressure and 0.08 for long term which shows that the heat capacity (subcooling) of the water film was of Low importance. Expert Review The experts agreed that the ranking of this phenomenon should be Medium during the peak pressure and long term periods of the LOCA, and Low for the MSLB
- blowdown as summarized in Appendix A.
i PHENOMENA IDENEFICATION AND RANKING Revision 1 - o:\3692nortwpf:1b-101497 October 1997
l 4 129 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary
- condition, correlation, code option, noding, input part. meter)
I ja,e JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I Ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c PHENOMENA IDLNTIFICATION AND RANKING Revision 1 o:\3692nonwptib.101497 - October 1997
4 130 4.4.8C Water Film Stability and Coverage ! The stability of the liquid film due to the effects of momentum, heat flux, effective thennocapillary forces, and surface tension has the potential to limit surface water coverage, and hence, the evaporative cooling (Reference 1, Section 7). Film stability was ranked Itgh whenever evaporative cooling was important, i.e., during the peak pressure and long term phases. Stability is N/A during earlier LOCA phases. Since PCS water is not available until later in the MSLB, it is ranked Low. PlRT RANTING N/A for LOCA phases - blowdown and refill HIGH for LOCA phases - peak precure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Resul" c ali " mits. Sen.sitivity Studies, Expert Review)
- Test Resui.
The 15 redah, (hd e a. x
- ion . 3 &r t+st$ 2tu S.3,208.1,216.1A, 216.1B, ; '2.L ".T 2, 207 2, a 4 297a c5ewed tnet the we. coverage, which directly affects t' Mti eva}nrabon Mu rerecr '. r.w w; g.,rtant. These tests confirm an importanw , al ';c w c.t p.....a when the shell is fully wetted, i.e.,
during the peak pressure and long term perivis of the LOCA. Sensitivity Stuales A sensiti rity study (Reference 1, Section 7.5.3) showed that a 50% reduction in water coverage (100% to 50%) resulted in the calculation of a 1.46 psi increase in the peak i containment pcssure. This calculated change is considered significant. This supports an importance ranking of High for the time periods when the shell is fully wetted, i.e., during the peak pressure and long term periods of the LOCA. Expert Review The experts agreed that the ranking of this phenomenon should be High during the peak pressure and long term periods of the LOCA, and Low for the MSLB blowdowm as summarized in Appendix A. PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692nortwphib-101a7 October 1997
4-131 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary - condition,' com!!ation, code option, noding, input parameter) ( 9 ja.c JUSTIFICATION FOR EVALUATlON MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,c
~
PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf:1b-101497 October 1997 _ )
?:
- '4-132
( EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 o \3692rmwpf:1b-101597 October 1997
4-133 I 9 ja,e PHENOMENA IDENTIFICATION AND, RANKING Revision 1 o:\3692non.wpf:1b 101597 October 1997
e ;34 4.4.8D Film Stripping The riser gas interacts with the film surface by momentum transfer (shear stress). However, the maximum air velocity is approximately 13 ft./sec upward, and the film surface velocity is 3 ft./sec downward. The gas-liquid velocity difference is too low to induce liquid surface instabilities that would cause drops or spray to be torn from the film, based on observations in the ETC flat plate tests at velocities up to 39 ft./sec. During the LOCA blowdown and refill phases, there is no outside film, therefore film stripping is N/A. For the peak pressure and long term LOCA phases and MSLB, film stripping whs ranked Low in importance. PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Test Results The Heated Flat Plate test results (Reference 26, Section 3) for tests 13 through 21 showed that the water film was not adversely affected by the countercurrent air flow (5.9 fps to 38.7 fps). This information supports an importance ranking of Low for the time periods when the shell is wetted, i.e., dunng the peak pressure and long term periods of the I.OCA and the MSLB blowdown.
Expert Review The experts agreed that the ranking of this phenomenon should be I ow during the peak pressure and long term periods of the LOCA, and Low for the MSLB blowdowm as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ ja.c PRENOMENA IDENTIFICATION AND RANKING Revision 1 0:\3692nortwpf:1b-101W7 October 1997
4-135
-- JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies)
[ ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS [- ja,e i 1
- PHENOMENA IDEN11HCATION AND RANKING Revision 1 o:u6%on.wpf:1b-101497 October 1997
4-136 l 4.4.8E Film Drag I The riser air flow friction resistance may be increased by the additional liquid film velocity. However, the riser friction accounts for less than 50 percent of the total PCS air flowpath loss coefficient (Reference 2, Section 10.1.4), based on hydraulic air flow path tests (Reference 10). i.pproximately 50 percent of the containment surface is wet at the peak pressure condition, the other 50 percent and the bafAe are dry. Relating the effect of friction to the wet area fraction permits the estimate of 0.5 x 50 percent x 50 percent = 13 percent of the total drag to be affected by the film velocity. Relating this to the effective veiocity increase from 17 to 20 ft./sec (see subsection 4.4.8D) permits the estimate that the film portion of the drag will increasa by 38 percent. The net effect is a 5 percent increase in the total PCS loss coefficient from 2.5 to 2.6, which has a negligible effect on containment pressure. Consequently, this interaction is ranked Low for time phases where PCS water is available and N/A othenvise. PIRT RANKING N/A for LOCA phases - blowdown and rehil LOW for LOCA phase - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Exped Rev.iew The experts agreed that the ranking of this phenomenon should be Low during the peak pressure and long term periods of the LOCA, and Low for the MSLB blowdown as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I l* JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,e PHENOMENA IDENTIFICATION AND RANKING Rev.aoU on36monwpt.Ib.10 m October 1997
4-137
- EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS
[_ ja,e - PHENOMENA IDENnHCATION AND RANKING Revision 1 o:\3692non.wp61b-101497 October 1997
4-138 4.4.9 IUser Annulus and Chimney Volume 4.4.9A PCS Natural Circulatinn Natural circulation through the external PCS flow path (downcomer annulus, riser annulus, and chi..iney) results from the buoyancy introduced by heating the riser air. Flow rates are established based on the balance of buoyancy and unrecoverable pressure losses. Unrecoverable losses are taken from the 1/6 scale air flow tests as discussed in subsection 2.2.4. The baffle geometry and baffle supports affect the unrecoverable losses in the riser, so the effect of baffle and supports is evaluated for PCS natural circulation in the riser. Daffle supports are a source of drag on the riser annulus air flow. Daffle support drag accounts for about 20 percent of the riser portion of the PCS air flow path loss coefficient. The baffle supports bleck less than 1 percent of the riser cross section area, and produce less drag than the georr.etry used to measure losses in the 1/6 scale loss coefficient tests. PIRT RANKING LOW for LOCA phases blowdown and refill hfEDit'hi for LOCA phases - peak pressure and long term htEDIUh1 for hiSLB DASIS FOR PIRT RANKING (Test Results. Scaling Results, Sensitivity Studies, Expert Review)
- Test Results The LST results (Reference 38, Section 2.3) for tests 202.2,204.1,205.1, and 206.1 showed that the riser air velocity had a small effect on the containment pressure.
However, since these tests had more extemal water (scaled) applied to the outside containment surface than AP600 and therefore, somewhat more subcooling heat removal, the AP600 may be mm sensitive to riser air velocity. Sensitivity Studier. Sensitivities in Reference 1, Section 10, show there is a low sensitivity of containment pressure to changes in loss coefficient. The lack of a high level of sensitivity is due to the self-correcting performance of the PCS, which results from the increase in buoyancy and heat transfer due to shell teniperature inacases. Increasing the loss PHENO 3ENA IDEN11FICATION AND RANKING Kevision 1 auo92non.wrtib-toi497 October 1997
4 139 1 coefficients tends to decrease the nser flow and energy removal, increasing the surface l temperature. A surface temperature increase tends to increase the driving force for ; energy removal, and reduced flow rates tend to increase the annulus average l temperature, and thus the buoyancy driving head. Additionally, the evaporation rate ! increases nonlinearly (as saturation pressure as a function of temperature) with a sheli temperature increase. Thus, buoyancy, (T, av evaporation rate increases dampen the impact of an increase in la coefficient on containment pressure, resulting in a low sensitivity.
- Expert Review The experts agr ed that the natural circulation should be ranked Low for the blowdown and refill periods of the LOCA and Medium for htSLB, but should be either High or Medium rank for the peak pressure and long term periods of the LOCA as summarized in Appendix A. The final ranking was based on the sensitivity studies which support a Medium ranking.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) ( l t Jax JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test , Experience, Modeling Guidance, Sensitivity Studies) l l [ jax PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA3692non.wpf*101497 October 1997
4 140 I I l Ja.c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c l'liENOMENA .1DENUFICATION AND RANKING Revision 1 o:\3692ncnwpf:1h101497 _ October 1997 _ )
1
- . . 4 141 4.4.0 Vapor Acceleration The evaporating film produces a steam ilux into the riser that has to be accelerated from the dowmvard velocity of the film surface to the upward riser velocity. The effective drag can be 2
estimated as rv /2, where the density, r is that of steam in the riser and the velocity, v is tha difference between the downward and upward velocities, or 16 ft./sec. The result must be 2 normalized to rv /2 for the riser mixture. The steam density at the top of the riser is less than 7 percent of the total density, so the steam will add approximately 0.07(16/13)2 = 0.11 to the total drag coefficient of 2.5. Thus the additional momentum is less than 5 percent and the net effect on containtnent is negligible. P.*(T RANKING N/A for LOCA phases blowdown and refill LOW for LOCA phase peak pressure and long term LOW for htSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that the ranking of this phenomenon should be Low during the peak pressure and long term periods of the LOCA, and Low for the htSLB blowdown as summarized in Appendix A. , llOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ ja.c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,c l'liENOMENA IDENTIFICATION AND RAfGING Revision 1 o:\3692non wpf 1b-101497 October 1997
4 143 EVALUATION MODEL TIGATMFNr OF UNCERTAINTY, DISTORTIONS [ ja,c PHENOMENA IDENTIFICAllON AND RANKING Revision 1 o:\%92nenwpf;1h101497 October 1997
4 143 4.4.9C Fog Fog can occur in the riser annulus if the evaporation of PCS cooling water causes the partial pressure of the vapor to exceef the saturation pressure. If fog forms in the riser, it will change the density of the riser gas / drop mixture. The density change results from three separate effects that accompany condensation: The density increase due to the increased liquid mass The. density increase due to the increased molecular weight of the gas, as air replaces the steem that condenses The density decrease due to the gas (alt + steam) temperature increase when hg g is released by the condensate formation Relative values of these three effects were evaluated at conditions corresponding to those in the riser at the time of peak pressure. It was postulated that drops arise from supersaturated-vapor and the relative magnitude of the three separate density change effects were calculated. The calculation showed the increased liquid mass produced a density increase that was approximately 10 percent of the net density change, the molecular weight increase produced a density increase that was approximately 10 percent of the net density change, and the release of h fg increased the mixture temperature and decreased the density by 120 percent of the net density change. The net change was a decrease in density with vapor condensation in the riser, that produces a net increase in the buoyancy and the PCS natural circulation air flow rate. Operation at temperatures other than design basis valuca are not expected to significantly affect the : elative magnitudes. The effect of fog formation in the riser gas is ranked Low during the peak pressure and long term time phases, based on the above evaluation, which shows that drop formation increases the PCS natural circulation flew rate. The effect of fog formation is also ranked N/A during blowdown refill, and htSLB since there is insufficient evaporation to cause drop formation under the assumed environmental conditions. PIRT RANKING N/A for LO'.A phases - blowdmvn and refill LOW for LOCA phase - peak pressure and long term N/A for htSLB PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA369 f.1 M ot497 October 1997
. - . . -J
I 4 144 BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Expert Review The experts agreed that the ranking of this phenomenon should be Low during the peak pressure and long term periods of the LOCA as summarized in Appendix A.
110W PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ ]** JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) jat EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS l ]"e 3 PHENOMENA IDENTIFICATION AND RANKING Revision I ce\3692non.wpf lb-101497 October 1997
... . , - _ . , , . - ,. . . - . . . . . . . - . - - . - . , . . - - .- - -- - - - - - - - - - ^ - - - - -i
4 145 4.4.9D Flow Stability The chimney and urper part ef .he shield building are large concrete structures that can cool the PCS air flow before it exits from the chimney, producing a negatively buoyant wall boundary layer, thereby reducing the natural circulation buoyancy forces and potentially affecting equivalent chimney flow losses. The issues associated with wall boundary layer and the potential for reverse flow directions are considered relative to flow stability in the chimney. Calculations performed for the scaling analysis show the PCS air flow loses only 2 percent of its thermal energy and condenses less than 1 percent of the vapor while passing through the chimney at the time of peak pressure. Consequently, there is insufficient energy removal to develop significant instability in the chimney, and there is little effect on the net buoyancy. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert ( Review) l Expert Review The experts agreed that the ranking of this phenomenon should be Low dudng all periods of the LOCA and for the MSLB blowdown as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ ja.c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) l Ja,e PHENOMENA IDENr1FICATION AND RANKING Revision 1 et\3692non.wph1M01497 October 1997
_____-____-__.. ~ 6146 - EVALUATION MODEL TREATMENT OF UNCERTAlhTlY, DISTORTIONS [ ja,c 1 PHENOMENA IDENHFICATION AND RANKING Revision 1 o:\3692ron.wptib 101497 October 1997
, 4147 .
4.4.10 Baffle The baffle receives radiant energy from the containment film and dry surface. Some of the energy is conducted through the baffle and rejected to the downcomer air by convection and to the shield building surface by radiatio t. 'Ihe dry baffle has a time constant of approximately 30 minutes, to will significantly lag the shell (with a 5-minute time constant) during the initial transient of PCS annulus flow. The buoyant air flow in the riser and chimney will be well-developed long before the baffle begins to heat the downcomer air and
; the shield. Due to the r,ignificant time lag, the phenomena discussed in this section are negligible during LOCA blowdown and refill, and so are ranked N/A for those accident phases. Since esternal cooling begins to take effect so late in the MSLB transient, baffle phenomena are ranked N/A for MSLB. !
4.4.10A Convection to Riser Annulus The baffle transfers heat to the riser gas by convection. The heat transfer to the riser has an effect on the buoyancy that may account for a significant part of the natural circulation
- driving force, so the importance is ranked Medium during the long term phase and Low during the peak pressure phase.
i PIRT RANKING i N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure , MEDIUM for LOCA phase - long term N/A for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensithity Studies, Expert Review) 1
- Scaling Results i
Energy scaling (Reference 2, Section 8.4) shows the convective heat transfer from the baffle to the riser accounts for less than 2 percent of the break energy during the long term phase, which supports a Low ranking for this phenomenon. However, external momentum scaling shows that the downcomer has an effect on the buoyancy that may account for up to 16 percent of the natural circulation driving force. Therefore, this phenomenon is ranked Meditun during the long term phase and Low during the peak pressure phase. PHENOMENA IDENTIFICATION AND RANKING Revision 1 as3692nonwrtib 101497 October 1997
~ _ . _ . . . . _ . _ _ . _ . _ . - . _ . _ . _ _ - - - - - . _. ._ _ - -
4 148 Expert Review Experts agreed that the baffle phenomena were not applicable during the blowdown and refill periods since there was insignificant heat transfer; and were of Low importance during the other periods when there was heat transfer. The final ranking was based on the scaling results discussed above. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) j i I l ja,c 1 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test : Experience, Modeling Guidance, Sensitivity Studies) I ja,e EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c l l PHENOMENA IDENT.'FICATION AND rah %ING Revision 1 ! oA3692nortwpf:1h101497 October 1997
. . . _ _ . __.~. , _ _ - - _ . ._.
4 149 4.4.10B Convection to Downcomer Annulus PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure MEDIUhi for LOCA phase -long term N/A for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results The baffle transfers heat to the downcomer gas by convection. Energy scaling shows the convective heat transfer from the baffle to the downcomer to account for less than 2 percent of the break source energy during the LOCA long term phase. This supports a Low ranking for this phenomenon. However, external momentum scaling shows that the downcomer has an effect on the buoyancy that may account for up to 16 percent of the natural circulation driving force. Therefore, this phenomenon is ranked Medium during the long term time phase and Low during the peak pressure phase. Expert Review Experts agreed that the baffle phenomena were not applicable during the blowdown and refill periods since there was insignificant heat transfer; and were of Low importance during the other periods when there was heat transfer. The final ranking cf Medium during the LOCA long term phase is based on the scaling results discussed above. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary wMition, conelation, code option, noding, input parameter) ( Ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692nortwpf.lb-101497 October 1997 r
4 150 [ 3a,c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF MIENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies)
. l'#
EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I jae PHENOMENA IDENTIFICATION /dID RANKING Revision 1 o:\3692non.wstib-101497 October 1997 m
4 151 4.4.10C Radiation to Shleid Building PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phases peak pressure and long term N/A for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results lhe baffle transfers energy to the shield building by radiation. Energy scaling shows that the effect is less than 1 percent of the break energy. The long time constant of the baffle, and the small heat transfer rate results in a ranking of Low for all time phases.
- Expert Review The experts agreed thet the baffle phenomena were not applicable during the blowdown and refil! periods since there was insignificant heat transfer; and were of Low import nce during the other periods when there was heat transfer.
110W PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja,c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,e PliENOMENA IDENTIFICATION AND RANKING Revision 1 0:\3692nonwpf.1M01497 October 1997
4 153 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 (r\3692ncawpf;1b.101497 October 1997
4 153 4.4.10D Conduction 'Duough Daffle The baffle is 1/8 in. thick steel v/ith a very small Blot number of 0.0004. Thus the baffle behaves as a lumped mass with negligible internal resistance for all time phases. Since conduction through the baffle supplies the heat that is convected to the downcomer, conduction through the baffle is ranked the same as convection to the downcomer annulus (subsection 4.4.1011) for LOCA peak pressure and long term phases. PIRT RANKING N/A for LOCA phases blowdown and refill LOW for LOCA phase peak pressure hiEDIUhi for LOCA phase - long term N/A for hiSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Sensitivity Studies Reference 40, Section 4.1 contains validation of the 1D conduction equations used in clime subroutines by comparison to theoretical solutions. Noding studies were performed for which the number of axial elevations, number of circumferential divisions, and radial conductor mesh were varied (Reference 1, Section 12.3.2.). The results of these calculations show that doubling the number of clime axial nodes, doubling the number of circumferential stacks, or doubling the number of conductor mesh points does not affect the calculated pressure transient in the evaluation model. Expert Review The experts agreed that the baffle phenomena were not applicable during the blowdown and refill periods since there was insignificant heat transfer. The experts disagreed on the ranking during the LOCA long term phase, as summarized in Appendix A. The final ranking of hiedium for the LOCA long term phase was based on the potential to degrade PCS natural circulation. IHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non wpf-1t>101497 October 1997
\ 4 154 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, crirelation, code option, noding, input parameter) I 4 ja c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja c , PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692rmwpf.1b 101497 October 1997 __ = = ___m _ _ _ _ _ __ _ . . _ _ . - - - - - -
4 155 . 4.4.10E Condensation on the Baffle Under conditions where the baffle temperature is below the dewpoint of the riser annulus air, condensation on the baffle could occur. Condensation on the baffle would decrease the baffle time constant by a factor of 20 (it would reduce from about 1000 to about 50), which could impact the relative importance of some baffle phenomena. An evaluation performed to support scaling analysis and using annulus average conditions shows that, for the limiting condition of high ambient temperatures, condensation on the baffle is not expected during any time phase of a LOCA. An examination of evaluation model results shows that WGOTHIC predicts condensation on the baffle only during about 3 seconds after PCS water is assumed to be applied, and only at the top where riser annulus steam concentration is highest, during the bounding DBA. PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term N/A for hiSLB UASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results For the conditions assumed for the scaling calculations, the model predicts no { condensation on the baffle for any time phase. Expert Review The experts agreed that the baffle phenomena were not applicable during the blowdown and refill periods since there was insignificant heat transfer; and were of Low importance during the other periods when there was heat transfer. HOW PHENOhiENON IS IMPLEhiFNTED IN EVALUATION htODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja,c PHENOMENA IDENTIFICA110N AND RANKING Revision 1 ow9:non wrtid 101497 October 1997
. . . ---a
i N l
)
JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test ! Experience, Modeling Guidance, Sensitivity Studies)
- I ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS i l '
l jax i I 4 l i l 4 l i 9 b L l PHENOMENA IDENTt9 CAT 10N AND RANKING Revision 1 o:\3692non.wpf 1b401497 October 1997
4 157-4.4.10F Heat Capacity of the Baffle i At quasi steady annulus conditions typical of peak pressure and long term LOCA phases, heat into the baffle is about equal to heat out of the baffle, so its heat capacity has no effect. During transients in the annulus, the heat capacity of the baffle could affect the rate of heating of the riser, and thus affect the initial annulus flow start up transient. The riser annulus is in fully developed turbulent flow by the time the containment shell is greater than 2*F above ambient. The time constant cf the wet shell is about 500 r.econds, as compared to the baffle time constant of 1000 seconds, and baffle heatup is driven by radiation from the shell, so the transient development of external flow is complete well before the baffle heats up significantly. Therefore, heat capacity of the baffle is ranked Low for the LOCA peak pressure and long term phases. PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term N/A for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that the baffle phenomena were not applicable during the blowdown and refill periods since there was insignificant heat transfer; and were of Low importance during the other periods when there was heat transfer, llOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) ( ja.c PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA3692nmwpf.1b 101497 October 1997
4 158 JUSTIFICATION FOR EVALUATION hiODEL TREAThiENT OF PilENOhiENON (Test Experience, hiodeling Guidance, Sensitivity Studies) I jax L EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I l'" 4 PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692nonwpf.1b.101497 October 1997 l
4-159 4.4.10G Leaks *1hrough Baffle Baffic leaks due to postulated missing or misaligned baffle plates, may permit some of the air flow from the downcomer to "short circuit" through the baffle to the riser, rather than following the normal air flow path to the bottom of the downcomer, up the riser and out the chimney. In .a forced flow system, the effect of short circuiting could be significant, depending on the relative area and pressure losses of the flow paths. However, the process that causes natural circulation is the relative density of the gas in the environment, the downcomer, and the riser. Short circuiting through missing baffle panels at the top of the downcomer will not reduce the buoyancy in the riser below the leakage point. In fact, if it is I postulated that the riser flow decreases, the gas temperature and buoyancy will increase, at least partially compensating for the leak, it is well known that a top to-bottorn pattern of circulation develops in rectangular cross section channels with a heated vertical wall, even without a partition (or baffle). Test data from Siegel and Norris (Reference 45) show a reduction of less than 50 percent in a heated wall Nusselt number when a rectangular channel with heated parallel sides have their open bottom closed (sides closed, L/D = 20). Thus, it is expected that even the worst case of baffle leakage has a minor effect. FIRT RANKING N/A for LOCA phases - blowdown and refill hiEDIUht for LOCA phases - peak pressure and long term W/A for hiSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that the baffle phenomena were not applicable during the blowdown and refill periods since there was insignificant heat transfer. The experts disagreed on the ranking during the LOCA peak pressure and long term phases, as summarized in Appendix A. Based on engineering judgment, the effect of baffle leaks is ranked hiedium for the LOCA peak pressure and long term phases. PHENOMENA IDENTIFICATION AND RANKING Revision 1 oA3692non wpf;1b-101497 October 1997 i
41(O ' liOW l'HENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I jae JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) l . jax EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I jax l PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non wpf.lb.101497 October 1997
4 161 4.4.11 Baffle Supports 4.4.11 A Convection to Riser Air The baffle supports are approximately 3/8-in. thick,18 in. high, U-shaped steel brackets with the two ends of the U welded to the shell. After the exterr.al surface or the shell heats up, it will transfer heat to the baffle supports that behave as fins to transfer heat to the riser gas. A comparison of the hA values (heat transfer coefficient times area) shows the relative magnitude of heat loss 'hrough the surface and the bracket. The cross section of the supperts 2 is approximately 13 in , and there is one support for every 11,088 in2 of shell surface. The heat transfer coefficient is approximately 3 for the dry shell and 50 for the support, giving a ratio of 0.02 for support / dry shell. This ratio is much less for the wet shell. Furthermore, the dry shell contributes only a small fraction to shell heat rejection. Consequently the ability of the support to influence the shell temperature is so limited that baffle support energy transport is estimated to be insignificant. Hence, tius process is ranked Low during all time phases. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert s Review)
- Expert Review The experts agreed that the ranking of this phenomenon should be Low during all periods of the LOCA and for the MSLB blowdown as summarized in Appendix A.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I 1**
\
' l'HENOMENA IDENTIF. CATION AND RANKING Revision 1 a:\3692nortwrf:1b.101497 - October 1997
.. . _ _ _ _ _ _ _ _ _ _ - _ _ -J
4 162 JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Esperience, Modeling Guidance, Sensitivity Studies) I ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS [ ja,c F PHENOMENA IDENTIFICATION AND RANKING Revision 1 0:\3692nortwp611 -101497 October 1997
4 163 4.4.11B Radiation from Shell After the extemal surface of the shell heats up, it will radiate to the baffle supports which will convect heat to the riser air and conduct heat to the baffle. Since the area of a baffle support that is exposed to radiation is normal to the shell surface, and the shell surface curves away from the supports, a conservative assessment of the potential effect of radiation to a baffle can be based on the areas of the two sides. The area of the two sides of the 2 supports is approximately 216 in , and there is one for every 11,088 in2 of shell surface. Smce the radius of the shell and baffle differ by only one foot, the corresponding baffle surface is 2 also approximately 11,088 in . Since a conservative assessment of the area blocked by the supports is less than 2 percent of the area for radiation, the supports block a negligible amount of the energy transferred by radiation from the shell to the baffle. As determined from scaling, radiation from the shell is less than 2 percent of the break source energy. Thus, radiation from shell to supports is ranked Low for all accident phases. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Expert Review !
The experts agreed that the ranking of this phenomenon should be Low during all periods of the LOCA and for the MSLB blowdown as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [
}"#
JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test - Experience, Modeling Guidance, Sensitivity Studies) I jax_ PHENOMENA IDENTIFICAllON AND RANKING Revision 1 oA3692nmwpf;1b-101497 October 1997
4 164 l l EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS [ ] I l PHENOMENA IDEN11FICATION AND RANKING Revision 1 a\3692non.wpelb.101497 October 1997
- 4 165 4.4.11C Conduction from Shellinto Baffle Supports A small amount of heat is conducted from the shell to the baffle supports from which the heat can be convected into the riser air or conducted into the baffle. Convection from baffle supports to riser air has been evaluated and ranked Low during all accident phases as discussed above. The resistance to heat transfer from the shell through supports to the baffle by conduction is much higher than the resistance from the shell to riser annulus gas by convection. Since convection from the baffle supports to ris(' ;as is ranked Low, conduction from the shell through baffle supports is also judged to be Low for all accident phases.
PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that the ranking of this nhenomenon should be Low during all periods of the LOCA and for the MSLB blowdown as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I l JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,C EVALUATION MODEL TREATMEM OF UNCERTAINTY, DISTORTIONS l l PHENOMENA IDENnFICATION AND RANYdNG Revision 1 ass,2nmwpt.1b 101497 October 1997
= =
f 4-166 4.4.11D 11 eat Capacity of BafSe Supports The heat capacity of baffle supports is evaluated for the potential to store energy leaving the shell by conduction or radiation. From scaling analysis, the amount of energy leaving the shell by conduction or radiation is less than 4 percent of the break source energy, while convective mass and energy from the shell represents more than 50 percent. A conservative assessment of the heat capacity of the baffie supports relative to the baffle can be made by comparing the mass of supports to the mass of the baffle, neglecting the mass of baffle stiffeners, which shows the support mass is less than about 13 percent of the baffle mass. Since the amount of mass is not negligible, the following provides an assessment. At quasi-steady conditions, the heat into the baffle supports by conduction or radiation is approximately equal to the heat removed by the baffles by convection, so that there would be no net energy storage in the supports. During the initial startup transient, heat absorption by b the baffle supports could delay the rate of baffle temperature increase or reduce the rate of riser aL temperature increase. A delay in baffle heat up would put less heat into the downcomer during the initial transient,_which would be a benefit for startup of PCS annulus flow. The effect can be quantified by comparing the baffle time constant ( t = p cp S / h, where p is the density, cpis specific heat,6 is the structure thickness, and h is the net heat ~ transfer coefficient) to the time constant assuming the support mass is added to the baffle mass as increased thickness. The time constant for baffle heating, about 1000 seconds, would be increased by 13 percent. Since flow is fully developed, turbulent forced convection in the annulus by the time the containment extemal temperature reaches 2'F above ambient, the increase in baffle time constant would not adversely affect air flow startup. A delay m the rate of riser air temperature increase could reduce the rate at which buoyancy driven annulus flow develops, thus reducing external heat removal. From the discussion in subsection 4.4.11B, the supports would absorb less than 2 percent of the radiant energy leaving the shell. An upper bound can be based on assuming that all the energy into the support comes from radiation, which shows that the supports would absorb less than (2 percent x 5 percent ), or 0.1 percent of the energy leaving the shell. Relative to the energy delivered dire:tly to the riser gas by heat and mass transfer, the energy absorbed by supports has a negligible effect on the development of the initial traisent. Therefore, the heat - capacity of baffle supports is ranked Low for all accident phases, t d PIRT RANKING LOW for all LOCA phases-LOW for MSLB PHENOMENA IDEN1&lCATION AND RANKING Revision 1 o:\3692non.wpf:1b 101497 October 1997
4 167 - BASIS FOR PIRT RANKING (Test Results, Ss.aling Results, Sensitivity Studies, Expert Review)
- Expert Review The experts agreed that the ranking of this phenomenon should be Low during all periods of the LOCA and for the MSLB blowdown as summarized Ln Appendix A.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, contlation, code option, noding, input parameter) [
]* c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies)
-J
[ ja c EVALUATION MODEL TREATMF'(T OF UNCERTAINTY, DISTORTIONS _[.
]* c PHENOMENA IDEN11HCATION AND RANKING Revision 1 on36 mon.wpt.1b-101497 October 1997 '
. le.
4-168 4.4.12 Chimney Structure R The following phenomena are ranked on the basis of energy scaling: A - Conduction through chimney B - Convection from chimney air C - Heat capacity of structure D - Condensation on chimney The chimney and upper part of the sbleld building are large concrete structures that can cool the PCS air flow before it exits from the chimney, producing a negatively buoyant wall boundary layer, thereby reducing tiv ;etural circulation buoyancy forces and potentially affecting chimney flow. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Scaling Results Calculations performed for the scaling analysis (Reference 2, Section 8.4) show the PCS air flow loses less than 1 percent of the break source energy by heat transfer, and condenses less than 5 percent of the break source energy while passing through the chimney at the time of peak pressure. Consequently, there is a small energy removal within the chimney, and a small effect on +he net buoyancy. This supports the Low ranking for chimney structure heat and mass transfer phenomena for all time phases, i Expert Review The experts agreed that the ranking of this phenomenon should be Low during all periods of the LOCA at:d for the MSLB blowdown as summarized in Appendix A. I PHENOMENA IDENIIHCATION AND RANKING Revision 1 o:\3692non.wpf:1b.101497 October 1997
6169 5 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary
- 4. g condition, correlation, code option, noding, input parameter) x-3 ja,c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience,~Modeling Guidance, Sensitivity Studies)
E ' [ ja,c 3 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS 23 I ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 c:\3692non.wpf;1b-101497 October 1997
4 170 4.4.13 Downcomer Annulus-4.4.13A PCS Natural Circulation w A generas description of this phenomenon is provided in subsection 4,4.9A. The unrecoverable pressure losses in the downcomer region include the inlet and turning loss and frictional losses on the baffle and shield building surfaces, Also considered in the
- downcomer are the losses due to the flow path which connects to the riser annulus, which have been minimized by the use of a curved vane based on separate effects tests of the flow ,
losses. l
- The baffle is heated by radiant er.ergy from the shell. The downcomer air is heated by convective heat transfer from the baffle, as well as by haat convected from the shield building surface, PIRT RANKING LOW for LOCA phases - blowdown, refill MEDIUM for LOCA phases - peak pressure, long term
- MEDIUM for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
Sensitivity Studies Sensitivities in Reference 1, Section 10, show there is a low sensitivity of containment pressure to changes in loss coefficient. The lack of a high level of sensitivity is due to the self correcting performance of the PCS, which results from the increase in buoyancy and heat transfer due to shell temperature increases. Increasing the loss coafficients tends to decrease the riser flow and energy removal, increasing the surface temperature, A surface temperature increase tends to increase the driving force for energy removal, and reduced flow rates tend to increase the annulus average temperature, and thus the buoyancy driving head, Additionally, the evaporation rate increases nonlinearly (as saturation pressure as a function of temperature) with a shell-temperature increase. Thus, buoyancy, (T, and evaporation rate ircreases dampen the impact of an increase in loss coefficient on containment pressure, resulting in a low sensitivity. PHENOMENA IDENIw1 CATION AND RANKING Revision 1 o:\3692nonwptib-101497 October 1997
4-171 Expert Review l The experts agreed that the natural circulation should be ranked Low for the blowdown and refill periods of the LOCA and Medium for MSLB, but should be either High or Medium rank for the peak pressure and long term periods of the LOCA as summarized in Appendix A. The final ranking was based on the results of the sensitivity studies described above and engineering judgment. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja.c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studios) [ Ja,c EVALUATION MODEL TREATMEB T OF UNCERTAINTY, DISTORTIONS [ ja,c PHENOMENA IDENTIFICATION AND RANKING Revision 1 m\3692non.wpf;1b-101497 October 1997 l
4-172
- u 4.4.13B Downcomer Annulus Air Flow Stability i
Since the baffle is heated by radiation from the shell, the downcomer air will convect energy from the baffle, giving rise to a buoyant boundary layer. Based on results of scaling energy ! and momentum in the external flow path, the downcomer accounts for a negligible fraction of energy and a small part of momentum in the PCS flow path. Therefore, the buoyant boundary layer does not significantly impact downcomer air flow. Downcomer air flow stability is ranked Low for all accident phases. PIRT RANKING LOW for all LOCA phases LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that the ranking of this phenomenon should be Low during all periods of the LOCA and for the MSLB blowdown as summarized in Appendix A. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) [ ja,e JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS ( ja c PHENOMENA IDENTIFICATION AND RAhTdNG Revision 1 o:\x92non.wptib.101497 October 1997 1
4-173 4.4.14 Shleid Building 4.4.14A Convection to the Downcomer A maximum heat load from the shield to the downcomer can be estimated by assuming the radiation to the shield from the baffle is all transferred by convection to the downcomer. Energy scaling shows that the radiation from the baffle to the shield building is less than 1 percent, which qualifies for a Low rankmg. Since PCS water flow is not available until later in the transient for MSLB, a Low ranking is assigned. Since the extemal shell surface does not reject significant heat until after refill due to its thermal capacity and relatively long time constant, cohvection to downcomer is ranked N/A for the LOCA blowdown and refill' phases. FIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phases - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that convection from the shield building to the downcomer annulus is of Low importance for the MSLB, however, the internal experts ranked it as Low for the LOCA peak pressure and long term periods while the external experts ranked it as Medium as summarized in Appendix A. Since convection from the shell to the riser (subsection 4:4.7H) is ranked Low during the LOCA peak pressure phase - and Medium during the long term, and convection from the shield building with a much lower surface temperature, should be less important, this phenomenon was assigned a Low rank. > HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja,c PHENOMENA IDENIw1 CATION AND RANKING Revision 1 o:\36 mon.wpt:1b-101497 October 1997
a
. 4 174 g I l-ja,c JUSTIFICATI' O N FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies)
I ja c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c PHENOMENA IDEN1w1 CATION AND RANKING Revision 1 o:\3692non.wpf 1b.101497 October 1997
4 175 - 4.4.14B Conduction Through the Shield Building The shield building is 3 ft, thick and has a time constant of 40 hours. Consequently, the Lmide surface will not experience the day-night temperature fluctuations or effects of sun exposure. Thus, heat transfer to the environment through the shield building is ranked Low for the LOCA peak pressure and long term phases, and is N/A during LOCA blowdown and refill phases, and MSLB. PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term N/A for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that the shield building phenomena were not applicable during the blowdown and refill periods since there was insignifiumt heat transfer; and were of Low importance during the other periods when there was heat transfer. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja.c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test
- Experience, Modeling Guidance, Sensitivity Studies) l ja,c PHENOMENA IDENHFICATION AND RANKING Revision 1 o:\3692non.wpf:1b-101497 October 1997
4-176 . EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS 1 [
\
ja,e l l l l l l i PHENOMENA IDENT1HCATION AND RANKING Revision 1 ) o:\3692nonwpElb-101497 October 1997 l 1
c- 4-177 4.4.14C Convection to the Envhonment Due to the time constant of the shield building, the outside surface of the shield building is expected to remain at or near the environmental temperature. Therefore convection to the environment was ranked Low for bcth the MSLB and LOCA events. The day-night average outside surface temperature will be less than the maximum technical specification value assumed in analyses. PIRT RANKING N/A for I.OCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term N/A for MSLB BASIS FOR PIRT RANKING (Test Re% Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that the shield building phenomena were not applicable during the blowdown and refill periods since there was insignificant heat transfer; and were of Low importance during the other periods when there was heat transfer. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja,c JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) [ Ja,e PHENOMENA'fDENTIFICATION AND RANKING
' Revision 1 o:U692nonwpf:1b-101597 Cctober 1997
4-178-I ja.c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I
+
ja,e , =- - PHENOMENA IDEN11HCATION AND RANKING Revision 1 c:\36monwpf 18101497 October 1997
4 179 4.4.14D Radiation to the Environment Due to the time constant of the shield building, the outside surface of the shield building is expected to remain at or near the environmental temperature, therefore radiation to the environment was ranked Low for both the MSLB and LOCA events. PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term N/A for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review)
- Expert Review The experts agreed that the shield building phenomena were not applicable during the blowdown and refill periods since there was insignificant heat transfer; and were of Low importance during the other periods when there was heat transfer.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary ccndition, correlation, co.ie option, noding, input parameter) [ ja,e JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja, PHENOMENA IDENTIFICATION AND RANKING Revision 1 0:\3692non.wpf.lb-101497 October 1997
4 180 EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS ' I jax i i PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf:Itw101497 October 1997
-- 4 181 4.4.15 Externai Atmosphere 4.4.15A Temperature The initial condition for the atmospheric temperature is assumed for safety analyses to be 115'F. The sensitivity of peak containment pressure to atmospheric air temperature is low because the pressure is primarily lindted by mass transfer rates that are functions of steam partial pressure differences and not temperature. Temperature was ranked Low during the peak pressure and long term phases, and was N/A for blowdown and refill because the PCS has no effect. Since PCS operation is not significant until later in the MSLB, a Low ranking is assigned.
PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Reviev0 Sensitivity Studies The sensitivity study results (Reference 1, Section 5.7) showed that a large decrease in the initial temperature from 115 F to 40 F subsequently decreased the LOCA peak pressure by approximately 0.1 psi, and the worst case MSLB peak pressure by less than 0.1 psi due to the large time constant for the heat transfer through the shell. Over the longer term, the lower external temperatures provided a 2 psi benefit in the LOCA case, which is judged to be low relative to typical values of temperature uncertainties. Expert Review The experts agreed that the external temperature is ranked Low for the peak pressure and long term periods of the LOCA and for the MSLB blowdown. PHENOMENA IDENTIFICATION AND RANKLNG Revision 1 on3692non.wpf;1b-101497 October 1997 l
4-182 HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, nodt ; input parameter) I jax JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeliag Guidance, Sensitivity Studies) I j.x EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS jax PHENOMENA IDENTIFICA110N AND RANKING Revision 1 o:\3692non.wpf:1b101497 _ _ _ - _____-- __ -_---- _. -=
i 4 183 4.4.15B liumidity The peak containment pressure is limited by mass transfer rates that are functions of steam partial pressure differences. The peak internal steam pressure is approximately 40 psia, while the ambient steam partial pressure is limited to the saturation pressure of less than 1.5 psia, in terms of steam pressure difference between the inside and outside of containment and riser, the maximum variation expressed as a fraction of the total can be used as a first order estimate of the containment pressure sensitivity to humidity. The resulting variation is less than 3 percent, so the ranking is Low during the peak pressure and long term phases. A sensitivi'.y to assumed inlet humidity shows relatively low effect on the assumed value in the calculation (Reference 1, Section 5.6). The phenomenon was N/A for blowdown and refill because the PCS has no effect that early in the transient. Since PCS operation is not significant until later in the MSLB, a Low ranking is used. See also subsection 4.4.9C for the effect of fog formation in the annulus. PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studier., Expert Review) Sensitivity Studies The sensitivity study results (Reference 1, Section 5.6) showed that a decrease in the initial humidity from 22% to 0% did not have any impact on the peak pressure since the driving force for evaporation (partial pressure of steam at liquid-air interface) is large compared to humidity. Expert Review The experts agreed that the external humidity is ranked Low for the peak pressure and long term periods of the LOCA and for the MSLB blowdowm. s PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:\3692non.wpf:1b-101497 October 1997
4 3g.. . HOW PHENOMENON iS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I jax JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I 1*' EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c . PHENOMENA IDENlalCATION AND RANKING Revision 1 o:\3e2non.wpelb.101497 October 1997-
4-185 4.4.15C Recirculation i The downcorv.c draws air from the environment through 16 discrete openings near the top of the shield building. Environmental disturbances, such as gusts, wakes, or "downwash" from upwind structures, and the downwash of chimney outflow to inlets on the downwind side (recirculation) can potentially affect PCS performance. I PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Sensitivity Studies It was shown (Reference 1, Section 6) that the worst recirculation will cause less than 15 percent of the chirrney outflow to be drawn through the inlets. The resulting effec > on PCS performance was calculated and found to produce a negligible increase in containment pressure. The effect of recirculation is therefore ranked Low during the peak pressure and long term time phases. The phenomenon was N/A for blowdown and refill, because the PCS has no effect that early in the transient. Since PCS operation is not significant until later in the MSLB, a Low ranking is assigned.
- Expert Review The experts agreed that recirculation is ranked Low for the peak pressure and long term periods of the LOCA and for the MSLB blowdown.
HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code option, noding, input parameter) I ja,c PHENOMENA IDEN11HCATION AND RANKING Revision 1 a:\36 mon.wpt:16101497 October 1997
4-18o JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies) I ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c 6 1 PHENOMENA IDEN11HCATION AND RANKING Revision 1 l o \3692non.wpf.1b.101497 October 1997
4-187 4.4.15D Pressure Fluctuations l The air flow in the environment can affect the PCS behavior by inducing pressure fluctuations on the PCS inlet and outlet due to large-scale vortex shedding from such upwind i obstructions as buildings, cooling towers, and terrain. 'Ihe effect of the worst case inlet-outlet pressure fluctuations from the wind tunnel tests were evaluated and determined to have-(conservatively) no effect, and most realistically, a beneficial effect on heat rejection. The reason is .' hat the wind positive nature of the external PCS design results in more positive benefit due to increased flow rates than the negative effect of the few-second cycles of zero to negative flow rates. The long time constant of the shell (about 5 minutes) buffers the containment pressure from such fluctuations. The fluctuations are wind induced, so only occur when the wind is blowing, which is also the time when the wind positive design assists heat and mass transfer. Thus the importance of terrain, structures, and wind induced pressure fluctuations on the AP600 containment pressure was ranked Low during the peak pressure and long term phases and was N/A during the short blowdown and refill phases. Since PCS operation is not significant until later in the MSLB, a Low rankmg is assigned. PIRT RANKING N/A for LOCA phases - blowdown and refill LOW for LOCA phase - peak pressure and long term LOW for MSLB BASIS FOR PIRT RANKING (Test Results, Scaling Results, Sensitivity Studies, Expert Review) Expert Review The experts agreed that pressure fluctuations is ranked Low for the peak pressure and long _ term periods of the LOCA and for the MSLB blowdown. HOW PHENOMENON IS IMPLEMENTED IN EVALUATION MODEL (e.g., boundary condition, correlation, code optien, noding, input parameter) I
}a,e . PHENOMENA IDENTIFICATION AND RANKING Revision 1 o:u6 hon.wp.:1b.101497 October 1997
-4 188
. JUSTIFICATION FOR EVALUATION MODEL TREATMENT OF PHENOMENON (Test Experience, Modeling Guidance, Sensitivity Studies)
I-ja,c EVALUATION MODEL TREATMENT OF UNCERTAINTY, DISTORTIONS I ja,c i l PHENOMENA IDEN11HCATION AND RANKING Revision 1 o:\3692non.wpf:1b-101497 October 1997
5-1
5.0 CONCLUSION
S
- l The phenomena identificetion and importance ranking of the phenomena for the' AP500 -
containment were completed for the LOCA and MSLB transients.' These transients were considered to be the most limiting events due to their effect on containment pressure. Ic is concluded that the mass transfer processes of condensation inside contam' ment and evaporation outside containment are the most important phenomena for reducing the containment pressure and transferring et ergy to the envirorunent especially for the LOCA event. The large heat capacity of the steel, concrete, and containment shell are also sigaificant in reducing the containment pressure. The MSLB event is not significantly affected by the phenomena outside containment due to the rapid pressure transient. The process for making these conclusions was based on results from test programs, scaling analyses, sensitivity studies, and engineering judgement. The results of this evaluation have been used to focus on the phenomena and models which have the most significant impact on containment pressure. The phenomena identified have been addressed in the evaluation model for the AP600 containment, as discussed for each phenomenon in Section 4.4. CONCLUSIONS Revision 1-3642w.wpf:1b 101497
61
6.0 REFERENCES
i l 1. WCAP-14407, }NGOTHIC Application to AP600, June,1997 (to be issued). - l
- 2. WCAP-14845, " Scaling Analysis for AP600 Containment Pressure During Design Basis Acci6:nts," Rev.1, June 1997 (to be issued).
- 3. WCAP-14812, Rev. O, " Accident Specification and Phenomena Eva'uation for AP600 Passive Containment Cooling System," January,1997.
- 4. 1000-P2-901, Nuclear Island General Arrangement, Rev. 8.
- 5. 1100-CC-902, Containment / Shield Buildings - Section B, Rev.1.
- 6. PCS-M3-001, Passive Containment Cooling System System Specification Document (SSD), Rev. 3.
7 WCAP-12665, Rev.1, Tests of Heat Transfer and Water Film Evaporation on a Heated Plate Simulating Cooling of the AP600 Reactor Containment, April 1992.
- 8. A. P. Pernsteiner, Condensation in the Presence of Noncondensible Gas: Effect of Helium Concentration,1993, University of Wisconsin Thesis, November 12,1993.
- 9. WCAP-14326, Rev.1, Experimental Basis for the AP600 Containment Vessel Heat and Mass Transfer Correlations, May 1997.
- 10. WCAP-13328, Tests of Air Flow Path for Cooling the AP600 Reactor Containment.
- 11. NTD-NRC-94-4138, AP600 Design Certification Test Program Overview, Rev 6, May 17,1994.
12. WCAP-13884, Water Film Formation on AP600 Reactor Containment Surface, February 1988,
- 13. WCAP-13960, PCS Water Distribution Phase 3 Test Data Report, December 1993.
- 14. WCAP-14134, AP600 Passive Containment Cooling System Integral Small-Scale Tests, August 1994.
4 REFERENCES Revision 1 3692w.wpf:lb-101497
6-2
- 15. WCAP-14135, Rev.1, Final Data Report for PCS Large-Scale Tests, Phase 2 and Phase 3, April,1997.
- 16. WCAP-13566, AP6001/8th Large-Scale Passive Containment Cooling System Heat Transfer Test Baseline Data Report, October 1992.
- 17. NTD-NRC-95-4561, Scaling Role in AP600 PCS DBA Analysis, September 19,1995.
- 18. NSD NRC-96-4790, Scaling Analysis for AP600 Containment Pressure During Design Basis Accidents, August 8,1996.
- 19. SSAR 6.2.1.1.3, NTD-NRC-95-4504, " Containment Structure Design Evaluation, Proposed Draft / Markups of SSAR, Sections 6.2 and 6.4", July 10,1995.
- 20. WCAP-13054, Rev.1, "AP600 Compliance with SRP Acceptance Criteria," January 1993.
- 21. SSAR 6.2.1, Rev.13, " Containment Functional Design", June 1997.
- 22. F. Kreith, Principles of Heat Transfer, Second Edition, International Textbook Company.
- 23. R. F. Wright, D. R. Spencer, F. Delose, " Reactor Passive Containment Cooling System Small Scale Containment Cooling Tests", ANS/ASME Nuclear Energy Conference, August 23-26,1992, San Diego.
- 24. E. M. Sparrow and R. D. Cess, Radiation Heat Transfer, pp 45-52,1978, Hemisphere ,
Publishing Corporation.
- 25. I. K. Huhtiniemi, Condensation in the Presence of Noncondensible Gas: The Effect of Surface Orientation, Preliminary Thesis (1990), August 16,1993.
- 26. WCAP-12665, Tests of Heat Transfer and Water Film Evaporation on a Heated Plate Simulating Cooling of the AP600 Reactor Containment, Rev.1, April 30,1992.
- 27. WCAP-14048, Passive Containment Cooling System Bench Scale Wind Tunnel Test, April 29,1994.
REFERENCES Revision 1 j 3692w.wpf:1b-101497
, 6-3
- 28. WCAP-13294, Phase 1 Wind Tunnel Testing for the Westinghouse AP600 Reactor, April 30,1992.
- 29. WCAP-13323, Phase II Wind Tunnel Testing for the Westinghouse AP600 Reactor, October 2,1992.
- 30. WCAP-14%8, Phase IV-A Wind Tunnel Testing for the Westinghouse AP600 Reactor, June 6,1994.
- 31. WCAP-14091, Phase TV-B Wind Tunnel Testing for the Westinghouse AP600 Reactor, July 19,1994.
32. WCAP-13307, Condensation in the Presence of a Noncondensible Gas - Experimental Investigation, April 30,1992.
- 33. WCAP-13353, Passive Containment Cooling System Water Distribution Phase I Test Data Report, Rev. O, April 30,1992.
34. WCAP-13296, PCS Watei Distribution Test Phase II Test Date Report, April 30, D92
- 35. WCAP-13960, PCS Water Distribution Phase 3 Test Data Report, Rev. O, February 2,1994.
- 36. WCAP-12667, Tests of Heat Transfer and Water Film Evaporation from a Simulated Containment to Demonstrate the AP600 Passive Containment Cooling System, Rev.1, April 30,1992.
37. WCAl'-13566, AP6001/8th targe Scale Passive Containment Cooling System Heat Transfer Test Baseline Data Report, Rev. O, January 1,1993.
- 38. PCS-T2R-050, Large-Scale Test Data Evaluation, May 1995.
- 39. NTD-NRC-95-4563, " GOTHIC Version 4.0 Documentation, Enclosure 3: User Manual,"
September 21,1995. 40. WCAP-14382, "}VGOTHIC Code Description and Validation," May 1995.
- 41. NTD-NRC-95-4563, " GOTHIC Version 4.0 Documentation, Enclosure 1: Qualification Report," September 21,1995.
REFERENCES Revision 1 3692w.wpf lb-101497 1
\
I (A
- 42. NTD-NRC-95-4563, " GOTHIC Version 4.0 Documentation, Enclosure 2: Technical Manual," September 21,1995.
- 43. NTD-NRC-94-4174, "AP600 PCS Design Basis Analysis (DBA) and Margin Assessment," June 30,1994.
- 44. Adrian Bejan, Heat Transfer, p. 421, John Wiley & Sons,1993,
- 45. R. Siegel and R.H. Norris, " Tests of Free Convection in a Partially Enclosed Space
' Between Two Heated Vertical Plates," Transactions of the ASME, April 1957, pp. M3-673. i REFERENCES Revision 1 3592w.wpElb-101497
_ A1 APPENDIX A SYNOPSIS OF PIRT EXPERT REVmW SYNOISIS OF PIRT EXPERT IEVIEW Revision 1 3692w.wpf:1b-101497
A2 . APPENDIX A Synopsis of PIRT Expert Review An expert review was performed on the " Accident Spwification and Phenomena Evaluation for AP600 Passive Containment Cooling System" report (WCAP-14812, Revision 0) issued in December,1996. The experts consisted of personnel both extemal and intemal to Westinghouse as listed below: Extemal Experts Per Peterson, University of California - Berkeley - Tom Femandez, EPRI A&TRT Sol Levy, Sol Levy Associates (EPRI A&TRT) Doug Chapin, MPR Associates (EPRI A&TRT) Intemal experts Larry Hocluelter, Consulting Engineer Gene Piplica, AP600 Test Engineering Manager Larry Conway, PCS Patent Holder Terry Schulz, AP600 Systems Design Engineer These personnel were considered experts based upon their knowledge of heat and mass transfer mechanisms and the parameters related to these mechanisms, and their understanding of the AP600 containment design. The experts reviewed the Revision 0 report independently of each other and then two groups (EPRI A&TRT and Westinghouse) met with the objective of reaching conensus. Attachment A-1 is a copy of the letter sent to each expert, requesting their review of the Revision 0 PIRT, The experts provided comments on the identification of phenomena and the ranking of the phenomena. The following four paragraphs provide a general overview from the expert review (four different reviewers).
"In general, we found the organization of the material to be quite good. The PIRT itself is logically presented and most of the phenomena that can be expected to occur during PCS operation is included. Our major concem is the rationale given in the prragraphs following the PIRT that justify the selection of each ranking did not always accurately represent the basis for the selection. For many of the discussions, insufficient detail was provided to logically conclude the importance of the phenomena being discussed."
SYNOPSib OF PIRT EXPERT REVIEW Revision 1 3692w.wpf It-101497
A3 "The revised documents are a significant improvement over the draft I reviewed in April, and at this point all of my comments are of an editorial or a simple technical nature. With the new structure of the PIRT, it is now much easier to conclude that all phenomena with the potential to influence the AP600 containment performance have been identified." t "This version of the report is a significant improvement over the original version reviewed in March 1996. We believe the authors have made a valid effort to address most of our previous comments. In particular the report is much easier to follow and understand, is much better organized and more complete than the previous version. The authors added important new information which significantly strengthen the content. The report still has several structural and technical shortcomings. "
"I don't think the report is organized as well as it should be to give the leader the confidence that we have an integrated program which will successfully provide the basis for licensing the AP600 containment."
The general feedback from the expert review was that the basis of the phenomena rank'ng was unclear in Revision 0 (this was addressed in Revision I with the addition of paragraphs in section 4.4 detailing the ranking basis including references to expert review). The specific expert review comments were compiled and filed in the Westinghouse calculation note system. The new phenomena identified by the experts, as shown below have been incorporated into the paragraphs describing closely related phenomena. additional sources of mass and energy release such as ADS was added to the Break 5 Source Mass and Energy Release (item 1 A) e evaporation within containment as the pressure decreases and the heat sinks release their energy was added to Break Pool Evaporation (item 5B) circumferential conduction in containment shell from the dry to wet exterior surfaces was added to the Shell Conduction (item 7F) shield building heat capacity was added to the conduction through Shield Building (item 14B) The ranking of the phenomena by the experts is shown in Table A-1. The differences between the Revision 0 ranking and the expert ranking are highlighted. When the experts agreed with the Revision 0 ranking,it was identified in the table as such. It should be noted that the set of eight external and intern-i experts were not convened to achieve a group consensus due in part to time constraints and schedules. Also, it was felt that the expert SYNOFSIS OF PIRT EXPERT REVIEW Revision 1 3692w.wpf.lb 101497
A4 review was only one of many bases for ranking the phenomena (others included testing results, scaling analyses, and sensitivity studies) and that consensus among all eight reviewers was not necessary. However, as shown in Table A 1, two groups of experts (extemal experts 2-4 and internal experts 5 7), did achieve consensus on their ranking.
, The extemal experts for the most part agreed with the Revision 0 ranking. The internal experts, in performing their review independently of each other, had more differences than the extemal experts in ranking the phenomena relative to the Revision 0 ranking. Therefore, the intemal experts met in two half day sessions to discuss and resolve their differences on the ranking. The internal experts were asked to explain why they ranked the phenomena as they had such that the other internal experts and the authors could understand the basis for ranking. Each of the experts had an opportunity to express their opinions and ask questions during these sessions. The authors provided information as requested by the experts. The experts ranked the phenomena relative to the other phenomena since not all phenomena could be ranked the same,i.e., phenomena were placed in a hierarchy of importance. For I example, condensation on the heat sinks was ranked higher than the convection and ,
radiation to the heat sinks. The experts used a combination of personal experiences and ! knowledge to rank the phenomena - from testing experiences like the Water Distribution tests and Large Scale PCCS tests, engineering judgment, results from various analytic studies, etc. In total,45 of the '3 phenomena in the PIRT were discussed during these two sessions. The other 28 phenomena in the PIRT were not emlicitly discussed since they were ranked the same as Revision 0 ranking. Where consensus could not be achieved due to differences in philosophy or viewpoint, the differing or dissenting viewpoints (either among experts or relative to Revision 0) are documented in the notes provided with fable A-1. The experts were provided the opportunity to review these differing / dissenting viewpoints. The phenomena ranking by the experts was used as one of the bases by the authors for ranking the phenomena. The authors blended the expert rankings with the other bases, which included test results, scaling analyses, hand-calculations, and sensitivity studies to reach closure. In some cases, not all sources for the ranking of phenomena were in agreement and these cases required some judgment by the authors. The specific ranking _ basis for each PIRT phenomenon is provided in Section 4.4. 1 l SYNOPSIS OF PIRT EXPERT REVIEW Revision 1 3692w.wpfdt>-101497
A.5 Table A.1 Summary of PlRT Expert Comments on Phenomena Ranking a J 4
- EXTERNAL EXTERNAL INTERNAL INTERNAL RANKING IN I
PIRT PHENOMENON EXPERT 1 EXPERTS 2-4 EXPERTS 67 EXPERT 8 PIRT ! NOTE REVillON 0
- 1) Steak Source
] A . Mass and Energy same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 H,na H H.H 1A B . Direct 6on and Elevation same as Rev 0 same as Rev 0 H,na,H.L,H HnaH.L,H H,na,H.H.H IB C . Momentum same as Rev 0 same as Rev 0 H,na H,k,H H,na,H,k,H H,na H,H H- 1C
- i. '.
l D . Density same as Rev 0 same as Rev 0 H,na,H k,H H,naH.LH H,na.H.H.H 1D E Droplettiquid Flashing same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 H.H,na,na.na ) (thermal) j 2) Conteinment Volume
' A
- Mixing' Stratification same as Rev 0 same as Rev 0 H,H,H,k,H H.H.H.k,H H.H.H H H 2A B . Intercompartment Flow same as Rev 0 same as Rev 0 Lk,k,M L,H H,H.H Ht Ii H 28 C . Gas Compliance same as Rev 0 same as Rev 0 same as Rev 0
{ same as Rev 0 H H H.H,H 2C 1 ! 0 Fog (circulation) same as Rev 0 EH,H k,na Lk,M,na L k,M,na LH.H.H na 2D E . Hydrogen Release same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 L L L L, na f) Containment Solid Heat Sinks (Steel and Concrete) A Liquid Film Energy same as Rev 0 same as Rev 0 (combine w/ (combine w/ L,M,H.H,M 3A Transport , condensation) condensation) i 8. Vertical Film same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 L. L L L L l Conduction C . Horizontal Film same as Rev 0 same as Rev 0 LM,M LL,k,M LH,H,H,H 3C ) Conduction , D . Intemal Heat Sink same as Rev 0 same as Hov 0 M,H M H EH,M,H M.H,H.H,H 3D
- Conduction i
E . Heat Capacity same as Rev 0 same as Rev 0 M.H.k,L,H M,H H H H RH.M.H 3E F . Condensation same as Rev 0 same as Rev 0 M H,M.H EH.M,H. M,H,H.H.H 3F G . Convection from sam 6 as Rev 0 same as Rev 0 Lk,M.L LM,k,L LM.M.M,L 3G Containment l H . Radiation from same as Rev 0 same as Rev 0 LL M.L LM,L.L LM M,M.L 3H
, Containment
- 4) initial Conditions
- i. .
SYN _OPSIS OF PIRT EXPERT REVIEW Revision 1 ! 3692w.wpf-Itr101497 E
.. _ _ , . - _ _ _ - . - . . _ _ _ . _ - _ ._ , , . , _ _ _ . , _ , ~ . . . . . - , _ _ _ . - , _ ,
a. j A-6 4. l EXTERNAL LXTERNAL INTERNAL INTERNAL MANIONG IN
; PIRT PHEN 000ENON EXPERT 1 EXPERTS 2 4 EXPERTS $ 7 EXPERTg PIRT NOTE W
REVISION 0 A . Inmat Temperature same as Rev 0 same as Rev 0 k,M,HM.H H.M.M M,H M M,H,H H 4 1 i B . Initial Humidity same as Rev 0 same as Rev 0 k,M,H,MH HM-M,MH M M,H,H,H 4 l C . Inmal Pressure same as Rev 0 same as Rev 0 k,M,H.MH H.M,M,M,H M,M H.H,H 4 l $) Dreek Pool i A . MixirpStratifcation in same as Rev 0 same as Rev 0 same as Rev 0 same ts Rev 0 LL,L.M,L 6A the Pool ' l B. same as Rev 0 same as Rev 0 LLL.M,L LLL,L.L L6,M,M.L SB j CondensataVEvaporaton C Convecten with same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 L.L.LLL Containment Vol. j D . Radiaton with same as Rev 0 same as Rev 0 same as Rev 0 same as Res o LLLLL j Conterwnent Volume i E Conducion in Pool same as Rev 0 same as Rev 0 LLLk,L L.LLL,L L,L LM L SE I 1 F Compartment F1thng same as Rev 0 same as Rev 0 LL.LL,L LL,LL,L LLLM,L 5F f 6) IMWST (A through F) same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 L,LLL L 6 j
- 7) Steel Shell l A . Convection from same as Rev 9 same as Rev 0 L,kkk,L L,M,M M,L LL L.L.L 7A
! Containment l 4
- 8. Radiation from same as Rev 0 same as Rev 0 LL,L,LL L,k,k,L.L LM,M M,L 78 Conia.,meni C Condensation same as Rev 0 same as Rev 0 same as Rev 0 B,H,H,H,H LH.H.H,H 7C
]- D . Inside Film Conduction same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 LLLLL E . Insede Film Energy same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 M.M.M.M.M { Transport ! F . Conduction through same as Rev 0 same as Rev 0 LB,H H.H 8,LH.H H LLH,H.H 7F
- q. G . Heat Capacity of Shell same as Rev 0 same as Rev 0 same as Rev 0 H,M,M,L,H LH,H,LH 7G I
H Convection to Riser same as Rev 0 same as Rev 0 LLL.M,L LL,k,M.L L,L,M M,L 7H Annulus I . Radiation to Baffle same as Rev 0 same as Rev 0 L,LL M,L LLL.M,L L.LM,M,L 71 J . Rachaton to Chimney same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 L.L,LL.L SYNOPSIS OF PIRT EXPERT REVIEW Revision 1 3692w.wpf;1b-101497
+
r,e--,, ,- , - - - , . , e ---m-m-~-+-- ,,o -----e-- a -
. - . _ _ . - _ _ _ . _ . . _ _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ _ __m._._._.-___.
t a i i A7 4 { EXTERNAL EXTERNAL INTENNAL INTERNAL RANKING IN PIRT PHENOMENON EXPERT 1 EXPERTS 2-4 EXPERTS 6-7 EXPERT 8 PIRT NCTE REVISION 0 K . Radiaton to Fog' Air same as Rev 0 same as Rev 0 same as fiev 0 same as Rev 0 L L.LL.L j itixture J l L . Outside Film same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na na,LLL 7L j Conducton 1 < M . Outokse Fdm Energy same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na.na,M,M,L 7M Transport N . Evaporaton to Rise same as Rev 0 same as Rev 0 same as Rev 0 ta,na,H,H M same as Rev 0 7N Annulus i 8) PCs Cooling wowr i i A . PCCWST Flow Rate ' same as Rev 0 same as Rev 0 same as Rev 0 sarno as Rev 0 na.na,H,H,M 8 I B . PCCWST Water same as Rev 0 same at, Rev 0 same as Rev 0 na.na,M,M L same as Rev 0 8 j Temperature 1 C Water Film Stabihty same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na.na,H H,L 8 1 and Coverage J D . Film Strippmg same as Rev 0 same as Rev 0 same sa Rev 0 same as Rev 0 na.na,L,LL 8 ! E . Film Drag same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na na.L.LL 8 } } 9) Rieer Annulue & Chimney Volume i l A . PCS Natural Circulation same as Rev 0 same as Rev 0 same as Rev 0 L.LMR,M LLH,H,M 9A l B Vapor Acceleration same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na.na,LL,L 98 i j C Fog same as Rev 0 same as Rev 0 same as Rev 0 sa.no as Rev 0 na,na,LL,na i D . Flow Stabihty same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 ' LL.L.LL i i 10) Baffle i
- A Convect <m to Riser same as Rev 0 na,na,LM,k na.na,LL,L na na,LLL na.na,LM,na 10 Annulus
- B . Convection to same as Rev 0 same as Rev 0 na,na,LL,L na,na,LLA na,na,L,M,na 10 f- Downcomer C Radiatiori to Shield same as Rev 0 na.na LLL na.na,LLL na,na,L,LL na.na,LLna 10 Building l'
D Conduction through na,na,Lk,na same as Rev 0 na.na,LLA na,na,LLL na.na,LM,na 10
- Baffle E Condensation same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na,na LL,na
! F . Heat Capacity same as Rev 0 na na,LL,na same as Rev 0 same as Rev 0 same as lov 0 4
SYNOPSIS OF PIRT EXPERT REVIEW Revision 1 3692w.wpf:1b 101497 4 2 g-we,+c -.,,r----,----..-~--v, - m,--w
, . , . e .-.-.=r -...--m+ w v. -- - -,2-i. ,,-,..-.--,www ., y- - - , - - - - , , ,,-..,3,-vmv.-,- ---v--,- - - . - - - - - ,
A-8 EXTERNAL EXTERNAL INTERNAL INTERNAL RANKING IN PlRT PHENOMENON EXPERT 1 EXPERTS 2-4 EXPERTS 5-7 EXPERT 8 PlRT NOTE REYlSION 0 0 Leaks through Baffle same as Rev 0 same as Rev 0 na.na LM na.na.LLL na.na,M,M.na 10
- 11) Baffle Supports A Convecton to Riser Air same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 LLL.LL B . Radiation from Shell same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 L.L.LL.L C Conduction from Shell same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 L.L L.L.L 4
0 Heat Capacity same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 LL.LL.L
- 12) Chimney Structure A Conduction through same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 LL,LL,L CNmney 4
B Convecton from same as Rev 0 samo as Rev 0 same as Rev 0 same as Rev 0 LL L,L L Chimney Air C . Heat Capacity of same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 LL L.LL Structure D . Condensation on same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 LL,L L.L CNmney
- 13) Downcomer Annulus A PCS Natural Circulation same ss Rev 0 same as Rev 0 same as Rev 0 LLjiM M L LH,H.M 13A B Air Flow Stability same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 L.L.L.L L
- 14) Shield Building A . Convection to same as Rev 0 same as Rev 0 na,na.LL.L na,na M.L na.na,M,M L 14A Downcomer B Conduction through same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na,na,LLna SNeld Bldg C . Convection to same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na.na,LL,na Environment D . Radiaton to same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na na,L,Lna Environment
- 15) External Atmosphere A . Temperature same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na.na,L,L.L B Humidity same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na,na,L.L.L C Recirculation same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na na.L.LL SYNOPSIS OF PIRT EXPERT REVIEW Revision 1 3692w.wpf;1b 101497
A9 EXTERNAL EXTERNAL INTERNAL INTERNAL l RANKINGIN PIRT PHENOMENON EXPERT 1 EXPERTS 2-4 EXPERTS 67 EXPERTg PART NOTE REVISloN 0 0 Pressure Fluctuations same as Rev 0 same as Rev 0 same as Rev 0 same as Rev 0 na.na.L.L.L Note: Bolded rankings signify differences with the PIRT Revision 0 ranking. Differences are explained in the following paragraphs Expert Ranking Basis: 1A) Break source mass and energy represent the driving force (it is the boundary condition like decay heat for PXS analysis) for each time period and should be ranked High even through there is a significant reduction in the M&E release between blowdown and long term for the LOCA. IB) Due to the signiiicant reduction in mass and energy release during the long term period of LOCA, the break source direction has Low importance. It was also noted that the ADS stage 4 provides a controlled break release location and direction once they open. 1C) Due to the significant reduction in mass and energy release during the long term period of LOCA, the break source momentum has Low importance. 1D) Due to the significant reduction in mass and energy release during the long term period of LOCA, the break source density has Low importance. 2A) Due to the significant reduction in mass and energy release during the long term period of LOCA, the mixing and stratification effects within containment have Low importance. Discussions indicated that since not much energy is being released into containment relative to the state of the containment volume, that mixing / stratification effects would not have a major impact. Also, dP/dt is negative during the long term period (post peak pressure). 2B) Expert 8 stated that intercompartmental flow was of High importance for blowdown (LOCA and MSLB) since it establishes the " initial conditions" for the remainder of the transient, but of Low importance for the remaining time periods of LOCA. Other experts did not feel that this initial condition was important. All agreed that details of flow paths were not important during blowdown since volume compliance or energy stored in containment volume was dominant and the time constant for heat sinks is long compared to the 30 second LOCA blowdown. It was agreed that gas content (steam or air in item 2A) was important, but the SYNOPSIS OF PIRT EXPERT REVIEW Revision 1 3692w.wpt1M01497
A !O rate at which it moves is not as important relative to the tiny. to reach peak pressure. 2C) Experts agreed that containment free volume was a more accurate description of the energy storage capability of the containment. All experts felt that smce the
" compliance" consists of gas properties which an well defined, and free volume, which is a known design parameter, that most appropriately this item could be ,
deleted entirely since it is not strictly a process. (It was also noted that the primary system volume does not appear in the PXS PIRT.) 2D) Extemal Expert Comment: Since fluid issuing from the DBA break is postulated to act as the fog source (broad drop size spectrum) and since fog thermodynamic and thermal properties are significantly different from steam or a steam / air mixture we would expect this item to be ranked High for blowdown since these properties could significantly lower the containment pressure history. (authors note that this most likely refers to item 1E droplet / liquid flashing) During long term cooling, it seems possible that fog conditions might form in a region cooled slightly below the dew point. Thus, the ranking would be Low. Intemal Expert Comment: The intemal experts agreed that fog was not a controlling factor on the containment pressure response during any time period of the LOCA and should be ranked of Low importance. Drops do not appear for the superheated MSLB event. 3A) Experts agreed that film energy was an element of the condensation process and should be combined into the discussion on condensation (items 3A,3D should be included in 3F to be consistent with traditional definition of condensation which includes the film resistance along with mass transfer resistance). 3C) Expert 8 felt that the upper surface of horizontal films would thermally saturate very quickly, therefore horizontal films were considered to be of Low importance during the transient. 3D) Expert 8 stated that initially the heat sinks play an important role (in blowdown and refill) but they decrease in importance as the heat sinks become saturated (peak pressure and long term). Expert 8 felt that heat sinks are the only mitigating SYNOPSIS OF PIRT EXPERT REVIEW Revision 1 3692w.wptib 101497
~ . - _ - - - _ - _ . - - .
A-11 fecture (besides containment volume), so it must ,e ranked High for blowdown. t Experts 5 7 felt that the heat sinks were of Medium importance during blowdown. 3E) All experts agreed that the heat capacity ofinternal heat sinks should be rated t same as conduction (3D). 3F) All experts agreed that condensation of internal heat sinks should be rated same as conduction (3D).
/
3G) All experts agreed that convection to heat sinks was less important than condensation. 3H) All experts agreed that radiation to heat sinks was less important than condensation.
- 4) Experts agreed that the initial cranditions increase in importance as the LOCA transient proceeds (from blowdown to peak pressure), is Medium importance for the long term; and is of High importance during the relatively long MSLB blowdown (600 seconds).
5ABEF) Expeits felt that the break pool had much less importance than the solid heat sinks and that the phenomena should be ranked Low for all time periods, except for the mixing and condensation which should be ranked Medium during the long term phan of the LOCA after the break pool fills with water. Th'. effects of cold water spilling into the break pool and ADS 4 actuation (during the long-term period) were considered. However, expert 8 felt that condensation on the break pool should be ranked Low for the long term period of the LOCA since the top of the pool is expected to be saturated.
- 6) The IRWGT is too isolated to affect the contal;unent pressure (recognize that liquid to core cooling is included in the M&E).
7AB) Experts felt that convection and radiation to the inside of the steel shell were much less important than condensation on the shell and should be ranked Low for all time periods. This is also consistent with solid heat sinks (items 3G and 3H). SYNOPSIS OF PIRT EXPERT REVIEW Revision 1 3692w.wpf:1b 101497 -
A 12 7CFG) Experts felt that condensation on the shell was of Low importance during the short LOCA blowdown period (30 seconds) except for expert 8 who felt that since the shell has a very large surface area, high thermal conductivity and is rather thin, it should probably be the most important heat sink during blowdown. All experts felt these phenomena were of High or Medium importance for all other time periods including the relatively long MSLB blowdown (600 seconds). The conduction through the film and heat capacity of the shell were ranked the same as 4 condensation, except for the heat capacity which is " consumed" by the long term and is ranked Low. 7HI) Experts agreed that convection to the riser and radiation to the baffle during all periods should be of Low importance except for the long term which was ranked Medium due to the higher dry fraction and, hence higher temperature of the shell. 7LMN) These 3 phenomena were not applicable during the blowdown and refill periods since water coverage was not initiated. After water coverage started (peak pressure period), the experts agreed that film conduction was of less importance than the film energy (experts felt that a more appropriate name for this film energy was sensible heat); and that evaporation was of High importance for the peak pressure and long-term periods. For MSLB blowdown, film energy and evaporation were rated lower than the LOCA due to the timing (600 seconds for MSLB versus 1200 seconds for LOCA peak pressure).
- 8) Experts agreed with authors on ranking for each of the 5 phenomena for the PCS cooling water. It was recommended that authors include a full description of the
, phenomena and parameters affecting film stability and coverage. 9A) Experts agreed that natural circulation through the riser was of Low importance during the blowdown and refill periods of the LOCA since there was an insignificant amount of heat transfer from the shell; but was of High importance during the peak pressure and long term periods when the heat transfer (especially evaporation - item 7N was highly important) was significant. However, expert 8 felt that it was of Medium importance during the peak pressure and long tenn periods since the LST tests did not show any sensitivity to velocity. SYNOPSIS OF PIRT EXPERT REVIEW Revision 1 3692w.wptib-101497
i A 13
- 98) Experts agreed that vapor acceleration (via water being evaporated from th shell) was not applicable during blowdown and refill since there was no water coverage; and was of Low importance during the other periods when there was water coverage.
- 10) External Expert Comment (items A,C): For MSLB, the text ranking is " Low" but is not consistent with the table 4-1 ranking of N/A. We believe the rank of Low is correct since the phenomena will occur but is not very important.
External Expert Comment (item D): The baff'e is 1/8-in. thick steel with a very small Biot number of 0.0004. Thus the baffle behaves as a lumped mass with negligible internal resistance for all time phases. A low Biot number says that conduction resistance is small compared to convection (and should be ranked as being low importance). Internal Expert Comment: The internal experts agreed that the baffle phenomena was not applicable during the blowdown and refill periods since there was insignificant heat transfer; and was of Low importance during the other periods when there was heat transfer. Leaks through the baffle are expected to be small but need to be quantified. 13A) Experts agreed that natural circulation through the downcomer annulus should be ranked the same as natural circulation through the riser annulus (item 9A). , 14A) Experts agreed that convection from the shield building to the downcomer was not i applicable during blowdown and refill since there was an insignificant araount of heat transfer from the shell; and was of Low importance during the other periods when there was limited transfer in the downcomer volume. SYNOPSIS OF PIRT EXPERT REVIEW Revision 1 369? v.wpf;1t> 101497}}