ML20212F773

Non-proprietary Assessment of Effects of Wgothic Solver Upgrade from Version 1.2 to 4.1

ML20212F773
Person / Time
Site:	05200003
Issue date:	10/31/1997
From:	WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML20212F765	List: ML20212F773 NSD-NRC-97-5396, Forwards non-proprietary WCAP-14968, Assessment of Effects of Wgothic Solver Upgrade from Version 1.2 to 4.1
References
WCAP-14968, NUDOCS 9711050087
Download: ML20212F773 (84)

Text

Mnwm?w&us;GWEM_s 3_;%

ERng,gyl Arm o:@

j hg yw wn: d * - :w p$DtQ$h%

Mtv;W SW;R,A:mc m

aMa Nm p!em$:97g);W%'

&s n r, y

41-u fgg MQ~p; hip w ww

=pk N

w.s #w%

&fy %jtv7:"D%Qfu%

MA M

&y A

8 Q%%geiW Jj,$;gyT4dw cznwn iT O e " y, -;s#:s&

-W A:ph-@Wwc x' -

ra,W W W # w / %

wgPng4;gwp Whre4ig,yMjg!O

,A vq T h

SppfM p!lJf figyp x h

A:%qm-m jd wN;bQ WW

& , . . 'My/1

%w[,,Mjg$nh y;

- am m w%n a w n(an n n 47 dM MjWpf  ;' n%

i  ?

q @vHp3 if $$$ pMpWW$p%Nhih4 p$

ut>sbg.pa*y ="o;ewy Lu.

k 4 . .n

,u. &,:-

mmynwgy .

s m e A$k 4~ '

M nPW,2::Q+W

$pf@@gp@@f fS hm m a #n43 vm*

1. A n

hl 6

i $  ?!ffSMd$!!

y%,h,hkeAm hh >

8N a

4 wsm hww$eww mu

$,qNMyiW@%g$xm.

- 4gdpHK e m %g ug&waese t p%%gApp ymmanwwnum dMk d m wp m

hig qlg;e;w mMg& ge wswa3wmes w egm

% @p e%y g4 e&lvPg Wm 40 NWae;fppfJi gee imew sw mumpg 44, PM

^

AMCK 05200003 ew

% M h M p+

kmme d M % h&[ k N +D Q M M 1

sgs enwamageng med g gg,gg_g& g _g gg s mm,g gv.

m-r----'----, ' " ~ ~ ' - - - - ' ' '

Ft; , v--({L ' Ui.,n..M#,

'-'m"-~~-'"~'tO '" [

' ' . . i . = rA T i,-,4,[. . . " . s >4 3 .M M ,l'*E e ' +

w ; g c q? "5 r3r e',--;  %,;'c ,- H->& v j_ 1.,

s . '. .' U;% =g-f

3,o4.;> l 2- -

g 3 , 5' ;d. t, t A g' g ) '.

j<

.ar n.'

' 7 t; 'v3' i,

• -s

• w . . , t,

, t T*M- L. 1 ; - ., tw' pM3 , ,t- ' l ?i 's'd :4 PW t > , * 'J: E

[' -

y b 7 c  : s op.

f g V

.$ r.

, ?O yO. 3y% f 3 } j , g%

c:

~ y1 j s gl .[ fJ g *

- 4 4.

, a' ' , "

pc p- ,

\

4'

?l i:.-

Q \\ t

,a-.sk t t,  ? -N f.s. ' ~ $ '* y  %  ;

3 '

) i . "?' U:M ' j -( ' L. 4h a 'I' '-a 4  !

d '

n ' h'

![ y

[' -

i ; r. h '

?' ,

.,,i)'S,_q.. (4' j .;

F 3[' ,

t.

gj.

. ' ' , di

~-- 'Q_

5-

'}

s -f g.

3 Q'Wstinghouse  : 'tk(

N6n Proprietary, Class 3 L i -

m 1

a' 3*.. 3 .,

,Q.

• y -

-j y

~

si}y- j _^ +

7 .

M..1 -- sv psg; e 4 3 < vc

( )

1 1g's 4 <l: 4 i 1 f g 4 4 - %D '

yj7+ yg 4

, 1 = -

, _ _ { --

1. ... n . , 3-_ ,

r ' 'p, c

• A 4J de t

.' :; . )

.' O\' 4 h -

,{ ..j _  %

t  %

,S 4i y), h ' ym I

S Te

,, ) e-F I

![ f b g 4 $( +*

g ._

4

' 7/t ,

v

.,=,4 f L

j.  %

i k ($ RV -- ' - ',,'l

[ [',d

& 2 n

k z '. ; v(-,:. , >~ , .j A A c

af.

9' , # 4

' f.N .f , A f

/

t' 8

W ["Ea.I-'" ."~ , ' ,,, . , , p s 1 4

...1"".$.

j

! , I-d [.7 , , , .- - 0 iM- '

} ) r 73 1 4 Ii s g a .N rg l:

~ ~

_-~~,~'

s l p- . 5  % ye w f'

a>

4 t 4

t f s gs e +

1 5 + ,

_yg , '_ ,J. , _a m' , . % s ,) o i $ -

' s n

4. .

. m St n y , .,\ ,

%m . . .<<G- e,,, + '

i f n%g- ^3 Gs '.w! ,'

ar y ,- .t s

r. >

b 1,a e -.

• r

?p , ,

i

, < rQ_g ;L .o .s . n, w ,** _ , ,

&; j , .. p s g 't  ;

1

-4 n

, , 8 t  :::

5 +yn 3 y' -

,',, ,' y-_.> ; ' zr

,,.{y i y .% e y -- t y

Y n,g.-  % ,s

- , } E^ '

( .

jQ- t : >? q ~ .l$ * * *g 4

j' ~

g,4 g

('

i.s4 mj $e - y*- h a

nw< # *

• s f,y, v; + t i.

r ua ' r  %* :4 e a h 47 fe

' 'h &! L'

':l-Y- -p:

r9, ,

4 - v< y i,

,s ( ,.,s g-~

1 1, .

• "r v ,-

-4g 7 'Y9 w .,,3 3

v.

+o w

, n, - n m A -+ , . s u3,r s' 4 +

,y ,

t g e

~,

o  % --

f q~m m '

s c: , .a w

T;.~O D M ,,., u, and&Effectb.oM

~u, E

m:s <

4

~

ess s

. - . - ,- F

'- , . .% .: * " < d- Es + '

g~.

W'q

.s e a .-

k g $ ) } 1 1,4V '-

-I k'lVIr-%,.tY'WGO"' THI'C' S l 'b.kYo vek-4

_ .,,,j_

., c 3 [;

.g 4 a

.y.

2, i r L

[

p _

b 4y

.iL 1-

?< ..f p

'1 J. , ?A ,, g. A 4 Y

$ 4.

%~, yy

~

9 sA , -

qUp, -

(;- 3 g,r'

> n. s p . 5 ad@fr, i g , ~g - , ,

y OC Q 4 >

~

pp J s M,~m > ggg Q g Qpg g ,g { g , y 3 s. y o g,

_ 3

% y,. , , a * . - r ,- ,

. . ~ .; y,y .

- . y -<('-

(. _ - 4 1.o .{. s mm.4.q (7q v ' ~^

w a ip .

r' i

r ;. a L i f.'1. :f,;.. y L *

, r ,

b_ x .

1 R.  ?

k

. ?

JV. [QJ I,(%{\ \ . ^*

et-R ~

g'( 4 JQ n &m a 1

• - , ; g(pL mLv a g' ,

t -

uruo - 7:a=r 4, 1

V, .

i .. .

" c. }3) - Wm 3 r

< y. q c

s a

• f .G .:~ n M' N '

• UM s et t wt. y C 4 Cs :'hw A a .1' q ' E>

4p %xx vp . 2; ,' ' na 1

s w ,-

,1 .

's

+: x , y~y

, r wL. y W.

8 yg

~

< 1 -

1_;y t, , A* '

s / <

r 7g- - g 4;:p: ..q 'H >

] >

dq $ ; 4 3

.t

. **7" j 3

i @jo.N.. =- ' li ? .uy1-3

,) D 3 Y,'

j yF

• 4, ym ij ', s (

+ =

y M, p t; 4 f ., .

s

.' :. .y f .,3 ' (g ird ,4.. !i &, t ,3 y

, i,pt

dy a Vr l 'y 1.,

i- ,

.t Q0 ': -' 7'_'

[*3 ' ,

j,t 4;M.a j sy> . (. ' s  ;

< a

-M '* /, @4.Q';n, gG ,% ? 1 $' ~ 4,,

) c I ,;a; ne L ; 5' 1

L. . u.

s

~

%~*

' ,~i' w

1 u.

r

, .a

,fi p o _

S i

1q.;3 S ~

a b

.. 4 fM )

a e @+ r

, u m %n , ~

s+

= i (4, ;. 4

? s V: A:

sr, v -

L e - ' Ji13 5.k . ..3p* ,

g' , _ < ~'xms, * , ?+ .g

$ U tI p. . .

3 A7 ,,

( f A Ih' 3 4p I ;I g

-9',3 , g-4 m;

, , 4

_ >i;. ** a ,.-.

. >Fi i

e'

%.--( n. -'.;&a

. -. . , y s. , ,

, e

)g kT t t '

.'l

.- 5 33 y' e.- ..t.

F o

k

s. '..>e n:5 ,

e .y .cM.

=_' i.f]O ( c; y d5 9  ; T

.'E 1 _ ~'

, s

,]

s' 5- L i

a

\

., ?

'.' h{. .'

,7 [p-f /j '

. - Q:Q'j i , p'N- , +

=y'4r'(b "g t -

k E --

pf.

1 h-L

v- f fj.

? 'I ' ' f[ 9^ $&  ?

,^

%@  ; 6

j. ,l %y g; gg w* * .e +o s

,'@<. .kW w. m,

+

' ? M.

o y 9. .. . . ,-w

%g+:e up', @$W.s -

4 n q * '

+  : 4 3 M,

p'f '; +

4 7; - g g q 1 &

. n *

, , ', y J44<9 '2 i

t;

~n-

.1 jy> ,-

41

' %.d A M' k ' #

, g ~.

• < -s .

y q[ -

g jp % i 3 s y e. *

,,.4

,,e t 2 '.1 f 4 j

• b { ' f, ' g d g-

~

is

1, 3 4 p

,s +

s ]w 5 a. st. zwe ,; . . + 5 A rs s s*,

t e. m:pyt

'g'., ..w -.!,

• u t:

7 ,' j p ,

t,,., -

g ( . ,

, q..,e a_  :

(b. #

. ..,-,.;# s I

y 7

.) $

k' 4

% g

- 4' '

p . ..

_ t k 4-

, , ) I

,*s. r a

i kl ,,,,-

m -

p /

r h.p gg We'stt [nlg hoju s ,+,s ms ,s , , s ..

e? Enfe r g-yc.S y s t e m s -

w-m ,

w_

o, v.p, ,,

q q a .? N '

M

<W&  % '

~,7 ' . q' Q+QQ"LX -

y t :s [e a n _ m, s

L' '

+ "

s g . bpmD.s - - % --

'gy:,T4 9711050097 971023 i PDP ADOCk 05200003

,f n VDH a g y%W n .g. , . . ~ ,, . 7<

n. .

l AP600 DOCUMENT COVER SHEET i TDC: IDS: 1 S Form $8202G(5/94)lo:V.1837n.frm:1b) AP600 CENTRAL FILE USE ONLY:

0068 FRM j RFSs: RFS ITEM s:

doc 0 DOCUMENT No. REVISION NO. ASSIONED TO PCS GSR 040 0 Page 1 of 2 ALTERNATE DOCUMENT NUMBER: WCAP 14968 WORK BREAKDOWN #: 3.3.2.20 DESIGN AGENT ORGANIZATION: Westinghouse i PROJECT: AP600

! TITLE: Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1

, ATTACHMENTS:

DCP #/REV. INCORPORATED IN THIS DOCUMENT REVISION:

l CALCULATION / ANALYSIS

REFERENCE:

1 ELECTRONIC FILENAME ELECTRON C FILE FORMAT ELECTRONIC FILE DESCRIPTION l 3837w.wpf WordP%ect 5.2 Text i

(Q WESTINGHOUSE ELECTRIC CORPORATION 1997 i

OWESTINGHOUSE PROPRIETARY CLASS 2 TNs document contains informaton proprietary to Westinghouse Electric Corporation: It is subrnetted in confdence and is to be used solely for the purpose for wNch it is furnished and retumed upon request. TNs doovment and such information is not to be reprodvoed, transmitted, disclosed or used otherwise in whole or in part without prior written authorization of Westinghouse Electric Corporaten Energy Systems Business Unit, subject to the legends contained hereof.

OWESTINGHOUSE PROPRIETARY CLASS 2C TNs document is the property of and contains Proprietary informat6on owred by Westinghouse Electric Corporation and/or its suboontractors and suppieers. It is transmitted to you in confideru:s and trust, and you agree to treat this document in strkt accordance with the terms and condisons of the agreement under which it was provided to you.

[ WESTINGHOUSE CLASS 3 (NON PROPRIETARY) a 4

. COMPLETE 1 IF WORK PERFORMED UNDER DESIGN CERTIFICATION QS COMPLETE 2 IF WORK PERFORMED UNDER FOAKE.

l EDOE DESIGN CERTIFICATION PROGRAM GOVERNMENT LIMITED RIGHTS STATEMENT (See page 2]

Copynght statement: A license is reserved to the U.S. Government under contract DE-AC03-90SF18495.

ODOE CONTRACT DELIVERABLES (DELIVERED DATA)

Subject to spedfied excepsons, diadoeu o of this data is restrteted until Sepiomber 30,1995 or Design Cerencation under DOE contract DE ACO3-90SF18495. whichever is later.

EPRI CONFIDENTIAL.: NOTICE:-1 C 2 0 3 E 4 O s O CATEGORY: AE BO CO DD-eO FO CARC FOAKE PROGRAM - ARC LIMITED RIGHTS STATEMENT (See page 2) 2 Copyright statement: A license is reserved to the U.S. Government under contract DE FC02 NE34267 and subcontract ARC-93-3-SC 001.

DARC CONTRACT DELIVERABLES (CONTRACT DATA)

Subject to speerfied exceptions, disdoeure of this data is restricted under ARC Subcontract ARC 93-3-SC 001.

ORIGINATOR SIG TE f

'J. Woodcock AP600 RESPONSIBLE MANAGER

/

TV

~

4 '

lOfGIQ APPROVAL DATE

-J. A. Gresham A lJ Cl

' Approval of the respono6ble manager segnrfies_ document is comp 6ete all required reviews are complete, electronic file is attached and document le released for use, cose.nen m t

.--- - . , _ . ~ . _ , - . - _ . - _ . . .

_ - - - . - - _ ... - ~ . - - - ... -

AP600 DOCU"ENT C;VER SHEET Page 2 Form 68202O(W4) LIMITED RIGHTS STATEMENTS DOE GOVERNMENT UMITED MlOHTS STATEMENT (A) These data are submitted with hmited rights under govemment contract No. DE AC03-90SF16495. These data may be reproduced and used by the govemment with the express hmitabon that they will not, without wntten permission of the Contractor, be used for purposes of manufacturer nor disclosed outside the govemment; except that the govemment may d solose these data outside the goverrvnent for the fobowing purposes, if any, provided that the goverrgnent makes such 6sclosure subject to proh/tNtion against further use and esdosure:

(1) This 'Propnetary Data' may be disclosed for evaluabon purposes under the restrictions above.

(ll) The "toprietary Data' may be esclosed to the Electnc Power noseardi institute (EPRI), electnc utihty representabves and their erect consultants, exclu$ng erect commercial competitors, and the DOE National Laboratortes undar the prohibibons and restnctions above.

(B) Ths notice shall be marked on any reproduction of these data, in whole or in part.

ARC UMITED RIGHTS STATEMENT TNs proprietary data, fumished under Subcontract Number ARC-93-3-SC-o01 with ARC may be duplicated and used by the govemment and ARC, subloct to the hmitabons of Article H 17.F. of that subocf1 tract, with the express hmitabons that the proprietary data may not be esclosed outside the govemment or ARC, or ARC's Class 1 & 3 members or EPRI or be used for purposes of manufacture without pnor perrnission of the Subcontractor, except that further @sdosure or use rney be inade solely for the following purposes:

TNs proprietary data iney be disclosed to other than commercial competitors c

• Joontractor for evaluation purposes of tNs subcontract under the teetncbon tnat the proprietary data be retained in confidence and not be further declosed, and suttoct to the terms of a non-esclosure agreement between the Subcontractor and that organizatson, exdud ng DOE and its contractors.

DEFINITIONS CONTRACT /DEUVERED DATA - Consists of doeurnents (e.g. specifications, drawings, reports) which are generated under the DOE or ARC contracts which contain no background proprietary dafa, EPRI CONFIDENTIALITY / OBLIGATION NOTICES NOTICE 1: The data in this document la sutsect to no conhdentiahty obhgations.

NOTICE 2: The data in this document is propnetary and confidential to Westinghouse Electric Corporabon and/or its Contractors. It is forwarded to recipient under an obhgation of Confidence and Trust for hmited purposes only. Any use, esc 60sure to unauthorized persons, or copying of this document or parts thereof is proNbited except as agreed to in advance by the Electnc Power Research Institute (EPRI) and Wesbnghouse Electnc Corporat on. Recipient of this data has a duty to inquire of EPRI and/or Westinghouse as to the uses of the informabon contained herein that are permitted.

NOTICE 3: The data in this document is oroprietary and confidentiallo Westinghouse Electric Corporation and/or its Contractors. It is forwarded to recipient under ar' obligation of Conf donce and Trust for use only in evaluebon tasks specifically authonzod by the Electnc Power Research Institute (EPRI). Any use, diodoeure to unauthortzed persons, or copying tNs document or parts thereof is prohibited except as agreed to in advance by EPRI and Westinghouse Electnc Corporabon. Redpient of tNs data has a duty to inquire of EPRI and/or Westinghouse as to the uses of the information contained herein that are permitted. This document and any copies or excerpts thereof that may have been generated are to be retumed to Westinghouse, directly or through EPRI, when requested to do so.

NOTICE 4: The data in tNs document is propnetary and conhdential to Westinghouse Electric Corporebon and/or its Contractors. It is being revealed in confidence and trust only to Employees of EPRI and to certain contractors of EPRI for hmited evaluabon tasks authonzed by EPRI.

Any use, diodosure to unauthorized persons, or copying of tNs document or parts thereof is prohibited. This Document and any copies or excerpts thereof that may have been generated are to be retumed to Westinghouse, directly of through EPRI, when requested to do so.

NOTICE 5: The data in tNs document is propnetary and confidential to Westinghouse Electric Corporation and/or its Contractors. Access to this data is given in Conhdonce and Trust ordy at Westinghouse facihties for hnuted evaluatson tasks assigned by EPRI. Any use, disclosure to unauthonted persons, or copying of tNs document or parts thereof 6s prohibited. Neither this document not any excerpts therefrom are to be removea trom Westinghouse facehties.

EPRI CONFIDENTIALITY / OBLIGATION CATEGORIES CATEGORY 'A*- (See Delhwed Data) Consists of CONTRACTOR Foreground Data that is contained in an issued reported.

CATEGORY T- (See Dehvered Data) Consists of CONTRACTOR Foreground Data that is not contained in an issued report, except for computer programs.

CATEGORY 'C'- Consists of CONTRACTOR Background Data except for computer programs.

CATEGORY 'D*- Consists of computer programs developed in the course of performing the Work.

CATEGORY T- Consists of computer programs developed prior to the Effective Date or after the Effeebve Date but outside the scope of the Work.

CATEGORY 7"- Consists of administrative plans and administrative reports.

.M _ _ _ _ _

4

O WESTINGHOUSE NON PROPRIETARY CLASS 3 WCAP 14968 t

l Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1097 Westinghouse Electric Corporation Energy System Beisiness Unit P.O. Box 355 Pittsburgh, PA 15230-0355 01997 Westinghouse Electric Corporation All Rights Reserved Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 oA3837w.non:1b/101097

I t Tatdo of Contenta List of Figu r es . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

, 1.0 In troductio n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 l

2.0 Assessmant of Code Changes on Evaluation Model Results . . . . . . . . . . . . . . . . 21 2.1 LOC A D E C LG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 2.2 MSLB...................................................24 ,

3.0 Applicability of LST Code Validation Conclusions (solver 1.2) to Development of Evaluation Modei (solver 4.1) . . . . . . . . . . . . , , . . . . . . . . . . . . 3-1 -

l 4.0 Impact on WGOTHIC Separate Effects Tests Calculations . . . . . . . . . . . . . . . . . 41 5.0 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1 6.0 4

Refe rence s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6- 1 Appendix A Summary of WGOTHIC Solver Changes from Version 1.2 to 4.1 . . . . .....A1 1

L Appendix B Comparison of LST Resuhs Using Solver Versions 1.2 and 4.1 . . . . . .....B1 '

i e

Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\383?w.non:1b/101097

_, ,y p- +-

r+-e-NPPJ 19Y "7 P' # #

ll List of f Jures

(

Figure 1 Noding Diagram for Wes' . Quarter Annulus Cross-Section Figure 2 WGOTHIC Calculated .%. Pattern LOCA Jan Momentum Dissipated in SG East Compartment at 8000 sconds (Version 4.1)

Figure 3 LOCA DECLG Calculated Pretsure Comparison of WGOTHIC Solver Version 1.2 and 4.1 Figure 4 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell-Volume 49 Figure 5 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell -

Volume 50 Figure 6 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell-Volume 51 Figure 7 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Ghell-Volume 52 Figure 8 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell-Volume 79 Figure 9 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell-Volume 80 Figure 10 LOCA DECLG Calculated Steam Pressure Railo Near the Steel Shell-Volume 81 Figure 11 LOC A DECLG Calculated Steam Pressure Ratio North CMT Cavity -

Volume 6 Figure 12 LOCA DECLG Calculated Steam Pressure Ratio South CMT Cavity -

Volume 104 Figure 13 LOCA DECLG Calculated Steam Pressure Ratio Upper East Steam Generator Compartment - Volume 4 Figure 14 LOCA DECLG Calculated Steam Pressure Ratio Lower East Steam Generator Compartment - Volume 107 Figure 15 LOCA DECLG Calculated Steam Pressure Ratio Upper West Steam Generator Compartment - Volume 5 Figure 16 MSLB Calculated Pressure Comparison of WGOTHIC Solver Version 1.2 and 4.1 Figure 17 MSLB Calculated Steam Pressure Ratio Near the Steel Shell Volume 49 Figuro 18 MSLB Calculated Steam Pressure Ratio Near the Steel Shell Volume 50 Figure 19 MSLB Calculated Steam Pressure Ratio Near the Steel Shell Volume 51 Figure 20 MSLB Calculated Steam Pressure Ratio Near the Steel Shell- Volume 52 Figure 21 MSLB Calculated Steam Pressure Ratio Near the Steel Shell- Volume 79 Assessment of Effects of WGUTHIC Solver Upgrade from Version 1.1 to 4.1 October 1997 on3837w.non:1b/101097 1

O lil List of Figures (cont)

Figure 22 MSLB Ca!culated Steam Pressure Ratio Near the Steel Shell Volume 80 Figure 23 MSLB Calculated Steam Pressure Ratio Near the Steel Shell- Volume 81 Figure 24 MSLB Calculated Steam Pressure Ratio North CMT Cavity - Volume 6 Figure 25 MSLB Calculated Steam Pressure Ratio South CMT Cavity - Volume 104 Figure 26 MSLB Calculated Steam Pressure Ratio Upper East Steam Generator Compartment - Volume 4 Figure 27 MSLB Calculated Steam Pressure Ratio Lower East Steam Generator Coinpartment - Volume 107 Figure 28 MSLB Calculated Steam Pressure Ratio Upper West Steam Generator Compartment - Volume 5 F igure B.1 Test 219.1 Vessel Pressure F y m B.T. Test 219.1 Air Pressure Ratio Dome 63 in.

Figuro B.3 Test 219.1 Air Pressure Ratio Elevation A Figure B.4 Test 219.1 Air Pressure Ratio Elevation E Figure B.5 Test 219.1 Air Pressure Ratio - Elevation F Figure B.6 Test 219.1 Helium Pressure Ratio Dome 63 in.

Figure B.7 Test 219.1 Hellum Pressure Ratio Elevation A Figure B.8 Test 219.1 Helium Pressure Ratio Elevation E Figure B.9 Test 219.1 Helium Pressure Ratio Elevation F Figure B.10 Test 214.1 Vessel Pressure Figure B.11 Test 214.1 Air Pressure Ratio Dome 63 in.

Figure B.12 Test 214.1 Air Pressure Ratio Elevation A

- Figure B.13 Test 214.1 Air Pressure Ratio - Elevation E Figure B.14 Test 214.1 Air Pressure Ratio - Elevation F Figure B.15 Test 216.1 Vessel Pressure Figure B.16 Test 216.1 Air Pressure Ratio Dome 63 in.

Figure B.17 Test 216.1 Air Pressure Ratio - Elevation A Figure B.18 Test 216.1 Air Pressure Ratio - Elevation E Figure B.19 Test 216.1 Air Pressure Ratio - Elevation F Figure B.20 Test 222.4 Vessel Pressure Figure B.21 Test 222.4 Air Pressure Ratio - Dome 63 in.

Figure B.22 Test 222.4 Air Pressure Ratio Elevation A Figure B.23 Test 222.4 Air Pressure Ratio - Elevation E Figure B.24 Test 222.4 Air Pressure Ratio - Elevation F Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

0 101 1.0 Introduction This report summarizes changes made in upgrading from WGOTHIC versions 1.2 to 4.1 and supports the application of lumped parameter validation conclusions (version 1.2 used in Reference 2) in development of the Evaluation Model. The effect of code upgrades on WGOTHIC calculations of separate effects tests (Ref. 3) is addressed, as is the effect on limiting containment design basis analysis (DBA) pressure transients for the AP600.

The AP600 licensing basis containment Evaluation Model consists of version 4.1 of the suite of programs (preprocessor, solver, and postprocessor) that make up the WGOTHIC computer code, as well as the input and methodology for developing the code input as described in WCAP 14407, Rev.1, 'WGOTHIC Application to AP600" (Ref.1), Section 3 of Reference 1 describes, in detall, the historical development of WGOTHIC version 4.1, derived from version 4.0 of the EPRI sponsored containment analysis code, GOTHIC. The changes from WGOTHIC solver version 1.2 to version 4.1 may be grouped as follows:

GOTHIC 4.0 upgrades and error corrections from previous versions GOTHIC 4.0 documentation updates GOTHIC 4.0 EPRI sponsored peer review and QA program

. Upgrades to ecvel subroutine Correction of coding errors in the WGOTHIC clime subroutines.

It is the purpose of this report to show that The methodology used in WCAP 14407, "_WGOTHIC Application to AP600,* (Ref.1) to develop the bounding Evaluation Model, is unaffected by the difference in code version and therefore remains acceptable.

The conclusions from the integral model validation based on the large scale test (LST),

related to lumped parameter code biases, are not affected by the upgrade to WGOTHIC 4.1, so that the code validation in WCAP 14382, "WGOTHIC Code Description and Validation," (Ref. 2) remains acceptable.

The separate effects test analyses that form the basis for selecting the passive containment cooling system (PCS) heat and mass transfer correlations, as described in WCAP 14326, " Experimental Basis for the AP600 Containment Vessel Heat and Mass Transfer Correlations," (Ref. 3) are unaffected by the upgrade to WGOTHIC 4.1.

It will be shown that the droplet model improvements and the correction to clime logic associated with partially wet climes are the primary differences between the two code versions that affect the calculated results. Other code changes did not have a strong effect on Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4,1 O tober 1997 oA3B37w.non:1b/101097

, 12 calculated pressure for the AP600 DBA. While noticeable differences were observed for the LST integral model calculations, the differences were small, and no single change introduced a large enough effect to cause a consistent deviation ootween the results.

The code changes are assessed for calculated pressure results for the limiting PCS DBA l transients, AP600 loss of coolant accident (LOCA) double ended cold leg guillotine (DECLG) i break, and the main steam line break (MSLB) full double ended rupture at 30 percent power, in Section 2.

Guidance from integral LST WGOTHIC model validation used in developing the bounding Evaluation Model is shown to remain valid in Section 3.

The supporting role of simple channel WGOTHIC models, used in selection of the PCS heat and mass transfer correlaiions,is shown to be unaffected in Section 4.

Appendix A provides a brief history of the evolution of changes of WGOTHIC solver from version 1.2 to version 4.1.

Appendix 3 compares code version results for selected LST tests and shows that the predicted pressures for the two versions differ at most by about 7 percent, and both versions calculate pressures higher than the rneasured pressure.

Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.nort1b/101097

0 0

21 2.0 Assessment of Code Changes on L <aluation Model Results DBA methodology development calculations reported in Reference 1 are performed using WGOTHIC 4.1. Based on an NRC request for additional information (RAI 480.1023), an assessment of the effects of error corrections made to the clime subroutines ard other changes in going from version 1.2 to version 4.1 are provided in this report.

As discussed in Reference 1, Section 3, WGOTHIC 4.0 was created to introduce consistency between WGOTHIC and the version of GOTHIC 4.0 that had been peer reviewed and brought under a 10 CFR 50 Appendix B QA program. Corrections to clime subroutines were implemented creating WGOTHIC 4.1 For completeness, this section discusses the not effect of the change in WGOTHIC solver from version 1.2 to version 4.1 on containment pressure DBA calculations, The major changes are summarized in Appendix A.

The LST comparisons of solver version 1.2 to version 4.1 arc discussed in Appendix B. The tests used superheated steam, so the drop model improvement had no offect on the predicted response, in the LST, the climes did not experience dryout, so the changes to correct the clime dryout error had no effect. Therefore, based on the results presented in Appendix B, it may be concluded that the net effect of the code changes on the containment pressure response is not significant for models that:

Are not sensitive to small changes in circulation patterns Are dominated by an imposed boundary condition, (imposed boundary conditions that can dominate the solution include changing steam flow rates, or actuation of the PCS)

Impose stratification using lumped parameter biases, as is done for an MSLB, as described in Section 2.2.2 The above conditions are met for the AP600 pressure DBA calculations. The LOCA DECLG model predicts nearly uniform steam concentrations in the containment (except for break compartments) throughout much of the transient (e.g., Ref.1, Figure 9-44). Therefore, small changes in circulation flow rates are not expected to result in a significant change to the calculated steam distribution, and thus, to the calculated pressure. Both LOCA and MSLBs are transients with rapidly changing imposed steam flow rates, so the response to the boundary condition dominates the transient calculation. The MSLB Evaluation Model imposes stratification using lumped parameter biases. Since the other code changes have been ruled out as possible significant contributors to the difference in the DBA response, the primary variation in calculated pressure between the two code versions is caused by code changes associated with the improved drop model and clime error corrections. This is confirmed in the following evaluation.

Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/1o1097

a 22 The method to evaluate 'he primary causes of variation in calculated pressure between the two versions was to peri .rm separate effects sensitivities to parameters or code changes that were selected as being 'ne most likely contributors.

As discussed in Appendix B, the covel subroutine was upgraded to improve the cell centered velocity calculation associated with lumped parameter fluid nodes. The cell contered velocity is only used in calculating heat and mass transfer in the extemal annulus. The resulting cell-centered velocities calculated by the two versions were verified to be similar, es it can be concluded that the differences between the velocities predicted by the two codes are not the primary reasons for the differences between calculated pressures. The following discussion summarizes the conclusions of the study.

2.1 LOCA DECLG Figures 1 and 2, taken from Reference 1, show the containment Evaluation Model noding diagram discussed in the following assessment of the effects of WGOTHIC changes from solver version 1.2 to version 4.1 on the AP600 LOCA model.

Figure 3 compares the calculated pressure using WGOTHIC solver versions 1.2 and 4.1. Note that the graphical printout is coarse during earlier time phases so that sufficiently smooth curves can be made for the longer term transient. Note also that the plots show version 4.1 as the solid line and version 1.2 as the dashed line, throughout. The two code versions calculate maximum pressures within <0.5 psi near 1200 seconds. Therefore, there is no appreciable change in the margin to design pressure due to the code changes. Long-term depressurization, beyond 10.000 seconds, is still relatively rapid for both versions; however, corrections to the analytical treatment of a partially wet clime at lower PCS flow rates result in higher pressures calculated by version 4.1 in the long term depressurization.

To simplify the discussion, the transient is divided into three portions:

1) 0-90 seconds, blowdown / refill

2) 90-1200 seconds, peak pressure

3) 1200 seconds to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, long term depressurization The following summarizes the primary causes of variation between results of the two code versions for the containment response to the DECLG LOCA transient.

Assessment of Effects of WGOTHIC l Solver Upgrade from Version 1.2 to 4.1 - October 1997 o:\3837w.non:1b/101097 l

, 23 2.1.1 Blowdown / Refill The pressure calculated by version 4.1 at the end of blowdown (about 30 seconds) is <0.3 psi lower than that calculated by version 1.2. The difference is a result of !:;s troplet model improvements incorporated in GOTHIC 4.0 (Ref. 7).

During the first 30 seconds, blowdown of the primary system is assumed to result in roughly 50 percent of the liquid portion of the transient being delivered to the containment as drops (Ref.1, Section 4.5.2.1). Near the end of blowdown / refill (90 seconds) and boyond, the effect of the droplet model improvements diminishes as the internal heat sinks reach maximum effectiveness well before the time of peak pressure.

2.1.2 Peak Pressure During the period from 90 - 1200 seconds, pressure calculated by version 4.1 increases to reach a peak pressure only slightly above that calculated by version 1.2. The primary reason for the higher calculated pressure is changes in steam concentration along the inner shell surface and in below deck compartments that contain the majority of internal heat sinks.

Representative nodes along the inner shell surface are nodes 49,50,51,52,79,80, and 81.

The comparison of the calculated steam concentrations for these nodes are shown in Figures 4 through 10, respectively. The dominant circulating compartments are represented by nodes 6,104,4,107, and 5. The comparison of the calculated steam concentrations for these nodes are shown in Figures 11 through 15, respectively.

From 90 1200 seconds, the calculated steam concentration alcq the steel shell is slightly less in version 4.1. This reduces the condensation rate on the shell. During the same period, the calculated steam concentrations in the intemal compartments are somewhat higher in version 4.1.

Since the internal heat sinks reach maximum thermal effectiveness well before the time of peak pressure (see for example, Ref.1, Figure 9-43), the slightly higher concentrations in below-deck compartments does not significantly affect the predicted pressure.

2.1.3 Long Term Depressurization During the long-term depressurization phase from 1200 seconds to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, the primary difference between AP600 pressure results is attributable to the clime error corrections. The correction to the clime subroutines affects clime heat transfer wh.n the clime is partially wet.

When the PCS water on a clime completely evaporates, such that there is liquid film delivered to the top of the clime, but no liquio

• r xiting from the bottom of the clime, version 1.2 considered the entire clime to be w .nis over-estimated the heat removed by conduction to Assesstnent of Effects of WGOTHlO S*er Upgrade frorn Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

l .

0 24 the extomal surface of the clime conductor. Version 4.1 specifically models heat transfer from both the wet and dry portions of the clime conductor Version 4.1 results in less calculated heat transfer from a partially wet clime than version 1.2. Climo dryout begins in both versions near 800 seconds. The effect is soon in vessel pressure some time later, after the thermal wave passes through the stool shell, and the effect bocomes more significant at the lower PCS flow ratos lator in time. The effect on pressure can be clearly soon in the increase in the predicted pressure of version 4.1 during this phase.

2.1.4 Assessment of LOCA Circulation and Stratification Blases Stratification blasos introduced into the lumped paramotor Evaluation Model for LOCA DECLG are calculated independent of the code (Ref.1, Sections 9.3.1.1 and 93.1.3). Therefore, there is no impact of code versions on stratification biases.

The use of the lumped parameter Evaluation Model to examine a rango of postulated circulation scenarios for the LOCA DECLG in Reference 1 was based on comparing results using WGOTHlO 4.1 to expectations based on the international databaso(soo Section 9.2.4).

The stoam concentration distributions, shown in Figures 4 through 15, show similar magnitudes and trends for each of the compartments. This shows that the offects of circulation pattom characteristics are not sensitive to the difforent code versions.

2.2 MSLB Figures 1 and 2, taken from Reference 1, show the containment Evaluat:on Model noding diagram discussed in the following assessment of the offects of WGOTHIC changes from solver version 1.2 to version 4.1 on the AP600 MSLB model.

Figure 16 shows the pressure transient for the limiting MSLB transient, a full double ended rupture of the main steamline initiated from 30 percent power. Note that the plots show version 4.1 as the solid line and version 1.2 as the dashed lino, throughout. WGOTHIC solver version 4.1 calculates a slightly lower pressure during the transient, with a peak pressure itterence of 0.5 psi.

Due to the low revaporization rate assumption in the containment Evaluation Model, there is a high degree of superheat in the DBA calculation, ranging between 40'F and 120*F throughout the transient. Therefore, the steam concentration and vessel pressure are not coupled via the saturation curve. This becomes apparent when comparing the steam concentrations throughout the containment versus the pressuro predictions for the code versions.

Assessment of Effects 0! WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 oA3837w.non:1tV1oto97 I

L

~

, 25;

- 2.2.1 - Evaluation of Effect on MSLB Calculated Pressure The variation in results is attributable to two effects:

The_ effect of the clime dryout modeling correction in version 4.1 The not e,ffect of code upgrades causing small differences in steam concentration distribution PCS flow is applied to the outside of the containment shell in the Evaluation Model at 337-seconds, which conservatively neglects the presence of water on the shell prior to achieving

{

steady-state coverage. At 350 seconds, the clime dryout models begin to be used in v5rsion . H 4.1. Examination of the results showed that less heat is removed from the outside of the shell 1 in version 4.1, consistent with the clime subroutine upgrades. Howaver, because of the -

relatively long time constant for the steel shell heat transfer, the effect of the extemal 4 boundary condition is relatively weak by the time of maximum pressure. However, it was also

- observed that there was reduced heat transferred into the steel shell from the inner surface in-version 4.1.

- The total heat rate into intetaal containment heat sinks was found to be higher in version 4.1,

= consistent with the lower calculated vessel pressure. These heat rate differences are consistent with the differences in predicted steam concentrations in _ volumes along the steel 1 shell (nodes 49-52 and 79-81) and in compartments below-deck (representative compartnient '

- nodes are 6,104,4,107,: and 5), shown in Figures 17 through'28 respectively.

In Figures 17 through 23, version 4.1 is seen to_ calculate a slightly higher steam concentration near the steel shell.- However, the reduction in heat removal from the outside of the steel .

shell due to the clime subrout'ne dryout correctior, leads to a decrease in condensation mass transfer in nodes near the shell. Thus, differences in the predicted steam concentration in 1 nodes near the shell are not the primary cause of the difference in pred'eted pressure -

l between he two code versions. The reduced PCS heat removal would tend to increase the calculated pressure.

- Figures 24 through 28 show ateam concentrations in the North and Soutt. CMT cavities, upper - 1 and lower East stean: generator compartment, and upper West steam generator compartment, -

as representative below-deck compartments. The steam concentration in each of the i below-deck compartmants is seen to be slightly higher in version 4.1 through most of the -

, transient.~ Because of the relatively short duration of the MSLB, mass transfer to below-deck

~lntemal heat sinks is a dominant pressure reduction mechanism.. Therefore, the net effect of WGOTHIC solver changes from 1.2 to 4.1 is a slight (0.5 psi) reduction in calculated peak

- pressure due to a small increase in mass transfer to the below-deck heat sinks.

Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 ~

October 1997 013s37w.non:1b/101097

, 2-6 2.2.2 Evaluation of MSLB Circulation and Stratification Blas i The evaluation in Reference 1, Section 9.4.2, describes the bias introduced in the MSLB Evaluation Model to bound the effects of circulation and stratification. The appropriateness of the bias is shown by examining the resulting steam concentration distribution for the case of

. the break node assumed at the operating deck (Ref.1, Figure 9-60). Results showed that vary little steam penetrates into the below-deck region in the model. SteLm access into the below-deck compartments in the model is govemed only by the volume pressurization. The reduced steam access is due to the momentum dissipation in the model, which eliminates momentum as a driving force for flow below the assumed break elevation. The initial underste.r' ding of the bias used for the MSLB was developed during code validation using version 1.2.

The steam concentration plots in Figuies 17 through 28 show that both WGOTHIC solver versions 1.2 and 4.1 exhibit the lumped parameter momentum bias such that the effects of kinetic energy driven circulation during the MSLB are biased toward a conservative pressure calculation. Therefore, the bases for MSLB circulation bias are not affected by the change in code version.

A Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097 l

o 2-7 99 100 91 93 M

wY 79 67

% 89 61 n g 14 o 33 4 +

95 52 36 28 20 12 16 44 60 88 95 m !3-to 36 94 51 35 27 19 11 15 27 43 59 87 94 3 to 35 l l

'O y 93 50 26 18 to 34 26 42 53 86 93 f = 8-to 34 92 5Y 49 33 25 17 9 13 41 57 35 92

2-to 33 g , __

5 4

/ s 7 , _

____ ____ ioi 108 107 -

1-to 49 I Figure 1 Noding Diagram for West Outer Quarter Annulus Cross Section Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097 t

2-8 (a,c) 4 l

Figure 2 NGOTHIC Calculated Flow Pattern - LOCA Jet Momentum Dissipated in -

SG East Compartment at 8000 Seconds Assessment of Effects of EGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997

- o:\3837w.non:1b/101097

29 Vessel Pressure (LOCA) ver 4.1 and 1.2 PR81 0018 g m- -m

~

ay

.,f

.g 18 N

a s

N- -

. s,

~

o , , , , ,,,,i ,,,,,,,,1 . , , ,,,,,1 , ,,,,,i, 10 100 1000 1e+04 1e+05 Time (sec) wooTwo sehevwsion 4.1

. . . . . . . . . . . . wooTac sehe m 1.2 Figura 3 LOCA DECLG Calculated Pressure Comparison of WGOTHIC Solver Version 1.2 and 4.1 AcaT,ent of Effects of WGOTHIC Sotver Upgrade from Version 1.2 to -.1 October 1997 o:( % 3 6 . M n:1b!101097

2 10 g - j v_ -

s ct l ~

g st~,

of o -

o , , ,,...I- , . , , ,,,,1 , , , , ,,,,1 , , , , , , , ,

10 100 -1000 1e+04 te+05 Time (sec) woorsic sav., v. mon 4.1 WGOTHIC Solver worsion 1.2

?

Figure 4 LOCA DECLB Calculated Steam Pressure Ratio Near the Steel Shell -

Volume 49 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

2-11 l

Steam Pressure RatioinVolume 50

' 9R50 DC24

-- - m- --a

~

o 3 -

e.

m -

o - a 5 - /

E /  %

& w 7  %- s 6 -- '

e , , , . ....I , , , , , , , ,1 .,,,,,,,6 , , , , , , , ,

10 100 1000 1e+04 1e+05

'~

Time (sec) .

1 WOOTHIC Solver version 4.1

. . . . . . . . . . . . - WGOTHIC Solver version 1.2 Figure 5 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell -

Volume 50 Assessrnent of Effects of WGOTHIC Solver Vograde from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

2 12 Steam Pressure Ratioin Volume 51 9R51 DC25

--  : n- -e

,g .

& m -

e o _

~

w _

/

/

b w

& , - / 'se~s o :/

-/

1 W

o o , , ,,,,,,1 , , ,,,,,i , ,,,,,,,i , , , , , , , ,

. 10 100 1000 1e+04 1e+05 Time (sec) 1 WOOTHIC Solver Asion 4.1

. . . . . . . . . . . . WGOTHIC Solver version 1.2 Ficare 6 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell -

Volume 51 Assessment of Effects of EGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 c:\3837w.non:1b/101097

2 13-Steam Pressure Ratioin Volume 52 9R52 DC28

. E- -a

.g -

a w -

o -

t , \ v ~

/ -

ci

-/

1_

m.

o _

o , ,,,,,,,1 , , ,,,,,,I , ,,,,,,,I , , , , , , , ,

10 100 1000 1e+04 1e+05 Time (sec)

WOOTHIC solver version 4.1

............ wcome soever version 1.2 Figure 7 LOCA DECLG Calculated Steain Pressure Ratio Near the Steel Shell -

Volume 52 Assessment of Effects of WGOTHIC Solver Upgrade from Version 12 to 4.1 October 1997 o:\3837w.non:1b/101097 i _

2 14 Steam Pressure Ratioin Volume 79 GR79 DC29

.- 4--m-3 -

a c o -

/

/

a

$ ' /

,, N /~s O

E -

e -.. -

i . ....:- , , , , , , ,~t , ,,,,,,,i , ,,,,,,,l 10 100 1000 1e+04 1e+05 Tl 1(sec)

T WOOTHIC Solver version 4.1

. . . . . . . . . . . . WGOTHic Solver version 1.2 Figure 8 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell -

Volume 79 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

2-15 l

l Steam Pressure Ratioin Volume 80 9R80 DC30 -

m- -a a -

,g .

E c c-

~

/

l d- ,

~

I I-m

%~J.

d-/

-I

~l et c>

)

.( c,i , , , , ,,,,i , , , , , , , , .i ,,,,,,,,i , , . , , , , , i 10- 100 1000 1e+04 1e+05 Time (sec)

WOOTHIC Solver version 4.1

. . . . . . . . . . . . WGOTHIC Solver version 1.2 Figure 9 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell -

Volume 80 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 oM837w.non 1."101097

e 2 18 l'

Steam Pressure Ratioin Volume 81 9R81 DC26

.- 4--e 51 _

8 ~

2e m i o '

@ _ /

I N g

g ._ j

, Ne~s

_I

~

N o .

a' ,,,,,,,,I , , , , ,,,,1 , ,,,,,,,1 , , , , , , , ,

i

} 10 100 1000 1e+04 1e+0S Time (sec) ,

l WOOTHIC Solver version 4.1

............ WGOTHIC Solver version 1.2 Figure 10 LOCA DECLG Calculated Steam Pressure Ratio Near the Steel Shell -

Volume 81 Assessment of Effects of EGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 -

o;\3837w.non:1b/101097

__________j

2o17 l

l Steam Pressure Ratioin Volume 6 9R6 DC22

-- :15- -s P N

. g

\

m - g ci

\

\

2 -

s (R

s '-

s g -

s g -

N 3 -

N,ew ,

E ~-

~

o , ,,,,,,,i , , , , ,,,,i , ,,,,,,,t , , , , , , , ,

10 100 1000 1e+04 1e+05 Time (sec)

WGOTH6C Solver version 4.1

. . . . . . . . . . . . WOOTHIC Solver version 1.2 Figure 11 LOCA DECLG Calculated Steam Pressure Ratio North CMT Cavity - Volume 6 Assessment af Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:0837w.non:1b/101097

2-18 I

L Steam Pressure RatioinVolume 104 9R104 DC23

.- = -e

.9 5

's 2e m 's '

g .

N E

A  : N o _

N f~s

, v E -

e , ,,,,,,i , , , , ,,,,i , , , ,,,,,i , , , , , , , ,

10 100 1000 1e+04 1e+05 Time (sec)

WoOTHIC Solver version 4.1 '

wooTmc soever v.e. ion 1.2 Figure 12 LOCA DECLG Calculated Steam Pressure Ratio South CMT Cavity -

Volume 104 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:W137w.non:1b/101097

I 2-19 Steam Pressure Ratioin Volume 4 9R4 DC20

-- m- -e g

\

= 1.

o

• w \

e  : N / ~s,

s M  :

e

n. -

6 :

y "_

~

o , , , , ,,,,1 , , ,,,,,,I , , , ,,,,,1 , , , , , , , ,

10 100 1000 1e+04 1e+05 Time (sec)

WOOTHIC solver version 4.1

............ WOOTHIC solver version 1.2 Figure 13 LOCA DECLG Calculated Steam Pressure Ratio Upper East Steam Generator Compartment - Volume 4 Assessment of Effects of WGOTHIC Solver Upgrade from Version 12 to 4.1 October 1997 on3837w.non:1b/101097

2 20 Steam Pressure Ratioin Volume 107 9R107 D019

-- __- = -s

N m

d -

3 -

~

& =

m -

B

• 8  :

e a- .

w 6 :

m

\

6 -

\

o .~ . ,,,,,,,1- , , ,,,,,,i ,,,,,,,,i , , , , , , , ,

10 100 1000 1e+04 1e+05 Time (sec) t WGOTHIC Solver version 4.1

............ woonec soever version 1.2 i

Figure 14 LOCA DECLG Calculated Steam Pressure Ratio Lower East Steam Generator Compartment - Volume 107 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

2 21 l

Steam Pressure Ratioin Volume 5 9R5 DC21

_ - E---E

\

\

ci

\

\

.g -

\

m .

N g j 6 :  % / ~s R -

E -

t ,

6 :

c, , , , , ,,,,1 , , , , ,,,,1 .,,,,,,,1 . , , , , , , ,

10 100 1000 1e+04 1e+05 Time (sec)

WGOTMC SoNet version 4.1

............ wooTuc sower version u Figure 15 LOCA DEDLG Calculated Steam Pressure Ratio Upper West Steam Generator Compartment - Volume 5 Assessment of Effects of WGOTHIC Solver Upgrade trom Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097 I

2 22 Vol. 81 Pressure Comparison . Ver 4.1 vs. Ver 1.2 PR81 DC26 g  : n- - e

- "~~

e -

~

,/

/

$ b -

E -

g a -

3 -

o ~ i,,,Iii i..,,IiiiiIi,, Iii<  ! 't i i 0 100 200 300 400 500 600 700 Time (seconds) x WOOTHIC Solver version 4.1

. . . . . . . . . . . . WOOTHIC Solver version 1.2 Figure 16 MSLB Calculated Pressure Comparison of WGOTHic Solver Version 1.2 and 4.1 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 oM837w.non:1b/101097

2 23 l -

l 4

Steam Pressure Ratioin Volume 49 9R49 D019

.  : n- -- e 5

o .

8 ~

~

E 9 O

j .

~

T U .

J C4 ,'

o o ~,,iiii.,iliiiiliiie I,,,ii ....Ii,,,

0 100 200 300 400 500 600 700 Time (sec)

WGOTHIC Solver version 4.1

. . . . . . . . . . . . WOOTHIC Solver version 1.2 Figure 17 MSLB Calculated Steam Pressure Ratio Near the Steel Shell- Volume 49 Assessrnent of Effects of WGOTHlO Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

_ __ u

l 2 24 l-l

[

Steam Pressure Ratioin Volume 50 9R50 DC20

_ :D-e S '- -

.g -

rf u! _

& w 6;-

1 N

o 4

o',,,,i,,,,i,,,,i,,,,i,,, i,,,,,i,,,,

0 100 200 300 400 500 600 700-Time (sec)

WGOTHic sow version 4.1

. . . . . . . . . . . . woonec sow version 1.2 Figure 18 MSLB Calculated Steam Pressure Ratio Near the Steel Shell- Volume 50 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:0837w.non:1b/101097 l

-2 25 Steam Pressure Ratioin Volume 51 9H51 DC21

. s- - e 8 *

~

c2 m _

m o .

~

v "~

C3 ,

N o -- -

f oI,,,,I,,,,I,,,,Ii,,,!,,,,I,,,,i,,,,

0 100 200 300 400 500 600 700 Time (sec)

WOOTHIC Solver version 4.1

. . . . . . . . . . . . WOOTieC Sh version 12 i

Figure 19 MSLB Calculated Steam Pressure Ratio Near the Steel Shell- Volume 51 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097 I

___ _ _ _ _ _ _ _ _ l

o ,

-2 26-Steam Pressure Ratioin Volume 52 9R52 DC22

.- - m- - e d -

(D d ~

w g ..

D o .

I

,l'*

o .. ili,, I ,ilii li,,,!,,,,l ,,,,

0 100- 200 300 400 500 600 700 Time (sec)

WOODUC Solver version 4.1 woome soiver version 1.2 Figure 20 MSLB Calculated Steam Pressure Ratio Near the Steel Shell - Volume 52 Assessrnent of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.nort1b/101097

2 27 l

l l-i

- Steam Pressure Ratioin Volume 79 9R79 DC23

_ - m- - e

~~

't

.j c2 e o -

E i v

~

"~

c3 ,

)

N o ._

1-P o , , , ,I,,,,I,,,,I,,,,t,,,,i,,.,,!,,,,

0 100 200 300 400 500 600 700 Time (sec)

. WGOTHIC Solver version 4.1 wGoTwee sever ver. ion 1.2 Figure 21 MSLB Calculated Steam Pressure Ratio Near the Steel Shell - Volume 79 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097 I

. . , . . . . . . . . . _ __9

2 28 Steam Pressure Ratioin Volume 80 9R80 DC24

_ =>-e co

.g -

et *

= _

g, .

~

w g ..

h .

o,, ,,!,,,,I ,,,Ii,ie Ii .iI.....Ii. .

0 100 200 300 400 500 600 700 Tune (sec)

WOOTHIC Solver version 4.1

. . . . . . . . . . . . WOOTHIC Scever version 1.2 Figure 22 MSLB Calculated Steam Pressure Ratio ' lear the Steel Shell - Volume 80 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

2 29 1

l Steam Pressure RatioinVolume 81 9R81 DC25

_-. - s- -- u -

M

.g -

~

d2 =

o _

W g -.

N a ..

o , , i i l i . . , 1 i a i .

0 233.3 466.7 700 Time (sec)

WGOTMC Scher version 4.1

. . . . . . . . . . . . WOOTMC Scher version 12 Figure 23 MSLB Calculated Steam Pressure Ratio Near the Steel Shell - Volume 81 Assessment of Effects of EGOTH!C 3 Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

2o30' Steam Pressure Ratioin Volume 6 9R6 DC17

. - s- - e d -

en 6 ~

w U ?

. - ~

a -

4 o',,,.I.,ii ..

I. . ..I,,iiii ,I ... l.,,.

0 100 200 300 400 500 600 700 Time (sec) wooTHIC Solver versioc 4.1

. . . . . . . . . . . . WGoTHIC Solver versiors 1.2 Figure 24 MSLB Calculated Steam Pressure Ratio No th CMT Cavity - Volume 6 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:k3837w.non:1b/101097

2 31 l

Steam Pressure Ratioin Volume 104 9R104 DC18

. :F-e

.g -

2

~

o ._

~

~

l

~

U" w

_ _ _ _ - ~ ~ AA e i .ili iili.,iIi .ili ile i*i!iiii '

0 100 200 300 400 500 600 700 Time (sec)

WOOTHIC Solver version 4.1

. . . . . . . . . . . . WOOTHIC Solver version 1.2 Figure 25 MSLB Calculated Steam Pressure Ratio South CMT Cavity - Volume 104 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:G837w.non:1b/101097

_a

c 2 32-1 I

S1eam Pressure Ratioin Volume 4 9R4 0015

. - m- - e 43 i

5. w 6

.f ___._ - -- ----- ~

u -

o i i i l . i f i . i l i i i i f i i i i i i e 'i iliiii 0 100 200 300 400 500 600 700 Time (sec)

WOOTHIC Solver version 4.1

. . . . . . . . . . . . WooTHIC Solver version 1J Figure 26 MSLB Calculated Steam Pressure Ratio Upper East Steam Generator Compartment - Volume 4 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 cA3837w.non:1b/101097

. 2 33 Steam Pressure Ratioin Volume 107 9R107 DC14

-p_g.

~

9 o "

~~

A -

1 oIiiiily ii1. e ilie iiliiie iiiSi lie i i 0 100 200 300 400 500 600 700 Time (sec)

WOOTHH: Solver version 4.1

. . . . . . . . . . . . WOOTHIC Solver version 1.2 Figure 27 MSLB Calculated Steam Pressure Ratio . Lower East Steam Generator Compartment - Volume 107 Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

,o -

2 34 Steam Pressure Ratioin Volume 5 9RS DC16

.  : s- - e E -

~

9 _

ci ei o -

M o ,,,,i,,,,i,,,,i,,,,i,,,,t,.,,i,,,,

0 100 200 300 400 500 600 700 Time (sec) 4

_ WGOTHIC Solyw version 4.1

. . . . . . . . . . . . WOOTHIC Solver version 1.2 Figure 28 MSLB Calculated Steam Preswure Ratio Upper West Steam Generator Compartment - Volume 5 Asessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 -

October 1997 0:\3837w.non:1tw101097

o 31 3.0 Applicability of LST Code Validation Conclusions (solver 1.2) to Development of Evaluation Model(solver 4.1)

The validation of WGOTHIC with PCS large scale test (LST) data comparisons (Ref. 2) was performed using WGOTHIC solver version 1.2. Biases inherent in the use of multiple lumped i parameter nodes to represent containment pressure performance have been documented based on industry experience (Ref.1, Appendix 9.C.3). The purpose of the WGOTHIC model comparison to LST data was to understand how the documented biases apply to AP600 and to develop guidance for bounding the effects of those biases. A scaling evaluation (Ref 4, Section 11.2) shows that the LST matrix is a reasonable database from which to develop an understanding of those biases for the AP600.

The purpose of this section is to assess the applicability of this validation program relative to the containment Evaluation Model which uses WGOTHIC version 4.1. It should be noted that data from the LST were also used for separate effects verification of particular models, as discussed in Section 4.0. This section addresses code validation of the WGOTHIC model with the integral LST data.

The results of the WGOTHIC LST integral model comparisons to the LST data have been used to understand the biases inherent in the WGOTHIC lumped parameter formulation, as they apply to AP600 DBA containment pressure analysis, and M develop guidance in developing a bounding Evaluation Model. The LST validation model used nominalinputs for geometry, initi% and boundary conditions, so that the effects of the code and noding used could be better isolated. The same WGOTHIC input decks that were used in the code validation program with solver version 1.2 were used as input to WGOTHIC version 4.1. The inertial length input was modified to account for the change in the ccvel subroutine in WGOTHIC version 4.1 (see Appendix A). This input change results in a similar annulus velocity for the two versicns. This was done to isolate the effects of other code version changes.

The following conclusions derived from WGOTHIC code validation using the integral LST data and biases to yield a bounding model have been usod in developing the containment Evaluation Model:

a) The lumped parc. meter calculations with the LST model in the LOCA configuration incurred compensating errors in steam concentration and velocity (Ref. 2, Section 8.2.1).

The compensating errors are addressed in the lumped parameter LST model by eliminating the effects of predicted velocities. The forced convection component of intemal PCS heat and mass transfer is shut off via input. Results from LST Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 c:\3837w.non:1b/101097 l

32 calculations using both solver versions 1.2 and 4.1 in Appendix B show calculated pressures higher than measured. The Evaluation Model also conservatively neglects ferced convection internal PCS heat and mass transfer effects (Ref. 5, Sections 4.4.3F, 4.4.7A, and 4.4.7C). Neglecting forced convection is conservative and independent of code version.

b) The lumped parameter calculations with the LST model in the MSLB configuration unrealistically stratifies the volume below the break elevation (Ref. 2, Section 8.2.1).

This characteristic, of models which use multiple lumped paranieter nodes to represent an open volume, is related to the momentum formulation used. The Evaluation Model, with the injection point in the node just above the operating deck, results in a conservative MSLB as a result of reduced steam access to the below deck heat sinks.

Such an approach makes use of the lumped parameter bias to penorm the calculation conservatively. As discussed in Reference 2, Section 9.4.2, results from the WGOTHIC solver version 4.1 Evaluation Model show this bias is correctly implemented.

Since the lumped parameter bias for a specific model is related to the noding chosen, the noding used in the AP600 containment pressure DBA Evaluation Model corresponds to the noding used in the LST s alidation work (Ref. 2, Section 6.4.2). Therefore, the biases and guidance developed based on LST models are applicable to the AP600 model.

Based on the assessment in Appendix B, the upgrade of WGOTHIC solver from version 1.2 to version 4.1 does not invalidate conclusions from LST integral model validation. Therefore, the development of a bounding approach to address lumped parameter code blasu in the Evaluation Model remains acceptable.

J l

Assessment of Effects of WGOTHIC Solver Vograde from Version 1.2 to 4.1 October 1997 0:\3837w.non:1b/101097

41 4.0 Impact on WOOTHic Separate Effects Tests Calculations The separate effects tests used to validate the selection of heat and mass transfer correlations have been evaluated using two approaches: 1) conversion of measured data to non dimensional parameters for direct comparison to correlation results, and 2) data evaluation using simple WGOTHIC once through channel models.

For most of the tests, where the variation along the test channel (in the direction of flow) is srnall compared to the parameter of interest, and sufficient data regarding measured boundary conditions are available, a linear variation along channel length may be assumed. The measured data may be converted to nondimensional parameters with hand calculations and compred directly to correlation results. Test series in this category are Eckert and Diaguila (Ref. 3, Section 3.2), Westinghouse dry flat plate (Ref. 3, Section 3.4), Westinghouse LST dry external heat transfer (Ref. 3, Section 3.5), Gilliland and Sherwood (Ref. 3, Section 3.6,10-cell FORTRAN program), Westinghouse flat plate evaporation (Ref. 3, Section 3.7), University of Wisconsin condensation (Ref. 3, Section 3.8), and Westinghouse LST mternal condensation data (Ref 3, Section 3.9). There is no impact of code changes on these separate effects test evaluations used in selecting heat transfer correlations.

For two of the separate effects test sdrias in the literature, boundary conditions (such as test section alt flow) were not provided, or there was significant variation along the length of the channel. For these tests, the WGOTHIC code was used to model a one dimensional series of nodes along the section length. The simple WGOTHIC models served the additional purpose of verifying the correct implementation of the selected heat transfer correlations. The two test series for which data evaluation was based on a simple WGOTHlc channel model are Hugot (Ref. 3, Section 3.1) and Siegel and Norris (Ref. 3, Section 3.3).

As further verification of the correct implementation of cu elations in the.WGOTHIC code, simple onca through channel models were documented as part of the Westinghouse OA program for the following test series: Westinghouse ., Mate, Westinghouse flat plate evaporation, and University of Wisconsin condensation.

As stated in Appendix A, code upgrades did not change the PCS heat and mass transfer correlations. None of the separate effects tests used as a basis for WGOTHIC models of PCS heat and mass transfer are signihcantly affected by the code upgrades. Therefore, the bases for selection of PCS heat and mass transfer correlations is unaffected by the upgrade to WGOTHIC 4.1.

Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

51 5.0 Conclusions Results of the AP600 DBA containment pressure analysis using WGOTHIC version 4.1 and version 1.2 are similar, and the small differences in pressure response in the DECLG LOCA and MSLB have been evaluated and understood.

The methodology used in Reference 1 to develop the bounding Evaluation Model which uses WGOTHIC version 4.1 is unaffected by the difference in code version and therefore remains acceptable. The drop modelimprovement and clime subroutine correction are the dominant effects on the Evaluation Model results (Section 2).

Since the conclusions from the integral model validation, related to lumped parameter code

- - biases, are not significantly affected by the upgrade to WGOTHIC 4.1, the code validation.

work reported in Reference 2 remains acceptable (Section 3).

The separate effects test analyses which form the basis for selecting the PCS heat and mass transfer correlations are unaffected by the upgrade to WGOTHIC 4.1 (Section 4).

l l

Assessment of Effects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 October 1997 o:\3837w.non:1b/101097

+

61 6.0 References

1. WCAP 14407, Rev.1, "WGOTHIC Application to AP600," July 1997.

2. WCAP 14382, "WGOTHIC Code Description and Validation," May 1995.

3. WCAP 14326, Rev.1 " Experimental Basis for the AP600 Containment Vessel Heat and Mass Transfer Correlations," May 1997.

4. WCAP 14845, Rev. 2, " Scaling Analysis for AP600 Containment Pressure During Design Basis Accidents," June 1997.

5. WCAP 14812, Rev.1. " Accident Specification and Phenomena Evaluation for AP600 Passive Containment Cooling System," June 1997.

6. AP600 Standard Safety Analysis Report

7. NTD NRC 95 4595, "AP600 WGOTHIC Comparison to GOTHIC," 8.A. McIntyre (W) to T.R. Quay, November 13,1995.

Assessment of Etiects of WGOTHIC Solver Upgrade from Version 1.2 to 4.1 Oc,ober 1997 o:\3837w.non:1b/101097 l

A1 Appendix A Summary of WGOTHic Solver Changes from Version 1.2 to 4.1 l

The AP600 licensing basis containment Evaluation Model consists of version 4.1 of the suite of programs (preprocessor, solver, and postprocessor) tha! make up the WGOTHIC computer code, as well as the input and methodology for developing input as described in the WGOTHIC application report (Ror.1). Section 3 of the WGOTHIC Application Report describes the historical develoriment of WGOTHIC version 4.1, derived from version 4.0 of the EPRI sponsored t ontainment analysis code, GOTHIC.

An assessment of the impact of GOTHIC code upgrades through GOTHIC 4.0, relative to WOOTHIC solver version 1.2, is presented in Reference 7. The following provides a brief summary of the recent development of WGOTHIC version 4.1 from WGOTlilC version 1.2 and summerizes the changes for convenient reference in this report.

A.1 Summary of Code Changes The changes from WGOTHIC solver version 1.2 to version 4.1 did not change the PCS heat and mass transfer correlations. The changes can be grouped as follows:

GOTHIC 4.0 upgrades and error corrections from previous versions GOTHIC 4.0 docurnentation updates GOTHIC 4.0 EPRI sponsored peer review and QA program

. Upgrades to ccvel subroutine Correction of errors in the WGOTHIC clime subroutines.

The observed variation in results from WGOTHIC version 1.2 to version 4.1 can not be attributed to a singla dominant change, The primary changes which can collectively affect calculation results of the magnitude evaluated in the body of this report and in Appendix B are the following:

The minimum value of the Uchida condensation heat transfer coeffic;ent was changed from 15 to 2, thereby reducing the condensation rate at low steam partial pressures.

The existing method for calculating steam saturation pressure as a function of temperature was replaced by a more accurate method.

1 Appendix A Summary of WGOTHIC Sohrer Upgrades from Versions 1.2 to 4.1 October 1997 o:\3837.wpt:1b/101097

!- Ad Tho wall dryout criterion used for intomal heat sinks was modified to allow water to

( remain on walls until the wall temperaturo is greater than Tsat (Total Pressure) rather than Tsat (Stoam Pressure).

The droplet drag and deposition models woro improved. This change affects the drop onergy exchange ratos.

. The climo dryout error was corrected.

Variablos in the underlying GOTHIC subroutines and in the WGOTHIC clime subroutines woro converted to double precision (GOTHIC versions prior to 4.0 had only used doublo precision whero judged to be most significant, such as in the matrix operations).

Several miscellaneous modifications woro made to improvo consistency and reliability.

Thus the changes made to create WGOTHIC solver version 4.1 and its documentation represent an evolution toward improvod solut!on, models, and rollability. The primary changos in going from WGOTHIC 1.2 to 4.0 are described in Section A.2. The additional changes in going from WGOTHIC 4.0 to 4.1 are described in Sections A.3 and A.4.

A.2 WGOTHIC Solver 4.0 Consistency with GOTHIC version 4.0 Some of the solution diffemncos in WGOTHIC solver from versions 1.2 to 4.0 are due to modifications of the underlying GOTHIC version 4.0 code made by the developer, Numerical Applications, Inc. (NAI). These code changes were made by NAl to correct deficiencies, including those noted in user reports, or to provide more realistic models. NAl tested each of the changes individually and confirmed correct implementation, and the entire change set was tested as a whole in the GOTHIC version 4.0 code release testing.

Many of the changes impi ented by NAl in GOTHIC version 4.0 had been incorporated into WGOTHIC version 1.2 by Westinghouse based on continuous dialogue with NAl. As a result of this close interaction, a number of code errors and upgrades identified early in the Westinghouse development program were corrected prior to configuring WOOTHIC version 1.2. The bulk of the upgrados resulting from that interaction were implemented by NAl in GOTHIC version 4.0. This process resulted in a limited number of differences between WGOTHlO solver version 1.2 and GOTHIC solver version 4.0.

A listing of differences between solvers in GOTHIC version 4.0 and WGOTHIC version 1.2 was provided in Referenco 7, Table 1, with an assessment of the reason for the chango and the effect of that difference on the WGOTHIC version 1.2 results. A calculation was performed to show the effects of diMerences between GOTF.1C version 4.0 and WGOTHIC Appendix A Summary of WGOTHIC Solver Upgrades from Versions 1.2 to 4.1 October 1997 o:\3837.wpf:1b/101097

A3 version 1.2 creating an intermediate configuration control version of WGOTHIC, version 1.2.2.

The assessment showed that the difference in calculated pressure resulted from the improved droplet model.

GOTHIC underwent an EPRI sponsored peer review. Reference 7 shows that the GOTHIC peer review is applicable to WGOTHIC 1.2.

WGOTHIC solver was upgraded from version 1.2 to version 4.0 to incorporate residual changes that had been made in the upgrade to GOTHIC 4.0, WGOTHIC is therefore consistent with the GOTHIC version 4.0 that was baselined under the NAl QA program.

A.3 WOOTHIC 4.1 Upgrade to ccvel Subroutine The covel subroutine was upgraded to improve the cell centered velocity calculation associated with lumped parameter fluid nodes. The cell centered velocity is only used in calculating heat and mass transfer in the external annulus. Tho resulting cell centered velocities calculated by the two versions were verified to be nimilar, so it can be concluded tnat the differences between the velocities predicted by the two codes are not the primary reasons for the differences between calculated pressures.

A.4 ROOTHIC 4.1 Correction to Partially Wet Clime Logic h performing verification activities with WGOTHIC solver version 4.0, it was discovered that the climes subroutines handled heat and mass transfer from a partially wet clime incorrectly.

4 The error in clime logic caused the PCS heat removal to be over predicted at the point of dryout. Changes were made to the Westinghouse clime subroutines to correct that error. The subroutines were modified to check for dryout; that is, the evaporating mass flux timet 15e clime area exceeds the liquid film mass flow delivered to the top of the clime, if dryout occurs, the code divides the liquid mass flow rato by the mass flux obtained from PCS correlations, and calculates the area which is required to completely evaporate the film. The remaining area within that clime is then assumed to have only dry heat transfer. Proparties, temperatures, and heat tmosfer for the dry portion of the clime are calculated assuming instantaneous quasi equilibrium, dry conditions. This approach conservatively neglects the heat capacity of the steel shell as its temperature increases from the cooler wet state to the hotter dry state.

Other minor code changes were made to correct some non-calculational problems, and WGOTHIC version 4.1 was created.

Appendix A - Summary of WGOTHIC Solver Upgrades from Versions 1.2 to 4.1 October 1997 oA3837.wpf;1b/101097

l A4 A.5 Conciualons The code modifications that have been implemented in WGOTHIC version 4.1 provide a more re!!able solution. Additional advantages of the change from version 1,2 to 4.1 are consistency with the EPRI sponsored peer review which resulted in GOTHIC 4.0 and its improved documentation, as well as the baselining of GOTHIC 4.0 under the NAl OA plan, 1

/

' Ispiidi: A + Summary of WGOTHIC Sotver Upgrades from Versions 1.2 to 4.1 October 1997

- o:\3837.wpt:1ti101097 -

.x . ,e

B1 Appendix B Comparison of LST Resuhs Using Solver Versions 1.2 and 4.1 B.1 Background The following considerations apply generally to the LST calculations presented:

The location of LST instrumentation is described in Reference 2. Figure 819.

Results from LST calculations show the calculated pressures using both solver versions 1.2 and 4.1 are higher than the measured pressures, when water is applied to the vessel shell as is done in the AP600 containment Evaluation Model.

LST calculations can be sensitive to code changes since the tests lack a flow path into the simulated steam generator compartment, leading to non prototypically high steam and air gradients between the above and below-deck regions. Because of the sensitivity of condensation rate to noncondensible content, small increases in calculated circulation fram the air rich vessel heel affects calculated pressure by as much as 7 percent in the LST. The circulation effect is shown to have much less influence on the AP600 containment Evaluation Model results in Section 4.

The ccvel subroutine was upgraded to improve the calculation of the cell centered velocity associated with lumped parameter fluid nodes. The cell centered velocity is only used in calculating the PCS heat and mass transfer in the extemal annulus. The input decks for the evaluations of the effects of code versions discussed in this report differed only in the inertial length input for the external annulus nodes. The resulting cell centered velocities calculated by the two versions were verified to be similar, so it can be concluded that the ccvel changes had no significant impact on the comparison of the code versions.

The LST noncondensible data is shown in terms of air pressure ratio (instead of steam pressure ratio as was done with the AP600), since the air pressure ratio was measured for tests 216.1, 219.1, and 222. 4.

The LST tests are superheated so the drop modelin;provement has no effect.

The climes do not expedence dryout, so the differences noted for LST results am not impacted by the clime dryout error correc'!on.

Appendix B Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997

_ _0:\3837.wpf:1b/101097 -

B2 The following sections summarize the differences in results of the WGOTHIC solver versions 1.2 and 4.1 for representative LST tests in both the LOCA configuration and in the MSLB configuration. The tests cover a range of boundary conditions. Three tests in the LOCA l configuration are examined. Test 219.1 has both a dry and wetted shell and has helium injection. Test 214.1 has natural and forced convection in the annulus. Water coverage for test 216.1 varies from 75 to 25 percent (in quadrants). Test 222.4, in the MSLB configuration, has an initial steam blowdown with an elevated 3 inch pipe.

B.2 Assessment of Variation in LST 219.1 Calculations Test 219.1 was performed at a constant steam flow and forced air cooling. The vessel pressure came to steady state under constant steam flow and a dry vessel (219.1 A). Helium was then injected and the vessel pressure allowed to come to steady state again (219.18).

The water was then turned on and the vessel pressure steadied to a tnird level (219.1C). The steady state water coverage for 219.1C was ( )b.c, The pressure transient for both versions are compared in Figure B.1. While the vessel is dry, the measured pressure lies between the pressure from the two versions. Once water comes on the vessel, the results of the two versio e w pse nearly on top of each other, and both are about 8 psi higher than measured.

From 0 to ~7600 seconds, version 4.1 and 1.2 are predicting approximately the same vessel pressure. At ~7600 seconds, the two versions begin to diverge. An examination of the steam flow boundary condition shows that, up to 7600 seconds, the model uses as measured steam flows whleh cyclically vary by about [ b "

, l .c, with a mean of [ )b.c lbm/sec. Frem 7600 seconds for a period of 9000 seconds, a constant steam flow of [ }D' lbm/sec was input. Once the input steam flow boundary condition became constant, small differences in circulation due to code version differences began to have an observable s'fect.

From -7600 seconda until the water is turned on (at about 34,000 reconds), the predicted vessel pressure for version 4.1 is lower (maximum of 7 percent difference) than both the version 1.2 predicted pressure and the measured test pressure. After the water is on, version 1.2 and 4.1 predict approximately the same vessel pressure (4.1 is slightly lower), and both versions predict pressure about 8 psi higher than measured.

The difference in calculated pressure between the two versions is consistent with the difference in air concentration distribution during the wet phase C. See Figures B.2 through B.5 for air pressure ratio and Figures B.6 through B.9 for helium pressure ratio. The air concentrations for 219.1 A and B, show that the vessel is somewhat more uniform for version 4.1 (more air above deck, less air below deck than version 1.2). For 219.1C, the air and helium concentrations for both versions are very similar. In both code versions, the Appendix B Comparison of LST Results Usir.g Solver Versions 1.2 and 4.1 October 1997 o:\3837.wpf:1b/101097 o

B3 trends in air and helium concentrations are similar. During the dry phase A, it is noted that the pressure predicted by version 4.1 is less than that for version 1.2, even though version 4.1 predicts more air above the operating deck. While the external surface is completely dry, the effect of different annulus cell centered velocities in version 4.1 is more pronounced. With a completely dry extemal surface, the resistance to vessel heat removal ir the annulus is dominant relative to the condensation resistance, so the different predicted steam concentrations in the two versions has a relatively weak effect. As can be seen in phase C pressure results, with water on the outs'de surface, the external velocity effect is nearly eliminated, and version 4.1 predicts lower pressure consistent with relatively higher above-deck steam concentrations.

Based on results from LST 219.1, it can be concluded that the net effect of the code changes is not significant for input models that are dominated by an imposed boundary condition, imposed boundary conditions that can be postulated to dominate the solution, on the basis of the LST 219.1 evaluation, include time varying steam flow rates, or actuation of the PCS.

B.3 Assessment of Variation in LST 214.1 Calculations The initial part of test 214.1 is natural convection (214.1 A); the second part is forced convection (214.18). The steam flow rate is nearly constant throughout the test, with a mean steam flow of [ ]b, Ibm /sec and cyclic variation of about [ )D C.Water coverage is measured at the elevation of the outside gutter where water is collected, and is given as a fraction of circumference that is wet. The water coverage is [ )b.c at the start of the test. During 214.1 A, the coverage is [ ]d.c, and during 214.1B, the coverage is

[ ]b C. Based on test observations, it is believed that the water coverage actually changed continuously from the beginning of the test to the point in time when coverage was measured to be [ )b.c. The input model used constant coverage values of

[ )b

• for 0-3000 sec., [ ]D- from 3000 - 9300 sec, and [ )b.c coverage from 9300 seconds and beyond.

The pressure transient for both versions are compared in Figure B.10. The pressure from the two code versions begins to deviate at the beginning of the transient. For this test the predicted vessel pressure for version 4.1 is always higher than the predicted vessel pressure for version 1.2, and after the step change in input water coverage, both versions are 2 to 5 psi higher than measured. The difference in relative code version results is not significantly affected by the change in external air flow when the fan is turned on (at about 9000 seconds).

The pressure results are consistent with the noncondensible predictions (see Figures B.11 through B.14). Version 4.1 has less axial noncondensible stratification than version 1.2.

Since the volume of the below-deck region is only about [ Ja b of the total volume, a change in air content at elevation F results in a relatively smaller change in air content above-deck.

Appendix B Comparison of LST Resutts Using Solver Versions 1.2 anc' 4.1 October 1997 o:\3837.wof:1b/101097

B-4 The relatively cons' ant input steam flow in this test may have allowed the differences in code versions to manifest itself in the air distribution early in the transient, and thue, in the predicted pressure. Both code versions predict pressures higher than measured once the input water coverage (at 3000 sec.) was set to the measured value.

B.4 Assessment of Variation in LST 216.1 Calculations Test 216.1 has a rektively constant steam flow with a mean of [

]b,c, and forced air cooling. The water was distributed over three quadrants for the first part of the test (216.1 A) and over one quadrant for th second part cf the test (216,1B).

The pressure transient for both versions are compared in Figure B.15. For 216.1 A, the predicted vessel pressures for version 1.2 and 4.1 are very similar with version 4.1 being slightly higher, and both being 4 psi higher than measured. For 216.18, version 4.1 predicted vessel pressure about 7 percent higher than version 1.2, with 1.2 being 8 psi higher than measured and 4.1 being 12 or more psi higher than measured. Version 4.1 predicts more air at elevttions Dome 63' and A, and less air at clovations E and F than version 1.2 (see Figures B.16 through B.19) illustrating that version 4.1 predicts slightly less stratification than version 1.2.

Examination of the plots for LST 216.1 shows again that with a relatively constant steam flow, the differences between the two code versions manifests itself early in the calculation in a slightly lower stratification gradient, consistent with the variation in pressure.

B.5 Assessment of Variation in LST 222.4 Calculations Test 222.4 has a steam blow down followed by a steady steam flow (222.4A); blowdown steam flow was a maximum of [ ]D lbm/sec, and the steady steam flow rate was

[ ]b,c. The steam flow rate was increased and allowed to come to

+ a second steady state (222.48) of [ ]b,c. Steam was injected upward into the vessel through a 3 inch diameter nozzle [ ]D'C above the operating deck.

The measured water coverage was initially [ ]b.c wet (01500 seconds), then

[ ]D' wet (1501 - 16,000 seconds), then [ )b.c wet (>16,001 seconds). The non condensable measurements showed this test has a relatively uniform axial air concentration, duo to the high kinetic energy of the jet.

The pressure transient for both versions are compared in Figure B.20. Version 4.1 predicted a lower vessel pressure than version 1.2 throughout the transient, although the differences are small during phase A. The differences between the noncondensible predictions for version 1.2 and 4.1 are small and have the same trend for all four elevations. At 15,000 Appendix B Companson of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o:\3837.wpt:1b/101097 e.

B-5 seconds, version 4.1 begins to deviato from 1.2, with version 4.1 having slightly loss air above the operating dock, consistent with the lower pressure in version 4.1. Both versions 1.2 and 4.1 predict significantly moro noncondensible axial stratification than the measurements (soo Figures B.21 through B.24), consistent with the lumpod parameter code momentum bias.

The stratification above the assumed break node inherent in multiplo node Nmped parameter models is apparent in both code versions.

B.6 Conclusions Based on comparisons : f solver version 1.2 to version 4.1, for LST calculations for 219.1 as compared to 214.1 and 216.1, the net offect may be less significant for code input models that are dominated by an imposed time varying boundary condition, (imposed boundary conditions -

which can dominate the solution include changing steam flow rates, or actuation of the PCS).

The lumped parameter momentum bias in open volumes represented by multiple lumped parameter nodos is apparent in both code versions. Thus, it is valid to impose stratification using lumpsd parameter biases as is done for MSLB, independent of code version, t

Appendix B Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 19E'7 o:\3837.wpt:1b/101007

4 B6 1

t b,c Figure B.1 Test 219.1 Vessel Pressure Appendix B - Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o:\3837.wpf:1b/101097

4 B7 b,c i

Figure B.2 Test 219.1 Air Pressure Ratio - Dome 63 In.

Appendix B Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 0:\3837.wpf;1b/101097 v .

s a - -

B8 b,c Figure B.3 Test 219.1 Air Pressure Ratio - Elevation A Appendix B - Comparison of LST Results Using Solver Varsions 1.2 and 4.1 Eober 1997 o:\3837.wpf:1b/101097

B9 b,c

\

Figure B.4 Test 219.1 Air Pressure Ratio - Elevation E Appendix B Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o:\3837.wpf:1b/101097

B 10 b,c Figure B.5 Test 219.1 Air Pressure Ratio - Elevation F Appendix B - Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o;\3837.wpt:lb/101097 ss _ _ . - _

B 11 1

b,c Figure B.6 Test 219.1 Helium Pressure Ratio - Dome 63 in.

Appendix B Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o:\3837.wpf:1b/101097

B 12 b,c Figure B.7 Test 219.1 Helium Pressure Ratio Elevation A Appendix B Comparison of LGT Results Using Solver Versions 1.2 and 4.1 October 1997 0:\3837.wpf:1b/101097

B 13 l

~

~

b,c Figure B.8 Test 219.1 Helium Pressure Ratio Elevation E ndix B - mparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997

B 14 I

b,c Figure B.9 Test 219.1 Helium Pressure Ratio - Elevation F Appendix B Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o;\3837.wptib/101097

B 15 b,c Figure B.10 Test 214.1 Vessel Pressure Appendix B Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o:\3837.wpf;1b/101097

B 16 M Prouvre RR Ewmen(X>6Y 1R

. ._83-.DC36, m

o -

.9 E .

6 -

d -

g h' - % ___--__3 o ', .,,i,,,,i,,,,i,..,i.,,,

0 4 8 ree <we) 12 16 g

Figure B.11 Test 214.1 Air Pressure Ratio - Dome 63 in.

~

Appendix B Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o:\3837,wpf:1b/101097

B 17 Air Pressure Ratio. Ewaton A 1R37

. ..---- ,0C37 co 6 -

6 -

w-e ..

, i ~-- ww------ 3,- - ...--- -

o .-

o '. ...t . , , ,i. ...i....i ..

1 4 5 12 15 g

Figure B.12 Test 214.1 Air Pressure Ratio - Elevation A Appendix B Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 0:G837.wpf:1tV101097

B-18 -

r

4 b

b Air Pressure Ratio,Elevaten E

- 1R66

. .;-- .OC38 .

K g '

E #

. f "'t,----

a . -

s j , L , %~ '

~s,.s-o -

L g  :

6 -

~

o -

o , , ,i,,,,i,.,,i,,. i ,,,

) 4 5 Tne(sec) 12 16 g

waome ts unw nim Figure B.12 Test 214.1 Air Prassure Ratio - Elevation E Apper.dv B - Comparison of LST Results Osing Solver Versions 1.2 and 4.1 October 1997 o:\3837.vvpf:1b/101097

B 19 Air Pmesure Rabo, Eevation F 1R46 DC39

.e-- un. - e q

. N ,' -

o -

%'s e

d

,~~,*s,  ;

1 ..

d -

~

N o -

i',,,i,,..i....i...

~*

g' 4 5 12 16 gg Tsne(sec)

Figure B.14_ Tcst 214.1 Air Pressure Ratio - Elevation F Appendir. B - CompaMn of LST Pea..Its Using Solver Versions 1.2 and 4.1 October 1997 c:\3837.wpt1bh01097 i

.

• i , m J

. B 20

~

b,c l

Figure B.15 Test 216.1 Vessel Pressure Appendix B - Comparison of _' ST Resu:ts Using Solver Versions 1.2 ar14.1 October 1997 o:4837.wpf:1b/101097

l .-

B 21 b,e i

s.

Figure B.16 Test 216.1 Air Pressure Ratio - Dome 63 in.

Appendix B Comparison of LST Results Using Solver Versions 1.2 end 4.1 October 1997 c:\3837.wpf;1b/101097 g.

a tv iy . .

, 4

. B 22 -

~

~

b,c l

. Figure B.17. - Test 216.1 Air Pressure Ratio - Elevation A Appwridix B - Companson of LST Results Using Solver Versions 1.2 and 4.1 October 1997 c:\3837.wpt:1b/101097

- B 23 l

i-b,c Id

- Figure B.18 Test 216.1 Air Pressure Ratio - Elevation E '

Apperdix B - Comparison of LFT Results Using Solver Versions 1.2 and 4.1 October 1997 c:\3837.wpf;1b/101097

l-'- B 24 b,c X

Figure B.19 Test 216.1 Air Pressure Ratio - Elevation F Appendix B - Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 0:\3837 wpf:1b/101097 a e -

.- B-25 l

1 b,c t

Figure B.20 - Test 222.4 Verwl Pressure

- Appendix b - Cornparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o:4837.wpt:1b/101097 gg . . .. . ..

o- B>26 b,c- ,

c

^

Figure B.21 Test 222.4 Air Pressure Ratio - Dome 63 in.

Apoendix B - Comparison of LST Resutts Using Solver Versions 1.2 and 4.1 October 1997 o:\3837.wpf:1b/101097 l 1

c. -

B :

t .-

t i

).

4 i

b,c -

!=

J 5 4

4 i-4 '

t-I 1

4 3

l t i

- Figure BM - Test 222.4 Air Pressure Ratio - Elevatiori A Appendix B - Comparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997-

. o:\3837.wpf:1b/101097; p -

9-

[' O.28-l l.

L b,c i

Figure B.23 Test 222.4 Air Pressure Ratio - Elevation E Appendix B - Cornparison of LST Results Using Salver Versions 1.2 and 4.1 October 1997 0:L3837.wpf:1b/101097 --

B 29 1 i'

l b,c-i

. Figure B.24 - Test 222.4 Air Pressure Ratio - Elevation F

. Apperdix B Cornparison of LST Results Using Solver Versions 1.2 and 4.1 October 1997

-o:\3837.wpf:1b/101097

I \l B 30 J

h d

Appsw4ix B - CsTperison of LST Results Using Solver Versions 1.2 and 4.1 October 1997 o:\3837.wpf:1b/101097 -

• 1 ,1 -

"