ML20134J319
| ML20134J319 | |
| Person / Time | |
|---|---|
| Issue date: | 11/07/1996 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-2082, NUDOCS 9611150149 | |
| Download: ML20134J319 (337) | |
Text
. . . .
L Official Transcript of Preceedings NUCLEAR REGULATORY COMMISSION p CRST-2082
Title:
Advisory Committee on Reactor Safeguards 436th Meeting
~
TRO4 (ACRS)
RETURN ORIGINAL Docket Number: (not applicable) $s*rI}j[6" 415-7130 THANKS!
Location: Rockville iryland Date: Thursday, November 7,1996 m
11 9 961107 T-2082 pop Work Order No. NRC-904 Pages 1-219 D0004 0 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1323 Rhode Islaad Avenue, N.W.
A:RS5Ef3 Sele:ai-0":'e fe Oi:10 30T1 Tree
i i
i R]
d DISCLAIMER '
PUBLIC NOTICE BY THE !
UNITED STATES NUCLEAR REGUIATORY COMMISSION'S ADVISORY COMMITTEE.ON REACTOR SAFEGUARDS j I
NOVEMBER 7, 1996 The contents of this transcript of the proceedings of the United States Nuclear Regulatory ;
Commission's Advisory Committee on' Reactor Safeguards on j NOVEMBER 7, 1996, as reported herein, is a record of the I
discussions recorded at the meeting held on the above date.
This transcript has not been reviewed, corrected !
! l and edited and it may contain inaccuracies.
I i
O !
NEAL R. GROSS j COURT REPORTERS AND TRANSCRIBERS l 1323 RHoDE ISLAND AVENUE, NW I (202) 234-4433 WASHINGTON. D.C. 200C6 (202) 234-4433 J
_.____...._.____.___.._.._...._._--..___..__..___...___.m_. _.
1 j 1 UNITED STATES OF AMERICA l
-2 NUCLEAR REGULATORY COMMISSION !
O 3 +++++ 1 1
4 436TH MEETING ;
l 5 ADVISORY COMMITTEE ON REACTOR SAFEGUARD (ACRS) l l
1 6 +++++ i 7 THURSDAY ,
I 8 NOVEMBER 7, 1996
{
9 +++++
10 ROCKVILLE, MARYLAND 11 +++++
12 The Advisory Committee met at the Nuclear 13 Regulatory Commission, Two White Flint North, Room T2B3,
'14 11545 Rockville Pike, at 8:30 a.m., Thomas'S. Kress,.
15 Chairman, presiding.
16 COMMITTEE MEMBERS:
17 THOMAS S. KRESS, Chairman 18' ROBERT L. SEALE, Vice Chairman 19 GEORGE E. APOSTOLAKIS i
20 JOHN J. BARTON j l
21 IVAN CATTON 22 MARIO H. FONTANA 23 DON W. MILLER 24 DANA A. POWERS j 25 WILLIAM J. SHACK NEAL R. GROSS !
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 u-
. . . _ _ _ _ _ _ . . -- _ . . _ . . _ - . _ . . _ . . _ . . - . . - _ = _ _ . _ . _ . -
2 )
I 1 ACRS STAFF PRESENT:
2 JOHN T. LARKINS, Executive Director O 3 ROXANNE' SUMMERS, Technical Secretary 4 SAM DURAISWAMY 5- RICHARD P. SAVIO 6 PAUL BOEHNERT 7 NOEL DUDLEY 8 MADHAT M. EL-ZEFTAWY 9 MICHAEL MARKLEY 10 AMARJIT SINGH 11 ALSO PRESENT:
i 12 JOE HOPENFELD :
r l-13 JACK STROSNIDER ;
?O
- U 14 STEVE LONG i l
15 JOSEPH DONOGHUE l 16 BOB JONES l
17 RICHARD PEARSON l
l l- 18 CLIVE CALLAWAY l 19 PATRICK BARANOWSKY l
l 20- ERNIE ROSSI l-21 STEVE MAYO 22 DALE RASMUSON 23 RON SIMARD l 24 TOM ESSIG 25 JAMES O'BRIEN l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N.W. j (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433 l
. _ _ . . . _ . _ . . . . . _ _ _ . _ , _ . _ . , . _ . . _ . - . _ _ . _ _ . - . _ _ _ . _ , _ - _ . _ - - _ . . _ _ , _ . _ _ . . _ . _ . _ . _ ~ _
s 3
l' A-G-E-N-D-A
)
, i 2 Acenda Item Eggg '
O 3 1) Ooenina Remarks by the ACRS Chairman 5 1
4 .1'.1) Opening Statement 5 l I l J
5 1.2) Items of current interest 6 ,
[ 6 2) Proposed Rule on Steam Generator Intecrity 8 7- 2.1)' Remarks by the Acting Subcommittee 8 !
l
~
8' . Chairman f
l 9 2.2) Briefing by and discussions with 11 l
10 representatives of the NRC staff, NEI,. !
I 11 .and EPRI regarding the proposed rule-
)
L l 12 'on steam generator integrity and an l l
lO
[
13 14 3) associated regulatory guide Risk-Based Analysis of Reactor Ooeratina l
l' 15 Exoerience
! t 16 3 .' 1 ) Remarks by the Subcommittee Chairman 107 17 3.2) Briefing by and~ discussions with.. 108 i
L 18 representatives of the NRC staff.
L l
19: regarding the staff activities 20 associated with risk-based analysis of 21 reactor operating experience, accident t
l '
22 sequence precursor program, development i 23 of risk-based performance indicators, i
24 and related matters 25 r
l NEAL R. GROSS !
j COURT REPORTERS AND TRANSCRIBERS !
1323 RHODE ISLAND AVE., N.W. l l
! (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433 l l
_ - _ , , ,_ _ - . -l
- .- . - - ~ . - . . . . . . . . . - . . . - . . . . - . . . . . - . . - ._ . . . . _ . . . . . .
l 4 l
1 A-G-E-N-D-A (CONT.)
2 Acenda Item Eggg i b 3 4) Revised Source Term for Ooeratina Reactors j
.j 4 4.1) Remarks-by the Subcommittee Chairman 158 S 4.2) Discussion of proposed report on 167 6 . plant-specific application of safety goals 7 5) Emercency Plannina f.or Advanced Reactors l l
8 5.1) Remarks by the Subcommittee Chairman 175 i 9 5.2) Briefing by and discussions with 176 10 representatives of the NRC staff 11 regarding a simplified approach to 12 emergency planning for advanced l l
13 reactors
,O 14 15 k 16 17 18 l 19 20 .
4 21 l
22 1 23 24 LO 2e NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234-4433
_ . _ . . . . _ - - - . - . _ _ _ . . _ , _ . _ _ _ _ ~ _ . . _ _ . . _ _ . . . - - . . _ _ _
5 !
I i
1 P-R-O-C-E-E-D-I-N-G-S :
2 (8:36 a.m.) !
O 3 CHAIRMAN KRESS: Will the meeting please come l
4 to order? This is the first day of the 436th meeting of .
5 the Advisory Committee on Reactor Safeguards. During ,
6 today's meeting, the Committee plans to consider the j 7 following items: one, the proposed rule on steam !
8 generator integrity; risk-based analysis of reactor '!
9 operating experience; three. revised source term for 10 operating reactors; four, emergency planning for advanced 11 reactors. And then we're going to' work on our reports.
12 This meeting is being conducted in accordance ;
l 13 with the provisions of'the Federal Advisory Committee Act.
14 Dr. John'T. Larkins is the designated federal official for !
15 the initial portion of the meeting. .
16 We have received no written statements or 17 requests for time to make oral statements from members of 18 the public regarding today's sessions. A transcript of 19 portions of the meeting is being kept. So it is requested 20 that speakers please use one of the microphones. And 21 before you talk, please identify yourself and speak with I i
22 sufficient clarity and volume so everyone can hear you.
23 We have some changes in the agenda that I'd 24 like te note before we start for everyone's benefit if I l 25 can find it. There will not be any presentations on Item L i NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. i (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
- i. 6 !
1 4 of the' agenda, which is the " Revised' Source Term for i l
2: 2 Operating Reactors." Instead, we'll just have a summary }
4 O 3 by our. subcommittee chairman of the subcommittee meeting. :
1
[ 4 And after that short summary, we'll use any time that's :
(
. 5 saved to work on one of our proposed reports, the ;
! i e
6 " Plant-Specific Application of Safety Goals." i 4 !
4 7 And Item 10 on the agenda, which is the ,
1 ;
i 8 " Report of our Planning and Procedures Subcommittee," has .
4 j
i 9 now been moved to Saturday. This slot will be used 6 so l l
j 10 to work on our reports.
11 CHAIRMAN KRESS: I do have a couple of items 12 of current interest. You do have a handout of items.of 13 interest. And there are a number of recent talks given by 14 Chairman Jackson that are very interesting, and I 15 recommend'you read them.
(
16 Oh, yes. Thank you. An announcement. I'd j i
17 like to announce that Jocelyn Mitchell is now the new Matt l 18 Taylor /Len Soffer. We're very pleased about that, 19 Jocelyn. Thank you very much. You have a little write-up 20 of a very, very short, brief resum6. It doesn't do her 21 justice. But I'll let you read that at your leisure.
22 I remind the members that they're scheduled to 23 be rebadged sometime today. So don't preempt that.
24 Let's see. We do have a number of letters. I 25 counted four on the list. They're all A with one A plus.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. i (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
7 1 So there's very littic distinction between them. The ons 7 ~3 2 that's A plus is mine, and it has an A plus on it because
( '
)
3 I may be leaving Friday night and we need to get it done 4 before I leave.
5 There's probably a fifth letter that I'd like 6 to put on the list. No. It's on here. It's the fourth 7 one. I didn't know it was on. We'd like to make a 8 Committee letter out of Dana Powers' write-up on the 9 direction-setting issue on research. So if you could look 10 at that, be thinking about it as a Committee letter.
1 11 MEMBER POWERS: I would appreciate feedback at i
12 any time on the numbers.
L 13 CHAIRMAN KRESS: Yes. Dana needs feedback on
,a k
i kJ -
14 it. And I think the feedback needs our own cover so that 15 people know what to look for.
16 MEMBER CATTON: Feedback on what, Dana?
17 MEMBER POWERS: It's the draft I put together 18 following our subcommittee discussion in Boston on the 19 Direction-Setting Issue 22, which deals with NRC's 20 research program.
21 CHAIRMAN KRESS: It's in this Handout 6.
22 MEMBER CATTON: I found it.
l 23 CHAIRMAN KRESS: Okay. Handout 6 also has 24 write-ups and comments on other DSIs. If members, other (a) 25 members, have comments on these or additional comments on l NEAL R. GROSS l COURT REPORTERS AND TRANSCRISERS 1323 RHODE ISLAND AVE., N.W.
j (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
l
ll - 8 ,
,{ .,
1 DSIs, why, we need to get those put together today.
l- 2 MEMBER CATTON: Which DSI is it?
l l'
3 MEMBER POWERS: Twenty-two.
4 CHAIRMAN KRESS: His is 22. I have no more 5 items of interest. Am I forgetting anything, Sam? Okay.
6 Are there other items members wish to bring up before we l
l 7 start? Anybody?
i l
l 8 (No response.)
l l 9 CHAIRMAN KRESS: Seeing none, I guess we can i
l 10 turn to the first agenda item, which is the " Proposed Rule l 11 on Steam Generator Integrity." This looks like it's 12 yours, Bob Seale. So I'll turn the podium over to_yor 13 VICE CHAIRMAN SEALE: Thank you, Mr. Chairman.
'O 14 This effort includes the preparation of a 15 draft rule and the supporting regulatory guide, w ich is l- (
! 16 specifically intended to provide a risk-informed, l 17 performance-based regulation for the industry to'use in 18 the management of its steam generators.
19 The scope of this effort includes the 20 characterization of the tubing in a generic steam 21 generator, the monitoring of the status of the tubes in l ;
22 the steam generator, and an analysis of the risk resulting l
~ 23 from steam generator tube rupture. This has been a j j 24 monumental effort requiring input from a broad spectrum of ;
25 experts in both NRR and Research. !
! NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
._ _. _ _ _ _ _ _ . . _ _ _ . . _ . . - _ _ . . _ . . . . _ . . _ . _ _ . _ . . . ~ .
9 1 The' industry has also been active in this {
2 effort. They've developed industry implementation 3 documents and a range of comments on the content of the !
- i. l 4 regulatory interactions that are implied in the reg guide. l l
5 Today we are first actually going to hear from 6 Dr. Jay Hopenfeld, who had requested the opportunity to f
7 speak to the Committee regarding a differing professional 8 opinion he filed.back in Jul of '94. Since that's sort of l 9 preamble to the current activities, we felt that would be ,
10 an appropriate place to do that.
11' Then we will hear: from the NRC staff. They've 12 been asked to specify the major areas of effort in their 13 preparations to date and note any specific areas in which l Ow ' 14 they're seeking ACRS comments or guidance, if you want to i 15 call it that. They've also been asked to identify areas )
i 16 of industry concerns, including those where the proposed l 17 rule'and/or the reg guide appear to be overly prescriptive 18 or where industry initiatives appear to be limited or 19~ discouraged.
20 Several specific questions that will be 21 covered include: one, how is risk factored into the 22 formulation of the rule? What is the impact of using CDF 23 versus risk in consequence evaluation?
24 Two, where is the evaluation conservative?
() 25 And this has to do with where you put the conservatisms.
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 I
10 1 And there are some rather unique or interesting aspects of 2 that choice that have been made.
(). 3 Three, what level of review is planned for j
4 proposed implementation documents submitted by the 5 industry? i 6 And,-four,.how are the probabilities of 7 failure. rates for one,'two, or three tube clusters 8 obtained?
9 And, five, what about the safety factor, which 10 happens to be a'three, on burst pressure selected versus 11 what the industry has gleaned from the ASME standards?
12 These are just some of the diverse scope.of efforts we j f
13- heard about.
14 We will also hear from the industry on some.of
.15 these questions. We have asked them to present their
'16 - approach to formulating a risk-informed, performance-based 17 rule, which they had presented earlier. ,
18 That kind of gives you a heads-up on what ,
1 i
19 we're going to hear about. So we'll first ask Dr.
l 20 Hopenfeld to share his concerns with us.
21 (Slide) 22 MR. HOPENFELD: Thank you, Mr. Chairman. I l
23 appreciate the opportunity which you give me to express my !
24 concerns, which I have briefly presented to you two years
() 25 ago.
NEAL R. GROSS i COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N W. !
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
[ 11 1 (Slide)
,- 2 MR. HOPENFELD: My main purpose is basically l
3 to inform the Commission that the public will be at risk
)
4 should this rule be approved.
i 5 (Slide) ,
l 6 MR. HOPENFELD: During my presentation to you 7 two years ago, the staf f told you that the 10-* CDP was 8 obtained on a preliminary basis and that much more 9 advanced studies had been conducted indicating that this j 10 number is very, very conservative, and it was just a 1
i 11 scoping type of a study.
12 I would like to tell you that this is 13 incorrect, it is misleading, and it is confusing. The im A
Cl 14 number that we have now is still on the books. It's 10-*.
15 Now, I do not want to leave you with the impression -- and 16 I speak to the best of my ability. I can tell you that.
17 I'm not trying to imply that anybody is trying 18 to mislead you intentionally. The results are misleading, 19 have misled, have confused, resulted in a letter from the 20 ACRS to the Commission that is confusing.
21 CHAIRMAN KRESS: Joe, you're talking about the 22 induced core melt frequency due --
23 MR. HOPENFELD: Yes.
24 CHAIRMAN KRESS: -- to all accidents involving 25 steam line breaks?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
-- - . .. - . . - - . ~ . - . _ . .- - . . - . - _ . . . . - .
-)
12 j
1 1 MR. HOPENFELD: That's correct, sir. I will 2 get into the details of the problem in one minute. 1 LO-3 The new item that wasn't discussed in detail i 1
4 has to do with severe accidents, where I have calculated I
5 four years ago that in a case of severe accidents or- 1 6 station blackout, the tube'will go first, instead of.the-7 surge line. What that really means to the public is that
'8 you get an early release versus later.
i 9 MEMBER CATTON: You calculated it?
10 MR. HOPENFELD: Yes.
11 MEMBER CATTON: Is it possible for you to 12 share that with us?
13 MR. HOPENFELD: It's possible. Not only is it' 04 14 possible. I shared that thing as part of the DPO. -All 15 the documentation was there. It was a scoping study. It i
16 was done in --
l 17 MEMBER CATTON: I'm interested, in particular, 18 in the calculations --
19 MR. HOPENFELD: Yes.
20 MEMBER CATTON: -- that led you to make that 21 statement, --
22 MR. HOPENFELD: Yes.
l 23 MEMBER CATTON: -- that the tube would fail 24 first.
l 25 MR. HOPENFELD: Right.
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS !
1323 RHODE ISLAND AVE., N.W.
(202) 23M33 WASHINGTON, D.C. 20005-3701 (202) 234 4 33 l
1 13 ,
1 MEMBER CATTON: Could you get that for me, 2 Noel?
- o 3 MR. HOPENFELD
- You have that document.
4 MEMBER CATTON: I may well. God knows where. ;
l l
5 MR. HOPENFELD: It's in the compilation of the 1
6 original DPO.
7 MEMBER CATTON: Can you get it for me, Noel?
1 8 I have it here? _
9 MR. HOPENFELD: I'll give you the reference.
l 10 I'm glad you asked. Can I continue?
l l.
11' VICE CHAIRMAN SEALE: Please.
l f
12 CHAIRMAN KRESS: Yes, please.
13 (Slide) 14 (Slide) l
! 15 MR. HOPENFELD: The basic. problem was i
16 identified in 1991. And what it really said is very, very 17 simple, that if you operate with through-the-wall cracks, 18 the probability of a multiple failure of tubes versus a 19 probability of rupturing the tube is much higher.
20 The whole concept is very simple. All it 21 really says is a leak before break. And all the test data l 1 22 shows that. So when you have a situation, when you 23 operate by definition or intentionally and knowingly with l 24 through-the-wall cracks, what's going to happen first?
25 They're going to leak.
1 NEAL R. GROSS COURT REPORTERS AND 1RANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
WASHINGTON, D.C. 20005-3701 l i (202) 234 4433 (202) 234-4433 l
14 1 Now, the main issue is how much they're going 2 to leak and at what time frame. That is the only issue.
O 3 Now, if you spend a lot of time, a lot of money studying 4 tube ruptures, that's fine, but it has really nothing to 3 do with the issue I'm talking about.
6 That was brought out in a very, very brief i
7 definition of what the problem is. I realized that when 8 you go for years and you get yourself conditioned to i
9 thinking in terms of a tube rupture, it's difficult 10 suddenly to shift into a different scenario, but that's 11 the reality if you allow it to operate with 12 through-the-wall cracks. And the only issue is timing and 13 how much. i 14 CHAIRMAN KRESS: Help me, Joe. You're saying 15 that the leak out of the tubes amounts to a LOCA. And if I 16 it's big enough, then you'll eventually uncover the core j 17 and reach core melt.
18 MR. HOPENFELD: Yes.
19 CHAIRMAN KRESS: So it's a function of how big !
20 that leak is?
l 21 MR. HOPENFELD: And how many.
22 CHAIRMAN KRESS: And how many leaks you have?
23 MR. HOPENFELD: Right. You're correct. In ;
24 other words, you take a crack. Okay? On the average,
() 25 let's say it runs at one gpm. If I have 600 cracks -- and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
15
- 1 that's what Westinghouse was allowed in the case of 2 Trojan. I don't know what they're allowing now, but let's ;
(' 3 say 600 times one is right there, is a tube equivalent, i
4 which is 600 gpm.
5 If-you take other flow rates because there's a 6 lot of uncertainties how you calculate these flow rates, 7 you can come up with 6,000. And this is only one aspect
'8 of the uncertainty.
9 I'm not trying to solve the problem.
10 CHAIRMAN KRESS: Yes.
11 MR. HOPENFELD: I'd like to convey to you that 12 what I'm talking about is a different concept of how to j 13'! analyze the accident. When you've spent thousands and
\~/ 14 thousands of dollars studying tube ruptures and what 15 happens to burst, that's. fine. But it's not a limiting-4 16 case.
17 It's fine to study a case and even simplified 18 case if it is a limiting case. This is not the limiting P
19' case. And that's why I want to really spend the time and j
20 tell you why it is not a limiting case. '
21 CHAIRMAN KRESS: These leaks, how do they show l
22 up? How are they detected?
23 MR. HOPENFELD: How do they detect? )
l 24 CHAIRMAN KRESS: Yes. Is it by water level?
l 25 MR. HOPENFELD: I'm going to get into that. l i-NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 2344 433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
_ . . . . _ . . . .. - . _ _ _ . . _ . . .. . _ _ . . . . - ._- - _ m _ . _ .
16 '
1 CHAIRMAN KRESS: Oh, you're going to get into f
2 the details of that?
O 3 MR. HOPENFELD: Yes.
4 CHAIRMAN KRESS: Okay. j i
5 MR. HOPENFELD: I mean, the partial answer to {
l 6 your question was followed up with another document, which !
r i
7 was three to four months later, going into how they leak !
i 8 and what causes this to leak. 1 i
L 9 The next. thing what happened was a brief 10 letter using the RELAP code and perturbing the heat l 11 transfer through the steam generator. What you have if i
12 you have intact tubes, the mechanism of heat transfer i 13 through the steam generator is natural circulation. 7 1 '. ,
l '- # .14 There are uncertainties, but it's not a big i I l 15 deal to solve that problem for what you need for severe-I 16 accident analysis. But now you're talking about having I '
17 leakage. Now you have a first convection superimposed on 18 free convection coupled. What happens is you're removing 19 more heat from the core. By removing more heat, there's i I
20 less heat on the piping to rupture the surge line. It's l 1
l 21 just that simple.
22 Now, it turns out that it doesn't really take i 23 much flow rates. And I think I used that kind of gpm.
24 That was enough to cause the tube to go first.
i
() 25 Now, that was done four years ago. And what's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
- (202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l , _ . . ,
17 j 1 the first thing they've'done? They threw the thing in the !
2 basket because'it just wasn't used because it's a ;
O 3. different concept.
i I
t
-J 4 CHAIRMAN KRESS: In this case you're talking i 5 about flow rate of steam?
6 MR. HOPENFELD: I'm talking about flow rate of
- 7. steam. l l
8 CHAIRMAN KRESS: Yes. {
i 9 MR. HOPENFELD: You're going to station -j 10 blackout. l 11 CHAIRMAN KRESS: Yes. j 12 MR. HOPENFELD: The pressure is something like 13 2,600. The steam generator.is empty. Now you go.into
\ 14 natural-circulation. When you have small tunes, that~ j i
! 15 thing' draws more -- I'd'like to get into this back --
t 16 CHAIRMAN KRESS: Okay.
l l 17 MR. HOPENFELD: -- because we can spend a lot 18 of time on that.
19 VICE CHAIRMAN SEALE: And we have to keep 20 going.
21 MR. HOPENFELD: Yes. And I have to keep
- 22 going.
l
! 23 Okay. What is the mode of a leakage which you i 24 could have? One is actually, obviously, you increase the ,
- 25 pressure in the crack. Then it's going to a higher flow l NEAL R. GROSS
' COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. '
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
l 18 ;
I l 1 area. Now, you don't have one crack. You have a multiple 2 network of cracks with a stress corrosion problem. It's
, O-H 3
l not one fatigue cram :. So you open it up, and there's 4 plenty of data to indicate that that is the case.
5 The next thing, what you have -- anybody who 6 has done a hydrostatic test, you know you run the test )
! 7 cold. And you don't get a leakage. Nothing happens. You '
l l 8 bring that thing up to temperature, and it leaks. And 9 that's known. Why? Because corrosion products are not a j i
10 l welding rod. They're simply not a welding rod. The 11 expansion of corrosion products is different than the I 12 expansion of bare metal. So this is an experience that 13 you can refer to.
14 The next is jet erosion. That's a difficult 15 problem to cope with because there are so many parameters.
16 \
that come in here. However, I'll pick up something at l l
17 random. Cracked boiler-incidents have been known to' occur i
18 because of a pinhole in the tube that would erode and cut I
19 off the adjacent tube. And then all the black liquor will l
20 just slump on the floor, and you have a big explosion. So 21 this happens.
22 The process of water jet machining, it's still 23 not in practice to a lot of degree, but it's in the 24 process of development. You have high jets that you can L ("% )
l (,,) 25 remove material on the order of seconds, millimeter per ;
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234 4 33
19 ;
i 1- seconds. .
g 2 The parameters are different. The pressures ;
%-) i 3 are different. But when you start putting-the numbers in :
I 4 there as well as the assumptions you come up -- ano you ;
5 can come up with erosion rates on the order of minute. j i
6 You can come up on the order even less. And you can come .
7 up on the order of hours. >
i 8 And in the case of a steam line break or the l
)
9 design basis accident, the scaled time here is on the 10 order of hours. In the case of the surge line, we're !
11 talking about on the order of minutes when you sit in that !
i 12 high temperature. :
13 Well, when I see somebody working and solving O 14 creep rupture problems and spending a lot, a lot of time i
15 on that, where the-problem is really controlled by the !
T 16 leakage, it's very difficult to comprehend.
17 MEMBER SHACK: Well, but in the particular 18 95-05 case, of course, the cracks are within the tube l l 19 support plate. So there really is no --
20 MR. HOPENFELD: Well, the cracks are not 21 within the tube support plate. Let me tell you why.
l 22 First of all, the tube support plate is not going to sit
, 23 there and wait for the steam line break without moving.
24 secondly, if you look at -- and even a small l
i ) 25 sample. And I looked at a trojan, where some of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
~ _ _ . . _ . . _ _ _ . - _ .__ _ _ _ _ . _ _ _ . . . _ . - _ _ _
20 t i
i cracks were - .I think there was one or two ecming out of j i
.p 2 the support plate. So all you need is just one squirt at i
' ?.d 3 high pressure, and you propagate it. And what is in the
'}
j
- 4 support plate -- and that's why it's so expensive to pull 5 those tubes because there's a lot of crud there. It does L 6 act like a welding material..
J 7 Now, is it going to behave like that when you l
\
f 8 move it around at high pressure? Nobody said it would. .i t- ;
l 9 But it's not to me to say. It's the guy who comes up with j
,- l 10 proposing the rule who should tell you that. He doesn't i 11 even have to tell you. He doesn't'even have to solve the !
1 12 problem. He should tell you what his assumptions are. I 13 was here for two days. I hadn't heard any assumptions.
O 14 I apologize. I get excited. But that's my.
15 nature.
16 CHAIRMAN KRESS: That's all right. We don't 17 mind.
I I
18 MR. HOPENFELD: Don't worry. I'm not going to 19 get a heart attack. I'm too old for that. ;
I 20 (Slide) t
- 21 MR. HOPENFELD
- Now, when we rule the stress l
- i
, 22 corrosion cracking, -- and that has been around, stress ;
I
. 23 corrosion cracking has been around, for many, many years I- 24 -- one of the basic rules you find in any handbook, any 1 25 book, any old-timer that has been around for many, many i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS f- 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
21 1 years that you don't take test data and extrapolate to 7
2 true life conditions. Nobody is doing that. Nobody has
( i
\"
3 done that. I went to ask FAA if they do that thing, and 4 they said no.
5 Now, you can do that for fatigue. If you have 6 a crack fatigue and you have a thick metal, yes, fine.
7 That's perfect. There's no problem. Even there in 8 certain situations you can't do it when you have a 9 multiple fatigue situation.
10 So that problem, what is being done here, 11 what's implied in that voltage-based regulation is that 12 this is allowed. This is the first time that this is 13 being allowed.
,g
\s 14 Let's take the next one. And that goes back 15 to the hydrostatic test. The PNL data, they have run a 16 set of flow rates on cracks. They all show either they 17 plugged after a while or they plug completely or 18 partially.
19 Now, are they going to be unplugged during 20 these tests? Who knows. But you've got to tell somebody 21 that you know that they won't because it affects the whole 22 sequence of the accident.
23 Now, let's go take the next one. I already i
l 24 talked about the jet erosion. You can make some
(, ) 25 estimates. There's data on turbine erosion. The NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
.-_. y 22 l 1
1 pressures are different. The velocities are a little bit
! 2 different. Materials are different. But you can make :
t 3 some estimates. And I've made those estimates. It l i
L 4 depends what assumption you make. j
-1 5 One problem you have is that the crosion rate l 6 is proportional to the cube or second power of. velocity. j i
7 It's very sensitive to it. And then it's proportional to !
i 8 the concentration of the particles. In the case of a j 9 design basis, you have super-heated liquid. So these are 10 going to be the particles that it's going to impinge.
11- In the case of the severe accident, you have 12 1,200 degrees delta between the core and the steam ~j
-13 generator. There's going to be a lot of stuff coming out O
k/ 14 of solution just because of the nature of law.
15 When you clue something -- I don't-know what 16 has negative solubility, but boron oxide is not one of 17 them. Why boron oxide behaves like aluminum oxide, which 18 is being used in water jet, I don't know. But those )
l
, 19 people that present it to you and tell you, "We've got the )
I i
20 problem solved, and we are studying on some minute details l l
21 of how the sensitivity of the code affects something,"
22 they should tell you about these problems.
l
) 23 MEMBER CATTON: Joe, the jet erosion is not 1:
4 24 the tube with the leak. It's an adjacent tube.
() 25 MR. HOPENFELD: Yes, obviously.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
1 (202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
23 1 MEMBER ~CATTON: And it sees the crack.
2 MR. HOPENFELD: Right. And that's what that O f3 crack boiler was, exactly the same thing. One sees'the ,
1 4 next one. That's where the British were when they were l 5 talking about Sizewell. That's what multiple tube rupture :
6 is, not some numbers that you pick.
I 7 In the last presentation, when they talked j l
, 8 about 10-* , somebody said that, I made an assumption about I
L 9 ten tubes cracked. There's nothing like that. These were I l 10 just an example. It's the concept. You see, once you get 11 a certain border, a certain line, you've got an.
]
L .
I 12 uncertainty. And that's fine, but don't ignore it. As 13 long as you tell the public what you have done, that's 4 O I 14 fine. )
I
! 15 But right now the 10-* because.of these.
16 uncertainties, the 10 says that once you agree to' operate l 17 with these cracks, through-the-wall cracks, and you cannot 18 quantify them and the technology is not there to quantify 19 them, what you're really saying is the whole thing is 20 driven by steam line break probability. And that's what 21 it is. And then when they told you here about 10 or 12 or l
22 whatever, it's just not true. That has nothing to do with 23 that.
24 Now, one thing I did forget to mention back, 25 that the issue originally in 1992 was classified as i
i i NEAL R. GROSS i COURT REPORTERS AND TRANSCRIBERS l
. 1323 RHODE ISLAND AVE., N.W. .
--(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 )
24 1 high-priority issue. NRR strongly objected to that. Th0y
,y 2 said, "Well, no. This is a trivial problem. We ought to
( )
RJ 3 just ignore it and drop the issue."
4 Now, on the books, -- this has been four years 5 -- that issue is still on the book, instead of being 6 prioritized.
7 MEMBER SHACK: What issue are we talking 8 about?
9 MR. HOPENFELD: One sixty. Can somebody help 10 me? It's 161 or 1,000.
11 MEMBER SHACK: Not a number. What's the -- ;
12 MEMBER BARTON: What is the description of the 13 --
I )
'/ 14 MR. HOPENFELD: Okay. The issue was what 15 happens -- let me take the previous slide, and I'll give 16 you the name of the issue.
17 (slide) 18 MR. HOPENFELD: The issue was how tubes crack, 19 what would be the core damage frequency when you have 20 leaking tubes during steam line break. It was classified 21 as high. Let me just get it here for you. Oh, here it 22 is. It is GI-136, " Multiple Steam Generator Tube 23 Leakage," September 28, 1992. This is probably the 24 longest generic issue on the books. It's still there. It
,/ m
(_,) 25 awaits a resolution.
NEJd. R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
25 1 In other words, when you go for a rulemaking, 2 the first thing you do, you set up the technical base. !
3 The purpose of that was to set up the technical base for-4 this problem. It has never been done. It's still there 5 on the books. j 6 Did I answer your question, sir?
7 (Slide) 8 MR. HOPENFELD: Well, these are difficult 9 problems. I'm not trying to minimize that. And I'm not 10 trying to say that you can solve them, but what bothers me i
11 the most when I see an effort of a year or two on thermal 12 hydraulic problems related to calculating temperature 13 distributions where you have an untold tons of crud in 14 there that will probably affect the distribution in 15 millions of ways, when you have a leakage that you don't 16 consider, and when I see it, at the same time taking this 17 thing into perspective, when you have all of these 18 uncertainties, you will run a sensitivity study. Wall, 19 the sensitivity study doesn't mean anything. You've got 20 the wrong model.
21 I'm done.
22 VICE CHAIRMAN SEALE: Any questions?
23 MEMBER CATTON: Do we have the opportunity to 24 hear from the people who are pushing for this rule on
) 25 these issues or have you already heard from them?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
,- - -....-. . . . - .- ~. . - . - . . . . - - .. - _ - - .- ~- . . - . . - . . ~
4 26 ;
~
1 VICE CHAIRMAN SEALE: Not specifically.
2 CHAIRMAN KRESS: Not specifically addressing l
V 3 these concerns. They are here and --
I i
4 VICE CHAIRMAN SEALE: Yes. Jack, would you l l
5 like to --
6 MEMBER POWERS: Maybe we could summarize i
I 7 exactly what the points of contention are here. We have a 8 contention -- j 1
9 MEMBER CATTON: Do;you have a summary slide, )
I 10 Joe? 4 11 MR. HOPENFELD: Basically it was the 12 conclusions, but I'll be glad to summarize to you in one '!
13 word.
N.
14 MEMBER CATTON: Well, not one word.
15 MR. HOPENFELD: Two words. The basic problem 1 i
l'6 relates'to physical parameters, which involve erosion, as I I
17 you know, stress corrosion, which are extremely difficult i 18 to characterize within the accuracy that you want.
19 They're okay for severe accident analysis to understand 20 what the problem is, to understand what the issues are.
21- That is fine.
22 Spend as much money as you want, and as long 23 as the good sign is behind it, it will add something to 24 it. But to go and tell somebody, " Hey, I can calculate 25 the" -- let me.just give you an example.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
27 1 You take the RELAP code for the steam
,e 2 generator for a relatively simple problem, very, very
( '3 i
3 simple problem, for calculating the heat transfer on the 4 secondary side. And you can't do it unless the code is 5 properly calibrated. I'm digressing.
6 But what I'm saying, the basic parameters are 7 not sufficiently known to allow you to claim that you can 8 calculate these things within that accuracy. The tube 9 rupture in the severe accident is 1,400 F. And you tell 10 somebody you can calculate within 100 degrees. It's 11 crazy. It's just impossible to do.
12 The point of contention, the basic knowledge 13 is not there to come up with the conclusion that you are
! \
kl 14 being presented with.
15 MEMBER CATTON: So basically -- I'm just i
16 looking at your viewgraphs -- there are three. One is the i i
17 increased crack flow area due to increase in pressure, --
18 MR. HOPENFELD: Yes. Okay.
19 MEMBER CATTON: -- crack unplugging, and jet
]
20 erosion.
21 MR. HOPENFELD: Yes, right.
22 MEMBER CATTON: Are those the three physical I
l 23 processes that -- l i
24 MR. HOPENFELD: That I can identify. I'm
.o
() 25 sure, Ivan, you can come up with better ideas, but that's l NEAL R. GROSS i COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
28 1
1 what I could, yes. !
3 2 MEMBER POWERS: It seems to me that there's
(' ~')
~
l 3 another contention here that ought to be resolvable. And !
4 that is that we have a crack networking, rather than 5 isolated cracks. I might, you ought to be able to look at l
6 tubes and tell whether you're getting networking or not.
7 VICE CHAIRMAN SEALE: Okay. i i
1 8 MR. HOPENFELD: Can I just take one minute i
9 because I really forgot? )
10 VICE CHAIRMAN SEALE: Sure.
11 MR. HOPENFELD: There was an item of iodine .
l 12 spike, --
13 VICE CHAIRMAN SEALE: Yes.
(' ' '
14 MR. HOPENFELD: -- which we have spent a lot 15 of time last time. And one of you members wrote a letter i
16 indicating there was a mystery factor in there. I'd like ,
i 17 you to know I read in the regulations that I've got that l l
18 mystery factor is still there.
19 CHAIRMAN KRESS: Dana, the summary, as I see 20 it, is, number one, that for the accident, the tubes will 21 fail first, instead of the surge line. And it's because 22 of the heat transfer hasn't been calculated correctly. It 23 hasn't included direct flow.
24 When the tubes fail, the leakage will be quite
/~'N
() _
25 a bit larger than we have assumed. Therefore, it will NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 200053701 (202) 234 4433
29 1 lead to core melt, virtually probability of one or l 1
p
-- 2 something. Therefore, the whole thing is driven by the
~
3 probability of the steam line break. Does that wrap it 4 up?
5 MR. HOPENFELD: I think you're 90 percent I 6 there. There's one thing. You see, the main contention 7 of the leakage that I was focusing wasn't really in terms 8 of severe accident because it's really too difficult.
9 It's not the leakage versus the burst. I go 10 with the Larson kind of analysis to figure out what the 11 rupture is but do the correct thermal hydraulic because, 12 you see, forget about the -- in other words, if you let it f
13 fail by creep rupture at this point because for the severe t
I
\/ 14 accidents, I think you can justify because the timing is 15 there. See, we're not talking about two or three minutes.
16 In the case of DBA, you may be talking about hours before 17 you really get the plant down. Okay?
18 So the timing is different. Severe accidents 19 I could see stretching my imagination. Okay. Let that 20 thing go and fail by Larson type of fatigue, creep 21 fatigue, but do the proper thermal hydraulic.
22 So you see what I'm saying? I mean, what 23 you're saying is correct. You really should look at the 24 whole thing. There's a limit to what you can do.
/'
( ,S) 25 MEMBER CATTON: What's missing I think is the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l i (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
30 1
1 increased heat transfer -- j 2 MR. HOPENFELD: Right.
O 3- MEMBER CATTON: -- due to the tubes that are !
4 leaking already.
5 MR. HOPENFELD: Right, correct. l 6 MEMBER CATTON: I don't know how to. convert ;
i 7 100 gpm into gas flow through the crack and what that does
)
8 to the velocities of -- !
9 MR. HOPENFELD: A hundred gpm oomebody'gave me !
i 10 from RELAP. I mean, what I did there, Ivan, I did -- !
11 MEMBER CATTON: But, you see, I think you made 12 a good point. If I have a leak in the tube somewhere and !
l 13 there's gas flow through the leak, that particular tube is
. O.
V 14 going to have.an enhanced flow --
15 MR. HOPENFELD: Right. That was the whole 16 point.
17 MEMBER CATTON: -- into heat transfer. And 18 what is it going to do?
19 MR. HOPENFELD: Right. And I tell you I did 20 100 gpm, 500 gpm on the RELAP.
21 MEMBER CATTON: Gpm usually is 100.
22 MR. HOPENFELD: But he gave it to me in terms 23 of -- I know. He gave it to me in terms of steam. So 24 that's the way it came out of the code. And I remember 25 the numbers. I'll be glad to check. It was 100 and 500 f l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
3 (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
31 1 that we used there. And that thing pulled enough heat so 2 the surge line was staying there, happy sitting there o
O~ 3 while the tube went at 1,400 degrees.
4 MEMBER CATTON: As I see it, there are four 5 things. And one is this what we're just talking about 6 now. What would a leak in one of the tubes do to the heat 7 transfer? How much would it enhance it? Should we worry 8 about it at all?
9 And that's a complicated process when you have 10 gas flow through a very small crack with a large delta P 11 because there's going'to be sonic flow somewhere, all 12 kinds of nonsense. P 13 MR. HOPENFELD: Right.
14 MEMBER CATTON: Second is the increased crack 15 flow area due to increased delta P. What's that s going 16 to do? And that plays back onto the first.
17 There's crack unplugging.
18 MEMBER SHACK: That's actually one thing you
-19 can calculate.
20 MEMBER CATTON: I understand. That's all 21 right.
22 And the fourth is this jet erosion. Given 23 that semething happens with the first three, are you going 24 to have a problem because you're blowing the crud from the
() 25 one tube onto the surface of the other tube? And what's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
.~ .- _._ ~. . _ _ . _ - _ . _ _ . _.. . _ - . . . . .. _- _ ..._
32 i 1 this' going to do?
1 s 2 I have seen some of this cutting that Joe's I 3 talking about. Actually, there was a guy at UCLA who was 4 doing it. And they can slice through rock and everything 5 else. It's awesome. ;
6 MEMBER POWERS: But, Ivan, that's clear that 7 sort of stuff is done at pressures like 40,000 psi.
.i 8 ' MEMBER CATTON: I understand, but, Dana,.let ;
9 me give you another example. j i
10 MR. HOPENFELD: Can I answer? !
11- MEMBER CATTON: In the Nike Zeus, at one time ;
i 1
12 they were going to use Canard-type controls. And they "
'i 13 tried to turn the flow through 90 degrees. It wasn't O !
14 anything like those kinds of pressures. And there it was 15 aluminum oxide. I mean, it just cut holes like you 16 wouldn't believe.
17 MEMBER POWERS: But there's quite a difference l
18 between water and aluminum oxide slurry.
I 19 MR. HOPENFELD: Can I answer your question?
20 MEMBER CA.TTON: This is gas flow particles.
21 MR. HOPENFELD: Can I make a comment?
22 MEMBER POWERS: No. I think -- .
l 23 MEMBER CATTON: Excuse me, Joe. Just a j l
24 minute.
25 It was gas flow with particles, and not all NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
33 l 1 that high speed either. Anyway, I think --
z .s 2 MR. HOPENFELD: Can I make a comment about the l
/
} i
!. I l
40,000?
3 It is true it is 20 to 30, but don't forget that j l
4 the velocity goes in the square root of P. Don't forget i
5 that the loading is very important, what concentration of 6 particles you have there. That's a very important 7 parameter, j 8 If you take a factor of ten, you take a square !
i 9 root, and then you talk about the timing over there, 10 you're talking about the' jet erosion is on the order of .
l 11 seconds. We're talking about minutes.
12 So when you put all of that thing together, 13 you say, "Ah, it depends what assumptions I made."
- a n
() . 14 Really, you hit the nail on the point. You really hit 15 that thing because it's the uncertainty. It doesn't say 16 yes. It doesn't say no.
17 VICE CHAIRMAN SEALE: It sounds like we have 18 another scaling problem.
19 MR. HOPENFELD: Well, it's more than scaling.
20 It's lack of --
21 MEMBER CATTON: No. It's scaling to the 22 problem.
23 MR. HOPENFELD: Right.
24 VICE CHAIRMAN SEALE: That's what I mean.
,s-() 25 That's what I mean.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE. N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
34 1- Perhaps we should let Jack Strosnider --
2 MEMBER CATTON: It would be interesting if O 3 they.could address each one of these individually, the i
.4 four. l VICE CHAIRMAN SEALE:
5 I'm sorry we hit you 6 cold with these. On the other hand, I'm sure you're aware 7 of the issues.
8 ME14BER CATTON: See, I think what we were 9 focusing on before was just'a part'of one of these.
10 CHAIRMAN KRESS: The heat transfer 11 coefficient.
-12 MEMBER CATTON: That's right.
13 (Slide)
- 14 MR. STROSNIDER
- Good morning. My name is I
15 Jack Strosnider. I'm with the Division of Engineering in i 16 NRR.
17 What I would suggest is that the staff go 18 through the presentation we had planned. I-believe that
- l l 19 some of the issues that Dr. Hopenfeld has expressed or i 20 raised are covered in that presentation. And if there are ~l l l 21 some additional questions, then we could try to address l
22 those when we're through.
i: 23 VICE CHAIRMAN SEALE: I believe we -- )
- 24 MR. STROSNIDER
- You also had a list of issues i
'25 --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 40005-3701 (202) 234 4 33 I
i 35 1 VICE CHAIRMAN SEALE: Yes.
. 2 MR. STROSNIDER: -- that the Committee was
~J 3 interested in.
4 VICE CHAIRMAN SEALE: Yes. Yes.
5 MR. STROSNIDER: We have tried to capture 6 those in our presentation. I'm sure you'll remind us if 7 we miss one. And I guess we have about 45 minutes. i 1
8 VICE CHAIRMAN SEALE: Yes, that's --
9 MR. STROSNIDER: So, unfortunately, we have to 10 go fairly quick.
11 VICE CHAIRMAN SEALE: Yes, i f
12 MR. STROSNIDER: But we'11 do the best we can.
13 (Slide)
). MR. STROSNIDER:
14 Just wanted to start off with 15 a brief introduction of what the staff's objectives are.
t 16 One is to provide the technical bases. We've had the 17 better part of two days' discussion in providing some of 18 the technical bases.
19 We mentioned during the meeting with the 20 subcommittees that there is a NUREG report in preparation-21 wh.ich will document these bases. Unfortunately, it's not 22 complete yet. So the Committee hasn't had the opportunity l
l 23 to see that yet. But it does provide those bases.
, 24 We obviously are requesting ACRS endorsement. ;
25 That's part of our process for rulemaking. We recognize NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l4 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
l
36 1 not having seen all of these bases that might be difficult es 2 for you to do at this time. So what we're really
.O 3 interested in is some feedback, what you can give us based 4 on the presentations'that you've heard, some areas that-5 you think right be of concern or that need to be addressed 6 in_particular. We'd like to see that in the fairly short
~
7 term if possible.
'8 Looking at the rest of the schedule here, of 9 course, we are having this meeting today. We plan on 10 putting the entire rulemaking package into the concurrence 11 process. And this is a point where I think we~would 12 really be requesting ACRS endorsement.
13 This entire package would include the draft 14 rule, the draft regulatory guide, the statements of 15 consideration that go along with that, the regulatory 4
16 impact analysis, and the report documenting the technical 17 basis. So it will be quite a package. We expect to have 18 that out probably in mid December. That's our objective 19 at this point.
20 We've scheduled a CRGR meeting in January. l l
21 And the Chairman is expecting to see this package in l
22 February with the goal of putting it out for public 23 comment. And I note that the endorsement we're looking j 24 for is to go forward for public comment at this time. So )
! 25 the -- i I
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
' 1323 RHODE lE'.AND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
37 1 VICE CHAIRMAN SEALE: Jack, you are aware of 2 the fact that our December meeting is early in the month? !
O- 3 MR. STROSNIDER: Okay.
4 VICE CHAIRMAN SEALE: And we don't have a 5 January meeting. l l
i 6 MR. STROSNIDER: Right. We'll have to look at !
7 that and see what it does to the schedule. );
8 VICE CHAIRMAN SEALE: Okay. i l
9 MR. STROSNIDER: I appreciate that )
10 information.
11 Okay. Public comment period and then, of 12 course, time to resolve the public comments. We're l
13 looking for final implementation in March'of '98. It's an
(
14 ambitious schedule because it is a very complex issue, 15 covers a lot of areas.
'k 16 (Slide) 17 (Slide) 18 MR. STROSNIDER: I just wanted to go in very l 19 quickly some of the things we were striving fer when we j i
i l 20 put the draft rule and regulatory guide together. Our l l
l 21 intent is that it be performance-based. And you'11 see 22 that we have the basic concept in this rule and regulatory 23 guide of establishing performance criteria and measuring 24 tube condition against those performance criteria.
4 25 I like to look at this in terms of what's not I
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
1 38 i 1 in the rule anymore. What's not in the rule is very l i
- s 2 prescriptive methods for doing inspections or repairing l t i i
\',)
3 tubes, that sort of thing. So that provides some 4 flexibility, which I'll talk about a little bit more 5 later.
6 Risk-informed; that is, both of these, 7 performance-based, risk-informed, that's consistent with 4 1
8 Commission policy on rulemaking. I want to underline or l 9 emphasize in form this rule is not risk-based. All right?
10 It's not purely risk-based. It is risk-informed. And 11 I'll try to explain what I mean by that as I go through 12 describing some portions in the rule.
13 In general we have performed risk assessments.
i
\_ ') 14 And we've used that to gain some insights. But we're also 15 looking at maintaining defense-in-depth and satisfying the
\
16 GDC, which goes along with that. ;
17 MEMBER CATTON: Just for my own benefit, when 18 you say " risk-based," I understand what that means. You 19 sort of carry it from an initiator all the way to an 20 impact on the public.
21 When you say " risk-informed," does that allow 22 you to uncouple it so now you can say, " Gee, this lowers l 23 risk. That doesn't. I'm risk-informed. Therefore, I'll 24 do this"?
,/ 3
( ,) 25 MR. STROSNIDER: To uncouple? I'm not sure I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
39 1 understand.
7x 2 MEMBER CATTON: Well, I see a number
\
C/
3 associated with the failure rate of these tubes.
4 MR. STROSNIDER: Right.
5 MEMBER CATTON: But I don't see it tied to a 6 risk statement.
I 7 MR. STROSNIDER: And I will discuss that when 8 I get to the --
9 MEMBER CATTON: Okay.
10 MR. STROSNIDER: -- performance criteria. !
11 We want this rule to be adaptable. I used the 12 word " flexible" earlier. What we're talking about here is 13 we want a framework that will allow addressing new types k- 14 of degradation. The types of degradation we're dealing 15 with today are significantly different than what we dealt 16 witi. in the late '70s and early '80s.
17 The technology; for example, an in-service 18 inspection, is much different today than it was at that l 19 time. There's ability to collect much more data. There's 20 ability to do inspections of the entire tube bundles 21 within an outage. And the technology is also much more 22 sensitive. And that's being improved. With each outage, 23 we see new degrees of sensitivity and abilities of the l
24 NDE.
(_) 25 So that's where I said earlier we don't want i NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
I 40 l 1 to be prescriptive. We want to allow people to use the
,w 2 best technology to address the particular forms of l L_)t l 3 degradation they're dealing with.
4 In this regard, we want to provide incentive i
i 5 to use that technology. I mentioned when we discussed !
l 6 this with the subcommittees this has been a particularly 7 difficult area to deal with, the reason being that if you j l
8 look -- and I'm looking specifically at NDE technology.
9 The way the technology typically evolves is i 10 that you develop the ability to detect defects before you !
l l
11 develop the ability to actually size them, which is what '
l 12 you need to do some sort of structural integrity !
13 assessment or whatever. j i
'/
)
14 So the position has been that if we detect !
15 indications or defects, that if you can't characterize j i
16 them, that you take them out of service. Now, that's i 17 resulted in a lot of tubes being plugged or repaired. .
e 18 So we've somewhat of a dilemma here for the 19 industry and for the NRC, but the regulatory position is 20 if you can't characterize it, well, you have to take it 21 out of service. That's what we've been doing, and that's 22 the proposal in the regulatory guide. So, yes, we want to 23 provide incentive, but it's a difficult thing to do.
l
[ 24 There are some other things in the rule. For
- (_)
25 example, it says you have to include NDE uncertainty in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
. . -- -___ -.. .-. , . . - __... - - ~ -. . - - . . . - ~.._
41 :
1 your calculations to demonstrating that you're meeting the l i
e 2. performance. criteria. Well, the more you can reduce that l
- (
3 uncertainty, the more tubes perhaps you can leave in i 4 service. So there is, we think, some incentive there.
5 Part of this rule that we think is very 6 important is that it be enforceable. We believe that in j 7 moving to performance-based regulation, we have to have ;
i 8 some clear enforcement guidance and objectives. And we ,
l 9 think that's important for the staff so that we can make i
10 sure that this is working properly. We also think that 11- it's only' fair to the industry that they recognize what !
12 the expectations are. f I
(Slide) 13
/ I 14 MR.-STROSNIDER: During the presentation with i
15 the subcommittees, we actually went through this paragraph l 16 by paragraph and described some of the bases and the 17- objectives. We obviously don't have time to do that l 18 today, but I did underline a few key areas that I thought 19 we ought to focus on. ;
j 20 With regard to applicability, we're saying ,
21 that this rule be applicable to all pressurized water 22 reactors. The rationale is that there are things that we l
23 believe need to be upgraded in the NRC's requirements:
L
[ 24 inepection,. scope, technologies, those sorts of things.
25 The industry is going way beyond that's in the i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D.C. 2000# 3701 (202) 234-4433 l
i 42 i
I current requirements. Our requirements were developed L ,
2 back in the '70s. We need to update those. And we want 3 to make sure that all licensees are, in fact, following 4 that philosophy.
5 Okay. With regard to performance-based, the {
6' main objective here is to monitor and maintain'the f l .
7 condition of the steam generator tubes consistent with 8 performance criteria. l l
9 Okay. Then we talk about what the safety ;
10 functions of the steam generators are. I want'to 11 emphasize here that conditions resulting from core damage i 12 severe' accidents are explicitly included in this rule. i i
13 Joe Donoghue is going to talk a little bit more about this j 14 aspect of the rule and the regulatory guide in his ;
15 presentation.
16 One thing that I also wanted to emphasize here 17 is-that with regard to the performance criteria, we're 18 saying they should be NRC approved performance criteria.
19 I emphasize this because it's different than what you see 20 in the maintenance rule, and that's one thing that people 21 always want to make a comparison of.
22 The rationale we have is that if you look in 23 the maintenance rule, those items for which the industry I
24 is establishing its own criteria are items -- well, let me L
I
() 25 say it this way. They're items that are already included NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
43 1 in the technical specifications, those components that are 2- important to safety.
7-~
i V 3 Steam generator tubes, likewise, are currently 4 covered in the technical specifications. You know, we 5 think it's appropriate, if you're going to go to this 6 performance based thing that -- approach, that the NRC t
7 should approve these criteria because this is obviously a 8 very safety important component given that it serves j- 9 function both as reactor coolant pressure boundary and 10 containment.
, 11 With regard to the performance criteria, we 12 have laid out the expectations that there would be 13 criteria established in the structural integrity, which is
. s
-- 14 basically looking at burst; operational leakage integrity, i
i 15 which addresses, I believe, the issue that Dr. Hopenfeld 16 was raising. And we're explicitly -- well, that along ,
17 with the accident induced leakage.
18 You're explicitly establishing performance i.
19 criteria to address the potential for not burst, but for 20_ leakage through a number of tubes. And I'll talk about 21 this more whe't I get to the performance criteria; and I'll 22 also mentior how this was dealt with in severe accidents.
1 23 VICE CHAIRMAN SEALE: Jack, I -- one'of the
.i 24 questions we raised in our list had to do with the NRC's
/~N
'\m,) 25 assessment or evaluation of documentation submitted by the j i
j 1
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
-(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433 i
i I
44 l
1 licensees --
7S 2 MR. STROSNIDER: Right.
! / i v !
3 VICE CHAIRMAN SEALE: -- to support their J 4
1 4 processes. Is that part of this -- would that be part of 5 this NRC approval on the performance criteria? That is, 6 the basis for the criteria may well be part of that l 7 documentation.
8 MR. STROSNIDER: No; I'm glad you reminded me l I
9 of that, because I managed to skip a viewgraph here. What J l
10 we're talking about here is reviewing and approving the l 11 performance criteria. j l
12 VICE CHAIRMAN SEALE: Yes.
13 MR. STROSNIDER: One of the policy issues and
/ T
\
\>i 14 an area that we were looking for some feedback on is --
15 it's this first bullet on performance based approach --
16 how much flexibility to give the industry. When I get 17 into the next page of the rule, what we're really talking 18 about here is we have approved -- we will have approved 19 performance criteria that the NRC has looked at.
20 Now, those performance criteria are not in 21 every case just as simple as going in and measuring a 1
22 defect and comparing it to an acceptable defect size. ]
l 23 VICE CHAIRMAN SEALE: Yes.
Y 24 MR. STROSNIDER: There's calculations (3
C) 25 involved, perhaps for the tube integrity. The first l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
' 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
45 1 calculations, you have to do some fracture mechanics p- 2 analysis. For the leakage aspect of this, there can be 3 some significant calculations or empirical approaches that 4 need to be developed.
5 And one of the main questions that we're 6 asking is to what extent does the NRC need to review and 7 approve those methodologies. Those are the methodologies 8 that are used to assess the condition of the tubes 9 relative to the performance criteria. For example, for 10 those of you who are familiar with Generic Letter 95-05, 11 there's an empirical correlation of voltage -- of eddy 12 current voltage amplitude to burst; also to probability of 13 leakage; and to leakage rate.
l 14 Now, the question is, and the industry's 15 proposal is that those methodologies would be developed in 16 industry documents and that they would not be subject to 17 NRC review and approval. And the question is, in a 18 performance based regulation, how much confidence do you 19 need in the way people are actually measuring against the 20 criteria.
21 Does it require NRC review and approval?
22 VICE CHAIRMAN SEALE: Okay, i
l 23 MR. STROSNIDER: And it's a question we're 24 raising that we think is a good area to get some feedback.
( ) 25 I would note, from the industry's perspective, if we NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
46 1 review and approve those methodologies, all right, a great fs 2 deal -- most of the flexibility et this rule is gone. We i
3 could have done that basically under the existing 4 regulatory framework through topical report review and 5 technical specification amendments.
6 The second area where we're looking for 7 feedback is, is the inclusion of severe accident and the 8 risk work within the regulatory framework. As I already 9 pointed out, the rule explicitly address -- includes a 10 need to address these areas. And Joe Donoghue is going to 11 talk a little bit more about that in his presentacion.
12 Thanks for reminding me. I wanted to point 13 those two areas out.
g j 14 VICE CHAIRMAN SEALE: We wanted you to.
15 MR. STROSNIDER: Going on to the -- this is 16 the second page of the rule. There's a section in here 17 which essentially has been described as programmatic 18 elements, and I think that's a fair characterization.
19 We're indicating that in order to manage tube degradation, 20 the a licensee's program should include these elements:
21 And it involves preventative measures to try 22 to reduce corrosion or degradation; in-service inspection; l 23 plugging and repair; monitoring of primary to secondary 24 leakage during normal operation. Okay, this is what we 73
( ,) 25 refer to as condition monL:oring. This is the requirement NEAL R. GROSS COURT REPf RTERS AND TRANSCRIBERS 1323 Rh0DE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
l
47 1 to monitor the "as found" condition against the
,q 2 performance criteria. 1 i t :
8
'")
3 And also, to do a forward look with regard to 4 do you think you're going to satisfy -- or do you have l 5 confidence that you'll satisfy the performance criteria at i
6 the end of the next operating cycle. Now, this is an 7 interesting thing here because current regulatory j 8 requirements don't really explicitly require that.
9 Now, obviously the plant has to maintain its 10 licensing basis. But if you look at the tech spec, it i 11 says plug at 40%. You're good for the next cycle. This 12 tells you you might have to adjust your cycle length or 13 your plugging criteria to make sure that you're going to i
i 8 V 14 satisfy repair -- the performance criteria at the end of 15 the next cycle.
('
16 Okay, and -- okay, so that's the operational 17 assessment. The last section down here where I have some 18 areas underlined really lays out some of the expectations.
19 These are almost performance guidance buried within the 20 rule and what we expect to happen up here. In particular, 21 I wanted to point out utilizing inspection or test methods 22 and analysis methods to provide a high confidence in the 23 assessment of the condition of tubing relative to the 24 performance criteria.
) 25 As you'll see in most regulations, this is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l WASHINGTON, D C. 20005-3701 (202) 234-4433 (202) 234-4433
l' 48 f 1 pretty general language. If you look in the regulatory i
i s 2 guide, we'actually provide guidance on how you qualify i i
J- 3 inspection methods. Right, and it references and ;
i.
2 l 4 augments, to some extent, the EPRI industry guidelines. l 5 okay, but we also provide guidance which l 6 indicates, for example, that when you do this condition ;
I 7 monitoring, that you should have certain probabilities and l 8 confidence levels in the calculations you're doing. All l
9 right, which dictates, to some extent, how you can do -- !
j I
10 what methodologies you need to use to show that you're 11 meeting the performance criteria.
f 12 I described earlier to the subcommittees that 13 the intent here, when you look at the regulatory guide, is 14 basically to draw a box in which the licensee can operate 15 within that, modifying repair criteria and cycle lengths, i
16 managing their tube degradation -- as long as they stay 17 within that box, that we will-feel comfortable that we 18 have a satisfactory approach.
19 Okay, one more comment in that area, because i l
20 you asked that we address some of the issues that the ,
21 industry has brought up. The regulatory guide associated i 22 with this rule is -- I guess I've lost count -- 70, 80 i 23 pages; something on that order. It's a very long 24 regulatory guide. l 25 It's been described -- some people have said NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
t 49 1 well, it's too prescriptive. I'd like to think that it i
2 has a lot of detail in it. The reason for that is that O 3 we're trying to establish those boundaries in which we're
{
4 comfortable that a licensee can operate. And this gets 5 back to the question of what do we need to review and 6 approve, i 7 When you look at the regulatory. guide, I think 8 you need to look at it with the idea of does this ,
9 regulatory guide bound the problem to such an extent that 10 if people satisfy it, we're comfortable with the way
'11 they're managing tube degradation. .
.t
> 3 f
12 And finally, again, we have license -- the t
i 13 requirement that licensees will take reasonably achievable :
14 measures to address basically the severe accident issue.
i ,
15 Again, Joe Donoghue will talk some more about the 16 approaches that are offered in the regulatory guide for i
17 satisfying that. -!
18 MEMBER SHACK: Jack, just coming back to this 19 loss of flexibility if you review those methodologies, now )
l 20 that you've been through 95-05 and you've developed the l
21 reg. guide and it's clear probably both to the staff and 22 the industry what you're looking for, don't you think that i
! -23 the review of those low level documents would be much 24 easier and faster than it was in the original development i (/"h
_,/ 25 of the IPC's and --
I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433 l
50 1 MR. STROSNIDER: Well, I think it's easier 2 once you've been through the exercise. Basically --
3 hopefully.the industry recognizes more what the 4 expectations are, and you sort of set up a process. I'm 5 not convinced that it's necessarily any faster. It -- and 6 part of that is part of what paces it. And a lot of-this 7 goes=back -- falls back on the industry.
8 You know, how quickly can they develop and 9 provide the data that are necessary to support some of 10 these alternate repair criterias. And a lot of these are 11 empirically based is what we're seeing. We're looking at 12 --
13 MEMBER SHACK: But that's true whether you 14 review'the document or not.
15 MR. STROSNIDER: Yes; and that's a good point.
16 This rule is not a panacea. This rule will not stop 17- degradation in steam generator tubes, and it won't solve 18 all the industry's problems. Okay, hopefully it will 19 provide a framework and some flexibility for them to deal
'20 with them in the most rationale manner. I 21 But it won't make the problems go away, and 22 the onus will still be on the industry to provide an 23 adequate technical basis for changes. And that's somewhat 24 difficult in terms of, like I said, developing the 25 necessary data bases.
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. 1 (202) 234-4433 WASHINGTON, D.C. 2000>3701 (202) 234-4433 j
51 1 I'd like to say this is the viewgraph that zs 2 makes it all clear. I'm not going to go through this in
("' )
3 great detail. I'd start off by saying ignore what's on 4 the 1cft side here temporarily and just look at the logic 5 that's involved in this part of the flow chart.
6 This is the way we visualize the rule and the 7 regulatory guide being implemented when it's in place.
8 I'll start with the plant shut down and performing tube 9 inspections. After the inspections are performed, the 10 data is collected on the condition of the tubing. You 11 perform a condition monitoring assessment.
12 This is where you compare the condition of the 13 tubes to the performance criteria. All right, are the
,a f 1 A
u- 14 performance criteria satisfied? If not, you inform the 15 NRC. You determine what appropriate corrective actions 16 are necessary, and you move into your operational 17 assessment.
18 If they are satisfied, you go directly to the 19 operational assessment. Here, you look at your cycle 20 lengthy; tube repair criteria; and you do the forward look 21 to determine if you have confidence that you'll satisfy 22 the performance criteria at the end of the next cycle.
23 Assuming you've satisfied that, you perform your 24 appropriate repairs, implement whatever preventive (n) 25 measures appear to be appropriate -- those could be NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
52 1 reducing operating temperature, changes in water chemictry
,y- 2 -- perform a risk assessment.
N]
I'm not sure that that's exactly where it fits 3
4 in in the flow, but as part of assessing the condition of 5 your generators, you need to look at beyond just design 6 basis accidents. Implement accident mitigation measures -
7 - this basically means that you've done your operator 8 training, your procedures are up to date.
9 Plant's restarted, the NRC gets an outage l
I 10 report. During operation, you monitor operational i
11 leakage.
12 MEMBER CATTON: Is this the risk assessment 13 part of 2.0? 2.0 is establish NRC accepted performance
/, h
\
\-) 14 criteria. It feeds 3.
15 MR. STROSNIDER: No, there's not a specific
\
16 performance criteria in the rule; but I think Joe Donoghue i l
17 can talk about how it is -- how we do envision it being )
l 18 addressed.
19 I would point out just a couple of things 20 here. What goes on over here -- it's basically guidance 21 on how to qualify your NDE systems. All right, how to 22 develop the assessment methodology that we talked a little 23 bit on how to do this assessment. If you look at the 24 regulatory guide, it was our intent in most of this --
(_) 25 with the exception of the NRC approved performance NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
53 1 criteria, the rest of this is addressed in the regulatory 2- guide.
V 3 Our intent was that it not be prescriptive.
4 All right, by that, what I mean is there's guidance -- we t
5 don't tell people what their water chemistry program 6 should be. In fact, here, we're referencing EPRI 7 guidance. We don't tell people what sort of NDE 8 technology to use, but we tell them how to qualify it.
9 So, it is specific in the way you go about 10 qualifying some of the methods. But it doesn't tell you 11 whether to use eddy current or ultrasonic, or whatever 12 method is most appropriate.
13 One other comment I'd make just with regard --
C)
!V 14 a lot of people perceive this rule as just a relaxation. j I
! 15 All right, and I think it's been pointed out that there j l- 16 would be the ability under this rule to perhaps leave 17 deeper flaws in service. And if you look at Generic l
18 Letter 95-05, even going up to -- through all cracks.
19 All right, but I'd like to point out a couple i l
l 20 other things in here. The performance demonstration -- i l
l- 21 the supplemental performance demonstration, qualification l 22 of NDE methods, is more stringent than what has been used 23 in the past. Also, this forward looking operational i
24 assessment is a step that, although I think people are --
25 licensees are doing it, they're responsible licensees, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
54 1 it's not something that's currently explicit in our 7 2 regulations.
3 So there are some things in here which are 4 more stringent.
5 MEMBER POWERS: Have you looked at the 6 relative risk worth of this rule?
7 MR. STROSNIDER: The relative -- for the rule 8 and after the rule?
9 MEMBER POWERS: That's right; for, say, a 10 representative plant -- say the NUREG-1150 for PWR's.
11 MR. STROSNIDER: Again, I don't want to steal 12 Joe's thunder; but, in fact, we've done a risk assessment, i
1 13 all right, but it really -- it's rather difficult to do a
('~T J 14 delta that we'd have a lot of confidence in. And when we 15 talked, for example, during the subcommittee meetings i
16 about flaw distributions, well, what's the flaw l l
i 17 distribution going to be after this rule's implemented at 18 a given plant?
19 It's very difficult for us to say because, you 20 know, we don't know what's going to be proposed. So, 21 we've done risk assessments to gain some insights as to 22 what we think are risk significant, and we've tried to
- 23 address those in our performance criteria and in 24 requirements for plants to assess their risk.
g
(,) 25 So, --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
55 1 MEMBER POWERS: Do you know what the risk-I 2 worth of improved tube integrity is?
O 1
L 3 MR.'STROSNIDER: Well, the regulatory impact I 4 analysis is currently under way. But I think -- actually, l l
5 if you go back to NUREG-0844 and some of the-earlier- I 6 stuff, we have some handle on that.
l- 7 MR. LONG: This is, of course -- excuse me; 8 this is Steve Long with the NRC staff. In trying'to look L 9 at the numbers we have now and just doing this in my head, 10 the risk achievement worth for tube integrity -- that is, 11 having tubes that have no flaws as opposed to tubes that 12 couldn't fail under any condition whatsoever -- should be 13 in'the high 10-5 range with the calculations we've done now lO
- V 14 for things that we think are roughly consistent with the L 15 rule'if it was adopted.
l 16 MEMBER POWERS: That's a frequency. worth, 17 right?
I 18 MR. LONG: No, I'm sorry; that's the frequency I
l 19 of core damage with, in this case, bypass because of the i-20 tube integrity.
i l.
L 21- MEMBER POWERS: It's unlikely to show up very 22 big in a frequency just because the frequency of a tube )
23 rupture accident is low alrecdy in at least a NUREG-1150 l
l 24 plant. But it might sho'.i up higher on a risk worth than a 25 frequency worth.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. !
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
56 1 MR. LONG: Okay, if I could do it all the way i ,esy 2 through to risk -- yes; we've had some debates on exactly
! I 3 what the source term would be for different parts of this.
4 But typically, trying to scale on NUREG-1150, NUREG-1150 5 had approximately half the risk being from steam generator 6 tube flaws and the other half being from some sort of 7 other bypass like a V&V.
8 What we've done is add a lot of reasons that l
9 you might have tube failure to 1150, and it tends to 10 dominate the V&V. So the worth with respect to risk from 11 what we've done p1obably -- for flaws again -- probably 12 would be somewhere on the order of 80-90%. l 13 MEMBER POWERS: Significant.
r~x i ' '/ 14 MR. LONG: It's the majority of the risk 15 probably to the public.
16 MEMBER CATTON: Dana, when they carry it all 17 the way to risk, they really don't treat from the ruptured 18 tube to the public very well, do they?
19 MEMBER POWERS: Well, --
20 MEMBER CATTON: They sort of just assume if 21 the tube ruptures, it's in the face of the public.
22 MEMBER POWERS: There are a lot of questions l
23 and debates on how well that particular aspect of it is 24 done. But lay initial interest is in the -- just how g
! _,) 25 valuable it is. It is my impression that NUREG-1150 l [
NEAL R. GROSS -
i COURT REf 0RTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
57 1 neglected decontamination on the secondary side of tha 2 steam generator, and it's my impression that that might be 3 significant.
4 I,JREG-1150 did try to understand at least the 5 uncertainties on fission product. transport on the primary 6 side and put in some sort of a distribution on that that's 7 arguable. If the secondary side estimates are very badly 8 wrong, then the risk importance of a steam generator tube 9 rupture is, of course, a -- it goes down almost directly 10 in proportion to the decontamination that you get on the 11 secondary side.
l l 12 It could be very significant. It's obviously 13 a research area that we are going to have to wait sometime O
l i
14 in the future to get sorted out, because it's a fairly !
l i
15 complicated decontamination problem and one that I don't 16 think you can do in your head.
17 You know, lots of people argue well, there's 18 decontamination or there's not decontamination. It's, I 19 think you could say, an example where thermal hydraulics 20 and scaling will come to the fore.
I
! 21 MR. STROSNIDER: I want to talk now about the I
22 performance criteria. There were some questions that the 23 committee wanted to address, and I think some of this will 4
24 also address some of the issues that Dr. Hopenfeld raised.
l 1
() 25 Let's start off first with structural integrity NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
- (202) 234 4433 WASHINGTON, D.C. 2000 # 3701 (202) 1344433 I
58
'l performance criteria.
- 2. And this is_ addressing _ tube rupture. All-O 3 right, and we have both the deterministic and the-4 probablistic approach. The deterministic approach we 5 referenced Section 3 factors of safety. !
6- CHAIRMAN KRESS: Is this an "either/or" or an i i !
7 "and"? l
)
8 MR. STROSNIDER: This is~-- you can use either j 9 one.
10 CHAIRMAN KRESS: Either one.
t 11 MR. STROSNIDER: If you look at the design 12 equations in Section 3, you'll'see that there's a factor I I
- 13 of safety of three for normal operating pressure, and 1.4
(
L 14' for main steam line break. There was a presentation by 15 the industry yesterday, I guess, which pointed out that t
j 16 this factor of safety of three is really on the ultimate L 17 stress for the tube.
i 18 And what governs failure is not the ultimate, 19 but it's something more like the average of the yield in 20 the ultimate. So in fact, although you believe that L 21 you've designed with this factor of safety of three, 22 factor of safety on rupture is really somewhat less.
[
23 All right,'and the industry has argued that l
a f 24 this nutaber could be reduced. There's an incentive for ]
() 25 the industry in this area in that one of the difficulties NEAL R. GROSS l,
COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON. D.C. 20005-3701 (202) 2344433 ,
1
. .. . - . - _. - - . _ - - . .~ . . - - .- - . - - -
59 1 they have is if they lower operating temperature on soms 1
s 2 units, the delta P goes up. All right, so in order to 1 3 lower temperature, which would be a good thing to do to 4 reduce the corrosion rate, they have to go plug tubes. 3 1
l 5 That doesn't make'them real happy. So they're ;
I 6 looking for some incentive in'this area. We're expecting 7 a submittal from them which lays out their rationale, and 8 we're going to have to take a look at that. We are also l
9 looking, however, for some consistency in how flaw 10- evaluations have been done in Section 11 and how they've 11 been done historically per Reg. Guide 1.121 which lays out L 12 our current guidance in this area. ,
13 So, we have an issue here that we need to look l ,
14 at when we get the industry's document.
i 15 Moving on to the probablistic portion of this,. !
16 the first criteria here, and I'll say this -- state this i
, 17 very simply. This is the current rate of spontaneous tube l
18 ruptu es in the industry. It's been trending down, I 19 would mention. It was at one point in time a little over 20 10-2, i
21 The industry gets some credit for doing l
l 22 inspections and correct things in order to reduce that.
. 23 In very simple terms, the regulator's view of the world is 24 I don't really want to issue a rule that's going to drive l( ) 25 this up, so the performance criteria is -- let's see if we l
NEAL R. GROSS !
court REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
60 1 can't continue doing this. It's just as simple as that.
2 Although, there is some more to it. If you O 3 tAke.that frequency of rupture -- all right, and this has 4 been put into the recently performed risk assessment --
5 you calculate -- I think maybe you can get some more 6 details on this later, but ycu get it in the low 10" range 7 of core damage with containment bypass.
8 And this is what I'm talking about in terms of 9 risk informed. Yeah, this is the current frequency that 10 we see, but we've looked at it in terms of a risk ~
11 assessment to say well, what does it mean. All right, so 12 we have some numbers. They will go through the reg.
13 impact analysis as part of what does that mean; is it an 14 acceptable level or not. I 1
15 Now, going to conditional probability of i
16 ruptures given a main steam line break, this addresses the 17 design basis accidents. Okay, and the other criteria for 18 one or more -- and you can see the criteria there. These 19 numbers came originally from NUREG-0844 which was -- dealt 20 with the generic issue on steam generator tube integrity.
21 They were utilized in a risk assessment at 22 that time, which was determined to provide an acceptable
'23 level of risk. In the risk assessment that's been done in 24 support of this rule, this number 5 x 10-2 was used as a
() 25 conditicnal failure probability. And again, it resulted l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS ,
1323 RHODE ISLAND AVE., N.W. l (202) 234-4433 WASHINGTON, D.C. 2000 # 3701 (202) 234-4433 I l
.. - . . ... - - ... . ..- . - - . ~ .-. -..- . ... - - . - .-. . . - .
f 61 1 in some frequency of core damage and containment bypass in the low 10-5 range.
l'"Gb 2-3 MEMBER FONTANA: Is that consistent with the 4 data with NUREG-1150 results that showed half of the risk 5 was due to steam generator' tube rupture?
f 6 MR. LONG: I'm sorry, I'm missing the context 7 of the question. This is Steve Long with the staff. Can
~
8 we -- you're asking is what consistent with what?
u 9' MEMBER FONTANA: Well,-he said that the l
I 10 numbers came from some early risk assessment.
l l 11 MR. LONG: I'm trying to make sure I know 12 which numbers you're talking about.
13 MEMBER FONTANA: Like 5-x 10-2 for one or more 14 --
r )
l l 15 MR. STROSNIDER: Yeah, these came from NUREG- !
16 0844.
17 MR. LONG: Okay, those three numbers are in 18 NUREG-0844, and they were estimates made on the basis of L
19 how fast things were assumed to be degrading at that time l 20 with a small number of flaws. When we -- those numbers l
l 21 were not included in NUREG-1150. They were --
l 22 essentially, steam line breaks were not part of NUREG-23 1150.
a 24 We have more recently tried to do an analysis 25' that would look at the risk again from steam line breaks i
NEAL R. GROSS ,
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
,nr -
62 1 with more current knowledge. And the 5 x 10-2 as a f- 2 performance standard in the rule was put into that risk b 3 assessment to see what the effect would be. The other two 4 values though for 2.5 x 10 2 for'two or more tubes, and 10-3
.5 for-ten or more tubes, because we don't think there are 6 very high frequency ways of getting the rupture of ---
7 well, in our case, more than about three tubes -- we 8 lumped things differently.
9 We lumped more like one to three tubes, 10 somewhere between four and about 15 tubes, and greater 11 than 15, and what we've done more recently. Merely, it's 12 only that the few tubes case that seems to be most 13 important to risk when you control the number at 5 x 10-2, 14 MEMBER POWERS: You've not found a propagation 15 mechanism?
16 MR. LONG: We didn't find a mechanism that 17 seemed to -- could not be self-limiting or to propagate at l \
l 18 normal temperatures -- you know,. normal operating j i !
19 temperatures. So that we went over this, I guess, l l
20 yesterday. The secondary depressurization mechanism j l 21 basically pulls down the primary pressure along with it j' l
22 until the secondary is pretty much depleted.
2
.3 And then the pressure on the primary starts to 24 rise, and that's causing the pressure differential to j o
() '25 increase. And at that point, you start rupturing tubes.
i l NEAL R. GROSS l j' COURT REPORTERS AND TRANSCRIBERS I
! 1323 RHODE ISLAND AVE,, N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
63 1 But, as you rupture them, you limit the ability of the l l
2 pressure on the primary to increase further. And it's O 3 just -- the LOCA is self limiting phenomena. We looked at ,
i 4 the main steam line break's ability to essentially yank on 5 the -- axiall'y on the circumferential cracks. l
.l 6- That has two factors. One, it looks.like a l 7 fairly small force; two, it is a fairly infrequent thing l
8 to get that kind of steam line break between the reactor i 9 containment wall and the MSIV's.. So that didn't look very 10 important. We looked at the circ. crack that.tries to 11 break the crack -- break the tubes around it.
12 That didn't look like it had a very high l
13 probability of occurring. And the thing that looked like 14 it might conceivable rupture a lot of tubes at normal
[
l 15 temperature would be some worth of an ATWS extreme high 16 pressure scenario where even though the failure of the i
17 tubes does turn the pressure around some, it's conceivable
- 18 that the pressure would go very high, and we dealt with 1
19 that on a frequency basis.
20 So, the things that are at normal temperature 21 which this is trying to address looked t. be pretty much 22 linear proportional to the probability that you'd fail one 23 or two or three tubes. And we took that in our risk 24 assessment as .05, which is what the rule is trying to 25 limit it to.
l' l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
r -- , -
64 1 MEMBER FONTANA: What was the scenario that.
2 1150 used to determine the impact of steam generator tube ;
'b) 3 rupture? '
4 MEMBER POWERS: It's a spontaneous rupture
- 5 with a human error to fail to recover all of them.
6 MEMBER FONTANA: Did that rupture propagate, r
7 do you know?
8 MEMBER CATTON: I don't think it mattered.
9 MEMBER POWERS: Yeah, I don't think it 10 mattered; but maybe -'-
11 MR. LONG: I don't remember any propagation of 12 the rupture in any of the 1150 work.
13 MEMBER CATTON: The assumptions were pretty l
(O
,) 14 gross as to what they would do once it ruptured.
J 15 MEMBER POWERS: The more important -- or the
\
16 thing that people focused on a lot in what I call the l l
17 Heny-Nixey sequence is -- which are spontaneous tube ]
18 rupture -- was the subsequent failure to recover. And l i
19 it's a human error. I think in every one of them, it's a 20 human error.
l 21 MR. LONG: There's usually -- a couple one is l-22 the failure to act -- to properly isolate the secondary
- 23 sides so that you stick open the secondary boundary and 24 the inability to not just depressurize below the secondary 3
25 relief value which is not too hard to do, but to have to l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
! (202) 234-4433 WASHINGTON, D.C. 200 15-3701 (202) 2344433 1
-]
65 j L1 get all the way down to where you're nct boiling in the 2 RCS before you depleted the IWRST.
O 3_ So-by the first error, you make your second -j i
4 action more difficult; and then there's a higher i
[- 5 probability of fail in the second action.
l r- 6 MR. STROSNIDER: If I can summarize, Steve, I L
7 think what we learned from our risk assessment is that l 8 basically the risk is dominated by'this first value. .The 9 rest of these -- number one, we don't believe the numbers 10 are necessarily that significant of the probability of 11- those kind of multiple ruptures.
12 But when youfgo through the analysis, it ,
i 13 doesn't change your overall risk significantly. I woCid
'O 14 point out a couple of things about this probablistic 15 approach. First of all,fyou need a significant amount of 16 data on the degradation mechanism in order _to be able to 17 do this. l 18 There's only one degradation mechanism right I
19 now -- probably ODSCC at the tube support plate --
20 addressed in Generic Letter 95-05 where you can really do 21 this. All right, the anticipation is that as data are l 22 collected, there might be more. Where you can't do it, 23 you need to fall back to a deterministic approach. ,
L 24 Nonetheless, we have put in the performance I /^
l ( ,g f 25 criteria some limitations on the potential for multiple NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
~. -- . . - . . - -- - - .. . - -.
66 l l
L 1 rupture. So again, it's risk informed, but we're also (
i s 2 using some of our judgement in terms of enforcing GDS and l lks) l s
3 defense in depth.
{
4 And in fact, you get to this next issue, --
l l L 5 and there was an industry presentation on this -- what ;
L i'
! 6 we've-indicated is that each -- for each degradation l
7 mechanism, it should not contribute more than 20% to this j l
l 8 conditional failure probability. We really came up with )
l .
l 9 this when we were working with Generic Letter 95-05. !
l l J
L 10 We recognized that if this was what we've ,
l !
11 included in our risk assessment and this is our goal, we l
l 12 shouldn't let one degradation mechanism take us right up ,
l l 13 to that limit. There are other degradation mechanisms i !
\~- 14 going on. We used some judgement here in saying well, l 11 5 let's reduce it by a factor of five and we're comfortable ;
l 16 with that.
l 17 The industry has suggested that'this could be l
18 handled in a different way. For example, you could sum up 19 all the known degradation mechanisms and make sure that 20 they don' t exceed the 5 x 10-2 In their presentation, 21 they indicated that if you consider ruptures from unknown ]
22 sources, things that you're r.ot managing, it could go as 23 high as .1 conditional failure probability from unknown f
24 sources and still, in their risk numbers, come out
() 25 acceptable.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234-4433
67 1 one of.the concerns the staff has and one of 2 the-things we're dealing with here is again, not just
.f s 3 being risk informed, but looking at defense in' depth. And f 4 I suspect if you look at their numbers closely, you can go ;
5 to a conditional failure probability of one and still show. -t 6 that, by their calculations, that the risk is acceptable.
7 We don't think that that meets GDC's or 8 defense in depth, and so we want to keep this number down ,
9 lower. There might be a better way of addressing this I 10 issue. We need to give that some consideration. But I 11' think the main point here is that we want to limit for ,
12 each -- you know, the contribution such that you're not 13 exceeding this value and leave some allowance for the !
iy U- 14 unknown forms of degradation.
15 okay, and'I emphasize again this is.for design 16 basis' basically the main steam line break is governing. ,
17- Now we move into I think a very important area I
18 which is the operational leakage performance criteria.
19- Operational leakage should not exceed the tech spec limit.
20 There's a recommendation in the reg. guide consistent with 21 industry guidance at this point. It's 150 gallons per i 22 day. Experience has shown that good primary to secondary I
23 leakage monitoring, good procedures for the operators to
, 24 respond to rapidly increasing leakage or to leakage i
() 25 reaching certain levels to be very effective in heading NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 433 WASHINGTON, D.C. 20005-3701 (202) 2344433
68 1 off potential ruptures.
2 This is - -and it's already included in the O 3 tech specs. What we're saying here is if you do exceed 4 that' limit, that we think it's -- we want to make it a 5 performance criteria so that we, the NRC - well, the 6 licensee first will have to assess why and determine if 7 any corrective actions are appropriate; and the NRC would 8 know about it, and we could get involved in looking at 9 corrective actions.
10 This also, I'd point out, tells you something
-11 about the integrity of the bundle. And this limit is much 12 tighter than it was Some of the original technical 13 specifications are up, I believe, around ten gallons per 14 minute. I don't know. I mean, this is down around .3 15 gallons per minute,_I think, or .1.
16 At any rara, this is a much tighter limit than 17 we had in the past with the addition of looking at rates 18 of increase. If you're meeting this, you have some 19 additional assurance that your bundle is fairly leak 20 tight. But you still need to address what's going to 21 happen under postulated accident conditions, right.
( 22 Now, the performance criteria we have are two l
23 fold, and this leakage is not governed solely by ruptures.
j 24 And Dr. Hopenfeld raised the issue of a lot of small i
() 25 defects leaking -- leaking, and that integrated leakage l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344 433
69 1 becoming significant. That's what this is intended to 2 address. The first criteria is that the integrated O 3 leakage should not exceed the total make up capacity.
4 All right, this is a judgement in which we 5 have looked at this and said we don't want to challenge 6 the operators with loss of -- essentially loss of coolant- J 7 accident in which they don't have the make up capacity.
8 All right, in addition, you obviously -- you j i
9 have a containment bypass situation, so you need to make j 10 sure that you're satisfying radiological. considerations.
11 Now, if you look in the regulatory guide, it talks about 12 how to do these calculations. There was a question 13 yesterday -- I made the comment that I think we have some )
j
(" ~ - .
\ 14 conservative methods for doing these calculations, and .
15 it's probably more appropriate for the people who put
-16 this together to address it.
17 But the point I was trying to make is that 18 when you look at performance criteria, you need a 19 tolerable -- that is, you've got to have margin in your 20 performance criteria. If you exceed the performance 21 criteria, you don't want it to be an unacceptable 22 situation. And you can do that by setting lower criteria i
23 or by doing the calculations in a more conservative manner 24 so that in reality, we don't believe that's what you're 1
() 25 really going to get.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433 l
70 1 All right, so there's a balance there. But I 2 think the people who put that analysis together could l
'~
3 address that better. All right, so this is addressing 4 under postulated accident conditions, potential leakage; 5 and it's limited by these two criteria. In addition, when 6 you look in the regulatory guide, you have to be able to 7 do these calculations.
8 You have to have sufficient data, and you have 9 to demonstrate that the c.orrelations are appropriate, and 10 you have to take into consideration uncertainties. And '
11 there are confidence levels with which these calculations 12 are intended to be performed. I mean, so this is to 13 address under postulated design basis accidents.
(. )
i / 14 And the issue came up about what under severe 15 accidents, and there's not an explicit performance 16 criteria here. But in the risk assessment that was 17 performed, we actually did some calculations similar to 18 what Dr. Hopenfeld was describing. A lot of this will be 19 documented in the NUREG report, which I recognize he's not 20 had the opportunity to see and the committee hasn't; and 21 we didn't go into this much detail yesterday and the day 22 before.
23 But in the risk assessment model where we 24 talked about tube failures, the assumption was put in the o
i j 25 model based on calculations we had dor.e that if you had a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
~ . . . ~ - - - - - . - _ . . - - - _ , - - ~ . . -. - . - - _ .
71 !
1 1/4 inch through wall indication, we analyzed the mass .
-2 flow rates,=we got data on gas impingement on these types !
O- 3- of, materials and the rate of erosion. It is -- can.be i
4 very high.
5 .We concluded that under those conditions, you j 6 could cut through an adjacent tube within tens-of minutes, 7 all right, which is significantly less than the times 8 we've been talking about the tube circ. line or other j 9 component -- primary system component failures. So the 10 assumption in the risk assessment was:that if you.had a 11 through wall crack 1/4 of an inch or greater in length, 12' that it went to tube failure.
~
13 All right, so when those risk assessments:were 14 performed, all right, the' potential for small crack 15 leaking and failing an adjacent tube was taken into.
16 consideration in that way. Now it's driven -- and I have L
17 to acknowledge it's driven by the flaw distribution that's 11 8 assumed and the frequency with which you expect you'll 19 get that-kind of defect.
20 It also tells you, I think, something else.
I 21 The potential to have large numbers of indications leaking j i
I 22 under the severe accident condition is not really the i
l 23 concern. Because if you have one, you're going to cut a l
24 tube and you're going to'go to tube failure.
) 25 So that -- those numbers he's talking about in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.. . . - . _- .. - . .. __-.~ . - - - . - -
a 72 1 the thermally induced assessment, those scenarios that 2 went to thermally induced tube failures under severe 3 accidents,.the assumption was made that you would cut 4 tubes, and conditional failure probability of one that.you 5 had that kind of leakage.
6 So that's been accounted for in the risk ;
7 assessment. We came up with numbers again in the low 10-5 8 range. And I think the risk people who did that
.l 9 assessment could talk more about it. One other issue 'l 10' which came up was the effect of leakage on natural ,
i 11 circulation or the thermal hydraulics in the secondary 12 side of the generator. q 13 I understand that in the calculations that !
t 1 l \-- 14 were'done, there was a primary to secondary leakage of 100 ]
i 15 gpm assumed. I didn't know all the details of how that j 16 affected, but -- so, at any rate, I guess the point to 17 make here is that we have, we think, in establishing these 18 performance criteria, addressed potential for tube
- 19. rupture, including multiple ruptures; potential for 20 leakage under postulated accident conditions, design basis 21 accidents in this case.
22 And in the risk assessment that was performed, p 23 we've addressed the potential for not really multiple 24 leaks, because what we saw there was that you're going to l
() 25 go right to a tube rupture if you get any significant l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 2000.x3701 (202) 234-4433 l_
. . -- - . - - - -- . . _ . - .. - . -= - . _ . - - . ,
73 i
1 leakage. So that's the way those issues were addresned. -i i
2 That concludes what I was going to present. I O 3 don't know if there's some questions that I missed.
i 1
4 VICE CHAIRMAN SEALE: Well, I guess we want.to .
l 5 find out what else Joe's going to say. l 6 MR. STROSNIDER: Yes, Joe Donoghue, Division 7 of Systems Safety Analysis, is going to provide some more 8 information.
9 Thank you.
10 MR. DONOGHUE: Good morning. Once again, I'm 11 faced with the same challenge I had on Tuesday, an 12 excessive slides to minutes ratio. So I'll try to go as l l
13 quickly as I can. I think you did ask us to try to r c 14 summarize what discussion I went through yesterday. I'm
- 15. going to leave that and just go through the basis for the 16 approach that the staff is following to address risk with l I
17 the rule. !
18 Now, I just put a couple of bullets up there, l
- 19. and there are some slides, I think in the back of your l l
l 20 handout, that point out the language in the rule that 21 we're using to address the severe accident risk that 22 relate the two function to containment of fission products l
1 23 and an outline of what the steps are in the regulatory 24 guide that we recommend to take to satisfy those things in ,
25 the rule.
I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.I
74 1 What:I want to spend the least few seconds or 2 minutes on'is the basis for-why we followed these -- this O 3 approach. We already-know that the rule is going to allow 4: operation beyond current repair limits. 'When the' staff 5 realized that, and having the mandate to do rule making in' 6 a risk informed manner, we started looking at the 7 potential'for thermally-induced tube rupture.
8 -We've mentioned-0844, how a pressure induced'-
9 - the risk of pressure induced' tube failure _ leading to 10 bypass was assessed previously. This was a new area for 11 us. It wasn't much background for us to start'with.
12 Now we've said this before, and I put it right
! 13 up front. The fundamental objective here for the staff l-l -14 was to maintain the defense in depth posture with respect 15 to the containment function of the tubes. This.gets into-I 16 the discussion of how well or how thoroughly we may have
~
17 considered the release after you have the -- once you have l.
L 18 the release path established.
L 19 We're focusing on a containment function of V l L 20 the tubes, and that's why down here I'm saying we measured j l
l 21 containment bypass potential against the subsidiary safety j 22 objectives. Right away, by doing that, what we've tried I.
l
- l. 23 to do is limit the variables and the uncertainties that go !
I 24 into the analysis of the safety analysis that's behind ;
4
() 25 assessing the severe accident impact.
l NEAL R. GROSS i COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
l (202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 I
.. _- . . . . . - . _ - . - - . . ~ - - . - - .. . .-
75 1 And although there is maybe -- you know, work jg e- 2 can be done, as we mentioned earlier, to try to assess V
3 this maybe more accurately, we thought it was more 4 appropriate to try to focus the problem to a point where 5 we knew we were meeting the objectives and then' satisfying .
6 the rule.
7 And the example of assessment that you heard
- 8. about yesterday shows that we have a bypass frequency 9 somewhat above the objective. But I tried to. point'out a 10 few times as strongly as I could yesterday that that r 11 result was only an example, that the bypass frequency 12 could lie over a range, and there's some plants that~could 13 be above and below that -- the mid to low 10-5 number for f%
14 containment bypass for thermal induced rupture. !
15 But realize there's another contribution.
16 Again, it's 0844 as in the reference here that talks about l
17 -- that lays out the analysis for the bypass frequency for 18 pressure induced tube rupture. That's another 19 contribution here. So from this work, we saw a need to 20 address this vulnerability in the rule.
21 And what we've done in the regulatory guide 22 again is the steps or recommendations that we think will
, 23 meet the intent -- the language in the rule. We've even 24 laid out the option that discusses trying to understand or
() 25 determine the' failure -- the conditional failure NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
76 1 probability for tubes by thermally induced means.
7, 2 We lay out what we have found to be the
!' )
3 uncertainties in that analysis, the problem areas that j 4 people should try to address when they do this kind of 5 work.
1 6 Any questions about at least the background 7 for the basis of our approach? I think it tries to 8 address some of the questions I heard at the end of the 9 discussion yesterday.
10 Now the rest of my presentation was really 11 just summarizing what we went through yesterday, and I 12 know you're pressed for time. It's up to you.
13 VICE CHAIRMAN SEALE: Go ahead.
(~h ,
'(s) 14 MR. DONOGHUE: Okay. I've only selected a few i
i 15 very high level pieces of yesterday's presentation just to ]
i 16 give you a flavor for what was talked about. The details, 17 again, are in yesterday's handout. l i
18 What we tried to lay out yesterday were the i
19 steps that we took to assess where we needed to build up 20 our knowledge on risk, and I've already mentioned that we 21 are focusing on the thermally induced tube failure 22 potential. The rest of yesterday's discussion were the l 23 components that went into that assessment.
24 MEMBER CATTON: Was the enhanced heat transfer O
$ ! 25 due to a leaker considered?
v NEAL R. GROSS
! COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4 33
. - - .. - ~ . . . . - . .. . . _ . -- . . . - . .
77 s
1 MR. DONOGHUE: We did some SCDAP/RELAP l 2 analyses early on at 100 gallon per minute and Sno gallon l
- -.O l
~3 per minute leak rates.
i 4 MEMBER CATTON: That wasn't the question.
5 MR. DONOGHUE: Well, I was just going to say I 1
6 had to go back and look at those assessments. We haven't :
7 included those in the risk assessment, but I have to go f 8 and -- to answer your question, go back and -- j
- . 9 MEMBER CATTON
- Okay.
i 10 MR. DONOGHUE: -- look at those analyses to !
11 see how it was -- how the heat transfer was addressed.
12 MEMBER CATTON: All right.
13 MR. DONOGHUE: And we started by talking about j l O 14 the layout of the event tree and how it'was quantified and ,
i 15 some of the problems that we saw in doing that on a ;
t 16 generic basis, which forced us in some ways to do this on ;
t 17 a -- what I'm calling an example basis. Thermal hydraulic l t
18 analyses, very briefly, was discussed yesterday because I 19 know you've had some opportunity to discuss that i 20 previously.
21 The flawed tube failure model and the flaw 22 distributions that we used, we laid out the -- not only 23 some of the details, but I pointed out that the flawed j 24 tube failure model seems like the highest confidence part l 25 of what we've done so far. The flaw distribution has NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON D.C. 20005-3701 (202) 234 4433
78 1 probably left the most questions open in our analysia.
,-~ 2 And then the tube failure probability
\ '/
3 calculation itself which fed back into the event tree to 4 give us a containment bypass number we discussed. And you 5 know, the numbers we were coming up with for containment 6 bypass again I'll repeat -- low to mid 10-' range for our 7 example plant.
8 And then it laid out -- at least outlined the 9 sensitivity studies that were done on the event tree 10 itself. It's a very recent thermal hydraulic sensitivity 11 studies that were done and some sensitivity studies done 12 using the probability calculation tool to assess the 13 effects of e range of temperatures on the tubes and a
- 14 couple of other variables.
15 And I ended my discussion yesterday just 16 listing -- and this just laid out in some' detail, but not 17 much more detail than this, in the regulatory guide the 18 areas where we saw people would really have to concentrate 19 their efforts if they were doing this kind of calculation.
20 Flow distribution -- again, we see significant 21 places where people could ask questions about what we 22 used. What we also see a need. to do on a plant specific l 23 basis is really hard to do a fit of a representative flow 24 distribution in any single plant.
( ,) 25 The event tree quantification -- there were NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
(202) 234 4 33 WASHINGTON, D C. 20005-3701 (202) 2344433 l
79 1 some qusstions b: cad on the 1150 information that wo us:d I 7x 2 for at least half of that tree quantification. And then i ) 3 we have an understanding that plant specific and design 4 specific features are going to affect not only the 5 quantification of the tree, but it's structure.
6 If a particular plant saw some peculiarities 7 of their systems and their procedures that could affect 8 that tree, that's something we understood people could l
9 include.
i 10 Now I want -- I had some discussions with the 11 staff just in the last few hours. It's probably 12 appropriate to strike the word qualitative there in a 13 sense that there were calculations done of other RCPB
,/ 3
\/ 14 components besides the surge line, hot leg, and the tubes 15 to understand if they were going to fail at these 16 conditions.
17 In fact, SCDAP/RELAP output was used to l 18 understand how valves, gaskets, and so forth might be 19 affected. My point -- by putting this bullet up, my point i 20 was that the next step to understand the probability of j 21 failure of those other components was not taken.
22 And I just want you to understand that, that l 23 what I meant here is we didn't do a quantitative analysis 24 in the sense of understanding the probability of failure
(~3
( ,) 25 of those other components relative to the tubes, surge l NEAL R. GROSS
! COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433 I
80 i 1 line, and hot leg.
2 Okay, and this is the final slide..I'll put up O. 3 for my presentation. We discussed the thermal hydraulic 4' results and we know that plant. design specific factors
'1 5 could affect. thermal hydraulic response under these kind ;
6 of events. Although some analyses were done.on something i
L. 7 other than Surry, which was our example, they were~not i 8 included in a risk analysis that we've done here.
9 'So on a plant specific basis, more detailed l l
10 analyses using those plant specific features ought to be ,
t 11 done. I put up there resolution of modeling' issues.
l L 12 That's been a discussion point with the committee recently 1
l 13 in several other meetings. l r
l
- t l 14 Very recent work that has been done.to address j
~
15 the uncertainties of mixing model parameters.and heat t i
16 transfer coefficients that are used or correlations.that l 17 are used in the code and so forth have just been l' i l 18 completed. And as I presented yesterday, those results l
L 19 have to be assessed to see what their impact is on the 20 overall risk.
21 MEMBER FONTANA: Are you still on design basis 22 space here or are you into severe accident space?
! 23 MR. DONOGHUE: No , I'm discussing severe 24 accident space.
25 MEMBER FONTANA: Okay.
( NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W. l l'
i (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 :
- i
_ _ ._ _ . .. . . . _ . _ . - .m_ ._ . . _ . _ _ _ _ . . . _ _ . . . . _ . . . _ _ ._.
81 1 MR. DONOGHUE: On the two performance model, 2 we knew on the outset --
although I said earlier that was
() -3 probably one of the tightest parts of our analysis -- that 4 there'were some gaps thac should probably be filled in.in 5 some future work, that it does not consider other failure.
1 6 .- modes, and we acknowledge-right up front that this'is a 7 burst calculation.
.8 We don't -- the test results don't tell us how 9 cracks are going to leak under these conditions and what.
'10 those effects-would be. It's also specific only to axial
.11 cracks. I know there's work going on to address the 12 circumferential cracks, but it's not been folded into our-l l
- 13 risk analysis at this time.
14 And then we have to understand that the creep t a 15 failure prediction means that it is necessary to
-16 understand to some degree'of accuracy the heat up rate; 17 not just the'end point temperature and pressure, but the .
18 heat up rate of the tubes.
i 19 Now I'll just end by saying that this is an 20 example analysis that we've done. We say right in the 21 reg. guide we don't consider it to be definitive by any
'22 means, but we think it gives us enough insight to say that l
L 23 there's a risk. vulnerability that ought to be assessed if t
24 we're making the changes to repair limits that we see A
Q 25 being done in the rule.
i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
- 1323 RHODE ISLAND AVE., N.W.
.(202) 2344433 WASHINGTON, D.C. 20005 3701 (202) 234 4433 I .
82 1 MEMBER POWERS: I would say the same thing I 2 said at the subcommittee meeting, that I think it's an 7-('~' /
3 extremely illuminating piece of work. It's illuminating 4 at least to me. But it is in fact not a risk analysis.
5 It is in fact a frequency analysis. And I wondered if 6 you'd given more thought to the question about going on to 7 carry it completely to risk.
8 Before you answer that, I'll just note that 9 the concern, of course, is that people could approach this 10 from a risk perspective rather than -- a licensee could 11 approach it from a risk perspective rather than a 12 frequency perspective. And yesterday, at least some of us I 13 were delivered a paper by some licensees on where they
-) 14 began looking at some of the risk aspects of this problem.
15 MR. JONES: I think, Dana -- Bob Jones from 16 NRR staff. In response to that, we have under way -- you 17 were at the last subcommittee meeting I guess - the PRA 18 general applications guides or risk informed general 19 guides which we'll be discussing with you further. At the 20 current time, our primary emphasis, we believe, is 21 properly done at the defense in depth kind of approach and 22 not the level three risk evaluations.
i 23 I think there is some questions that will 24 probably be something worthwhile seeking during public ,
r~N l
( ,) 25 comment period as to what extent should we allow use of l NEAL R. GROSS l l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., NW.
WASHINGTON, D.C. 20005-3701 (202) 2344433 (202) 234-4433 l
6
83 1 risk. But at this point, I think that would be a bit of a 2 leap for us at this point.
3' MEMBER POWERS: I mean, that sounds like a 4 pretty good idea just to include it as an area that you'd 5 like some public comment on. Because I think it'is a 6 leap, and I think it's a leap that takes place in slow 7 motion. I think there's a lot to do there. But it would 8 be interesting to get public comment back on that.
3 MR. JONES: We agree.
10 VICE CHAIRMAN SEALE: Any other questions 11 right now? I have the sense that we really do need to get 12 the details of Dr. Hopenfeld's analysis that accompanied 13 his differing professional opinion. And we also need to 14 get the information from the staff on the calculations 15 they did as a part of their assessment where they had 100 16 and 500 gallons -- was it, per day?
17 MR. JONES: Per minute.
18 VICE CHAIRMAN SEALE: Per minute, I'm sorry, 19 yes -- leak rate and what that did to the heat transfer.
20 MEMBER CATTON: There's two parts to this, 21 Bob. One is you can easily calculate the enhanced heat 22 transfer coefficient in the tube. What's difficult to 23 address is how much of the hot fluid it's going to suck 24 off from the plenum because they make certain mixing
() 25 assumptions by the way they nodalize this thing.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
84 1
1 And they've tuned it to that, and this is vary l l
7 2 different. I
( 'i 3 VICE CHAIRMAN SEALE: Yes.
4 MEMBER CATTON: How are they going to treat 5 that?
l 6 VICE CHAIRMAN SEALE: And then ultimately, of 7 course, they assess what that might do in modifying the 8 failure times of the other components of the hot leg for 9 the tube or whatever.
1 10 MR. JONES: We have to go back and look at i l
11 that. And I think all of our memory's a little weak on l l
12 it, but my initial reaction of memory says that it didn't 13 seem to have a big effect; otherwise, we would have been
, a, (s' 14 studying it much further over the last eight, nine months. I 1
1 15 So , I don't believe we saw much difference.
16 But we need to look at some of the questions 17 on what happened in the heat transfer, how was it modeled. ;
18 But I don't think it affected the timing much, but we will i
19 go back. I l
20 VICE CHAIRMAN SEALE: I must confess that I l 21 remember -- i l l l 22 MEMBER SHACK: The EPRI analysis says it l 23 drives the temperature up about 20% for 100 gpm leak. !
1 24 MEMBER CATTON: I really would like to know
',) 25 how they know that. That's a complicated mixing problem NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
85 1 at best. It was treated poorly and tuned to the data
- 7. - 2 which you could argue gee, that's okay; but now you come l 1 i
3 along with something entirely different. I One of those 4 tubes with 100 gpm leak or 500 gpm leak is sucking on that !
l 5 lower plenum with a certain amount of strength.
6 That is very different than anything we've i 7 looked at. And the question is, how did they treat it? I 8 don't think they did at all.
9 MEMBER SHACK: Yes, but that only drives you 1
10 to the conclusion that they should have considered this l
11 problem, which they had, in the rule. Not that they l
12 compute the conditional probability very well, but you 13 have to address it somehow. l
/ \
! )
N/ 14 MEMBER CATTON: You come to certain !
15 conclusions based on the analysis that has led to time to 16 failure in this race between the various pieces. This 17 changes the race. Is it important? I don't know. But it 18 sure as hell ought to be reflected somehow. Either write 19 it off with some kiv> af logical arguments, or go take a 20 look and see what it does to you.
21 I don't think you can -- I mean, I think Joe 22 is right.
23 MEMBER SHACK: But see, on a generic analysis 24 side, you know, there are lots of great big details here.
,m (v ) 25 I mean, I think the point to this analysis was to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C 20005-3701 (202) 234-4433
86 1 demonstrate whether there was a problem that ought to be.
! .,-s .
2 considered or not, not to analyze any particular situation .
N.) 3 which has to be'done on a plant' specific basis, i 4 MEMBER CATTON: If 100 gpm is important, then !
L 5
l this should be looked at; or else you have not evaluated 6 it. .And I think it's that simple.
7 MR. JONES: But let me say, Ivan, your 8 characterization that this is one tube pulling 100 gpm.is l .9 very unlikely. What we would anticipate this to be -- ;
10 MEMBER CATTON: Well, that's why you're,doing !
11 risk. Argue it away based on riek. ')
i
- 12. MR. JONES: What I'm saying, from what we've l
-i l 13 seen and what I remember -- and I think the materials guys l
14 would have to do a better job of this than I can -- is y 15 what.we're looking at to get 100 gpm is an integration of
\
' 1 16 lots of flaws with little leakage.
17 And so, I' don't think you have the large flow l
18 differentials between one tube and the bundles of tubes 19 you're thinking about as a general rule. I mean, I'd have l 1 20 to think some mcre about it, but the characterization that i
21 it's one tube, I just want to say that's -- we would j l
22 consider that to be very, very unlikely.
23 MEMBER CATTON: And that's fair. And I think I
24 that's a valid argument not to have to consider it. But I j
) 25 think you should make it.
4 NEAL R. GROSS j COURT REPORTERS AND TRANSCRIBERS L
1323 RHODE ISLAND AVE., N.W.
L (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433 l'
4 .__..-.~....__.__,...,~._.._.___..__.-_____...-...,_m._
87 1 MR. JONES: I agree. I don't disagree.
2 MEMBER CATTON: You know, I just did a quick O 3 calculation. I get a Reynold's number close to a million l
( 4 in the tube with that kind of leakage.
'5 .VICE CHAIRMAN SEALE: 'I-think under the- .!
6 circumstances, it might be worthwhile for-us.to take a 7 short break. I know, and then hear the industry after the 8 break, if that's all.right with you.
9 CHAIRMAN KRESS: We're supposed to break at 10 10:45. How long is this going to take?
11 MEMBER CATTON: Are we taking a break?
12 . MEMBER FONTANA: We don't know yet. ,
1 l
13 .VICE CHAIRMAN SEALE: All right, I'm sorry; j-
['
r-r
- 14. I've been overruled. Let's go ahead and ask the industry 15 people.to make their presentation. I'm sorry. I'm sorry, L-I l
- 16. we have another thing coming up that requires that we keep 17 the original timing, so go ahead.
- 18 MR. PEARSON: I'm Richard Pearson from 19 Northern States Power Company -- a steam generator
'20 engineer at the Prairie Island' Nuclear Plant. -And I'm 21 here to kind of summarize our position and try to respond 22 to the couple of questions on what the industry's position
- '23 is on the rule.
24 And we put some of this stuff together last 25 night and other meetings were done. Before I start to get NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 3 1323 RHODE ISLAND AVE., N.W. i (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433 l
. .-. - . - . - .~ . . . . . - - _ - - - . -. - . .
i 88
' ~
1 into details on this and we're looking at trying to changa {
2 limits and looking at relaxation and -- Prairie Island is .!
)
3 in a farming community, and sometimes people want to know 4 how I treat nuclear power. I like to use the gentle bull 5 syndrome.
l 6 When-I was growing up on a dairy farm, a bull 7 was a necessity. But the farm journal every year had an ;
8 article about farmers who turned their backs on gentle h
9 bulls and were maimed or killed. And so, I just use that !
1 i
10 analogy that yeah, we can control the gentle bull, but we 11 can never forget what we're dealing with.
12 .And so, we are locking at some changes here in' !
13 regalation. We, the industry, are not forgetting what
[D
, \~ / 14 we're dealing with. i 15 The industry need has been for some relaxation !
16 in the repair limits on steam generator tubing. In 1993, 17 we presented.an option to Mr. Russell and to the NRC on 18 how to perhaps change the regulatory framework in order l
19 that we could implement the concept of steam generator-20- degradation specific management.
21 We were looking for relief obviously on the l
1 \
L 22 repair criteria. We were looking for timely relief; and j l
23 timely, of course, is a relative sort of thing. But we i 24 were concerned both with the industry resources and the
() 25 NRC resources and had proposed a lead plant group concept NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
. 1323 RHODE ISLAND AVE., N.W.
[ (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 r
89 1 at that time to try to implement the voltage based
-7 ,s 2 alternate repair criteria.
\
3 At that time, the concept of rule based 4 regulation for steam generators was proposed. And the 5 industry's response to that is that we neither agree nor 6 disagree with that. Industry today would not feel bad if 7 the rule went away. I think that may have been -- may not 8 have been clear yesterday that -- who was pushing for the 9 rule.
10 We, the industry, were not pushing for the 11 rule. We were pushing for some change in the regulatory 12 environment. However -- and along that line then, the 13 industry was looking for the flexibility to determine
_/ 14 their own repair limits.
15 Industry would kind of like to have t
16 technically based criteria, and we look at the voltage 17 based criteria that's kind of an arbitrary voltage limit 18 set, whereas the industry has presented a technical basis 19 for what the limit could in fact be.
20 And so -- now, that's kind of the industry j 21 position then on the rule situation. Now, we can't ignore 22 what's happening, and we are in fact working with the NRC l
l 23 and have had meetings with them on the issue of the steam l l
24 generator rule. Our vision for how that would be
/^N t
) 25 implemented is similar to what the NRC has been discussing NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
. - _ . . _ . . _ . . . _ . _ _ . _._ ._ _ _ _m m _ __ _ _ _ -.. . _ . - _ . . - .
(
- 90 l 1 and maybe the'1evel of' detail is different. j i
i 2' But if there's a rule and there would be a {
) !
, 3- regulatory guide, we would want the regulatory guide to I
4- endorse an industry document which:would have all the l
- j. 5 details in it similar to what the maintenance rule'had !
6 done. We would prefer that the industry document have :
i 7 methodologies endorsed in it that would allow us to= .
l 8 implement the procedures needed within industry.to ensure i
9 steam' generator tube integrity, and they involve several i 10 procedures and documents that.are already out there, but 11 they are the steam generation examination guidelines. :
i 12 There's some topical report -- there's one out !
13 there on the OD stress corrosion. cracking at the tube !
(~h i 3s / 14 support plates. There's a procedure being -- out there on 15 the data base to support that and the protocol ~ assigned 16 with that, and we've been working with the NRC on how this 17 data base can be updated and whether we can update it-18 routinely without their prior review and approval, 19 assuming we do it within the methodologies that have been 20 agreed to.
-21 And so, ideally, we would like the review to 22 stop at this point from a pre-approval standpoint, and we 23 implement here; and then, of course, the NRC always has 24 the option to come in and inspect how we are implementing.
() 25 NEI has put out a document -- I believe it's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4 33
r 91 1 been sent to the NRC -- on performance based regulation,
,- 2 and that document is NEI 96.04. Some highlights from that t
%~')
! 3 document is that in performance based regulation, the 4 regulatory oversight would be commensurate with the degree l
5 of achieving the performance criteria.
6 As long as the plant is meeting the 7 performance criteria, the regulatory oversight would be 8 minimal. And obviously, if you don't meet the performance 9 criteria, then you would expect additional inspections.
10 Along with that, we'd like the NRC to kind of approve and 11 tell us how we're going to do something, and the industry 12 would figure out what to do; and the industry would put 13 together the detailed implementing procedures, and the (3
14 responsibility then applies to the industry for properly 15 doing that.
(
16 And that's a great responsibility. And l l
17 sometimes you might say it's a great risk, because you 18 could be inspected on a routine inspection and find out 19 that you really messed up in your implementing procedures.
20 But ideally, that's the way we would view performance 21 based regulation.
22 We would like to be able to develop the repair i 23 limits based on pre-approved NRC methodology, and that we 24 could then implement without prior approval. Again,
( ,/ 25 mention that there's some risk involved in that. And this l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
92 1 obviously is one of the high level issues involved with 2 the steam generator rule.
()
3 Let me skip to the next slide just to kind of l 1
I 4 talk about that a little bit more. Or skip two' slides, i l
5 rather, to what I call the plugging criteria box. And !
6 that's been alluded to by the'NRC. As an example, in the 7 operaticnal assessment, as it's kind of written right now, 8 it ideally requires the sizing so that you know how much 9 your degradation is growing in order to predict where it's 10 going to be at the end of the next cycle.
11 Sizing as required by the draft reg. guide 12 requires a validated technique which we've given you some 13 detail on yesterday. And without validation of the (m- 14 technique, you need to plug on detection. If you're !
15 plugging on detection,.you probably don't have growth rate 16 and an individual given degradation mechanism. l 17 If you don't have growth rate, it becomes l 18 difficult to make the operational assessment. It almost 19 maybe becomes impossible to develop an alternate repair 20 criteria, and it could lead you to an extreme case of no l
l 21 cycle length -- zero cycle length because you can't 22 predict growth.
23 Now, generally, owe can say if we didn't detect )
24 it one cycle and then we go on to the next cycle and find
- 1) 25 it, that it must have been okay for that cycle. But the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433 l
- _ ~ - . . . _ . _ - - .. . . ~ . . - . .. . .. .. . ---
i 93 l 1 questions do periodically come out and say well, if you 2 didn't protect it the first cycle and now you find it, how O 3 do-you know it didn't all grow in the last month? ;
1 i
4 And we don't without inspecting every month. j
)
5' And so, there are some things here that require judgement
.6 and maybe some leaps of faith a litele bit'and become very 7 difficult to implement if we're looking at trying to-L 8 implement the draft reg. guide in what I' call the spirit j
- 9 of e.xcellence versus the minimum requirements.
L 10 If you're under a performance regulation where j l
L 11 prescription was not there to plug on detection precisely,. 1 1
12 you could construe a scenario where somebody's trying to l
13 develop growth rate as saying well, I've got ten of these
\~ / 14 indications that are really small; maybe I've got one tube ,
l 1 15 pulled, so I have some idea of what small means; maybe ,
t 16 making some judgements based on available other data, and )
l 17 there's getting to-be a lot of data on what eddy current I
l 18 does in fact tell us, but it doesn't meet the rigorous 19 requirements.
i 20 Would a plant be willing to take the risk l- under performance based regulation of, for example, i L
22 leaving in ten of those cracks -- and I think these things 23 have~been proposed by others sometime in the past -- in
- i
- 24 order to develop the growth rate? And could a plant
() 25 develop some sort of risk of leaving those ten cracks in NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
! - _ _ l
94 1 service, for example, in order to develop that growth 2 rate?
/7 ,)
'~'
3 And, would a plant manager risk his career to 4 do that sort of thing? But under a true performance based 5 regulation, perhaps that's an option that would be 6 available to a utility.
7 The other difficult .hing about growth rates 8 is that the eddy current has gotten extremely sensitive 9 compared to how it used to be. The analysis has gotten a 10 lot better and everybody's doing a dual analysis using two 11 teams. We know there's still problems with that, and 12 we're investigating ways to even improve that performance.
13 But we're losing the ability to say well, it's uJ 14 here this cycle; let me look at last cycle's inspection 15 results and see if I can find some of them that were there 16 that weren't called. And that ability is slowly going 17 away from us, and so some of that information that might 18 have been used in the past for growth rates is slowly 19 disappearing.
20 So that's one of the issues that is, I guess, 21 a high level issue in our mind. Okay, let me go back to 22 the slide before that then which was severe accident 23 issue.
24 MEMBER SHACK: Do you think that's a decision
,a
) 25 that the utility should be able to make on its own?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
95 1 MR. PEARSON: This one about leaving some (g #
2 cracks in service that aren't adequately -- are rigorously 3 sized? Under performance based regulation, as NEI has 4 proposed and the industry has proposed, yes.
5 MEMBER FONTANA: I would think a performance 6 based regulation ought to allow the acceptance criteria at 7 a level that, if the criteria is violated, you don't 8 impact the health and safety of the public. In other 9 words, they've got to be two steps away from it.
10 MR. PEARSON: I agree there; yes, yes.
11 MEMBER FONTANA: And the question is, is there 12 a mechanism of reasonably high probability where cracked 13 tubes could lead to failure propagation under release to e i f
O 14 the environment -- significant release to the environment?
15 And if that is not a hi gh probability -- if 16 that is not a high probability, then one would say sure, 17 go ahead and leave cracked tubes in there. But it is 18 something to be concerned with, and then you can't do it, 19 I think.
20 MR. PEARSON: That's right. Under the current 21 regulation, you can't. l 22 MEMBER FONTANA: Or even in a performance 23 based regulation.
]
24 MR. PEARSON: And as is proposed by the draft
/N
() 25 reg. guide, we can't.
NEAi. R. GROSS COURT REPORTERE AND TRANSCRIBERS 1323 RHODE ISL1ND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
.. . -- - - . - . . _ . - ~ - - . - -. .-. ._ .
96 1 MEMBER FONTANA: Can't. !
.- 2 MR. PEARSON: And maybe that's one !
' \ !
3 prescriptive item in there. I guess we in general agree i
4 that there's lots of detail, and how we implement that ;
5 detail is up to us. But there are lots of details. That :
6 plug in detection you could call a very prescriptive item- i 7 looking at it'from a performance based standpoint.
2- 8 MEMBER SHACK: But in a more general sense --
- 9 I mean,.when you have performance criteria that are not -- ;
. l 10 you know, the yardstick is kind of a stretchy, rubbery '
! 11 thing, do.you really believe that you cannot have the NRC
- 12 review the methodology that you used to -- how you're !
4 3
- 13 assessing whether you're meeting those performance 0
- 14 criteria? .
l 15 MR. PEARSON: We would like agreement on the ;
16 methodology. And maybe we don't agree on what methodology 17 means, right. We think the performance criteria like the 18 structural limits are conservative. We have these 19 factors, you know, of safety in there of three times 20 normal operating pressure or 1.43 times the main steam 21 line breaks. l 22 So, if you've approached one of those things, 23 you've approached a criteria; you haven't approached the 24 actual failure of the tube. ,
25 MEMBER SHACK: No, but just how you compute NEAL R. GROSS !
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
97 1 those then demonstrate that you're meeting those criteria f~s 2 involves an awful lot of steps mostly in the NDE that are
( )
3 not, shall we say, transparent. j 4 MR. PEARSON: That's correct. We're very l
5 dependent on NDE to tell us what's there. And we're very !
6 dependent on developing correlations where we don't have 7 them to give us the sizing capability or -- you know, 8 we're not necessarily going to size the indication, but 9 we're going to have a correlation between the eddy current 10 parameter and the structural integrity of the tube and the 11 leakage integrity of the tube. j 12 And I don't know that we're complaining -- no, i 13 I'm not complaining about the details the draft reg. guide k-) 14 has there except for the supplemental performance 15 demonstration.
16 MEMBER SHACK: Well, that may be a matter of 17 interpretation.
18 MR. PEARSON: Right, yes.
19 MEMBER SHACK: Well, again, you think that 20 they shouldn't have the opportunity to review that 21 correlation when you develop it?
22 MR. PEARSON: The NRC always can review 23 whatever they want to. I guess the --
24 MEMBER SHACK: Should review.
iN! 25 MR. PEARSON: Should review? Ah, yes. Should l
- NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
i (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
98 1 they review it before you implement it? Under performance 73 2 based, as we would try -- an ideal case; no, they wouldn't I !
\ /
3 review it before implementation. Would a plant want to 4 take that risk? You'd have to talk to the individual 5 plants to gain that insight.
6 VICE CHAIRMAN SEALE: Okay, let's move along.
7 MR. PEARSON: Okay, severe accidents, as I see 8 it, are put into design basis by the draft reg. guide and 9 rule. That's a big policy issue from the perspective of 10 the -- well, from the NRC and from our standpoint. Very 11 big step -- the industry right now disagrees with that 12 kind of step.
I 13 The industry feels that severe accidents ought
( ,
i 1
'd 14 to be handled in the severe accident management space. It 15 looks to us that you could interpret the draft reg. guide i
16 as it is right now that every utility has to address 17 severe accidents for steam generators regardless of 18 plugging criteria they're using.
19 We're not sure that that's the intent, but it 20 can be read that way whether they have brand new steam 21 generators or old steam generators. And so that's one 22 clarificatic;. that we would certainly want -- utilities 23 would want from that perspective. We understand as all 24 the discussion has been going on, there are large (q,! 25 uncertainties dealing with trying to quantify severe i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIDERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
99 l :
1 accident effects. !
2 And we and the NRC acknowledge that the non-3 degraded tubing will fail under severe accident conditions
, 4 or.some probability of that. And we do have severe ?
5 i accident management guidelines going in place for. steam-L 6 generator. tube failure under severe accidents. And it was e
1 7 just kind of funny when I talked-to our severe accident ~
8 person -- their engineer at the plant about well,lwe're .
s
~
9 looking at trying to determine whether tubes fail-under -
i !
.10 severe accidents or not,'and we think maybe they don't.
11 And he said well, you can't do that;.we're l
12 already planning on them failing. So, the industry is i r
13 cognizant of steam generator tube failure under severe O- 14 accidents with the result in containment bypass.
i l >
15 That's our position on severe accidents. It's ,
l 16 a big step to put it into design basis which is what it f i
17 looks like in the draft reg. guide and the rule, and'it I L
18 should be handled under severe accident management space. :
1 19 CHAIRMAN KRESS: Which part of the design :
- 20. basis do you consider to be a severe accident part that 21 they've put into the rule?-
22 MR. PEARSON: The fact -- well, to use an t
23 alternate repair criteria at the plant level would be
! 24 considered a nadification, plus the licensing l C
/"~ !
25 requirements. When you do modifications, you're required ,
1 a
f
! NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 2344433 l
I
100 1 to assess all design basis accidents. We're not required
,, 2 to assess severe accidents when we do modifications at the
( )
\'
3 plant under the current environment.
4 CHAIRMAN KRESS: And you're saying the new ;
l 5 rule requires that?
6 MR. PEARSON: The way the draft reg. guide is 7 written, yes. Because there's a couple of alternatives.
8 Because one, if you don't meet -- if you're less than the )
1 9 10-' value for challenging the steam generator tubes for l
l 10 induced tube rupture, then you don't have to go into the l
l 11 very detailed analysis that we in the industry can't agree 12 on what's accurate and what's good enough and that sort of l 13 thing.
i p)
\_/ 14 And we can't even agree on what the acceptance 15 criteria are, let alone what are the right assumptions and 16 what are the uncertainties. But yes, somebody at the 17 plant looking at that would say I've got to do a severe )
l 18 accident analysis for steam generator tube rupture for a l 19 new alternate repair criteria.
i 20 MEMBER FONTANA: Of course, some day on a !
21 total risk based environment, it really wouldn't make a 22 distinction. They'd do the assessments and decide to cut l
23 off what you're going to address and what not address on 24 the basis of what the risk is.
(-,
25 MR. PEARSON: Right, I agree.
(~.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
- . - . - . . - . - . . . - - - . ~ . . ~ . .- .. - _ - . . - - . - .- --.. .. . _ .
)
101 ,
'- 1 1 MEMBER FONTANA: But you're not there now. l 1
, 2 MR. PEARSON: We're not there now. At the 2
( 3 plant level, we do in fact use risk assessment to make j
-4 decisions, whether it be what we call the probability risk l l
5 assessment for aux. feedwater -- it was done at Prairie !
l 6 Island ten years ago and they made hardware modifications )
l 7 because'of that risk assessment, j l
l l 8 Or, the IPE's that have been done and we've .
l l
9 made some operating procedures changes already because of l l
10 the results from those assessments. So, yes; in fact, the {
11 utilities are using risk information to make the changes.
12 The last slide here was the --'what I like to I 13 call resolvable issues. And we'll have to resolve the O
\~ / 14 other ones out of the -- but maybe lower level' issues that
- 15 we think are resolvable at the technical level.
16 We've had a discussion on the deterministic l 17 assessment of safety margins and specifically looking at 18 the arena when you do have burst correlations available to 19 you and that we're trying to remove at that point the 20 three times delta P requirement.
l 21 We've given you a basis for the conditional 22 probablistic criteria from our perspective on the l 23 probability of steam generator tube rupture during a main 24 steam line break. We've been working with the staff --
() 25 some of our consultants have on certain levels associated 1
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
..,.,m
102 l 1 with in particular the deterministic critoria and how to l
,s 2 keep from stacking them all up.
i
) ;
3 And we, the industry, are looking at, you 4 know, how would you have met the rule for what we consider 5 -- or used to consider any way -- a very simple 6 degradation mechanism called cold leg thinning which is a 7 wastage in one single spot or -- we're doing some work on 8 that to see what's involved in meeting all the draft reg.
9 guide requirements for that kind of issue. I 10 The ISI remains a big issue; has been for l 11 years. We've been making improvements. The concept of 12 the validation versus qualification is a big one for the 13 industry. An extreme case becomes cost, dose, time
( )
x/ 14 intensive if you wanted to meet the spirit of what I call 15 excellence or standard of excellence for validating a 16 given eddy current examination at your plant.
17 And so, the ISI community is working hard on i
18 trying to find some middle ground with that; and they've 19 got a meeting planned, I believe with the NRC to discuss 20 that. The dose assessment -- we had presented to the NRC 21 alternative to the iodine spike issue. The mystery factor 22 is -- Dr. Hopenfeld called it -- we did some really good 23 work there.
l l
24 Some questions came back, I guess, from the t
k_) 25 contractors. We, the industry, have not answered those.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., NW.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
103 i
! 1 Mainly we haven't answered them because we ran out of i
I 2 money, and we weren't sure how important they were. Our l 7-s s b 3 people will be going back to look at the draft reg. guide r 4 as it's now itemized and see whether we want to pursue f l
5 that anymore. That needs further review.
6 Okay, to summarize, the rule -- well, when the :
7 utilities look at it, it looks onerous. There are 8 additional requirements there above and beyond, to some i 9 extent, what we're doing. Those are additional 10 requirements. . Expectations are changing. A-lot of what 11 you've heard about -- particularly in the area of 12 condition monitoring and operational assessment is being 13 done by the utilities. :
A
- 14 I don't know whether it's being done because 15 that's the new expectation, or we're afraid of '
16 enforcement, or we think it's good engineering practice. ,
17 As it's laid out, those assessments make sense to us, and 18 we're doing them much more rigorously than we have in that 19 past. And so, a lot of the stuff that the rule is going 20 to put in place -- would put in place is in fact now 21 happening.
22 And so there's quite a few t'aings we do agree 23 with the NRC, and I just wanted to itemize some of the big 24 issues that we still disagree with.
i
() 25 MEMBER FONTANA: Does everyone here know what NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
- r. .- , _ - .. . . .- .,_
, 104 l I
1 the iodine spike is? Because if they do, I'll find out
-s 2 off line.
' < s i
' '! 3 VICE CHAIRMAN SEALE: I think Dana can give us 4 a short lecture on iodine spike.
5 MEMBER POWERS: Yes, I guess I understand what 6 the iodine spike is. j 7 MEMBER FONTANA: Okay, tell me later then.
8 MEMBER POWERS: Okay.
9 VICE CHAIRMAN SEALE: Any other comments or i
10 questions? j 11 MR. PEARSON: I guess I've got one comment 12 from Clive from NEI.
13 VICE CHAIRMAN SEALE: Yes, sir. Please come
'(_// 24 to the microphone and identify yourself.
15 MR. CALLAWAY: Yes, Clive Callaway with NEI.
16 Yes, I'd like to respond to Dr. Shack's question about the 17 appropriateness of the NRC's review of the correlations.
18 What we have proposed is that we would develop an industry 19 document that would lay out in sufficient detail the 20 methodologies we have used to develop the industry's 21 repair limits.
22 So the correlations, while maybe not directly 23 reviewed by the NRC, they're going to be based on one pre-24 approved performance criteria. And also, they will be
/~N
( ,) 25 based on an endorse methodology which will be captured NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
105 1 within a generic industry document.
,s 2 Now, I --
('~')
3 MEMBER SHACK: Now was that the industry 4 document that's the parallel to the reg. guide or a 5 different document?
6 MR. CALLAWAY: Right. Well, the general 7 methodologies that would be endorsed by the NRC would be 8 captured within that parallel industry document. Now, the 9 industry document will indeed -- could reference detailed 10 handbooks that would really lay out the how to's of how to 11 -
develop these correlations. i 12 And with the ODSCC effort in 95-05 we've used f 13 as a model, we are d.eveloped pre-approved protocols for V/ 14 the data bases that would develop these correlations. So 15 it's not like the industry is going out without guidance 16 from the NRC how to develop the repair limits.
17 But under a performance based rule, what the 18 NRC should be concerned about is meeting the performance 19 criteria. That's the focus of the regulatory oversight 20 would be the outputs of our programs. And if you look at 21 the regulatory guide, it is very prescriptive in telling 22 us how to develop the data which we need to analyze our 23 tubes.
24 And right now, there is a delta between where
( ,) 25 we are with our ISI guidelines and where the staff wants NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
106 1 us to do. And I think we have to resolve that delta ao 7- s 2 reasonable men in order to come to an agreement that the
( I
( /
3 quality of data which we receive from ISI inspections is 4 good enough to develop the appropriate data bases.
5 So, once we resolve that -- that is, once we 6 gain the confidence of the NRC that we are indeed having 7 the quality of data, then we can go on to develop the 8 correlations based on endorsed methodology that will be 9 captured within an industry document.
10 VICE CHAIRMAN SEALE: Thank you. Does that 11 help you? Any other comments or questions? I think -- we 12 want to think you very much for your presentation.
13 Appreciate it. I think this -- I think this pretty well
- t. 1
's > 14 winds up the program. We gave you a list of things, Noel, 15 that we would like for you to get for us.
16 I think we still have some issues that have to 17 be resolved before we can come to a determinacion in a 18 couple of areas. I will ask the members of the committee 19 to think about what it is we would like to put in a letter 20 if we're going to write it, and hopefully today we'll get 21 back to that so we can begin to prepare.
22 And I'm going to now recess for 15 minutes.
l 23 Thank you.
24 (Whereupon, the foregoing matter went off the
() 25 record at 10:58 a.m. and went back on the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., NW.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
107 l' record at 11':16 a.m.)
- 2 CHAIRMAN KRESS: Let's get started again.
3 The next item on our agenda is the risk-based 4 analysis of reactor operating experience. George 5 Apostolakis is our subcommittee chairman on this. He
, 6 isn't here, and I wasn't at the meeting, the subcommittee 7 meeting, so I'm going to turn the_ running of this part 8 over to you, Bob, if that's all right.
9 VICE CHAIRMAN SEALE: Okay. We had a 10 presentation, whenever that was --
11 (Laughter.)
12 -- Tuesday actually.
13 MR. BARANOWSKY: October 30th.
14 VICE CHAIRMAN SEALE: Yes. And we received a 15 briefing on AEOD's program.
16 I have just a very few comments. The big 17 concerns that appear before the subcommittee at the time 18 had to do with both the origin and the use of the common 19 cause failure database in what we're going to hear about, 20 some questions abou* proprietariness, and so on. There 21 are also questions about what would be required to apply 22 the database, which is generic in the way in which AEOD l'
23 does it, to the particular characteristics of a given l
- 24 plant.
25 There are other, more detailed questions, but NEAL R. GROSS court REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON. D.C. 2000 # 3701 (2m) 2344433 L:
108 1 I'think among others those were two that really stood out 2 as being substantial, and things that the committee might 3 be especially interested in.
4 With that introduction, I'll turn the floor 5 over to Pat Baranowsky.
6 MR. BARANOWSKY: Thank you. I'm Patrick 7 Baranowsky, Chief of the Reliability and Risk Assessment 8 Branch in the Office for Analysis and Evaluation of 9 Operational Data.
10 For the benefit of committee members that 11 weren't at any of our subcommittee briefings, or those of 12 you that are new and haven't heard us talk about.these 13 activities before, I'm going to give a little bit of 14 background. Then, I'll describe our plan forlthe risk-15 based analysis of reactor operational experience.
16 That will include discussion about the 17 accident sequence precursor program, system reliability 18 study efforts, the common cause failure database and 19 software, and then I'll spend a little' bit more time 20 talking about risk-based performance indicators, because 21 that's an activity that is beginning to get started now,
-22 the development of risk-based performance indicators, and 23 it's one that the Commission has_been interested in and 24 would eventually like to hear the views of the ACRS on
() 25 this. And certainly we want to make sure that we're at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 2344433
109 ,
1 1 least starting off in a direction that we all agrea upon. i 2 We'll talk a little bit about what we think
- \
\ / 1 3 the future interactions could be on these topics, and then i 4 hopefully we can discuss an ACRS letter. 1 5 I think it's worthwhile to recognize what the 6 mission of AEOD is, and I've put together a few bullets on l
7 this chart which describe the overall mission of AEOD with 8 respect to the collection and analysis of operational 9 data, and then a few more bullets that are specific to the 10 Reliability and Risk Assessment Branch. Some of the key 11 words here are AEOD performs independent assessments, it 12 analyzes and evaluates operational data, it collects the 13 information and disseminates it. l
\ l K/ 14 The Reliability and Risk Assessment Branch has 15 an overall mission associated with basically three areas.
16 The first one is to basically analyze historical trends 17 and patterns associated with reactor operational data. l 18 The second one is to extend that analysis to look at some 19 plant-specific implications and performance associated 20 with analysis of that data. And then lastly, my branch is 21 responsible for identifying data sources, data collection, 22 databases, to support the agency in general and activities 23 associated with the first two bullets that I identified 24 that my branch is responsible for.
()
,m 25 In our prior briefings, we talked to the NEAL R. GROSS
, COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (N2) 234-4433
. . . _ . - - . . . . . . .. .. - - . - . - . ~ . .. -
110 1 Commission -- in 1995, in 1996. They were particularly
,- 4 2 interested in how we might take the kind of risk-based
'L] 3 analyses that we were doing, particularly with system 4 reliability analysis and accident sequence precursor work, 5 and develop that or evolve it into some sort of 6 performance indicators.
7 So after our last meeting, the Commission ,
l 8 specifically asked that we go to the ACRS and have them, l l
9 the ACRS, take a look at our program and our ideas on 10 where we might be going with risk-based performance 11 indicators. And so that got us really into the last two l 12 subcommittee meetings that we had. One was in July and
- 13 the other one was about a week ago in October.
D 14 Prior to that, we had a couple of short, half- j 15 hour to one-hour briefings of this committee on different 16 aspects of our program, but I don't think they came i
17 anywhere to being as comprehensive as the discussions that 18 we had with the last two subcommittee meetings. l
! 19 Prior to those subcommittee meetings, we had 20 provided a fair number of reports. I've listed them --
l 21 VICE CHAIRMAN SEALE: One minute. I want to E 22- just make sure I caught you on one phrase there. j l
l 23 MR. BARANOWSKY: Okay.
, 24 VICE CHAIRMAN SEALE: You did say " risk-based r
25 performance indicators."
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433 I
. . ~ . . - . . - . - -- . . . . _ - . . . . ~ . ~ . . - _ . -
111
-1 MR. BARANOWSKY: Yes.
2 VICE CHAIRMAN SEALE: Okay.
't 3 CHAIRMAN KRESS: Good things.
4 VICE CHAIRMAN SEALE: But they're not 5 informed. )
6 CHAIRMAN KRESS: Risk-based,- he said.
7 MR. BARANOWSKY: They're informed, too, I l 8 hope.
9 VICE CHAIRMAN SEALE: I'm slow this morning.
10 MR. BARANOWSKY: Okay, i 11 MEMBER POWERS: You prefer that to core damage 12 frequency informed.
I 13 MR. BARANOWSKY: I like it because=it's l
, 14 shorter and easier to say.
15 (Laughter.)
16 MEMBER POWERS: But less accurate. Much, much 17 less accurate.
\ \
l l
18 MR. BARANOWSKY: Okay. But we did provide the l
l 19 ACRS with quite a few reports, a number of analysis 20 reports on system reliability studies, the common cause 21 failure data system had six volumes to it, the accident 22 sequenca precursor reports. There were also several 23 methodology reports, some aspects of which were discussed
! 24 in our July subcommittee meeting. And then we provided 25 our overall plan of what we're trying to accomplish in the NEAL R. GROSS
. i COURT REPORTERS AND TRANSCRIBERS ]
1323 RHODE ISLAND AVE., N.W. '
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
=
112 1 risk-based analysis of reactor operating experience. I 2 This work is also described in the PRA .
O 3 implementation plan under items 3.1, 3.2, 3.3, 3.4,.and 4 3.5. Basically,'you can see that these topics line up i
5 fairly well with the sort of mission objectives that we 6 identified in the earlier viewgraph.
7 The plans are conceptually stable at this 8 point, but they're undergoing tweaking in terms of the l
9 phasing of the work and scheduling, and so forth. So 10 we'll probably have a revision to our program plan in a
- 11. couple of months.
12 The next couple of viewgraphs I hope will put 13 some additional context to our program. One of the 14 important things is that we're trying to base the methods 15 and the applications that we're performing on a practical 16 ability to collect and analyze data, so that we're not !
i 17 research oriented, although we will take methods that were f 18 developed through the Office of Research or elsewhere and 19 modify them or develop them a little bit more, so we can 20 apply them in our projects. But we're not trying to come 21 up with whole new methods that people haven't thought 22 about before.
23 Most of the methods that we're using are the i 24 kinds of things that have been applied in a number of PRAs j
() 25 or reliability types of applications in the past, and many NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
p 113 i I
1 of them.have been used for regulatory purposes.
2 Also, I think it's important to note ~that we !
3 have put together a program that we think is based on a 4 reasonable decomposition of risk into its constituent j 5 elements, especially those elements that line up well with ;
l 6 PRAs. And this next chart shows the conceptual-way out of j i
7 our program. !
8 As I mentioned,-we look at both industry f i
9 trends, plant-specific trends, and just analysis of data i 10 that feeds into those trends, and so on this chart what :
11 you see at the top is if you summed all of the risk, of I i
l 12 course, _ from all of the plants, in theory you would have ,
l i 13 an industry risk profile. And each plant's risk could be l TN
\/% 14 broken down into elements, which we've broken down here I
15 pretty much the way that PRAs break the risk elements down 16 into containment failure probability, core damage 17 frequency, and health effects.
18 Now,-most of our work is oriented toward 19 analyzing aspects of risk related to core damage 20 frequency, although we will be looking at some elements of l 21 containment system performance in the future. Exactly l
- 22 what is practical to look at at this point I don't know, !
i 23 but we do know that we can look at core damage frequency a l
24 little bit easier because there is data available, and
() 25 models that are a little bit more tractable to work with. l NEAL R. GROSS i COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 - WASHINGTON, D.C. 20005-3701 (202) 234-4433
, w
_ _ . . . _ . _ _ _ _ _ _ . . _ _ ._ - . _ _ . - . . _ _ . _ . . -. . , . ~ , _ . . . - - . __
114 )
i 1 And the elements of those models are shown in i 2 the last'line of'this chart, in which we look at ;
O -
V 3 initiating event frequency and safety system reliability. I 4 Common cause failure probability, of course, feeds into ;
)
5 safety system reliability. But when one does a PRA, in !
l 6 many cases it has done sort of an adjunct to the safety -
7 system reliability analysis, and the same'could be-said of I 8 the human error probability, j 9- So we can take this kind of a logical breakout !
1
)
10 and look at both vertical and horizontal. cuts of the t I
11 operational data to learn about individual components and ;
12 systems or about plants and industry risk characteristics, j 1
'13 as would be evidenced by the data, the operational data,
]
14 that's available. r 15 And thus, we have put together a number of !
16 specific programmatic activities -- the accident sequence.
l 17 precursor program, which I'll diccuss-next. No , I won't 18 discuss that.next. I will discuss that shortly.
19 Initiating event frequency evaluations, including the l 20 frequency of some special concern safety issues like i
l 21 service water system events and fire event frequencies, a 22 number of reliability studies, which I will talk about 23 later also, common cause failures.
24 We have performance indicator work, which is 25 both risk-based and traditional. The risk-based is under NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
_ .- -4
- - . . . _ - = . -
I 115 i i
i development. A number of data systems, which I'm not !
r 2 going to talk much about, but include LERs and other f l
\.sl 3 reliability. data. And yet to be developed is a human ;
i l
4 performance element, which we'll be working on over the l I
5 next six to 12 months. But the framework for performing !
l 6 human reliability analysis is a little bit more up in the !
l I
7 air, less agreed upon if you will amongst the PRA j l
8 community, and wa're looking to take accepted state-of- i 4 the-art approaches and apply them to the analysis of the 1
10 data as opposed to doing the fundamental research to l 11 figure out what is the proper framework for doing 12 analysis. t 13 So I think the PRA community is coming to
\ 14 grips with this issue, but I'm not sure whether we're at a 15 stage now where I can categorically say we'll be able to 16 come up with a specific activity related to this.
17 Although in the context of risk-based performance 18 indicators, which I am going to discuss next, I can talk 19 about how we can derive some human performance insights.
20 The performance indicators, it's important to 21 recognize, are one element of the NRC's assessment of j 22 licensee performance, and it's usually an element that is 23 used in support of the senior management meeting process. l 24 The senior management meeting process involves a lot of !
. () 25 inputs, performance indicators being one, but things like l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234 4 433 i
- . . . - . . . . - - . . - - _ - - . . - - - . - . . _ . - . ~ _ _ . - . . .
i 116 .j l
1- SALP, regional assessments of plants done by the regions, 2 the accident sequence precursors get fed into that O .3 consideration, enforcement actions, and as you're probably j
4- aware there are a number of meetings that are_ held prior j i
5 to the senior management meeting in which the performance !
6 judgments are sort of finalized, and in those meetings all l 7- of these factors are one way or another brought to bear in j 8 the discussions and decisions as to how one judges the )
I 9 performance of licensees. j l
10 In June of this year, the Commission' issued an i l
11 SRM in which they asked the staff to.look at ways to l i
12 improve the objectivity and consistency of the senior 13 management process. AEOD has been tasked to take a look 14 at the senior management process and the' types of 15- objective performance measures that are used in that 16 process.
17 The risk-based' performance indicators are a 18 modest part of that activity, and so I want to make sure l l
19' that the committee understands that this isn't the only 20 element that one would use in determining what is a good 21 performer or a poor performer in a plant. But it's just i
22 one of the factors that would go into the overall decision i
23 process.
f 24 AEOD expects to have some work completed on 25 this with the help of our consultants, Arthur Andersen, by NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234-4433
117 1 the end of'this calendar year. And I would expect that we ;
2 will be discussing our findings with the ACRS. Would that l 3 be correct, Ernie?
4 MR. ROSSI: Yes. I'm sure that we will do 5 that, and Pat is right that that work is due to be ;
6 completed in December of this year.
I 7 MR. BARANOWSKY: Okay. So the current l 8 performance indicators that the Commission has endorsed 9 are listed on this chart -- automatic scrams while ;
10 critical; safety system actuations; significant events as 1
11 determined by a panel composed of NRR, AEOD, and Research 12 folks; safety system failures; forced outage rate; ;
i 13 equipment forced outage rate per.thousand critical hours; f
14 collective radiation exposure; and something called cause L 15 codes, which is basically a compilation of all of'the 16 causal factors that go into reportable events.
17 So if an LER is issued, and there were two or 18 three failures, let's say, involved, and each failure had 19 a different cause, then there might be, say, two or three 20 cause codes fed in, which might be related to design or 21 engineering or operator performance, and so forth. !
l 22 MEMBER BARTON: . Pat, how are you handling on- -)
23' line maintenance when the safety system is taken down for 24 -- it's in a window that is allowed by tech. specs.? How j 1
3 25 does that affect this data or --
J ;
l l
NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
. - .. - .- . ._. . . . . - . . .. - . ~ - . - . . _ . . . _ - - . - -
118 l 1 MR. BARANOWSKY: Okay. This data in the l l
I- ,
2 current. set of performance indicators is not handled at d' 3 all, but I can talk a little bit about how we're handling i 4 it or proposing to look at it with some of the either 5 system reliability results or the risk-based performance
)
6 indicators.
! 7 VICE CHAIRMAN SEALE: How long is this cause 1
8 code list? ;
1 9 MR. BARANOWSKY: It's about six. Do you 10 recall what they are, Steve, all of them? l l- i l 11 MR. MAYO: I didn't bring the list with me. i I
l 12 It's got six or seven, t
.13 MR. BARANOWSKY: I may have ther here. l O
14 MR. MAYO: It's six or seven. It includes' !
! 15 things like administrative errors, procedural errors, [
16 operator errors, design issues, maintenance problems, so i
-17 there is a list of them. We gave them to.you at the-last 18 briefing.
l l
! 19 VICE CHAIRMAN SEALE: So it's still a pretty l
20 coarse -- {
l 21 MR. BARANOWSKY: It's a very -- 1 22 MEMBER BARTON: Is there a subset in that
, 23 cause list? You mentioned procedures, inadequate i 24 procedure, didn't follow the procedure. You know, there i 25 could be a whole bunch of --
i.
ll NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
119 1 MR. BARANOWSKY: The causo code is --
2 MEMBER BARTON: -- subsets under each major
- i 3 cause. Do you go into that?
4 MR. BARANOWSKY: Here they are. Here we go.
5 No , that's not the right one. Wait a minute. I have too i
6 many backup slides. There it is. Those are the cause l 7 codes. I 8 So it's a coarse cut. I mean, the idea was 9 that they would be at a fairly high level to indicate, in I i
10 a sense, programmatic areas to look at more closely, i 11 VICE CHAIRMAN SEALE: It also helps you get i
12 statistics in a hurry. !
13 MR. BARANOWSKY: Yes.
Es 14 VICE CHAIRMAN SEALE: It's a big bin.
15 MR. BARANOWSKY: Yes.
16 The candidates for risk-based performance 17 indicators would involve some initiating event frequency 18 characterization, reliability and availability of risk-19 important systems, and I think I would throw components in 20 there too. I could see, for instance, a valve reliability 21 indicator as being a possibility.
22 We would hope to use accident sequence 23 precursor type analyses to put the so-called significant 24 events into a risk-based context, and some sort of an p
() 25 indicated core damage frequency type of indicator that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
i (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
120 j 1 might pull together system reliability and initiating :
2 event information in a model that reflects the l
.O O 3 configuration of a particular plant.
4 None of this is developed at this point. It's j 5 just ideas that we are looking to pursue, and we're hoping 1
6 that the ACRS will either agree or suggest some !
.7 modifications. But my understanding is that this is the l 8 kind of thing that one could logically identify from I 9 looking at risk and trying to-identify performance ,
l 10 indicators that are associated with risk.
11 VICE CHAIRMAN SEALE: If I try to come off the' 12 wall with one, one that kind of -- you know, I have the 13 feeling it would be indicative of something, but I'm not 14 sure it would be easy to quantify. But what about the 15 time it takes to recover from an event?
16 MR. BARANOWSKY: Yes. l l
17 VICE CHAIRMAN SEALE: You know, there are some 18 plants that run forever, but then when they finally shut I 1
19 down they have a hell of a time getting back up. Now 20 that's a different situation somewhat. But, you know, it 21 has to do with the real -- what's the word I'm looking 22 for? Anyway, you understand my --
23 MR. BARANOWSKY: I understand what you're 24 talking about. There are probably a number of things like
-(~N
( ,) 25 that that one could suggest as possible indicators. The NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
. .. . . . . . . . .. .. -. . - . - . - -- . - - . . . - =_- -
121 ,
i 1 hard part for us would be to figure out, how do we relate ,
i 2 them to risk?
73 l U .3 VICE CHAIRMAN SEALE: Yes. [
4 -MR. BARANOWSKY: And the reason that's 5 important, at least for what we're doing here, is w'e're ,
i 6 trying to not just do a bean count, for instance. We're i 7 trying to put some sort of a severity or a consequential l l
8 measure in there in addition to counting things. The risk !
l
-9 equation is basically consequence times frequency. We l 10 don't just want frequency. We want to bring consequence l
11 in .' l 12 And there is different ways to bring it in, j l
13 but-if we can't connect it somehow to risk, I don't know f
O 14 how to get a perspective on all of these things, which 15 doesn't mean it wouldn't be put into the mix, because i
16 there is going to be a lot of things in'the objective 17 measures of performance that are outside of a strictly l i
18 risk-based framework. And maybe over time we'll figure ]
l 19 out how to put them into that framework, because people :
I 1
20 are looking, for instance, at organizational factors and 21 how do those things relate to risk. I mean, we know it 22 does, but we're not able to have a simple model I think to l
l '
l 23 show that all the time.
24 VICE CHAIRMAN SEALE: Well, do you score
$ /"
25 things differently if you have a failure and you maybe
]'
, I
- 1. NEAL R. GROSS 1 l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
! (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 122 1 have to fall through a couple of subsequent steps in an I
g3 2 event tree before you get to mitigation?
(' )
3 MR. BARANOWSKY: Yes. I think you do score 4 things differently, but that doesn't mean you ignore !
5 things that are not the most significant, because they 6 could be the building blocks to more significant events 7 down the road.
1 VICE CHAIRMAN SEALE:
8 Yes.
9 MR. BARANOWSKY: You just need to be able to 1
10 put them in context, so that you don't say, "I had an HVAC 1
I 11 actuation, and two diesel generators failed, so those are '
l 12 two counts." Well, they're not really the same, but I ;
l i
13 might have a lot of small things occur which are ;
(N r f V
l 14 indicative of problems building in an area. And through a j 15 risk model, I would say, "Hmm. I seem to be seeing 16 degradations of diesel generators before I see multiple 17 failures." j 18 VICE CHAIRMAN SEALE: Yes.
1 19 MR. BARANOWSKY: And so we need to figure out 20 how to deal with some of that stuff, and that's just --
21 the more you go down, it's sort of a preemptive look, if l l 22 you will, at performance. And we're just going to look at 23 those kinds of things, and hopefully we'll discuss them 24 over the next year or so as we try to look at the data
,/
'() 25 that's available and what kind of insights we can get out NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
I 123 1 of it from a risk-based performance indicator point of- j I
2 view )
3 So I think it will be a good thing for us to i 4 explore in detail with the subcommittee and then come.back
.l 5 to'the full committee.
6 CHAIRMAN'KRESS: Let me ask you about the j 7- fourth bullet'there, the integrated CDF indicator.
8 MR. BARANOWSKY: Yes.
9 CHAIRMAN KRESS: Do you have in mind there 10 something like a living PRA that looks at plant 11 configuration and projects the core damage frequency?
12 MR. BARANOWSKY: It would be something like 13 that, but limited. For one thing,-there are some-things. l
. 1 14 that aren't changing much in time that we're not' going to 15 be able to look at indications on, like the' frequency of 16 earthquakes. So some external events --
17 CHAIRMAN KRESS: Right. .
l 18 MR. BARANOWSKY: -- we're just going to say, 19 "They are what they are" --
20 CHAIRMAN KRESS: Okay.
I 21 MR. BARANOWSKY: -
"for the site as j 22 characterized." But in terms of how do the kinds of j l
23 equipment performance elements that are talked about a lot 24 in plant performance reviews, how do they feed into the 25 risk picture, I think we might be able to put something NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
-124 1 together there using models like we have in the accident 2 sequence-precursor program, which are basically high-level 3 or simplified models that just capture what we know to_be 4 the dominant risk characteristics of the plant.
5 They're not really good for doing design
! studies and making lots of operational decisions.
6 But if l
'7 you want to just put things into some context that you've j 8 already determined are important, you'can use them for l
l 9 that. You can't use them to screen things that you l 10 haven't got.in the model, obviously.
L 11 CHAIRMAN'KRESS: So this would be a-measure of j i
12 how well.the particular plant is maintaining basically a l
13 design basis?
14 MR. BARANOWSKY: Well, it's more than design I
15 basis. It would be the operation of the plant, the l 16 maintenance of the plant. Design basis --
17 CHAIRMAN KRESS: Yes. It would be indirectly 18 reflected in down -- how to service things that were a 1
19 part of the design basis but are suddenly out of service '
20 or that are allowed to be out of service. But --
I l 21 MR. BARANOWSKY: They could be out of service l
22 because they're allowed to be out of service. Also, it l
l 23 could be out of service because they don't meet design i
24 requirements.
25 CHAIRMAN KRESS: Okay. Yes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
f (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
125 1 MR. BARANOWSKY: And then were uncovered 1
2 through some sort of a design review. A new test could be i 1
3 performed which reveals that this capability of the !
4 equipment wasn't what we thought it was. That would be 5 reflected in there. l 6 CHAIRMAN KRESS: Okay. Now, I'd like to also 7 ask you about the third bullet. Now, I view that as sort l 1
8 of the Bayesian approach to looking at actual things that ,
9 are happening at the plant, and seeing what the actual j i
- 10. core damage frequency status of the plant looks like.it j 1
11 may be as a result of all of the operating events, that-12 you project onto core damage. ,
i 13 Am I reading that one right? '
A b 14 MR. BARANOWSKY: Is that the third bullet? l l
15 CHAIRMAN KRESS: The third bullet, yes. l 16 MR. BARANOWSKY: The third bullet is meant to 17 be a way to go and look at the most significant events 18 that have occurred over some period.of time. !
I i 19 CHAIRMAN KRESS: For a given plant. j 20 MR. BARANOWSKY: For a given plant. As 21 opposed to what are the core damage frequency trends >
! i 22 indicated by the overall performance of equipment and j 23 people. That's the fourth one. So it's a sort of a --
i 24 CHAIRMAN KRESS: The fourth one is --
i O
Q 25 MR. BARANOWSKY: -- how many of these big hits i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHoDE ISLAND AVE., N W.
l l (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
126 1 do you have?
l ,m 2 CHAIRMAN KRESS: The fourth one and third one k
3 may be related, but loosely.
l 4 MR. BARANOWSKY: They're related, but the 5 difference is in one case I could see sort of a time-l 6 dependent curve, and in the other case it would be a 7 listing. In fact, I'm going to show you something in the 8 accident sequence precursor part of the discussion that 9 will be like what we're talking about here.
10 CHAIRMAN KRESS: Yes. And the other two 11 bullets are pretty obvious, I think.
12 MR. BARANOWSKY: Yes.
13 VICE CHAIRMAN SEALE: Do you keep track of
( >
14 maintenance backlogs?
15 MR. BARANOWSKY: We aren't keeping track of I
16 anything yet to do this.
17 VICE CHAIRMAN SEALE: Is that a potential 18 performance indicator?
19 MR. BARANOWSKY: It's a potential performance 20 indicator, but it's --
21 CHAIRMAN KRESS: It's hard to connect that 22 to --
23 MR. BARANOWSKY: Is it a risk-based one? I'm 24 not sure.
r%
, ) 25 VICE CHAIRMAN SEALE: I understand that, t
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
l 127 !
1 CHAIRMAN KRESS: Now, there -- !
2 MR. ROSSI: Jumping back, one of the problems j 3 with the maintenance backlog is that different. plants keep 4 track of that and count it-in very different ways, and so l 5 it's not something that you can easily compare from plant
=6 to plant', without us getting deeply involved in
.7 understanding how they do it at each plant or --
8 VICE. CHAIRMAN SEALE: The first derivative 9 might be interesting, though.
10 MR. ROSSI: That would be perhaps, yes. But 11 even there, they can combine various things into one work 12 order and that kind of stuff. So it's something that can 13 be very hard to compare, and even trend I guess.
14 MEMBER BARTON: Don't count the number of work 15 requests. Count -- a common basis may be number of .j
'16 manhours of backlog work.
17 MR. ROSSI: And then, of course, you have to i
1 18 look at the ones that are really safety significant and l 1
19 those that are not. So it's not an easy thing for us to i
20 use as a performance indicator. ;
i 21 MR. BARANOWSKY: But these are factors that 22 would go into the overall set of performance measures that 23 might replace some of the ones that I showed you in that 24 earlier list that are currently being looked at.
25 VICE CHAIRMAN SEALE: We could go on on this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234 4 433
128 1 forever, but I have one other I wanted to ask you. Whsn !
I 2 you go through and you look at events, do you also try to I,,)
3 identify when the -- if the plant had had a similar event 4 in the recent past?
5 MR. BARANOWSKY: That's something that we have 6 done as part of our performance analysis in AEOD in 7 looking at operational data.
8 VICE CHAIRMAN SEALE: Yes.
I 9 MR. BARANOWSKY: Yes. We do that right now. l 10 VICE CHAIRMAN SEALE: Okay. )
11 CHAIRMAN KRESS: I still want to ask you, the j 12 only -- the second bullet there -- risk important systems 13 reliability -- is the only one I see that could have any
,x x 14 connection to containment reliability. I suspect it could 15 have --
16 MR. BARANOWSKY: Yes.
l 17 CHAIRMAN KRESS: -- some risk important 18 systems have to do with containment.
19 MR. BARANOWSKY: We could have containment 20 spray systems. We could even have something to do with 21 containment integrity. !
22 CHAIRMAN KRESS: Would it be a good idea to 23 have a containment integrity, like the CDF indicator up 24 there? Can you --
,/ y i ) 25 MR. BARANOWSKY: If it's prcictical --
l l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
. . _ . . . - . . . - _ . _ . _ _ ._-_m. _ _..._.__...-m- -- _ __
129
-1 CHAIRMAN KRESS: -- do that?
2 MR. BARANOWSKY: -- it would be a good idea, 3 yes.
4 CHAIRMAN KRESS: Yes. Okay.
5 MR. BARANOWSKY: Because that really gives --
6 when you talk about risk, you need to know about 7 containment in addition to core damage frequency to really 8 have a risk perspective.
9 CHAIRMAN KRESS: If you had both of those 10 covered it would be better.
I 11 MR. BARANOWSKY: It would be better, and I l 12 think our ideas are if we can do it we would like to. At 13 this point, we're not sure how to do it. But when we have O i t
V 14 some ideas, we're going to come back and talk to you about 15 it. l 16 Let me move along, skip the next viewgraph, 17 because it just shusa the relationship of some of-the old ;
i 18 indicators to the new ones, and go to a chart that we put 19 together at the request of the subcommittee in October, in 20 which the issue was raised of, well, what are some of the !
21 attributes of these performance indicators?
22 And so we tried to put some thoughts down on 23 paper. These are by no means cast in concrete. But I 24 know a number of the subcommittee members said, "Well, go 25 ahead. Give a try at it. See what your thoughts are."
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE IStAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234 4 33 l
130 l
1 And so we think that the risk-based performance indicators :
2 need to be related to plant risk by accident sequence 7s I i' ')
3 logic and risk hierarchy.
l I
1 4 Now, the risk hierarchy is important because l l
5 you can look at different levels. You can look at core i 6 damage frequency, go down a level; system performance, go i
7 down a level; equipment performance, go down a level; !
8 equipment degradations that -- maintenance, whatever. You 9 can keep going down levels, and you have to figure out !
l 10 what level makes sense. ),
11 And one of the factors that go into that is: )
12 what data is available and what is the -- sort of the ,
13 response time of that particular element? I have to have, I
/~N 1
t
\> 14 in a performance indicator, the ability to identify 15 changes in a timely manner. I don't know if that's six 16 months, a year, or five years, but I do have to come up l 17 with some indications that aren't so stagnant that I can 18 never really get a measure of how performance is changing.
i 19 Obviously, the indicators need to be 20 objective, and they either should be directly measurable 21 or we should be able to calculate them in a fairly direct i
22 manner. And so these are the kinds of attributes that we 23 think have to be overlaid on our look at what are 24 candidates for risk-based performance indicators. And if (x
i s ) 25 anybody from the ACRS has some more ideas, we're NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
. .-_ .- - . - - . - - - _ . _ _ .- ._ - . . - . . ~ . _.. . ..
131 1 interested in it. We don't need to, you know, resolva'it g 2 now. But as time goes on, we'll talk about it more.
i 3 Also, in that meeting we'said we would 4 identify what might be some of the steps in a-project to i 5 develop risk-based performance indicators, and so we said, !
6 "Well, maybe the first thing is to come up.with some 7- attributes and see if there are data source, identify the 8 candidate indicators in a less abstract way." ;
9 In other words, try and put some definition to !
i 10 them, perhaps conduct some trial analyses using.the )
11 candidate indicators, and get a broad review, not just of ,
a i
12 the trial analyses but the whole thing, and that would i
13 involve making it available for the nuclear industry, !
(
Y 14 staff, ACRS, Commission, public. So I think they are j l
15 going to be fairly important, and we don't want.to have 1
16 any misleading indicators, because these will have a' l l
17 pretty high visibility.
18 And so that's the kind of program that we're 19 going to lay out as we go forward here.
20 okay. I'm done now with risk-based L
21 performance indicators, and I'm going to talk a little bit j
)
l 22 about a few of the items that I have listed on this 'l l l 23 accomplishments chart. In particular, I'm going to talk j i
24 about accident sequence, precursor program, some of the j r
() 25 system reliability studies, and the common cause failure l NEAL R. GROSS l l COURT REPORTERS AND TRANSCRIBERS ,
j 1323 RHODE ISLAND AVE., N.W. l l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
132 1 database.
,_, 2 The accident sequence precursor program, for
/ \
\
'~') 3 those of you who are relatively new to this, has some 4 objectives. And the most important one is to identify and l 5 rank risk significance of operating experience, or at 6 least if it's not the most important one it's the initial 7 one that we had when we first implemented the program. l 1
8 But we're also interested in seeing if there
]
9 are any generic implications to this analysis. It is a 10 supplemental input to plant-specific performance 11 indications. We could check PRAs to see if the 12 characteristics of the accident sequence precursors are l 13 like those that are identified in PRAs in both the k_-) 14 frequency and the elements that go into the sequence.
15 And it may provide some sort of an empirical 16 measure of industry risk. That's a controversial one. I 17 over the last few years, there have been 18 several important accident sequence precursor events. I 19 have listed five of them here. I'm not going to go over 20 them in any detail. But as an example of my earlier 21 response to Dr. Kress on would we have the accident 22 sequence precursor results listed separately, I could 23 imagine -- for instance, I think you see Wolf Creek down 24 here twice. So we'd say, "Uh-huh. In the last ,hree 7
( ,) 25 years, Wolf Creek has had, you know, a couple of the more NEAd. R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
133 i l
1 important accident sequence precursors."
\
i
,-s 2 See Haddam Neck down there. Some questions 1
~
3 about Haddam Neck's design. In fact, there is a motor l I
4 control center -- MCC-5 -- that shows up as being very l 5 important in their PRA, and it showed up as being very 6 important in a couple of incidents that occurred at the 7 plant. l l
8 The Wolf Creek event that occurred in 1994 was l 9 interesting, not only because two of them occurred at Wolf )
10 Creek, but because it had an intersystem LOCA i 11 characteristic to it that was a little bit different than 12 anything we had seen before. This was an event where they l 13 had been shut down for 28 hours3.240741e-4 days <br />0.00778 hours <br />4.62963e-5 weeks <br />1.0654e-5 months <br />, and due to an i
(~'i !
k 14 inappropriate alignment on the residual heat removal 15 system they drained -- and some would say blew down -- in 16 about a minute, 10,000 gallons of water from the reactor 17 coolant system into the refueling water storage tank.
18 Some analyses indicate that there were about 19 three minutes away from being in a situation where they 20 would have made their ECCS inoperable and would have 21 voided the core in about 30 minutes. So it was a pretty 22 interesting event, one like we hadn't seen before.
23 And I'll just mention the Catawba 2 event is l
l 24 just another station blackout incident in which we had a
(-) 25 loss of off-site power that was brought on by some faults NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 2344433
134 'I i
1 in the main transformer, and then one of the diesel :
2 generators was out of service at the time. You had d 3 mentioned that before. And if they had lost one more r
4 diesel generator, they would have been in some significant I
5 trouble.
f 6 So we can come up with a characterization of 1
i
!' 7 the most important events like this, and then the next
[ .
8 viewgraph gives the kind of trending plots that we could j i
r 9 in theory put together. I'm not saying these.are the ones !
I 10 we would propose, but in the upper right-hand corner what [
l 11 we have~is a trend of all of the accident sequence !;
12 precursors.
13 Now, this is for the industry,-and you can see t
O 14 that the number of accident sequence precursors has been l
15 coming down, and we have some confidence that that's a j l
i ,
l 16 true trend. j I l l 17 on the lower left-hand -- )
18 CHAIRMAN KRESS: What is meant by the word
( 19 " predicted" there?
l l 20 MR. BARANOWSKY: By the word what?
21 CHAIRMAN KRESS: " Predicted" on that curve. i 22 MR. BARANOWSKY: Which one?
23 MEMBER POWERS: Upper right-hand corner. l 24 MEMBER BARTON: Your solid line on that graph.
l 25 It's called " predicted."
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1'823 RHODE ISLAND AVE., N.W.
(202) 234-4433 WA"sHINGTON, D.C. 20005-3701 (202) 234-4433
. - . - _ . . _ - . -__ __ . _ _ - - . . . -- - . . . . -. ~
I 135 l' MR. BARANOWSKY: That's the curve fit, and l
2 it's what we would predict to occur if the curve' fit was',
3 you know -- 1 4- CHAIRMAN KRESS: I see.
5 MR. BARANOWSKY: -- the best estimate.
6 CHAIRMAN KRESS: Yes, I-understand.
7 MR. BARANOWSKY: It's the --
]
1
.S CHAIRMAN KRESS: .It's a poor choice of words, l
9 I guess.
10 MR. BARANOWSKY: I accept that.
l l'
11 CHAIRMAN KRESS: Okay. ]
12 MR. BARANOWSKY: The one on the left shows a 13 breakdown of accident sequence precursors over time into D.
L 14 core damage frequency bins. This is a curve that'we l
15 provided to.the Commission, and I believe it goes to the' !
16 OMB every year, as one of the measures of regulatory 17 performance. 'And on the right, the lower right, I broke j 18 out a curve here that shows just the most significant core 19 damage frequency occurrences.
20 And in this case, it's interesting to note l 21 that we keep on seeing ones and twos of these every couple j 22 of years. And as I indicated from my prior chart, 23 sometimes they have some unique characteristics to the !
i l 24 events that we hadn't thought about well before. And so I i 25 think it's one of these elements of reviewing operating l-NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
. .. - . - . . .. - - . - . -. . - ~ ~ .- . -. - . - - - . . -- - - - - . ... - ..-
136 1 experience that provides a good feedback of what.might bs 2: important and what we.might focus on.
O 3 In many cases, there have been either 4 information notices or bulletins that have come about as a 1
5 result of findings on'these analyses. .]
l 6 CHAIRMAN KRESS: That trend on precursors you l 7 have there,-would you say that's an indication of what a 8 good job NRC is doing?
9 (Laughter.)
10 MR. BARANOWSKY: I'think you have to say that 11 it just indicates where the industry-is, and I don't know 12 who is doing a good job.
13 CHAIRMAN KRESS: Hal Lewis would have liked
( <
14- that-answer.
~
15 MR. BARANOWSKY: Yes. In fact, that's what we i 16 say to the OMB. We don't say this is a measure of NRC's 17 regulatory excellence or.what licensees are doing.
18 MEMBER POWERS: Maybe you should, i
.19 MR. BARANOWSKY: I don't know.
! 20 (Laughter.) I l 21 CHAIRMAN KRESS: Well, Hal didn't like it when l
'.l 22 they did before.
'23 MEMBER POWERS: Well, Hal is not dispensing i
24 the budget and OMB is.
25 (Laughter.)
I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4133 WASHINGTON, D.C. 20005 3701 (202) 234 4433 l
137 1 MR. BARANOWSKY: Okay. Another aspect of our j
,s 2 program that we discussed with the subcommittee is the
( )
%~/ 3 system reliability studies, and here what we're trying to 4 do is pick a number of risk-important systems and come up 5 with fairly simple reliability models that are data 6 driven, so that we can use either actual demands or 7 demands on systems that are as close as possible to actual l 8 demands, take failures and unavailabilities associated 9 with these demands and analyze what the reliability 10 implications are, look at trends, put some uncertainty 11 onto it.
12 And now, one of the important parts is to 13 compare our results with what people have been finding in g~
i i
/
\- 14 PRAs and IPEs. And even though we have a fairly sparse 15 database because of our uncertainty analysis, we ought to 16 at least be able to capture, within our uncertainty 17 bounds, what people are getting in PRAs and IPEs. And for 18 the most part, we do.
19 Sometimes we have a really good agreement with 20 what some of the PRAs and IPEs are saying. In other 21 cases, we don't.
22 CHAIRMAN KRESS: What is the status of this 23 reporting rule that --
24 MR. BARANOWSKY: Okay. I was going to cover
) 25 that in a minute --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
138 l l
1 CHAIRMAN KRESS: Oh. You were? Okay. !
,y 2 MR. BARANOWSKY: -- if I could. 1 i )
3 So we're using primarily LER data here, so
- e. it's -- what we have to do is -- l l
l 5 CHAIRMAN KRESS: Oh. You're getting this out 6 of the LER data.
7 MR. BARANOWSKY: Yes. We have -- that's why 8 it's very sparse data. We have to cover many years --
9 CHAIRMAN KRESS: Yes. i 1
10 MR. BARANOWSKY: -- in order to get even any j 11 kind of indication. But it's what's available.
12 CHAIRMAN KRESS: Yes. Okay. f 13 MR. BARANOWSKY: And let me just show you some
> 1 N/ 14 of the kinds of things that we have been finding for some 15 of the studies that were already completed. We have done 16 high pressure coolant injection on boiling water reactors, 17 diesel generators for plants that report to Reg.
18 Guide 1.108, isolation condensers and reactor core, 19 isolation coolant systems for boiling water reactors.
20 Now, we did these systems because they are the 21 easiest to do. The data was reported well in LERs, and we
, 22 think we can get an unbiased count of demands and failures 1
23 if we carefully go through the LERs.
24 MEMBER BARTON: What if you are uncertain n
() 25 about what you've seen? I've seen some poorly written l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., ILW.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
139 1 LERs.that I wouldn't' be able to find the data that you're l
t 2 looking for. !
O 3 MR. BARANOWSKY: Yes. In some cases, we'run l l
4 into that, and we try to resolve it internally if it's !
5 important. Sometimes it's not-that important, and we can l
6 have one or two mischaracterizations because we're looking l I
7' at numerators and denominators. j 8 Also, we have methods for treating the !
1 9 uncertainty in our ability to classify the data. So our )
?
i 10 uncertainty bounds can get larger. And then lastly, if j i
11 it's not too old, we.can go back to a licensee to find out ;
l I
12 about it. But it's pretty hard to go back and ask them !
a 1
13 about something that occurred six or seven or eight years j i
14 ago. A lct of times a system engineer who wrote the l l ,
15 report is not available. [
16 MEMBER BARTON: Right.
17 MR. BARANOWSKY: So that's a problem. I would-18 have to agree.with you on that.
19 Even so, I think one of the interesting l
l ,
20 columns on this chart is the one that says " consistency j 21 with PRAs and IPEs." And for the most part, we're finding ;
22 good consistency. In some cases, we're finding that the !
( 23 estimated unreliability is higher or lower, and one reason j
, i 24 that it has been higher or lower is licensees have used l
25 generic data in some cases, and in other cases they have l l'
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 L
_. . . _ . _ _ . . . _ _ . _ _ - . . . _ _ . _ _ . . . _ . . _ . _ . . . . _ ___ _.m_. .., _ . . ,
140 '!
1 used a lot of demands that were for, let's say, less [
2 challenging tests than would be associated with a real l
( 3 demand on the system.
i
}
4 I guess it's possible that our estimates could I 5 be in error, but if it is.the use of unbiased data for-6 actual demands, I'm hard pressed to find out why it would 1 l
7 be inaccurate except for the items that Mr. Barton-f 8 mentioned where we might not have interpreted the data 9 correctly.
{
i 10 The other thing that is interesting is we are j
11 seeing that some of the failure modes and the way they 12 operate these systems under real demands is a little l
13 different than what is assumed in some of the PRAs. And-l 14 I'll mention the RCIC, which was kind of a dramatic 15 example of this, where in a PRA the assumption is you 16 start and run the RCIC and it runs for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. That's !
l 17 its mission success criteria. f J
18 Well, what we see in the real world is RCIC i 19 starts and stops many times during certain evolutions.
20 Sometimes one goes into recirculation mode, and other 21 times it just starts and stops to fill the vessel. So 22 there is -- the nature of the challenge is a little bit 12 3 different, and the failures and failure modes that 24 occurred that we identified in.our study were not exactly l
- - 1
- g / 25 the'same as those that we saw in a number of the PRAs.
NEAL R. GROSS ;
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234 4 433 l
4
- . . - - -. .- .~ . . . - . , . - - ._ . _ . - - ~
141 1 Interestingly, the quantitative results warc s
,,) 2 very.similar, so it may be-that the data was not
\'# 3- classified quite right in some of the PRA cases, or we .j 4 were just lucky. I don't know which.
5 We plan on doing work on auxiliary feedwater l 6' systems, high pressure safety injection, reactor I I
7 protection' system, and, if possible, the low pressure t 8 safety injection systems. And we presented some of these ;
i 9 results to the ACRS full committee before, and we'll come '
10 back again at different points to show you what we're l
11 finding, j l
12 Now let me move to the common cause failure 13 database. The common cause failure database is based on 1 O
\w I 14 what I would call current state-of-the-art methods. The 15 NUREG 4780 that was developed in the mid-1980s identified 16 a scheme for screening and classifying common cause 17 failure data, and the data work wasn't taken any further 18 than that. And so the data that was available was I think 19 through about 1984.
20 Is that right, Dale? '83.
l 21 And we felt like, gee, here we are with l 22 another 13 or 14 years of experience. We ought to not be 23' using a database that's that old to try and draw ;
l 24 conclusions on common cause failures.
() 25 So we took the existing methodology. We NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
t
-142 l 1 developed protocols for applying it, trained people to j f
-2 screen the data and code it, and employed a number of i O 3 nationally and internationally recognized experts in the
.f f
4 area'of common cause failure to advise us and to resolve l
-i 5 different technical matters that arose during the l i
6 preparation of the database. l 7 .So the database now provides a scrutable data r
8 source -- one like we really don't have for component- i
.9 failures for general liability analysis at this point. j
'10 There's a lot of quality control applied to the database. !
-11 It's easily updatable, and I think it provides a better !
i i
12 understanding of common cause failure mechanisms and j 13 causal factors that one could ever possibly get by a ;
O 14 casual or even a dedicated screening of the data through 15 ad hoc methods.
)
16 VICE CHAIRMAN SEALE: Pat, I think it's -- a.
17 significant step in this has been the access to the NPRDS 18 database. And I understand that there are some 19 limitations on what you can do with that in some cases 20 or --
21 MR. BARANOWSKY: Right.
22 VICE CHAIRMAN SEALE: -- or I understood there 23 were anyway. Is that still the case?
[
24 MR. BARANOWSKY: Yes. Now, some of the data 25 is derived from NPRDS records as well as from LERs, or any I
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
i 143 '
1 other source that we can get common cause failure
,s 2 information. Now, NPRDS is proprietary. It's restricted t' 1
'~
3 in its distribution by the Institute for Nuclear Power I
4 Operations, INPO. And so we have to, when we distribute l
5 the common cause failure database, we actually have two 6 versions of it -- the proprietary and the non-proprietary l 7 version. j l
8 The proprietary version we think is the one 9 that is necessary to do plant-specific common cause 10 failure analysis, because it has the descriptive material 11 that one needs to look at in order to understand the 12 applicability of events to particular circumstances. !
)
i 13 VICE CHAIRMAN SEALE: Are all of the events in
( i
\/ 14 both databases, but the material provides the explanation 15 not in the --
16 MR. BARANOWSKY: Right. The --
17 VICE CHAIRMAN SEALE: -- non-proprietary -- l 18 MR. BARANOWSKY: -- the proprietary 19 information is the descriptive material --
20 VICE CHAIRMAN SEALE: Okay.
21 MR. BARANOWSKY: -- which I think is really 22 critical to understanding an event. And although one 23 could take the database without that descriptive material 24 and just blindly calculate things, there would be some f
(j .
1 25 question about its applicability to a specific case.
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
l
144 1 MEMBER POWERS: And I still think we need to g 2 discuss.this topic, because I think it's very bothersome 3 when you're having.to use a proprietary database in some 4 sort of regulatory context, simply because there's so many 5 people with an interest here that don't have access to i 6 that database.
7 MR. BARANOWSKY: Of course,-the data is 8 available to all members of'INPO, which means all U.S.
9- nuclear utilities.
10 VICE CHAIRMAN SEALE: I think Dr. Powers' 11 concern is that there are members of the public who have a 12 strong interest in the problem who don't have access to l
13 the database.
14 MEMBER POWERS: There are going to be vendors.
i 15 that don't have access to the database.
16 VICE CHAIRMAN SEALE: Do you mean component 17 vendors?
i.
l-l 18' MEMBER POWERS: Component vendors, sure.
19 There are going to be A&E firms that don't have access to l
20 it.
21 MR. BARANOWSKY: Yes. I don't know how many 22 of them are members of INPO. Some are, and some have 1
23 access, and.they may even pay for access to the database. 'l 1
. i
'. 24 It has some commercial value, so it's an 3ssue that one -
j 1
25 could' discuss, and I think it's a legal issue if nothing l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
- - .- ..- , - _ . . ,. -, . . . - . . - - - .-,-,.--.-.o
. _ _ . . _ ~ ~ . ~ .
l i L 145 !
1 else.
l 2 VICE CHAIRMAN SEALE: Yes, it is. j O 3 MR. BARANOWSKY: The database -- -
, 4 VICE CHAIRMAN SEALE: And it hasn't been~ easy, f l'
5 by the way, for the NRC to negotiate the access that they i
6 have. So -- f 7 MR. BARANOWSKY: Oh. I can vouch for that. I t
I I l 8 (Laughter.) i i
9 VICE CHAIRMAN SEALE: And I've heard some of {
10 it from the other end, too, so I --
11 MR. BARANOWSKY: No. I mean, we can spend j l b 12 literally a year or more negotiating minor changes in the :
i 13 contract, and I'm not' talking about, you know, a whole new I l
14 big deal.
i 15 VICE CHAIRMAN SEALE: Yes. t
- l. 16 MR. BARANOWSKY: They are protective of their l l i l 17 proprietary information, j i
18 The database includes a number of systems, l
l 19 component types, and failure modes, and I've shown a table ;
i t i )
20 of most of the systems here, and I think all of the ;
i
- l. 21 component types and failure modes are listed. There are i 22 1,368 records in the database that are related to common l i
23 cause failure. Not all of them are everything failed at j 24 once. In fact, only a couple hundred are of that nature, !
l 25 but the others involve things like one or two components NEAL R. GROSS :
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
f.
146 g i failed and another one was degraded.
2 It covers the period 1980 through 1993, and we !
l[~T.
\_s/
i t
p 3 plan on updating this through 1994 or '5, Dale?
f i
4 MR. RASMUSON: '95. ,
I 5 MR. BARANOWSKY: 95. What, this coming ,
i l
6 summer? Summer of '97? ,
t
! 7 MR. RASMUSON: No. The work will be finished -!
l 8 at the end of December, f i
9 MR. BARANOWSKY: Okay. So we'll have it up 10 through 1995, the end or December. The analysis (
i 11 capability for deriving some common cause failure !
l 12 parameters involves the method called the alpha factor I 13 method or the multiple greek letter method. We could put !
\
k 14 other methods in there, but those are the ones that are l 15 pretty commonly used in the United States, and we are. l 16 using the alpha factor method ourselves.
17 And the database provides really a lot of' 18 flexibility for users to screen the data and modify input 19 data for specific applications of the data. Now, what 20 we're doing is not generating an analysis that is meant to l 21 be an input for a PRA. Our job is to analyze the l 22 operating experience and draw some generic conclusions 1
i 23 from it. That's like the number 1 in our mission, j 24 remember?
) 25 And so what we're looking to do is perform an i
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
m . -
147 1 evaluation of the data, and some quantification of it,
,y 2 that allows us to characterize what we're currently seeing i t
'\_/
3 in terms of common cause failure information. And there's l
4 an issue as to whether or not that characterization is a 5 generic characterization for the nuclear industry, and l
)
l 6 whether some of the uncertainties that we are able to l
7 calculate are indicative of the actual variability that 8 might occur throughout the industry in terms of the l l
9 distribution of these parameters, if one were to look at 10 them plant by plant in some sort of critical way. I l
11 I would not expect that we're going to try and l l
12 do any kind of plant-by-plant analysis, because frankly i
p_
13 that's a pretty resource-intensive activity where one goes
( )
k 14 through the data and specializes it for the particular 15 application of each plant.
4 16 I can't honestly say I know how much the data 17 would vary plant by plant, but I do know that most PRA 18 analysts will start out with a crude assessment of the 19 common cause failure parameters to even see if it's an 20 important input to their PRA. And if it is, then one 21 might go back and further refine the analysis. That seems 22 to be appropriate because this is a resource-intensive and i 23 an expert-intensive kind of activity.
24 It does involve a lot of careful review of the
/N
(_,) 25 data, and it's not as easy as just calling down all of the l
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
! 1323 RHODE ISLAND AVE., N W. l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 )
I j
, . . ~ ~ _ - . - . - . _ . - ._- _ - -- .. -- . . . ~ ~ . - -
148 1 pump failures and generating a pump failure alpha factor 2 and saying, "Therefore, that's right for my plant." I !
3 think you.can get an industry-wide number for that, but it .
4 may or may not be appropriate for a particular. l 3
5 application.
t 6 So what we're planning on doing is making the ,
7 database available, generally. Right now it has a limited j 8 distribution -- NRC staff, INPO. f 9 Anybody else, Dale?
10 MR. RASMUSON: A few contractors.
11 MR. BARANOWSKY: Okay. A few contractors. !
12 Since it's the most comprehensive compilation 13 of data available on common cause failure, we think that >
O !
'V 14 it should be made available to the nuclear industry. And 1 15 we will continue to look at our analysis and l 16 characterization of the data, and then after we're !
i 17 satisfied that we've understood a couple of the comments 18 that were raised at the.ACRS subcommittee, and have ,
13 addressed them, then we'll issue some operating experience ,
20 analysis reports. -
3 21 So that's sort of where we are on that one.
-1 22 MR. MARKLEY: And Pat, this page covers a i i
23 couple of things that were extensively talked about the i
24 other day, and one being the alpha factor and using single 1
25 numerical values as opposed to ranges. And then the other l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 2000 # 3701 (202) 234-4433 J
\ l i
l , .
1 149 1- thing, the plant-specific versus generic --
2 MR. BARANOWSKY: Right.
3 MR. MARKLEY: For the members who weren't 4 here, are there any other things you want to add to that 5 discussion related to the points that were raised or -- I 6 mean, I think you covered it to an' extent, but --
7 MR. BARANOWSKY: I-don't know. Maybe we could-8 find Dr. Apostolakis here and he could help us out a 9 little.
10 VICE CHAIRMAN SEALE: He's not.
11 MR. BARANOWSKY: Okay.
< 12 VICE CHAIRMAN SEALE: -But there were several 13 questions he did raise. Oh, he is?
(D' V 14 MR. BARANOWSKY: We addressed a lot of the 15 issues of how were things entered into the database, and l
16 then we had some discussion on was the uncertainty l
17 characterization indicative of the distribution in the 18 industry as opposed to more or less a characterization of 19 how well we just know the mean, I think. But George --
20 MEMBER APOSTOLAKIS: Well, I think you just )
1 i
l 21 said that you are planning to continue thinking about the 22 analysis methods.
l l 23 MR. BARANOWSKY: Yes. i
\
24 MEMBER APOSTOLAKIS: So maybe we shouldn't 4
) 25 discuss this issue now.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433 l
-150 1 'MR. BARANOWSKY: Yes. 'We'do want to continue 2 the dialogue on that, but -- and we don't need to narrow O 3 it down right here, but we want to make sure we understand 4 the specifics.
5 The one thing I will say is that I think we 6 have an obligation', once we've reviewed and analyzed the 7 data, and we have' eliminated any technical-problems, to?
-8' making some statement about what we see in the data. And 9 so --
' 10 ' ' MEMBER APOSTOLAKIS: So what you are proposing-11 is to issue now the database.
12 MR. BARANOWSKY: Yes.
13 MEMBER APOSTOLAKIS: And wait until some lO 14 technical issues are resolved before --
-i i
L .
l- 15 MR. BARANOWSKY: I think we're talking about i i
16 not issuing Volumes 5 and 6, which were the. plant-specific )
i 17 application and the generic analysis of the data.
j 18 MEMBER APOSTOLAKIS: So subject to Dr. Powers' 19 question being resolved on'the issue, I have no problem 20 with that.
l 21 MR. BARANOWSKY: Okay. So -- yes, so we 22 wouldn't want to hold up the database. ,
23 VICE CHAIRMAN SEALE: And that's going to l
- 24 broadly expand, or at least add two more years or three .l 25 more years on it -- two more I guess -- in another few I NEAL R. GROSS !
( COURT REPORTERS AND TRANSCRIBERS l t 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 . WASHINGTON, D.C. 20005-3701 (202) 234-4433
. l
151 1 months I guess.
.p 2 MR. BARANOWSKY: Right. So it's --
G 3 VICE CHAIRMAN SEALE: Great.
4 MR. BARANOWSKY: I think there's a lot of 5 interest in the international community, which we're ,
6 trying to figure out how to make it available there, too.
t 7 MEMBER APOSTOLAKIS: Do I have access to that !
8 database, by the way? :
9 MR. BARANOWSKY: You probably do as an ACRS '
10 member. I think ACRS has access to any proprietary data, f 11 right, that the NRC hfs? !
I 12 CHAIRMAN KRESS: Yes.
13 MEMBER POWERS: But you can't use it in your !
~
14 work, though. ,
i 15 MR. BARANOWSKY: You can't use it, but you 16 have access to it. !
17 (Laughter.) !
18 MEMBER APOSTOLAKIS: Unless I present 19' something as an ACRS member.
20 MR. BARANOWSKY: Right.
21 MR. MARKLEY: But you still have to maintain 22 the integrity of,the. proprietary material.that was 23 entrusted'to you. l 24 VICE CHAIRMAN SEALE: Okay. Let's --
1 25 MR. BARANOWSKY: Okay. I'm going to wrap up j 1
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
! 1323 RHODE !SLAND AVE., N.W.
l (202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i.
.._. . _ _ _ _ . - . _ _ _ . _ . _._. _ _ _ . . ~ _ _ _ . . _ _ _ . _ . _ . . _ . - - _
152 i i
i pretty quickly here.
2 VICE CHAIRMAN SEALE: All right.
3 MR. BARANOWSKY: We have a number of other 4 reports and activities that we did over the last year or-5 so -- service water system. study, fire experience study --
6 ar.d what we want to do with some of these things is at -
7 appropriate points come and talk to the ACRS and the .l 8 subcommittee I think on plant operations or PRA. And 9 these are the kind of things that we could talk about. f 10 And in the future, we're going to make some f
r 11 improvements in the risk-based performance indicators j i
12 which we discussed earlier. The' accident sequence 13 precursor program -- we're working on improved models and 14 data and timeliness. We discussed the models with-the .
15 subcommittee, and I would assume we'll have probably 16 subcommittee meetings on some of these Level 2 and 3 and -
'17 fire models that are being put.together.-
l 18 We have planned to update the loss of off-site ;
l 19 power database. I just wanted to mention that one in 20 particular, because the database and analysis of that data 21 is about 10 years old at this point. I ought to know; I 22 did it. So I think it's time to update that work.
i.
23 And we have a component reliability report
- 24 which we're working on that allows us to do some generic 25 analysis-but not plant-specific work. And then we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
1 153 *
'l regularly will update things like fire event frequencies I
2 and service water system performance, because there are 7--
3 generic issues associated with that information, and the ;
)
4 people who do the generic issues want to see what the l 5 actual operating experience is showing.
6 And then we expect to come back to the 7 subcommittee or the full committee regularly, maybe every. ,
i 8 six months, maybe three months, I don't know, but what 9 we'd want to talk about is analyses that we have completed i 10 and show you the results, and key developmental things 1
i 11 like the performance indicators in particular. We want to l
12 go through in steps instead of getting to the end. !
i 13 We'll have both subcommittee and full l
('
14 ccmmittee meetings, I assume, and what we're going to do 15 is talk to the staff to try and arrange something so we i
16 can_ plan on it over the next year or so. 'J 17 So let me just summarize. We have a plan to 18 analyze reactor operating experience that I think is 19 pretty comprehensive, and it provides good risk focus for 20 that analysis. It looks at both industry-wide trends and i 21 seeks to also evaluate plant-specific performance. It 22 looks at significant events through the accident sequence 23 precursor program and compares results with PRAs.
24 We can do more. And, in fact, with regard to f'h
\ ,) 25 the accident sequence precursor program, we're requesting NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
~ - - - . . - . . . .- .-
154 1 some additional resources in order to go back and look at j f-- .
2 accident sequence precursor results and compare them to 3 PRAs. If the committee thinks that's a good idea, it 4 might not be bad to mention it in your letter, because 5 there is a business of prioritization of resources, and 6 we're ready to do it. We think we have the resources to 7 do it, but you can never tell when these things are going 8 to be eliminated if the priority isn't understood to be 9 important .i 10 The risk-based performance indicators are in 11 the early stage of development. We just wanted to show 12 you our thinking and see if you agree with the' kind of 13 thinking that we have on it at this stage. We think that 14 they will be an improvement over the current indicators in 15 several areas and that they will be objective.
16 The system reliability methodology and common 17 cause failure database reports -- we provided a lot of 18 information to the ACRS, so we try to keep you informed. l l
19 And I don't know if we're deluging you with information or l 20 not, but I guess it's better just to provide the 21 information. And I thought we had a very good set of
~
22 discussions _over two days with the subcommittee on these 23 topics. ;
24 And I expect that we'll be talking about our
( 25 significant developments as time goes along. That NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234 4433
155 I completes'my prepared remarks on this topic. And I just ]
j W s 2 want to mention that what we were talking about in terms l 3 of an ACRS letter-was one that basically.found the thrust
~4 and direction of our activities to be appropriate, scope 5 reasonable, and, in particular, that we're taking the )
6 right steps with risk-based performance indicators.
7 VICE CHAIRMAN SEALE: I sense that. I also 8 have the impression'that.-- a comment to the effect that 9 the plant experience represents.a very dearly bought
. 10 record. And the people who paid the' bill have^not only
- 11 been the NRC but, more importantly, or even more ;
12 significantly, the utilities themselves. But you don't 13 really get full value from that database unless it has.
14 been analyzed with the risk focus, and that's what you're l 15 doing. i
-i
'16 MR. BARANOWSKY: Yes.
17 VICE CHAIRMAN SEALE: And I-think that's a 18 very important part of closing the loop, if you will, on l 19- that experience.
l-l 20 Are there any comments or questions from any p
21 other members of the group?
22 MEMBER BARTON: Pretty good.
23 CHAIRMAN KRESS: Yes. I'd say it's right now.
j- 24 MR. BARANOWSKY: Someone did ask about the 25 reliability data rule. I might just give you a one-minute L
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISt.AND AVE., N W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 2344433 i.
,v, ,m. - --- - -
.m _. . . _ _ _ _ _ . _ _ . - . _ _ _ . .__ _ - _ . __ . _ . .
I 156 1 synopsis on that.
7.,
2 CHAIRMAN KRESS: Yes, I was wondering what the b 3 status of that was.
4 MR. BARANOWSKY: Reliability data is important 5 because what we're getting from LERs is so. sparse and 6 spaced that we can't'see trends occurring quickly enough.
7 And also, it is hard to get plant-specific results that i
8 have enough certainty on it that we're not sure we're off 9 by some amount.
10 So the reliability dat e that we're looking 11 for, and had put the rule out, would help there. And what 12 has happened over the last several months is that INPO has l
13 offered to allow us to take a look at their safety system
_f s 14 performance indicator data to see if it could be used in 15 place of what we are calling for in the rule and be )
16 provided to the NRC under some voluntary umbrella, much 17 like NPRDS.
- 18 Whether it would be proprietary or not, I'm 19 not completely sure, because I have heard NEI folks say 20 before the ACRS that it wouldn't be proprietary, but it is 21 currently proprietary at INPO. And what we're going to do i
22 over the next couple of months is look at that data and 23 see if it could be used, or supplemented, to provide the !
l 24 information necessary to do reliability analysis.
25 It is not a reliability data set per se, so'it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
157 l
1 may need.some modification. And we expect to complete i
2 that evaluation probably in the end of February or in (O
l 3 March, and it.will also include some meetings with 1
l 4 industry to go over why we think it has some weaknesses, 5 and then prepare a Commission paper with a proposed 6 position on that data, the voluntary approach, and 7 rulemaking. So this spring I think --
l 8 VICE CHAIRMAN SEALE: So you're evaluating and- .j i
9 getting ready to negotiate. !
l 10 MR. BARANOWSKY: Right.
11 VICE CHAIRMAN SEALE: Okay. Very good.
I 12 Any other comments?
l 13 I want to thank you and your colleagues.
14 Again, we are very happy to have you come talk to us any 15 time you are prepared to do so. Our experience is that 16 you usually have something to say when you do come to see 17 us. So we look forward to doing it again at the next 18 convenient time.
19 MR. BARANOWSKY: Thank you.
20 VICE CHAIRMAN SEALE: George?
21 MEMBER APOSTOLAKIS: It's not for Pat.
22 VICE CHAIRMP.N SEALE: Oh, okay.
23 MEMBER APOSTC.LAKIS: Is this list of l-l 24 comments / questions from the PRA subcommittee?
25 MR. MARKLEY: I put those together, George. l
! l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 2000# 3701 (202) 234-4433
158 1 MEMBER APOSTOLAKIS: Are they a part of the
- 2 record now?
)
3 MR. MARKLEY: No. They were just for the 4 people's benefit who did not have a chance to attend the 5 meeting.
6 MEMBER APOSTOLAKIS: For the members only.
7 MR. MARKLEY: Right.
8 VICE CHAIRMAN SEALE: We will adjourn until, 9 let's say, 1: 00.
10 CHAIRMAN KRESS: Let's recess.
11 VICE CHAIRMAV SEALE: Yes. I mean, recess.
12 Sorry. We're recessed. Thank you.
13 (Whereupon, at 12:27 p.m., the proceedings in
(_)
U 14 the foregoing matter went off the record.)
15 CHAIRMAN KRESS: Can we get started again, 16 please? The next agenda item is the revised source terms i
17 for operating reactors. Our cognizant subcommittee 18 chairman is Mario Fontana, so I'll turn the floor over to l
19 him.
20 MEMBER FONTANA: I'm going to present a review 21 of what took place yesterday.
22 The Severe Accident Subcommittee met on the 1
23 afternoon of November 6, 1996. Members in attendance were l l
24 Dana Powers, Bill Shack, Tom Kress, Bob Seale, and myself. l
,n
) 25 I'm Mario Fontana, chairman of the committee. Cognizant l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
i 159 i i staff enginear was Jit Singh.
,-~
, 2 The purpose of the meeting was to gather
'~
3 information on the proposed SECY paper concerning the 4 approval of license applications using revised source 5 terms at operating reactors. The Subcommittee heard from 6 representatives of the NRC staff, the Nuclear Energy 7 Institute, EPRI, and Entergy Operations, Inc.
8 Some background. During the September, 1994, 9 ACRS meeting, the Committee discussed the proposed final 10 version of NUREG-1465, " Accident Source Terms for Light 11 Water Nuclear Power Plants" and issued a report to l 12 Chairman Selin.
13 The report " urged that the risk implications I )
k/ 14 be evaluated and consideration be given to allowing 15 current licensees the option of using the timing 16 assumptions in a proposed source term without performing a 17 complete source term reanalysis."
18 The report also stated that the Committee 19 believed that using realistic source terms could reduce 20 operational plant risk.
21 On November 15, 1995, NEI submitted EPRI 22 report TR-105909, " Generic Framework for Application of 23 Revised Accident Source Term for Operating Plants" to the 24 NRC for review. Four licensees have submitted licensing
(~~^;
\ ,/
' 25 amendment requests, which the staff is reviewing as part l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
160 1 of a pilot program. In a Staff Requirements Memorandum,
~s 2 SRM of July 2, 1996, the staff was requested to return to
(
k
)
' '/
3 the Commission for guidance as it proceeds, and not to 4 proceed down a path of de facto exemptions if rule changes 5 are necessary.
6 In responding to the SRM, the stafi prepared a 7 proposed SECY paper concerning its proposed action plan.
8 Dr. Richard Emch presented the staff report on the 9 proposed SECY paper, which primarily responded to the NEI 10 Generic Framework Document which was submitted for review 11 in November of 1995.
12 The final NUREG-1465 was issued in February of 13 1995. The NEI Generic Framework document submitted to NRC g
k_sl 14 for review in November of 1995. The CRGR briefing in 15 October of 1996, and a Commission paper due in November of 16 1996.
17 Plant changes were classified into four 18 groups: Allowable leak rate changes, isolation valve 19 timing changes, filtration unit simplification, and 20 mitigation system actuation timing.
21 Four principles of the GFD, the generic 22 framework document, are one, existing licensing basis is
! 23 acceptable.
24 Two, complete implementation of a new source
/-~
() 25 term as a substitute for existing licensing basis is NEAL R. GROSS
( COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
, (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
l 161 <
1 acceptable. But under the conditions that you need an 2' integrated' assessment and removal of accident mitigation I
- 3. hardware.
4 .Now in the early version of the SECY paper-5 that we received, the removal of-hardware was frowned on.
6 But now it appears it is more agreeable, that it is 1
l
-7 justified.
8 Third, is selective implementation of the 9 revised source term is acceptable =for' timing only 10 applications. But if those calculations are necessary, 11 the use of the entire revised source term is needed. I L12 Fourth, dose calculations-based on existing 13 models.and limits are acceptable. A new analytical 14 framework, including TEDE and any two-hour dose evaluation 15 period may be necessary. Part 100 does not require a.
16 specific source term, but references TID 14844.
l l
17 Now there is a possible need for 2n11e making 18 and exemptions. That is for phased timing of the revised l l
19 source term, the existing part 100 requires dose 20 evaluations of two hours immediately following onset of 21 release. There are additional radionuclides in the l l-j- ~22 revised source term. Existing part 100 requires i
l 23 evaluation of whole body and thyroid doses.
l 24 Now there is a rebaselining effort going on.
i
() 25 The intent is to use Grand Gulf and Surry. These are two NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
i 162 i
1 representative plants. They are also NUREG-1150 planto, 1
- 2 which allows _ comparison with significant prior work. l
\/ ,
3 Those consequences of a new source term, !
4 sensitivity studies, and the effect on other' design and ;
i 5 licensing basis impacts, and the assessment of-severe !
l 6 accident impacts would be required. l 7 The pilot plants are Browns Ferry, Grand Gulf, 8 Indian Point 2, Perry, and Oyster Creek. The schedule is l
-l 9 to complete baselining and the decisions on the need for- .
i 10 rule making and exemptions by mid-1997, and complete pilot i 11 plant reviews by late 1997.
12 The staff and the industry have converged '
13 somewhat since the early draft of the Commission paper !
I O- 14 that had been given to us. An important convergence is l
15 that some removal of equipment, such as charcoal filters, f t t l
16 may be agreeable, whereas the initial draft prohibited l 17 this on a basis of maintaining defense in depth.
i l
18 A key convergence is that the staff may now i
19 agree that a rule making may not be necessary to implement j 20 improvements based on the new source term. The final cell 21 on this probably will be within the Office of General- l 22 Counsel. !
l r 23 The staff briefed the CRGR and indicated that i 1 24 the proposals _were received well, but CRGR wanted to l(::) 25 review the cases later when implementation is about to
-l, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS ,
1323 RHODE ISLAND AVE., N.W. l
! (202) 2344433 WASHINGTON, D.C. 2000.)-3701 (202) 234 4433 i l
163 ,
i l 1 start. ,
l l l l f- g 2 Ms._Sheri Mahoney presented the "Entergy ,
\~s/ !
3 Viewpoint on Revised Source. Term Implementation." The key ;
4 points of Ms. Mahoney's presentation is that rule making l
5 is not necessary. It would cause a great delay in 6 implementing safety improvements in plants other than the ;
i i 7 pilot plants. [
! 8 A particularly important point was that the-9 safety benefits that could accrue from elimination of the ,
10 automatic isolation function of selected primary and .
11 secondary containment penetrations, particularly.those
- 12 that do not communicate directly with the containment 1
13 atmosphere, such as instrument air lines, fire water i
(~~ \
14 penetrations, and service water penetrations. [
15 At present, these are actuated on a LOCA i
16 signal. In practice, the operators must manually override 17 the automatic closure after the fact to allow performance. ,
i 18 of important safety' functions. This is distracting at a l 1
19 time when operators should be concentrating on !
l 20 accommodating the event. Entergy's proposal would be to )
21 allow manual isolation from the control room as plant 22 conditions dictate.
23 In concluding, Entergy requests that the ACRS l
l 24 endorse early implementation of the revised source term 25 through 10 CFR 50.59.
i NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASCNGTON, D.C. 20000-3701 (202) 234-4433 l
164 1 Mr. Kurt Cozens presented the HEI position
,- 2 regarding implementation of the revised source term. The f ;
~#
3 NEI approach would be to perform parallel calculations for 4 Part 100 using a new source term, but following Part 100 5 requirements. That is, whole body and thyroid doses for 6 zero to two hour release, primarily for regulatory 7 compliance and a TEDE and maximum release in any two hours 8 for technical information.
9 Parenthetically, our committee has been 10 concerned that performing a calculation for zero to two 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> with the NUREG 1465 source term, which has a more 12 realistic release timings, might miss a time window during 13 which significant release could occur, in principle.
) i
\/ 14 Dr. David Leaver presented key elements of the 15 NEI/EPRI " Revised Design Basis Accident Source Term 16 Methodology and Applications." This was an update, the 17 presentation was an update on the EPRI report TR-105909, 18 " Generic Framework for Application of Revised Accident 19 Source Term to Operating Plants."
20 This work started in support of the Utilities 21 Pequirement Document for advanced light water reactors and 22 wac extended to apply to operating plants. Three pilot 23 plant applications were submitted in 1996 and several more l
l 24 are expected in 1997.
,7
- ( ,) 25 The four principles summarized previously in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
165 1 Emch's presentation were modified and two additional ones 2 added. A_fifth one to apply to source term for TEDE 1')
3 calculations for control room dose to meet GDC 19, and the s
-4 sixth, to give considerations to the integrated safety-5 impact of aspects of the plant licensing basis as affected 6 by the revised source term with associated plant. changes. l l
1 7 The~ updated framework document methodology '
8 calls,for use of NUREG 1465 for timing, chemical form, and l
9 release magnitude, to address removal methodology 1
-i 10 generically with main focus on airborne and aerosol l 11 removal. And to assure that removal methodology is 12 sufficiently conservative that'when combined with 1465 13 releases and conservative DBA methodology and existing
! s 14 licensing limits, a level of conservativism is maintained.
15 That is, considering the changes that may be put into L 16 place.
l 17 Completed applications include Browns Ferry,
- i
- 18 (that's a BWR Mark 1), Perry, (BWR- Mark 3) , -Indian Point l
l 19 2, (PWR large dry containment), Oyster Creek, (old BWR I
l 20 Mark 1). These are in progress -- no. The Oyster Creek 21 one is in progress.
I 22 -For Browns Ferry, plant design changes which !
l l
23 were evaluated included: increased main steam isolation
!. 24 valve leakage, fission product holdup and retention in
- (jO 25 main steam lines and main condenser, which was made NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISt.AND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
166 1 seismic and therefore could be counted as operable.
3 2 No credit for reactor building charcoal. No
( }
3 credit for control room emergence ventilation system 4 charcoal, and PH control is under evaluation.
5 For Perry, plant changes analyzed were:
6 elimination of the MSIV leakage control system, increase 7 in isolation valve leakage, aerosol retention in the main 8 steam line with no credit for condenser, delay in control 9 room recirculation, reduced credit for control room 10 charcoal, charcoal filters, and increase allowable 11 containment bypass leakage.
12 For Indian Point, removal of post acciaent in 13 containment filters, HEPA and charcoal. Elimination of x' 14 spray additive This would include plant changes to 15 achieve PH adjustment by trisodium phosphate in baskets in 16 the sump. Control room ventilation was modified, would be 17 modified to pressurize the control room.
18 In all the above analyses, the resulting doses 19 were well within requirements. A key point is that 20 existing regulations are adequate for application to 21 operating plants. Sxemptions are not required. I am 22 still talking about Dave Leaver.
l 23 Now my opinion is that the use of the new 24 source term for these applications should be encouraged
,, y
(,,) 25 because of the greater realism compared with TID 14844 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTCM, D.C. 20005-3701 (202) 234-4433
_ . _ . . ___m... . . _ . _ . _ _ _ . _ _ _ . _ _ _ . _ _ _ . _ _. . _ _ ..__ _ _
167 [
i 1 numbers. Howsver, patching the source term onto existing l i
s 2 regulatory requirements is inferior to performing best !
\ !
3 estimate level III PRAs and evaluating impact of changes !
4 on a much more logical basis.
5 It was decided that this was a very important 6 issue that will be followed through the Source Term l 7 Subcommittee and the Severe Accident Subcommittee. The t
8 next meeting between the staff and the Joint Committee i
1 9 will be scheduled when sufficient work has been completed 10 to warrant a progress review. 'Jit Singh will coordinate' 11 with the staff for implementation. That's the end of my t 12 review.
13 CHAIRMAN KRESS: Very good, Mario. Do any of (b
14 the members that weren't here at the subcommittee meeting 15 wish to have any clarification of what Mario has 16 summarized or feel pretty confident? Okay.
17 Thank you, Mario.
l 18 VICE CHAIRMAN SEALE: I have a question. I 19 thought about the decision or I guess it was really a 20 consensus that we reached, +. hat we not write a letter at !
! I 21 this time.
I
- 22 MEMBER FONTANA
- Correct.
i i l \
! 23 VICE CHAIRMAN SEALE: And I was concerned that j i
' 24 after thinking about it, that we may be remiss in not 25 writing a letter, because this is an instance which you NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS
( 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
_ . _ _ _ _ . . . _ _ _ _ . _ . _ . - . - _ _ - _ _ ~ _ . _ _ _ . _ _ . _ _ _ _ . . .
l 168
-l
~1 know, in a way is as fundamentally important as the idea !
2 of perhaps moving to risk informed regulation. j O 3 Namely, the Commission, the Department of i
t 4 Energy, and the industry have all spent a lot of money in j i
5 putting the source term issue on a much firmer technical' !
- 6. footing than it was when TID 14844 was written. I wonder i 7 if we shouldn't encourage some immediacy in putting that l l
8 into the basis for regulation rather than to continue to 'l 9 rely on things which were at best a best estimate or-an- ,
i 10 estimate, anyway. They weren't e.ven a best estimate. !
t 11 They were a set of numbers. I guess vapor pressure is the l l
12 closest correlating factor.you could figure out for~them. !
13' Anyway, I just wonder if it wouldn't be i
l 14 worthwhile to point out that when better data are' {
15 available to serve as the basis for regulation,-that-there <
l !
16 shouldn't be -- there should be a more deliberate _ effort- !
i 17 to get those data to be the basis for the regulation. l 18 CHAIRMAN KRESS: Well, that's r. good point,
! 19 Bob. I find the timing of this a little unfortunate
, -20 because this to me would have been an ideal place -- this i
21 would have been an ideal place to check out the proposed L 22 new way to implement risk basis to the regulations.
23 But when we heard a few meetings ago from Gary 24 Holahan the proposed method of doing this was someone j 25 comes in for a requested change in the licensing basis, l !
l NEAL R. GROSS i COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
169 1 that they have a way to do that on a risk basis, thio 1
2 would be an ideal test to that system to me. They are not 7- s
.Q 3 quite ready for that yet. Yet there seems to be some 4 urgency for getting this through at this time. So we have 5 to balance.
6 Really the correct way to do it is to look at 7 the effect of these changes on the overall risk status, 8 and whether or not the increase in risk that one gets, 9 where there's nothing clear that you get an increase in 10 risk, whether that is still an acceptable increase taking 11 in view the great benefit that one gets from it. That's I
12 precisely what that technique was intended to look at, 13 those sort of things.
14 So I would have loved to have this been a 15 pilot test of that. Apparently, we can't wait that long 16 because there's too much of an urgency for wanting to get 17 this approved.
18 What do you think about that thought, Mario?
19 MEMBER FONTANA: There were several thoughts 20 in there. There's no reason why you can't go ahead and l 21 implement something like this and then do the kind of 22 analysis that you are talking about to determine how a 23 more fundamental risk analysis would have rated the t
! 24 changes.
() 25 CHAIRMAN KRESS: There's one reason, to see --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
170 1 wa don't know whether or not you meet your acceptanco 7w 2 criteria. You are guessing.
! i 3 MEMBER FONTANA: Yes. Well, there are things 4 that I haven't mentioned here. It looks as if the changes 5 are not -- the results and doses that you calculate and 6 things like that are well within the standards and about -
7 CHAIRMAN KRESS: Yes, but those are in design 8 basis specs. I'm talking about a real risk analysis.
9 It's difficult to connect design basis based to real risk 10 based.
11 MEMBER FONTANA: The question is, it looks 12 like these are real improvements and should you hold them 13 up or do you wait for some PRAs to be done. Why can't you
( )
'/ 14 do them and do the analysis later. If the analysis shows 15 that -- I think it's very unlikely to have subsequent 16 analysis which is going to use increase to risk 17 significantly to the point where you had to do something 18 about it.
19 CHAIRMAN KRESS: Are you making a judgement 20 that the risk analysis is supposed to give you the answer.
21 So I didn't know what the answer to that is.
l 22 MEMBER FONTANA: What I am saying is that I ,
23 don't think you want to wait several years. You go ahead )
24 and do what you want to do, but go ahead and start the l gy
'(_ ,/ 25 wheels rolling to get these going now.
l NEAL R. GROSS ;
' COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 2344433
171 1 CHAIRMAN KRESS: That's one approach, ycc.
f-~ 2 MEMBER FONTANA: It all costs money. It's G'
3 parallel analyses is what they are.
4 CHAIRMAN KRESS: Once you have a rule in 5 place. Well, if you go ahead with the concept that a rule 6 making isn't required for this, then you will probably be 7 better off because you don't have to change the rule.
8 MEMBER FONTANA: That decision probably is 9 going to end up in the Office of General Counsel I think, 10 because it's not a technical decision.
11 Well, the question that Bob raised was should 12 we be writing a letter. Yesterday we thought maybe we 13 didn't need a letter. The staff said well, they didn't kJ 14 feel like they needed a letter.
15 So to proceed, I don't know whether a letter 16 is really necessary to speed things up or not.
17 CHAIRMAN KRESS: Anybody else?
18 MEMBER FONTANA: At some point we have to 19 write.
20 CHAIRMAN KRESS: Have any thoughts on whether 21 we need a letter or not?
22 MEMBER POWERS: It was just my impression that i
23 the planning and what not was still rather preliminary, 24 that we're just getting started. I didn't feel I had much 7
(_) 25 to contribute to your --
NEAL. R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
! (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
172 1 CHAIRMAN KRESS: In the way.of a letter.
2 MEMBER POWERS: You know, what to say about O 3 the plans at this point, because they are still kind of 4 feeling their way around here.
5 I thought they had resolved some of the major.
6 hurdles between themselves and the industry to the best of 7 their ability at this stage. I didn't think I could 8 contribute much.
-9 CHAIRMAN KRESS: So your feeling is, we still 10 don't need a letter?
11 MEMBER PC'4ERS: Yes.
12 MEMBER FONTANA: The letter would be kind of 13 an attaboy because I don't think it would say to do 14 anything significantly different other than you know, it's 15 legitimate doing the PRA, but we keep saying that anyway.
16 CHAIRMAN KRESS: Well in fairness, I
~
l 17 understand they plan on doing that.
18 VICE CHAIRMAN SEALE: Well I just -- my point 19 was really more to the fact that this money has been 20 spent. It just seems to take an inordinantly long time to 21 develop not only the process but the will to fight the i 22 process to incorporate the results into the regulatory --
23 MEMBER FONTANA: Well, nobody seems to want to 24 tackle the rule making.
[ 25 VICE CHAIRMAN SEALE: Yes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
[ 173 1 MEMBER FONTANA: If you don't tackle them --
2 VICE CHAIRMAN SEALE: It will just stand there :
I 3- forever. ,
i l
4 MEMBER FONTANA: -Stay there forever. I
'3 VICE CHAIRMAN SEALE: If you don't rain on it, j I
6 it won't wash away. ;
7 CHAIRMAN KRESS: Well, do we have a position !
8 on whether a rule making is needed on whether the 50.59 9 process is sufficient? !
10 MEMBER FONTANA: I don't know enough to tell. !
11 CHAIRMAN KRESS: That would be the basis for, !
i 12 in that area, if we had a position on that. [
13 MEMBER FONTANA: My position would be to let l C- 14 them try and see what happens. i i
15 VICE CHAIRMAN SEALE: Let them try the 50.59?
i 16 MEMBER FONTANA: Yes. !
17 MEMBER POWERS: In light of our due process ;
i 18 for preparing letters and what not, this is an area where 19 if a member or a group of members have ideas on what a l
l 20 letter should look like, that they should prepare a draft i i
21 that we can discuss at the next meeting. ,
22 CHAIRMAN KRESS: I think that's an excellent i
23 suggestion, Dana. Let's leave it at that then. I think i
24 that's a good way to do it. !
! m i 25 So we won't have a letter at this meeting. f i.
l NEAL R. GROSS ;
1 COURT REPORTERS AND TRANSCRIBERS L 1323 RHODE ISLAND AVE., N.W. !
WASHINGTON, D.C. 2000# 3701
(. (202) 234-4433 (202) 2344433 {
_ ._ q 174 .!
i I
1 VICE CHAIRMAN SEALE: It's not necessary.
l 2 CHAIRMAN KRESS: So at this time, we are --- i O ~3~ MEMBER FONTANA: Am I directed to draft i
-[
4- something for the next meeting? l 5 CHAIRMAN KRESS: No. Only if you want.to. ,
6 Other members'can feel free'to give you suggestions.on !
t
. i 7 what not to.go into it if they feel like the letter is i l
8 needed. l 9 MEMBER FONTANA: I might draft one anyway. f i
10 CHAIRMAN KRESS: 'Okay. If you do draft.one, l
11 we need to get it to the members befr.re the next meeting.
12 At this. time, we are at a point in the agenda i
13 where I think we-can make use of some extra time.to work' -
( 14 on one of our letters.
15- (Whereupon, the foregoing matter went off-the L !
- 16. record at 2:03 p.m. and went back on the 17 record at 3:20 p.m.)
18 CHAIRMAN KRESS: Okay. I think we'11 go ahead l
19 and get started a little early on the next agenda item.
l I
20 It looksLlike I can give this one to you, Bob 21 Seale. Is that right?
22 VICE CHAIRMAN SEALE: I guess. I must say 23 that if leadership has as a hallmark the ability to i
- j. 24 designate or delegate, you guys sure got caught on one 25 record spot this time. That wasn' t a recorda' ole remark NEAL R. GROSS court REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
i l (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 l
I , _ _, _. - _ .
175 1 anyway, so --
.q .2 _All right. We're going to hear this afternoon
, b l
i 3 about emergency planning for advanced reactors. _This is 4 an issue of course that many_of us have had a great deal l
l 5 of interest in. There were some questions that remained l ;
! 6 about some of the detail I guess back when we were doing I I
l- 3 l 7 or we looked at the ABWR and syste.n 80 plus, that was :
8 something that was in the bag that went to the applicant.
l 9 So now apparently, we are going to hear a little bit more 10 about NRC's evaluation criteria.
1 11 Who is going to start off for the Commission? t l
l 12 Staff, okay. ;
13 Very good. Sir, we'll let you have at it.
14 Then we are going to hear some comments from'the NEI as 15 well.
\
16 MR. ESSIG: Before Jim gets to his j 17 presentation, my ncme is Tom Essig. I am the section l
18 chief for Emergency Preparedness and Environmental Health 19 Physics at NRR. We have in your package, there are three l l
l 20 introductory slides. We do not plan to go through those l
)
21 in any kind of detail, except to note that this started as l l
l 22 an issue for the advanced plants. Mostly they were non )
1 l
l 23 light water reactor designs, MHGTR, et cetera.
24 The issue that we are going to talk to you ,
25 about today has been reduced in scope based on a February
! NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS I
1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
176 1 1995 memorandum to the Commission, wherein we narrowed tha
! ,._x 2 Commission's question that they had asked us to look at
( )
'~
3 for EP simplification to the passive and evolutionary 4 designs. That's really what we would like to talk to you 5 about today.
6 So with that, Jim will begin with slide number 7 four.
8 MR. O'BRIEN: Actually, I'm going to start I
9 with slide one. l 10 MR. ESSIG: Okay. Slide one, l
11 MR. O'BRIEN: Describe the overview of what I I i
12 am going to talk about today, and then we'll go right to !
13 slide number four. i
<- 3 ,
N-) 14 I have got a lot of slides. I understand I 15 have about 25 minutes to go through all of them, so I am 16 going to try and -- I think it's important that we hit 17 each one, but I'll do it in perhaps less detail unless you 18 guys have questions on them, we'll stop and we'll go into 19 more detail.
20 The overview of the presentation is going to 21 be I am going to discuss -- well, actually, Tom already 22 discussed the background of the Commission's request for l
i 23 evaluation of EP for advanced reactors. I won't go into 24 any more detail on that.
ry
( ,) 25 I wanted to highlight what the objective of
! NEAL R. GROSS
( COURT REPORTERS AND TRANSCRIBERS
! 1323 RHODE ISt.AND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
- .-- = _ - . . .. . . .. - ._ . . . - . .
177 1 our evaluation was. I am going to talk about the steps 2 that we performed in our evaluation. I'll discuss the j 7-s V 3 preliminary conclusions on EP simplification for advanced 4 reactors. And lastly, ;o over the response to ACRS j 1
l 5 questions that you provided us in I think it was 1994, l
6 when we briefed you on this issue, and in particular, on a 7 CE 80 plus calculation, pad calculation.
8 Now I am going to go to slide number four. j 1
9 The objective of che evaluation was to respond to the 10 Commission's request in a 1993 SRM to submit to the j 11 Commission recommendations for proposed technical criteria l.
l 12 and methods to use to justify. simplification of existing 13 emergency planning requirements for advanced reactors. I ,
14 have highlighted the words recommendations, technical 15 criteria, and methods. That's what we are.looking at in L
i 16 our evaluation. That's what we have presented here.
l 17 What we did in our staff evaluation. First we b 18 reviewed the basis for current EP requirements as 19 presented in NUREG-0396. We looked at some of the 20 implications from updated risk and source term studies.
i 21 We reviewed Commission and NRC's decisions on challenges 22 to the EPZ and the basis for those decisions, because they 23 provided more body to the basis.
- 24 We looked at industry reports provided on EP
) 25 simplification for passive and evolutionary reactors. We NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. ,
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
.. _.. _ _ _ - _ _ .. _ _ _ ._ _ _ . _ _ _ _ _ _ . . . - m . _ _ . .
178 :
1 evaluated the implications from passive and evolutionary 1
-- 2 PRA results. Lastly, we looked at'what EP requirements 3 could possibly be reduced or simplified. f 4 A lot of slides are on technical basis for- l l
5' current EP. We believe those basis are very good. The l l
~
6 report was I think very well done, the NUREG-0396.
7 NUREG-0396, the study was started in 1976. It l
-B was performed by a task force.of NRC and EPA staff. The i 9 purpose of the report was to provide a basis for federal, l 10 state and local government emergency preparedness l 11 organizations to determine the' appropriate degree of 1 12 emergency response planning efforts. The issues addressed
' 1L3 in the report.were the size of the EPZ, emergency planning ;
l
. 14 zone, the timing of releases, and characteristics of
?
15 releases. !
E16 The most important issue that was addressed 17 was the size of the EPZ, because that determines the areas !
18 over which the planning should be carried out. It really L
L 19 drives a lot of the other issues with emergency planning.
i l 20 When the task force looked at how to determine j 21 what the size of the EPZ was, they looked at several 22 rationales upon which emergency planning could be based.
, 23 They looked at a risk rationale, a probability rationale,
- 24 cost / benefit rationale, and a consequence rationale. The
. () 25 rationale chosen is very important because it drives the I
! NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 433 - WASHINGTON, D.C. 20005-3701 (202) 234 4 33
4 179 ,
1 1 criteria that was developed and the methods that were j l1 f3 2 developed. i
$ i l Ns / ;
, 3 The criteria that was chosen, or the rationale ;
r .
t 4 that was chosen was -- maybe I need to spend a little more l 5 time on the slide before. I'm sorry. Was the consequence ;
i
~
6 rationale. ;
i 7 They looked at risk. They determined that i i
8 perhaps the degree of planning should be comparable with 9 the degree of planning that's made for similar events that i l 10 occur, not nuclear type accidents. They determined that ,
i 11 there was no risk basis for those type, the emergency _ ,
8 12 planning and degree of it for those type events, so they
! 13 discounted the risk rationale.
i i(~N ;
i N ') 14 Secondly, they discounted the probability E
15 rationale for the same reason, that there is no -- that 16 society would accept probability or no emergency planning i 1 l 17 for more probable events, but for emergency plann- I or 18 for nuclear events, there's a perception that has to be )
I 19 involved and taken into account. It's the perception of )
L i
! 20 the risk and what could be done that was driving the l l 21 requirements, that should drive the requirements. j
! l 22 Lastly, they looked at cost / benefit. Again, !
23 they had problems with using that rationale, addressing I l
24 what the cost of EP is and moreso, addressing what the
() 25 benefit of EP is. l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 2000>3701 (202) 234-4433
180 !
I 1 So the task force chose the consequence _
f ,3 2 rationale. They thought that was a good one to scope the ,
bu -] 3 areas over which the planning should be performed. b 4 VICE CHAIRMAN SEALE: It would be interesting ;
I 5 to know whether you picked one basis and then discarded 6 it, and-then the next'and discarded it in series, or 7 whether you listed all four and then went down the list. l 8 That's just a jive, I'm sorry. 2 9 MR. O'BRIEN: I actually probably think that 10 it was the risk first.
11 VICE CHAIRMAN SEALE: Yes.
12 MR. O'BRIEN: Okay, the criteria. This is 13 important because this is what the Commission requested us 14 to look at, technical criteria and methods.
'15 The criteria that was used in 0396 and is.used 16 for current plants was the EPZ should encompass those l 17 areas where projected dore from a design basis accident l 18 could exceed protective action guidelines or essentially 19 one rem whole body, five rem thyroid.
20 That was the subject of a briefing that was 21 performed in 1994. If you have any more questions on l
22 that, I can answer that.
23 The consequences -- the second issue is the 24 consequences of the less severe Class 9 core melt y ,) 25 accidents. The EPZ should encompass those areas where the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
181
.1 consequences from these Class 9 accidents could exceed EPA
-g- '2 PAGs.
3 The second criteria'--
4' VICE CHAIRMAN SEALE: Remind me. What are the j o
-6 .MR. O'BRIEN: Okay. There's one rem, whole 7 body --
i 8 MR. ESSIG: It's actually total effective dose 9 equivalent now.
10 VICE CHAIRMAN SEALE: Yes, I understand. -It's 11 the TEDE now.
12 MR. O'BRIEN: Yes. It's TEDE now and SEDE.
13 VICE CHAIRMAN SEALE: But it's the same level.
14 MR. O'BRIEN: Yes.
15 VICE CHAIRMAN SEALE: Okay. l l
16 MR. O'BRIEN: .In 0396, there's actually more 17 of a range, one to five rem, whole body, and it's now more 18 specific to one rem TEDE.
19 EPZ should be of sufficient size to provide 20 for a substantial reduction in early. severe health effects 21 in the event of a more severe Class 9 accident, j l 22 Lastly, the detailed planning within EPZ was i
! 23 expected to provide a substantial base for ad hoc )
l-24 expansion of response-efforts in the event it was l l
25 necessary.
! 1 L NEAL R. GROSS l
! COURT REPORTERS AND TRANSCRIBERS l 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4 33
L i
182 1 The methods. The 0396 task force did a study i i
2 to determine the fraction of plants which exceeded PAG i
3 levels at various distances from DBAs when they evaluated l l ,
! 4 FSAR data to come up with that number.
5 Why don't we go right to the next slide. We r
l 6 can come back to this one, Tom. The next slide shows the 1
7 results of this evaluation. You can see that the whole
, l 8 body dose and the rems, one, two, three and so forth, ,
l 9 distance from the site, and areas where 50 percent of the ;
l i 10 plants did not exceed this dose level. Here's where 10 1
l 11 percent of the plants or 90 percent of the plants did not 12 exceed this dose level.
l 13 Down here is 10 miles. You can see for one A
s 14 rem at'about maybe several miles, 90 percent of the plants l
15 would not exceed that from a DBA accident, type accident, l ..
16 and all the conservatisms that go into those calculations.
17 The next method was the test where it uses i
18 data from the Reactor Safety Study and input into the CRAC 19 dose assessment model to in turn to obtain the probability 20 of exceeding certain dose levels, 200, 50, 5 and 1 rem L
j 21 doses as a function of distance.
22 We might as well go up to that slide, to show. l 23 what the results were of that. Some of you might be ;
24 familiar with this curve. It's been used in a lot of EP
() 25 evaluations.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 2000 # 3701 (202) 234-4433
183 1 Essentially it shows that the consequences R 1
. . I s 2 drop off at about 10 miles. This was interpreted. Now j 3 it's.not exactly where it drops off. This curve is a hand
-4 interpolation of data. That's what it' indicated.to the 5 task force, how it was used to develop the size of the 6 EPZ.
7 Okay. We might as well go into the next 8 slide, because this shows this aspect of this curve in a 9 little bit different fashion. This'is also from the 0396
-10 study. It shows the probability,.the same as the last 11 curve essentially shows the probability of exceeding-12 certain dose levels with distance from the site.
13 The difference between this one and the last r'N ,
14 one is that the source terms, the accident types have been !
1 15 broken out into what essentially is on thisicurve is core l 16 melt accidents with release via' melt through of the 17 containment.
i 18 This one is core melts with accident, excuse
~19. me -- with release to the atmosphere, directly to the l l
20 atmosphere.
21 over in this one, you can see the whole body i l
22 dose for five rem, which was PAGs at the time that 0396 i
-23 was, and one rem, what they are now. It essentially goes
! 24 down at about 10 miles. That can be looked at as the less
!'( ) 25 severe core melt accidents, which was one of the criteria NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 ;
.- - - - - ~ ~ .- .-. . = . _ - - _ . - - . _ . .. . _ _ - . . -
184 ;
i 1 that'was ' coked at in 0396.
2 'This over here is the more severe core melt !
O
& 3 accidents. Obviously the consequences are much greater, 4 especially when you look at much larger distance. These l 5 are one and five rems going cut there that large distance.
I 6 So the criteria for the more severe core ~ melt 7 accidents was not to exceed the PAGs, but to prevent 8 severe health effects by having the emergency planning. i 9 Now we can go onto the next slide. It {
10 explains that in more detail. This is also from the 0396 11 report. This describes or indicates how EP with benefits. i 12 get out of EP as far as the mean number of early 13 fatalities at selected radial distances. This is from
'\
( 14 zero.to five from the plant, this is five to 10.
15' The number one here corresponds to no 16 emergency planning. The number two - let's go right to 17 the evacuation. The number five here corresponds to 18 evacuation at five hours, with a five hour delay. The 19 number seven is evacuation with a one hour delay.
20 So it gives you an idea of the benefits you 21 get out of emergency planning. Obviously you get the most 22 benefit from zero to five miles. You still get a good 23 benefit from five to 10. When you get down to 10 to 15, i
24 you get much less benefit.
O Q 25 Number two, now if you go back to number two, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(' (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
185 ,
1 number two was for sheltering, in the bsnefito you get in l
2 sheltering. You do get quite some benefit from
.( 3 sheltering. You see when you get down here, it doesn't I
I 4 matter so much if you evacuate or shelter. You.get about 5 the same kind of protection to the public as far as early j 6- health effects or fatalities.
7 Go onto --
8 MEMBER FONTANA: If sheltering is a six hour 9 exposure, it says here --
10 MR. O'BRIEN: Yes.
11 MEMBER FONTANA: Does exposure include moving 12 them out from where they are sheltered to somewhere else 13 after six hours, presumably you-have to go over O'
k-s/ 14- contaminated ground?
15 MR. O'BRIEN: Yes. You would have to go.
{
l 16 MEMBER FONTANA: Is that in there?
17 MR. O'BRIEN: I'm reading to see. I'm not i
- 18 posacive. I believe it is, but I am not positive. )
19 MEMBER FONTANA: It would be logical.
- 20 CHAIRMAN KRESS
- Why did the number of f ,
l 21 protected fatalities increase in the 15 to 25 miles? l l
22 MR. O'BRIEN: This is because of the size.
23 This is a 10 mile radial distance. This is five mile
! 24 radial.
- fg
, \q_,) 25 CHAIRMAN KRESS: I nee. You are adding in a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBEF, 1323 RHODE ISLAND AVE., N W.
i (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
186 1 lot to your distance.
7-s 2 MR. O'BRIEN: That's correct.
G 3 The conclusions that we have drawn from using 4 all this information was that, and I actually didn't 5 provide as much information as I probably should because I 6 didn't provide information on the ingestion pathway, but i
i 7 at the same kind of considerations went under that.
8 The conclusions that were drawn were that a 10 9 mile radius is appropriate for the plume exposure in the 10 planning zone, and a 50 mile radius was appropriate for 11 the ingestion pathway emergency planning zone.
12 I am just going to really briefly go over the i
l 13 other issues that the 0396 task force looked at. That was '
/s i 14 the timing of release and the characteristics of release. !
l 15 Essentially, they concluded that a release could occur as 16 soon as a half hour after initiation of the accident. .
1 17 They also considered the most likely 18 characteristics of the release would be gaseous material l 19 and volatile solids. The audience for this report was 20 emergency planners that wanted this type of information sn 21 they can decide on what they needed to do for their 22 emergency planning.
23 Next slide goes onto the next phase of the 24 evaluation. Actually, I am skipping over, not providing n
( ,) 25 information on I believe two aspects of the evaluation NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.. .. .. - - . . ~ . - .- . - - . . . - . . . - . - -- - - -
187 l
1 that was before in order.to conserve tima. Those-anpacto- j i
2' that I am skipping are the review of'the'NRC's decisions l O 3 on:the:EPZ size, and I have some backup _ slides on that.if.
4 we get into that, and the evaluation. implication from l 5 updated risk _and source term studies. l l
6 Now I was going to move right on to looking at l 7 the advanced reactor submittals. In'short, I did not plan j 8 on speaking too much to these submittals because NEI was 9 here to I'm sure describe them in more detail than I l I
10 could. I i
11 Essentially there were three main submittals. .
I 12 One-was in May of 1993, which was the URD, EPRI URD. .The
, 13 December 1993,.EPRI submitted'a report on technical 14 aspects of the advanced light water reactor. Both the URD ]
15 and this report they provided what you.would look~at as ,
L i 16 technical criteria and methods. That's the way the.URD l i
17 was structured for emergency planning. i 18' Those aspects, the technical aspects, did not 19 actually -- said these are the technical activities which
- 20. would allow you to take advantage of simplification for L
L l- 21 EP. But they did not go into what is the detail, what- _ '
22 kind of simplifications would be allowed. These were the 23_ technical criteria and methods, that if you met then you
- 24 would be eligible for the simplifications.
() 25 Essentially at this point, the EPRI report NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l _
188 1 says that NUMARC, or I believe may be NEI at that tim 3,
,3 2 would take the ball from there and come up with what
( l 3 simplifications would be possible and deal with any of the 4 policy issues involved there.
5 In January of 1995, NEI came in and briefed us 6 on their concept of EP for advanced plants. Once again, I 7 think that I will defer to their presentations for that, l 8 as it may or may not be the same as it was back then.
i 9 Okay. Now we looked at some of the PRA 10 results that were provided in the safety analysis reports 11 submitted for the design certification reviews. You have 12 got to remember, early on, we switched over from looking 13 at the advanced advanced reactors, the MHGTR, the PIUS and l
- g
(_ ') 14 so forth, to looking at the evolutionary and passive.
15 That's what this study, we're dealing with today focuses 16 on.
17 We looked at the core damage frequency from 18 the internal events. We looked at the core damage 19 frequencies. We looked at the early containment failure 20 and bypass frequencies which would really drive the 21 accident perspectives on what these would mean as far as 22 EP. Obviously they are very low on one to two orders of 23 magnitude below what the current 1150 study showed.
24 We also had a contract out to Brookhaven
, p
(_,) 25 National Laboratory to do some max calculations looking at NEAL R. GROSS i COURT REPORTERS AND TRANSCRIBERS
! 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
189 1 the source term data that was provided in the safety 7s 2 analysis reports, the evolutionary and passive plants, and 3 running through the MACCS code and coming up with 4 something that's similar to the 0396 dose versus distance 5 curve.
6 This is the probability of exceeding. I 7 should have put it up here. This is the probability of 8 exceeding a 200 rem dose with distance from the site. It 9 included NUREG 0396 curve, for a comparison with the 10 system, CE System 80 plus calculation that was done by 11 Brookhaven.
12 Again, you go essentially towards the low 13 probability, but notice also that the consequences come
'w / 14 and start decreasing at about the same level as what 15 occurs for the NUREG 0396.
16 MEMBER POWERS: You wouldn't expect anything 17 different. I mean the drop off has to do with things like 18 weather and dilution, which is independent of the plant 19 design.
20 MR. O'BRIEN: And source term. And perhaps 21 with the release. It's mainly source term. Yes, you're 22 right. If you have a similar source term, you are going 23 to get a similar shape. The only thing you do is offset 24 it by the probability of the release to begin with.
(
(Q,j 25 You see that for the ABWR. You see it also NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
190 1 for the AP600.
7S 2 Okay we looked at what EP requirements may be t i
\ ,/
3 candidates for simplification. Of course the major one 4 would be the size of the EPZ. A reduction in size of the 5 EPZ would be a major reduction in EP requirements.
6 Another one, or the next most beneficial or
\
l 7 major simplification would be in the off-site notification ;
l 8 requirements, if they were reduced. The timing or the 9 scope of where they would need to be, promptly or within ,
i 10 15 minutes as they now are needed to form. l l
I 11 Thirdly would be the exercise requirements. l l
12 If you reduce the exercise frequency, which is a big cost 13 to the industry.
k- 14 Lastly, we looked at perhaps staffing l
15 requirements. You need to get to people onto the site as 16 quickly as you do right now. We have in EP requirements 17 requirements for on-chift staffing levels and 30 minute 18 responders and 60 minute responders.
19 After all this work, we looked at -- we came 20 up with conclusions. The conclusions are really premised 21 on -- take the first one, which is emergency planning 22 should be maintained as an essential element in the NRC's 23 defense-in-depth philosophy.
24 MEMBER POWERS: How did you arrive at that 7
) 25 conclusion?
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433 l
. ~ ~ - . - . -. - - - - - . , . - - . . . . -. .- ._, .- - . - - -
191 1 MR. O'BRIEN: I think it's more of a policy 2 issue than anything.
3 MEMBER POWERS: It seems to me you can arrive 4 at that conclusion or a policy issue by saying emergency l 5 planning is part of the structure of defense-in-depth and 6 I'm going to have it come hell or high water. Because 7 considering you can do all the calculations you want, I 8 have it there in case you are wrong.
9 MR. O'BRIEN: That's correct. !
10 MEMBER POWERS: And I have a tremendous 11 backing for that kind of a position because if I look at r t
12 things like the NSAG guidelines to safety, they say i
\
?
13 emergency planning, it's part of defense-in-depth. It's a l
l L L14 structural part. It doesn't come from any calculations. l l l t
15 It's part of the structure.
l !
! 16 Or you can come about it, come to that as a I 17 matter of policy from some sort of a calculation. What if 18 you are wrong about your calculations and how wrong can 19 you be. Do I gain anything by having emergency planning
[
l l 20 in the event that you are wrong.
l 21 That would seem to be the.two ways to come i.
22 about it, and that are obvious to me you could come to 23 that conclusion that you need this as part of the defense-j 24 in-depth.
) 25 MR. O'PRIEN: Yes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 2000 # 3701 (202) 234-4433 i
i
192 1 MEMBER POWERS: I guess what I am asking is' 2 which one'did you -- how.did you come by it? Is it just 3 part of the structure?
4 MR. O'BRIEN: It's part of the structure.
5 MEMBER POWERS: So.I mean it does not matter 6 what kind of a reactor I have. I mean use your 7 imagination, any kind of reactor, I am still going-to have 8 emergency planning.
9 MR. O'BRIEN: Yes.
10 MEMBER MILLER: Independent of the curve on 11 page 18, which says the probability of exceeding 200' rems 12 at one mile to a System 80 is what, 6 X 10 to the minus 8.
13 MR. O'BRIEN: That's correct. We didn't look 14 at --
)
15 MEMBER MILLER: Design a plant where it's 6 X [
16 10 to the minus 12, we're going to have the same 17 philosophy.
18 MEMBER POWERS: It wouldn't matter because 19 you're -- that's correct, because you are saying 20 structurally --
21 MEMBER MILLER: I'm not sure I believe 6 X 10
- 22 to the minus 8. l l .23 MEMBER POWERS
- Structurally you are saying 1-i 24 what if they are wrong.
25 MEMBER MILLER: How much are we going to pay
[
i NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344 433
193 1- to be wrong.
j~ . 2 MEMBER POWERS: That's another part of the
() 3 question he's going to answer later on. Right now it's 4 whether you have it or not as part of defense-in-depth.
5 MEMBER MILLER: So if I look at all these j 6 numbers and take those numbers,'and then I look.at the 7 numbers on page 13, for probability of early fatalities, 8 it's somewhere on the order of three to four, if I -
9 understand'those graphs.
10 MR. O'BRIEN: Yes. j 11 MEMBER MILLER: So if I have 1,000 of these 12 plants operating in every 100,000 years, how much money i J
13 would I want to spend to save three to four lives. Does j" i
\~s)/ 14 that kind.of thinking ever take place here? q i
15 MR. O'BRIEN: No.
16 MEMBER MILLER: I'm not really against the l 17 policy. I am just trying to put numbers together here.
18 MR. O'BRIEN: Yes. But I think in part that 19 part of the issue is not only the numbers issue, but the 20 public perception and acceptance, which --
21 MEMBER MILLER: Well I understand. Public 22 perception is a major issue.
23 MEMBER POWERS: But if you're operating from a 24 position of structural defense-in-depth, it does not
,g T) 25 matter what you calculate. You can calculate a tenth of a l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
. . . -- - . - . -- - .. . ~ - - -. - -. . . - - - .- .- - ..
1 194 l 1 person every billion years, because I have created a 2 defense-in-defense structure in case you are wrong. So it 7-sy V 3 literally does not matter what you are calculating.
4- MEMBER MILLER: Well, and of course we don't 5- have any uncertainty of ours put on these graphs on page 6 18. I wouldn't believe anything you put on it anyhow.
7 MR. O'BRIEN: The second conclusion was the 8 rationale for EP requirements for advanced plants should 9 be'the same as for the current plants. That is, 10 consequences.
11' In reality, I didn't bring up this point, but .
1 12 it's tempered by probability considerations, which is kind j
. 13 of a vague term, but it has to do with the idea of the
(~%
k- 14 more severe and less severe core melt accidents and the ,
15 different protection levels that you have there and the i 16- decrease in the doses with distances.
]
17 So if we believe that rationale is still
-18 warranted for advanced reactors, the technical criteria 19 which would be developed would be a similar technical 20 criteria based on not exceeding PAG doses for the DBA type 21 accidents and also the less severe core meltants, and not-22 exceeding our -- I don't remember the exact words, but not i
- 23 being.able to take protective actions to prevent severe l
- 24 health effects at those areas that were more the I . .(g
- ,/ 25 consequences for more severe accidents.
i NEAL R. GROSS l l COURT REPORTERS AND TRANSCRIBERS )
1320 RHODE ISLAND AVE., N.W. - ;
(202) 234-4433 WASHINGTON, D C. 20005-3701' (202) 234-4433 1
- - . - . - . . - . . . . _ - .-~ ._-. - - -. - .- .- - --. . . - - . -
195 'l; 1 Those same technical criteria see a reason for i 2 changing that for advanced reactors. t O- 3 The methods is fairly straight forward. It's j f
}
4 the same type of methods. You have your source term and {
5 you do your dose calculations. We essentially already did 6 those for the advanced reactors to find out what the. .l 7 consequences would be with distance. We showed one of the i 8 examples, in.all the. advanced reactors have similar ~
i 9- consequences, waa would be expect it from the source terms 10 that are available for release. i 11 The fifth conclusion is, and you may hear more
! 12 on this, was the elimination from the industry. You might 13 hear some issues on this. Elimination of accident types I\ 14. from consideration based on risk analysis is not
- i. 15 recommended because of the uncertainty associated with 16 these evaluations. l t
17 Then-I quote essentially what was stated in 18 the SER for the ABWR which talks about the implications of l 19 low core damage frequencies. I could read it to you, but l I
-20 essentially i.t says you can not take them literally. l 21 MEMBER FONTANA: That presumes that we're l 22 talking about evolutionary or light water reactor. See up l 23 front here you say-you started with HGTR. We're not 24 talking about those any more.
25 MR. O'BRIEN: That's correct. We shift gears NEAL R. GROSS !
COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N.W. l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433 j
1 196 l i
1 here.
2 Okay. I am going to be next going onto some (7-
'~' )
3 of the ACRS questions. If anybody has any more questions 4 on the previous stuff, we'll wait. l l
5 Okay. These were challenging questions. I 6 think that you will find that the responses in some 7 respect do not answer the question directly. It gives l
8 information.
l l
9 VICE CHAIRMAN SEALE: Rather you impeach the 10 question, is what it amounts to. We understand. I mean 11 it's not a criticism.
12 MR. O'BRIEN: Okay. The first question to 1
13 ACRS. Actually, this was posed in the aspect of
<T
( 4 k/ 14 developing a firm risk-based criteria for EP. ACRS j l
15 thought that the criteria that was provided for EP, the 16 ones that I have discussed, not exceeding DPA doses from 17 the less severe and DBA accidents and the shoulds and the l
18 caveats, the qualitative type language, was that EP 19 deserved better essentially, is what a comment came back l
20 from ACRS. j l 21 In the process of establishing a firm risk l base for EP, the ACRS believed the staff should look at 22 23 these questions and develop answers to these questions.
24 The first question was, what level of risk is
() 25 being accepted for currently operating light water NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISl#dD AVE., N.W.
(202) 234 4 33 WASHINGTON. D C. 20005-3701 (202) 234 4433
_. _ . _ _ _ . . . ~ _ _ . _ . . _ . . _ _ _ _ _ . - . _ _ _ . _ _ _ _ . _ . _ _ . _ _ . _
! 197 l
l 1 reactors with their existing EPZs. The first bullet is 2 not an answer. It's a statement. The size of the EPz is 1
O 3 not based upon a defined risk level, but rather on the 4 potential consequences of an accidt.nt if one'were to ]
l
, 5 occur. :
! I l- . !
l 6 So in some respects I am saying that we did i 7 not' base it on a risk level, but then that doesn't answer ,
- . 8 your question'. So we looked at perhaps a way to in some. .j
! l i
9 respects answer the question by providing information from j i
10 NUREG 1150 on the issue. That's on the next slide. i l
1 11 NUREG 1150 evaluated an impact on the risk as l 12 indicated by etarly fatalities for reactor year. Looked at 13 the impact on that risk of various emergency response
!O 1:
14 options for the five plants that were studied.
15 MEMBER FONTANA: Is this for the whole i
16 spectrum of accidents weighted by probability? ;
17 MR. O'BRIEN: That's my understanding, yes.
18 MEMBER POWERS: If the safety goal were 19 applicable to individual plants, which I understand it 20 isn't, but just say it.was, where would it show up on-
- j. 21 these plots?
22 MR. O'BRIEN: That would be B-6, Tom. This is 23 also from NUREG 1150. I'm sorry, this is. individual early i
s 24 f atalit.y, but I think it may answer your question.
25 Whereas this one has the -- this one is a number of
- . NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS ,
1323 RHODE ISLAND AVE., N.W. j (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4 433
, . . - ~ . . . - . - , . . - -, _ .- - -. _ _ _ _ . - - , _ l
-- - _.-..- -..~.- ,. .-._.- .-.- -- .... .
198 1 fatalities. The other slide that we were looking at, 2 number 20, page 23. But the safety goal or objective was -
3 on the individual risk level.
4 I'll describe the slide a little bit. Early; ;
5 fatalities per reactor year for the five sites. This base 6 case corresponds to 99.5 percent evacuation from zero to i
7 10 miles. The evaluations. performed for these plants 8 takes into account site specific meteorology, site 9 specific population density, site specific evacuation 10 speed. You see a variety of what would you call the risk 11 level at these different plants if this was the-12 characterization for risk that you used. You see a-13 variety of benefits of the EPZ. I I
14 Number one here is 100 percent evacuation j 15 where the base case was 99.5. Number two would be zero :
I 16 evacuation, but you do have the relocation from high 17 contaminated areas.
18 You see the benefit going from two to the base 19 or the rise. So this gives you an-indication of the 20 benefit for emergency planning.
21 So the question arked what is the acceptable 1
22 level of risk with the current EPZ size. This is the i 23 answer that we -- or this is at least part of the answer.
1 1
24- The next question the ACRS asked us, is this !
l 25 level of accepted risk appropriate? If not, what should i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N.W.
(202) 234J433 WASHINGTON, D.C. 20005-3701 (20M) 234-4433 l -
. _- , . _ - . . - - -- _ _ l
199 l' it be? Once again, that's a very difficult question. It i
/. 2 goes to the first one. Since we didn't answer the first i !
s'
~
3 one exactly, we certainly couldn't answer this one exactly l 4 either. l 5 Just stating the safety goal and policy l i
6 statement, that the Commission established goals that 7 broadly. defined an acceptable level of risk of electricity I
8 generation via nuclear power. It does not define l l
9 acceptable risk level for an~ individual plant. )
l 10 As you guys are more than aware, the policies l 1
11 regarding the use of risk information are under l
12 consideration.
13 Question number three. i
/~h
- 14 MEMBER FONTANA: Can you go back to the slide i
15 with the vertical lines on it?
I 16 MR. O'BRIEN: Okay. j 17 MEMBER FONTANA: Let's take one of those 18 plants. For example, what you've got -- let's say the 19 probability of a large release is 10 to the minus six.
20 Does that mean that if you get that release, you are going 121 to get about 1,000 fatalities? l l
22 MR. O'BRIEN: Yes. What you do is you have l l
23 the frequency of the release. It's really the frequency l l
24 of the number of fatalities. But you are right. It would j
( ) 25 be 1,000 times.
i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
_ _._ _ - _ _ _ _ _ _ ._m. - , _ . . _ . - _ . - . _ . . . . . _ _ _ . - _ _.._ .. . .-~._._ . . _ . . _ _
t 200 l
l 1 MEMBER FONTANA: All right. In other words, I I 2 just did the calculation backwards than what was done to l
O v <
3 get that, basically. -l l
4 MR. O'BRIEN: Yes.
5 MEMBER FONTANA: Okay. ]
l 6 MR. O'BRIEN: The actual details in there, I i 7 don't know the actual details or the exact details, but it j 8 appears that there would be a lot of dependency on the 1
9 exact type of release characteristics, timing of release, 10 and so forth. It's just not the overall number, but 11 rather the specific individual sequences types of 12 releases.
13 MEMBER FONTANA: See the thing that's a little 14 not strange, but look at Peach Bottom and look at Grand 15 Gulf. I think Peach Bottom has a higher population around i 16 it than Grand Gulf does.
'17 MR. O'BRIEN: That's correct.
l 18 MEMBER FONTANA: Yet the numbers are about the t
19 same.
20- MEMBER SHACK: But you're evacuating 21 everybody.
l
! 22 MEMBER CATTON: Look at the responsiveness of l.
! 23 some actions at some plants relative to others. Sequoyah 24 looks like nothing much matters. Where you look at Peach 25' Bottom, and one is -- going _from the base case to one is a i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
201 1 huge change. Zion, nothing much seems to matter.
<~s 2 MEMBER FONTANA: No. Base to one is almost l
I 3 the same.
4 MEMBER CATTON: On Peach Bottom?
5 MEMBER FONTANA: 99.6 evacuations as compared 6 to 100 percent evacuation. Those numbers ought to be 7 pretty close to each other.
8 VICE CHAIRMAN SEALE: Nobody's there. ;
9 MEMBER FONTANA: Okay, well thanks. I just 10 wondered.
11 MEMBER CATTON: Okay. I guess it depends 12 where the people are. If you evacuate where there are no l
13 people, it doesn't matter.
77
\ )
'O 14 MR. O'BRIEN: Question number three, another 15 difficult question, but it relates to the other two, so
\
16 once again, the answer is probably less than satisfying.
17 For the advanced plant designs, what would be 18 the size of the EPZ based on a level of risk comparable to 19 the accepted value? We didn't define accepted value 20 before, so we can't really answer that question. What are 21 the implications of this result?
22 So rather than answering that question i 23 directly, once again, we talked about emergency planning 24 requirements being based on the risk of an accident. Not 7m,
() 25 on the risk of the accident, but on the potential l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
202 1 consequences.
2 Then lastly, provided a bullet on how the t
( 3 Commission addressed this issue in a statement of 4 consideration for the part 100 siting criteria. I'll read 5 it out to you.
6 "In summary, the next generation of reactors 7 are expected to have risk characteristics sufficiently low ;
8 that the safety of the public is reasonably assured by the i
9 reactor and plant design and operations itself,.resulting 10 in a very low likelihood of occurrence of a severe 11 accident...Nevertheless, the Commission concludes that l 12 defense-in-depth considerations and the additional 13 enhancement in safety to be gained by siting reactors away 14 from densely populated centers should be maintained."
15 So it's really a re-affirmation of the :
1 16 defense-in-depth policy, the proceedings that were .;
17 recently -- I don't know if they have been issued, but ,
- 18 they are very close to going out in the Federal Recister.
19 That ends my presentation. I can take any 20 questions or --
21 VICE CHAIRMAN SEALE: Any additional 22 questions? I think we have kind of grilled you as you L 23 went ale >ng. Any comments from the audience at this point?
24 MEMBER POWERS: I guess you still have the 25 question, the third question. Have you tried the little !
i L NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 2000 #3701 (202) 234-4433
203 l i
1 exercisn'of saying -- maybe it's a hand calculation th t I l 2 can do myself, but saying I can look at an advanced plant 3 on the Zion site, for instance, and look at its 4 probability to have an accident, and come away with the .
l 5 conclusion that changing the EPZ one way or another is not l i
i 6 going to make any difference whatsoever. l 7 MR. O'BRIEN: We did do a little exercise on I
i i 8 that. It wasn't as satisfying as doing it, because we 9 actually did it on a hypothetical site with 100 people per ;
10 square mile rather than throwing it on a site which l l
11 probably would have been more representative in a one to I
12 one check against one of these other sites. l 13 But we put it on a site with 100 people per
()
' '- 14 square mile. We saw something similar to what the other 15 pictures were. We saw a decrease in the number of '
i 16 fatalities.
17 Now it depends of course on the probability of 18 release. When you get down to a certain level, then 19 there's no expected fatalities to begin with because they 20 are so low. We concluded that if you change the number of 21 people, there's a lot of variabilities that could have 22 been done. We did not complete the work. We were ;
23 actually hoping we would have a chance to do it, put these 24 plants on different sites. We didn't look at that, but we i g s.
\,f 25 thought that that was an interesting result to get, but we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
204 1 didn't think it would change our conclusions on the study
,s 2 so we didn't complete it.
/ )
3 MEMBER POWERS: It probably wouldn't have.
4 VICE CHAIRMAN SEALE: Any other? Well, thank 5 you very much. I think we'll ask the industry now to make 6 their comments. Mr. Ron Simard is going to speak for NEI.
7 MR. SIMARD: Good afternoon. I am Ron Simard 8 from NEI. What I would like to do today is give you a 9 sense of the industry approach to defining emergency 10 response appropriate to advanced 2eactor designs, and to 11 give you a brief status report on where we stand with 12 respect to a petition for rule making that would propose 13 probably minor changes to the regulations and the io)
N/ 14 underlying regulatory guidance that set these emergency 15 planning requirements.
i 16 I guess what I can do is refer to the slide 17 number 20 that Mr. O'Brien used. Remember slide number 18 20, which gave the conclusions regarding the approach?
19 One of the things that I would like to reinforce today is 20 that in looking at the requirements for advanced reactors, 21 we agree with the staff that we ought to follow the 22 current approach because it has served us well for almost f 23 20 years for now.
24 So we think that despite the fact that the p
i ,) 25 advanced reactor designs we are talking about have j 1
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 1
_ . _ _ . _ _ . _ _ _ . _ . -. . . _ - . _ - ~ - _ . _ . _ _ _ _ . . _ . _ _ _ _ . . . _ . . _ ._ ..
i 205 l l'
i i superior core damage prevention and they incorporate l- .
1
- 2 severe accident design features to mitigate releases, that i 3 it's still prudent to have planning to provide a' defense !
> )
4 in depth in the event that there is a release. 1 I
l 5 So we agree with the basis that Mr. O'Brien.
1 J
6 outlined. That is the approach that we are taking as we l
l l 7 look at applying-the same sort of approach that.you have L I l' 8 just heard about to advanced reactors. j l !
9 In other words, the' basic principles we'are 10 looking at is that this' defense-in-depth means that it's 11 prudent to have an area where you might need to take 12 prompt actions. A good guideline to use as a sort of a 1
13 trigger for you to project when you are going to need'to l 0 14 take action is this protective action guideline.
15 In other words, given a hypothetical release, a.
16- what is.the projected time at which you might exceed the-1 17 protective action guideline at that site boundary. l 18 So following that same approach, what you just 19 heard is that in NUREG 0396, which-was issued in 1978, and 20 which used the reactor safety study, the.Rasmuson report 21 from 1975, which relied on work in severe accidents that l 22 was done in 1972=or 73.
f .23 Following that approach that we just talked 24 about, on the basis that you wanted to minimize the L
() 25 probability of exceeding these protective action NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS !
1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
, - _~ _ , _ ,
I 206 ,
1 guidelines, 10 miles seemed to be a prudent value to pick I l
-c 2 for current light water reactors. But there was l
. 3 recognition that for the. smaller reactors like Big Rock or- l l
4 La Cross or for gas cooled reactors like Fort St. Vrain, ,
5 you would expect a smaller value. In fact, that was set .;
6 at five miles.
i 7 But knowing what we know today,-and applying i
8 that same approach to the advanced reactor designs, common 9 sense might lead you_to expect that you would come up with
. :UD a different number, that it would be less than 10 miles. i 11 Because the size of the potential release, based upon the 12 source term, is going-to be smaller than what was assumed j 13 in 1978. The probability of release will be less, and the !
\I 14 delay time before this release will also be different for i r
15 these advanced designs. I think this committee has ;
16 discussed the design features for advanced reactor 17 sufficiently to have some appreciation for why intuitively 18 you would expect that.
l l
19 MEMBER POWERS: But maybe my intuition is 20 different. Maybe my. intuition is T. This is an advanced 21 reactor, it's a very different design from anything else.
22- I have got no operational experience with it. Lots of l 23 things about plants look good on paper and don't look so l
i 24 good in there. Calculational experience is really zero.
25 I can be just as wrong about this reactor as I am about NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
207 1 the light water reactor. So maybe I don't -- my intuition l
r~N 2 then comes to saying no, ought to be about the same.
l(
3 MR. SIMARD: Well, obviously intuition isn't I l
l 4 enough to proceed. What I would like to conclude with j l 5 today is give you some sense of the approach we're taking f l- :
6 to provide a sound technical basis, at least the ,
l 7 equivalent, if not more complete than the basis that was l 8 used for the. current reactors today.
9 So if you are comfortable to some degree with !
10 the approach that was used for current emergency planning,. :
1 11 that's the-same approach that we are following. j 12 Now the inputs, the methodology will be the !
L l
! 13 same. The inputs will be different because the' design l L
l ) 14 features are different, the knowle6J e of severe accidents !
l . i l 15 is different. But we are still following the same- !
! l 16 approach. We are applying the concepts of risk. We are .I I
17 considering probabilities of various accidents in the same i
18' way that they were considered originally, and sticking 19 with the same basis because we just don't have a better 20 basis. I mean this one has served us for 20 years.
21 MEMBER POWERS: What happens is that if I come 22 in and I say I want defense-in-depth because I believe in t
23 the fallibility of man in his calculations, it seems to me 24 it doesn't matter what you calculate. It doesn't matter 25 what you adjust your risk to be. I have already said I am i
i NEAL R. GROSS
[ COURT REPORTERS AND TRANSCRIBERS t 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 2000>3701 (202) 234-4433
~.-
l 208 l 1 discounting that. That's why I am going to this defense-
,.q 2 in-depth. I may use it for other things, but when it OE 3 comes to_my final barrier of protection, I am not doing it 4 because of what you calculate. I am doing it in case you 5 are wrong. And there's no input.
6 MEMBER FONTANA: Dana, excuse me. I think 7 they did. I think the 10 miles comes from eyeballing that 8 0396 curve. You see that curve comes down. If you draw I
9 it at 10 miles, it isn't all the way to the right and it 10 isn't all the way to the left. In fact, I think Bob 11 Bernero did it.
12 So if one were to redo the very same-13 calculation but with different numbers coming frcm a more j (O_/ 14 modern plant, you get the same kind of curve. But chances 15 are, if you draw the line at the same spot, you are going 16 to have something less than 10 miles.
17 MEMBER POWERS: No. 5 don't think so. I !
1 18 think the turndown has strictly to do with distance.
19 CHAIRMAN KRESS: Yes. It has to do with the ;
i 20 plume characteristics.
21 MEMBER POWERS: Yes. I mean it's everything l 22 after the release has taken place. It doesn't matter what j l
23 shape the bottle was that let loose things. It only I I
24 matters how far out you went.
() 25 MEMBER FONTANA: You are looking at a i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. i
! (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
5-209 1 different curve, i- ,
2 MEMBER CATTON: So for a given plant, things O 3 are linearly proportional to the release.
b 4 MEMBER POWERS: .No. If you are looking at i
, 5 shape --
6 MEMBER FONTANA: Look at page 11'of the prior 7 presentation. That'has a dose and a source term inherent l
8 in it.
9 MEMBER POWERS: But the fact that it turns 10 over at a certain distance, with geometry.
11 MEMBER FONTANA: The shape here would be the -)
i I
12 same, but the dose would be different.
13 MEMBER POWERS: It's different, but if I say
/
3 14 look, I don't believe your vertical axis --
15 MEMBER FONTANA: Well if you don't believe me.
16 MEMBER POWERS: I am asking what if you are 17 wrong about that vertical axis and I don't doubt you are 18 right about the horizontal axis. Then I'put it out at the 19 same place no matter what bottle you put in the center.
20 MR. SIMARD: Well ultimately I think you have 21 put your finger on what ultimately drives your decision, l
22 you know,-your level of comfort. Ultimately, once you-23' have seen the technical basis, you have to come to a 24 decision about what constitutes an acceptable level of l
( 25 risk.
i NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W. l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
1 210 3
'l Right from the beginning, you can't avoid tha
- 2 R word. I mean if you agree that what we are trying to do 3 is avoid reaching these protective action guides, then 4 right away you are buying into the decision.that well, one ,
5 rem, whole body, TEDE is an acceptable level of risk. f 6 So'you are quite right. Inherent through this j l
7 are'various assumptions about risk. When you are all. .;
8 done, and when we are done with what we are doing now, we !
1 9 will have the same sort of thing. We will have a set of !
10 curves.
l 11 Now if our intuition is correct,- and.'if we are l l
12 to proceed, because we won't proceed unless there is.a t
13 sound technical basis. If our intuition is correct, you i .
( , i 14 will see that curve lower down. As a matter of fact, f
-15 there might even be some indications that.this curve I
\
16 wouldn't even appear on this scale. That to be able to
'17 pick it-up, you would have to change the. scale.
18 But the point is, this is the approach that 19 exists today. It's the approach we are following. Unless 20 there is a significant difference, unless this curve 21 shifts, then your point about ultimately this evolving j 22 into a judgement decision means that we won't proceed. .i; 23 Unless there is a really clear technical advantage to :
)
24- considering a policy shift, it would be foolish to propose 1 25 a policy shift.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W (202) 234-4433 WASHtNGTON, D.C. 20005-3701 (202) 234-4433
1 211 -!
J 1 Attached to your handouts, I put a copy of a 2 letter that I thought might be useful because in that
{ .
3 letter and the attachments to that letter, there is a 4- pretty good statement by the staff of why these three 5 factors are important, as well as some.other explanations j 6 of what might actually change, you know, in terms of the 7 requirements, whether it's the sir.e of the zone, whether 8 it's the nature and the timing of the actions. So I l 9 attached that to here_because I thought it might'be
~
10 useful. j i
11 -Getting back now to the point we were making -
'12 about why we are even on the tack'that we are on here, is 13 that emergency planning was addressed right in at the~
14' beginning, right from the beginning of the design-15 requirements.for_these. advanced reactors.
16 Back in'the EPRI utility requirements 17 document,.was the requirement for the designer that what 18 the desioner produced had to satisfy this criterion. That 19 you wouldn't exceed that one rem TEDE for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> at a 20 distance as close in as half a mile.
-21 So what I am trying to do with this is give 22 you a sense of what-made.us even begin the effort that we 23 are on now, and why we think that it's productive.
24 'Now if in fact we are right, and if in fact 25 that curve does shift, and if in fact you follow the same NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
212 l 1 reasoning that we have applied in the past, then what you !
l
, 2 find is that the area in which you need to take a rapid ;
I i l t i 3 response, in other words, the area in which you need to be 4 able to take quick action because there's a chance that 5 you might reach those protective action guides, that area 6 would be smaller. It would be less than 10 miles.
7 Now what I am showing here is that the 8 approach that we are on right now, because if in fact the 9 technical basis supports reducing the size, the area in 15 which you need to take a rapid action, it may not be 11 prudent simply to say well, we'll just scale back.
12 Everything we have today that goes out to 10 miles, all I 13 the requirements, we'll scale them back down to this. You
\m >' 14 might still want another area in which you continue to pay 15 attention and work with the off-site authorities. So 16 we're calling that an awareness area.
17 What I have tried to do in the next two ,
18 viewgraphs is characterize for you what is different about !
l 19 these two areas. The response area, again, it's like the l 20 emergency planning zone today. It's determined by dose at 21 the boundary and the probability of exceeding the 22 protective action guide. It's the area close in, where 23 you want to have provisions for prompt notification and 24 immediate action following an accident.
n
( ,) 25 What we see is that within that area, you NEAL. R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4 433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
213 -l 1- would continue, the licensee would continue to have a 2 complete set of planning-action, similar to_the emergency =
0 3- response plans that licensees have today-. l l
I 4 But-rather than simply stopping there, we i
5 think it.would probably -- not probably, we definitely -l
-l 6' think it would be prudent though even beyond that, to !
7 define a set of_ actions that are appropriate in an area !
8 where_the radiological effects, if any, would be a lot 9 smaller'and would take place over a longer time frame, say l 10 greater than one day.-
l 1
11 Again, the size and. shape here would be based 1 1
]
12 on factors other than dose. It was dose that was guiding ;
13 us toward some judgement as to what would be a response j O 14 area. But here you might want-_to take'other factors into i
L 15 account. Emergency planners like to use the phrase egress I t
16 limitations or the. meteorology, or just you know, the 17 local jurisdictions. In dealing with the local 18 jurisdictions at the time of licensing, that size and 19 shape would be site specific and would be~ integrated with 20 the capabilities that the local. jurisdictions already have
-- 21 to deal with non-radiological emergencies. I l l
- j. 22 So the emergency planning that takes place in
{
23 this area integrates with the plans already in place to l
j 24 respond to fires, floods, and hurricanes, except that now, 25 you have got the licensee supporting them by providing the I
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l.
! (202) 234 4 33 WASHINGTON D.C. 20005-3701 (202) 234-4433
_ _ . . . _ ._.. _ __ - . . _ _ . . . _ _ , _ - . _.l
214 1 dose assessment, the field' monitoring that.would be unique 2 to a radiological event.
O. 3 Now that is the outline of what we are' heading 4 toward. At this point, if there are no more questions on 5 how this might look,' in other words, if this is to lead to j 6 a petition for rule making, what I would like to do now is )
7 explain what we'd be asking for, what would change to i l
8 reflect this kind of -- j 9 VICE CHAIRMAN SEALE: Yes.
10- MR. SIMARD: Briefly, there are two things 11 that would change. There are regulations, and then there 12 are the underlying guidance documents. ,
l 1
13 In the regulations, the reason I said the ;
14 changes would probably be minor is there are a few I
15 regulations like part 50.47 or appendix E to part 50, that 16 mention 10 miles. They are fairly explicit.
17 A second thing that might change is again, in 18 Appendix E to part 50, there's a reference to being able 19 to provide notification to all the public within that 10 20 mile area, within the plume exposure, EPZ, within 15 21 minutes.
22 Let me give you a third example. In several 23 of the regulations, there are references to the types of f 24 plans that a licencee needs to submit. So he needs to-i 25 submit not only his own on-site plans, but also various NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
215 l l
1 plans that the off-site agencies have. There might ba ,
l ,y 2 some changes there if in fact the technical basis supports I i 1 3 this approach.
I l 4 So the changes to the regulations would 5 probably be fairly minor, and would probably cover three I
6 or four -- I'm looking at Mr. Essig. Three or four I l
7 regulations maybe that address emergency planning, 8 contents of applications, conduct of exercises, for 9 example.
10 Then as far as the regulatory guidance 11 documents, I can think of two that might change. There 12 are evaluation criteria that are very detailed, for 13 example, in NUREG 0654. Very detailed checklists about
'/ 14 the nature and the timing of the actions that the licensee 15 has to take. You might expect some of those to change.
16 Similarly, there's another NUREG that deals j 17 with the licensee's manning of his emergency support 18 facilities.
l 19 So that is where we are heading now. But 20 here's where we are today. EPRI has the lead in doing the 21 work to establish the technical basis in recreating the 22 curves that Mr. O'Brien showed and following that same i
23 approach. They are fairly close to done.
24 As part of that work, what they are doing is
, (
(,) 25 they are taking each of the three designs that are now l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHOOE ISLAND AVE., N W.
(202) 234-4433 WASHING'lON, D.C. 2Go,6; (202) 234-4433
l l 216 )
L l 1 before the NRC for certification. They are doing spscific l
2 verification that in fact those designs do fall within a l7s
\~) 3 sort of -- a common response. In other words, that you 4 can define a response area size that would envelope all l 5 three designs, so not only are they verifying that utility 6 requirements on emergency planning, but they are also i
! 7 doing design specific verification, and they are 8 recreating that NUREG 0396 analysis.
9 The results should be done fairly soon. We 10 have targeted the second quarter of next year, assuming i 11 again that that technical basis comee out and we feel we L 12 are arguing from a really sound case. We would then 'f l
l
> . -s 13 proceed to identify the regulations and the regulatory 14 guides that would then be changed. We would submit a l 's /
15 petition for rulemaking.
l i \
16 At that time, I would assume we would appear i 1
17 before this committee again. l I
l 18 VICE CHAIRMAN SEALE: Are there any questions?
I 19 MEMBER FONTANA: I have a statement. It might l 20 not even be relevant. But I have always been kind of 21 concerned about approving sites without any consideration 22 of the kind of reactor you put on it. It seems illogical.
23 It seems.that if you have an advanced reactor j
- 24 which you know, has a much lower probability and much
() 25 lower source terms, you ought to be able to go to a i
i NEAL R. GROSS
! COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234-4433
_ ._. _ _ - _ _. .~. . _ _ _ _ _ _ _ _ . _ _ _ . . _
i 217 i 1 smaller site. So why approve sites ahead of time. Well, 7g 2 that's a d.!fferent argument I guess. Forget it. It just
' O 3 doesn't make sense.
. 4 It looks to me in looking back to this curve I
i 5 on figure 11 that he had, that like maybe a factor of five l 6 or 10 reduction in the source term can make a significant 7 difference, even if you accept th'e same consequences. It 8 can make a significant difference on what your EPZ is.
9 MR. SIMARD: Yes, sir. It does. You know,
~
10 it's premature to get into the technical results that EPRI 11 is coming up with. But I think it's fair to say that 12 calculations have been done. The source term moves that 13 curve up and down, as I said, even potentially off the t~
14 scale of that chart.
15 VICE CHAIRMAN SEALE: Unfortunately, there's 16 another reality that is as potent,_perhaps, as the one 17 that Dr. Powers made regarding the idea of defense-in-18 depth being the motivation for emergency planning zone. j 19 That is, that by hook and crook we now have in place a I
20 methodology for funding emergency response capabilities in 21 regions around nuclear power plants. l 22 That capabilit- has turned out to be a very 23 valuable asset to communities that have had-hurricane or 24 tornado or other kinds of non-nuclear disasters of one
() 25 sort and another. So I have a real question as to whether NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
218 1 or not your most -- it's not a question, it's an
-m 2 observation. That you most formidable protests will come
)
3 from people who are interested in keeping their emergency 4 preparedness capability and don't really care whether the 5 nuclear power plant is there or not.
6 MEMBER BARTON: Then it's not a technical 7 issue.
8 VICE CHAIRMAN SEALE: It's not. That's right.
9 I mean it's gratuitous as far as this committee is 10 concerned. But just feeling the breeze, as it were, you 11 get the feeling that those are the guys that are really 12 going to give you the hard time.
p.
13 MR. SIMARD: Well, yes, sir. It might very i \
\/ 14 well be that another advantage to this two zone concept is 15 that in this area here, the licensee might choose for 16 goodwill with the community. You know, he might choose to i l
17 continue to subsidize ambulance, hospital, whatever.
18 But the point is, licensee resources, NRC 19 resources, FEMA resources, spent on inspection, !
I i
20 enforcement, the biannual exercises and all that, you !
21 know, need not be as resource intensive as they are today 22 in that area. But I think you raise an excellent point.
23 VICE CHAIRMAN SEALE: Yes. And it may well be 24 that putting things in that context may put enough of a
('
(_)) 25 different spin on it that you might get some more l
1 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 219 ;
! i i favorable respon=e. [
j q 2 But I don't think there is anything that we ;
l 'V 3 need to plan to do at this time. We thank everyone for !
I i j
l 4 giving us.a heads up on this. As you say, if you come i
I 5 around with a rule making proposal, we'll probably see you i 6 again, f 7 MR. SIMARD: Yes, sir. Thank you.
8 VICE CHAIRMAN SEALE: Mr. Chairman, I'll turn ,
9 it back to you.
10 CHAIRMAN KRESS: Okay. At this time I think 11 we can dispense with the transcription and work on our i
12 reports.
13 (Whereupon, at 4:32 p.m., the proceedings were V 14 concluded.) ,
15 16 17 18 19 20 21 22 l
l l 23 24 !
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 2000 # 3701 (202) 234 4433
i r
3 CERTIFICATE This is to certify that the attached proceedings before the United States Nuclear l Regulatory Commission in the matter of:
Name of Proceeding: 436TH ACRS' Docket Number: N/A Place of Proceeding: ROCKVILLE, MARYLAND were held as herein appears, and that this is the original -
transcript thereof for the file of the United States Nuclear l
Regulatory Commission taken by me and, thereafter reduced to l
- (A) typewriting by me or under the direction of the court reporting company, and that the transcript is a true and l
l l
accurate record of the foregoing proceedings.
1 l
[ 11 . ,
. 14EL k ORBEff NINER Official Reporter Neal R. Gross and Co., Inc.
I l
l o \
i '
U I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVENUE. NW l (202) 234-4433 WASHINGTON. D.C. 20005 (202) 234-4433
{
-. .. - .- .- -. . - - - - . . . . - -, ~ - - .. . ~ - . - . .
1 INTRODUCTORY STATEMENT BY THE ACRS CHAIRMAN 436TH ACRS MEETING, NOVEMBER 7-9, 1996 kJ THE MEETING WILL NOW COME TO ORDER. THIS IS THE FIRST DAY OF THE 436TH MEETING OF THE ADVISORY COMMITTEE ON REACTOR i SAFEGUARDS. DURING TODAY'S MEETING, THE COMMITTEE WILL CONSIDER )
THE FOLI4 WING:
(1) PROPOSED RULE ON STEAM GENERATOR INTEGRITY (2) RISK-BASED ANALYSIS OF REACTOR OPERATING EXPERIENCE (3) REVISED SOURCE TERM FOR OPERATING REACTORS (4) EMERGENCY PLANNING FOR ADVANCED REACTORS 1 (5) PROPOSED ACRS REPORTS THIS MEETING IS BEING CONDUCTED IN ACCORDANCE WITH THE PROVISIONS OF THE FEDERAL ADVISORY COMMITTEE ACT.
DR. JOHN T. LARKINS IS THE DESIGNATED FEDERAL OFFICIAL FOR THE INITIAL PORTION OF THE MEETING.
WE HAVE RECEIVED NO WRITT.EN STATEMENTS OR REQUESTS FOR TIME TO MAKE ORAL STATEMENTS FROM MEMBERS OF THE PUBLIC REGARDING TODAY'S SESSIONS. A TRANSCRIPT OF PORTIONS OF THE MEETING IS BEING KEPT, AND IT IS REQUESTED THAT THE SPEAKERS USE ONE OF THE MICROPHONES, IDENTIFY THEMSELVES AND SPEAK WITH SUFFICIENT CLARITY AND VOLUME SO THAT THEY CAN BE READILY HEARD.
p-? l I WILL BEGIN WITH SOME ITEMS OF CURRENT INTEREST.
1 HE HAVG SOME C H A N G&s ~ro TWG R c ,e n 3> g ,
TkGn6 HtuL NOT BG A n y pggse nrp nc as OH .T TE M 4,
R6 v us &.1) SOURc6 'TG R M PO R.
OfGaarsas R E Am es "
IotsT2 AZ) , *TH & .S U B C O N He?'rd E C H A ut. M A r/ W I Lt. PKova G A 8RIEF R E POR T. A F 7& c. 'D+ G SoScoMMs7TGG CHAWMANE R & f% A T", kl G H o t.L Oo.scuJs eog PR o pes s.D REpog7 OA fL ANT- .3ficI FIC f% f f L e C A 'rf o d Of S APS ry Gl O RL.s "
. .T re m Io , RE PoR T 0 F 7'*
- PL ann o a c.
'a m q @ .Sc H G.DvLe> L fRocSDUREs $U6CeMMfr&B, Poe FAe. day H A.s 8ss a no v s.,p TO
$ A TU R D ** Y. Th ls 77 H & Sw7 &*LL lb & UJ S 1) TC) .2> t.se o ss PRoyo.52.o 19 ea.s .R S p o s .:r .
[ %I if 4
- ' f.
- .
- .- -l ),'f llf ab:.Y.^j. . . ) '
'Qp
- _ ;- '}'
,q' . : a o y . , ,
O,.,, m.c.m' ,;;2:y aw.. .; O.: ..u;fk*h'.r n.. ,. .;-.g1 =; ['.n'4l *...-..,,u'".-
nf'2 ~ : : ;a c,
u,.:;. . l- n: .
\'i. A .
sg.N: . - r) :
&.Y.;;\d-l:*p.fift
..~,'o,., *
,~ 3.::.;:z -
. n., ,y,.,s . :m.,4.:.v.n : , ,,.. .; .ty%y.d+, ..=g..g :y ng,,;, yw t
. a 3. n. %;;q -W :i% . ' .\
..). ;:~...:
- - .- .:: % a .m: ~ p; y . : y '. nn ,g . . , ..:; . >m.. .
y."N]r
. .", ~ s....,. : : , .y* * :. n W;.e..mv ,*. ;,:i' : :e.m:.:. ,,.r._.., . .
- ...' cf.:, .. sy:.W' .:t.,t p;.
g;.~;t:
.}'
y
. . . .s
. .: . 'Wx : n . : I a , t, ,P. < .. ,- , . . -. n
...7 ' -;s:.oN y . .
0.s ..': ..:. ::. ' ...R - .:? : .., :',-
'. . ~ , .; e ..>
- 1 .
s ..';'... ' . ;, 7: -
-@1. w.- ' . u
- .-" . . . .. , w : .%
c.
- p
- * - y e,
}D L t a ..'*l
- Y: VY:. t .k
,V \ . . :3 'h:4 ':; ?. 'k'[.? .%.*
s...
. 1 %..',s'
- Y,h. $ II !*.,re- ?.;
- 5.
.'.'b
- ? ?
.e.- : *
'- u. A' $: ;
'b ^:: ..
y -W..,. , -
/ 2' c:s :y.,* :1 m;. ?7 '%. :.44' .:;.if.,.y
. ' * . s.2, ;h t: *. '<:q ::. J: :. , . :. , l :?'
- 3 ~ ? * ' 3
.::y. . . }y!,.l ~ ,e-g, * .&y r.,v 7 3..
.r
. .!. ;le , .y .: . o, . ("c.'3
- :s,').. ....;>
.i . , a. . . l ;,. .: e
. ., ; ; M a' ' - V.wf.
,w.,*
g .:::+3.,-. ..' .n., ,:Q::* -
$ " . ' .. b0::::
u.;p;Q'.'. :;'.w' .e,
?..
y:..;n,
'".7
?.: w';lF', ~.-p&: .1.y.'
.c
- =l* 4' f;s:t. , ~. ;.
.a :- . p: ....h-;'.,...'. ^, *T,y'.;#'.'.-.
- ; . . .p~p ..:...?.,,.*J..,r.
.a ': a'l- n.++. . , , . :g,; . ;Q.
- :,, .,- --..,..u Ss s,% 0, f' r j e ,:.,%.i.'
i O
- ..< t .; . 'c. . ,'.~.. v. ,,-
.1 ,vt 4y .as_. .~ .:;. c' . - . . c ,..,..:y i. .;,. ,. y. . .'s.%.;'"-*.':. ;-, ..;. . ' .,~p,/...*:.*
- ;^
!< : ;;as : ; . C ,. ' ,.:.
'; -; '. c':c.-. -.,: ::Q \ ': ;' .
...%'.,.n',:.,.~j...a.-. .t:.f.g . =.
e,;, . - p 4 . e' :,:,.v. '1 1 -.;,..w .: -,y- ;
t
. ,. 3.*
.;: a . ,,
- t...t~.+ ,*:.
' '. .* .'.,..,.,'s.- ,. j.- .,, Y,T-r
- ' :, n l,,. , ,
e-..
g ;,: '
- ,~-- ,'< ^l b ..
,'l. a :,
. . . - .. ' '.). b'. 0' i:' f _W !.:.
,;,f',;l,hl' :_
L. .j~ ;. T.._ e ' . d- ;, ' ; '* .' % f .l$.l,: )* }.' ."',', : z.*
- .; :.; *f,_'.1 q:;,i 'l 7lpf: .?ff:c..g t
s' , ' , .:1
,,:y.:f:..[.Q' < -
,' :
- h, yo,,'. ... \ :- .s A ..z;?.. .'. & . L;;_j.;;
y,
.h_g[:;. _. ;;;},
., .{.
,,. ,._* % ' . . :.;,'.Q;
.- : ' %.j. ' . .i
,n
- c. ;f . '..' ;[' , 'l^.i' ?.' f,), .,
._.b ;;. :._ v jv :*.'- . :* Q li.'.;!; . ',. ,g , . .}.m c. , ",: p':,
?. lA . Q. ' l'r'j ,- ' l .~ ',';{. Y.x. :., 'sd.:
. : ., , . . , . , , . .' w. .;; . .
- b.' ;;, .i.. :*- : . ..g. j ; _ l .;%.".._
.',G. l,'-.
.:.3;..
[eQ-,-:'.. . '.* : '-l?.ha':,
^*. . - :.. : " ; ?,? -..y '_ t > ?, * .: ,,,," ; ; ; . 7 .I. 'e':l . .
- ' .".._ +. !,..u Q:
A.' :.' .;,_' [ .:'.{ .. : .[' f,;,:.q-l j . . _,.] ;'j .l,
\
? 1. ? 3.-*
'Y '. } ' ^ . ' 's ," . y ;4. ,'*~-l ['y' '. ^ L *; *'*. ,.f',^~. : .* -
m . .: -
9 '. .) l:_ is '.l h 3 . , ;.n.l :, . ' -
V,
- ,1. ( l.,..-
\ . . ,
- ' .?,L ' q .l f;:' ,l;. A _ * : ,, :< * . ' ';. -:. '{ 1 l l a . '- . , . l
'. l.=l ,~' ? ._.':.l ';* . ;. ..=*), .;; ,- f' T:. _ L NT ; ': _
,. ?-
_^gl.'),,
< ::) . f.) & ' ,,
r.;;;.'.l'\ -: ', (,E:h - l:. Il -d : '.,, .5
,:. . f. 'n-lJ .* W M. *. . .. ,' ... f. ... ,.p :: 'i :,,' .-,I$:(.i., \ . 's. . e.:
== __*, .'t'- n r
,'.y, ,l- . ;:-"q*';', , * $ ' '..[
,..e. .J ....-..-;'.n -_t.- ;*, ; s.:& , ','.,;_,-I' ..
M .
s .::- .y.'. ~,.. . :- . :
- * ,'s:
-:- ^,. . :W . - a;; :.;'Q.
0.c.p:..'a...i..'..-'
^'
.:.:l , _'::-l_'.-(.~,.'*'l:.?.,.'
,,,,i',_'.;'.'.**.v;- .'
- ..b '. ., '~., ,.fl', :' >,' .'
.n.:,-
\ l.
n-
= -
y: i- .
,,,m .- .. . . . -
. v. '1s ..V'.;..Y,'..,.e.
, ;l ',
.'~-l..i- c , _ q.;M .g .,y , .
. ';;- W . M :[ \l..,.; -\;; *.; ,.c.. - . i.l::'s;;. :, ;rg %'_ ' ' ..:*. , . .;, j :;,. q: Q, .; , ._ f . .'.u;._:7,;j,);:
, .., v-
- . ,. ."..; r ;.. +: . :.a'. ; . ._.,.';'.
, e...
b,. _: . _a. ,;. . . . ,w: '
,_ . G . '; , :s
- - <.,.,y*.;
il .- . :. . , .e':.i j: '.;
. . , ^(,. .,
, , ; . . m:, . z. . '. >
- ...
- ;.g .3
, . . .g
- a...:.
s . :.,:,.a.. . . V:'
.; n: ., .'; .* _.v. :
-. . , _ , , . , ...; -.q.,.,,
4.:...;,
..... ..); , ;<
...j...,
' , ...,.;...;._....... : *: :;- s
'.:<. .; <.. .- .a... .. n *:' ?. w
.,..~.'. :*
..;. l s~ .,v,,.:
h . ':( ; ._f:y ?. .l..; ,h,7 Y f. '^ .f i
' h , l .'. ' '. ' !^ *
.r }' l l, r ;&. . . ., ^' 'l.Y
'l. f _ ~ _ ,. . .f
'.;l{,*
, ' ;'. .' ';_ ' :'; . Q. ll' ;,:'j_ . :
, :' - i ',. .,
l: _ lq~ ~.., lb '
- N .- v*
-';; ' , _j' .
l .'. >fl. ' . . , , . ' . ~ ' . . L
_,..C.'4's.d'='. -
- l. ; ; ? ._ ,, . ,,...._. . ,(( y ;;' . f ' . **' ! ,{ _., %;; .
- j ;l, .1. . . , , . - -f, ' [.' [ '
.._...',s .. . iT
^
- %_ . ..t. s
,I%a , - } , V. . ' ' . , ' ' ' , 'd .. ; . ; ,; ; ' _'
- a ' *- ' . ' ' ...'...'
'. } ' ill'?..N '. ,. I, y
' :.;y;l.- ,; - , .. . % , a.P k ~ "' s ' .f. 'r!.[.. r .e ' . '.l
- ll,. ;; .1 p:: : ;i;.' z *', .G.:L-'?.
- ' ? i ' f :l,. ; ,;
,u..:.'.- -k. 3,. .: 'n ;. : , '...i. ,,
. . r.y.,..,
. ; _ _ } , f_ :. ' ' # ;Y;;.t ..[n: Q' :. (. ;:;,j . .;' ." ; ; .,
< - =
j -:s *' k .. . . ..','
.: .;. ..; T . .y :';,., e jl.l.' { _ ::.,p : - 4l..;,1,:.",':. . '. [.'._ :. ,l,:',* _'_ 3%. . .*."s ; ,, __
. .l : : [_.,,_.m .
, . . . ' .' _j f; ' ,' ' ' _ %_
,(*k, ,,. .;' ..: g, . .; ' '- { .a ... .',; . .g..:
.l, e'. , ,,
'u;
.9.e.1. . .; . -'; ,,,* ,-,, .; '..pl5 i.; *: ., -. .. ,,.,d'...':S, ' .:t:;.,"::','* ;9., .. '#. ,*,. ,'... ', -'m'.-#
h- ',' . ;.'
.e
, .. , - . . ' . . ,,,u.,'.,..;;',c..,*. , , , . ., $ .. *- : - -'<'i',,,..,;, ...
- d, ., .s ".,'D. ,
f.. . c.t ,; : ;
' .' .',//yl'
,;..{..,,-: .'N - *
* ... ' ., ',e *'.b..'.',-[
.'[.
y" . , : ,,)'..
, - [;.:,,...j., . . - .' [, ';-; '. : .:'..'6...;~....'.,,
hy ,[ 'y '.;$e . ' :r.U.."f.'k' l ., ' '.
-[. ' . #. . 'E \ ! ,/ l . - .: - ,.1.'.
*/ , '
. ."[. *
. f. [*...,,;,s..;.'...,....',
-( *'[ ". '. t
.=..'(,.,! ,,.'.. ,.... :?
[
. .: . [ f.
* ": .) '> ..s.-,f,M. .,..,;.-*,,v'.
,, - [.i a ;,.... . . , . . ../ " ;, . . ,'*'Y,..'. o
- ,...,;.; ?%.,- . . . 'i '. ..V.glb*/s?.
. p .
'g .
;- ' *' u. ? ' 3*; ' ,
. l *' t ' : ,I.? !'.;'_ '
', h i. ; ill:. ' :' ..i: ':. :.. 40;:
. Q { f, ,l'f'L1'!,' 'd 'k; ;f . 4 ;.o. -f.f*.e._, ,_. ';? ':f' ,";* +lc : f; ;[
- ll,. lk . ' , . ' '.*,.;' *-q'll?:
i .~ . ..; M- . ' ' , ' ' .. _
'; d.;l : Y,~'. : '.5 . . ' ' '
*1.. $ ^ k.*k
~
^l -
'. : ;. ,: .Av, ,~_ ; ,; ' . :e ,. .. r.: V . :,.:.* . . ~!'::'; . -*.' ,_;:,. . . _ ,_' ,_..i'- :
. '_ ; _ e ' ..?'
-: . t :. n ; ,a. .
*..q ' . ,; y*.n. :
?..-~: N .,* '.e;' ;-. ' . %'.
.... ..s
,._';.' .'.: > i u. **.
- T .
-* .;L -.* . c.
. .*- . ' ,^ . ' . . . , ..%'.
l',, Y . ;' l; . .. -: ':'.' ' f i ;,. :,* * ; ; :.'5 ' {, ', L .
,'l; -::N f.: ' L; Y .
.- '.',' , ; g a pl:: . . .'))'.' !. ; ...: . , ;. .;'_: ; '; ll . __,,!.';,..l.,L,'...,T',._,'.,', :r;_;**. , ' .-l ' \
'.j
.: _..._l_.y
- ll l * * ':,._l;g
. ;. :;;';9. . ,; ' ' , .; [ - ,,-'3:.."'_. y :*.,'_;,7 1[
.:,....[:
. ..l. . .'. .,, .:... ;j .,, . ,,.,, 5,:l . ~.,,'. .. ,5 . ;;. $ -'.:' 5 ..
. 6,e : . r .$
?. b, '. ' .. . 1[ .'.
. . . ..... 4
. . . , w . . . ,; ,. .
y Q c o m ri Q Q f. y,y i g ; j._.....:. p Q <: J
- v. . . . - .n..... f .f . g. .. .s.j%f_
-v;. ~
fgyjQ, '- .. n . .
. - ., ,}' ,' .N.'
y ;; ; _ l,.. j; :: . ;: '_'.:.; . :n
'~*.
[, '.
s
. ' . .' ,l..
) ,,Q,..*:<'
sf , _ ; * *.'l N D, -S us : *:;. e.
' $:.Y..l .,.. .;;-?.Y . jf'__; }D ,je .. e> . . , ,z _ .s ._. ;, _ W. ,y )'. .y '.'. f '., l.. 3,- $ 'O. ._ ,')r
.4, ,y.l'.;.,0, .1fc , . . ..,,..,.-',.y .c y' .; 1x:
; ,, : , ,{ { .;
..y. ..<- . e t . ;. . ;
.u . 3f . ,. ,-
X. .s . T .; . l Lj .:s. . . ' n % . ; * " ' ' . Q; ~. < ' ^ ' g; _.i ;. ' - _1* -' . - ...-.nn :, ; 2: ', _j .,'.- ..h.~.
._. ;' .' " ;l.. ,'_... :)?.: [' .Q':n,,
.: r k
- _ ., . . ,;;. ?..'Q:
_ . q ,: em -. 1 r . ~ v,. '.s *'. ; ._ , _- _'L< ;?),.y 3.. y ; l A- h :. ). ;..,o ... . ..:lw',' .'?..J., ' ' ; . .' ..w ...
.*;;R i ~ e
;. .a:= * :.N,...~.,..,f..'
; J.,n . .: f-lG; . '. ' .y
. . ' . , ' , . .Q .; e. .la :
) ; v , .: ;;.g'-.{..
'. , . .S. ' f. C ' k,[
. . . . - ',;<*'.:.' . . . ., n n . ,* ! ] .
~ :.. ;.. ; ..: }, L
.. .:b . .
4 . - Y , " , .
' f.9 -.; c. .v..:c.
.. . - .....i.. .....y...,.,
t.1. ..:<.. m,. . . . - - s-
.. . , g::+, . .
. ~r . ..: , - -
#., f.tr.,., . , D).b. Y'? , ?i
.f. .'. . , f,.;h. a .'
[.hw.: ., . .p2.n;* _., # lf. . .L. %'. l a v
.s. , - u> ........ . .- ;f.;7:. ' _ .%:;. .v......,.9..-
.Q... . .:u,p..y .,.,-
? : I,' ' f,. ?.
k N' f.l A. .. j:,' f '
. :* I
.n.,..n.1 ;. :
..q.1.L. ;;. .%
4r; 3;.,>u,_ .. " . ,n ^ l sq . . x. . r.. 6, 4.: ;l s.. * ;, .a .: .: ;. ; . .s 'i. 1 ;:
..: ':: fl: . .c . .. ;_, %. *
,".l..
_i
...y.. La.,.y .::%. ';; .
, , . .. . . , - ,_ ; -y,. . .a ; . ..a. ;
',.v* ( ,..
c..sa,. a.;.;.. . e. , ,.
. ., r,r.-1 ; .. ; .,
- v. z.,'
3 . . ... . ..,
.. . . a. ., :.. . r .
' 1..1i j . ;?. :s$.4. u .
. i,3 '. - -
r :%a m - - _ . . , .- . . :..
.r "$. . :-.;~,-. . .
- ....{..
.... < - .,,_;._...u.
4 _' ..: *:::s.' . 3 , ' :: R;:' .: , . a.g - %. 5;e. , n. :v i
. .t .; .. ', * . O. :; .. ?L. f.l.s.,.: ._ :,i v.
.% . q -}' y' ' .: !...'.. : ,.
;'..;& . ', *:l.d *i .i
'.' ?,l i.' ' _' ,; ~..-(.%....' .
. :. y -) . . __
- .: ,: l.
^
-,(? f,3 ":. a..I.* . cy : ;j; ..lf.:.i f..?.
. 9. . s h?:/:. l{..o[:w. l.'.&,
- =
i
' i :. .,.: ::{ . .,' . .;.;.r*
.o S. :%. ;. .. ; ' . . ; .f':f. : :.ly; ;Q,:.f.,; .(..v 9","a;
; lr: 7..v..
- y~
. . . . - - yk;.'
k:f;5:::q t +..:. i. .m - .::..l. ' ..;;
. b, (: ..:;z,'h,J. ...; .,b.
f;
. ..; , '.: :. .. '['?.6. '..$,?lg. .
- '. n(, .. ' u. :.; .p.i:V "g; . ;:4/. y.*
- . .. v : ,. s.
P. .i.; . s., .. %. .:'s. <:s - p. ,.. . . . . . . . . :.','...~.... : 7.. Q..y,. .,q:* ~
. .,-s'..,.......
c . .. . . . . :._. . . ; .;
- 9. y . ..c. .1, : ;; .it ;.;. ..- ty... . ;.._..,,.f.u,:c.,,..
,~ .i ~. .....e,- . y. . ,w. . . .
-gn .n. .,c: .. v _;; . . - : .,.; n. . . . .
, .. g.
.6,
.., i
- r.- ..
's. .c ;;..:. ' %' .' .,:.
., - /.>c'. : .- 1 .;, .s . . ;i i-. ^:" *
'.#.*'*:.,...,e;
,V:.
1,., .
?e+ . d, ~ h .. f. . . :. ,.;. : A * ::".- a -,j s: : - .,; .b:f.
.e .
3> ,s .J _, : ? g.J ' ,. . c ', s.<' : - J. T. ,
- e - ' . . n.s;e , %,
&, :, ' f' .. ;M j:, :: *
, <;& ; ;;.,;r. :.*.. . :b. . ". ,? . . " : ; '.; t - y;. : ; .'n,. l ;' . :'o;x_ !.' c_.1
- u. -
T. ' y ; ' ].: .%
- .'.! : - ' .% .l . *.; ;.:
.*_G!{ p. .
?, , .,,.~.:l ,
;5.
, . .; ;p)l j ~.?;_'b l. . ? O &
. .; i,: ?.z.s .;! ;_.a .,: ,> Qf'.:*.
e,.v- . t ...f.~.,.
' ;: :j.<_ '. ;.; '.; :
'. .r['l.,.. ~ _ ;,. . %.Q.. . , q :,,' '.f .;.. ; ? m.: : r . '[,* ...'. * :l. ._ ' f.', . . , . s,_ . ;[. ., , '
"< M y ,,*,'._.::'y:l .,' V
. gJ'r.. y M , :.- , .' ...e,..,y.'......,..:,..r'
! y:
, , . . .,. ~ , , w.. . @8 '...t@j,.,.,. -.e- . - . , . . . . .,..,....'9 ' , '; *: :., . s,;
- ' .?' . . . ' O';3
, , c .
^ ,
^
.,. ;I , . < . .
.' N-.:. '. [s:l .. __; .. q,;.: .1., . cin . lG;;.. - . :c n. .;i:;;-Zl,;.h:;}%._ . . ' . j,'< ' % y . ".' .. ' .; Q L;
% Y. ?;; . ; ! .p '4.;; .', ;_ ' ,,l g -
- q. . . ;_ :: ;. ~
9:. .- (79.-..m..n:.
- . . . /' Q.. ; u .q. a- '. -
.'.,\ t..
." .e ;, '.':.,. . .*.;, \:y'r. ; y:-' . ; el:v: p{. '. ,. , . :. ;; .
rt ... 4, ,, ,: '. .: .: ._ ~..;.y.; :; ;.,., . z -a",
- ~
.;; ....: .: ....; ae n . .,. ;; ,* . . .;. - .
- e. . e, . . ' ....;.
- <: ... .. ., . . . . . . . . ,'::. ..'. 5..
.e s r.y. ..
,f....".,t. 7 ;. .,,....,:.,.~:<.> :_ ;..s. t. _. ,p.; n
. . , .: :.,,.u...'.>u..
E , V". ;, . Si'.
. f, . e
- _?:. ,- *-; P. . ,*....g0 -
.l :.:..y-" . ', % ' m.C. .::.;.i , , ..- < ::.. ,). ;: d:. . ., r:
O.. :.':.& . " : :. ;. :. . .+ .-'. . . . ^- . . .':'..T J ':. , .:.'~ _
- , .';- >^: ,. . i : % );'
..a,.. +-: .:
Y .- .-
.. .., . ;? a.:.*
.7. I;i. . ;,~ . *, ;,', .'y'. .i;,. _: \ [.;;{. : .} 'd;; , u f. _,v,,,;, '-;,_ y' . /.: 3::;.. : .' ..;.[.[. .:. . ;f,,;sl, y .'.:., . . , Q,;,'y . : ' '. V . :.
_ . j'-; ' _ : i. : y,;.
;.:.j' :; ? .:;_ ;y ;.j
_ j.
'_ ^ ';M.
\ ?_. . , -, ; ,.3. . . *. . ' ' , j:tu.,. .
'; e
-. s _ ;; . .. . ,:,; ,4. . . .q f. . . ,..t
.. , . } :, , ;.
y. p _ '; ' ,lq.: .*.;..........
.: z '; : ; g. ,w;e:' . :. ;iG '.: :. ,.' ... .s ..
.';py, ; h. , ...
.g. '.;;; ' . g: r. .. + ': ._
..;p ., ,.;..L.~3
- e ;.. .. . .. : ::V> , .s,,...,.3...~
s
,i , . . ,:,, ~,, :g l.
k-'
>- .- ~, ,.
a
..,,~.
,.l( .. ;.,- I'- - .. . ,-- . ,s , :,, ... ."?... ,'. , .t"-
t
- ,-: 4.,.. .
<-,.b'*',1',,- 3
- - : . L :, - '.'.,..d._,.;* , . ' . ;,~ i '. .-./.: l .'g..'[ ..s,, . , $$ g.
...'....j.',',:. , ,p','_ +
, , ]on ;'.7n r i
- f. . _,i , ")' . _ ' f .f:;e- .'.,;. . . .
~....; ..
...%: ':~..,. , .
- - > , .f E' ... ,, ..
j.O . '..
g .,
..\
- l. p..'. ;; ' v,,'*;',a' ~ %;~., . .' ' ' ' ,,t b: ', l,.i mp .s.',.l . j;'",.J'- .., W : ; .' ; , b ; M,\ e ; ,',3: ;. " . .Q. 'A .
$ : /,-!.. . &**z ?.$ ?'.N-
.,.- ;,O. ..' ~_lg ".":_'_'.;l ',f:\'$ . W: ..',',1
', . .. 9. ; ' Y ,l- '^ ^ F. E,.
Q . .a r.
Y'..
a.';s.. _,;;n-. :.-;f.k . . t. , : . . , . 3.:: .. ; , ';i :: s 9 v. c. .. . .; > ; ,
.: , - ,cL * .b', ~ s, . % : , .4 : ., ,,c . Q,:;;: ,. ;v . ; ,*, . f. ?_ r..- .,: - . . . ~ .. ., '. ,, ;We %. rg ';.,.*H. . .;J.* .
L~
....'...~'.;_..;,_..'.:
. r,. a %
M' ,; ,a,s' ,., .. ..a.r _ q' ,' ....'..._. *<'
-l #4 :. ..,.;:. f.i,:....'_ .. ; . :
- v.: s : :, .. , ;,
- _ .
- .a *'
-+.9-L .y 4 ' . .,:*;
, , . . ; .. . . . . ~.~ . . q' .; ;. . -
'*,; .; .. 9 ..
- f. . . . ' .b' {. ; .ya. . -. '_':.:..l j .,[-' -:
- q . :l' s
.K ;: "y
;" . a; c , . . ,'. . _ _' 3. " I . - <
T_f.~'$ . ::.. +,
.g, ;- y,
- :q _ :. n y ,' ;; . : fy
, .; i : ,
?. s, .-
. ..r...w, a; ;. ; ; .;9,.
... n.....a:
- . ../.3.,;..,. ,,';3.9.. .y:..:r._ :u'; ,, .;, f,.; c, .l,:. . i;c. . g .?: . ._ :
=---
.: a. s-, ,., ..
~ . s. . 9,
. , [
. 2,::,,, ' . . = ,;:. _:
.,c : . a* , . , . , ,<..'.,,..;., , a . e.- <.c. <. : w..
,.;',.,;.<..,,,..s......:t_.,.;A. ,- - + '.**:
; : s e
.,..,.:..,;_.v.,:....-
- ...s
.m.
g
..:'.... o . .;_u. ,:.. t..;;f .*
., .. sp ..
r...u.,. v: :.,*. ..,: * ... .-< c .-:.: , . , .. ,., : .- '.. i ,.: ..- i.
.4, V i , . ,, .-. . . . ...:.m,,..
, . , ' . .p,,,3..,,.:9...-
's-,
'L .;.: . .$ . . , >f,; %.r;n :* . ' 4 Ev .. -
_. : . o.
- . - . . . . ~ ..a
- :. a,-
.,.: 1. r. . -
.y," *:: ,
) ' f. q , , . ; ' ' '#
, J' . : , . ,-,.; ,'.L,,N f.. 3. *. N.d.q 3'... -
'; '. .. -l s . . fv : S::- . ly'+ .s. '4 * ,-l..,;*:. 4]. *. ;Y, r-
' y :a*..', ;y,. ,r' g ,4'%* :t y*,'. . : .; .. . - , . a .. ; ,.c' ; . t ,,
; _ ::G. _ j;
. ~ ' ' p. ' . r * .l' .l. (. /(::,' !, l l, . ' .. .; ,m .-i".:;4 ;;-' .:2 ;," -*e.;.n, . ;
'.; u .; , og g ,', : ,- .*,. w,-
ru ,.. :
. c; ..
r, .j:, , ;;..
. . y' ," .2 s. . ' , , . '; _\,,% v , yy ~.lg ', -> -Q; .. . q? (, , s:, .. , ; ';q ' .. : .y ; :,'r. .. :: v * :
~ _
" ; , . . ~,;' -
. s ; , ; <. b l.;.l . . Y ::,':. ...:
'u y::< .'& * . W . M .1 ' y;...l ';. , : ;
;.o-
- .., 7.. .i' f,- Ci . ,, - i'.? .%ls
- ',' :, , ,::l:w. . 4 . ' ./:;.;..' .. ; .% , ' '. .. . . :: 1. .y . '.. v.: . . r,; '_.q ' J.
p'_.' .y t;y ;' ;.-. .Le. , . %: ? .,. .
':~ ~ ? '_
'.~l c
.:' ' : , ;, ] , ...a. -: O. .; . l:9 f f ' '. .[;[ly . .:;in ' . : ',' ,.:(n, ( . ,f ' 'Q.. : f. d. . '. 3 ;:
y , "-; . . _; >:
}. .e ~ }y. ' V ' ); ;,
=- . '.,-=;. .,L.p.>y.',.Q
- ;.,. l : .
1 .
' j' . 7 ;; p-j: .,..,,.g'.;5*,.; .% .g. ,. . p, ;, ,,9 9 . l. -
,e.M .T ' 'l.:. .} ,< .l;,';, ;+..,s,,
,';l } , * '. i .#,;. . , , ., _..- . . ,?, .;. ;,,, s w I's:_ :. k
,"9.... : ;
.. .;^;;
)* .'.
' %,. ;, j..-l
'. . ' - -'- .1,,-.j.,;_ j.l.:;
F. . ! . ' . .'. ,c;4. . - e t_ . 'f,!, ,,, 'l. e ,. ' j . ;. 3 r ..<.s; a.Y l '. '. ' . 1,.:g, - 1 ;, ' z-l:',W .';.;. . ,'.,Y .L.~..i: l: :1l'Q ; +n' .: .c.,I\ *$ T ' p - ~...'r..-(.e .. . . . .. p .F.......,,j;.-_.':.:t
;_. . ;. *(.,; , ; : t.,' .;; . . ,. .. .:
s..
~:. : ._,..s...-,.. . .:.,.',p',
- e. ; *. .g2.'l'..'. i ' ,'.~ ' . c ,, , . y . , . ,v P.t . .. ,. ::..';... .s.-v..y:
,s s,:
. . 4_,, :.,.. n,.. ; .,..,~;,r,....-9 ~.
-(N,. ' a:p.
s
. .:. - s . : . L..;. ' ' . :l. .?, ,llll,k, ~ ;* :9 ..y '.-
+. ._..n. '.. q . ,,: ,:. ..*..,e,.', , .. _ . . ..:
;.,..,_...;._..,...q,;,-
c-
- .y -
, ' . - ,lg: . ;
...c..- .. ;f . fl.,.
p.,+.,}.'..
+ c v ..c. :
.se ;:V
- __ ,.s . ,. . ...<.
- <---n..y
..q -<
s .,~.pa-.'. g; o
; ;,, . . ' < : ;.: e, 2.. ; .c:r,..,
,1- . n,c .: .. .:. ,; '~: ..-
,.;p
,-- , ..y: .,
= . -',;r; ; ;:- - ., . , , , . . . . _,.
) -
'. ;', ; 4. z .
- _<u,._,[.. , i., , _ ^ g: .; _,,
~._.__~q_.,t.
.- ' ' . . .: 2 e..;;1:lq::;,;,,...-.. : < .
,%, '*3 . , . ,i . : , .';.4
; ; a < . .' ;;;, . ;:*;',., , -; ~, g;;; , . ,r2,.-... ? ,;
Y ;, .<_ ~; ) : '.:; < ' .: . .: '. s,, " . - ; r .** ". . y >- y L ' :.;.' ..:_ : ..,,. ;,.:..;*,;,.. . b,; . ,;. .,'j,.,.',.- " " , .y,..' _ .; . _ . o.s.
., _ .y y _
- v. : w ' . . . .. , y: ,;, ,,; + .. * " _, % ,: -y
.s 2 '
* ;l t , "' * \ ' _s ._._'_ $ ( ;).i. . .'r L ';; ;:2, : ' l -,o'.b ~; . .. '.. , 1._ *.Q.
- ^- .. . .' g.,'.
', ! .'G,; ,::.~ g_n '.b ,, ].E O. .W '. . .L..yu; !.,. \.. , :W.,f:.:"., , . . , :; j, ' . a: ,* .'~ '. . ,j $,0.,' . 'll,3,s
~;
<. ;.j:n!'.N ,. ,s:. ;. ; y' i,'.?
,.t..
- '_'.1 L
wi
---ii--. .
__ _ . _ . _ _ _ _ _ - _ . . -.____m , _ . _ _ . . . _ - _ . . . . _ b "
* *;; . : d' i;N * *?.; N .fG '(.f.\ ; 'I'. "Y ... . '.Y ,*', ^ - '
[ ,,. ,,
'Wp{'l,j'yl:y ll'!'- if';: ..'V.t, 4 f 9
.I ' h . j o.'.{Q. Q, lll5. ,4 .l; ,.?.:. ,, *h - {, *:$ !N. b ,;. . 'lc b :
..Y W '.,l'.'l} f;Y0 h ., lk h.?' h .'![f, : $.h'".' Y;f. .'.I&_$,_(:'Y. _ -,' :f'; }.* ll\ I'k '. l N ' .,t,.,'.b' %,.,"r r}lQ @Y 'if'y! Y;,
i f ' - }'- + y: ,, '
. p,[. , ':', ,".,'g - [e , ., , .. [,,, (;', ' . .';.; Q. > ' .. .,, ?9 [q , ' . ' ;.
, . , - .,,'t t ,, e, ,. . .
J. . ; " ., .Kf ; , , , ,I, t * , , , ' ' ?'ll4 .'.). .,'; f ' ' y' .% l . ;. , ' Y ' ! $ ., ,y _ . 'l } y;, ' . f f, :9 .'.).,.f.yj: .) J. , _' ;.s )_ ,%. ;. . ;; ;:. ..). ,p 's [;, ' ,{ ., h.,"}c j g'. n a: ';- Q.t,gQ. c
' f&: .:. y.js '; j.','-Qp .
h }" /, . Id' ,, , j' i ' l . . f'
, ,;, -. .j-'f ' ';. ,: ' ,.. f' ?.r ., '.J l [ [:) . ; . ' i.
f '3 .d;fk [ ? 9 E,', ' ' [ a '. '.N**'... M,U,[ hf:7. ! E,. 'N',';; j '*.
* .('.y Jf ,'.t
-f.jy-
'.._.:..y"..e' ... . . '. - ,, .; , w kf
'.. n. r : -' ,". : y WL; c : -'
; , -,:., " *; ., : ..;. . . _a. ::_ y _. '. - .
) 4
.: 2.
. [; :
;.,' , ,y;y:
. '.;._. .s,'1,>
;<;', ,, a *: . ',) , ;
- : , _ ' '.;e . *$f.. ;,; .
f,a'.; , . : \ c_ :.';* :)_ .y ,g ,' :.) s ;;>. ' f.;. ' ' i
..'i..,','.m. ..(,........,,'..;. -t .:. ,.,+1.:,,...,6 a. '%-';-.,,,3
..,.;....,y.'__.,:t*-<
.- . ;. S , ... . ,, ..
-.- : - .; ; >- -t n : .* .s.-.
. .. 5,, ..p .
' , . :.,c. ....a. ,l.... . ; ; ... ..y,,*.'...... .f.~'.-
5.'_,,,_..';j.;,',;._., *
- s. . , ; , ;
. . . .y : ? .e q .
., I' . . . {J , ', , , f; i. * . .
-t a' _ J.' '
,I ._ 7 ' . .; ? .' . [.f, , . , [' g l lJ, [, . ' i ,' ,,'
.,0,"
g j; .. , . M b'.j,..:..,.('-.,,.
.:. ' ;:, .; -i .; . l: ...,'.*:e, y =;., ., ., ' ',. . ;:. '"*._:";,r.. j3. %g 8, ,9 ,.], . , . . . .....d.- , c. S . ,; ,i.*;e . [ -;. ,
, n .y ,<;. . :,, ':, .y 1' ' .','7.. v.. . ; +; .',?'*
d '8'
-) . . ."?:e ';:. , 7
', f.,,
. ' ,1:.
g,.,
.,,.. c. ,; . .! :l J. ;.".. . $ "}., *
- 7.i - ' W. .A . . ? :. ...,f';.. '-.;;;;:r.',
J. ! ' .
- . .;; .t . . ; ;. r . ' : L ., , .. - :w.,y f. g. p,y 4,':: 7. ;p ._ Tf, 'g -ll;O..:
- p . ' y:, ,_S r...,
% 'y
; e< ; ;.) l: ',_. ('..~~,-. :% 7 ..: - A.gl. ..' c
.. i ..
3
; : :t L ( ~ : . . .. . i . -:' - .' -c '
.* _. _ .y ;. [ ' . :l',, T. M ; n, ' . d:s..'~;,l ;<.y*l:,; ,i
!a.,:(}'[ q fY . Y ;l,::+.l,. '.Q'
" l KI Y. f " e , b '.:l.;l:. . ?[Ql:l.f.i'i .b [ U, ( ,,_::'*v:,A $ j . W ' I. "; : t ' . ..);. (,f'. ;1ll:,:l,'.f,j,,,[
,',..?.';* _,s...-...,
~ ,
';' ?: , ' ' ' b 1;
- * . :f' , , ' .':1;l)' ! N f._* ?' ' ,' I ' * ^r .l k, ' . If . ,.'{ :l .,, *l *\ . j Y'
< Yshtl. .h .' i
.:.,*,):: ;'i; ., _ j. .': .. ( ',::;:v(,)3 .3. - ^ '; ..:u . .:
,I,: f....,c l;;
2
.. .. %s :f,',.': ,:.,',,. .~ .; .,-[:;, .; :..'_. ' '; ;.. % ,
q ,,, c... a' . g.:;: _?,.
. . ..,.9, . ,,,.d. %.'5
;, ! ,;.;.; ;:.y y' .'.. ',' 3. :
% ,, :. 'e . .:. y 7 .'. "*
/,' , ':' [* . ~. . . ".I. **Q
'T ;;.,, . .: * .,.['. N : ,V. . . , . . .
['I, L E 7 #. I ,'w . ' ':'. "' '
- , ; o
{ 0;' .d ',! ,. ...;;)/ e
'o jcf . -g!, .M ,*," .' -.:. ;r , ',y
- f'- .-: , .~'. <',4
.I .; a'; -
* *s.,. ?. ", f : l, 1
1.
' ', :. j ',,': ' L',$,,a, ; ; ;.'... .. ,.. 1.; . . , ;. '; '(. '., ,.../. .,;f.
.f..., , n+. .3.e..,...,ja . .,# b; .; . ,;l ri* ,g. 4. ',
i
;;' u.
,,.; .,'u.;..-..l'- ;,.
. 3.;.-. , .:,,,.. :,,.,.,. .
. .- ,j . , : e.,
i...-,)>;.., e'd; ?. . ' , . . : r
- ' /: '. . * ..,,,: ,.,Jg,,,..',,s.,,
. %p. .;*- . .,. ,...+r ' ,-
!-r ....q 7 . . ., .n. :, .,s 4
-.4* .,.. c .
. .1
~
;- '( ,
. . . , . , -;....,.*f., .; . .;. r y,
'_] :-l ,;,. ._' - : . ..;:. L,.,.l; . . < . :, j_ __ ,'Y," *ll :."; ,' . , . (;..X'.' .-
\
*; [' . ,; ' *y'. ' :
. ' . ,:'- ..~ '. ", .., 'v*.;_ : . ,'.,':: :.3* ."' q'
;,.*,,'.....~:='.'..,'y.. : [,. ; i;; ' . "[:.
'. .','.'.L
\ ; s ..'.'c. l:;. . .
- : o .- ;; , .. ,,.m.'n l %.. ,. ' t y ? '-
- t".z.;,.,,,...L,.,,'_.W;.',._., ),7 '. i.w.,
g
.,.l,..e,..._.,4._ ; 9. . ,, .:'.'.'. ' , ' , ' .,_;, . ); ; ;
,p ,
- n. . , 4
,.n k..
. ,: ..., , _ . . - _ ' . . .A ls. _.,,
,.,- .;; 'q i>- . .
r9..y,_ * ,l . ' .: :.. . ' ,. s*l ,.
, . .4. . * , .. .:.;...c....g..5 . ..,
1 i ,. -
?.
, , ., , ,.; ; ;,. ,...c...., . , . . :; :
..t :, . , , . .,...,,c .o; ;,:4.,., .- ....,%.; . . . ' -..,;;.......,.i....!;.
.c u,..
. , , , , . ,:,. . . . . . .J., .. ,, . .-i > . .. ( 3. ,.
W-". :
.. :q: ,
.Q ,y , ' '- W ., :- ': < m.- \, ,7. w: c/ 2',.';'*,'- , , e
'.L
-' :._. .'.' ';.J. : <[l ;_. l: . . , , f., .,. ,. :. . . ., ,_ "
.k..
..."v.
.* . ,; :. < ~ . ., : g.;u -: e. : -.*., p,.;.,-.. _;-. ~ ; _._';,...q..... . :. , .,3.,y _ j. ~.;. .'.: , . . lsm y.:: ; .,, ,, ..-. .> ..w..d. -> _ ..t.;;. :
y
- : b .L. ' ,, --
. ;-;':.". .'.n.;; > L ,. " . ' .G., ;q x.' , :.\,' , . . .
s. r' .- - . @ . . .: . :,...,
- ,_ ' . . . . q. ... _,m. ,.~!- .. , :;l _
.u 7
, _ ,, r . .;v..' i,_ ,l,. 4,.' /k * .: 'i s .r.:
_ t ; r,.. . ,3' g ,, .. ;. v . g ,.7 ._ ?y .r,.a ,. . l' ,, : ,'i .' , h ' \. l l _:',.
.*.,s
.,,,.'.a'..' _.',,).;.'..'l* : *
, .. * \
, . . . . . ., . n, sc. ..;' t l r ,,. ' : .
' '.' .'..; .; :e. - .o_.di,'.
- .:, ;.:nc,,;;,_.-
...g.
g ,,. ::L ( , ; ,., J'. ?.
. .'+ ; .',...',l .:.
. . . .l,
. : :: :. . :. _ ? . ; < .' .,
,;:p:*' . ::_ k.f . ' '.e :. .. .a,
. , . . , ,. : , c: . . .;.,.:.
): . . ' . . ;' ' W,,4 -g .' .,;.'.",,y-, ir,;," . : . .~..-
1_ 8
.W.
. ,. .1,'...- . b^'. ti 4
< .: .. f ,u : ,- e 3
l
..g..-..3...'-.
4,.
- q. _;c,. . : .. :'..: .;.
z,-/ . . .9,.,... : v.. . 7 , ;uu' , :.,.3, w.,_,,....g. c
'<.,J,:<..~.s.
' . /, . , ,
e..
.- ,c..-
< .,..g..,.::;; . :.. , x '.e. 4,
.r ,. ..,. . p . ;;,
;, y
,,, .; %,,'Y.
* . . .;s.-). s.:,..,-.',..ar'.1'"'.':,
- y. . . .: . . . . " ',
r .'.'s\;,. :, ...,,v..
. c c ... v f.
<, o
- .['l.' ; ..; -l . (,' f f. ;j; . , ' . ' . 'l; ' : l .. _ l _.L.'._', b N;.v.:. '
l ' . . .; l' , l , .. , ;, :) . . (I ',.,._
';; b: :
;vl.l - :, -:?. '. .?' .' !. : N 'l*
y .' ,J; ; r, , . , _ . _ , , _ ,, g
. .. .) ,, .,;_ :.:.;.:.; ; . -
c .;,y . ', .
., ',f.,. ,..,.,,,,(.: _ p. l,?',. - .7 l
L
. . .y , . j **.t.
..,r . .; ' ;. .,;.. . . p;", . : ',.'4 * .;- ?
'. 7.,.._..J.,
{
,..; : ),
_. ' . " . u' ';_...i'.;',1,..,
-v
- x. . ,, < . i , ;c ,' -* ** * ,;' ,', , -
;..? - ' . . . ;. .* ;, ',...sl4.., ~. '. * , . _ ,. _ . .;; '-
.,,,r . . ' , - . . . . . ;.#: :. ;,;',, ,;- 'b . :
,",'.;Qh'.,
! / ' ., s e- . . , , , I % ,L')
. ,. , - :...s ,: : .;. ., . ;.,.. 1_r ; ' , ,
, ' . " , . , [ .,;p ,s ;; . ' ' - . . ,t , , , - , , , ,,,. C ;. , ' . ' l[
1 ( ., i. ....,,,,,.g....y.:;.. :' g',,.c.
, . ::'s., . - ,"., %.
4.,,'...,.y,;:,;;.',..'..,,,.,;* . . . ..,s, .j; - _ * ;-
- , , ,,' .-; .s ... . -y...,., :. 3: . , . , .,
d* : . ;_ * ' ' ' , ' ': P n, , ./ , , . V * ' _
- _ ;'l e
.v.
J* :
' ' . ' , ..~..;.' . ..
. y -' 'a,s - _-, .;); . ~,' 'y.
y'*.',~',',,p._ z; f;
., ....%.d -,..si ?; . ' . .. ., , . '.,' ...;-.,y,t.--.. s, . -:. .'..
e :
...'.h.=.,.c'
*,.'a.....,'.
i
, ....)
., l . a v -_, -
,,'j; * ; o-s.
';.-.e- . g. . . .;.',.-;*,f._: .e ..
, - _ . . .m,,-','
J :
'; . ". : . :, .. i- :t ,;_.' .:. ;:. ; : c.
.a . .
' ;:,.., .; t. -
_ ;% y . :'.,$. . .~ .' ,..- =. Y s.y , ,. -#.,'....4.
- ; h. .: : ~'- ; _ _* ' O. '. ' }. ' ' ' .; .se ^Y. ^
., . .p' .'. N .~ , , ;,) , ., .,s . .g . . . ."..,,.;l.,',,' .- .'; : ' l i.,*:.;*
I i>', _:)l, ..e ..y ;%,. : ,. ' ~ ' , :
- a; d-
'. ;l ..,.%.
*, , , ; pl . y' , , ,- y '._,..i,.;,..
'.-..a- ..,
' . 4* ' .i. .c.,_ ,o-
- _ .q 1 _,','Y'. ',,'J.#
, ,1 .- .l(. : ,l ' ,);' i ' I :' '
C, . [ ','. ',", . .):* ; . * . . l. ' F 9 s*, . '.... _, ' b ;[,' ,
. <.-(.
'.y - ,
j j ..s. ',w.. -J-5(:., J j ',;'.i i / N '
,, .,. ,j , . - .,.; . . ;; .,.,1 . , Y. ',
,?.,;, v,.."'i'c;.' ... , .; .
- E.[
,g...._. ,. . . . ..- ,q ,,q c,;.;;
;. o .
- p. -..... ....;-
.,,..,t' .,,.i:.,_.s.,r,....,-.;'
v '. a .
.. .;; l.- . .,: ' ,- ; . c. ._ 2 . .dl. :.. ;? . . ; . ' .'
7..,. . ,, .
. ,.: . '. ..',._m.
, , v. :
f ,i
^ , . . . . - .
'. 1 *; .';' -s', * , . i,.'..,..*!<
* ,,,e,. .- ll-
...i).' - . ' ' - ..
- f. . ".i . . . _; A : .. ;* -
..s9_ e . . ;, ,;., '
. . , , . --!s-
.- ', . -- ,- , . . *. .p, l,.. .
,.c ,%,.,,
c,.,.. . ,, *., , ~ . ' M.... *:. ... - -'
., ,y , ;.
, <'..; *- ,a , ;-
3 l r -
': '- * ..i *
.,c.:-
~..c,,..
. ;; r . - *
, e .
ie c' l* . , '
'*)--
I. ..
' ,.;,' ..- a. .
,I. p ,' + l^ \' .' _ ' ; l' .. ( .: .,; [$
','l _ ' : - . ':
ff, _.':.,l ' ' 'd x , , ', _ ) c ; S . .. I . ' ' ' . , [ ,l. . ' a -
,l "'
- ' , ' r:d.l ,. . , i, 4 ' i ' ;, i
',l.',h','.,},,..;,,' ('. ll' :
. - [..,j ' , a l E .,. * , 'l L G : .! 'A '. lq.i. f '
.n.,;,...... .l. ' ; . : ,, '-':,'>' .- ^ . '; s '.l ' E ' .:. , . . - ' '2; '. , -'.a , '..t';, . . ' ? : ? ?. '.
*.\' . ' . , . .,: v'
.- *. . *. l'n' . a.h ' '
,s
.4 . .
~..,*
l 1,u/ .y 7,"- .,', ,.,- ,.I;.k,'.',.[*m
'. ? f ,; , .- . , , ,.;. .
; ,; s. t .
' . '.. - . .p _ . ; ,,
p
,i3.,
l , ., _ n' .. ' -
.,1,,., _ ;- '. ,,a.,.,.
.h ,' . p f.;.:. . , , :. ,i..:f t. ?:l f , ,6
, . ..:.t. , . , , .s,,'.',..'.-
,'+ i,,
, ,',;... : [ - , , .
, .) :9
' ,f f ...i., ..
c o . ;_ . ;;; .l .l'
; ' .f.a . .*'*, '; ' .s, .;' ;'..'..e.
.,,y'.
.. i
.. .f,. . ,..,'.; . . . .,< _. , s;_.*,, ,. . ,;.,;.,s."..'.
e.
- ;'...- ,,su _.q4,. , . , _,:,,
;_,, .;; . .i .. ;,% . - ., t.-:. . . .g.. ? . .'. .': :.e. -
. .#' 's
- y. . ,....g
,...w
, ., , , , ^:
,a
- :u- .: ..
- .'..t
'.r
.- - - - l ; ., ..; ,
. 1
- 7. :- .,: .
5
. , , . . . . . ~ , . . . . _ , _ >;;c ; ->
r:.. 8,_. ,
,) Nl " I,. .'. 'k I;,,l'( ;;.3,.c
.;; - . ,. l: . '.,, .pll
, ,...'l .. __'ft, W
* ...'..,.,0,.
s.n.,. s 'i,'f.,l t. . l;- :: c: -.- . -
.a . .,, -
., :b b f ' .. a -
hl .;,. _ : ' j ;V '.5.1, ; *
^
? l :- :- ,. . -
(- I
\. ' l . ;,b '-... ?;'.? . :"' . l ' _ . , - ~ ,: l ,l ,
,f. ;....%.2'*-ll * '.Y.? ....; 1 .I., W; . f $.t '. v. f '- ' .1,s '. i. .I:
,;-l,, .] .. :. E ' 'u .
,, . ., . - i _ .; a.'!
b ' - f .i; -' ql; .
. ': s ' f f lll ' ,
b.y,',."..
- . f y i.- '.s',,.,
< ^. . _ 'll ..'l *
- l ,i. W :
6._ ,s. ; .?- ,
...,..l
..L :' .
' i. !
'l ' . ,.: ;. _.' ; y ' '. " ? ', 'l '.{l , ..
. ' ,l , . . i.Y ,,;, . '
,'.-l.. . .-. ,. '.!:.,c. 9,
' l..l:5 . ; g ; lf:,3 . :.;..
- 3. l:3,.. . .;;,. !A -. ..; l i' .
.... n ;, i -f. ,...
,e.,_
- n. . .: . t';%3. , ., . . . . . , ' -
,.,...,e,- .
; .g, l; ,,J. , . . ..,
. . . . ; p; _ , _.n.' ', j . ; . ; , , ': ;'. , :%
.L...*, ; ;:- l . .e' u
,:' . ' 7: e:
, ;% d,. ),' ,,', f. ; '!p"'.,~;;.)
;;'i, ?? ,.N ? f 4 ', p.';, .,, .'. 7, q;'..*)' ," _.;: /. ;;' . ' t; ' :
aL_,T; '. . y,.
;- { j ' g .% *;*q "Dl,: \
<- ,[: . :* ..., ;. ',:r ,
.i j , ;7 ,_ S',r: -. '.- ;i. ; .'.;. c, ; ,'. :e ; .'., ;;._.p
,..,n .
'.Yf . *
- . . ," . - :) '7. , ": .,.;.. : ._ Q - . , ,Li ' : ; . x. ;:',.,_v,,:, .. , , , , ;, . vg. : , b.y-I . \ . : , :,,? .-l ' i ' ' ' ; :;
.; j n '. , f.j. , . t &,',.; ,,
'.,4..,.,,..,....
.. : , ; )' J. ; :
- r:; v .l,jo -
a.,,_ /, ) _ c'
.'.p'- a.,,. l;.e lj:1.' ' '* . < - .
','y...
.s ,......_.e<, ,"....,"g' . ~ -
.y l 4 :
v:.
, ;%.. 'a.:;,, 4 )-,L, _
, :y..
,,'. ,g,.c.\ ,,qc ..> ,.;1.5. ;k,.1_. .g..,,_ .,.,; 9_.
,s". ; , . . ,; ),
. . ; .?.: . ; ;
.m"cJ',:- . .a ...sa ..'.('.-.i,3.;.,.,,..,_.,,;,G
. ,. . ..i.>,.-3.,,..c'( .'.
e, . ,.. 1
.O,s , ' t T. ;' ;,.t._. ,:,. 1 ,. .
. . . a.\,.3 5
I
,,e,s- ; g ..i,.+: ...,
,..g*..t.,%,a....~..,p
.e'. ; r. . . .u-
*..' ' ( ', ,. , .; _' i
, :l;c.:
, .Ay * . ;;p , . , Q
. v.
l: ,A) :- [' :. , ;,l* , g .
- mQ'
, : x.4 '. . , ,y,y; ..:*.y-f
- ,.; .v -.,-. } :~ . yt .v - . py, ' q , 4_ . .g 4 n, ;;,g ' . .') '.'r.o -y< 4' .' .( ',. , n.,. h*l% : p 'j ,
;;;: . , Qb, 5;f.)- flM; '.V - u !.
r , '""M y .
...t ,
.,,Q,
- .c ". . c
- , 2 ;
",.,-l'.@,' : - '.: i::a.;:, ay:,',
n :., .. s : ';; - y ', ;;. .:,. n
; ' ,g y 1. _._ - .
..l: W >N :.W:,.,;._h
;g c' rp l ,g ' '. ',a'"
,~_ ,_ ,l' r:._ .q%.,
i6 A ';;3 .,1..%.'f.;l. _ H'y ',y M. .c.E Q '::; ,. .h v.*; Ww;.
.:*.. .; sa. w' u'
,u.
; 'w..r,n c e *x
- .c
- :..'.,.,.:,...t.
l
. m ..r :.v :y ,';,',
- ,.,J.'<~. n
. . ' .y . -g- V .< < , ,
< :,. -g't.Q. *p lp.
. .,y, . .y
-r s . o .' r *. ; s ,.-
*; m.c::;nl .,
g".,. .s',qll'.: A . L. q.; ,)'eg ;;g, : y.,J". . x J ~ ;. y ;','
.u,
;rp' ::. .-( . f. w. '
.y: L,q ;.. .. .n.
, W'w y ;,'s :: .
4 f,* ,.;y,, ,s.:.. 'A rps".s. ,:. p
',%'v,44.,[, %.,
.i . ;ty: : . m . S. j, M h . y' l . . . . ' , , . .o _
. , . : ?_?
.- ..... 't
.p\,,. . ul 4U "c Y. . . ' ' ;. .p c , ...y.
.. . .x, > . <.
I
' :, ' * ~
.. . . . .n p . vyQ.,j..'i '. 'llt ' ,we. ' M.b., ;;; ; , M .'a :i:,' 6< s'W :o'. h's. . . ,H9 . u.
v' %, 1.;; . : e
. ;".. . .; , :J:c .R *, ;,l. ' !i ?, f,~ ,k. :: .x,g' W'ql.,;.(l'!'?: : :<
, ' ?
a;
- l.
. "Q,'.n._ ':h.,, ?. .g _ : 5,4.;,Q @p. ;;;:V: . %
f..l'r.%w -;;4: . .. .)Q:,s
. ,; , v.f\.',.
\ . . . . s. ,s', e !? C. :)v ;"') m.9; 9 ,kQ :.y.. o. .:. n. ,Q,'o,.
- r*. #. .:,. y; , o:. v ?,..' &. &.:-.':; yl ;b.f.l.~.:..W . ?.Q t.;
l 3:. >y . ! , . L.::7 :., b' ;g l:;:,s'y*., n, vy .. :c: : ,;; t Q g b"dy' b.s: <4 :)y.;M.c r ' ,:a,,q. y
.s ,*l;a6*&,a%..;. ;..'A ;;;',,; k:,, :.
- . . n ,n ..
y:;y. ' Jg~ .-i. 9 . ;;
.. u.: :,,9, _2 .f:a,a -. ,,,
c.'
* ;;. HjQ?,g,:
.9 * . 3:. g . .;.f. .' L,y .4.]
h,.'.<;.. l . ?Q.- . , .
.s; t v 'y . y , .S ' D ; ' .h ' '. , ' %u.. .:, ;., v :.. '; ' ,.y9.v'.?. Q,b vn,
, t ;,
r .. o rV. ;.; .v:;. n.
",.:,'. g ,*.. .?. .:-%;
I{ ^'l.g,h i.gfp.' .:l:. F ,I.~,; y 'pt;;lp)"^,, }.{ ,.b,' Ji,_' * .~:;:',,'.h'.j{'.Q~. 1 ~; ' X'.
. W' y:)y,p 0.lp.
r
. ,'.i.k 5.%
f )*}A.: % l, ,.,I~l. l_ A[ P.llh ',U L:' (~ ,. l j,i ijO L
~
;j . ,. ,?i ,.\ ), s.:.: [:' :", ; :, $ . '.T [_\ _ ,a'I:.s .:.h^.l ^l.l.? l-l .:a $; g f ^ T.&h:.J). / g'YA, l . , =.. .
- .;., f'{} ~,;,'.Qh% ^j. ; il .n yl 3 },.. q'll,Qj.;'h f Q ,,p^'.. .
- rl
- .} '".D
* . . . '. l ,j .;;' 'f ;, ' ' '
l . - __ . -: .
.i
^l ., _ , ;.; q . , ,, ': :; ;.'..' , ' . ;, ; . ly',?. '.. . .l ..; ., . ' ,", v:+y:
l
',;"-.l :. ,' .',.<
e...... s-
;,, .), . . {;
e;. ,.,,;
..'_,,.;.;._.f.-
-" . , ,.;.\'y' .p.. :,e'7,e..a..
I
' _ _ )3.",..
, '; y Q. ,;.,* Q, rd. :.;b),
r 4. ' ,' .f .,t . . ('; * .,.f. 'p:,;; .y :? s 3 > +;: g. ? n :
* * . , . ,9 ' :. ..;".::",,,., .,. .;:,1 '
< .: . .c: .: - e e ... p; i. . .' v, ;w' . .f. . n.:<; ,; e . t... ;
- 4. . -4.,;,..,',5,....,...c.-.c,4,.=..:.~,.2,., ..r..-...v..- ...a..
m *y -
>+
s,. ,.
.f.:,..,,, .', : : :.:. .
.a .~$;. . . .g....:"i;.,.J, .,..: ....,. ; ' _r,., . %. _. .(c, :p;<. ... ., . 'p. . . .
~
m;. ,.e.. s. j'
.. . . ~, -
. ~
.. % :' ,p'i r .;c_ . 9. .. Y, f .,,_ ,: '.f.. u. :. .t.s. <: . . .;. ..::l , ,L:.',.4.,. .. .l.*
..m e ,.,;'.,.:c..'; r^:'\'.'- , . ,. ., ,.. .' ,_. '.g'.,L.
.. : $.. ,r g .
,. ,ie .Q , . -V . ; ,.' j.. p .;. ,.l .. ':..'l.~ . ;,-l.._,; .' ,_. .. : y:,h
.4;;;. . * ,_ p ;. . ,.'
a t,;. . . ,
'C ' * , '
- ,, ;,J.,: ;; -
' , . , ' w; ;
.:g* a;; . L '.) .;,;[L
.'C.. ;;n, + . , . . . , v, ('. ; ; ' , s. ' J ,- ',;- ). ;&.;.,;i. ' "'; . .'
. ...;. ;.l..' s . g,. , J ['u f'., '.; , ', , '( ? : :s:;,,- ;.,,, ..__: ~W. (l;,;,:. '%.,'.:*ls- )',, ,: p .: ';, ._,,.., ;'* \ _': _ .'L.; -l.; ,'. 'D.b,..,.: <?\ :. 'l .,,.>.? ' x.; " _?' ,'~.'.L 'T'Q.,.,';'j:y .{
*._.;;,'*....~.7,v.. : * :'; q I*.. . . . ....,.1 Y}; :' ,,::;,y l y;.
.% l:. .f
- a * '
4 9 ' s . . t.%. ;.
.' .,*,, .. ; , . ,;,\ .~,. 2.g
%' : .:-' i. .".
- i;;.c_ ,: ., f
_ ,j. : \;. '.s.'.:.') *?' ' l. :: . .' ;.y- ^. : 'yl ' y ,.l.., l'."',.. ;- '. . , 1 : .*.y .v. , d -: - . . . ;'
, . ..,,,,. .; ..':,..s'.:.,
, , . .,:, ; '. : y -: . ,,W: ., ,,..,;..
,~. . . <.:.eI;9 . . .; . ; . ,, , .; .- ^',,:!,,_.
. .~. .. .. .:.,. .y.-.>;:. ;. g,, ',n .;;:, ;,', . . ;._ .., ... , . ._.,4.,.,, . ,
,,. ;, c-
- y. .. .. m .. . t. s. c... .g.. :+ . . .'., . . ; . .q,. , ,.y s ,;.* . . .;f.,.&,s - ,. : ..,,, ,.,s, . } &.g. ,;,
6,, :: ., 9 - j. ;. . ~. . . . . .. .. ; . .;; y . ,. :y _: . . r. ; -'.y,;...", ,:v v ,%: '. :.9._:.,...' ~ m ;, ,.c ..s..... .,- .: b. . .. : ,3, '; . : . _v
;: ,;': Sy,.; Q< .. _,; ;.. ;;. - ,, . f._ u . ;.,:..: W, :, >. ,_ _ ,. ,; ,;
,,h' ; y' . s 4 :.. - *.s 4 .,e 4'.. L'..** 4- - .. , y
. . ' . . - .. , ,, .y , . , <?
~ .: ~. . \ '. ~ :.< ...3 e.._ - . :' ,,. .
- . . = . ..,.-
.' .,'..:y."'
. ., ., ' ,, [ 3, t. . '- . _ . :: . ; [l. .,_ : .l ' ' z ; {, . ', ,; j, , . . .. . -fs; . . " ' ;.. ..-'ll ' :] :l : :
-l
', .. .. .r '.9,'............._.,
- , . ' . .:- :., . , ,'::., ' l} ' , ', } f y '. ,
t.
. ~.. , ,. . . - . . . -;m.
. _,. .-._ ._. . ... ;.. ,: .:. .._.:.,' ; . :.y ': ; ;; : . _ ;; , l-
.s
; ..; q;,. :. .. . ..-4.,.,
- .- :. .. .:(.
- v. .- ;
. -. .(;. ..,, a . :; ~ ~-
i
. . .+ . . . n.:
...v;. .. . . ,
_ . ._ . . . ,v,, . . . c :- s-w.
.: . ',,:. . .~ .;;
..p. . . , , . . , .
4, .
',..,Y..;...
~
~
't **:t$_ .
!;N .!.;,-'.._q,. ,$ I_ ('; & .,- - ., ' , ' . ,.f.... . ,' ".; * & : ...: -- n, A. ? ' ' ' , , * ; , , t
- ' l _.:'a , ,'., . , l 5. *-,ec . l fl. N . 4,' .', t '- ' , . ' . ** - ..v' l. O' . .' .;%-...: : i;.. . . . h.'.'j:1.'e y
..\....',_f
.....a: " . : ;,,, y. ;; , ..... , ,;,a !' .* :],V . -lt-
. '. \L ..."...~..?.'.,e ,..::.,t....
('...-.- .;l...- vra , .; , i o .
+, .- .- > - .'
'.v-
. ,,c.,,,y'. d, '. '!,'.': . ' '.c , ',.i . . . ,: . . .' + -: .., .,fc,.,. . ..' y' :);n'*~- :
t' ; : -- l 4 s. e z
.,, ,',[::. ,. A . -' .f, .: ..; j, ]. *. . ;; , '. . ' ,. '? ;: 8 , . ' , . f i,(3p
- e. ...
. L j '-
,k ;k . . . , .l l. !. l '. . . . . a.. . ': .,l. ' , , .. .
{ C;. . l _ d4 (,7,;,'.".*,;..' . . .. . ll ,?_ "j '. . l. . ' ,;ff. .,' N.? l 59:" N . ig:r.l . j.,l r , ; i $;
,, . N .l:; - \
g -*: c, Q&, l,{$ . :..glf. .': N $. . Y,*.k...h'. li,; .;; y , . ;, ,, j, , l*.O l ! ;,.e ,e u_- :,, v. - -
, y
,,.. .. .t r,c. . , ;r . .,. ,;', . .-
+ , .
, , /.; ..; y , ' . . .( ' '; ..-*:
g,, i .
'}+,..;:". ,...%;..
y ..i,..,.. 4,n... a V .:.':,' {* :; . .. , , y.],,.' .[ .s .s,,7. yy:_..; y;%c, i.. . ,.:*.,'..: . ,.n.e'.' , . . -i3.;;b,,;Q;;,.. ,,.,' ;. ...z. ,[id,;. .* g(,_ *,.; i .
).,:h. ,k ;' o,. ;[,. .,.,y
\;;.[ '. ; f. :.+a>i.v ...,. ..
- l. j. e, 7
.; ,;Q h:- * .,.: .t ;;
's' W
..,ff*e .. . ' .' . , * .
.-*-. *s . -;.'. r'*,h....,,
Fg L. .-..'k.a. y , ',c '~n1-- 'I*,g *. s'
;q.' i.' , ;; !c[/f ..y U:'. .... b. .m
' . ' s-i,... , . _ . ' , p,
.y,f '7.. .?,..
)- ,,[_'..' [.,. . ; . \ ,,. .*,Q .c -d
. \ T:.. .. -:.' h.5. .;; ...... i. :[ *
' ".;* s '-1. '
, 'p. , , . l.' ..i.-
l....,,__ ,, , . _ . . . _ _ . . _ . . _ _ . _ _ . . . _ . _ . . . , _ . _ _ _ _ _ _ _ - . _ _ _ _ . _ _
_ _ _ _ _ _ - - - - - - - - - ~ ~ ' - _ _ - - - - - - - - - - - - ~ ' - ' - ~ ' - ~ ~ ~ ' ~ h
.N
- 4. :
\, ::
lz , '. ,j: :g';., ', t, '.j. i.V- P. ,'g . T'Il ;. . ,i; ..[y, _l. 7 .. .y Q;.,.<. ,,.. J.f". m- .r
., g :.;..
;, l'*l,l '. ;. . f 't: . '....,,, av.ba .3 ,.p . . *';,...,.,.c.. ./
;,, , - : im *
- r. .3 , .
'# . , , e '. , ? , , ,,
i.;l_, 1;
.__. _',l 3. ,. ... , ,
. t,
- 3 ;, .
l,Q '
- l' . ,';, .(f.l, ' .[,,Q's.7.;,.
..i.,....
; .,. f.r.Q,., @ , ::ff;l;. . . Q; f,.
,Q y}ll .("i_l' : Q 'A , . .;; .: ; , ,;,Qfj. 'l. ." 4. Y;).:. ; . *, ::. . , '
' fr.q'::.>
^ 'E j :
e .: . .d [_, , ..:.
; l;, .i M'.?.:$ % i eL p.~ L ./ ' b
;;ll'{}: [; i., s. .[.: *; ;;::% q _
3 .l :j,N{ . Q ;.l'i. f';{ ;l$ .f. _ n ..(\.;.; .l',
'.. . .y. y ;l'_ ;
jr:;_ q.y . y, 3, _ * ,
,, l.
T.~ f._,
- l, ~,j i j
- *I '"..:': .;;TM ff:::'i :: h . ..., . ', Y , ' lt l ?, ,'j:' h . '
* '. * ;;',. q G : , .; :' ? ,;;: ;.)
y 7 .,.;- ". ;:h. f e,.
;
- _*. y,4,;;
. . ; .s -
. 'y ., .?'Z. n .J G .K y. -
,y - * .;, $ *
, 'y.,
,s, p'...,9."*.,:
f...'3,._[\,]'.,&"',Ql-
,,, ; 'j .. , '.
- . '; '7 ,," M f.
. ..;.' ;% ,.:, .',:[-. l' . , ; . :, -./? - - 'l l (( , . ,b: A , L lq. l:g
. r .--;. ';:l.-.'.;,.L*..*~.f : );',:. gm.a,
* ?: .
.' , . u: , :- ,_':
1 , , , , , V
.! I. . g. . a* , .. .. ' ' .:: . ' ..: l: %. .g . : , ..: &. ? . , ,
'. ,' ,. ' . , . . . . . , ,: .f,y ':f ) '; 7.; . :;/, : ' ., ' . _
,o, ; ..
'.y ; : . . s. ..;
.:..'.I. .Y
, i ;.j , ' ' .. l . - l'. - l[, y...l o~.,
'l ^ [.kf .} ..k.1 '-
. y .; y _
y,_;,_.,..,,:,_. ,.., f K.; E
-(
, ';, , :.z, -
ac
.{ -.,, e,'
9.'. 7l7 ( e y, .. c.,. .
- .. ': ... ,) i .
. . . .. . . . * . . ,-. . ,._ .i. .';" :.j_ ;';, :.:. ::, , ' .':, A .. ./- ' . - .,,,. :! ; - , ;. . . ; , . . , _> . , . , , , ,
.u ' , ' . . . ?' * .i
;, ;, ; 1 ., l,, ? ; ;:; . , '
y,.';,zu: y
.y: q!( e . ?' .> :
. . ~:._ L. , ;r, :: q., : . <; : :
f .;.<. .: g) ' u. . .:. . . r y,, _.
. .; .jeni , ., ..-p'
) ; ". {...b,..- ':. -[ .-;['.-;. ;.'; N'. ,'.?+',
l
~
. o' ,f,~'- - ';'; ,;.'. ' 9[.
,,,., , 'U'., },g
~,. . ;p, . :' e, < ; T :,,' , , , ^* 1 ; - *,
..f.>{ f ::jg::('.f'
< . ; g .l-* * . . . . . . ,..;..~~y...;
y, ',- . i :2 ,' :. : - ' .; . ?, .,}. -
'.i , ".:' '
' l _;,', ,,; ,. .,y 3., . [. u ' ;t
., . ';.;; n y-llj. ,!,. Q.' , . , . ._ . :; [ , . ' . %. . . , , f. . l '
.,f.: F. ,1. .,. . - l. , ? - . :
..f. ) -l '[ .:'y :.]. I,;';'...
; N _' b ' ,l'
.c U' 4 *. ,, ( .; :.:) .,
,. _, l,;.'. ,._,.: k,:j,; lj"$),,, :'; .' R(_ ". _ .;
;_;,, ,,y,_.. ., _ .;' . .;:; r . '.F . . .
; _; 9 f. i;.' ',, . '.C '.:
. ', . ' ~ ]; :: , Y N,L> .Y , -l $.
,,l."*- g3 . ., , , - -
.. , , v . 'Y_ , ;l, . _ d . . ::._,[.'_,
, ' ' _ . , . l : v, q.3 _ , . . , ,in-, . L :: lS. - ~f'Y\. I ., ' C;'_ lr.,f i G
.: . .. , .:. { . ?,l '.*~. ,?
.s. .'.;,;.' _ .; . , ., .h
- j ' c: '* ,.
. '- .'_ j .. y ;,(?*il* , "l'
#;1 .l . ' ' A ';' 'u 0
'g l { . ' L .' S' 'l ~.,. < ; :
A .
'_ .' &; ._. ; ., i ; ;..L y,,
Q ' _ _ ; ;1.G'l' -l ' h, J 'f' I l: 4'l ,r , . :
*; ,, : - l ,' < '
. A.; l* . i :. ' ' , ' '
? .::. ~ _:\
~
- _ . , 'l .i; ^ ;.,... ','b,, .
*- '. , ; y, *,,,; . ,'. .j. ;~ . c... .,'.,.'l ,:: ' j
[..;, ,' , ,, j.j( ' ll/ . . '.-"
- 3^ Q' ',,,(' ,,' r', ;3;; . .;, yfl:::k. y. -' _ L. j. 1 jll;'f.. ' .,
. _ , ' . ~ .: '.
;'d ; ; :" s'
;L., . .;,.
._, [
' &'_ w3-'.'.. ."'.r .
. . ."',*':... . [. 4.. l':.,Nt '
[Y'; ; [.;, I , ; '. ,. [ j [ ,.].I s ( '. *y *
.r .:: :. .
' . . ' , )' 9. '
'[ * - [.l ' ' h *. <. '.%.jl .
ll _. 4: j' ' *
'l ' *:."..'.',-
,3
; s ,, ; ;:. ,
,; ;. ; ^l : .l., Y
' .'. ". ; ~ '.: ',$' 3_ . ; ; '
.y;- Q o. ' (":- . ,; * 'g', }i; y
- ' ' ~,;)
'...L; ' '
*, 'l .':* : , -_ :' f, :.li: ;';
. _ x; ._ , .c.,
.;: ". .u ',i .
...n.'. l ;; ),,
.l ,' . .. : .i'.
.. ,s
'l ~ ', ?.- .- l ..' ..;si; {.;:;.4;L'
.,:, .. ( . g' ,...;, ..
- v. '. ., , n
..,h h . . ' ' ,' 5
~ a '. ;; .s
;;:\:$ i;
.l. .
..'1- , , 3. . :; c. .,'.y
, i. J !;,' ' l
' : - ' .;' b . v
.,;, ., . :. . ._.y....',. Q *j, . _ ;" .) ~ !: '
(. . ' ~,; _ ., ,, 4;, '.h f _
. _;{' j ",. .,: . :; :*v* ?. e. .:.' ':u. '-'?.?;; .. . _ . ~ '. . -.-.. .' . : .: '7
+i[. :,- ; ;}.
n, .' '
' ' . . s
; ,. . . q l . l ; _ _ c. . ,:, , .:._ .. ~ , ,, . . -
f .. _.:,'_',, ..c._ i, .y.: : ;- {; p } . , , '.
; y .: . ,' ; +" , n. .e _ ; .. : ,:; . . . .. . ze , , * . . ">x . e,.;,, ';;;,l , ,, .,
N ' f \ f,.!.
"..I...:,% . . ' . , ,; f:';; . ?
, ' '= '*
- '. 'l ,_:, , ' ..
'l ;.Q: "
{' ' . i f. . , .-
' : ;* : 2: . ,;,
; .. *., * :; .. i ' ..-l",",'.: :., , . : ,,~ le
"; ~-
' :l ^;': e >' f ,[p, , -';* _ ' , ;;. ' , '. >
- f. ';' .:..[ ;j_* ,,l ' ?:'[) . ' ':, .}.:_L ' _? :~ ' '. \:' , ' . ',' ": i
- N. l, J. ...'.. -
...7',- .y.. ,6 f
- lc 3. , .
;,. 7 ..,i < 1. d'-
j i;. - . .' : { );[
- m. R . ,< ?. "
f,.. .. .,, ,g ' , , ..,. - l..,,
, .;n:: . .~
y , . .' ; y_
- j , * , ;,) ;'j . ' *y,;. i, t ,
- . ,9
_( L ' , ;l.; c 'A,' t
;a , . , . . -
; ., }J'.,*.'.- ,'; ,, . .
.;,',.' ' )\'.if. : l'. . <,
.y. '., ( r. . ;, , .
": ,' N; ;-
'.l .L
.g..
f.;* ; w
- y ;; .;. . . .'. . '. ; r s.
- _[ ' .; ' .
.e ' : .- *e n
.g
" . :, ';* q , ;; '; '.._,*;,:,\,'.., ,' p.I' s .>\
p ,,',',--l
. . d .__; ;; l ;. y . '_ .l-'; ' .' .,' .'.;
';'. . n . .. _g_ l/, - ',
,- e: - _ f._ . .
- 7. .
* ~![ }. ', (. ,*- - '
,- d. . ' ..: - __ '(
+. , .4., _ ..,. -
.-. . -:l.p, t ,- . ,: ._ , . . _ .; ;; \.. ,,.. ; y:. -
y , - .,_..i_ ,..;<'.'.' ;.i
'r_, ,' ; .l'; -
.r
.- ,r.... , -* j
. . T.w.?; 1
'e d'
[ , , ,y' .,-;,;;, ,, ;;.
-' . : a _ ,
<a
.m
._. ' ; r i. . , , y .;.o;', . , " f7
,..;', .s-
.l.
- : : ; .1. q,; .: q[~ _l'*S,
, ; ' , ;; - ~
l s: "
. , [* = ) ,
- .7 .f, ';' ,. ~ i; ' .
,- . ?. f j,, s'
.;..,.s.'
,;-' f ',;,,'
5, , . ' , . ,~j
=,; - ,
~ ." ' ' . , _. nal,
.f u.. y ., , .. , . , . , . .
,,,\',
g. J - < -_,/ ,' , h *. .,
., ;, .;.. . , , E ';. - 7 . i.
4 .. ,. ,,
,.c .4. , .:
, y* . I .,
,.'F.,.,,,,.. ,, . . , . y . ;. . . , :
,,t .
c,. e.',.( ,,. 5 *
,I.,
I.
...'.;v"',
_, a
.....,,., . , . p 3. .,, , 3 , . . .
,, - .,c.,1,.
; . ' q;.
,-': ,.,., . :e . ". g . ; ' ,
', ' , * :0 l
- f ' <.(
p"
- , c.
, '7 ,,
i ' :
-qv.4 , "l\ ' .. l } *: .; ' .c'" f.l I;, ..7, .
.g., ..
f...(
.f' '.&. '. ,
. . . .. :? ; .- -
l ' f'
.o.
4 J. Y .
,.k. -., Y p .q f- ;
l"; _u'p'
,f.. . _, : . ': 'Y:f S., p.p' .;;;:.
i ,
, \
. . . . . . ....e .
. ;), "..r_ _
~ i',S,-
) g;<
.c ,s
. _n .. .. , .;.., : .y. .[.: _
y ,
, v:' _ _ _ ; . . . .
c :t.
}:_f. :
f.. s, k ' I ,J', *
', . . I [.
, , - 3 .
.. ..,...;,, 23
' J J .g , '.". , 7. s.;: .;;,, , . , _
; j
..[. h l , - @, [ ',' l ' 5 l f 3; f . s
. . :[ ** ; i< s . , . . . .! , ,. ,(', l', '.,'"
t-
.a . h,>;,*g,_
.;* ,. Q e,fy ..;.
t l:{. ,. t.s
'[
. .g.<. /'[ '
p-c t
;,- 9 - .' .
'l'.:J
,1;h ,, h' p:t>,,
e ,. : y:, e
- . e.) . -.ll i:. ,'h
.].' .; , < ; I. 'c
';g')
[
,? '
s v .w c : . f. i 'i .h _ .s 'w r_y pp. F. g *. X..;y't h
,f E. p , Ql:".
- ' /
l ;f,_.- h g,
?
; . .s.'Wr ! 1i '. n .),'
.qs' k.f' ._-'~ .
,;i
.v,,
'};e qQ, y-.y . g
;)
- . ..L; ft .. . ,
vl..: - . a; .
- a. .; ' !+ ; .&. , L's..pa :,,
', *j' b.Q. V,y.'q;tu-s..
,. L.. .
' O ' '
g;o), 4' ;Y ,.
,d.{.. . :. ; '% * :. .9 ,,"~:e,.),'L- '3: ?,-t - .;::q ,
. ':}' [ ;
.; 'J.O. 9 Q J..:s + M/
...,2.-
*: : . _r _
, ' ..- -4:<:) 't
.a+..,
?:b *
;'~ ;
c ' y'
. ,.1 y . .;;.T p _. :',. ;;; .
;* ns :,
e'j l..:
.>. .. ' ::y- ;
- _j.. ./.
,,n 't,i ; . '..( 1" .': 'E l% -..h, - ' -
- ;'. ... f. :4j' L
- " \, n.. n :;.r
. ' . ~l s
1l, ;[, ? ' .; ?
. . ;., c f ', '
i.q:1 A:V,\ ' m., .'
,) P.e&'l. " . '.Ms.? l
. ';; w i.
;a. s.. ': ;..- i c ,m,l; n _' ., '. .':., .. .1
.y (
.y. )..! , y >[j .., 0 . / . ' .
~.
l{.' f',l,,
.m ;\ . 4
'.,~}}