ML20196A303
| ML20196A303 | |
| Person / Time | |
|---|---|
| Issue date: | 11/19/1998 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-3058, NUDOCS 9811270111 | |
| Download: ML20196A303 (185) | |
Text
{{#Wiki_filter:i. - ,,f .. -. g.L... ~ . :.,,.i y.. p.
- .7
...y L.... _ I: _'l $ l_.f ?'JO,...: li '. .t.5. ..., 9, ~ ...L.
- .4
~.* h ~;. : M kWl$.4.$ i \\ ~ q: 4q jQ}4+ W, 3.s a a.e . cap .t . -4 .s c - ee. f.n:.t,:...
- c. g.a.gm 3 -
fl- . l*hh. A c 'c I '. ;R=.h. 9 .. 4g + -. q;. q 3 .,.....&g.
- (
1.?. ;:> L
- ..,..
- 2 C'.'.' i }.g.
,f m " '[ l. f +,, ; ' I ',3 f ^ ' X, y . A.c; :) lf).f. % j.;' f Q,; y 'r p ,(J I.:[ ^+.y e .g... g. .r .. x .st z; ; ~- !.%:.J . :.., ~..:.. .s,. s L. % ;~u , " 7 ' .,.' '"C".: b~,,. l ') 3 y,.. '.a; f *.,1. f. ' ]{ J, _ ? ' }.V ' ' - l $ % [ - Q. ' : 4.. q '. A.: u-l ; 3~ p* y ..:e,..- v-s 8 i,. g ..,,. - ~ .. y y g.; A MI.,.:. g g 9 " n, {i., p:. l . MI: %l.fi.L5 ~ ' % b.. :.h ; V
- f ~ f '.
.[ : /- ' ~'. ?: [. c 'Q%~ { 6
- 3. ?
+ " ^ j ) * * '
- [ - '}j.
~ w&.+
- m. m,.
- g. ~.. 4 y y + '
- )
. ',} Q...j:p If
<~
<~:
_.. [..
..a
. q.
wy, c i.-
.... ;. '. ; N 4 p.;,ew 1 '
' -.. ~ 1 ' -
Q.yWl M,, pyya.
.s
,' :,;g.n. i' t
s u..
..,. 3.-,
s...
..,V
,,...y..
y.y :,
~,~s 4-
.,.
.~ gy;q.y.
~
.c h
l
- p ': y._,,., _ 9.. '. o.
. l _._.
']3.
'.k =.l ?,;?
- p
- nl:n,Q. g.f
^ ( w,r. .(' - 1 ~ l t >., ?,. ; A .., /,.. ;.s. i. ., y ;.w 4.", '. :e... .n . p 1 .n' _Y ~ ',,,,, , "M,rmi.,% .,q."- = ns4 ,q k T > [, ' ;.
- D 7
%~ "M .Q. gg&g'.Q.A*f V,.:: f '.7,8;c t: .y Q' l :4.,., . ;.1 ' 6.~- .y lh5 h j '.i r Nfi%;}., A-f .R ~- ^ g;...\\ W -,_.. lhe ' } Q;u,ff t[ ~ d ; .. _.~' y~ cy9' -- Q W p,, - ^, . '... '. ', f:].Q9[ ~ J .a. . s
- . ' K. (g
.G t,M,ph'% ,r. g . ~. 2l ~ [, t. .ff ,,[ ".ky ' L. 1 7_ ?. y. . q., '.. J.: l f _
- P m:.m m.
.y 3- .s., r..... w-. .e 1 t ..-...g 3, \\ .?. W,.. ...%^
- L V*
?:M.;.i .nl:.,g, W ' L, .. c c?
- P 5
m.,W. hr p'$' 8' . se' - l',' i,\\,.<5-*..',' ' '. + E @&g.R v . %l R - - ' ~ -
- 6 J v'.
.D. .;. & _;l.,;. e-
- . ' f. '., s 3.x,; ;
.::. ; ;.y:t ; e ,.l, .t ,i, ~ . ~ f;.p. : t-e.,. + ( (; -_.,
- i. -
1 .~ ';.,..: ;,. - n 1.. .g.4.. ' c J,;,v ' a s'., ..g.4, .v.- ' V-. g n y. ..n ? ;..g p. 1,; x. .,3, <....y. y j [ - d _.:, z. .- ~ , +. u..,.. J'.. .s ... : (.. s .. ;3 - s r... . 3, ,r- ,;?,,,l l f. ) 3 l lv. . - :n a
- .s -
L. o.. s .i.,
- / n
- ; w ~ ; -4. (( (..;. f.L q:}.1,,
. ; q - ; ;:; c.,, ; m n v. a i . ;:. ~.: s :
- h. J,,
^ [.' f.;
- * '. J L
- ' ^ ' ' ".. ld.h.. ; i.. :..
V -..g.... .,..i.h. s] h l:. ,.z. ........ u. .. :.. ( a ^
- r..
c.i n.v. <:. k, , y y, o.> - :./ y t r 1 t E. 1.: - i*'... N O N. i: " $, /w."w.; 0 I 9A11270111 981119 1 ~ . G ) "S ~ .? ? '.. c f ~ h N +.' '.V:44 PvR ACRS n ' 9.,, 4., - T-3058 PDR . ',.. H. - c '..;.'.. . '.' '. 1..,: :.n: c;....e e. Q: '9 o
- n. S......J._
-,M. ,_. ;l..i 4 (.,. ) i r s ,V %g }, ~
- v,
$.,>' i,' " 7, s ~ 1;g ' ,.. g # 67.. 9 X ;~'l.:.T.',,, ,. r g.f'$.,.".,;" i N. .g..;
- s..
' f;.S.. " G c. ': # :j.[ J: y. ~.; b-v
- \\..,
7 '; j' q:" -,.. 3 gg ,1. - :.u ...../.,... . y _, {'- p .i-f g y. , f!D.l l y'.;p.;f ' ; ' ' '.',:.~.y .q - y,'# - ',' l ' ~ g j, ~. L, % ' ;<.' ;,. L ...,4 a , d ';,
- ' A
- .g ;. 3.,: q ;,..
3. _. ; ;.. /_ 9,g. ...,n....- w 4.h.' 'f -(. J.. :. m.. ;.:
- h..
V. .* ). { k '- L : ',' . 'l$ ) ' :.. b. ?.:: ~ .c L' - @Q;.%: _ ),,.. ; g'
- ...vi - ;3
... 'p..q..,,. ~ ~ y; .~.
- p. ^
- ~ L l _3: ? l_ Q}, y - ' '4 ~..4 ',:. -. ~ .v.,- . 1.y. ; +.. n;' :;>:...- ... p,;;;. ~ : . ;. : j.n '"):.. ; 7.- 3- { : [. ? 6 -c S v~s.. &y. n. .~ - v .:.., ~.
WM ;,.f<h k k ^.1 p, y, n_ y. , p o.s_ _ ..,^s 2< j i U m# sc 3 m, o m,.& n -m-L Ei :$, u. ........,,..'..t., s. r 3,,,. s >gjf 1P?, F,, ~.... ;7/OFFICIAIMRANSCRIPTJOF PROCEEDINGS ,i c n ~ ~ c c
- -kT ij, I'r
. -. 9 J- .. 4'r' S. I ' f ',r < j. g rw . 1 ,s J %?,, W AMNUC.LEARYREGULATORWCOMMISSION (w$l,~ [i:y):M [Fr,?>n - 1 ~ rNu \\ >w t Q ~, % MDVISdRY COMMITTEE ON REACTOR SAFEGUARDS i n ~- ~ t 7 . g 'E' -l ,f,. v.,;gW4. q ig 9 w;, MO;P E NOS E S S I 0;N 4 l e!_) : m,,, )'" e.~. { 3 y, ?? ?!Nj 'e f] 'I A g .,s .,g^ 9%;. fVfJ ' ^; .a > ' q -- e v s u Q '4 wpn M g;;n W s ',uW !a 4# " t c ev W ..J s t T;;:lhM %lq ,.,lTitlen,E < '. S O_,ACRS. MEETING REL khl h b + Y' l ^ s F T?' n ~, , h.,,, r ,.g j >.
- L K 1,-
- 1 g >
'
- l: Y 55
-6 ? y h ]b. e 1 r i- 's - T, s is'> C @wmK
- w' t LTHERMAleHYDRAULIC PHENOMENA -
c sw 4' x, m&nQfHQ ' ';. Q_,_ h s s r - s> <e w.;p; r, ? ' - l ;gh ,' H Y + '1i t ...t _s.p,. aj.n, n; j a <.<,,s an, 4 ~ Ro4 <xcas) u. ci 2 .F [4 A+ )/4 s g s- 'y + RETURN ORIGINAL g TC B XHTTE . a:% m + i
- j' M/b I-2E26
~,! w,. s
- 1., <
n. t [' [., p W> v f7)p'% h? g ,..'.3 V~ sh v - /M;.'>'y g"g 415-7130 s THANKS! C s M E we:* G.. E W ',~ ?' u, 3: E"y ~JW,e/ lDocketLNo.F. J - wr u v + ^ w , w s'u 1 i43" ,ygg'. 7 r t ~ :J5 'y 4'. ' g d ' M /, - - <- + 3 3 i f, v,Y ,.,g y ? .g__ s %:p ;p;w' pg ;, a ,i i n-pa ... ~ m y/ : 4.; t n \\;15 % ;s u. s a A GA g;w.- r 1_ m D. ' " u -~+,,f,,, y x m ~
- g.ni ',;p,p;m % yP ' r
- v. ?
s P q :.,;
- P r
i 1 ' W ;r 3: ,.%c J:- J (,, p - s., s 4 f ' lb{ &q;l,! e l {h 'UY '( ^ m m 1 , e p$ i c, tW6rliOrder No;:f 1A, SB-300-551' dj.f@+ % m u' .h s p &g %
- '.y-s w
vc n' i c.r e n% W-1 x g
- f. 4fM. ',
3., -l$'k l':gl :, * +
- 3.,
s D g c. j-s ., ~m.. m ..>r-s .#3 C}N;%Q<,hc
- g d LOCATION:l '.
iRockviHe,MD - ?;+.e a; ( y, ,m g' g:% O l'< ,fDATEi ( Thursday,Ncrember 19,1998 PAGES:1 - 37 ' f
- a. m A,1 c Jc M t W M.;3 y
C s 9811270111 981119 122 - 152 ' Dygg4 PDR ACPS 6 o it' r' T 7058 PDR 4 plAl2 7 ",'m %. f g g, .C7 yk gf q. o -a + e i y f, 4 e m % 3 e,, %,'
- ANN RILEY & A OCIATES, LTD.
y v y y@M's_ 1 %.p g'
- 1025 Connecticut Ave.,NW. Suite 1014 6
i Washington, D.C. 20036 N: nM (202) 842-0034 %sp ~sw
- w s'
~ ~ ~ ~ U_h 7 ';; 9 TC6 so)y-Reh an,, - y I i
(~V) DISCLAIMER UNITED STATES NUCLEAR REGULATORY COMMISSION'S ADVISORY COMMITTEE ON REACTOR SAFEGUARDS NOVEMBER 19, 1998 The contents of this transcript of the proceeding of the United States Nuclear Regulatory Commission Advisory r'(T Committee on Reactor Safeguards, taken on November 19, 1998, j as reported herein, is a record of the discussions recorded at the meeting held on the above date. This transcript had not been reviewed, corrected and edited and it may contain inaccuracies.
1 1 1 UNITED STATES NUCLEAR REGULATORY COMMISSION 1 m (w-) ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 2 3 4 ACRS MEETING RE THERMAL-HYDRAULIC PHENOMENA ) i 5 6 l 7 U.S. Nuclear Regulatory Commission 8 TWFE Building, Room T2B3 ) 9 Two White Flint North j l 10 Rockville, Maryland l 11 12 Thursday, November 19, 1998 13 s 14 The subcommittee met, pursuant to notice, at 8:30 I 15 a.m. qj 16 17 MEMBERS PRESENT: 18 GRAHAM WALLIS, Chairman, ACRS 19 THOMAS KRESS, Member, ACRS 20 21 22 23 24 25 [ ') ANN RILEY & ASSOCIATES, LTD. (_ / Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
2 1 PROCEEDINGS Q)h I 2 [8:30 a.m.] 3 DR. WALLIS: It is now 8:30. The meeting will 4 come to order 5 This is a meeting of the ACRS Subcommittee on 6 Thermal-Hydraulic Phenomena. 7 I am Graham Wallis, the Chairman of the 8 Subcommittee. 9 The other ACRS member in attendance is Thomas 10 Kress. 11 The ACRS consultant in attendance is Virgil 12 Schrock, and we're hoping that Novak Zuber will come by. 13 The purpose of this meeting is for the 14 Subcommittee to review the Westinghouse methodology for best m() 15 estimate small-break loss-of-coolant accident analysis using 16 the W-COBRA / TRAC code. 17 The Subcommittee will gather information, analyze 18 relevant issues and facts, and formulate proposed positions j 19 and actions as appropriate for deliberation by the full 20 committee. 21 Paul Boehnert is the cognizant ACRS staff engineer 22 for this meeting. 23 Portions of this meeting will be closed to the 24 public to discuss Westinghouse proprietary information. 25 The rules for participation in today's meeting ANN RILEY & ASSOCIATES, LTD. A/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
_._.______._.___..._._._....m_. L 3 y 1-have been announced as part of the notice of this meeting l () 2 previously published in the Federal Register on November 4, 31 1998. 4 A transcript of the meeting is being kept. It is i 5-requested that the speakers first identify themselves and 6 speak with. sufficient clarity and volume so that they.can be 7 readily heard. / 8 We have received no written comments or requests i 9- 'for time to make oral statements from members of the public. ( 10 Now, we have a timetable for this morning which 11 calls for a break at 9:45, and I note that the real i l 12 substance of the presentation, which is the code i ' 13 ' improvements and their validation, is to occur after that 14 break. () 15 So, I would do my very best to make sure that we 1 16 start that break on time or, if anything, before it's .17 scheduled. 18 I note that we have Novak Zuber with us now. 19 I'd now like to proceed with the meeting, and I'd 20 like to call up Mr. Robert Kemper of the Westinghouse
- 21 Electric Company to begin his presentation.
22 MR. KEMPER: Good morning. I'm Bob Kemper from l 23 Westinghouse, and I'm going to give a brief introduction to l- ) 24 kick off our session today, j 25 I'll be followed by Dr. Art Ginsburg from Con 2 L ANN RILEY & ASSOCIATES, LTD. I i Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 9
t 4 i 'l Edison, and~we'll be speaking to the best estimate I ) 2 -small-break LOCA analysis methodology that we're developing 3 for three-.and four-loop conventional PWRs. i 4 At Westinghouse, we remain committed to. developing i 5 -improved methodologies for LOCA accident analysis and i 6' methodologies that are improvements on and go beyond i 7 Appendix K. l i 8 .Our intention is to have the capability to better 9 meet customer and regulatoi/ needs and have these tools 10 .available so that we can better understand new issues and I 11 evolving issues with models that have a good basis. 12 Over the years, our development in LOCA has 13 included a number of things. I 14 We originally got into COBRA / TRAC for two-loop () 15 PWRs with UPI. This is back in the days of SECY 83-472, and 16 there were some issues raised by the staff at that time and 17 were addressed under that methodology with a COBRA / TRAC 18 model for UPI plants, the two-loop upper plenum injection 19 units. 20 Later on, we went into best estimate large-break 21 LOCA technology, and in 1996, we received approval from the 22 staff for a large-break best estimate methodology for three-23 and four-loop plants, and this was based on the CSAU 4 24 -methodology that had been provided in the period between l 25 1983 and 1996. I O' Court Reporters ANN RILEY & ASSOCIATES, LTD. l 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
5 1 DR. SCHROCK: Excuse me. You emphasize 2 large-break in that presentation. Does that imply that the 3 SECY 372 was a large break or a small break? 4 MR. KEMPER: That was really a -- sort of an 5. amalgam of a best estimate and Appendix K approach. It 6 enabled you and encouraged you to use best estimate-type 7 techniques, but Appendix K restrictions were still plied 8 back in those days. So, it was -- 9 DR. SCHROCK: Well, I don't 'hink you've answered 10 the question. We're here, I thought, today to talk about 11 best estimate methodology for certain PWR plants for 12 small-break LOCAs, not for large-break LOCAs. 13 MR. KEMPER: That's correct, sir. 14 DR. SCHROCK: Okay. So, at this point, it's () 15 unclear to me whether your second item on that slide implies 16 that you were dealing with small-break LOCAs. 17 MR. KEMPER: No, this is just a brief history of 18 19 DR. SCHROCK: Well, this is the point. I mean we 20 have limited time. Why do we want to fill up the time i 21 talking about large-break LOCAs when the focus here is -- 22 MR. KEMPER: The real bottom line here is that, 23 having worked through large-break, we're now looking into 24 small-break. 25 DR. WALLIS: That's the main message, but I would 1 [) ANN RILEY & ASSOCIATES, LTD. A_/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
6 1 like to ask you what you mean by best estimate -- () 2 MR. KEMPER: Okay. 3 DR. WALLIS: -- so we understand the rules here. 4 MR. KEMPER: Well, our methodology is -- for 5 small-break is likewise going to be from the CSAU approach. 6 DR. WALLIS: Well, best estimate -- I've seen best 7 estimate analyses which then invoke conservatism, and that 8 seems to be incompatible -- two incompatible approaches. I 9 mean either it's a best estimate or it's conservative. 10 MR. BAJOREK: I think what we mean by best 11 estimate is the inherent conservatisms in plant operation 12 and models are identified and removed. We've gotten rid of 13 things like the decay heat and other models, for example, in 14 the heat transfer, which causes very large PCTs in and by () 15 themselves. 16 We do get a conservative estimate of the 95th 17 percentile PCT by a statistical combination of the 18 uncertainties, then. 19 DR. WALLIS: Yes, that was my understanding, too. 20 So, we're looking for best estimates and not conservative 21 assumptions. Thank you. 22 MR. KEMPER: The program we have ongoing is a -- 23 involves us at Westinghouse, Consolidated Edison, and EPRI, 24 with the purpose to develop and license a small-break LOCA 25 best estimate methodology that will be applicable ANN RILEY & ASSOCIATES, LTD. s_. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
7 1 generically for three-and four-loop plants and to perform a '(). 2 first plant-specific application for Indian Point Unit Two. 3 Our goals for today's meeting are to introduce the 4 program to you so that you get a feel 4or and know what 5 we're about here, to give you an overview of the 6 methodology, the fact that it's a CSAU approach, which Dr. 7 Bajorek will address. 8 We'll go through the small-break LOCA PIRT. Dave l 9 Shimech from Westinghouse will speak to that. ~10 Then we'll summarize things that we've done to 11 COBRA / TRAC to enable us to pursue small-break LOCA analysis, 12 some new models and the validation that we intend to do, and i 13 Dr. Katsu Ohkawa will present that part. j 14 Then I'll outline some preliminary results we have (%. (_) 15 for Indian Point Two and our thoughts on the methodology 16 that we will use to obtain this 95th percentile PCT. 1 17 DR. WALLIS: You said there's validation you 18 intend to. The additions are presumably based on some 19 evidence, so there's some pre-validation. 20 MR. KEMPER: That's right. We have been, you 21 know, looking at these facilities. At this point, we're on 22 the verge of freezing the code and doing the validation that 23 will be our submittal. 24 DR. WALLIS: Presumably, it's better because of i 25 evidence from the past that indicated that something better b'4 ANN RILEY & ASSOCIATES, LTD. O Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
8 1 was required or could be performed. [ ) 2 MR. KEMPER: Yes, I think you'll see that. \\) 3 DR. WALLIS: So, there is sort of a trail of 4 evidence that says this thing you intend to do looks better 5 than what you did before. 6 MR. KEMPER: Well, we believe it is, and you know, 7 of course, we're interested in obtaining feedback and 8 comments from you. 9 So, that's all I had. 10 Dr. Art Ginsberg would now give a perspective from 11 the utilities' standpoint about this program. h,. 12 MR. ZUBER: When did you start the program? 13 MR. BAJOREK: We started almost immediately after 14 the best estimate large-break program. So, we've been at (O,) 15 this for better than two years now. 16 MR. ZUBER: What is the manpower involved? 17 MR. BAJOREK: We've generally had about four 18 people full-time on this, in some cases a fifth. 19 DR. GINSBERG: Can you hear me? Okay. 20 Thank you for allowing me to address you this 21 morning. 22 I found that, in doing the large-break best 23 estimate LOCA, it was useful to give a presentation on why 24 this is important to a utility and why is Con Edison 25 supporting this project, and so, I thought I'd do the same O~ ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
9 1 thing here for the -- when~we sdart off the~small-break 2 project. 3 Con Edison has been involved with EPRI and 4 Westinghouse in developing best estimate methodologies, and 5 our. goal here is to provide a superior product for best 6 estimate analyses. t l -7. We are coming to NRC and ACRS early in the 8-licensing process for two reasons. 9 Number one, we think it's better for the overall 10' licensing process, and the other thing, in my mind, is I 11 want to get insights from experts in the field, because I 12 really want to get this to be the, you know, landmark 13 project, landmark code, landmark methodology, and we did the l 14 large-break between 1989 and 1997. ) 15' In 1994 and '95, I started thinking about why not 16-do this for small breaks, and there was a lot of questions 17 about that, because small-break, typically, you had a lot 18 .more margin than large-break, why were you doing this, and 19-I'm going to try and give some feeling on why we thought it 20 was important. 21 First of all, for a utility -- this may sound a 22 'little trite, but safety is our number one goal. If we l l 23 don't operate the plant safely and continue to show safety, 24 we're not going to make it for very long. It's very, very I 25 'important to us to maintain safety, and we think using the r i 1007 RILEY. & ASSOCIAIES, LTD. I Court Reporters i 1025 Connecticut Avenue, NW, Suite 1014 i Washington, D.C. 20036 (202) 842-0034 i-
m 10 j 1 best estimate or realistic methods for LOCA is one of the (/~m) 2 keys of providing new insights into safety, and that's why x_/ 3 we're very interested in it from a safety perspective, and I 4 mentioned Reg Guide 1.157, and that's what -- at least from 5 the utility standpoint, we've been trying to make sure that 6 this project follows that. 7 What are the benefits in terms of safety? 8 Well, better understanding of the accident to 9 allow us to concentrate on real safety issues. One of the 10 by-products of this best estimate is higher peaking factors, i 11 which will allow you to load fuel more towards the center of 12 the core, which reduces vessel fluence, which is good from a 13 safety point. 14 We also want to provide more margin for the n'( ) 15 operators in operating of the plant. The more margin you l l 16 provide them, better operation you have. l 17 From a regulatory standpoint, we want to I 18 demonstrate even more margin, the 2,200 PCT limit, less i i 19 reporting requirements than we have with current hybrid t l 20 models, and we feel that we do have encouragement from the 21 ACRS, you know, throughout this project, and the NRC, and we l 22 appreciate that. l f 23 One of the big debates that we had in the industry i l 24 was what kind of code to use. A lot of people wanted to use l 25 existing two-fluid codes and try to modify them with j / ANN RILEY & ASSOCIATES, LTD. 'N Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 N s ' (202) 842-0034 s
11 1 different models, and we felt very strongly at Con Edison, /~ ' O) and the people at Westinghouse also felt that way, that we 2 3 really should use the same methodology as the large-break 4 best estimate. 5 So, we insisted that, for this project to go on, 6 'for us to support it, that we would use the COBRA / TRAC code, 7 and I guess we'll find out later on -- I guess the 8 advancements in computer technology is one of the things 9 that's helped us do that, because in the old days, older 10 computers, it would take so long to make these runs, it was 11 almost impractical, but we think know, even though it takes 12 a long time to run these codes, that, you know, since this 13 is the best code to use, we want to use it. 14 In terms of increased flexibility for the utility (O,) 15 operators, more -- one area is the training area, more 16 efficient training of the operators, better understand the 17 transient so he -- to predict how he would have to react in 18 the -- in a small-break LOCA, and one of the important 19 elements of this program which we're going to talk about 20 later is operator action. ) 21 In the large break, during the injection phase, 22 operator really doesn't have that much to do, it happens so 23 fast, but in a small break, it's a longer process, and 24 there's a lot more things that he has to do, and we felt it 25 was very important that, in this project, we took into / ANN RILEY & ASSOCIATES, LTD. (_ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 I Washington, D.C. 20036 (202) 842-0034
.._.. - - _.. -.~. ~... t .12 i 1_ account the effect -- the potential effects of operator 't )
- 2~
action, mistakes or faster action, slower action, and'we are 3 -including this in the project. 4 We have done a lot of runs in our IP2 simulator to ~ 5 -- with Westinghouse people coming up and trying to see what 6 the operator actually does to try and get a better feel for 7 how to put him into this analysis, and I had an invitation 8 to the NRC, and I also have the same invitation for members 9 of the ACRS, that at some future day, I would like to set up 10 a meeting up at Indian Point and have you people come up and i i 11 -- you know, I don't -- if you have or haven't done this { J j 12 before, but go through a day on the simulator with us and 13 see how the operators actually operate during an accident. l l 14 Anyway, let me just keep going on. i G( l / 15 We think this is a major advance for the industry, 16 and we also think it will help us economically at the 17 utility, because it will help us do fuel management i 18 improvements, reduce reactor trips, and we also are looking 19 for a potential' upgrade of our plant again -- we've had one 20 upgrade in the early '90s -- and also, we're looking for 21 potential life extension. 22 So, that's some of the -- in the package, I gave 23 you a little information about the IP2 plant, but I'm not 24 going to go over it now, because -- in the interest of time, t i 25 but I really wanted to answer -- someone had raised this e i } ANN RILEY & ASSOCIATES, LTD. j Court Reporters l 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034 i i
l-l 13 l 1 question. j I-() 2 We're very serious about this project. We're i 3 ' spending aflot of money..Of course, we think we can improve 1 4 the safety, we're going to improve the' economics of a plant, .5 and this is kind of an overall organization chart. 6 One change, Dr. Bajorek is now teaching at Kansas '7~ State, and Bob Kemper, who gave you the first presentation, 8-is now going to be the technical lead, but we've kept Steve 9 on the project for the continuity and for his insights and 10 abilities on this, and we have, you know, a large project 11 team, but what I wanted to point out is that we do have 12 senior management from both Westinghouse and Con Edison to 13 oversee this project. That's how important we think it is 14 for the utility. And also important is we have EPRI L ( 15 involvement. .They are one of the co-leaders of this 16 project. l 17 DR. WALLIS: What is the EPRI involvement? ) 18 DR. GINSBERG: EPRI involvement is financial and 19 project management. l 20 DR. WALLIS: Is it technical? l 21 DR. GINSBERG: Technical review. l 22 DR. WALLIS: They're not involved in helping you i l 23 figure out what to actually put in the code. 24 DR. GINEBERG: No. Primarily Westinghouse. 25 DR. WALLIS: Okay. l [\\- ) ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l t
_m._....__. 14 1 DR.. GINSB3RG: Although when we review the Con /\\ 2: O Edison and EPRI review reports, we have regular meetings, 3 'you know, and try to ask probing questions, and if we can 41 improve it at all-we will do it, but it's primarily 5-Westinghouse. i 6 Just some general feelings of the type of benefits 7 you can get by increasing margins. s l 8 I mentioned upgrading power plants, operating 9 plants at higher temperatures, more efficiency, reduction in l l 10 required --'at least analytically required SI flows and RHR 11-flows, makes it easier for plant personnel to do balancing 12 operations during refueling operations, wider bands on 13 accumulators, using new fuels. 14 One thing that we found very useful in the large () 15 break which we wanted to extend to the small break is we've i 16 added time'to the diesel start. 17 Now, you could take that either as starting the 18 diesel or the actuation of all the valves, and recently, we 19 had a problem where there was an issue in the industry, i 20 thermal-binding, pressure locking thermal binding, where a i 21 lot of valves had to be worked on. 22 We were able to change the actuators in our valves l 23 within changing the whole valve, by taking credit for 24 increased time for this SI sequence to occur. I l 25 Also, it's more accurate modeling of the accident, i. 1 3( ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034 =
15 i 1- .using the_ kind of conventional Appendix K approaches, /T 2- ' combine different effects, and they may be competing ' Q/ 3 effects. By doing everything realistically, we think we get j 4-a better feel for how to really' analyze the. accident. l 5 DR ~. KRESS: These are benefits you've'already had 6 because of'the large_ break? 7 DR. GINSBERG: We haven't taken credit for all of -8 them yet, because some of them -- for example, like the l 9 accumulator level, yes, because you can -- that's just 10 really basically a large-break phenomena, but something like 11 peaking factors, we haven't, because we have to do -- finish 12 the small break. 13 DR.-KRESS: Small break is the driver in that one. i 14 DR. GINSBERG: Yes, in that one. Well, they all { () 15 are. We have done the large-break best estimate; we have J 16 -taken care of things from a thermal DNB standpoint. The big 17 key that we're missing is realistic small-break best 18 estimate. I 19 Now, these are general advantages of large-break 20 and small-break best estimates. 21 What I wanted to show next -- and I won't go over i 22 it, because it's kind of detailed, but I gave it to you to 23 look at -- specific benefits that are looking for in this l 24 particular project, and I'll just briefly go over them, in 25 the interest of time, unless you have questions and you want 4 ANN RILEY & ASSOCIATES, LTD. Court Reporters 4 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 4
._.__.__._.._..-.__._....._-._.._.__..-._._.__..y 16 1 Eto go through it in a-little more detail, but higher peaking / ) 2 factors, red'uced flows. 3 This is one interesting thing. As plants have 4 gone to longer cycles -- for instance, we're a 24-month 5 . cycle -- the drift errors that you have to put into the s6 instrumentation are very large, and what that's done is 7 reduce the margin to tripping. 8 For like a low-pressurizer pressure trip or a 9 low-pressurizer SI actuation, there is not as much margin as 10 there once was, and we want to demonstrate analytically that 11 there should be more margin, and that's another big 12 advantage for this best-estimate LOCA approach. 13-And I have two pages of that, but like I said, you ~ 14 can look at that, and if you have any questions, you could () 15 always ask me, but in the interest of time, I'll just 16 MR. ZUBER: Just a short question. 17 DR. GINSBERG: Sure. 18 MR. ZUBER: Overall, without summing all those 19 numbers, what is the gross amount of saving per plant? 20 DR. GINSBERG: The fuel management -- it turns out 21 to be the biggest economic savings. We think you could save 22 $2 million a cycle, like four fuel elements by utilizing the 23 -- other benefits like power upgrade -- this would be only a 24 small part of it for the overall power upgrade, but nower 25 upgrade could be worth -- I guess I can't -- I don't know ANN RILEY & ASSOCIATES, LTD. (_ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
- - -. -. -... -.. -. ~. -.. -... -. - ~. - -. -... -.. -.. -.. _. -. -.... _ 17 L 1 exactly, but you know, if we can increase the power of { r. M. L2 b: . plants 5.' percent or 10 percent, it's quite significant. .3 DR. KRESS: By having a good small-break best i 4 estimate code, you can show that you still meet the d l 5 regulatory requirements by -- at a higher power -- 6 DR. GINSBERG: Yes. 7 DR. KRESS: for all of the spectrum of things 8 you have to look at. -9 DR. GINSBERG: Definitely. One of the things that 10 we'll-get into, which I kind of found interesting, is, you 11 know, we always talked large breaks, small breaks, what 12 about intermediate breaks? We think, for the first time, 13-we've been able to cover everything. ( 14 DR. KRESS: You can look at the whole spectrum [ 15 with a single code. 1 16 DR. GINSBERG: Yes. That's the way I see it, and ] I 17 I think that's great, because we always have questions, you 1 18 know, what about the intermediate breaks, what about, you 19 know, going from a small break to a large break, can you be '20 consistent? 21 DR. KRESS: You're required to determine which 22 break is the limiting one anyway, aren't you, in the 23 regulations? 24 DR. GINSBERG: Yes, but it's kind of interesting i j 25 that the industry has gone like large break /small break. i i i I l 4 i ANN RILEY & ASSOCIATES, LTD. i Court Reporters l 1025 Connecticut Avenue, NW, Suite 1014 d l Washington, D.C. 20036 1 (202) 842-0034 l i' i
18 1 DR. KRESS: Yes, I know that. \\ \\ ['T 2 DR. GINSBERG: So, I think we can answer that once %f 3 and for all, I hope. That's a good part of this. 4 DR. WALLIS: You mean large break and small break 5 do not overlap? 6 DR. GINSBERG: Typically, in the industry, the 7 methodology has been very different. The codes have been 8 different. l 9 DR. KRESS: It's hard to compare the two results. 10 DR. GINSBERG: When you're doing conservative 11 Appendix K, yes, but when you start getting into best l 12 estimate, I think -- at least I've seen it so far -- it 13 looks like you have a much better idea of being able to j 1-4 cover the whole spectrum. { t gg (_,) 15 DR. WALLIS: Well, the naive observer might assume 16 that a good code would do both large and small and it has 17 the right physics. 18 DR..GINSBERG: I'll let Steve go into that more, 19 but in terms of -- if you have one code that you could run i 20 forever on a computer, yes, but in the past, a small break 21 really stretches out so long, the numerics and everything, 22 it's just kind of difficult. 23 So, that's why I guess there's been -- and another 24 thing, too, is -- this was a debate in the industry. Most i 25 plants, for small break, have a lot of margin, 2,200. Why i i l ^ ( ANN RILEY & ASSOCIATES, LTD. 2 \\,, Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
19 1 do this? I hope I give you a little feeling for why I think / 2 'it's important.to'do_it. \\ l '3 It's a consistent approach. You're always going 4' to get.more margin, which is -- for the utility, it's great, 5 . we could always use more margin, and so, I think the 1 .6 combination of those factors'-- that's why I wanted to do 7 it, and that's why we convinced our management that it was 8 important to do. 9 The only other message -- I know, in the industry, f f, 10 the resources for reviewing these things aren't all there, 11 and it's going to be hard to review many different i 12 submittals. l 13. What I want us to show today is that we are ahead ) 14 of everybody else in terms of what should be done. We're t 15 trying to make it as, you know, good a product as possible, l. ( 16 and I-hope _that we will get the review time. 17-DR. KRESS: We view margins from a little 18 ~different viewpoint than you do. 19 DR. GINSBERG: Excuse me? t 20 DR. KRESS: We view the margins a little 21 differently than you do. We saw the margins are there 22 'already, and what this allows you to do is use more of them, 23 use them up, and do so, hopefully, in an acceptable manner. 24 DR. GINSBERG: I agree with you. 25 DR. KRESS: That's what we look at it. c i 5 ANN RILEY & ASSOCIATES, LTD. Court Reporters i 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 1 (202) 842-0034 i a
i 20 1 DR. GINSBERG: I agree with you. If I didn't say \\ (n-) 2 it that way, I agree with you, because you start saying -- 3 people are questioning whether you're reducing the margins 4 by -- 5 DR. KRESS: Yes. You're not changing the margins 6 at all. 7 DR. GINSBERG: -- demonstrating -- 8 DR. KRESS: You're better characterizing them and 9 hopefully be able to make use of those margins in an 10 acceptable manner, and the question is what's acceptable. 11 DR. GINSBERG: I agree with that. 12 DR. WALLIS: I would hope we're not just using up 13 margins, that better understanding might lead to better 14 operating strategy or something which will actually increase l' 15 the margin. v 16 DR. GINSBERG: I hope so, and that's why this 17 operator action is very interesting, and how we handle that. 18 DR. WALLIS: So, are we ready to move on? l 19 DR. GINSBERG: I think so, yes. Thank you very 20 much. 21 DR. WALLIS: Thank you. 22 DR. SCHROCK: Could I make a point here? It seems 23 to me, since he's already stressed the limited review time 24 available for this thing, that the process needs to be a bit 25 more efficient than it has been historically, and one of the [~ \\ ANN RILEY & ASSOCIATES, LTD. Nl Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
.~ 21 1 things that struck me in reading the ACRS letters on the /~') 2 meeting on -- or the conclusion of the large-break exercise %s' 3 was that the staff was asked to prepare some guidelines for 4 future reviews of this kind. 5 It would seem to me that reviewing those again, 6 having us understand them, having the industry and NRR all 7 together on what it is we're dealing with in that respect 8 would tend to streamline the process. At least I would hope 9 it would. 10 There was also in there the suggestion that 11 Westinghouse was going to provide comprehensive 12 documentation to cover the basis of the large-break best 13 estimate methodology, and it would seem to me that having 14 that at hand would also be useful as we go through this r I 15 exercise. 16 There are some other items in there, but I'll not 17 belabor that. 18 It just seems to me that setting the stage for how 19 it's going to get done now on this go-around might save a 20 lot of time. 21 MR. ZUBER: If I just may second what you said, I 22 think, for a very efficient operation, which you would like 23 to have and ACRS would like to have, it's mandatory to have 24 good documentation. 25 In the past, Westinghouse did very lousy on that, [} ANN RILEY & ASSOCIATES, LTD. \\- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
22 1-and that was part of the problem, and hopefully, you will . [~} 2 learn from mistakes and you change your course in this LJ 3 activity. I do urge you. Otherwise, we waste time again. 4 DR. WALLIS: I'd like to stress, too, we need the 5 documentation well ahead of time, so we can review it before 6 you come with a presentation. We don't want the 7 documentation a week before or after a presentation, that 8 kind of thing. 9 MR. LANDRY: Mr. Chairman, Ralph Landry from the 10 NRR staff. 11 If you can remember those comments when I give my 12 presentation, I think I'll try to answer them, lay out what 13 our plan is for this review and other code reviews, what our 14 schedule plan is, and what our approach is. ,o ( i - 15 DR..GINSBERG: Just as a final comment, I'll say s/ x 16 that we are committed to helping with this documentation as 17 much as we can to do what you want. We agree with you. 18 MR. BAJOREK: Good morning. My name is Steve l 19 Bajorek. Currently, I'm an assistant professor at Kansas 20 City University, in the Department of Mechanical and Nuclear 21 Engineering. 22 What I'm going to do this morning is to provide a 23 general overview of our technical approach. I think, as 24 you'll see as we go along, that we are trying very hard to 25 incorporate a number of the lessons learned in our l l /^ ANN RILEY & ASSOCIATES, LTD. , \\m, Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
23 1 large-break code development exercise to try to make sure () 2 that the small-break development and its licensing and as we '% ) 3 bring it to the ACRS goes much smoother and more efficiently 4 that it had before. 5 As we mentioned earlier, we started this work 6 about two years ago, right about the time when the 7 large-break development was completed. 8 We thought a lot about how the development had 9 gone, and we went initially to a review of NUREG/CR-5249, 10 the original CSAU document, and also some guidance that had 11 been provided in Reg Guide 1.157 by the NRC in order to help 12 us come up with a better approach for this. 13 We decided that, based on the generality in the 14 original CSAU approach, our best approach was to follow CSAU f i 15 through the small break, and we started with developing an %/ 16 initial road-map of how this methodology should look. This 17 helped us find where we are going to have to sink a lot of 18 time and effort into the development, and it also helped to 19 clarify the types of things that we should expect as we come 20 closer to the statistical methodology. 21 Starting off with CSAU, the two things that were 22 different from what we had done before, first we are going { 23 to assume the scenario is a small to intermediate-size break 24 somewhere in the primary coolant system. 25 Perhaps more important or more novel is that we O ANN RILEY & ASSOCIATES, LTD. b Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
24 1 are going to assume that operator actions could occur during [~)' 2 the course of the accident. \\/ 3 To start, we went and we reviewed the emergency 4 operating procedures. 5 As Dr. Ginsburg mentioned, we went to Indian 6 Point, we watched.the operators go through the simulator 7 training for a number of small-break scenarios so that we 8 understood which consequences could really affect the 9 accident, and we've been using those in the course of our 10 sensitivities to gauge the effect of the operators. 11 Of course, Indian Point is the power plant that we i 12 have selected for this study. j 13 We spent, initially, a fair amount of time 14 developing a PIRT for the small-break accident. ) 15 We realized, after we looked at doing this, that 1 16 this was the first time anyone had developed a PIRT for a 17 conventional Westinghouse plant. There had been work done 18 for a B&W plant and a significant account for small break in 19 the AP600. 20 What we did -- so, we developed the PIRT. We then 21 went to an outside group of consultants to get their 22 comments. We incorporated those into the final PIRT. 23 Dave Shimech will discuss the PIRT itself and the 24 process in a little bit more detail following me. 25 MR. ZUBER: I have a question. In that letter (/) ANN RILEY & ASSOCIATES, LTD. A-Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
m. _,__m__., ...m __..mm.-. ' 25 1 that you wrote to ACRS, you mentioned that you had to -2 develop a new PIRT because the one developed by NRC was not 3 applicable. '4 .MR.'BAJOREK: That was for large-break. 5 HMR. ZUBER: Well, I know that, but they also 6' developed one for small-break. 7 MR. BAJOREK: For the AP600. 8 MR. ZUBER: No, before the AP600, it was my 9 understanding they were working on -- 10: DR. SCHROCK: It was a B&W plant. 11 MR. ZUBER: A B&W. 12 MR. BAJOREK: Yes. 13-MR..ZUBER: Okay-. 14-There is also reference here to work that was done () .15 in Los Alamos. Which one is that? 16 MR. BAJOREK: This was the work that Brent Boyack 17 had done, again looking at AP600. 18 MR. ZUBER: Okay. 19 MR. BAJOREK: And I want to mention, in part of 20 that process, we did talk with Dr. Boyack and also Gary 21 Wilson, and he did help us come up with some ideas on how we 22 could better formulate our PIRT. 23 When we finished the PIRT, we found that we could 24 summarize those processes that were ranked highly in terms 25 of eight treatable uncertainty parameters, and I apologize, [U\\j ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 -Washington, D.C. 20036 (202) 842-0034
26 1 on the copies that you had, I left out perhaps the most () 2 obvious one. 3 We had break flow, mixtura level swell, the fuel a rod mcdels. We found that, in looking at the small-break 5 uncertainties, we've had to retain a number cf the 6 uncertainties that were there in large-break. They haven't l 7 gone away, and they still have to be addressed. 8 Horizontal flow regimes, referring to the 9 transitions, CCFL in the hot-leg elbow, steam generator 10 hydraulics, including flooding in the tubes, condensation 11 within the steam generator tubes, the loop seal clearance, 12 which is the one thing that makes this significantly 13 different than the small-break PIRTs for AP600, degraded 14 pump performance, and condensation as it occurs in the () 15 loops. 16 DR. KRESS: What's the significance of the word 17 " treatable" in your title? 18 MR. BAJOREK: These were overall categories that J 19 summarized where, in our PIRT, high occurred either for all 20 phases or would have a very significant effect on the 21 transient and were such that we would have to assess the 22 models that predicted these events, develop some type of a 23 variation of that model in order to assess uncertainties in 24 these models when we propagated them at full-scale. 25 DR. KRESS: Actually, that should read, then, O ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washinston, D.C. 20036 (202) 842-0034 l l l
27 1 uncertainty parameters;that should be treated -- 2 MR. BAJOREK: Yes. 3 DR. KRESS: -- rather than the word " treatable." 4 MR. BAJOREK: I tried to retain -- this is the 5 same terminology that was in the original CSAU, and I think, 6 again, there were seven or eight of these types.of 7 parameters. .6 DR. WALLIS: Could we say, for example, when you 9 say mixture level, then mixture level is presumably - 10. predicted by having ome sort of a bubble rise or something 11 and you have a velocity and some uncertainties, range of 12 velocities, and that's all figured into the uncertainty, 13 then. 14' MR. BAJOREK: Yes. I mean uncertainties in the () 15 vertical inter-facial drag, the bubble size, the bubble 16 break-up are somewhat globalized in what we call the mixture 17 level. 18' DR. WALLIS: So, eventually, it's uncertainty in 19 some physical parameter. 20 MR. BAJOREK: Yes, sir. 2.1 DR. WALLIS: Yes. 22 MR. BAJOREK: Step four, select the frozen code. 24 This and step six is a bit of a circular pattern. 24-We started off with COBRA / TRAC Mod 7A, which was 25 the version that was approved for large-break. ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
1 28 1 After we went through the PIRT and did some ['} 2 initial testing with this version of the code, we found it %J 3 to be seriously deficient for a number of the small-break 4 parameters. 5 MR. ZUBER: Such as what? 6 MR. BAJOREK: Flooding in the hot-leg elbow. The 7 break flow model had to be changed significantly. We have a 8 bias in the models for mixture level. We're trying to 9 incorporate those. We're still working on condensation 10 models for the steam generator and in the loops. 11 So, what we've done is we've gone to the PIRT, we 12 did initial testing, it gave us an idea how the code behaved 13 without changes. 14 We've been going through, adding new models, () 15 testing them as we have gone, and we'll come up with a 16 COBRA / TRAC-SB, a small-break version which will have all the 17 models that we've identified from the PIRT developed and 18 added into the code, and this will be our frozen code from 19 this point on. 20 MR. ZUBER: Do you think it will degrade its 21 capability for large-break LOCA when you change so many 22 things? 23 MR. BAJOREK: No. That's a bit surprising, but 24 what we have done to the code, I think, as we'll see, is 25 we've gone to a lot of work to almost essentially eliminate ANN RILEY & ASSOCIATES, LTD. A Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
29 1 the TRAC part. [) 2 We've taken advantage of the COBRA formulation and V 3 extended this through the loops. 4 Conceivably -- and we would have to check this and 5 test it -- a user could model the large break as large 6 break, use the code, get his answers, use the small break 7 and use those models to model the same plant. 8 The nodalization will change. Some of the models 9 will be active in small break but are of no consequence in 10 large break, prediction of a horizontal stratified flow, for 11 example -- it's gone in large break but very important in 12 the small break -- to give us a continuum from small break 13 through an intermediate to large. 14 I think, as Bob will show, he'll show a figure, (~' t 15 PCT versus break size, and we'll see a fairly close merging x 16 of the types of PCTs we were getting with large-break with 17 this upper end of the intermediate spectrum. 18 There's some difference, because we have changed 19 some of the plant conditions, the peaking factor and the 20 power, but by and large, it looks as if, at this point, we 21 would be able to go through the intermediate and merge with 22 the large-break with the same code. 23 DR. SCHROCK: What is the time-table for freezing 24 the code? 25 MR. BAJOREK: We're hoping to freeze the code by [\\ ANN RILEY & ASSOCIATES, LTD. V Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
-l l L i L 30 1 1 the end of'the year. 2: DR. SCHROCK: By the end of '98? [ 3 MR.'BAJOREK: Ninety-eight. 4 DR. SCHROCK: That implies that all new work to 5 modify models is already completed. l. 6 MR. BAJOREK: It's in progress at a bit of a 7 hectic pace, but we're -- we think that the break flow i 8 model, as Dr. Ohkawa will point out -- we think we have a l .9 . handle on that one now. 10 Other models which have an effect -- 11 DR. SCHROCK: That's a little bit of a-mystery for 12 me to understand why break flow is dependent upon whether l 13 you're dealing with small break or large break. The i-14 phenomena in itself is at it is. It may take you into o 15 different territories, but -- l 16 MR. BAJOREK: I think that's the concern, is that 17 our upstream conditions can be much different for small 18 break than in large break, and we've had to make j 19 modifications to account for that. l 20 DR. SCHROCK: Well, this brief description that i i 21 was handed to us so far indicates that, through the j 22 blow-down, what you have is sub-cooled blow-down. That p 23 seems a little surprising at first reading, but maybe it's l 24 true. But certainly, sub-cooled blow-down is important in 25 the large-break LOCA, so that's not taking you into any new t JJRJ RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
31 1 territory. [ 2 So, it-is rather unclear, but this area is clearly l 3 one where we'need to see good supporting evidence, I think, 4 for the model changes that you've.made based on something 5-that'didn't exist previously. 6 MR. BAJOREK: Okay. 7 DR. KRESS: I was intrigued by your comment that 8 the nodalization size will change between large break and 9 small break. ) i 10 .To me, that's a major change in the code, believe 11 it or not, and do you have some sort of criteria for a magic 12 size break where you're going to make this change in ) -13 nodalization, or is it a different set of nodalizations for 14 each break size, or are we going'to hear more -- ( 15 MR. BAJOREK: It is going to be a different i 16 nodalization for small to intermediate. I E17 Now, our plans at this point would be to use this 18 code and nodalization fcr that range. 19 DR. KRESS: So, the whole range would have the B 20 same node, but it would be a different set of nodes than the 21 'large break. 22-MR. BAJOREK: That's correct. 23 DR. KRESS: There would be larger nodes for the 24' small break? 25 MR. BAJOREK: More detail in the loops, actually .[ ANN RILEY & ASSOCIATES, LTD. ' \\. Court Reporters 1. 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202)~ 842-0034 L r t nx--- ,m-,. me w, w- ~~r-n -u--e
y 32 1 smaller' nodes in the loops than would be done in the () 2-large-break.model. 3' We had some comments earlier about the 4 documentation,'and I think this is one place where we can 5
- apply a lesson learned from large-break.
6 I think we-made-a mistake in large-break by 7- -submitting the documentation piecemeal for review, then 8 realizing, after later-simulations, that we had to make 1 9 model changes, resubmit documentation, and that confused i i -10 everybody. 11 Our plans are to submit a four-volume set i 12 simultaneously, at the same time, and to make the submittal 13 complete and stand alone. l 14 The volume one, which will document the () 15 'small-break version of the code, will contain all of the 16 models and correlations that's used in the small-break 17-version, not just the deltas, the things that we added in 18 and modified between the large-break and the small-break. 19 That way, a reviewer will be able to pick up the 12 0 code and know everything that is there and not have to trace 21 through some other documentation to find these changes along 22 the way, t [ 23 Volume two will be the validation. 24 Volume three will be another area where PWR l 25' uncertainties, many of which we talked about in the 4 ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 l (202) 842-0034 i o
33 1 large-break dev lopment, they still apply in small-break. [JY 2 We'll include these in volume three so that a 3 reviewer will not have to search back through other 4 documentation to try to identify those uncertainties that' 5 are going to apply to our methodology. 6 .And volume four, then, will have the statistical 7 methodology. 8 MR. ZUBER: What is the time schedule? When do 9 you plan to have this delivered to NRC? 10 MR. KEMPER: This is Bob Kemper. The second 11 quarter of 1999. 12 DR. WALLIS: Steve, we have to move along if we're '13 going to stay on time. We're falling behind. g 14 MR. BAJOREK: Okay. I'll skip step six, because () 15. we're going'to the small-break version. 16' We've used our PIRT. Let me zip over a couple of 17 these. We've identified from the PIRT the small-break and 18 integral' test facilities that we should be simulating with 19 the code. 20 We've established a consistent set of nodalization 21 guidelines. So, even though it's different from 22 large-break, we've gone to the separate effects, the 23 integral effects, and the PWR and imposed these 24 nodalizations on all tests. 25 Determining the effect of reactor input O ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
34 1 parameters, significant is that all of these parameters that [^) 2 we identified as being important to large-break must also be \\ / 3 addressed as part of small-break. 4 However, we have a couple of additional ones now 5 for small-break -- main steam safety valve set-points, aux 6 feed injection. Indian Point and others have an asymmetric 7 pattern -- two receive aux feed, two don't. This may affect '8 the loop seal clearance time. So we need to incorporate 9 these. 10 Some of these will have different sensitivities. 11 The initial conditions will probably have virtually no 12 effect, although they had an important effect in large 13 break. So, we'll be accounting for those. 14 We've been doing sensitivity calculations. We've
- A l(
) 15 been doing these for the better part of a year, with x_/ 16 versions of the code which are approaching the frozen to 17 look at break location, orientation, loss of off-site power, 18 which determines essentially whether the operator can have 19 an effect or not, in addition to things like power shape, 20 MSSV, and other parameters that affect the boundary 21 conditions to the small-break accident. 22 The sensitivity uncertainty analysis -- one of our 23 goals is to try to make the statistical methodology not 24 quite as complex as it was for a large break. That will 25 make it easier to review and also make life easier for us. ('} ANN RILEY & ASSOCIATES, LTD. (_,,/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
_ - _ - - ~. -...-. 35 1 But we do anticipate the approach to be similar in (} 2 complexity to large-break LOCA. Using on the order of 30 to 3 40 COBRA / TRAC runs, we'll be propagating the uncertainties 4 in the global models, where we identify the bias. We'll.be 5 incorporating the fuel rod model uncertainties, develop 6 response surface to the combined uncertainties of the global .7 and the local models, and then use a Monte Carlo code to I 8 sample the'se response surface to get a statistical 95th 9 percentile PCT. l i 10 DR. WALLIS: Now, do you go back again if you find f l 11 that a particular uncertainty is more important than you 12 thought or needs to be investigated more? Do you go back 13 and do some more runs? l 14 MR. BAJOREK: I think the answer to that is yes. () 15 If we do our job right and if we find a certain parameter l 16 being more effective than others, it should show up more l 17 often in the run matrix, t 18 So, things like MSSV set-point or top-skewed power I t 19 shapes which tend to give you worse answer -- we'll look at i 20 those over their range more closely than initial RCS. 21 DR. WALLIS: That's encouraging. You're not to l 22 have a sort of fixed-run matrix which you just run but run l l 23 it and then look at it and say we'd be better off we did l-l
- 24 some more in this area.
l l 25 MR. BAJOREK: In general, no.
- ' ("'s ANN RILEY & ASSOCIATES, LTD.
tJ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 1
36 l 1 DR. WALLIS: That doesn't happen? / 'T 2 MR. BAJOREK: l V I don't think we'll be doing that. 3 I think you're referring to, I guess, what I would call the 4 German sampling approach, where you look at your matrix and 5 then add on as the parameters go. 6 We think that we're going to have a run matrix 7 that's fixed, and hopefully, in the development, we've 8 identified all of the parameters that are most -- that most 1 9 dominantly affect the peak cladding temperature. E10 DR. KRESS: Could you refresh my memory on the 11 95th percentile? Is that confidence level on the mean, or l 12 is that the 2-sigma-like uncertainty on -- of the overall 13 uncertainty? 14 MR. BAJOREK: I'm not sure I understand your ) 15 question. It's a 95th percentile with no confidence l 16 interval. 17 DR. KRESS: Of the total uncertainty. 18 MR. BAJOREK: The total uncertainty, that's 1 19 correct. 20 DR. KRESS: That was my question. 21 MR. BAJOREK: So, to conclude, we are going to 22 develop a best estimate small-break LOCA methodology. It 23 will be a CSAU style of approach. It will account for ( 24 propagation of model uncertainties, and it will account for 1 l 25 PWR operational uncertainties, things like power shape MSSV /T ANN RILEY & ASSOCIATES, LTD. (_,/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l l l
. ~. - . ~. -. - ~ -.... -.. - - _ - - -. _. - -.. 37 1 set-points. l ( 2 We are using COBRA / TRAC. We have made l 3 modifications to the code, and we're continuing to make
- 4
. those at this time. 5 We believe that, when we are completed and freeze 6 - this~ code, it will be applicable to those process ranked L 7 highly in the PIRT. I .8: We have tried to put-these in in such a way that 9 it will not perturb the large-break performance of the code, ] 10 and we think that, with additional development, it would be ) ~ ~ 11-possible to.use this' code for the entire spectrum of breaks, 12 from small: through intermediate to large. l 13 Thank you. 14 DR. WALLIS: Thank you very much. i 15
- MR. BOEENERT:
I show now we're going into closed i 16 session, so I need to ask everyone who doesn't have an 17 agreement with Westinghouse to be here to leave. 18 Transcriber, we'll go into a closed session 19 portion of the transcript. 20 [Whereupon, at 9:23 a.m., the meeting was resumed 21 in closed session.] '22 23-24-25 ANN RILEY & ASSOCIATES, LTD. i \\~. - Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 i Washington, D.C. 20036 (202) 842-0034 i - -sw++- w y-t 4 e +--- my -m m yr -,nmp7 --e1=
i 122 1 OPEN SESSION l U[ ) 2 [11:43 a.m.] 3 DR. SCHROCK: I have to be careful not to say 4 minor. This proprietary thing is laughable. Minor 5 parameter. I won't say that anymore. 6 DR. KRESS: You can't say that. 7 DR. SCHROCK: I can't say it. 8 DR. KRESS: It's the value, it's what those 9 consistent of that's propriety, not the term itself. 10 MR. LANDRY: Okay. My name is Ralph Landry. I'm 11 the NRR staff person responsible for this review, and what 12 I'd like to do is just give you a few brief comments 13 regarding the status and scope of the thermal hydraulic code 14 review, this one in particular and the code review in r-( 15 general. 16 This is the second time we have talked with 17 Westinghouse regarding the best estimate sma.11-break LOCA. 18 Westinghouse came in to us in July and gave us a briefer 19 presentation than we had today of what their intentions 20 were. 21 Westinghouse, accompanied by Consolidated Edison, 22 informed us that they are working with EPRI to put together 23 the best estimate small-break LOCA, and they plan on using 24 Indian Point 2, Unit 2, as the lead plant for this 25 submittal. [~ ) ANN RILEY & ASSOCIATES, LTD. \\/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
.. ~.. _. _ _. _ -. - -. _ _ _ _ _.... _ _ _ _ _. _ _ _ _ _. _ _ _. _. _ _. - _ -.. - -. _. _ = _ _,.. 123 1 At that time, they told us that.they had two 2 objectives, to prepare a best estimate small-break LOCA .3 . methodology that's applicable to Westinghouse three-and 4-four-loop PWRs and to do plant-specific application 5 calculations for Indian Point 2, same material which they 1 i 6 presented this morning'and informed the subcommittee about. 7-At that point, Westinghouse said that they planned l. L 8 on the documentation being in a stand-alone form. This was i l .9 a concern to the staff because of recent reviews which we've 10 done of codes. 11 We, along with the committee, were unhappy with 12 having to go from document to document, collecting materials 13 and tryingEto piece together the material that covers the 14 code property. () 15 So, we had some rather heart-to-heart discussions 16 with the vendor in July regarding our concerns about 17 documentation, and they informed us that, yes, they will 18-make it stand-alone, as they reiterated this morning. 19 So, we're still waiting to see what the 20 documentation looks like, but we have hopes for it. l L 21 Regarding the comment earlier about what will be 22 the criteria to determine what is good or how good is good j f 23 enough, this has been a concern not only in the U.S. but L within the international community, looking at codes, for at 24 i 25 least 15 years that I've been involved in it. ANN RILEY & ASSOCIATES, LTD. Court Reporters 4 i 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
124 1 It's never been adequately defined, and I know ('h 2 that, when the large-break assessment LOCA was reviewed, 3 part of that discussion was what are the acceptable 4 criteria. 5 At this point, the staff, in conjunction with the 6 staff in Research, are preparing a standard review plan for 7 code review. 8 That effort is underway, and there will be some 9 presentation made in discussion at the December subcommittee 10 meeting, and I would like to defer further comment on that 11 at this point, because that stuff is still in the 12 preliminary and developmental stage. 13 At this point, the staff is aware of three codes, 14 three thermal-hydraulic transient codes that are coming in (~h (v) 15 for our review. 16 The RETRAN-3D code is being presented by EPRI and 17 by the RETRAN users group. That documentation is in hand at 18 this point. We have reviewed the documentation, and we'll 19 be going through that with the subcommittee in December. 20 The code documentation has been provided to the i 21 subcommittee or to the ACRS staff. If you haven't received 22 that, you should be receiving it soon. 23 MR. BOEHNERT: They will get it soon. 24 MR. ZUBER: What is the split between users of 25 RETRAN, RELAP, and COBRA / TRAC? i [~} ANN RILEY & ASSOCIATES, LTD. (_/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l L
l 125 1 MR. LANDRY: The split? (} 2 MR. ZUBER: Yes, the numbers. I mean in the 3 industry. How big is the group? 4 MR. LANDRY: RETRAN-3D is a transiant code. At 5. this point, we are aware of approximately 24 utilities who 6 are supporting or involved in the RETRAN-3D submittal to the 7 staff. 8 That has not been tied to a lead plant, but it has 9 been submitted to the RETRAN users group under the auspices 10 of 24 utilities who have said they plan on using it. 11 We are aware that a version of RETRAN is under 12 review for application at Westinghouse, which is to replace 13 the LOFTRAN code. That material is being reviewed, also. 14 DR. KRESS: Transient means it has to have () 15 neutronic capability? 16 MR. LANDRY: Yes. Well, it has -- this version -- 17 I don't want to get into all the details of this version, 18 but this version will go beyond the point kinetics 19 neutronics. 20 DR. SCHROCK: Transient signifies reutronic code? 21 DR. KRESS: Normally that's a code word for that, 22 because it's to deal with the transient sequences. t 23 MR. LANDRY: It s a non-LOCA code. The other two 24 codes are LOCA codes. Siemens is planning on coming in with 25 an S-RELAPS best estimate large-break LOCA, and as we've [] ANN RILEY & ASSOCIATES, LTD. \\m / Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
126 1 been talking about this morning, Westinghouse ie ccming in n [%.]) with the W-COBRA / TRAC best estimate small-break LOCA. 2 3 MR. ZUBER: And the utilities? How many utilities 4 are using RELAP versus -- 5 MR. LANDRY: Well, I don't know how many utilities 6 in total, but this is Siemens who will be coming in with the 7 S-RELAP, the Siemens RELAP5 large-break LOCA. That will be i 8 tied to a CP&L H.B. Robinson reload as the lead plant. 9 As we've talked about this morning, the 10 W-COBRA / TRAC is tied to the Consolidated Edison Indian Point i 11 Unit 2 reload as the lead plant application. 12 DR. SCHROCK: I missed what S -- j 13 MR. LANDRY: Siemens. 14 DR. SCHROCK: Siemens. Okay. (,7 15 MR. LEVIN: Ralph, this is Alan Levin from the l i 16 staff. ) 17 Siemens has about eight PWR licensees that use its 18 analytical. methodology, and they -- all of their codes are 19 currently based on RELAP4, and they're planning to upgrade 20 all of them to RELAP5, S-RELAP5, as the best estimate 21 large-break LOCA. 22 DR. SCHROCK: What have they used? I mean what 23 version of RELAP5 did they begin with? 24 MR. LANDRY: That will be this version, is the 25 RELAP5. We haven't seen it yet, so we don't know exactly. /~ ANN RILEY & ASSOCIATES, LTD. (s h/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
127 1 MR. LEVIN: I think it's RELAPS, Mod 2, but we (m) r 2 haven't seen it yet, so we're not entirely sure. 3 MR. LANDRY: Most of this is preliminary. The 4 only documentation that is in to date is the RETRAN-3D. All 5 this other information is preliminary, what we've been told 6 is coming. 7 So, we've put together an action plan for 8 reviewing these materials as they come in. 9 The approach that we intend to take is perform an 10 acceptance review. 11 If you remember back in the early to mid '70s, 12 when we were receiving so many license applications a month 13 at the AEC, then the NRC, that we couldn't handle them all, 14 we went to what we called a mini-review or an acceptance l 15 review in which we would review the documentation, the SAR, 15 determine if it was complete enough to allow it to be l 17 docketed, and then begin the formal review. 18 We have decided that what we want to do on these ( 19 codes is do an acceptance review of a similar nature, that l l 20 when the documentation comes in, we go through the i 21 documentation to determine do we feel that documentation is 22 adequate to perform us -- to allow us to perform a review? i 23 If it is, then we would accept for formal review. 24 Now, that acceptance review is planned to be a 25 four-week process. We'll spend four weeks going over the [ ANN RILEY & ASSOCIATES, LTD. \\- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
I 128 i 1 documentation to determine that is adequate, there is enough . p ). ( 2 material there, it is sufficient to perform a formal review. 3 This is not an approval of the code. This is 4 simply that, okay, you've described it well enough that we 5 feel we can now review this material. 6 If it's not, then we will reject the documentation 7 with a statement of what is inadequate, where does it need 8 to be strengthened to provide sufficient material for us to 1 9 do this review. 10 I'll get to a schedule in a minute. 1 11 DR. WALLIS: There's another question. One is how l 12 good is the documentation? How good does it look? The 13 other question is how well are you prepared to do a review? 14 Do you need to do your own research on some of these issues Gk,) l 15 in order to be knowledgeable enough to make a decision? 16 MR. LANDRY: We are using three or four members of i 17 the staff who are experienced code people to perform these 18 reviews. l l 19 We have a large number of people on the staff that l 20 we will call in to assist, but we are planning at this point 21 to use two thermal-hydraulicists, a person who is reactor i 22 physics. 23 DR. WALLIS: It seems to me you need to do it i j 24 earlier, because if Westinghouse is struggling with modeling 25 some physics in the code and then says we think we've done .[ ANN RILEY & ASSOCIATES, LTD. ' \\s-Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
129 .1 it, the best person to assess that is probably someone who ry (v, 2 has him or herself struggled to model the same physics and 3 understands what the tricks and difficulties are. 4 MR. LANDRY: All the people on the staff involved 5 in this have been involved in code development work. 6 DR. WALLIS: With these specific phenomena which 7 are now particularly important? 8 MR. LANDRY: We hope that we have it thoroughly 9 covered. There are other people in the staff -- we won't be 10 using only three or four people. Those are the core people. 11 We will be relying on some of the people in Research and 12 other people on the staff who have experience that we can 13 call in. 14 If we find that there's a particular model we y,) 15 don't have enough knowledge of, we will go to somebody, 16 someone in Research, like a Joe Kelly, who may have the i specific knowledge that we need. 17 18 DR. KRESS: I presume part of your review will be 19 to look at models, new models and even old ones, and try to 20 assess whether or not there's enough database to actually t 21 establish an uncertainty. 22 MR. LANDRY: That's correct. l 23 DR. KRESS: And if you can make -- do you have 24 some sort of feeling or criteria for how you'll make that i 25 judgement, say this one doesn't have enough, you have to go
- T ANN RILEY & ASSOCIATES, LTD.
A-) Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
I 130 1 back and get us some data, or there's enough data exists ,m '(v) 2 here, we can accept your uncertainty distribution there and 1 3 allow it to be used as part of the best estimate 4 methodology? 5 Do you have some feel for how you're going to make 6 that judgement, or is that just maybe an expert opinion 7 based on your staff? 8 MR. LANDRY: I think, Tom, it will be a 9 combination. We will be relying on some of the staff 10 statisticians to assist us in determining is there adequate 11 statistical base. 12 DR. KRESS: Well, that's partly what I had in 13 mind. 14 MR. LANDRY: We do plan, within this group, to f3() 15 pull in two of the staff statisticians to assist us, also. 16 The emphasis of the review on these codes, because 17 of the limited time and staff availability, of course, is 18 going to be on the new material. 19 We don't intend to go back and re-review all of 20 W-COBRA / TRAC. 21 What we want to do is put the emphasis on j l 22 reviewing the new material for the small-break LOCA best 23 estimate COBRA / TRAC and then touch on the old material 24 where, as we discussed this morning, if models are changed 25 or affected, we would have to look at those areas to make i i /~'T ANN RILEY & ASSOCIATES, LTD. kl Court Reporters s-1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
131 1 sure that they have not altered the results from the [' s_-)' 2 previous model or the approved model. { 3 DR. SCHROCK: Is that a subject that you've got 4 good agreement with Westinghouse on? Is it going to be easy 5 to identify so-called new material, or are you going to have 6 to wade through the whole thing to find a paragraph here and 7 a paragraph there? 8 MR. LANDRY: Part of that is going to be we have 9 to interact with Westinghouse when they are prepared to 10 submit the materials. 11 Now, what we've done on another application, the 12 one that is in, is we've had the applicant go through and 13 bar along the side of the pages the material that is new to 14 readily identify it. rm (,) 15 MR. ZUEER: This brings really a concern. I think 16 I mentioned it to you. To have a document, you can have a 17 good document, per se, one, or you can have four poor i 18 documents. Doesn't mean, if you get everything in one 19 package, it's going to be satisfactory. 20 So, one thing is to be really clearly written, but l i 21 the thing which I am really concerned is that here we have 1 22 some phenomena which we are really -- we are not important 23 in large-break LOCA, which we never address in this best 24 estimate space, and they will commit four reports in end of 25 the June, and then you find out -- you may find out, well, [O ANN RILEY & ASSOCIATES, LTD. (,. / Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
132 these experiments are not adequate or this was not really 1 i [ ~.l 2 validated correctly. %/ 3 This then puts a glitch in the process, and it may 4 be more advantageous if you could identify these new areas 5 and have that information in terms of either a report or 6 something where you can make a judgement early enough, 7 before end of June, when you get these four documents, 8 because at that time, it will be too late to make a change. 9 For example, if you find out that these jet 10 condensation experiments were inadequate, why do we have to I 11 pass a judgement in June if the infurcation they have now 12 can be shown it's adequate or not? -13 It's kind of what you put first, I mean which 14-problem you address first, and my suggestions would be look fs(,) 15 at the new material and try to get in advance information 16 before these four documents come in. 17 MR. LANDRY: That's one of the reasons why the 18 staff is encouraged that we have been interaccing back and 19 forth with these vendors and applicants at an early stage 20 before the material is submitted, so that we can express our l l 21 concerns to them and they can come lack and talk to us about l l 22 where they're going with the documentation and the l l 23 materials, so that we're interacting back and forth at an 24 early stage. 25 One of the things that we would like to do is,
- 'l ANN RILEY & ASSOCIATES, LTD.
(- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
.~. ._..~... ---,-.- -. - -... -. -. -. ~. - 133 1 'where'there has been prior peer review of correlations and r' ( 2 models, staff would like'to make use of some of that peer 3 review, and that, of course, means is it what we feel is a 4 good peer review, is it international, what type of review 5 has been done, so that we can shorten up out review. 6 This is.not-abrogating our entire responsibility 7 for the review to a peer review group, but simply has a peer 8 review of a-correlation or model been performed and can we 9-make use of some of the results in our-review? 10 MR. ZUBER: Let me ask you. Years ago, Research ~ 11 was conducting an international program, led to some 12 . reports, conclusions which models were recommended, which 13 models were adequate or inadequate. Is this available to 14 you, or can'you use it? 15 MR. LANDRY: That is part'of what we would also 16 consider as part of a prior peer review -- 17 MR. ZUBER: Okay. 18 MR. LANDRY: -- materials of that nature which say i 19 that.they have looked at a model or a correlation and this 20 group feels it's adequate or inadequate, can we make use of 21 that material? 22 DR. WALLIS: Now, you will not yourselves run 23 COBRA / TRAC. 24 MR. LANDRY: That I was going to get in just a 25 minute. I l l-( Y ANN RILEY & ASSOCIATES, LTD. , \\- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
134 1 DR. WALLIS: Are you going to talk about that? ,a ( ) 2 MR. LANDRY: Yes. I don't have it on here. This x). 3 is just part of how we're going to do the review. 4 This morning you heard that what Westinghouse is 5 coming in with is a CSAU methodology. We also want to keep 6 the door open, if someone else wants to come in with what 7 they call a best estimate or realistic modeling which is not 8 full CSAU, we want to keep our minds open to what are they 9 proposing in the way of support of their modeling so that 10 we're not locking ourselves in, saying CSAU is the only 11 methodology we'll accept. 12 DR. SCHROCK: It seems to me it was described as 13 an acceptable methodology, and people have the right to 14 create other acceptable methodologies, but that's different /h < i,/ 15 than this fuzzing of what one means about best estimate 16 versus evaluation model, and your last item there gets back 17 to what Dr. Wallis commented on at the beginning of the 18 meeting, and that is that there seems to be some confusion 19 about what is meant by best estimate. 20 I mean it's either best estimate or it isn't best 21 estimate, but we went through a lot of discussion on that 22 when we were reviewing the large-break methodology, and I j 23 Westinghouse has, in various ways, brought in conservatism 24 into this -- what they call a best estimate apyjroach. I 25 don't think you should be accepting that as best estimate
- /
il\\ -} ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
_ ~ ~ - -. _.. _. - -.. - ... -. - _ ~..- 135 1 approach. It isn't best estimate. I don't know where it 2 fits under.the rules that you have to work under, but it's 3 not best estimate. 4-MR. LANDRY: Well, we're trying to be open to what j 5 Jus proposed to us and determine is the modeling acceptable G that's being presented? 7 DR. SCHROCK: Being open to how you evaluate best 8 estimate methodologies, other approaches than CSAU of 9 evaluating a best estimate methodology -- that's one thing, 10 but being open to accepting approaches that are midway i 11 between best estimate methodology and EM methodology, that's f 12 a different thing, and I think you're making a mistake to 13 get into that area. 14 MR. LANDRY: Well, we haven't seen it yet, so (_, 15 we're just trying to keep an open mind at this point. 16 DR. SCHROCK: That's just my personal opinion. i l l 17' 'I'm sure it's not shared by all. i '18 DR. KRESS: Well, I think that goes to the point 19 of how you deal with the uncertainties. l l 20 The key element of the best estimate approach is 21 this 95 percentile, and if you're applying.conservatisms 22 that you can be sure that the 95 percentile you'll end up 23 with is, indeed, a bounding number, I think best estimate 24 could incorporate conservatism, because I think you're going l -25 .to end up with conservatisms anyway when you choose your I 4 ' ANN RILEY & ASSOCIATES, LTD. N-Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034 d. s v
-. ~. ~. . - - -.-...- - _.-. ~ 136 l' distributions that you end up with getting the final result 2 anyway. 3 S o,' I think it's inevitable you're going to end up 4' with conservatisms even in your best estimate approach. 'I .5 don't knowlhow you can avoid it, frankly. 6 MR. LANDRY: Okay. 7 Now to get to Dr. Wallis' comment, we have 8 informed each of the vendors, applicants that we require 9 that they. submit a copy or a running version of the code as 10 .part of their application. 11 Our intent-is to have a version of the' code 12 in-house so that--we can put it up on our work-stations and 13 we can do some of the what-if calculations also, that if we want to tweak a model and see how it responds, we can do so. 14 p ' (,). 15 We are insisting that-they provide to the staff 16 the source code, and if they have a binary that they would 17 like to give us for a particular work-station -- we have 18 virtually all work-stations in-house -- then we could take 19 that' binary and run it immediately or we will compile the .20 source' code for our available work-stations. 21-DR. KRESS: Will you also do some bench-marking 1: 22 . ith other codes like RELAP? w t l l 23 MR. LANDRY: Any calculations we do, if we do 24 bench-marking, it would be with our own in-house, not with 25 .another vendor's code. 'b-ANN RILEY & ASSOCIATES, LTD. \\- I Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 I Washington, D.C. 20036 (202) P42-0034
...m . _.,. _ -. _. _. _. _. _. _..._.=. _ _ _ _._ _. _ _ __ _.. 1 137 l' -DR. KRESS: Yes, that's what I meant. /% ( ) 2 -MR. LANDRY: If we decide to do bench-marking 3 calculations, confirmatory-type calculations, which we have t 4 always done, we would do them with our own NRC codes, not 5 .with another vendor's code. 6 DR. KRESS: That's-the new code you haven't 7 developed yet. 8 MR. LANDRY: Right. Code XYZ. 9 .DR. WALLIS: I think this is a very important i 10 development, that instead of just taking pictures from 11 . Westinghouse, which, in some way, inevitably, have to have 12 been selected, you now -- I hope you will really follow o 13 through -- will have the ability to run their code and to 14-try it out and see where perhaps it doesn't do as well, 15 because the only way you knew before was if Westinghouse ' 16 actually showed you those results. 17 MR. ZUBER: It's really a good idea except there p 18' is a kind of time delay for the users to become familiar . 19 with the code. R20 MR. LANDRY: Well,.yes, there.is. We do have some 21 . advantage there that, in the group, we have people who are 22. experienced with RELAP5, we have people who are very 23. experienced with TRAC,~so there is a familiarity with the
- 24-base codes for these submittals.
12 5 MR..ZUBER: That's a good idea. 1 ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 i (202) 842-0034
138 1 MR. LANDRY: So, we do have that familiarity 2 in-house; Research has familiarity with the codes. We have 3 staff here who may not be W-COBRA / TRAC experts, but they are 4 experienced TRAC runners. So, it's not a completely ncw 5 learning curve. 6 The schedule at this point is quite fluid. 7 The RETRAN code was supposed to have been in last 8 August, did not come in until September, so we've had to 9 modify the schedule and we haven't even discussed with them 10 this schedule at this point, but you can see -- and the 11 W-COBRA / TRAC, you can see, was planned on submittal in 12 February of '99, which was based on our July discussion with 13 Westinghouse and Consolidated Edison. 14 That, we found out this morning, has been slipped 15 to the second quarter, so that's at least a three-month 16 delay in there now. 17 But based on that information, you can project out 18 to what these break points would be on the Fchedule. 19 We plan on a four-week acceptance review for the 20 code, and prior to issuing the letter of acceptance, we want 21 to come to the subcommittee, discuss the documentation with 22 you, and get your feedback before we say yes or no. 23 Now, on RETRAN, we've already issued a letter to 24 EPRI accepting the code for review. We had a time 25 commitment there, so we had to go ahead before we could get ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
139 1 on a schedule with the subcommittee. (m) 2 DR. WALLIS: This is just accepting the code for v 3 review. 4 MR. LANDRY: Only accepting the documentation for 5 the formal review. 6 DR. WALLIS: Well, is that the time when they 7 submit the running version of the code for you to experiment 8 with? 9 MR. LANDRY: We're being flexible on the submittal 10 of the date so that, after we accept the documentation, 11 then, if they want to submit the running version of the 12 code, we'll accept it at that point. 13 During this four-week period, we don't anticipate 14 doing any running of the code. So, we don't need the code IO (,) 15 at that point. 16 After we have issued an acceptance letter, we will 17 then start preparing our requests fcr additional 18 information. 19 DR. WALLIS: Excuse me. One way to make sure that 20 you get this running version of the code is to tie it to 21 this acceptance letter. 22 MR. LANDRY: So far we've had no difficulty or 23 resistance. l 24 DR. WALLIS: It's still nice to have the running 25 version of the code and accept that, too. I think that (8'-) ANN RILEY & ASSOCIATES, LTD. g Court Reporters ) 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
j 140 1 should be part of your acceptance, because you might find () 2' that it's more difficult to run than you thought. 3-MR, LANDRY: That's a good point. We'll discuss 4 that with the. applicants. 5 Prior to issuing the RAIs, we would like to come 6 to the committee a short time before the RAIs go.out to .7 express our concerns with the review and to determine if you
- 8 have any concerns that we haven't covered so that when we 9
issue RAIs on these codes, we issue one set. 10 We don't want to go into the five-to-10-year 11 review mode where we're going back and forth and back and 12 forth. We would like to issue one set of requests for 13 additional information, so we'd like to get the feedback 14 from the committee immediately before we issued those 15 requests. 16 MR. ZUBER: My experience with this RAI was that 17-the answer from Westinghouse was most often very inadequate, 18 and you have really to enforce it that you get adequate 19 answers, because some of it goes back and forth, just making 20 more and more paperwork without resolving the issue. So, 21 there should be some response in the quality of the answers 22 they give to you. 23 MR. LANDRY: Our whole schedule is based on the 24 presupposition that the results coming back will be 25 adequate. If they're not, I think Art might have something O ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
l 141 1 to say. He might be unhappy, also, if we're not happy. 2' So, each of these codes'that's tied to a utility, 3 to a licensing action,-I would expect that ihose people f 4 would keep after their vendor, also. 5 MR. BOEHNERT: Ralph, I was just wondering. l 6 S-RALAP5 -- that February '99 date -- is that still 7. reliable,_or you don't know? 8 MR. LANDRY: We haven't been told anything t 9~ different. 10-MR. BOEHNERT: Okay. 11 MR. LANDRY: Nobody has told us -- well, this-12 morning, we found out different on the W-COBRA / TRAC. We 13 haven't been told different on the Siemens submittal yet. 14 So, that's all we can deal with. ~15 That may be soft, also. We just don't know yet, 16 formally. 17 MR., LEVIN: This is Alan Levin again. 18 'Siemens may be slipping a little bit, a couple of ' 19 months. As Ralph said, it's still very, very fluid. ' 20 Siemens.has got a lot going on aside from just that, getting . 21 code documentation up to par. 22 DR. WALLIS: Just make sure there's no year 2000 23 problem. You submit a change in the code and the code -24 ' thinks you submitted it 100 years ago or 1,000 years ago. t (' - 25 MR. ZUBER: I thought that they wanted to finish i ANN RILEY & ASSOCIATES, LTD. Court Reporters 4 y 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
.~ .~.-. -... _. ~ _.. l l 142 1 everything by December. l } 2-MR. LEVIN: I'd also like to make a comment in 3-response to Dr. Zuber's remark on RAIs. 4 The problem of getting inadequate responses on 5, RAIs, in general, in reviews, has been recognized for a long l t 6 . time, and there are~NRR-wide things going on as'part of our 7 overall trying to do things better, faster, smarter, that s 8 kind of thing, to improve the process of issuing RAIs and l 9 getting adequate responses the first time around, things 10-that we did successfully on the AP600, like sitting down on l I 11 the telephone and discussing them with Westinghouse before i 12 they went out, so they knew what was coming, they could ask i 13 any questions that they had ahead of time, and reach an 14 agreement as to what the scope of the response needed to be i l 15 that we would find to be acceptable, and those are the kinds l-16 of things that I would anticipate we would be doing as part ) 17 of these reviews to try to improve the efficiencies. 18 MR. ZUBER: Good. Good. 19 MR. LANDRY: You'll also notice that, after the 20 responses are in, we will allow a certain period of time to 21 review the responses, and we will draft our SER. Before we '22 issue the SER on each of these codes, we also would like to 23 come back to the committee and discuss the SER and our l 24 position on the acceptability of the code before we go on 25 the street with an SER. i i i ) ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
143 1 So, we've added'in these schedules, break points ]} 2 to discuss with the committee, the acceptability of the 3 documentation, the concerns, technical concerns with the t 4' code, the acceptability of all the responses in preparation l 5 for the safety evaluation report. 6 What we are trying to do is work closely with the 7 committee to keep you informed all along the way of what the 8 code looks'like and where the staff is going with the review 9 of the code, so that we don't come all the way up to an SER, 10 then start all over with-the committee, that we can work 11-hand in hand throughout these reviews. [ 12 DR. KRESS: Which of those ACRS lines would you i 13 envision would be the full ACRS versus the subcommittee? ~ i h 14 MR. LANDRY: I would guess the SER would probably l( 15 be the most important one. 16 DR. KRESS: Because you have both the subcommittee ..Y 17 and a full committee. l 18 MR. LANDRY: I just put ACRS but really l 19 envisioning the subcommittee. 20 DR. KRESS: I was thinking we might have that also f 21 for the RAIs, too, but I don't know. 22 MR. BOEHNERT: It's really our call Tom. I mean 23 we can make that decision, decide what we want -o do, how we 24 want to handle it'. l 25 MR. LANDRY: That is up to the subcommittee, the t 5 /~h ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
_. ~ __..___.-.-.m 144 1 committee, how you'd like to break this out, but we're just t( ) 2 trying to.make sure that.we keep in close contact with the 3 ACRS or the subcommittee throughout this review. 4 DR. WALLIS: Now, the SERs tend to be very bland 5 documents. We simply say we have reviewed this and 6 everything is'okay, and the SER isn't the place to go into 7 the history and say why do you think it's okay, perhaps, but 8 we need to know why you think it's okay. 9 So, we need to see sort of the trail that led you 10 to decide that it was okay. Rather than just waiting until 11 we see an~SER and saying, well, this doesn't tell us why it 12 was okay, we hope that you'd tell us all the way along the 13-line, so that when we get to the SER, we thoroughly 14 understand why it was okay, i 15 MR. LANDRY: Well, we feel that this process will i 16 help with that,.so that when we are ready for RAIs, you'll 17 be plugged into the RAIs, you'll see what our concerns are. 18 When we have the draft SER, we'll be back to discuss it with 19 you, so that you'll hear why we find the responses 20 acceptable. l 21 DR. WALLIS: You've done very well. You've 22 finished just on time. I think you have finished off. 23-MR. LANDRY: That's having two clocks. 24 DR. WALLIS: But you choose the right clock. It's 25 like the right coefficient in the code, you know. t !( ANN RILEY & ASSOCIATES, LTD. 4 Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034 i
. - ~,. -. ~. -. _. - i 145 .1 MR. LANDRY: The 95 percentile on the clocks? 2 ~ 'DR. WALLIS: Now we have'a subcommittee caucus f 3 -scheduled. Presumably, the purpose there is to hear from 4 members of the subcommittee.and_ consultants on how they. j 5- ' respond to what we've heard this morning. 6 I would like to ask the consultants, do you have 7 anything additional that you would like to add or that the i 8 presenters today should go home and remember? 9 DR. SCHROCK:
- Well, I. guess what I heard about the 10 modifications in the code give me the uneasy feeling that
-11 it's too superficial and that the result that arrives for 12. NRR review isn't going to be fully acceptable, and that's 13 mainly in such areas as thoroughly evaluating what model '14-changes are available for the fix rather than seizing on ) 15 something-based on a very limited review of those things, 16 and then, in terms of the database Lhat's used for '17 validation of the models that are finally selected, it's 18 clear from what we've seen today that there is more database ~ 19 available on some of these areas than has been cited, and 20 so,.I don't think it's going to turn out to be adequate on 21 that basis. 22 I think the schedule is very optimistic with the 23 three-month delay even to imagine the this new code is going 24 to be adequately reviewed both by NRR and by ACRS by next 25 summer, I think is pretty optimistic, but certainly wouldn't ANN RILEY & ASSOCIATES, LTD. I Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 ~ (202) 842-0034
146 1 want to argue it shouldn't be tried. ~ ' [ ')'\\ 2 But I don't have other criticisms of it. I think L 3 seeking the best estimate methodology for small-break is 4 long overdue, and I'm glad to see it coming. 5 I'm just a little disappointed it isn't being done 6 with a bit more resources behind it and with more 7 determination to make it something that's really going to be 8 lasting. 9 DR. WALLIS: Thank you. i 10 Novak? 11 MR. ZUBER: First, I would.like to commend the 12 industry in the event of deciding to go with the best 13 estimate approach for small-break LOCA. I think it was wise i 14 and overdue, and I was really pleased to hear all this (O 15 information from Mr. Ginsberg about the benefits of this f 1 16 approach. I think this was very informative, and thank you. ] i 17 And I found this type of rationale and approach efficient, i 18 and hopefully, they will pursue it, although I doubt they 19 will meet the schedule. 20 Now, I have a comment, and I think here I second 21 some of the comments that Virgil made. l 22 I doubt that you can really put his together by 23 the end of next June. What I'm really concerned is that -- l 24 my concern was always poor documentation. This is a track l 25 record which my experience of 10 years with Westinghouse was e O ANN RILEY & ASSOCIATES, LTD. \\/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
147 I 1 confirmed. 1 () 2 I don't think by putting four documents together 3 and giving them in one package will address this question. 4 This takes a managerial decision within Westinghouse to 5 produce good technical reports, j 6 I think, no. matter how you package it, whether l l 7 early or. late, it takes technical management to decide that 8 I want to have a good document and enforce it,.and I think I L I l 9 this is the only way we shall see good documents, and maybe 1 l ~ 10 industry can then provide this incentive to Westinghouse, j i l 11 I would also suggest that there are some new 12 phases in this small break which were not addressed in large J 13 break and in the PIRT documents, and an example was.this jet i 14 condensation, j 4 15 I an not concerned about the' experiments for which 16 we have enough'information and documentation, like i 17 small-break LOCA, we are running INL and so on. 18 What we need is really good reports, maybe advance 19 report, before the end of June to give us information on 20 this new important phenomena which we are really not 21 assessing, used to validate that code before, and I would 22 ask Paul or something if they can procure such documents on 23 some of these important phenomena which were not addressed I L 24 in previous studies to give us some information before June, i. l 25 so at least we see what database is being used by the i ANN RILEY & ASSOCIATES, LTD. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 i Washington, D.C. 20036 (202) 842-0034
F 148 1 industry in this process. [~'T 2 If the database is adequate, fine. If it's not lV l 3 adequate, we could save some time to tell them, look, this i 4 is a shortcoming, you better address it in January and 5 March, not wait till June. 6 So, I think there should be a balance to address 7 the important phenomena, then to_see what database we have. 8 Is it a new database? Is it available? For example, I 9 don't know these French experiments. 1 10 If you want to use this to validate a code, you 11 can provide this report ahead of time, so we can assess it 12 and you can save some delay or headaches in June. 13 But I have some additional -- another comment is 14 how you determine the most limiting case. I think you O I j really have to document and make a good case. Just by using 15 16 a code which is not adequate to- ,'rticular phenomena, w 2. 17 you cannot decide it's important cr not 18 So, I think this should be also addressed early in 19 time, so you put this problem to rest, so it won't come 20 later on. 21 And I may have some other comments and put them in 22 the letter to you, Graham. 23 DR. WALLIS: Thank you. 24 Yes, I'd like to receive letters from both 25 consultants. , [D ANN RILEY & ASSOCIATES, LTD. l \\_s/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
149 1 Tom. [} 2 DR. KRESS: Yes. Thank you. \\s / 3 I would like to second the last comment, at least 4 of Novak's, that I think you really do need to have strong 5 justification for what is your limiting case, and you need 6 that up front. 7 I think this is a good start. I like what I've 8 seen. It looks like you guys have got the right plan, and I 9 don't know whether the schedule is optimistic or not, but 10 you might as well have an optimistic schedule to s. -t with. 11 I like seeing this TSU used, the PIRT up tront, 12 and I think we may -- like Novak, we may need to see more of 13 the database that supports the new models as soon as we can 14 and what the new models actually are in their physics, so ' O t ,7 that's important we see that early. 15 16 I think the issue of what the appropriate 17 nodalization is just has never been addressed very well in 18 my mind, and I'd like to see more justification on why this 19 particular nodalization chosen for the small-break LOCA is 20 the right one and that we know how it affects the final 21 answer in terms of uncertainty. I think that's pretty l l 22 important. 23 I guess there is a question that's really outside 24 or peripheral to this that I might address to the staff 25 rather than to this exercise, and that's -- we are dealing [(-] ANN RILEY & ASSOCIATES, LTD. Court Reporters / 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
150 1- -here in deterministic space, where-we have'the design basis ) and we're trying'to figure out what the actual margins are 2 3 that we have to' deal with, margins in things like peak clad 4-temperature and other things. 5-I've never.really seen anything that-tells me why 6' those margins are.there, what is.Eul appropriate margin to 7-have, how much of it can we eat away into, and you know, 8 this is what, ultimately, we're going to be dealing with. 9 The use of the code is to do that, and eventually, we're -10 going to face that issue, and I'd like -- maybe the staff 11: could give that some thought. 12' Just what is an acceptable margin we'll allow it 13 to eat into? 14 That has to do when you get actual applications () 15 themselves. 16 So, with that, I think it's a good start, and'I am 17 anxious to continue the review, and I think we're on the 18 right track. I particularly like the staff's proposal for 19 how they're going to approach the review. I think that's a 20 'really nice way to do it. 21 DR. WALLIS: Anymore remarks from Westinghouse or 22 the staff? 23 MR. KEMPER: Well, I'd like to thank everyone for 24 their attendance today, and I think, from our perspective, 25 we've -- this has accomplished what I was looking for in h ANN RILEY & ASSOCIATES, LTD. A/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
151 1' getting some early comments and feedback. {) 2 In particular, we'll strive to make available the '3 COSI report and'other things that comprise the database that 4 you may not have seen before at an early date to help l 5! facilitate the process. l l [. 6 .DR. WALLIS: Well, it's clear that we have a nice 7 beginning here. We're actually talking about things that 8 are important. l 9 You've got a very good impression, I think, of the l 10 kinds of questions that we're going to be asking in the 11 future, and I'd repeat what you've heard so much today, that i t i i 12 we need some vary good documentation, needs to be in good e l 13 shape early, so it doesn't'have to be fixed up over a period L 14 of time. I - 15 We need to be sure that you have a good database, 16 because all these theories have to be compared against some 7 realities,.and you need to justify why you've picked a 18-particular choice of model to fix up parts of the code which L 19 you think are not adequate. 20 I think we saw that today, that you can't just say [ 21 this is the model, that you really need to look and say here 22 are these various models and this is why we've picked this j 23 particular one for this particular purpose. L 24-So, we didn't get today into the validation of a 25 these, but that's going to be very important down the road. i 3 ANN RILEY & ASSOCIATES, LTD. 1 _/ Court Reporters c. 1025' Connecticut Avenue, NW, Suite 1014 Washington,'D.C. 20036 (202) 842-0034 l:
- - -. ~. . - ~. - i .152 i 1-Without validation, none of these computer predictions [ /.\\- ~ 2 really go very far. f 's-3 So, I feel you have a great deal of work to do, 4-and we're all looking forward to seeing it when it comes. 5 With that, I will go by the clock which is fastest 6 -and adjourn the. session. 7 Thank you very much. i 8 '[Whereupon, at 12:28 p.m., the meeting was 9 concluded.] f i 10-11 12 13 ( 14 (/ 15 16 l 17 i 4 '18 19 20 21 22 23-l 24 l 25 l I ANN RILEY & ASSOCIATES, LTD. Court Reporters ~ 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 1 i.
REPORTER'S CERTIF"CATE This is to certify that the' attached proceedings before the United States Nuclear Regulatory Commission in the matter of: o NAME OF PROCEEDING: ACRS MEETING RE THERMAL-HYDRAULIC PHENOMENA DOCKET NUMBER: PLACE OF PROCEEDING: Rockville, MD () were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and thereafter reduced to typewriting by me or under the direction of the court reporting company, and that the transcript is a true and accurate record of the foregoing proceedings. \\ w on Hundley Official Reporter Ann Riley & Associates, Ltd. O
~ i IN i~RODUCTORY STATEMENT BY THE CHAIRMAN OF THE i THERMAL-HYDRAULIC PHENOMENA SUBCOMMITTEE ] 11545 ROCKVILLE PIKE, ROOM T-2B3 ROCKVILLE, MARYl.AND l l NOVEMBER 19,1998 j j i-l The meeting will now come to order. This is a meeting of the ACRS Subcommittee on Thermal Hydraulic Phenomena. I l am Graham Wallis, Chairman of the Subcommittee. i l The other ACRS Member in attendance is Thomas Kress. The ACRS Consultants in attendance are Virgil Schrock and Novak Zuber. I j The purpose of this meeting is for the Subcommittee to review the Westinghouse methodology for best-estimate small-break loss of coolant accident analysis, using the I WCOBRA/ TRAC code. The Subcommittee will gather information, analyze relevant issues j l and facts, and formulate proposed positions and actions as appropriate, for deliberation by the full Committee. !O Paul Boehnert is the Cognizant ACRS Staff Engineer for this meeting. ). Portions of this meeting will be closed to the public to discuss Westinghouse Proprietary i_ information. The rules for participation in today's meeting have been announced as part of the notice i of this meeting previously published in the Federal Register on November 4,1998. 1 A transcript of the meeting is being kept. It is requested that the speakers first identify 3 themselves and speak with sufficient clarity and volume so that they can be readily heard. i J We have receised no written comments or requests fnr time to make oral statements from j members of the public.
- r (Chairman's Comments-if any)
We will proceed with the meeting and I call upon Mr. Robert Kemoer of the Westinohouse Electric Comoany to begin. [ DOCUMENT NAME: G:BoEHNERTVNTR11-19.PB) O l
l i lr 1 l' r [( . ACRS THERMAL HYDRAULIC PHENOMENA SUBCOMMITTEE MEETING l WESTINGHOUSE BEST-ESTIMATE SMALL-BREAK LOCA ANALYSIS l' NOVEMBER 19,1998 I l ROCKVILLE, MARYLAND l PRESENTATION SCHEDULE l l CONTACT: P. Boehnert (301/415-8065) Subiect Speaker Time l l
- l. Introduction G. Wallis 8:30 - 8:40 am
- 11. Westinghouse Presentations
- Introduction (Open)
R. Kemper/ 8:40 - 8:55 am G. Ginsberg l l l . Overview of BE-SB LOCA Methodology S. Bajorek 8:55 - 9:15 am lO ^99<o cn - Coa i teacy with CSAU and General Design Criteria (Open) L SB LOCA Scenario and Phenomena D. Shimeck 9:15 - 9:45 am identification and Ranking Table (PIRT) (Closed) l BREAK 9:45 - 10:00 am WCOBRA/ TRAC Code: Version for.BE-K. Ohkawa 10:00 - 11:00 am SB LOCA (Closed) f (a) Summary of Code improvements i (b) Separate Effects Tests for Validation l (c) Integral Effects Tests for Validation i H = Preliminary BE-SB LOCA SimuWons R. Kemper 11:00 - 11:30 am and Statistical Methodology Appi.cach for Indian Point 2 (Closed) + Summary Remarks 11:30 - 11:45 am i LO 4 - ~
Presentation Schedule 2 ib. November 19,1998 111. NRC Staff Presentation (Open) R. Landry 11:45 - 12:15 pm
- Comments on Scope, Status and
. Schedule of Staff Review IV. Subcommittee Caucus (Open) 12:15 - 12:30 pm
- Comments on Meeting Presentations i
Follow-on Actions V. Adjourn 12:30 pm l l l O O
s O ( STAFF COMMENTS: STATUS AND SCOPE T/H CODE REVIEW ACRS THERMAL / HYDRAULIC SUBCOMMITTEE NOVEMBER 19,1998 Ralph R. Landry Reactor Systems Branch Division of Systems Safety and Analysis 4
- O
2 O I I WESTINGHOUSE BE SBLOCA 1 l Initial Meeting - Ju;y 20,1998 1 l Indicated Joint Program with Con Ed, EPRI Lead Plant -Indian Point Unit 2 O bi" "* * ~ BE SBLOCA methodology applicable to Westinghouse 3 and 4 loop PWRs Indian Point Unit 2 plant specific application Documentation is to be Stand-Alone O
. _ ~. _ _ 2 -.4 3 O T/H CODE ACTION PLAN j Codes { RETRAN - 3D: Transient S-RELAP5: BE LBLOCA i WCOBRA/ TRAC: BE SBLOCA Anolications ~ RETRAN - 3D Utility user group S-RELAP5-O CP&L - H.B. Robinson reload WCOBRA/ TRAC Consolidated Edison - Indian Point Unit 2 reload i Review Anoroach Acceptance review - four weeks to assess documentation Review emphasis - new material Verify previously approved models in range of applicability Utilize prior peer review of models/ correlations Use subset of CSAU methodology O nservatism where applicable
l. j s. l. 's 4
- O
[ l t 1 MILESTONE 5 RETMN 30 5-RELAPS HC08M/IEC Submittal September 1998 February 1999 February 1999 kceptance Review October 1998 february 1999 February 1999 ACR5 December 1998 March 1999 March 1999 r kceptance Letter hvember1998 March 1999 March 1999 Os ACR5 March 1999 October 1999 November 1999 ulsisswd April 1999 November 1999 December 1999 ElResponses June 1999 April 2000 May2000 l DraftSER August 1999 June 2000 August 2000 ACR5 September 1999 July 2000 September 2000 final $ER October 1999 October 2000 December 2000 i i !,O i (.- i
i io 1 Westinghouse / Con Edison / EPRI 4 1 Best Estimate Small Break L,0CA Methodology Development Program TECHNICAL APPROACH Stephen M. Bajorek Assistant Professor Kansas State University Department of Mechanical and Nuclear Engineering (785) 532-2607 bajorek@mne.ksu.edu November 19,1998 O SB_ACRSJ!ov_19_1998 1 h
a I GENERAL OVERVIEW OF TECHNICAL APPROACH f j CODE SCALING, APPLICABILITY AND UNCERTAINTY (CSAU) i APPROACH TO BE FOLLOWED FOR BE-SBI,0CA 1 >> BASED ON REVIEW OF NUREG/CR-5249; CONCLUDED THAT THE CSAU APPROACH IS GENERICALLY APPLICABLE TO { MANY TYPES OF PROBLEMS ? j >> NRC HAS PROVIDED SOME GUIDANCE APPLICABLE TO BE-SBLOCA IN REG. GUIDE 1.157 l lO BE-SBLOCA " ROAD MAP" HAS BEEN DEVELOPED. j l i i f i i i I 4 O SB_ACRS_Nov_19_1998
A FIFMENT1 SPECIFY SCENARIO ] SELECT REQUIREMENTS FROZEN 4 AND CODE 3f CODE CAPABILITIES SELECT NPP 2 PROVIDE COMPLETE DOCUMENTATION 3g CODE MANUAL USER MANUAL IDENTIFY & RANK PROGRAMMER'S GUIDE 5 PHENOMENA (PIRT) 3 DEVELOPMENTAL ASSESSMENT MODEL AND CORRELATIONREPT. I t DETERMINE CODE APPLICABILITY 6 FTFMENT2 ESTABLISH ASSESSMENT ASSESSMENT 7 MATRIX AND RANGING OF PARAMETERS T DEFINE NPP NODALIZATION < f f COMPARE CALCULATIONS COMPARE CALCULATIONS VS. SETS USING NPP VS. IETS USING NPP SET 8 IET NODALIZATION ^ DATA DATA fj DOCUMENT DOCUMENT (_./ I I ODIN CHANG YES NO BIAS & DETERMINE CODE 9 UNCERTAINTY,4 AND EXPERIMENT ACCURACY UNCNRT INTY ^ 4-DETERMINE EFFECT OF SCALE 10 4--- DETERMINE EFFECT UNC INTY 4-OF REACTOR INPUT PARAMETERS II AND STATE T PERFORM NPP SENSITIVITY CALCULATIONS 12 r-- 7 13 I ADDITIONALI COMBINE BIASES 4- - - -{ MARGINIF l m AND UNCERTAINTIES ^ WARRANTED g FI FMENT 3 t ___; TOTAL UNCERTAINTY TO S CALCULATE SPECIFIC SCENARIO 14 UNC A AN LYSIS IN A SPECIFIC NPP Figure 3. Steps and Elements of CSAU Methodology
{ 4 O 1 a CSAU ELEMENT 1: REQUIREMENTS AND CODE CAPABILITIES l l STEP 1: SPECIFY SCENARIO >> SCENARIO ASSUMED IS A SMALL TO INTERMEDIATE SIZFsD LOCA DUE TO A PIPE BREAK IN THE PRIMARY COOLANT SYSTEM. >> OPERATOR ACTION IS ASSUMED TO OCCUR BASED ON PLANT SPECIFIC F. ops (EMERGENCY OPERATING PROCE-DURES). STEP 2: SELECT NPP >> INDIAN POINT UNIT 2 STEP 3: IDENTIFY AND RANK PHENOMENA (PIRT) >> PIRT FOR SHLOCA WAS DEVELOPED. O SB_ACRS_Nov_19_1998
l llO SMALL BREAK PIRT
SUMMARY
i l HIGHLY RANKED SMALL BREAK PIRT PROCESSES CAN BE SUM-MARIZED THE FOLLOWING : " TREATABLE UNCERTAINTY PARAMETERS": l l BREAK FLOW N\\KkM6 Lf Mi FUEL ROD MODEL HORIZONTAL FLOW REGIME I STEAM GENERATOR HYDRAULICS Q LOOP SEAL CLEARANCE PUMP PERFORMANCE CONDENSATION l !O i SB_ACRS_Nov_19_1998
) i O STEP 4: SELECT FROZEN CODE WCOBRA/ TRAC-MOD 7A, REV. I USED AS THE STARTING POINT >> MODIFICATIONS MADE TO MAKE WCOBRA/ TRAC APPLICA-BLE TO SMALL BREAK LOCA PROCESSES BASED ON CON-CLUSIONS FROM THFs SH PIRT AND INITIAL V&V EFFORTS >> THE SMALL BREAK CODE VERSION CAPTURES ALL RECENT REVISIONS TO THE LARGE BREAK VFsRSION PLUS NEW SMALL BREAK MODELS & CORRELATIONS. STEP 5: PROVIDE COMPLETE DOCUMENTNflON (WCAP-14936) O > VOLUME 1 = SMALL HREAK MODELS AND CORRELATIONS 1 VOLUME 2 = SMALL HREAK CODE VALIDATION VOLUME 3 = PWR UNCERTAINTIES AND SENSITIVITES I'OR SBLOCA > VOLUME 4 = BEST ESTIMNFE SHLOCA STATISTICAL METIIODOLOGY THE SUHMITTAL WILL HE COMPLETE & STAND-ALONE l O SB_ACRS_Nov_19_1998
O STEP 6: DETERMINE CODE APPLICABILITY WCOBRA/ TRAC-MOD 7A,REV.1 WAS REVIEWED AND CONSID-ERED NOT APPLICABLF, TO SMALL LOCA PROCESSES. NECESSARY TO ADD SEVFRAL NEW MODELS & FEATURES TO ACCOUNT FOR SMALL BREAK PROCESSES RANKED HIGH IN THE PIRT. SMALL BREAK VERSION: WCOBRA/ TRAC-SB O l I O SB_ACRS_Nov_19_1998
O CSAU ELEMENT 2: ASSESSMENT AND RANGING OF PARAMETERS STEP 7: ESTABLISH AN ASSESSMENT MATRIX CODE ASSESSMENT NEEDS WERE BASED ON SB PIRT GUIDANCE SEPARATE EFFECTS TESTS SELECTED TO DETERMINE BIAS AND UNCERTAINTY DISTRIBUTIONS FOR IMPORTANT SMALL BREAK MODELS O INTEGRAL EFFECTS TESTS SELECTED TO DEMONSTRATE OVERALL CODE PERFORMANCE AT VARIOUS SCALES AND TO PROVIDE A MEANS TO ASSESS COMPENSATING ERRORS O l SB_ACRS_Nov_19_1998
O STEP 8: Define Nodalization for NPP Calculations Compare Calculations for SET /IETs Using NPP Nodalization i > ESTABLISHED A CONSISTENT SET OF NODING GUIDELINES > USED FOR ALL SETS,IETs, AND THE PWR MODEL MODEL IS SIGNIFICANTLY MORE DETAILED IN THE LOOPS AND STEAM GENERATORS THAN LARGE BREAK MODEL O l 1 1 O SB_ACRS_Nov_19_1998
O ELEMENT 3: SENSITIVITY AND UNCERTAINTY ANALYSIS STEP 11: Determine Effect of. Reactor Input Parameters & State > ALL OF THE PLANT RELATED UNCERTAINTIES IDENTIFIED IN THE LARGE BREAK LOCA METHODOLOGY STILL APPLY TO THE SMALL LOCA: i >> POWER SHAPE >> PEAKING FACTOR >> SI TEMPERATURE >> ACCUMULATOR TEMPFsRATURE >> ACCUMULATOR INITIAL PRESSURE >> ACCUMULATOR WATER VOLUME >> RCS INITIAL PRESSURE >> INITIAL TAVE O ADDITIONAL SB RELATED PLANT UNCERTAINTIES ARE THOSE THAT AFFECT THE STEAM GENERATOR: >> MSSV SETPOINT >> AFW INJECTION ASYMMETRY l O SB_ACRS_Nov_19_1998
O STEP 12: Perform NPP Sensitivity Calculations > SENSTIVITY CALCULATIONS PERFORMED TO INVESTIGNFE: >> BREAK LOCNflON >> BREAK ORIENTATION >> LOSS OF OFF-SITE POWER AND OPERATOR ACTION >> POWER SHAPE & PEAKING FACTOR i >> MSSV SETPOINT & UNCERTAINTY >> PUMPED SI FLOW & TFMPERATURES O >> ACCUMULATOR PARAMETERS (T, P, V) >> STEAM GENERATOR (SGTP, AFW) >> WCOBRA/ TRAC GLOBAL MODELS O SB_ACRS_Nov_19_1998
ELEMENT 3: SENSITIVITY AND UNCERTAINTY ANALYSIS STEP 13: COMBINE HIASES AND UNCERTAINTY STEP 14: TOTAL UNCERTAINTY TO CALCULNTE SPECIFIC SCENARIO IN PWR APPROACH WILL BE SIMILAR IN COMPLEXITY TO LBLOCA. MODEL UNCERTAINTIES WILL BE PROPAGATFsD IN WCOBRA/ TRAC-SB SIMULATIONS OF (FULL-SCALE) PWR FUEL ROD (LOCAL) UNCERTAINTIES WILL BE ACCOUNTED FOR. O RESeONSE SURvACE(S) DEvEtOeED TO REeRESENr uNCER. TAINTIES i MONTE CARLO SAMPLING TO DETERMINE 95th PERCENTILE PCT O SB_ACRS_Nov_19_1998
i 4-i O CONCLUSIONS d i ~ (1) BEST ESTIMATE SMALL BREAK LOCA METHODOLOGY WILL FOLLOW A CSAU APPROACH. i >>> ACCOUNTS FOR PROPAGATION OF MODEL UNCERTAINTY l >>> ACCOUNTS FOR PWR OPERATIONAL UNCERTAINTIES \\ )i O (2) WCOBRA/ TRAC WILL BE USED AS THE CODE FOR BE-SBLOCA. MODIFICATIONS HAVE BEEN M ADE & ARE IN PROGRESS TO ACCOUNT FOR PROCESSES RANKED HIGH IN THE SB-PIRT. 4 I i i !O 2 SB_ACRS_Nov_19_1998
. ~. _ _. - _ _ _ V BEST ESTIMATE LoCA ANALYSIS PROJECT i O con EniSox/EeniewsSTixonouSE - 9x& pitum & MM!% Large Break 1989-1997 -k [ Small Break Initiated 1996 W i
- 1. Safety
- Best Estimate (Realistic) - Regulatory Guide 1.157: " Safety is served when decisions concerning the limits within which nuclear reactors are permitted to operate are based on realistic calculations." - Benefits: Better understanding of accident Concentrate on real safety issues Reduce Vessel Fluence Better Operation through more Operating Margins O "**"'** 'r 10 CFR 50.46: More margin to 2200 PCT limit Less reporting requirements ACRS Encouragement
- 3. Demonstrate Margins to Safety Criteria Using WCOBRA/ TRAC g g P187 P &
- 4. Increased Flexibility For Utility Operations 0 >a4M NN
- Training: More efficient training of operators / - Testing and Performance: Wider range of testing Wider surveillance bands and test intervals - Fuel Management: Longer Cycles and Increased burnup Reduced spent fuelinventory
- 5. Major Advance For The Industry
. Schedule of Future IP2 Plant Improvements: Fuel Management Improvements - 2000 Trip Reduction - ASAP Potential Power Uprate - 2002 Potential Life Extension - 2002
i-INDIAN POINT 2 i O . 4 Loop 1000 MWE WESTINGHOUSE PWR i . 3083.4 Mwth NSSS POWER ) . FULL POWER OPERATION IN 1974 l . STRETCH POWER IN 1990 . 24 MONTH FUEL CYCLES I ! O . NOW IN ITS 14TH FUEL CYCLE i i l ^ .l k a 4 i i s
GENERAL LARGE BREAK BENEFITS t . PARAMETER BENEFIT Uprated Power Level (3216 MWth) Increased Thermal Margin at High Temperature Future Uprate Higher Peaking Factors Ultra Low Leakage Fuel Mgt. Reduced Core Peripheral Power Reduced Vessel Fluence Longer Fuel Cycles Less Spent Fuel Discharge Axial Blankets Reduced SI, RHR Flows Increased Recirculation Flow For Improved ECCS Testing Demonstrated Margin for Potential Pump Degradation More Accurate SI Flow Balancing der Bands on Accumulator More flexible accumulator setpoints OFA/ VANTAGE +(ZIRLO) Fuel Management, High Burnup, Fuel At High Burnup Use of Zirlo for corrosion resistance Additional Diesel Delay Time Additional margin for valve response and loading sequence Full range of combined More accurate modeling of Containment and Accumulator accident Temperatures High Steam Generator Thermal Margin Tube Plugging Allow plugging without reanalysis / regulatory action O. Reduced RCS Loop Flow Thermal Margin Allows for degradation of pump flow
O O o l l I C. W. Jackson i S. D. Rupprecht Con Edison c'y Westinghouse i Senior Management Sponsor Senior Management Sponsor i i I i J. C. Bass S. Ira Manager, I,0CA Manager i Integrated Services Customer Projects i I I A. P. Ginsberg i S. M. Ilajorck I,. Agee Con. Edison 4 Westinghouse EPRI I Technical 1, cad Project I, cad Ifeat Tramfer,I evd Swell Hedets l West Orvidentation.Flandbig 14054 Analysh g I i Project Team L I).Shimeck K.Ohkawa N. Petkov D. Golden M. Y. Young R.R.Lauteam rwn u.d4 wFT Amstysis bisp Seal Oeer m w ure.*a se*are, rem < w rr c.= =s* s see al sva resens.eiosis tvR Calculate.no seas. Mdisoas. avn >: vans.iins, liest Estimate Sill,0CA Project Organir.ation / Imp _med/Imene/bajores/Inunker/simfiles/SDLOCA_Fign c_forOm_Fel
.c t O a rd Benefits From Best EstimateDall Breik LOCA R&D Project Value From Applicabic EIIcci on Operations Savings Estimated ossedsNs0 Date of Planned APP""*'d W Implementation Technical Specification, Evelmenon Operat.mg Procedure or j Analysis Limit Deeed NRC APPnna'/S**"at d I m ese Amendment Parameter (s) changed Old New/ Proposed licms Reduced / Avoided [ I Peaking Factors
- Less burnabic poisons with Save $26,tNN) on cost F,
1.7 1.8 Lower enrichment for 12 fuel to enrich per assembly Fq 2.5 2.7 assemblics $E,(NN)/ cycle t Purchase 4 less Assemblics Save $2 million on i fuel purchase f i Reduction of Required 800 gpm 400 gpm Less How required from the 2 Reduced surveillance r St Flow minimum Si pumps and maintenance [ Less pump strain (tested once Eliminate 1 major i per cycle) overhaul every 10 l [ Recirculation line may be years i increased in sire to expand $750,tMN)/cvent f test criteria Save I day / cycle in Si pump Ikiw balance at $2tX),0tMI/ day t i Reduce Pressurizer Low Reduced probability of Avoid safety system Pressure Setpoint: reactor trip or Si on a rapid actuations Trip 1928 psia INIKI psia couldown 0.75 SORS /ycar Safety Injection 1833 psia 18tK) psia $1,875/ year Reduce Required Aux 300 gpm 2tR)gpm increased aux feed availability 1 SORS,0.15 LERs Feedwater Flow Decreased test failures $4,750/ycar increase Time Bands 5 sec 20 sec Decreased failure of ISOR for Delay and Isolation surveillance tests and $2,500/ year of Main Feedwater ' associated maintenance j Page - 1 i
O Benefits From Best Estimate inli Bre k LOCA R&D Project Si lireak isolation Tests valves every 11 hick valve If ilow reduced so only 1 Si i SOR,2,500/ year MOVs 851A,B other outage open; tests pump needed in future, need Eliminate tests $xxx/4 eliminated for these valves is eliminated years increase Setpoint 1080 psia 1110 psia Eliminate inadvertent Si by 3 LERs (i SI) Pressure for MS safety 1095 psia 1125 psia reducing posibility as a result $6,750/ year relief valves 1110 psia 1140 psia of exceeding setpoint 1125 psia 1155 psia 1135 psia 1165 psia increase Margin on 2tM) F 4(R) F Allows broader operation Eliminate analyses to Peak Clad Temperature because margin on PCT is show will not exceed better defined PCT for operating Shows fuct failure is less problem likely Reduced FCU Reduced VC activity for fuel charcoal and HEPA failure event filter loading Show RCP seal failure 2tNR) F PCI' 1500 F PCT Seal leak in a failure is below Resolves an open is not a limiting cwnt small break limit NRC unresohrd Avoids implications on design safety issue for IP2's for a station blackout event case Power Uprate 3071 M Wth 321re M Wth liigher gmwer limit More generation Code Availability Conduct analyses in-house $lo,tMKl/ cycle in external consulting I References-1. Technical SpeciGcations - section 3.10.2, p. 3.10-1 i 2 Technical SpeciGcations 3. Technical SpeciGcations 4. FSAR 5. FSAR 6. FSAR Page - 2 f
d f4 O ADVISORY COMMITTEE ON REACTOR SAFEGUARDS THERMAL HYDRAULIC PHENOMENA SUBCOMMITTEE I Best Estimate Smoll Break LOCA Analysis Methodology for 3-Loop and 4-Loop Plants Robert M. Kemper Westinghouse Electric Company (412) 374-4579 Dr. Art Ginsberg Consolidated Edison Company of New York, Inc. (212) 460-4331 0
O INTRODUCTION WESTINGHOUSE IS COMMITTED TO THE DEVELOPMENT OF IMPROVED METHODOLOGIES FOR LOSS OF COOLANT ACCI-DENT (LOCA) ANALYSIS LONG TERM LOCA DEVELOPMENT PLAN HAS INCLUDED: >> "SECY" METHODOLOGY FOR 2-LOOP PWRS WITH UPI 1 >> HEST ESTIMATE LARGE BREAK LOCA FOR 3-LOOP AND 4 - LOOP PWRS (CSAU) >> HEST ESTIMATE LARGFs HREAK LOCA FOR 2-LOOP PWRS O wiTH UPI(CSAU) >> AP600 HEST ESTIMATE LARGE BREAK LOCA (CSAU) OUR INTENTION IS TO CONTINUE ADVANCED T/H CODE DEVEL-OPMENT TO MEET CUSTOMER AND REGULATORY NEEDS, AND TO UNDERSTAND NEW AND EVOLVING ISSUES 4 HEST ESTIMATE SMALL HREAK LOCA IS THE NEXT STEP O NRC_MTG_SB_BELOCA_ Ira
e ! O BEST ESTIMATE SMALL BREAK LOCA PROGRAM d PARTNERSHIP AMONG WESTINGHOUSE, CON EDISON, EPRI PURPOSE IS TO DEVELOP & LICENSE A BEST ESTIMATE SMALL BREAK LOCA METHODOLOGY THAT IS GENERICALLY APPLICABLE TO CONVENTIONAL THREE & FOUR LOOP PWRS l - AND, TO i j-PERFORM A PLANT SPECIFIC APPLICATION FOR INDIAN POINT UNIT 2 O O
.=
- o 1
i GOALS OF TODAY'S MEETING TO INTRODUCE THE BEST ESilMATE SMALL BREAK LOCA PROGRAM TO OVERVIEW THE W BEST ESTIMATE SMALL BREAK LOCA METHODOLOGY CSAU-BASED APPROACH 4 ^ SMALL BREAK LOCA PIRT TO SUMMARIZE SMALL BREAK-RELATED ADDITIONS TO,WCOBRA/ TRAC Mod 7A 4 MAJOR NEW MODELS l EXPERIMENTAL DATA FOR CODE VALIDATION TO OUTLINE THE PRELIMINARY INDIAN POINT 2 PLANT RESULTS AND THE METHODOLOGY TO OBTAIN THE 95th PERCENTILE PCT i TO OBTAIN COMMENTS FROM THE SUBCOMMITTEE MEMBERS 4 a t 4 O}}