ML20134J751
| ML20134J751 | |
| Person / Time | |
|---|---|
| Issue date: | 02/06/1997 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-2090, NUDOCS 9702120277 | |
| Download: ML20134J751 (120) | |
Text
.
Officirl Trcnscript sf Prsccedings g
NUCLEAR REGULATORY COMMISSION AcRST-2 090
Title:
Advisory Committee on Reactor Safeguards 438th Meeting TRO4 (ACRS)
RETURN ORIGINAL TO BJWHITE Docket Number:
(not applicable) s/S T-2E2s 415-7130 THANKS!
Location:
Rockville, Maryland O
Date:
Thursday, February 6,1997 A
7SPCECopy~ Ref: air
~0'n'e',"le0".".e;',0mnglgg Work Order No.:
NRC-994 Pages 1-107 i
NEAL R. GROSS ANrr CO., INC.
Court Reporters and Transcribers
() $
1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005 (202) 234-4433 R'
'n,I
- g12ja;797o206 g
jj jh t onn19 T-2090 PDR
O DISCLAINER PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSION'S ADVISORY COMMITTEE ON RSACTOR SAFEGUARDS FEBRUARY 6, 1996 The contents of this transcript of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards on February 6,
1996, as reported herein, is a record of the discussions recorded at the meeting held on the above date.
This transcript has not been reviewed, corrected and edited and it may contain inaccuracies.
'l i
i O
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RilODEISLAND AVENUE,NW (202)234-44.3I WASIIINGTON, D C, 2000$
(202)234-4433 j
1 1
UNITED STATES OF AMERICA 2
NUCLEAR REGULATORY COMMISSION i'v 3
& + ++ +
4 438TH MEETING 5
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS (ACRS) 6
+++++
7 THURSDAY 8
FEBRUARY 6, 1997 9
+++++
10 ROCKVILLE, MARYLAND 11
+++++
12 The Advisory Committee met at the Nuclear 1
13 Regulatory Commission, Two White Flint North, Room T2B3, 14 11545 Rockville Pike, at 8:30 a.m.,
Robert L.
- Seale, 15 Chairman, presiding.
16 COMMITTEE MEMBERS:
17 ROBERT L.
SEALE, Chairman 18 DANA A.
POWERS, Vice Chairman 19 GEORGE E.
APOSTOLAKIS 20 JOHN J.
BARTON 21 IVAN CATTON 22 MARIO H.
FONTANA 23 THOMAS S.
KRESS j
24 DON W.
MILLER l
C
(_,
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
2 1
ACRS STAFF PRESENT:
p 2
JOHN T.
LARKINS, Executive Director k
3 ROXANNE SUMMERS, Technical Secretary 4
SAM DURAISWAMY 5
2AROL A.
HARRIS 6
RICHARD P.
SAVIO 7
PAUL BOEHNERT l
8 NOEL DUDLEY 9
MADHAT M.
EL-ZEFTAWY 10 MICHAEL MARKLEY 11 AMARJIT SINGH 12 13 ALSO PRESENT:
,: (
)
- /
I s 14 FRANK COFFMAN I
15 MARK A.
CUNNINGHAM 16 MARTIN J.
VIRGILIO 17 WAYNE HODGES 18 HAROLD VANDER MOLEN 19 TIM COLLINS 20 WARREN LYONS l
l 21 22 23 24
/~T (j
25 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 3
1 A-G-E-N-D-A ex 2
Acenda Item PAGE (V
\\
3 1)
Openina Remarks by the ACRS Chairman 4
4 1.1)
Opening Statement 4
i 5
1.2)
Items of current interest 5
6 1.3)
Priorities for preparation of ACRS 6
7 reports 8
2)
Subcommittee Report -
7 9
Thermal Hydraulic Phenomena Subcommittee 10 3)
Subcommittee Report -
24 11 Subcommittee on Instrumentation and Control 12 Systems and Computers 13 5)
Shutdown Operations Risk 47
,_s
/
\\
14 5.1)
Remarks by the Subcommittee Chairman 47 15 c.2)
Briefing by and discussions with 50 16 representatives of the NRC staff 17 regarding issues associated with 18 shutdown operations risk 19 20 21 i
22 23 24 i
f%
k 25 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
4 1
P-R-O-C-E-E-D-I-N-G-S 2
p (8:34 a.m.)
(_ /
3 1)
OPENING REMARKS BY THE ACRS CHAIRMAN 1
4 1.1)
OPENING STATEMENT I
5 CHAIRMAN SEALE:
Good morning.
The meeting 6
will now come to order.
This is the first day of the 7
438th meeting of the Advisory Committee on Reactor j
8 Safeguards.
During today's meeting, the Committee will 9
consider the following:
ACRS subcommittee activities, 10 future ACRS activities, shutdown operations risk, 11 reconciliation of ACRS comments and recommendations; and 12 proposed ACRS reports.
13 This meeting is being conducted in accordance
(\\ )
14 with the provisions of the Federal Advisory Committee Act.
15 Sam Duraiswamy is the designated federal l
16 official for the initial portion of this meeting.
l 17 We have received no written statements or 18 requests for time to make oral statements from members of 19 the public regarding today's sessions.
A transcript of 20 portions of the meeting is being kept, and it is requested 21 that the speakers use one of the microphones, identify 22 themselves, and speak with sufficient clarity and volume 23 so that they can be readily heard.
24 First of all, I'd like to apologize to start
(_h
(~ /
25 with for any problems that we might have because of my 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 5
l l
1 working my way into this slot, but we'll get there.
And I
,f-'S 2
know that I'll have the good humor of everyone to help me U
3 as we go along.
4 1.2)
ITEMS OF CURRENT INTEREST 5
CHAIRMAN SEALE:
Among the items of current 6
interest, we have four candidates for consideration as 7
possible members of the Committee.
These gentlemen 8
generally have capabilities in the thermal hydraulics 9
area.
10 Two sets of interviews with members of the 11 Committee are scheduled for today and two tomorrow.
That 1
1 i
12 will require some of the members to be out of the meeting l
13 for short periods of time during the day today.
I hope
]
7-
\\
)
14 that we can take care of that transition without too much 15 difficulty.
I do urge everyone to be sure that you do i
16 take the opportunity to take to the candidates.
17 There is a packet of bio information that you j
18 have available to you.
Also somewhere else there was a 19 set of questions that were made available to us that are 1
20 the questions that are being covered by the screening 21 group which has been set up by the Commission.
And so we 22 don't need to worry about those.
I would urge you to take 23 a look at that list.
Do we have it?
Well, they'll pass 24 it out to you.
And those are things you don't need to
(~h
(-
25 worry about in that process.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
6 i
1 Today it's Virgil Schrock and Jose Reyes from i
fs 2
Oregon State that are going to be in the group.
s V
i 3
This evening at 6:30, we will all reconvene at l
4 Positano's for a last chance to take a shot at Ivan as a 5
member of the Committee.
And we need to plan to do that.
6 So we will also plan to adjourn for the evening from here 7
at 5:30.
So you might keep that in mind among your plans.
8 1.3)
PRIORITIES FOR PREPARATION OF ACRS REPORTS 9
CHAIRMAN SEALE:
We've got a fairly full plate 10 this week.
There are already three draft letters 11 available to you.
And I want to compliment the authors of 12 those letters for having those ready in such a prompt way.
13 Hopefully we will be able to do this sort of thing a
/".,%
+
t 14 little bit more regularly.
And, as a matter of fact, I
)
15 want to have some more comments to make about the process 16 of doing letters when we get to that.
17 We also have a couple of reconciliation items 18 from Congress that we will get to later.
19 Oh, by the way, among our letters, so far 20 everything has got an A on it except one.
And that's an 21 A++,
the letter to Congress, which we have to get out in 22 February.
23 MR. DURAISWAMY:
Yes.
24 CHAIRMAN SEALE:
So that's the reason for that
(_/
25 status.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
7 1
Are there any other comments that anyone would
,- g 2
like to make before we get started?
John, we're about to
)
3 get started.
Were there any comments you wanted to make 4
before we got started?
5 EXECUTIVE DIRECTOR LARKINS:
No.
6 MEMBER KRESS:
We're not going to overlook any 7
problems that we may have due to your transition here.
8 CHAIRMAN SEALE:
Not at all, I'm sure.
9 MEMBER KRESS:
We're going to look for them.
10 CHAIRMAN SEALE:
I understand.
I understand.
11 That's fair.
12 Okay.
Well, Dr. Catton wanted plenty of time 13 this morning.
So I've tried to n;ake it available for him.
14 The first item is a report by the Chairman of the Thermal i
15 Hydraulics Subcommittee having to do with the review of 16 the Westinghouse AP600 test program.
)
17 MEMBER CATTON:
Thank you.
18 2)
SUBCOMMITTEE REPORT -
19 THERMAL HYDRAULIC PHENOMENA SUBCOMMITTEE i
i 20 MEMBER CATTON:
In that I'll be leaving the 21 Committee, I'd like to take the opportunity to give you a i
22 broader picture in just the subcommittee meeting I'm 23 reporting on.
24 The AP600 thermal hydraulics differ in some m
I ).
\\
(_
25 respects from existing plants.
These differences are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
8 1
primarily in the passive thermal hydraulic features they p) employ to deal with various loss of coolant events that 2
eV 3
could occur.
4 The ACRS became involved very early in the 5
review process because of these new issues and some 6
concern about the computational tools that were to be used 7
both by the staff and by Westinghouse.
8 In particular, what drove us was that both l
l 9
Westinghouse and the staff through their INEL contractors 10
-- and I'm not referring to NRR here -- argued that there l
11 were little differences.
And if you look at some of the 12 early reports that were put out of Idaho, they saw no need 13 to do anything to any code.
This sort of rapidly changed.
i r
14 NRR was unconvinced and made a case for 15 high-pressure, full-height testing.
Westinghouse offered 16 SPES, and Research offered ROSA.
We reluctantly agreed 17 about SPES but disagreed about ROSA.
And I think our 18 views have been confirmed, but that's another story.
19 Westinghouse argued from the outset that they 20 only needed information about long-term behavior because 21 all breaks were turned into large breaks and their 22 computational tools were adequate for the task.
23 The need for data for this phase led to design 24 and construction of the OSU facility.
I think it's called 25 APEX.
The OSU facility is probably the best designed and v
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 9
1 1
anstrumented facility in the U.S.
And in my view, its 1; g-)
2 only close competitor might be the BETHSY facility in C/
3 France, in Grenoble.
4 Some of the results from the OSU facility 5
clearly dw.'onstrated how sensitive the thermal hydraulic 6
behavior is to small design differences.
It was clear, at 7
least to some, before any data became available and 8
confirmed when it was in hand that condensation, thermal l
9 stratification, small differences in buoyancy and drag 10 forces lead to difficulties in simulation.
11 Both NRR and NRC have the wrong types of codes 12 for the job.
This doesn't mean that they can't get it 13 done.
It just means it's going to be far more difficult t
\\-
14 and computationally expensive than it needs to be.
15 The actual design has been a moving target.
A 16 series of small changes, ranging from addition of a sparer 17 to the CMT to removal of the balance line between the 18 pressurizer and the CMT and timed opening of Stages 2 and i
19 3 of the controlled opening of the primary system, the 20 process has evolved to something that I believe can be 21 approved.
This doesn't mean that it has been shown to be 22 approvable.
Unfortunately, the bringing together of all 23 the information has not been well-done.
24 According to Part 52, a case must be made for b)
(_/
25 the fidelity of the predictions of AP600 behavior.
In 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
10 1
meeting with Westinghouse about six months ago, maybe a r~
2 little longer, -- Paul would know the date -- we told 1
(~,g]
3 Westinghouse that they needed to address the issue of the 4
completeness of the data set in a meaningful way.
5 Our meeting on 17-18 December was to address 6
this issue.
To address completeness, one needs to bring 7
to bear analysis, experience, and common sense in some 8
kind of a scrutable way.
9 I believe the following steps are needed:
10 first, a clear statement of what is though: to be 11 important and what is not with documentation of the basis; 12
-- this is often called a PIRT -- second, scaling studies 13 to demonstrate that the various facilities, both SET and
[s1
\\'/
14 IET, will yield the data needed to verify the ability of a 15 code to predict the phenomena highlighted by the PIRT.
16 The scaling needs to be bottom up for SET -- that's 17 separate effects tests -- and top down for the integral 18 tests.
19 The effort by Westinghouse to date is 20 incomplete.
Westinghouse touched on all of the needed 21 elements but feel short in bringing it together in a 22 scrutable way.
23 Of the many questions that we raised during 24 the meeting, only a few require further technical work in rs 25 my view.
Unfortunately, many answers to the quest.ons 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
11 1
concerns raised by the subcommittee were by reference to 73 2
as yet unwritten reports or unavailable, unreferencable
\\%s) 3 documents.
4 The large-break LOCA is tre ted in a best 5
estimate manner.
And early work by Westinghouse is 6
referenced.
Unfortunately, the documentation referenced 7
was never finished, and commitments to do so have not been 8
met either by the staff or by Westinghouse.
The 9
commitments to make the documentation scrutable and 10 complete appear to have been forgotten.
And I think you 11 probably recall some of the efforts that we had to go 12 through to get them to even agree that it was needed.
13 I'll come back to this.
I,,h
\\/
14 Many questions were addressed by referring to 15 unreferenced RAIs.
And, Paul, I got it right that time.
l l
16 I wrote it down so I would.
17 These RAIs are sometimes a series of questions l
18 and answers that create a convoluted and difficult path to l
19 follow given that the reader knows of their existence.
l 20 Once the NRR reviewers are happy, they move on.
Little l
21 attempt is made to correct the original documents or I
22 cross-reference the RAIs to the submitted documents.
23 We told Westinghouse that we were unwilling to 24 move to the next step, which I'll describe in a moment, 7\\
k_
25 until the process was complete.
We relented but noted NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
12 1
that if something turns up, which I doubt, it's a fs 2
Westinghouse problem.
!w_-)
3 Once the integral and separate effects data 4
sets are found to be complete enough, the process is 5
finished.
The steps that follow this are to:
- first, 6
obtain a description of the code that's detailed enough to 7
assure that all important phenomena are modeled.
1 1
8 Now, they use several codes.
And the only one 9
that we really know about is COBRA / TRAC.
The i
10 documentation that we had in hand for the NOTRUMP code, at i
11 least what I had, was almost 20 years old.
12 It turns out at the outset they had planned to 13 use COBRA / TRAC, which is a best estimate tool, for the 14 whole process.
But it got too damned expensive for the 15 small breaks.
So what they did is they said:
Okay.
We 1
16 will use COBRA / TRAC for the large-break LOCA, and we'll 17 use NOTRUMP for the small-break.
Now when you get out to 18 the long-term cooling, they went back to COBRA / TRAC, which 19 in my view is a mistake.
20 The second step in this process is to make 21 appropriate comparisons of the code in the separate 22 effects test data to show that the predictions have l
23 sufficient fidelity.
24 The final step is to make appropriate
(\\_ /
25 comparisons of code predictions in integral test data to 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
13 1
show that the predictions will have sufficient fidelity 7-~g 2
for AP600 evaluation.
In my view, a suite of codes is V
3 needed to carry out such a process and that the use of one 4
code to do it all is an expensive way to do business.
5 But, again, it's Westinghouse's choice.
6 Again, I'd like to emphasize the need for 7
scrutable documentation by referring to the lessons 8
learned from the recent Maine Yankee problems.
The 9
document of most interest is a memo from Dorman, Cubbage 10 and Cotta to Holahan dated 5 December 1996.
And if you 11 look about midway through it, you'll find there are some 12 recommendations made.
13 It's recommended that a guide for
.s 14 documentation of ECCS methodologies be developed; --
15 that's first -- two, RAIs and responses be included in 16 approved versions of vendor topical reports; and, three, 1
17 sample applications of codes and methodology should be 18 submitted for approval.
There were others but not 19 relevant I believe to the AP600 review.
20 These recommendations have been made by the 21 ACRS in the past.
And if you look through letters and 22 ) minutes of meetings for the past 10 or 15 years, the ACRS 23 has been pushing very hard to get the staff to create some
'A sort of a structure for these kinds of reviews.
And it I
-]
s_/
25 never happened.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 2344433
14 1
I believe the ACRS should continue to be rs 2
proactive in this area.
And, further, I think the
(
)
~~j 3
Westinghouse effort should be required to meet these 4
recommendations.
It is probably more important that they 5
do so for plant certification than for a Yankee-type 6
problem.
7 I'd be delighted to answer any questions about 8
our subcommittee meeting.
I only took ten minutes.
9 CHAIRMAN SEALE:
Yes.
Mario, do you have a i
10 question?
11 MEMBER FONTANA:
What do you think it would 12 take to get all of these things, how much time?
13 MEMBER CATTON:
I don't know.
I find it
/' ')
\\
14 really strange that some of these things that appear to me 15 to be relatively simple seem to be impossible for an i
16 organization like Westinghouse.
But I don't know.
17 I think that if you put two or three people in 18 a room and locked them up for a couple of weeks, they 19 could get it done.
It's mainly a matter of organization.
20 I would say 95 percent of the issues that were 21 raised during the subcommittee meeting were answered by 22 reference to something, but nobody quite knew where it 23 was.
In some cases, it hadn't been written yet.
And 24 there was no referencing anywhere.
So what you had was A) k 25 words, a transcript, and an incomplete document.
ss 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 l
1 I don't think how long it would take is really I
r~x 2
important.
I think they should do it when you're about to 3
put on the table a certification for a plant.
4 MEMBER FONTANA:
Your key thing that you said 5
was that you don't think much more technical work is i
6 necessary, it's a case of putting it together.
Is that 7
correct?
8 MEMBER CATTON:
That's right.
I don't think 9
they need to go out and run more experiments.
10 MEMBER FONTANA:
Yes.
11 MEMBER CATTCN:
Now, there were a few 12 technical issues.
One had to do with the passive heat 13 removal system and whether or not there was sufficient I-sT
%J
\\
14 area and whether or not boiling in the pool would impact i
15 it.
I think these questions can be addressed.
16 MEMBER KRESS:
Ivan, you said that you thought 17 a suite of codes was needed.
18 MEMBER CATTON:
Yes.
19 MEMBER KRESS:
If you look at any thermal 20 hydraulic code, that really is a suite of subroutines.
If 21 you have enough of those subroutines or the right kind, 22 why couldn't that be viewed as a suite of codes?
23 MEMBER CATTON:
Well, it certainly could i
24 except that when you do that, you build something that's O
25 so complicated you can never figure it out.
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 l
16 1
MEMBER KRESS:
It becomes inscrutable and --
7-2 MEMBER CATTON:
Yes and becomes very difficult
(_3
)
3 to follow.
There's no reason you can't.
The problem is 4
the way codes like TRAC are put together is they have a 5
particular time-stepping algorithm.
And they have a 6
certain way of doing business with the nodalization.
7 That's what has to change, and that's the heart of the 8
code.
9 If you backed up and started over, you 10 probably could put together a code that could have complex 11 representation when needed and simple when needed, but 12 that ain't the way it was done.
And that's the problem.
13 MEMBER KRESS:
Okay.
O 14 CHAIRMAN SEALE:
Any other comments or 15 questions?
16 VICE CHAIRMAN POWERS:
Ivan, you've talked 17 primarily about thermal hydraulics in the reactor coolant j
18 system.
What can you tell us about thermal hydraulics 19 within the containment?
20 MEMBER CATTON:
That is a little more bleak.
21 At the outset, Westinghouse -- well, for reasons that are 22 unknown to me, it seems that the standard operating 23 procedure is to run tests, then see if they are any good.
24 And that's exactly what was done with the containment.
(-
(m,/
25 They ran some scale tests, and they got some 1
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON. D C. 20005-3701 (202) 234-4433
17 l
1 data.
Then they tried to do calculations.
The f-'s 2
computational tool is GOTHIC.
N 3
MR. BOEHNERT:
Right.
4 MEMBER CATTON:
GOTHIC runs so slowly when 5
they run it in the mode that's needed to represent the 6
environment within the containment that they really 7
couldn't do the calculations.
8 VICE CHAIRMAN POWERS:
When you say "the mode 9
that's needed," what is that mode?
10 MEMBER CATTON:
A finite difference of some 11 sort, lumped parameter, something.
See, the difficulty in i
12 the containment is stratification.
And stratification i
13 impacts the heat and mass transfer to the boundaries of 14 the containment volume.
And unless you properly represent 15 that environment, you can't predict the behavior of 16 stratified flow.
17 MEMBER KRESS:
Is this thermal stratification 18 or --
19 MEMBER CATTON:
Thermal stratification.
And 20 you have also steam stratification.
And if you have any 21 inert gases, they're going to stratify as well because you 22 have a convective process that carries this stuff up.
And 23 the steam condenses out on the wall.
l l
24 The measurements that they did were m
25 inappropriate to measure these kir.ds of things.
So they 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
18 1
really don't know what they had inside.
There were 2
scaling difficulties with the experiments that they ran, g-]s
\\~
3 simple things like:
How much flow should I have in order 4
to represent what I would actually see in a full-scale 5
system?
They just flooded it, and there was no 6
relationship between their flooding and what you might 7
expect.
P MR. BOEHNERT:
There were also some 9
atypicalities in the test facility as well.
10 MEMBER CATTON:
Yes.
That's right.
Some 11 atypicalities you can live with, and others you can't.
If 12 the film evaporation on the external surface is important, 13 then somehow you ought to be scaling that film evaporation
(
\\
\\-
14 process properly.
Either that or you have to have a 15 damned good model that represents the whole thing.
- Well, 16 they have neither.
17 That's my understanding that they're going to 18
-- I guess Jack Kudrick is the NRR person who is 19 responsible for this.
20 MR. BOERNERT:
That's correct.
It's in his 21
- branch, 22 MEMBER CATTON:
Sandia, by the way, through --
23 I forget the person's name -- did a really good job of l
24 evaluating the deta.
I think they're coming to the
[)
(_
25 conclusion that they'd better put sprays or something.
j NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
19 1
I don't recall where they have really come 2
down on that.
We haven't heard in quite a while.
As a n!
!O 3
matter of fact, Dana, it's probably time to ask them.
I 4
guess you have responsibility for containment.
5 VICE CHAIRMAN POWERS:
Not in this area.
I 6
have to recuse myself.
7 MEMBER CATTON:
Somebody ought to ask about 8
what they're doing.
9 CHAIRMAN SEALE:
Okay.
10 MEMBER KRESS:
We'll ask.
11 CHAIRMAN SEALE:
Thank you.
12 MEMBER CATTON:
It's been quite a while.
13 CHAIRMAN SEALE:
Any other questions?
\\
14 (No response.)
15 CHAIRMAN SEALE:
Well, what's our next step 16 now?
17 MEMBER CATTON:
What's your next step?
18 CHAIRMAN SEALE:
Yes.
19 MEMBER CATTON:
I think the next thing is 20 We.. tinghouse is going to come back with -- and I think 21 they have.
We received a new table of contents for the 22 report that addresses Step 1 of the process.
23 And I've taken a brief look at it.
And I 24 notice no mention of the RAIs.
So I told Paul to let them
(
I(
25 know that there needs to be an appendix.
And in 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
20 1
appendix, there should be all the RAIs.
And they need to (N
2 be properly cross-referenced to the text.
And when
)
v 3
they're ready, we go forward.
4 Now, in parallel, what they want to do is they 5
want to start looking at the code validation process, 6
which means you need to take a look at the codes to make 7
sure it properly represents all the important phenomena.
8 That's one of the next steps.
9 I think that's a fairly simple task.
10 MR. BOEHNERT:
Well, the next meeting, 11 actually, they want to meet with us on long-term cooling.
12 MEMBER CATTON:
Long-term cooling is a major 13 headache.
'~'
14 MR. BOEHNERT:
That's why they want to meet 15 with us.
16 MEMBER CATTON:
The problem is they're using 17 COBRA / TRAC.
COBRA / TRAC was a code that was written to 18 evaluate large-break LOCA, deals with very fast 19 transients.
20 And I don't know.
If any of you have done 21 numerical work, you know that you have to use different 22 kinds of algorithms for slow transients than for fast.
23 Well, they're using a code that has the algorithms for 24 fast transients for a slow transient problem.
And they're
(~'s k_
25 having major headaches, and they're coming up with all 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 21 1
kinds of convoluted ways to deal with it.
q 2
The problem you have, or at least that I have,
\\
/
v' 3
is that you can't tell them what to do.
It's their 4
busincas and their money.
And you have to listen to all 5
of this nonsense and hope they get to an answer that mr.kes 6
sense.
They probably can if they spend enough mone) and 7
time.
8 MEMBER KRESS:
The only problem is for running j
9 time; right?
10 MR. BOEHNERT:
They're going to something 11 called a windows approach, where they take --
12 MEMBER CATTON:
There's more to it, Tom, than i
j 13 that.
If you take the codes with these kinds of
{\\
14 algorithms and you start doing some -- you can't do proper 15 convergence testing.
16 So it's not like when I take a nice problem 17 and solve it with a CFD.
It's not the same.
18 MEMBER KRESS:
But usually you don't have a 19 convergence problem with these slow transients long term.
20 MEMBER CATTON:
You shouldn't.
21 MEMBER KRESS:
You could, though, you're 22 saying?
23 MEMBER CATTON:
You could because the 24 time-step is for a different kind of problem.
And so you g"N (j
25 have a tendency to build up errors and all sorts of things NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
22 1
as you move forward.
You don't know if you're going to 2
get true convergence unless you test it.
And how are you 3
going to test it?
4 They have a problem.
And I guess that's why 5
they want to talk to us.
6 MEMBER KRESS:
It's not really a convergence 7
problem.
It's converting to the right value.
8 MEMBER CATTON:
Now, they're arguing that they 9
will and that it's really none of our business if they 10 choose a poor way to do it.
11 MEMBER KRESS:
They might have a point there.
12 MEMBER CATTON:
I agree.
I agree.
But when 13 you choose the wrong method to address a problem, it just O]
\\N-14 makes it a major headache to try to figure out if it's 15 being done right.
So in a way, it is our problem.
16 MEMBER KRESS:
So what are we going to look at 17 out at L.A.
in the --
18 MEMBER CATTON:
Oh, when we go to Los Angeles?
19 MEMBER KRESS:
Yes.
20 MR. BOEHNERT:
That's another issue.
21 MEMBER CATTON:
Next week.
22 MEMBER KRESS:
That's a different issue?
23 MEMBER CATTON:
That's the research effort in 24 the same arena.
[' /
s, 25 MR. BOEHNERT:
What you're going to look at in NEAL R. GROSS COURT REFORTERS AND TRANSCRIBERS 1323 RHCOE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
23 1
L.A. is the RELAP5 code applicability work, where they're p
2 taking RELAP --
N.]
3 MEMBER KRESS:
Oh, that's the RELAP stuff?
4 MR. BOEHNERT:
Yes, RELAP.
5 MEMBER CATTON:
Westinghouse is trying to get 6
a subcommittee meeting scheduled for sometime in March.
7 MR. BOEHNERT:
Right.
I need to talk to you 8
gentlemen to see if I can --
9 CHAIRMAN SEALE:
Okay.
Fine, Paul.
10 All right.
Is there anything else that we 11 need to bring up at this time?
12
(:No response.)
13 CHAIRMAN SEALE:
Okey-dokey.
The next thing
(
)
\\-~'
14 we have is a report by the Subcommittee on I&C Systems and 15 Computers.
This involves the National Academy of Sciences 16 Phase 2 report and also our review of the -- also some of 17 the things on the review plan.
18 I understand you're going to be assigning 19 members' or confirming members' responsibilities for the 20 review of pieces of this thing, Don.
21 MEMBER MILLER:
Well, with the support of the 22 Chairman, yes, we will --
l 23 CHAIRMAN SEALE:
Okay.
Well, we look forward 24 to hearing what you have to say.
O
( _)
25 And I'll remind some of you that you have 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
24 1
date at 9:15 to begin the discussions with one of the rx 2
candidates.
I
\\
O 3
MEMBER MILLER:
And I have an appointment wi '.h 4
one of the commissioners at 9:30.
So I'm not going to be 5
able to make one of these assignments.
6 CHAIRMAN SEALE:
Well, we'll move you around, 7
then.
Okay?
8 MEMBER MILLER:
Okay.
9 CHAIRMAN SEALE:
All right.
Do you want to go i
10 ahead and get started?
j i
11 MEMBER MILLER:
Yes.
12 3)
SUBCOMMITTEE REPORT - SUBCOMMITTEE ON l
13 INSTRUMENTATION AND CONTROL SYSTEMS AND COMPUTERS
/-s\\
\\\\ ']
14 MEMBER MILLER:
As a bit of background, 15 between the meeting in December and the meeting in 16 February, I felt that it would be an opportunity to kind 17 of sit down and reflect upon some of the I&C issues that 18 have come about as we have discussed it in various 19 subcommittees.
20 So I pulled together the memos and the reports 21 and so forth and started a memo which kind of grew like a 22 camel built by a committee and kept evolving.
Finally I i
23 had all of the issues written down, which evolved into 24 that memo you all received dated January 23rd.
[)
x_
25 We also had a PRA Subcommittee meeting, which NEAL R. GROSS j
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433 i
1 25 1
was shortened from two days to one.
So I took the
(~S 2
opportunity of initiating an informal meeting with some k) 3 staff members in the I&C area, both from Research and from 4
NRR.
And that did occur on January 29.
Dana was in town.
5 He joined me, which I appreciated his input and 6
collaboration on that meeting.
7 Now, the issues are as I documented in the 8
memo on January 23rd.
And they involved, as I said, a 9
number of things that various members of this Committee 10 have brought up during our discussions, including items 11 brought up in our meeting with the commissioners in j
12 December.
13 I'm not planning to review that memo here f-
' ~]
\\
14 unless you have questions.
I'd certainly be glad to 15 respond to them.
As you can see, it's just a potpourri of 16 issues that were brought up.
17 In addition to that, I added an issue, or at 18 least an item, that came out of a report that I had the 19 opportunity to review over the December time period.
20 So, with that, we had a meeting on January 21 29th, which I thought was from my perspective productive.
22 And I'm certain Dana would be happy to comment also.
23 In the meantime, as we came to the January 24 29th meeting, the National Academy study or report became r~%
k_
25 available on January 22nd.
And I received it I guess NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
26 1
about that time period as well as the rest of the g-2 committee.:. embers.
And, of course, being the chairman, I
())
l 3
dived in it right away and made some sort of review before 4
the January 29 meeting.
5 It turned out a number of the issues that I 6
put together in my memo kind of naturally moved into the 7
issues that were raised in the National Academy study.
At 8
least some of them did.
9 So, with that, the meeting of January 29th 10 became one of discussing the memo I put together as well 11 as my comments and my view of the National Academy report 12 as well as the staff's view of the report, which many of 13 them also had had the opportunity to at least have a ti i
14 first-cut review.
15 So, with that, we went into the meeting and 16 went through a number of issues and came out of the 17 meeting with some commitments by the staff.
And I'm going 18 to review those commitments as we approach the National i
19 Academy review, which will be presented to us at the March 20 meeting.
21 And also, as heads up, we're going to deal 22 with the final standard review plan update for Chapter 7 23 at the May meeting.
And between that, there will be a 24 subcommittee meeting to deal with a number of these Og (j'
25 issues, including folding the National Academy study into NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 j
i 27 1
any additions or changes or recommendations for the fw 2
standard review plan.
i 1
%J
\\
3 That kind of gives you a background, a kind of 4
an overview of where I believe we are with the I&c.
And 5
I'd be glad to stop now for a moment if there are any 6
questions.
Otherwise, I'll go on and discuss some of the 7
commitments that have come up that came up out of the 8
January 29th meeting.
And I'd also stop and see if Dana 9
has anything to add to this.
10 VICE CHAIRMAN POWERS:
Well, I think, like 11 you, that I got a great deal out of our meeting with the 12 staff, understanding how they were responding to the 13 National Academy report and what they thought were issues
(
i
1
=
14 they could handle.
15 I think it's important to understand that the 16 Academy report is not a carved stone tablet from the 17 mount.
It has its own sets of strengths and weaknesses an 18 difficulties or impossibilities in some cases to guide the 19 staff.
And so it can only be treated as a set of 20 suggestions.
21 MEMBER MILLER:
Dana gave me and the staff 22 some insight on National Academy-type I guest efforts.
l 23 You've served on a couple of those, I assume?
l 24 VICE CHAIRMAN POWERS:
I have.
O(_,)
25 MEMBER MILLER:
I served as a subcommittee NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W, (202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
28 1
member on one, but I'd never been on a final review, where g3 2
you kind of pull everything together.
tv) 3 And, as such, Dana also gave us, then, some 4
insight on, as he just pointed out, the strengths and 5
weaknesses of a National Academy report.
And, as I said, 6
they're not cast in concrete.
So we're not obligated to 7
do anything with them.
Certainly they're recommendations 8
and suggestions.
9 With that, we went through that.
And so I've 10 got a memo I'm putting together, and we'll send it out.
11 Mike and I will send that out sometime in the next day or 12 two.
And I'll just summarize the memo, which summarizes 13 the meeting of January 29th.
14 The first thing we did do is we have at least 15 tentatively scheduled a subcommittee meeting for April 17 16 to 18, which would deal with the National Academy study 17 report as well as the standard review plan.
18 CHAIRMAN SEALE:
When is that subcommittee 19 meeting?
20 MEMBER MILLER:
That would be April 17 through 21 18.
22 CHAIRMAN SEALE:
Okay.
23 MEMBER MILLER:
And here I'm just -- there's a 24 list of I say comments or commitments as this meeting took
(-()
25 place on January 29th.
Dana noted that formal methods, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
29 1
which were discussed in the report, are out there but are f N, 2
not being used.
But he did request the staff to provide a Q
3 discussion regarding this matter and why they are not 4
being used.
And they agreed to do that.
5 The staff agreed to provide a discussion on 6
the use of software development tools and how the proposed 7
guidance would assess the various tools used in licensee 8
submittals.
That issue was one I had raised in my memo.
9 And it's also embedded within the National Academy report.
10 The staff agreed to provide a discussion 11 regarding diversity between their view on what it is and i
12 what it is not as in regards to I&C systems.
And, again, 13 that issue was brought up in the National Academy report
(")
14 as well as comments that I brought up in my memo.
15 The staff agreed to provide a copy of the EPRI 16 topical report on commercial off-the-shelf software.
We 17 discussed the staff safety evaluation report when it's 18 available.
19 I have available in my library the report as 20 of July of
'96.
The report is being updated.
And 21 apparently we do have an updated version.
22 MR. MARKLEY:
It's coming.
23 MEMBER MILLER:
We will have one coming which 24 will be probably pretty close to the final version that
)
(_/
25 the staff agrees to and in a sense endorses through an NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
30 1
SER.
(""
2 Again, that issue of COTS was a major or was U
3 one of the technical issues that the National Academy 4
study addressed.
And they also referred to the EPRI 5
effort in collaboration with the staff.
6 The staff agreed to provide a walk-through 7
presentation to demonstrate acceptance criteria and how 8
they apply throughout the process as in guidance, which is 9
provided with the SRP.
Dana and I both emphasized that 10 the staff should lead us through I guess as best they can 11 in a tutorial or instructional type perspective, that we 12 all need to be educated, so to speak.
13 I had taken the opportunity of opening up the f3; )\\
14 SRP again and saying:
Okay.
If I were a user of this and 15 trying to follow it through, could I indeed see where the l
16 acceptance criteria are provided and baw we test the i
17 product as well as do the process because that's been an 18 issue this Committee has raised on a number of occasions?
19 And in doing that, I prepared more for myself 20 than anybody else kind of a guideline of what I would do.
21 And that guideline document we provided to the staff.
And 22 I'm certain they're going to shoot holes in it and give me 23 some feedback.
24 And I believe that guideline has been provided
/k-)
25 to the Committee.
Right, Mike?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON. D C. 20005-3701 (202) 234-4433 I
31 1
MR. MARKLEY:
Yes.
p 2
MEMBER MILLER:
They have that?
G)
I 3
Again, you have to take that as notes that I 4
made as I went through the process of saying:
If I were a 5
user of the SRP, how would I look at it?
And where would 6
I see process versus product and testing of the product?
7 So if it's a little bit sketchy, the staff is 8
going to fill in the gaps and hopefully make it readable 9
for all of us on the Committee.
10 CHAIRMAN SEALE:
Did you also examine it as a 11 victim of the SRP?
12 MEMBER MILLER:
Victim.
13 CHAIRMAN SEALE:
That is, an applicant that is (v) 14 presamably making a proposal they hope will meet the 15 requirements of the SRP.
16 MEMBER MILLER:
I tried to look at it as a 17 potential user, yes.
18 CHAIRMAN SEALE:
Well, the user is the 19 reviewer.
20 MEMBER MILLER:
Okay.
Oh, I see, from that 21 perspective.
I suppose I was looking at it from both 22 perspectives, yes.
23 CHAIRMAN SEALE:
Okay.
I thought you probably 24 were, but there is that distinction.
G 25 MEMBER MILLER:
Again, I didn't spend as much NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
32 1
time as a reviewer would nor as a potential user would, 7-~s 2
but I tried to trace it through from beginning to end.
3 And it turns out that by looking at a couple 4
of the documents that we have been provided, Chapter 7, 5
OAA, and the BTP-14, really are the key documents.
And 6
when I came out, I felt the acceptance criteria were 7
pretty well laid out, although it's a little more 8
complicated than maybe it could be.
Other than that, I 9
felt it was fairly well laid out.
10 VICE CHAIRMAN POWERS:
I thought one of the 11 most interesting features of our discussions was the 12 central role that the branch technical position plays.
13 And when you first approach this document, you say, "Well,
/~%
i' (\\ ')
14 what's important is the SRP elements."
And you don't 15 realize how central this branch technical position, it 16 seems, a second tier document really is.
j 17 MEMBER MILLER:
Yes.
The BTP-14, as I went 1
18 through this exercise, I gained more appreciation for it.
19 I think the staff did a good job with the BTP-14, but they i
20 probably should have done a better job in emphasizing the 21 importance of that document to us as a member of this 22 Committee.
So we emphasize they should walk us through 23 how they would use this as guidance and emphasize the 24 importance of that document and some of the key issues Gk,)
25 brought up in that document.
s NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W.
(202) 234 4433 WAShlNGTON. D C. 20005-3701 (202) 234-4433
33 1
It is a matter of I had to go through it a l
2 second time to begin to appreciate where that document fit
~
(3) w./
3 into the entire scheme.
As Dana points out, it's almost 4
the primary document to look at.
And the SRP is in a 5
sense supportive.
6 We discussed for those who might worry about 7
consensus standards the fact that OMB Circular A-19 8
basically requires or -- how did we put that, Mike? --
9 almost requires us to use standards when they're 10 available.
11 MR. MARKLEY:
It's a government document that 12 tells agencies to go out and do these things, but it was 13 later promulgated into law.
So it is a requirement.
14 VICE CHAIRMAN POWERS:
Well, what you don't 15 want to do is overestimate the impact of Public Law 16 104-113.
It says, "Use consensus standards when they're 17 available and suitable."
It is not a straitjacket.
It 18 says, " Don't develop a standard simply because you want 19 one in an area when one already exists in the consensus 20 process."
If you have a situation that's unique and 21 different and whatnot, you're not constrained to avoid the 22 standard, the sub-generation of regulations.
23 MEMBER MILLER:
Yes.
24 MR. MARKLEY:
I think the primary purpose of fh
(,)
25 it is really just so that government agencies and offices 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
don't recreate things that already exist and waste 2
government money.
It's more an efficiency in government f-ms b
3 document than anything else.
4 VICE CHAIRMAN POWERS:
You also have to look 5
at the title on the act.
And it is a technologic transfer 6
and advancement act.
It is simply reiterating a truism we 7
all know, that a certain amount of standardization across 8
and between industries facilitates technology transfer and 9
the advancement of technology.
10 Looked upon that way, I think it gives a lot 11 of understanding of what the legislature was trying to 12 accomplish.
13 MEMBER MILLER:
The next item, Dana requested 14 the staff provide a briefing on the Halden reactor project 15 at a future meeting.
And this should be added to the 16 future activities list for this Committee.
17 VICE CHAIRMAN POWERS:
I'll just interject 18 here that I've had the benefit of looking at a trip report 19 from some of the NRC staff who attended the Halden 20 meeting.
And I found -- a very good trip report, by the 21 way -- that there are an awful lot of things going on at 22 the Halden project that addressed topics of interest to 23 this Committee.
And what I asked if the staff could brief 24 us on in this particular area, what was going on at Halden f)%
(_
25 for our information and that we may want to pay closer NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C 20005-3701 (202) 234-4433
35 1
attention to this Halden effort.
1
/~}
2 NRC has increased its involvement in that L/
3 whole Halden effort recently.
And it is an international 4
consortium of researchers that are producing information 5
at least that's pertinent to a lot of the topics that we 6
address, both in the digital I&C and in the fuels area and 7
in the human performance area.
8 In addition, it seems like the gatherings that 9
they have annually seem to attract a lot of allied topics 10 to be addressed at various workshops.
So it looks like as 11 we move toward a more international research effort in 12 reactor safety, that maybe the Halden is an area that we 13 should keep our finger on the pulse a little closer.
,s
)
i 14 CHAIRMAN SEALE:
You'll remember several 15 months ago we had a presentation on human factors that 16 included a fairly large component from some work that had 17 been done in control room behavior from Halden.
We'll 18 have to put that on the future projects list and see what 19 we can get that will --
20 VICE CHAIRMAN POWERS:
Apparently at this last 21 meeting there were also some discussions of risk 22 assessments during low-power and shutdown operations.
23 CHAIRMAN SEALE:
It would be useful.
l l
24 VICE CHAIRMAN POWERS:
So it looks like l f.
(_-
25 there's an awful lot of information being brought together NEAL R GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
36 1
both as part of the research program again but also being 2
brought to the table by the participants in that research g S.
V 3
program.
4 It's clear that a lot of our research is 5
becoming international, rather than self-directed.
We 6
need to take advantage of this evolution.
7 MEMBER MILLER:
It would sound to me like we 8
need almost a routine periodic report of the Halden 9
program.
10 VICE CHAIRMAN POWERS:
Well, I think it's my 11 impression that this program goes out of its way each year 12 to reveal to its participants all that's been discovered, 13 that they have a periodic recording effort and maybe we
,_()
14 want a little more active participation.
15 CHAIRMAN SEALE:
Yes.
16 VICE CHAIRMAN POWERS:
And maybe we want a l
17 little more active participation.
18 CHAIRMAN SEALE:
Get on the circuit.
l l
19 MEMBER MILLER:
How long do those meetings 20 last?
21 VICE CHAIRMAN POWERS:
They're several days 22 long.
23 MEMBER MILLER:
It might be prudent that one 24 of us go and attend one of them.
p(_)
25 In direct regard to the National Academy Phase 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 37 1
2 study, we proposed the following actions.
And the staff f~
2 has agreed to do that.
First of all, the staff and we, j
(
I w
3 Dana and I, agree that many of the points raised in the 4
National Academy study are, in fact, addressed, either 5
directly or indirectly, in the Branch Technical Position 6
14, again raising our appreciation for that document.
And 1
7 it was not clear whether -- it was our assumption the 8
National Academy study did not have the branch technical 9
position available when they did their report.
10 An issue that might be controversial coming 11 out of the National Academy study is the use of PRA and 12 quantitative reliability numbers for software in digital 13 I&C systems, which we as a policy, as NRC policy, are not 14 going to do it this time.
15 The staff agreed to look at that issue and 16 felt they could handle that issue.
And that will be part i
17 of their report to us in the April subcommittee meeting.
18 VICE CHAIRMAN POWERS:
And I felt that the 19 Academy report did an incomplete job of defending that 20 position.
I think when we hear about that Academy report, 21 that really ought to be a topic that we try to understand 22 better how they came to that conclusion.
23 MEMBER MILLER:
I would assume you and I and 24 other members of the Committee and George will spend some
(~-
(,
25 time when they give that report on that issue.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
38 l
1 VICE CHAIRMAN POWERS:
I think it's just i
s 2
important for us to understand because this is a step that
! l
\\
l L.)
3 cuts the Gordian knot if we can take it.
I mean, this is 4
the issue from what a lot of us -- the pressure to have 5
the Academy report was borne.
Our concern that applying 6
probablistic failure criteria were penalizing digital 7
methods unfairly.
i 8
And they're coming along and saying, "No.
You 9
can do that."
In a practical sense they're saying you can 10 do that.
And I'd like to understand very well why that is 11 because it does, as I say, sever the Gordian knot that we 12 found ourselves in.
l 13 CHAIRMAN SEALE:
I should comment that at the 10
-)
\\
14 request of the subcommittee chairman, we have invited 15 former member Hal Lewis to be here at the presentation in 16 March on the National Academy study.
And he has accepted.
17 I think that we can expect that Dr. Lewis will 18 also be interested in that particular aspect of their 19 findings.
20 MEMBER MILLER:
The March subcommittee meeting 21 should be interesting, to say the least.
22 CHAIRMAN SEALE:
Yes.
23 MEMBER MILLER:
That probably from my l
l 24 perspective of a first-cut review of the entire report is l
/~
'(_T) 25 probably potentially the most controversial issue or
[
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 l
1 uncertain issue, recommendation, so to speak, made in the i
2 report.
(n) w/
3 The next item, the staff agreed that for the 4
proposed final SRP Chapter 7 review to forward a document i
5 identifying their points of agreement and disagreement 6
with the conclusions and recommendations that were made in 7
the National Academy Phase 2 study as well as their 8
proposed resolution and recommended actions.
9 During their meeting, we went through the 10 National Academy report kind of issue by issue and looking i
11 at each of the recommendations.
And I formed some 12 opinions of my own at that point as well as staff members.
13 We kind of compared notes.
We tentatively had
/,_\\
\\")
14 I would say a' reasonable correlation on the points we felt 15 needed to be further addressed and further studi d to be 16 taken place on those issues.
17 There were a number of points raised or 18 recommendations that we felt were more or less validating 19 what we were already doing, a number of recommendations 20 that were made that in a sense confirmed that the 21 positions that we're moving into are probably the right 22 way to go, a number that really just were ones that we 23 didn't think we had to deal with, so to speak.
So the 24 National Academy recommendation kind of fell under three p)
(_
25 categories:
potentially controversial; validating; or 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
40 l
1 a sense stating, "Here's the position of the industry, and l
2 you don't need to do anything with it," or a position of n)
\\v l
3 software engineering.
4 And following the lead of the Chairman in 5
trying to facilitate more efficient letter writing, I have 6
made assignments of members of this committee to look at 7
the National Academy study report and look at the 8
particular sections and then provide me some feedback.
So 9
those assignments are made.
10 You're going to pass out that list; right, 11 Mike?
12 CHAIRMAN SEALE:
It has been.
13 MEMBER MILLER:
It's already been passed out?
g!")
14 So if there is any debate on that list, I would appreciate 15 feedback or comment.
16 CIAIRMAN SEALE:
Yes.
There's a sheet here 17 you've 1een given.
And it has 2 sets of lists on it:
one i
18 from the PRA Subcommittee, 5 items; and then Dr. Miller's 19 assignment of 11.
20 I believe everyone is committed to do 21 something here.
22 MEMBER MILLER:
Yes.
This is not limited to 23 the subcommittee.
This is the entire committee.
24 MEMBER KRESS:
These all are the March
,r~)
d 25 meeting?
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
MEMBER MILLER:
Yes.
p 2
CHAIRMAN SEALE:
Okay.
v) 3 MR. DURAISWAMY:
The members can take a look 4
at the whole report, but this is especially assigned.
You 5
are encouraged to look at the whole report.
6 MEMBER PILLER:
Yes, everybody is encouraged 4
7 to look at it.
You know, this is our report.
This report 8
or this study was initiated by this Committee.
So in a 9
sense it's our report.
Certainly the staff is going to 10 provide a lot of support in dealing with the entire 11 report.
But we, the ACRS, initiated it.
We I suppose are 12 obligated to make certain we do a pretty thorough study of 13 it.
f
\\
~
14 And as the chairman of the subcommittee, I'm 15 going to certainly be.
And I have already take a fairly 16 serious look at this report and have drawn a number of 17 conclusions.
And they're still evolving.
By the March 18 meeting we'll have a more thorough study from my 19 perspective.
1 20 So, even though I'm assigned to only one or 21 two parts, I'll take responsibility in a sense for all of 22 it.
But I certainly am hoping the rest of the Committee 23 members will support us on various sections here.
24 CHAIRMAN SEALE:
Well, I think that we owe
\\_-)
25 ourselves, in one way or another, a serious consideration NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W (202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
42 1
of these assignments.
We can't expect to approach these s
2 problems in a unified and task effort way unless we do our
(
%.J 3
share of the task.
And so I would urge everyone to be 4
sure to come prepared to meet your obligations here.
5 MEMBER MILLER:
One thing further.
As far as 6
looking at the March meeting -- and I think Frank Coffman 7
here is in the audience.
He might want to make a comment 8
on who is going to attend the meeting in addition to the 9
chairman of the committee.
10 Frank, do you have any idea on that yet?
11 CHAIRMAN SEALE:
Yes, Frank.
Thank you.
12 Thank you.
13 MEMBER MILLER:
Thanks for coming.
14 MR. COFFMAN:
Yes, sir.
Right now the entire I
15 panel has been invited, but we only know that Larry Damon 16 will be attending and that Nancy Leveson will not, will be 17 unable to attend, in addition to the chairman.
Are there 18 others, Leo, that --
19 CHAIRMAN SEALE:
And that date is again when?
20 MR. COFFMAN:
March 7th.
21 MR. DURAISWAMY:
Well, it's a full Committee 22 meeting.
23 CHAIRMAN SEALE:
Yes, that's during the full 24 Committee.
r
(,)%
25 MR. DURAISWAMY:
Right.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234 4433
43 1
MEMBER MILLER:
And we're going to have what, gs 2
two or three hours on that?
3 MR. DURAISWAMY:
Yes.
4 CHAIRMAN SEALE:
Okay.
And then are there 5
subcommittee meetings subsequent to that?
6 MEMBER MILLER:
The subcommittee meeting for 7
I&C will be in April, April 17th.
8 MR. DURAISWAMY:
That is a look at the 9
proposed and final SRP and the reg guides and the BTPs, 10 those kinds of things.
11 CHAIRMAN SEALE:
Okay.
12 MEMBER MILLER:
Right.
13 MR. DURAISWAMY:
Bob, I think the members' 14 assignments, individual members, it will come at the 15 future activities, but I think I put a specific date, I 16 think February 21st, for the members to give you feedback.
17 So if you want to put together something.
It will come in 18 the future activities.
19 MEMBER MILLER:
Okay.
So not only have we 20 made assignments.
We have made a deadline, drop-dead 21 dates, on these.
22 CHAIRMAN SEALE:
Okay.
1 23 MEMBER MILLER:
February 21st, of course, 24 we're also scheduled to have a subcommittee meeting on s_)
25 PRA.
So many of the members will be here.
l l
NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS i
1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
44 1
MR. DURAISWAMY:
That's why I picked the date.
j3 2
MEMBER MILLER:
I figured you picked that V
3 date.
Thic sa) I can go around and collar people who are 4
here.
So tnose attending that meeting may certainly bring 5
your reports.
6 CHAIRMAN SEALE:
Anything else?
7 MEMBER MILLER:
I believe that's all unless we 8
have comments or questions.
9 CHAIRMAN SEALE:
Well, I had one question.
10 MEMBER MILLER:
Okay.
11 CHAIRMAN SEALE:
You said that this Committee 12 was the originator of the National Academy report.
I 13 certainly would agree that the Committee had a lot to do
,_s 14 15 MEMBER MILLER:
Stimulator.
16 CHAIRMAN SEALE:
-- with the stimulation of 17 the report.
On the other hand, of course, it's Researth 18 that is the official --
19 MEMBER MILLER:
Yes.
20 CHAIRMAN SEALE:
custodian of this.
One of 21 the characteristics of these National Academy reports 22 occasionally is that they seem to attract extreme measures 23 to keep them alive once they've been put together.
And 24 they don't die very easy.
O k_)
25 Do you think this report is going to have a 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
45 1
clean, crisp end of life?
fS 2
MEMBER MILLER:
I'll answer that by saying I N_]
3 don't believe and I think the staff -- one conclusion out 4
of the January 29th meeting was we don't believe there are 5
any of what I call showstoppers.
6 CHAIRMAN IiEALE:
Okay.
7 MEMBER MILLER:
And I think that answers your 8
question.
However, I think the report is a document which 9
can be valuable.
For example, in the human factors area, 10 I felt that section was quite interesting just from the 11 viewpoint of it gave me a lot of education, put together a 12 lot of things that I had not understood.
13 I thought that section, for example, I would 7,
k'~'l 14 go back and read a couple of times.
Other sections were 15 not quite as tutorial, so to speak, but they were 16 interesting reading.
I think it's a way to put together a 17 lot of issues in an area.
l 18 But, as I say, there are no recommendations 19 except the one we already discussed that I think is going 20 to have a life beyond the report.
21 CHAIRMAN SEALE:
But the National Academy 22 activity itself will end at that time?
23 MEMBER MILLER:
Oh, I think so, yes.
I don't 24 see any reason why it won't.
I thought you were saying:
k-)
25 Are there issues --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
46 l
1 CHAIRMAN SEALE:
No, no.
I realize there will
(^g 2
be follow-up internal in Research and other places likely.
()
3 MEMBER MILLER:
I guess I can't fully comment l
4 on that because, as you point out, Research is really the 1
5 caretaker or the administrator of that issue.
But I think 6
Frank and others would assume it's going to be over as of 7
the March meeting.
8 CHAIRMAN SEALE:
Staff, want to make any other 9
comments about the status of the report?
10 MR. COFFMAN:
Just to clarify.
11 CHAIRMAN SEALE:
Yes.
12 MR. COFFMAN:
The contract will close the 13 National Academy activity.
(D 14 CHAIRMAN SEALE:
I understand that.
I
~
15 understand that.
16 MR. COFFMAN:
The staff will be providing the 17 response to the recommendations.
l 18 CHAIRMAN SEALE:
Yes.
19 MR. COFFMAN:
Perhaps the ACRS may want to 20 consider responding to the NAS.
21 CHAIRMAN SEALE:
We'll certainly have some j
22 comments for you.
And likely you may wish to pass those 23 along.
24 MEMBER MILLER:
Would we not write a letter on
(/
25 this report?
NEld. R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
47 1
CHAIRMAN SEALE:
Yes, I think we should.
Q
's j 25 we deal with that?"
We ended up choosing one of 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
l 54 I
1 So there was some sort of a comfort level r
2 there, but that's --
( )w)
\\.
l 3
VICE CHAIRMAN POWERS:
Well, I think it was at l
4 our last meeting that a speaker from AEOD pointed out to l
5 us that at Operational Mode 4, an incident had occurred 6
which set new standards as far as its conditional core 7
meltdown probability was.
And, yet, Operational Mode 4 by 8
your screening analysis came out to be a relatively 9
low-risk mode.
10 Now, with deference to Dr. Kress and his 11 accurate comments about stochastic processes, this causes 12 a great deal of discomfort to me.
13 MR. CUNNINGHAM:
I assume you're talking about
/
N
\\
1 14 Wolf Creek.
15 VICE CHAIRMAN POWERS:
Yes.
16 MR. CUNNINGHAM:
Yes.
Okay.
At any rate, we 17 went through this study.
And we ended up a couple of 18 years ago finishing that part of the work and publishing 19 NUREGs.
This is actually part of what was published in 20 terms of the PRAs in NUREG/CR-6143 and 6144.
Those were 21 the very last pieces that came out in July and October of 22
'95.
23 What I have done is sort of set the stage for 24 what I'll call the initial work in terms of low-power and l rN k
25 shutdown.
I'm going to switch over, and Marty is going to 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
talk a little bit about the rule development that has r~T 2
occurred since then.
And I'm going to come back after YJ' 3
that and talk a little bit more about research that has 4
occurred and may be worthwhile in the future.
5 (Slide) 6 MR. VIRGILIO:
Okay.
I joked a little bit 7
about being up in the Northeast.
The last time the staff 8
met with the ACRS I was leading an inspection team up at 9
Millstone Nuclear Power Plant.
10 That was in May of this past year.
And at 11 that point in time, we briefed the staff on the rule that i
12 we were developing.
And, in response, we got a letter 13 dated June 4th.
And I've summarized in the bullets here
'\\,)
14 some of the points that were included in the letter, where 15 the ACRS agreed with the staff that:
shutdown risk is 16 important and we need to focus on it, that the pr fosed 17 rule would likely lessen risk, that the rule you 18 recognized was based, in part, on engineering judgment.
19 And I will not minimize that as a significant part based 20 on engineering judgment.
And the ACRS at that time 21 suggested that we would get together again after we 22 reconciled public comments.
23 The approach that we're on is to develop a 24 revised rule based on comments that we received the last rs (s_)
25 time we published the rule, go back out for comments, and 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
56 1
then come back to the ACRS.
l 1
l r~s 2
(Slide) i 1
iO)
\\
l 3
MEMBER BARTON:
What's your timing on that?
4 MR. VIRGILIO:
I'll get to the schedule in j
5 just a second.
(
l 6
We responded to the ACRS letter.
We agreed, l
l 7
as I said, that engineering judgment was a significant 8
piece in the development of the rule.
We acknowledged 9
some limitations in the Surry and Grand Gulf studies and 10 offered that the regulatory analysis that was being 11 completed to support the backfit would provide additional 1
l l
12 insights.
We also recognized that a Level 3 PRA would be helpful but not necessary to support the rulemaking.
l 13 c)
I 14 (Slide) i 15 MR. VIRGILIO:
Specifically to answer your l
16 question, we're on a schedule now that I've outlined here 17 to have a completed package developed for rulemaking at 18 the end of March.
i 19 This is a slip in the schedule that we 20 presented to you back in the May time frame.
One of the l
21 things that we recognized back as we were about ready to l
22 publish the rule and reg guide was that there were some 23 areas that needed additional work.
And so we dropped back 24 and re-scoped what our efforts were staffed up with about rm
)
25 a dozen people who were working on the rule pretty much NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASil:NGTON, D C. 20005-3701 (202) 234-4433
57 1
l 1
1 full-time at this point, and at this point myself l
2 included, and put forward a new schedule.
And this is (b
%J
\\
l 3
currently the schedule that we're working on.
i 4
So following EDO concurrence, we would submit 5
the rule to the Commission.
We're currently on their 1
l 6
schedule for a Commission meeting sometime in the August 7
time frame.
I don't think we've pinned down the exact I
8 date yet.
And following that Commission meeting, we would 9
go out for public comment on the rule.
10 Then, of course, you go through the process of l
11 reconciling comments.
One of the milestones at that point l
12 would be to come back to the ACRS and discuss with them 13 how we've reconciled the public comments and then go i
l
\\-
14 forward and issue the rule if we had full support of the 15 Commission and the ACRS.
i 16 (Slide) 1 17 MR. CUNNINGHAM:
Let's turn back now, then, to l
18 the research element of this.
I wanted to mention --
1 l
19 VICE CHAIRMAN POWERS:
Before we advance to I
20 the research activities, Marty, can you give me some l
I 21 understanding of:
Is it the regulatory analysis that l
22 you're going to use to convince yourself that you're not 23 just inflicting a large number of requirements on an 24 already heavily regulated industry with hope that you
[ j)
(
25 actually address risk-significant items and not just NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
58 1
correcting the deficiencies and vulnerabilities that have 2
been observed in the past?
,S L) 3 MR. VIRGILIO:
That's -- sorry.
A little 4
distracted here.
5 VICE CHAIRMAN POWERS:
Sorry about that.
6 MR. VIRGILIO:
An invitation of the Chairman's 7
office.
A pleasant distraction.
8 Let me just -- Mark, can I borrow the 9
projector a second?
10 MR. CUNNINGHAM:
Sure.
11 MR. VIRGILIO:
I think maybe we back up and 12 try to answer the question.
I had a couple of backup 1
l 13 slides.
j
,.s
(
's i
.4 (Slide) 15 MR. VIRGILIO:
What we're looking at is the 16 cost-beneficial safety enhancement.
Clearly the burdens 17 that we imposed have to be justified in a trade-off 18 evaluating the risk and evaluating the cost.
19 There are burdens to be imposed, but I think 20 when we look at the regulatory assessment, it is telling 21 us that some burdens are justified.
Burdens in terms of 22 extending outages?
I don't think so.
And I say I don't 23 think so because the regulatory analysis isn't complete, 24 but we recognize the significant impact.
If you add a day
,,\\
$.s,)
25 or add two days or add a week to an outage, you have to 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 l
59 l
1 have significant risk benefits.
2 It is a cost-justified safety enhancement.
-~
%/
3 We're not talking about adequate protection here.
We're 4
talking about justified costs.
And, in addition to the 5
regulatory analysis, we fall back on engineering judgment.
6 And engineering judgment is coming from, in part, the 7
offense that we're seeing.
8 We recognize we just put together a short 9
list, but it's significant when you step back and think of 10 the events that have occurred over the last several years 11 that have gotten -- they have been recognized for their 12 risk significance, Haddam Neck most recently.
13 I think you all were briefed on that event,
,_I)
\\#
14 where we were really surprised and a little bit 15 disappointed, but, again, it renewed our emphasis on the 16 need for this rule.
And that's --
i 4
17 VICE CHAIRMAN POWERS:
I'd just comment that l
18 it also renewed my understanding of how complicated it 19 must be to formulate a rule in this area.
20 MR. VIRGILIO:
In part, it is.
And, in part, 21 it is because what we're trying to do is formulate a 22 risk-informed, performance-based rule.
And what we're 23 doing is we're looking at both preventing and mitigating 24 events that occur.
p)
( (_
25 From our risk insights, both from the Surry 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 l
l 1
and Grand Gulf studies and from the regulatory assessment r^N 2
we have done, we recognize that there were loss of RHR I
1 D
3 events, there are events that cause loss of level control, 4
there are draindown events, and there are loss of power 5
events that lead to loss of decay heat removal or some of 6
the secondary systems that support that.
7 These are key risk insights that help us focus 8
on where we want to have the additional measures of 9
prevention and protection.
Why we are saying this is and 10 why I fully believe it's a performance-based approach is 11 because what we want to do is outline very objective 12 performance measures for licensees, operators actually in 13 the power plant to focus on to ensure that they have g1 t
l
\\
/
14 adequate decay heat removal, to ensure they have inventory 15 control, that they control pressure and reactivity.
So it 16 is performance-based in that regard.
We'll lay out clear j
17 objectives, and we'll expect licensees to measure those 18 objectives and to adhere to them.
19 We will also ensure that they have reliable 20 means of mitigating events should they occur.
And, again, 21 here's where these risk insights come into play.
We focus 22 on those events that all of our risk studies are telling 23 us are important.
And we'll have reporting requirements.
24 What I want to do is make sure it's clear to n
w-25 everybody what the relationship between this rule and the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005 3?O1 (202) 234-4433
61 1
maintenance rule and also the industry guidelines that tre fx 2
now out there, NUMARC-9106.
So we're not trying to b) 3 duplicate what's out there.
4 In a sense, a lot of what's out there today is 5
what's needed.
In one sense, we're codifying a lot of 6
what's currently being done by the good plants.
7 VICE CHAIRMAN POWERS:
Let me ask you a couple 8
of questions here.
You brought up 9106.
Is that factored 9
in into the risk aesessments that you have done?
10 MR. VIRGILIO:
The way we're doing the 11 regulatory assessment, there is what we call the base 12 case, which is without voluntary actions.
Then we look at 13 industry voluntary actions.
And then we look at the rule i
\\)
14 case.
So we cut it three ways.
15 VICE CHAIRMAN POWERS:
What I'm interested in 16 is:
Your insights, your risk i* sights, the quantitative 17 risk insights that you gained, were those obtained with 18 these voluntary actions in mind or not?
19 MR. VIRGILIO:
Yes and no.
Let me --
20 VICE CHAIRMAN POWERS:
That covers the range 21 of possibilities.
22 MR. VIRGILIO:
It's a complicated question and 23 a complicated answer.
If I look at what we got from 24 research in terms of the insights from Grand Gulf and 25 Surry, we recognize that some of those voluntary actions NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
62 1
have, in fact, been incorporated into the procedures and 2
programs that the licensees have had.
7\\
V 3
And so, even with them incorporated, there 4
were still insights, but we also recognize that in some 5
cases; for example, in Surry, why we get different nunbers 6
when we run our risk assessments and look at what Mark 7
provides from the Surry study and then pick apart and say, 8
"Well, why are these different numbers?
What are the 9
insights that are coming out of this?"
10 You recognize that in some cases, what Surry 11 has done is they have implemented the voluntary actions.
12 And in some areas, they have been able to suppress the 13 risk.
(~h
\\~
14 But we have to back out or step away and 15 realize that not all plants have done this.
Most of the i
16 good plants have.
Most plants have done something like 17 this, but not all plants have taken it to that level that 18 Surry has in that regard.
19 So yes, in that regard, we recognize that the 20 risk insights are telling us the importance of these 21 procedures and programs that prevent events that shutdown.
22 They are telling us the importance of having the 23 mitigative features available.
24 VICE CHAIRMAN POWERS:
And on these risk
(.
()
25 insights that you have, the ones that you have in detail NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
63 1
come from shutdown operations.
Do you feel like they s
2 translate to low-power operations or do you have to f
k 3
extrapolate or interpolate between power operations and 4
shutdown to get these insights or just how do you go about 5
getting it for low-power operations?
6 MR. VIRGILIO:
It's not easy to extrapolate.
7 And where I'm having a little bit of difficulty, quite i
8 truthfully, is in understanding how Wolf Creek happened 9
because that was an event that was sort of outside of 10 where we're studying now.
11 And I think that we have regulatory, good, 12 sound, regulatory, requirements that cover those modes of 13 operation.
And I think the licensee failed to comply with r'#
14 those requirements.
15 MEMBER BARTON:
I would agree.
16 MR. VIRGILIO:
So in a sense we're not really 17 going back and looking at power operations.
We have drawn 18 the line.
We're looking at cases where the plant is shut 19 down, where they're in cold shutdown and refueling as 20 prescribed by the tech specs.
And that's where we're 21 focusing our attention.
22 MEMBER BARTON:
Dana, what I think makes this 23 even more complicated is if you look at plants that have 24 used the NUMARC guidance and have signed up for EPRI, the n(,,
25 ORAM program and stuff, and have pretty well done a risk NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
64 1
assessment of their shutdown and refueling activities.
f-s 2
I think you find that, still following that,
(
)
v 3
the human error factor factored in, a lot of these events 4
happened because people did something that they shouldn't 5
have done.
Even though there was in place a risk 6
assessment with some backup and defense-in-depth and all 7
that was there, somebody still did something they 8
shouldn't have done which caused an event.
9 So that has to complicate the process you're 10 going through with the rule.
11 MR. VIRGILIO:
It makes a purely 12 performance-based approach a little bit more difficult i
13 because we firmly believe at this point in time that you 1
14 have to have features to prevent events from occurring.
15 I don't think it's enough just to monitor and 16 control parameters.
What we have to do is increase the 17 operators' awareness, the licensees' sensitivity, and 18 prevent events from occurring.
j 19 And I agree with you.
A lot of them are 20 driven by operator error.
There are some equipment l
21 failures that lead us into the events, but I personally 22 believe that operator errors are a real significant 23 contributor.
24 MEMBER BARTON:
I agree.
(,/
That's about all I had.
I'm, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
65 1
unfortunately, going to have to leave.
(~
2 VICE CHAIRMAN POWERS:
A higher authority.
\\ss 3
MR. VIRGILIO:
I leave you in good hands.
Tim 4
Collins is at the table.
Warren Lyons is in the audience.
5 There's a lot of good people here.
They could answer any 6
questions that you might have with regard to what we're 7
doing.
Please excuse me.
8 CHAIRMAN SEALE:
And we trust that they will 9
be able to answer the questions that are going to be 10 addressed to you.
11 VICE CHAIRMAN POWERS:
And let me just say 12 that I have appreciated you being here and I've 13 appreciated you taking time out in January to come talk p-
's'-)
14 with me.
15 MR. VIRGILIO:
Thanks a lot.
16 (Slide) 17 MR. CUNNINGHAM:
The discussion we were just 18 having about operator errors and things is a good lead-in 19 to what we have been doing in terms of follow-up research 20 items.
This is in a sense follow-up to these studies.
21 One of the things that we saw fairly early on 22 as we started to do these PRAs was that we looked at 23 operational events that were occurring in the plants.
And 24 one of the concerns we raised was that the types of human (h
\\_)
25 errors that were occurring in these low-power and shutdown NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
66 1
events were different than the types of human errors that es 2
we were seeing in operations.
G 3
And what we did then was to say, "Well, we're 4
going to proceed with the PRAs as best we can, but we're 5
going to follow up with an additional program that's going 6
to try to expand the boundaries of PRA human reliabiljty 7
analysis."
8 VICE CHAIRMAN POWERS:
Mark, you just a set of 9
words that's just intensely interesting to me.
And so I 10 look to you as a well-respected expert in the PRA 11 community to educate me a little bit.
12 MR. CUNNINGHAM:
Okay.
13 VICE CHAIRMAN POWERS:
Is it, in fact, true,
-s 14 which I believe it to be true, that we are better off to 15 do a PRA as best we can than not to do a PRA?
16 MR. CUNNINGHAM:
It's an interesting question.
17 MEMBER BARTON:
Where's George?
18 VICE CHAIRMAN POWERS:
The answer had better 19 be yes.
l l
20 MR. CUNNINGHAM:
I'm just trying to think 21 across the cases.
In a general sense, I would say yes, i
l 22 we're better off.
I'm just trying to think.
Now I'm 23 going to pick an extreme example and try to decide as --
24 VICE CHAIRMAN POWERS:
I'm not asking you for
(~)
, (_)
25 an absolute judgment.
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
67 1
MR. CUNNINGHAM:
Yes.
2 VICE C
'001 POWERS:
Your feeling is that 7 S A
us 3
we're better off, that you gain so much from the 4
discipline of thinking that, even if you can't do things 5
very precisely, the methodology has built within it a 6
forgiving nature for your ignorance or an accommodation 7
for your ignorance?
8 MR. CUNNINGHAM:
It has an accommodation for 9
ignorance.
10 VICE CHAIRMAN POWERS:
So that you gain 11 insight, --
12 MR. CUNNINGHAM:
Yes.
13 VICE CHAIRMAN POWERS:
-- even if it's not 14 perfect?
^'
15 MR. CUNNINGHAM:
Yes.
And, all other things 16 being equal, people have the money to do things and all 17 that stuff, yes, I think in general.
18 CHAIRMAN SEALE:
There's an interesting 19 orthogonal cut in that.
And that is that the IPEs 20 themselves represent, among all of the people that 21 conducted them, a highly variable set of quality control 22 standards for the PRAs that were done for all of the 23 different plants.
And, yet, I think you can generally say 24 that everybody who did one learned something that made (3
,/
25 that exercise worthwhile.
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 68 1
It's a very different look at it, but I think 2
it certainly supports that position.
j p
NJ 3
MR. CUNNINGHAM:
Yes.
I think that's right.
4 I was thinking more of trying to think of an extreme 5
example --
6 CHAIRMAN SEALE:
Sure.
7 MR. CUNNINGHAM:
-- and asked the question for 8
9 VICE CHAIRMAN POWERS:
I'm not interested in 10 mathematical proof.
11 MR. CUNNINGHAM:
No.
12 VICE CHAIRMAN POWERS:
I'm interested in an 13 in-general kind of answer.
')
14 MR. CUNNINGHAM:
In general.
I was thinking 15 of sabotage, and I was trying to say:
Would it be better 16 if you tried to do a PRA on something like sabotage than 17 not to?
That's perhaps a little more --
18 VICE CHAIRMAN POWERS:
Arguable and has been 19 argued.
20 MR. CUNNINGHAM:
Yes.
It's been argued with 21 this Committee and others, yes.
22 At any rate, so we have started a program, l
l 23 started a program a number of years ago to try to broaden 24 the scope of the types of human errors that we build into
(.
(,/
25 PRAs.
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
69 1
VICE CHAIRMAN POWERS:
That may be in the most 2
unfortunate wording, " human errors that" you " build into
,e')
G 3
PRAs."
I would hope you don't build any human errors in 4
5 MR. CUNNINGHAM:
Okay.
Poor choice of words.
6 That we explicitly reflect in the PRAs or something like 7
that.
8 That work isn't finished yet.
We're going 9
through in the spring an initial demonstration of it with 10 the Grand Gulf plant and try to understand how best to 11 build the model, how to construct the model so that it can 12 be practically applied in a PRA.
Then we're going to go 13 back and see how the results differ from expanding the f_
14 scope of the human reliability analysis.
15 A couple of other things underway were to look 16 at the methods that you could use for analyzing the impact 17 of performing preventive maintenance at power versus at 18 shutdown.
i l
19 Obviously there are risk trade-offs that occur l
20 if you decide to maintain a piece of equipment while 21 you're at power versus shutdowns.
And what equipment you 22 maintain during shutdown may be different depending on 23 your risk insights from that.
This work will basically be 24 done about the end of this fiscal year.
(3 x_/
25 The third thing is a feasibility study of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
~
70 1
looking at developing more simple models of low-power and l (~g 2
shutdown operation that could be used in events
' N-]
3 assessment.
4 We talked about the Wolf Creek event and the 5
AEOD st udies of that a little bit ago.
One of the things 6
that made their study more difficult is they did not have 7
a routir.e, if you will, or have something in their office 8
at the time, a model of the risk model of the low-power 9
and shutdown operations.
10 So we're looking at the practicality of being 11 able to do that; in addition, to extend in a sense the 12 traditional accident sequence precursor models that they 13 have for normal operations.
7-(
/
14 CHAIRMAN SEALE:
Are you going to include in
'~'
15 those models or that approach an attempt to look at 16 various stages of shutdown:
hot shutdown, cold shutdown, 17 transitions, and --
18 MR. CUNNINGHAM:
Part of the feasibility study 19 is to look at that, yes.
20 CHAIRMAN SEALE:
Okay.
I know that there's 21 considerable interest in exactly what the implicatienc of 22 those different levels are.
23 MR. CUNNINGHAM:
Yes.
I l
24 CHAIRMAN SEALE:
And that would be incomplete (s
i
)
l
(,)
25 I think unless you did it.
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
71 1
MR. CUNNINGHAM:
Yes.
That's righ' The same f-~)
issue applies for analyzing the impact of maintenance.
2 I
\\v 3
CHAIRMAN SEALE:
Sure.
4 MR. CUNNINGHAM:
And you don't maintenance 5
equipment during mid-loop operation.
There are better 6
times during shutdown to do it and that type of thing.
7 So these are three of the examples of the 8
things that have been going on since the Grand Gulf and 9
Surry studies were finished.
10 (Slide) 11 MR. CUNNINGHAM:
Then, to look a little more 12 in the future, then, the Committee has raised the issue 13 some times before about the need to proceed and do
[_ '\\
\\~'
14 something like an 1150 for a larger number of plants for 15 looking at the full breadth of operations, of shutdown 16 operations and things like that.
It's taking this and 17 going on closer to the scope of the 1150 study, which is 18 five plants, internal events, external events, that type 19 of thing.
20 When we thought about that and were trying to 21 deal with the Committee's concern about this, we had a 22 number of considerations we had talked about as 23 alternatives for dealing with the issue of risks from 24 shutdown operations.
g
(_)
25 One was to pursue specific method NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 2000 5 3701 (202) 234-4433
l 72 1
applications, development applications or development r3 2
activities in areas that we saw were particularly crucial (v) 3 to shutdown operations.
4 The human reliability analysis is one I've 1
5 already talked about a little bit.
Another question 6
that's come up is the need for having fire methods for use 7
in PRA.
And that came up more in the general sense of not 8
just shutdown operations but PRAs in general.
And this 9
Committee has talked about that issue before that we may 10 not be doing as good a job in fire methods as could be.
11 The human reliability work is a good bit 12 underway now.
We are just getting started in the area of i
13 trying to look at how we could better do fire PRA work, if
,r 3
(~)
14 you will.
15 Another alternative is in a sense the 16 alterative that was suggested by the Committee at one 17 time, which is:
Do we go on and do additional 18 demonstration PRAs?
And filling out the 1150 box or 19 matrix, if you will, is one way to do that.
20 A third alternative --
21 VICE CHAIRMAN POWERS:
Not necessarily the 22 only way.
23 MR. CUNNINGHAM:
Yes, not necessarily the only 24 way.
25 VICE CHAIRMAN POWERS:
It has some advantages l
v 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 l
73 1
in that you would give yourself a better description of lq 2
the spectrum for a subset of plants.
'V 3
MR. CUNNINGHAM:
Yes.
4 VICE CHAIRMAN POWERS:
On the other hand, it 5
is by no means obvious that the criteria that were used to 6
select the 1150 plants for power operations is the correct 7
criteria to use for the equivalent work in low-power and 8
shutdown operations.
9 MR. CUNNINGHAM:
That's right.
That's right.
I 10 It's one way to do it, but it's not necessarily the best j
11 way.
i 12 Another option, of course, would be to say, 13 "Well, we feel comfortable enough on how to do shutdown s
l
\\'"}
14 PRAs that we could develop a guidance document to say:
l This is the way to do low-power / shutdown PRAs."
I don't l
15 l
l 16 think we're at that point right now based on our 17 experience and what we have seen with other PRAs.
18 Another alternative, of course, is to say the 19 industry can go out and do -- somebody has raised the 20 question more than once:
Well, is the outgrowth of all of l
21 this IPEs for shutdown operations?
I'm not sure we're at 22 that point either.
I 23 VICE CHAIRMAN POWERS:
It has always struck me 24 that the technology is in a very primitive state A(j 25 relatively.
And the ability to receive an industry IPE at 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 74 1
NRC and then review it and do more than just weigh it is a
(~g 2
little low right now.
I mean, your own understanding is N -]
3 limited.
i 4
MR. CUNNINGHAM:
That's right.
In both cases, j
5 I think with the industry and with the staff, that our 6
understanding is not something where we think we should go j
7 out.
And we don't think it's the right way to do it, just j
8 go out and do a set of IPEs for these.
It's just --
9 CHAIRMAN SEALE:
Don't ask for something i
10 unless you're prepared to receive it.
l l
11 MR. CUNNINGHAM:
Yes, that's one way to look 12 at it.
Yes, sir.
13 VICE CHAIRMAN POWERS:
On the other hand, if rs
(
i 14 our colleague from MIT were here, he would pound the i
15 table, I suspect, and say, "But the industries are doing l
16 shutdown PRAs."
17 MR. CUNNINGHAM:
Yes.
We (1, that was my next 18 point here, is that there are a number of things going on 19 in the industry.
As I understand it, there are shutdown 20 PRAs that have been performed for some of the plants.
l 21 More broadly, someone mentioned earlier there i
22 was work going on, for example, with, I believe it is, the 23 EPRI-sponsored work on ORAM, which is a risk management 24 approach.
I wouldn't necessarily call it a PRA, but it's O(_)
25 a way to manage the outages and things like that.
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
75 1
that figures into the -- that's a factor that comes into l
2 our discussions of what research needs to be done.
! g's)
%-)
3 The last point on this slide is that there's l
4 been a number -- or there's a fair amount of work being 1
5 done via CSNI to look at internationally what is going on 6
in low-power and shutdown safety studies, if you will.
7 It's looking at what improvements are being made to plants l
l 8
internationally, what types of methods are they using to l
l 9
drive them to make safety improvements.
10 There are PRAs that have been done l
11 internationally for shutdown operations, so we're told, 12 and things like that.
There is a --
i 13 MEMBER KRESS:
Do we have a delegate that
(
l
'~'
14 takes part in that study; NRC, I mean?
i l
15 MR. CUNNINGHAM:
Yes, we do.
Joe Murphy has i
16 been the traditional -- this is being done under Principal 17 Working Group 5 of CSNI, which is the PRA group.
Joe has 18 been the delegate for that committee for a number of 19 years.
He's now the Chairman of PWG-5.
And I guess that 20 makes -- I've become sort of the delegate, if you will.
l 21 There have been a couple of studies that CSNI 22 has published in the '93 '94 time frame in this area.
23 They're updating this work now.
And they're supposed to 24
-- we were in a meeting of CSNI in September, I guess it
(~h
(_)
25 was, of last year.
And they were about finished with a f
l 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
1 76 1
draft report on whatever, their update from the 1993-1994 (3
2 time frame.
V 3
I haven't seen that.
We're supposed to be 4
back meeting with them again in another month or two.
And 5
I think that's on the agenda to discuss.
6 VICE CHAIRMAN POWERS:
When you encounter this 7
situation where you have technical altern=tives, you can 8
do this or you can do this, you can spend your money this 9
way, do you do something like a formalized trade study?
l 10 MR. CUNNINGHAM:
We do a trade study, but I I
l 11 wouldn't even come close to calling it formalized, if you l
l 12 will.
It's more of an informal discussion of the pros anl l
l l
13 cons of thint' and that sort of thing.
14 VICE CHAIRMAN POWERS:
So there's not an 15 attempt to do something like a multi-attribute utility 16 analysis?
l 17 MR. CUNNINGHAM:
No, no.
There's not, not in i
18 the traditional sense of those definitions, no.
It's all 1
19 very informal, if you will.
20 VICE CHAIRMAN POWERS:
The Department of 21 Defense has certainly found that to be fairly valuable in 22 its decision-making on the acquisition of systems in 23 making decisions that it is cod _fied within the rules of 24 the Department of Energy that thou shalt do that.
It may O)
(__
25 be honored more in the breach than in fact.
NEAL R. GROSS COUFsT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C 20005-3701 (202) 234-4433
77 l
l 1
Certainly other agencies have found that on l
2 these substantive issues a more formalized decision-making gS 1 %.)
3 process has been valuable because it allows you to put in l
l 4
perspective biases that otherwise get accounted as truth.
l 5
Do you have an opinion?
l 6
MR. CUNNINGHAM:
Personally I would like to 7
see a process of deciding what research needs to be done 8
made more formally, if you will.
I think it would help us 9
focus the research and give us a lot more credibility in 10 the outside world as to how we decide to do one thing or 11 another.
12 We have studied some of these things over the 13 years.
And one of the concerns that seems to be a nagging
(
)
\\#
14 concern is that they are not practical.
I wasn't aware of 15 what DOD has been doing in this area.
Maybe they have 16 come up with a better way to do this than we have seen or 17 I have seen anyway.
18 VICE CHAIRMAN POWERS:
The Department of 19 Defense has an extremely formalized process of identifying 20 mission needs and then having a fairly independent 21 assessment of alternatives for meeting that mission need 22 on fairly rigorous kind of trade studies between various 23 alternatives before they go to the development stages --
24 MR. CUNNINGHAM:
Yes.
O i,/
25 VICE CHAIRMAN POWERS:
-- of a preferred s
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
78 1
alternative.
2 MR. HODGES:
This is Wayne Hodges with the g~g U
3 Office of Research.
4 I will say it's not totally just sitting 5
around and having some informal discussions.
I mean, 6
clearly in what Mark will propose to management needs to 7
be done, he has some trade-offs that he does there.
But 8
there is an office-level discussion.
9 We go through some criteria.
There are 10 relatively few and fairly general things, like:
Should 11 this work be done in the Office of Research versus 12 somewhere else?
Try to rank its importance from a 13 regulatory standpoint and from a safety standpoint, those k-_)
14 types of things.
t 1
1 15 So there are criteria of that type that are 16 applied to it across the office, where it's competing for l
17 funds against projects in other divisions or whatever.
So 18 there is some of that that goes on, but I'm not sure it's f
19 as formalized as you are trying to refer to.
20 VICE CHAIRMAN POWERS:
And, of course, highly 21 formal systems are useful only on big decisions, not i
i l
22 myriads of decisions.
That would just tie you up, and 23 you'd never get anything done.
24 MR. CUNNINGHAM:
Yes.
That's right.
Clearly O(,,)
25 the Defense Department's scale of decisions is much NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
79 1
different than the ones I deal with on a normal basis.
,r')
2 VICE CHAIRMAN POWERS:
You'd be surprised.
t i
L.,/
3 MR. CUNNINGHAM:
I'd like to think that.
4 (Slide) 5 MR. CUNNINGHAM:
At any rate, in a sense that 6
kind of gets us into my last slide here.
Some of the 7
considerations we've brought into play as to looking at 8
the alternative of expanding the 1150 matrix, if you will, 9
to go out into a number of plants and do the full 10 low-power and shutdown operations, internal and external, 11 for all of those plants.
12 I guess there was -- obviously the cost of 13 that is considerable.
These PRAs were quite expensive.
I__sT 14 They were expensive by the budgets I had five years ago, 15 and they would be --
16 CHAIRMAN SEALE:
Humongous.
17 MR. CUNNINGHAM:
They would be humongous or 18 overwhelming by the budget that I have today probably.
19 VICE CHAIRMAN POWERS:
I'm not going to pin 20 you down on budget.
I'm going to ask you to nod.
I was 21 asked to estimate what I thought it would cost.
My guess 22 would be to do your completion of the 1150 PRA matrixes 23 about $13 million.
Am I wildly high, low, or about right?
24 MR. CUNNINGHAM:
You're about right, yes.
(D (m,/
25 Part of 1150 was that we did it in a sense 2 or 3 times NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. O C. 20005-3701 (202) 234-4433
80 1
because of the considerable external review.
And that's
(~T 2
an element in what 1150 cost, but you're about right in
\\
]
v 3
the --
4 CHAIRMAN SEALE:
There was a steep learning 5
ramp on 1150.
6 MR. CUNNINGHAM:
Yes, there was.
Yes.
7 VICE CHAIRMAN POWERS:
And I think there would 8
be a steep learning ramp in the expanding to --
9 MR. CUNNINGHAM:
Yes, that's right.
That's j
l 10 right.
11 VICE CHAIRMAN POWERS:
I would not want to 1
12 underestimate the learning curve here.
13 MR. CUNNINGHAM:
Yes, that's right.
So $13 7
(
/
14 million in terms of my budget today, that's basically 15 about 2 years worth of my budget.
So it's a strong factor 16 in the consideration.
17 VICE CHAIRMAN POWERS:
In that regard, time 18 and money, there's a trade-off.
19 MR. CUNNINGHAM:
Yes.
20 VICE CHAIRMAN POWERS:
Is there in thinking 21 about this -- NUREG-1150, there was a great deal of 22 pressure to get it done in a finite period of time.
When 23 you think about an expansion of the matrix, were you 24 thinking of a similarly intense activity or is it equally (D
(_)
25 valuable to do a more drawn-out activity?
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 81 1
MR. CUNNINGHAM:
I suppose I had in mind when i
<~s 2
we're thinking about more of a drawn out activity where 3
you could probably save some money and things, but it 4
would be a three or four-year process or something like 5
that.
6 It's probably what was sitting in the back of 7
my mind when I was thinking about this.
8 MR. HODGES:
When Mark says that's two years 9
of his budget, I mean, that's all of his budget.
10 VICE CHAIRMAN POWERS:
I understand what he 11 was saying.
l 12 MR. HODGES:
There's nothing else but that if 13 he does that.
'-]
14 VICE CHAIRMAN POWERS:
I understand what he l
15 was saying, yes.
16 MR. CUNNINGHAM:
That's right.
17 MEMBER APOSTOLAKIS:
Now, in 1150, a 18 significant item, expensive item, was the expert opinion 19 elicitation process, wasn't it, and figuring out what to 20 do?
21 MR. CUNNINGHAM:
That was a significant 22 contributor to it, yes.
It wasn't by far the biggest item 23 or anything like that.
24 MEMBER APOSTOLAKIS:
No.
But would you have m
i
(_
25 to repeat that here, you think?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
82 1
VICE CHAIRMAN POWERS:
No question in my mind.
fs 2
Now, could they do it more efficiently, and did they learn k.)
3 something from how to do that?
Absolutely.
4 MR. CUNNINGHAM:
Yes.
5 VICE CHAIRMAN POWERS:
I am certain they would 6
do it more efficiently.
But I'm willing to bet that 7
enough substantive technical issues would come up that you 8
wouldn't have to get an expert elicitation.
And, in fact, 9
I think that's one of the great values of 1150 that it 10 gives it an imprimatur that's hard to ignore.
11 MR. CUNNINGHAM:
Yes.
In terms of the cost of 12 expert elicitation in 1150, I would probably say the 13 biggest cost associated with that was the fact that we did p)
!\\_,
14 it two different ways.
15 We did it one way and had a lot of costs 16 associated with building up that information base.
And 17 then we basically said in peer review that it wasn't l
l 18 adequate.
And so we had to go back and start enew.
19 That part of it we could avoid now.
I think 20 we have a much better feel for how to do quality expert 21 elicitations now.
Still, it's not cheap.
But I tend to l
22 agree with Dr. Powers that there are some of these issues 23 that are going to almost drive you to have to deal with 24 expert elicitation because of the complexity, among other (3
'q )
25 things, and the lack of information in a number of areas.
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 83 l
1 MEMBER APOSTOLAKIS:
But would the number of gs 2
issues be the same?
I mean, at that time you had to
\\
)
1
\\
x./
l 3
formulate issues, decide what to do and all of that.
- Now, 4
with all of that experience, it shouldn't really be that 5
much.
6 MR. CUNNINGHAM:
It's hard for me to say 7
w;.her it would be dramatically different or not.
8 VICE CHAIRMAN POWERS:
I think that if I l
9 follow my analogy for the DOD system, that any attempts to 10 shave costs in planning effort along that direction would 11 be rejected soundly.
12 Your historical database is this.
You just 13 have to use that as your planning basis in thinking about:
\\',)
14 What does it take for time and costs on this study?
15 MR. CUNNINGHAM:
I was going to say one of the 16 other concerns that I guess plays into it in terms of the 17 cost of it is that the way we performed the PRAs for these 18 PRAs was a very detailed, extensive study going down into 19 a great deal of detail in the plant operations and design.
20 Whether that was necessary now in hindsight is in my mind 21 a fair question.
22 Clearly PRAs are being done on low-power and 23 shutdown operations by a number of utilities 24 internationally.
And I'm quite sure they're not at the
/
\\
(_)
25 level that we're talking about here.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234 4433
84 1
Where the right level is I'm not quite sure.
,m 2
VICE CHAIRMAN POWERS:
And I think that's a s
(
1
'v' 3
very difficult question because I worry that too high of a 4
level, you start missing these things that are all of our 5
problems in shutdown.
6 MR. CUNNINGHAM:
Yes, that's right.
7 VICE CHAIRMAN POWERS:
And those are very 8
detailed sorts of things.
9 MR. CUNNINGHAM:
That's right.
That's right.
10 So that's a tough one.
11 MEMBER KRESS:
You were pretty strong about 12 saying you would need the expert elicitation judgment 13 process.
I understand that's needed to get uncertainties
,,_ \\
1 14 and you need uncertainties to have a distribution and 15 actually get a mean value.
But do you think a pure PRA 16 that does the Level 2 only with just using the codes to 17 calculate the core damage frequency and the containment 18 failure probability would not serve a very useful purpose i
19 in this case?
And wouldn't it be much, much less i
20 expensive than going through the full uncertainty 21 analysis?
22 VICE CHAIRMAN POWERS:
Earlier I asked Dr.
23 Cunningham if a PRA done as well as you could with 24 existing technology wasn't better than no PRA at all.
And im 25 he said that with certain limitations on the extremes, 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
85 1
that yes, that would be the case and that you would gain r~
2 insights frcm there.
' ]y And since he's far more expert in the area of 3
4 PRA than I, I'm going to defer to his judgment in that 5
area and say that, indeed, a crudely done PRA is better 6
than no PRA at all.
7 And that's what you're saying here, I think, 8
is that you do as well as you can with the tools you have.
9 Doesn't that give you insights that you probably don't 10 have right now?
11 MEMBER KRESS:
Or can't I get enough 12 information to make risk-informed decisions with that?
13 VICE CHAIRMAN POWERS:
And I think the answer
.s
[
h 14 is probably.
I also think that in this area, that during 15 shutdown operations, that you've got to be relatively 16 careful about what you declare as success criteria and 17 that they're different than what you have used in power 18 operations.
19 And then when you go to the Level 2 analysis, 20 it's different than what we have.
Now, can you use what 21 we have now and still derive something useful, 22 understanding that there are probably the physics and 23 chemistry you're not modeling?
24 I guess that's a decision that you have to im 25 make out of this.
I welcome Mark's judgment on these NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
86 1
things.
7m 2
MR. CUNNINGHAM:
As I said, if I were going to V) 3 go off today and -- well, this is another way of saying 4
it.
If I were going to go off today and start some PRAs 5
on low-power and shutdown operations, what I have in my 6
mind is that the way we did it for Grand Gulf and Surry 7
was probably too detailed.
How much we can back off from 8
that, it's difficult.
9 As I say, the industry has PRAs.
We have the 10 events assessment work that I talked about.
That's 4
11 ancther end of the spectrum.
The question for us there I
12 is:
How much is good enough to be able to be done in 13 events assessment?
That's an issue we're still wrestling O) 14 with.
15 VICE CHAIRMAN POWERS:
Before you undertook 16 1150, you had sponsored a lot of PRAs.
17 MR. CUNNINGHAM:
Yes.
18 VICE CHAIRMAN POWERS:
And those PRAs in their 19 earliest days got lots of scrutiny on the Level 1 aspects 20 of it so that you had a good idea of how detailed to do it 21 for power operations.
By the time you got to 1150, --
22 MR. CUNNINGHAM:
Yes, that's right.
23 VICE CHAIRMAN POWERS:
-- you knew how far to 24 go.
You don't really have that, I think, for shutdown.
(,1 25 So if you were going to this mythical undertaking that you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
87 1
imagine, is the situation where you'd have to walk the 2
learning curve on the Level 1 and you said, "I'd do it
,y i
)
v 3
less detailed"?
Would you go through and peer review that 4
proposal and kind of take baby steps through this 5
mirroring what you had to do from about 1974 until 1150s 6
in the late '80s or something like that?
7 MR. CUNNINGHAM:
That's right.
It seems just 8
like yesterday, but I think that's right.
And that's one 9
of the things that --
10 VICE CHAIRMAN POWERS:
Longest 100 years of 11 your life.
12 MR. CUNNINGHAM:
That's right.
13 CHAIRMAN SEALE:
Time flies fast when you're
/~ \\
r i
14 having fun.
15 MR. CUNNINGHAM:
Yes, that's right.
16 MEMBER KRESS:
Mark, on this question of costs 17 and what needs to be done, if you were to only do Level 1 18 PRA to get a core damage frequency and with uncertainties 19 in that; whereas, the uncertainties were only propagated 20 through the model and from distributions on the input so 21 that you might use some sort of expert opinion to get 22 those but you might already have those, would that 23 substantially change your estimate of how much this would 24 cost?
It seems to me like they would be much cheaper.
O V
25 MR. CUNNINGHAM:
It would be -- I don't know 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 88 l
1
-- a good bit cheaper, yes.
l 2
MEMBER KRESS:
And if one could say something
%_/
3 like, instead of a core damage frequency of 10'*, we're i
4 really worried under these circumstances about core damage 5
f requencies c f 104 Then you'd have something very 6
useful, it seems to me.
7 MR. CUNNINGHAM:
Okay.
Yes, I guess so.
The 8
right level of detail, or even the Level 1 part of it, 9
it's -- I'm uncomfortable with the level of detail that we 10 did here.
It's just I think it was overkill.
11 One of the things that's clear is that we 12 haven't thought much about that issue of:
Well, what is 13 the right level of detail since the completion of these 1
,s
l 14 studies?
We have been focusing on other aspects of the 15 technical issue, notwithstanding that we have a rule going 16 forward separately.
17 The technical issues that we tended to focus 18 on were more -- human reliability assessment was clearly 19 the big item, if you will, because that was something we
]
20
-- I can't remember if you were in the room or not.
We 21 talked about that.
We saw it early in the low-power and 22 shutdown studies, where we were just Jeaving out important 23 human errors.
l 24 MEMBER KRESS:
I wasn't here for that.
(
(,,/
25 MR. CUNNINGHAM:
Okay.
And that was --
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 89 1
VICE CHAIRMAN POWERS:
Mark, let me ask you a 2
question.
When you think about "Let's just do Level 1 on g-~3 b) 3 this problem" and then we think about the insights that 4
you've got out of what you've done so far, isn't one of j
l 5
those insights that the risk relative to risk at power 6
operations is not well-reflected to the core damage 7
frequency?
8 MR. CUNNINGHAM:
I think that's correct.
I 9
think one of the important insights coming out of tb n'
10 studies was the very different containment performance in 11 shutdown operations relative to full-power operations.
12 VICE CHAIRMAN POWERS:
And, in fact --
13 MEMBER KRESS:
That's why I suggested backing
' ')
i 14 off on the acceptable CDF that you deal with to a lower 15 level because of that insight.
16 VICE CHAIRMAN POWERS:
You're coming to a 17 conclusion because if you just look at CDF, it reflects j
18 poorly the actual risk when you compare it relative to 19 power operations.
20 MR. CUNh ~NGHAM:
That's right.
Even i the 21 Phase 1 studies, one of the criteria we had for deciding 22 what were the important operating states was containment 23 performance right from the beginning.
24 It was not just core damage frequency.
Core b
q,/
25 damage frequency and containment performance was the key NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
90 1
issue because in some parts of shutdown operations, you r-'g 2
have basically an intact containment.
'\\)
3 In other aspects, you have containment that 4
can be open.
You can have the suppression pool drained in 5
boilers and things like that.
And that plays a key role 6
in the decision.
7 VICE CHAIRMAN POWERS:
I think what you're 8
telling me is that if I have an IPE on power operations 9
that gives me a core damage frequency and I lack anything 10 for the low-power and shutdown operations, that there is 11 not a facile way to multiply my CDF by some number and add 12 it to that CDF to give me a total of core damage 13 frequency.
7_
('-~)
14 MR. CUNNINGHAM:
I think in the context of the 15 regulatory guides that we have been here talking to you 16 about, we certainly have kicked that idea around.
As I 17 said, it would be nice if we could do that because it 18 would give us a much clearer definition of acceptance 19 guidelines and things like that.
It's not a 20 straightforward ratio, if you will, or something like 21 that.
22 MEMBER KRESS:
I think it is on core damage 23 frequency.
The problem is if you're trying to talk about 24 an acceptable core damage frequency along with an f3 i
(/
25 acceptable containment failure probability, there's where 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
l
91 1
you have the difficulty.
I mean, you can --
2 VICE CHAIRMAN POWERS:
I left out a sentence.
f-g 3
MEMBER KRESS:
Yes.
Okay.
4 VICE CHAIRMAN POWERS:
It's that acceptable 5
core damage frequency.
6 Let me formulate one.
Simply by multiplying 7
here, I've got to have red and black CDFs here, two 8
different colors.
9 MR. CUNNINGHAM:
Again, I might be more 10 comfortable with the PWR and some sort of a ratio because 11 of the prominence of mid-loop operations as a sore thumb, 12 if you will, in the shutdown operations.
With boilers, 13 it's a little more difficult to see something like that.
(~>\\
14 VICE CHAIRMAN POWER.S:
One always worries that l
15 it may be that it's these two particular plants.
In the 16 broader spectrum, you might not be this concerned.
17 MR. CUNNINGHAM:
That's correct.
That was one 18 of the concerns from the beginning, I guess more so with 19 Grand Gulf.
Grand Gulf was a BWR-6.
It had some 20 additional features beyond the normal, if you will, BWR l
21 and things.
That was always a concern of how that was l
22 going to -- that might buy us the results that we had.
23 MEMBER KRESS:
See, the reason I'm arguing the 24 way I'm going, Dana, is that I see this slide up here,
/^
(,)/
25 which to me almost rules out doing really good Level 3 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
92 1
elicitation process of the NUREG-1150 type.
But maybe we 2
should think a little more about acceptance criteria on g-ss t
)
v 3
core damage frequency and back off to a Level 1 on this 4
issue of shutdown risk and --
5 VICE CHAIRMAN POWERS:
We haven't allowed the 6
speaker yet to advance from the first or the second line 7
of his slide to his conclusion.
8 MEMBER KRESS:
Okay.
That's true.
I've 9
already read it.
10 VICE CHAIRMAN POWERS:
I'm dying to hear how 11 he reached this conclusion.
12 MEMBER APOSTOLAKIS:
I have a question, 13 though, on the first.
We hear that the industry, both
(\\
k-14 here and abroad, has done detailed low-power and shutdown 15 risk assessments.
Do you know how much of this matrix, 16 how many entries of this matrix, can be filled by the 17 results of these PRAs so you don't have to reproduce them?
18 MR. CUNNINGHAM:
Let's back up a second.
The 19 matrix that we're talking about here was one way to 20 address the question of shutdown risk.
That was an issue 21 that was raised by the Committee at one time, which was in 22 a sense to stylize the answer expand 1150 to cover the --
l 23 MEMBER APOSTOLAKIS:
Right.
24 MR. CUNNINGHAM:
Okay.
That's one way to deal O(_)
25 with that.
My point in the slide was that I'm not NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTGN. D C. 20005-3701 (202) 234-4433
93 1
comfortable with that for a couple of reasons.
With that 2
specific definition of how to consider low-power and LJ 3
shutdown risks, on that redo, if you take it in the j
4 context of redoing 1150 or expanding 1150 is very costly 5
keeping with the ground rules, if you will, of 1150.
6 A technical justification aspect of it is I'm 7
uncomfortable with that today because of the work that we 8
have going on in human reliability analysis that I think 9
is an important element of any detailed PRA that you would 10 want to do in shutdown operations.
So that's the middle 11 part of this.
12 I'm trying to work through this slide without 13 I
\\
V 14 MEMBER APOSTOLAKIS:
No.
But what I'm saying 15 is that you did limited-scope studies.
16 MR. CUNNINGHAM:
Yes.
17 MEMBER APOSTOLAKIS:
Others probably have gone 18 beyond that.
19 MR. CUNNINGHAM:
Yes.
20 MEMBER APOSTOLAKIS:
Have you collected all 21 this information that others have generated to see how 22 much more we know now?
23 MR. CUNNINGHAM:
That's really I get to my 24 bottom line here.
p(,)
25 MEMBER APOSTOLAKIS:
Okay.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, O C. 20005-3701 (202) 234-4433
94 1
MR. CUNNINGHAM:
Something we have talked r3 2
about is in a general sense of going back and looking at (v) l 3
the other PRAs that have been done.
The specific answer 4
to your question is no, we haven't.
We haven't not gone 5
back to do that.
6 The last bullet was intended to be something 7
to say maybe that's a good idea to do, that it's to go 8
back and look at the PRAs that have been done, to look at 1
9 the international experience, the CSNI work that I had 10 mentioned, to look at how risk is being factored into l
11 outage management by the utilities, to look at the 12 operational events that have occurred since these were 13 done and kind of come back and do a reassessment and say:
\\ ')
\\
14 Is there specific research that now is merited in the area i
assment?
15 of low-power and shutdown risk l
16 In a more general sense, what I was going to 17 suggest is that we take the task on of doing that, of 18 trying to compile this information, either via collecting j
19 reports or workshops or a subcommittee meeting or 20 something like that, and come back and lay it out on the j
l 21 table as to what's out there and have the industry -- you 22 know, EPRI has been very involved in this -- and have them l
23 come in and talk about it and use that as a basis to say:
24 Is there something else that's missing?
Is there a piece A)
(_/
25 of this risk puzzle for shutdown operations that's missing 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
95 1
that merits research by the Office of Research?
2 MEMBER APOSTOLAKIS:
I think that's a r')
QJ 3
worthwhile approach.
4 MEMBER BARTON:
Good approach.
5 CHAIRMAN SEALE:
Yes.
6 MEMBER SHACK:
It sort of boggles my mind, 1
t nougil, that you're ready to have a rule on this at the 8
end of March when you haven't done this yet.
9 MR. CUNNINGHAM:
Well, the rule is trying to 10 deal with some specific concerns that have been raised and 11 that deal with containment performance, that deal with j
l 12 mid-loop operations and things like that, 13 The question is:
Does the staff feel like it 7,
(s)
14 has to wait for this type of information before it 15 proceeds with that rule?
And I think the staff does not 16 want to wait.
17 I think the safety issues out there for 18 mid-loop operations, for example, and things are 19 considered to be serious issues and we want to proceed 20 with it.
I think we have enough information to proceed 21 with the elements of the rule as they are.
22 CHAIRMAN SEALE:
Could I ask a question about 23 some of the characteristics of that rule?
I'm very 24 intrigued.
We've seen initiatives on the part of the
(_/
25 utilities to apply risk assessment methodologies in a wide NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
96 1
variety of areas.
2 And we have here a situation where we say:
fx 3
Well, part of our difficulty is that we can't afford the 4
time line that would allow us to apply that risk 5
assessment to these problems but that clearly risk 6
assessment could be very helpful in resolving or in 7
prioritizing these issues and putting them in their proper 8
perspective.
9 Do you have in the rule any accommodation for 10 or acceptance of a risk assessment initiative by a utility 11 to address these issues, as opposed to the prescriptive 12 rule that you're talking about?
13 MR. CUNNINGHAM:
I'm going to defer that to
,._h i\\\\ '/
14 Mr. Collins.
Since Mr. Virgilio had to leave, Tim will --
15 MR. COLLINS:
This is Tim Collins from DSSA.
16 There is nothing in the rules specifically to 17 take into account an industry-initiated risk assessment, 18 if you will.
19 But I would disagree somewhat with your 20 characterization of the rule as being prescriptive.
We're 21 deliberately trying to make the rule a 22 performance-oriented rule, which takes us away from the 23 prescriptive aspects.
24 I think what we're going to come up with is a
(_)
25 little bit of a hybrid:
a little bit of prer.cription 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
l 97 1
a little bit of performance elements.
But I believe that 2
we're really trying to minimize the prescriptive aspects i
,f S
\\-)
3 of the rule.
l 4
CHAIRMAN SEALE:
But the idea of an 5
alternative approach by one or two or whatever number of 6
people who might want to come in and do a risk assessment 7
to prioritize the way in which they attack these things, 8
there's no accommodation for that.
i 9
MR. COLLINS:
Well, no.
I think the rule 11 n
10 aould accommodate that, in fact.
Like I say, there may be 11 some elements which are prescriptive.
Now, that would not 12 accommodate somebody's risk assessment as we see it now.
13 But in general the performance aspects of the rule would t
\\-
14 accommodate that the way we have it written.
I mean, 15 we're trying to hit it that way, at any rate.
16 CHAIRMAN SEALE:
We have this whole problem 17 of:
How do you bring risk into the game as a substitute 18 for what is already there, as opposed to an add-on to 19 what's there?
And this is kind of a case where you maybe 20 can address that question up front, rather than trying to 21 figure out what the Band-Aid looks like on the back end to 22 modify it.
23 MR. COLLINS:
See, one other thing is as far l
24 as what's already theTe in regulatory space, there's not a l
(~%
l l (_)
25 whole lot that's --
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
98 1
CHAIRMAN SEALE:
But in other areas, there are
,ex 2
things that are there.
I
\\
V 3
MR. COLLINS:
Well, I don't know what you mean 4
by "in other areas."
5 CHAIRMAN SEALE:
I'm saying at the present 6
time when we look at other technical problems, we ask 7
ourselves:
How do we bring the risk assessment to bear on 8
that problem?
9 It seems we always wind up adding additional Ichings, rather than substituting for what's in the present 10 t
11 regulatory requirement.
12 MR. COLLINS:
I guess I'm not really following 13 what you're saying.
! ()
14 MEMBER APOSTOLAKIS:
Does the PRA add l
15 requirements or not?
Is that really what it comes down 16 to?
17 CHAIRMAN SEALE:
Yes.
That's right, the idea 18 that --
l 19 MEMBER APOSTOLAKIS:
Is the PRA used just to l
l 20 add requirements?
21 MR. CUNNINGHAM:
No.
22 CHAIRMAN SEALE:
That's what I'm saying, see.
23 MR. LYONS:
Warren Lyons.
24 In the process of the entire development of A())
25 the rule, we started back in essentially November of '94 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
99 1
to develop a risk assessment.
Of course, in many areas we r
2 didn't have the data that we wanted.
And in that sense it (s
3 was not as quantitative as we would have liked.
4 And, as Marty mentioned earlier, we are 5
continuing to modify the regulatory analysis, the PRA that 6
we have done, to come up with better insights and perform 7
additional sensitivity studies.
8 With this little bit of an introduction, we 9
have found the PRA work that we have done to be highly 10 valuable as we went ahead and developed what we believed 11 was necessary.
12 There have been several instances in which we 13 have seen insights that, in hindsight, if you will, were 14 obvious.
But without the PRA in my judgment, we would 15 have missed them.
16 For example, when we looked at the results of 17 our PRA and our release calculations, we found the PWR 18 containments were not as effective as from an engineering 19 standpoint we thought they would be.
So we went back and 20 we asked why.
Both we and the industry were basing 21 containment closure operations on loss of RHR.
22 Well, the contributor that this clearly 23 flagged for us was loss of inventory would be taking 24 people by surprise and resulting in an inability to get n(_,)
25 the containment closed.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
100 l
i l
1 We consequently modified our approach in the 2
rulemaking activity to take this into account and doubled 3
the -- I'm sorry -- cut in half the failure to get the 4
containment closed.
5 So we are using the PRA techniques.
We are 6
finding they are valuable in combination with our 7
engineering judgments as we have gone forward in putting 8
this together, both the rulemaking itself and the guidance 9
in the implementation.
10 CHAIRMAN SEALE:
Well, in a sense I guess 11 that's fine, but I'll het you there are some insights that i
12 a utility might be able to bring to the table, too, if l
13 they sat down and looked at the problem.
('_s.\\
V 14 And what do you do with those insights?
Is
]
15 there a way ju which they can be brought to that utility's 16 approach to satisfying the requirements?
17 MR. LYONS:
I would agree that utility 18 insights are also very valuable.
And as those of you have 19 been associated with us in the long term know, we have 20 gone out to the industry on a number of occasions.
We 21 have, in effect, crawled through the plants during 22 shutdowns and observed the way they were doing things.
We j
i 23 have shared information.
We have obtained a number of 24 highly valuable feedbacks.
l' ~)
\\_-
25 We do have some awareness of the procedures s
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 101 1
that they use and how they apply PRA in the conduct of 2
their outages and, in effect, how those provide the kinds 7-~g U
3 of feedback that are very similar to the example that I 4
just provided during our rulemaking activity.
5 So we're not doing this in a vacuum, if you 6
will.
7 VICE CHAIRMAN POWERS:
Mark, you've skipped 8
this wonderfully provocative conclusion that says 9
" Completing the matrix is not technically justified."
If 10 it's not technical 3y justified, it surely is not 11 cost-effective.
12 Can you summarize again the thinking?
13 MR. CUNNINGHAM:
The aspect of it, the 14 technical justification of proceeding with, again, 15 defining a fairly stylized approach to this, which is 16 redoing 1150 or expanding 1150, che technical 17 justification concern I have is that we shouldn't proceed 18 with that while we're trying to better model the human 19 performance in these accidents.
20 VICE CHAIRMAN POWERS:
What you're saying is 21 that you have to do a lot of the homework that was done 22 for operational PRAs over 10 or 15 years.
It just hasn't 23 been done yet.
You're just not --
24 MR. CUNNINGHAM:
Yes.
l p/
(_,
25 VICE CHAIRMAN POWERS:
-- in a technical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-443?
102 1
position to do the job.
2 MR. CUNNINGHAM:
Yes.
That's what I mean 73
\\_
3 here.
4 VICE CHAIRMAN POWERS:
So that a plant which 5
said, "My ultimate goal is to, for instance, do 6
NUREG-1150" -- I don't want to make that too precise 7
because it doesn't have to be a carbon copy of 1150 --
8 would be preceded by a variety of steps that still need to 9
be carried out.
10 MR. CUNNINGHAM:
Exactly, exactly.
l 11 VICE CHAIRMAN POWERS:
Have you formulated a 12 plan to get to the state of nirvana like that formally or 13 is this, " Gee, I have a hard enough time getting the first t
I i
14 step done.
Now we're" --
15 MR. CUNNINGHAM:
No.
We have talked about 16 this not so much in the context of shutdown operations but 17 in the broader context of methods development in PRA that 18 said we've had an item in the budget, at one time anyway, 19 that basically said -- I forget if it was 5 years or so or 20 7 years or something after we finished 1150 or after we 21 finished these key methods development activities, we 22 would go back and apply them in an integral sense in a 23 reevaluation of one or more of the NUREG-1150 plants to 24 say:
What is the overall impact of the methods
,,(_,)
25 development over the last 5 years or so?
In a sense, 1150 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
103 1
was that in relation to the reactor safety study in a
<s 2
sense.
3 When WASH-1400 was done, they recommended that 4
it be redone in about the 1980 time frame.
We were kind I
1 5
of slow in actually doing it, but --
1 6
VICE CHAIRMAN POWERS:
Well, I think that's 7
not fair.
I think the proper statement is they didn't 8
realize the magnitude of the problem.
9 MR. CUNNINGHAM:
That could be, too.
That 10 could be.
And there were some other things that came 11 along in the interim that kind of distracted us.
12 VICE CHAIRMAN POWERS:
Got your attention.
13 MR. CUNNINGHAM:
Yes, got our attention.
/^T
(
/
14 So this has been in the budget.
I will admit 15 that when we look at the long-term resources in the office 16 and others, it's always a big question.
Is that 17 necessary?
When is the right time for it?
18 It's an expensive proposition.
And is that 19 the right place to put the money?
That's a constant or an 20 annual discussion, I guess it is, in terms of the budget, 21 our long-term budget projections for the PRA research.
22 Should that be an element that says we're 23 going to go redo this X years down the pike, if you will?
24 Usually I don't fare very well with the budget people or r^N
(_)
25 the decision-making process in that that --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
104 1
VICE CHAIRMAN POWERS:
Well, it provokes me, 4
n 2
then, to ask a question.
Would it be of use to you and i
i
\\/
3 the entire office, not you in particular, --
4 MR. CUNNINGHAM:
No.
5 VICE CHAIRMAN POWERS:
-- to acquaint the 6
commissions with the needs in this area?
7 MR. CUNNINGHAM:
If the Committee would like 8
to do that, I'm sure the Commission and I would be happy 9
to hear about it.
Yes, sir.
10 VICE CHAIRMAN POWERS:
You've completed your 11 presentation?
12 MR. CUNNINGHAM:
Yes.
Maybe I'11 reinforce 13 the end of it.
I think we have some learning to do as to O
V 14 where we are, where the industry in general is in terms of 15 low-power and shutdown PRAs.
I think that would be 16 worthwhile to have some sort of forum where we could 17 discuss that and see perhaps where that could lead us in 18 terms of what we do in terms of other research activities 19 and things like that.
20 MEMBER APOSTOLAKIS:
So you plan to do this or 21 at this point you're just saying it's a good idea?
22 MR. CUNNINGHAM:
If the Committee thinks it's 23 a good idea, I think we would go ahead and try to do this 24 over the next -- I don't krow -- sometime between now and Ok) 25 the end of the fiscal year or something like that.
If the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISI.AND AVE., N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
105 1
Committee is not terribly interested in it, then maybe the rx 2
j k
)
3 VICE CHAIRMAN POWERS:
Perhaps it's useful to 4
tell you the Committee is acutely aware that there are 5
needs for research in this area and that their methods 6
development would have to precede any full-blown 7
NUREG-1150-style effort and did not in any sense think 8
that this was an easy chore.
9 MR. CUNNINGHAM:
Yes.
And on the technics 10 side, I think there is -- we need to go back and look and 11 see what's there.
That's a legitimate question.
And it's 12 more of a matter of the timing of when we would do that 13 relative to some other things probably.
There's probably l' )
~'
14 a better way to characterize how we would proceed with 15 this.
16 VICE CHAIRMAN POWERS:
The Committee 17 definitely does not attempt to set the management 4
18 priorities and doesn't try to dictate the time scales for 19 doing things.
20 MR. CUNNINGHAM:
Yes.
21 VICE CHAIRMAN POWERS:
We're not very skilled 22 at that, to begin with.
And we're much better at saying 23 what the needs are than how you get there.
24 MR. CUNNINGEAM:
Yes.
Okay.
(
\\
\\_./
25 CHAIRMAN SEALE:
We cause more trouble than --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i
l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C, 20005-3701 (202) 234-4433
106 1
VICE CHAIRMAN POWERS:
Than we're worth, yes.
2 Well, thank you, Mark.
That's --
b,~3 3
CHAIRMAN SEALE:
Are there any industry 4
comments?
5 VICE CHAIRMAN POWERS:
I guess we have no 6
additional --
7 CHAIRMAN SEALE:
There was no industry 8
comment?
Was there anything anyone else on the staff 9
would like to --
10 (No response.)
11 CHAIRMAN SEALE:
Well, I guess we're through 12 with that, then.
And we --
13 VICE CHAIRMAN POWERS:
We are far from
("%'-]
\\
14 through.
15 (Laughter.)
16 CHAIRMAN SEALE:
We have just begun to fight.
4 17 VICE CHAIRMAN POWERS:
We have enjoyed you so 18 much it looks like we're going to get to have your company 19 many, many times here.
20 MR. CUNNINGHAM:
Thank you very much.
21 MEMBER APOSTOLAKIS:
It was very, very useful, 22 though.
It was really very useful.
23 CHAIRMAN SEALE:
Yes, it was.
We appreciate 24 your response on relatively short notice.
p-
' _/
25 MR. CUNNINGHAM:
Certainly.
's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
107 1
CHAIRMAN SEALE:
Would you like to discuss the
,m 2
letter a little bit since we have about ten minutes here?
(v) 3 VICE CHAIRMAN POWERS:
As I said, I think in 4
our new procedures, that we're not going to try to --
5 (Whereupon, the foregoing matter was concluded 6
at 11:52 a.m.)
7 8
9 10 11 12 13 fm I
14 4
15 16 17 a
18 19 20 21 22 23 24 i(
25 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 CERTIFICATE This is to certify that the attached proceedings before the United States Nuclear j
Regulatory Commission in the matter of:
}
Name of Proceeding: 428" ACRS 1
j j
Docket Number: N/A Place of Proceeding: ROCKVILLE, MARYLAND were held as herein appears, and that this is the original i
transcript thereof for the Sile of the United States Nuclear Regulatory Commission taken by me and, thereafter reduced to
.I typewriting by me or under the direction of the court i
I reporting company, and that the transcript is a true and accurate record of the foregoing proceedings.
l.
Ak M RBETT RINE'R Official Reporter Neal R. Gross and Co.,
Inc.
1 O
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RiiODEISLAND AVENLT.,NW (202)234 4 433 WASillNOTON, D C. 20005 (202)234-4433
INTRODUCTORY S1AfU.ENT 3Y THE ACRS CHAIRMAN a
438TH ACRS MEETING, FEBRUARY 6-8, 1997 73 THE MEETING WILL NOW COME TO ORDER.
THIS IS THE FIRST DAY
_)
OF THE 438TH MEETING OF THE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS.
DURING TODAY'S MEETING, THE COMMITTEE WILL CONSIDER THE FOLLOWING:
(1)
ACRS SUBCOMMITTEE ACTIVITIES (2)
FUTURE ACRS ACTIVITIES (3)
SHUTDOWN OPERATIONS RISK (4)
RECONCILIATION OF ACRS COMMENTS AND RECOMMENDATIONS (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.
7y WE HAVE RECEIVF;) NO WRITTEN 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.
I WILL BEGIN WITH SOME ITEMS OF CURRENT INTEREST.
qb b
si
o o
o il
,f
<y United States l
- %.,,,)
Nuclear Regulatory Commission RISK ANALYSIS OF ACCIDENTS IlNITIATED DLRING LOW POWER l
AND SHLTDOWN' CONDITIONS 1
Presentation to Advisory Committee on Reactor Safeguards Mark A. Cunningham 415-6189 Martin J. Virgilio 415-3226 Division of Systems Technology Division of Systems Safety and Analysis Office of Nuclear Regulatory Research OfHee of Nuclear Reactor Regulation l
February 6,1997 l
l 1
i
- -. - - _ - - -1
\\
O O
O l
i l
Low Power & Shutdown Studies l
BACKGROUND i
e ORIGINAL INTEREST IN LOW POWER & SHUTDOWN RISK WAS BASED ON BNL AND FRENCH STUDIES, WHICH INDICATED THAT CORE DAMAGE FREQUENCY DURING SHUTDOWN MODE COULD BE SIGNIFICANT.
e PROJECT BECAME PART OF USNRC'S RESEARCH PLAN IN RESPONSE TO THE CHERNOBYL ACCIDENT IN APRIL 1986.
l s
PROJECT WAS JUST BEGINNING WHEN THE VOGTLE PLANT LOST ALL VITAL AC POWER AND RHR SYSTEM WHILE AT MID-LOOP IN MARCH 1990.
l e
PROJECT WAS DIVIDEb INTO TWO PHASES IN ORDER TO PROVIDE EARLY SUPPORT TO SEPARATE VOGTLE EVENT RESPONSE BEING CONDUCTED BY THE OFFICE OF NUCLEAR REACTOR REGULATION.
[
2 t
.-------n -.---.-
O O
O
- i Low Power & Shutdown Studies i
l s
e PHASE 1 PERFORMED A COARSE SCREENING ANALYSIS OF ALL I
NON-FULL-POWER OPERATIONAL MODES AT ONE PWR (SURRY)
AND ONE BWR (GRAND GULF).
RESULTS WERE NOT PUBLISHED AS l
NUREG/CR REPORTS BUT WERE SUMMARIZED IN PHASE 2
[
APPENDICES.
e PHASE 2 ESTIMATED THE FREQUENCIES OF SEVERE ACCIDENTS INITIATED DURING TWO PLANT OPERATIONAL STATES BY PERFORMING A REALISTIC PRA FOR THE DOMINANT OPERATING STATE OF THE TWO PLANTS STUDIED UNDER PHASE 1.
THIS i
RESULTED IN TWO PRAS:
j SURRY DURING MID-LOOP OPERATION (DONE BY BNL).
GRAND GULF DURING COLD SHUTDOWN (DONE BY SNL).
e RESULTS PUBLISHED IN:
i l
NUREG/CR-6143 (JULY 1995)
NUREG/CR-6144 (OCTOBER 1995) 4 l
i 3
l l
t i
O o
o
- t Low Power & Shutdown Studies t
SHUTDOWN RULE DEVELOPMENT l
I LATEST DISCUSSION WITH ACRS MAY 21, 1996 ACRS LETTER JUNE 4, 1996:
i EVIDENCE SUGGESTS SHUTDOWN RISK IS IMPORTANT f
PROPOSED RULE WOULD LIKELY LESSEN RISK i
I RULE BASED ON ENGINEERING JUDGEMENT j
ACRS WILL COMMENT AGAIN AFTER PUBLIC COMMENT l
RECOM4 ENDED LEVEL 3 PRA FOR SHUTDOWN OPERATIONS LIKE NUREG-1150 i
4 i
Low Power & Shutdown Studies i
5 1
l SHUTDOWN RULE DEVELOPMENT CONT STAFF RESPONSE JUNE 28, 1996:
AGREED THAT ENGINEERING JUDGEMENT SIGNIFICANT ACKNOWLEDGES LIMITATIONS OF SURRY AND GRAND GULF STUDIES l
REGULATORY ANALYSIS WILL HELP RISK INSIGHTS f
LEVEL 3 PRA NOT NECESSARY, BUT MAY BE DESIRABLE i
I i
5 r
'i
O O
O Low Power & Shutdown Studies SCHEDULE Item Date Complete development of rulemaking package 3/31/97 Office concurrences 5/07/97 Committee To Review Generic Requirements (CRGR) 6/30/97 concurrence EDO concurrence and submittal to Commission 7/31/97 t
{
6
O O
O
- l Low Power & Shutdown Studies j
1 5
FOLLOWUP RESEARCH ACTIVITIES s
e HUMAN RELIABILITY ANALYSIS t
e IMPACT OF MAINTENANCE AT POWER VS.
SHUTDOWN e
LOW POWER / SHUTDOWN MODELS FOR EVENTS ANALYSIS
(
7
't d
O O
O Low Power & Shutdown Studies t
r CONSIDERATIONS IN RESEARCH PRIORITIES TECHNICAL ALTERNATIVES:
ADDITIONAL METHODS DEVELOPMENT i
e FIRE METHODS DEVELOPMENT e
HUMAN RELIABILITY ANALYSIS j
ADDITIONAL DEMONSTRATION PRAS l
GUIDANCE DEVELOPMENT l
INDUSTRY-SPONSORED PRAS i
INDUSTRY ACTIVITIES RISK MANAGEMENT PRACTICES i
i INTERNATIONAL ACTIVITIES:
CSNI SURVEY 8
i i
l
Low Power <& Shutdown Studies 1
I l
i i
CONSIDERATIONS IN RESEARCH PRIORITIES (CONTINUED) l COSTS:
COMPLETION OF NUREG-1150 PRA " MATRIX"
- VERY b
HIGH L
CoMCLuSIONS:
COMPLETING MATRIX IS NOT TECHNICALLY-JUSTI-FIED NOR COST-EFFECTIVE APPROACH I
REASSESSMENT OF NEEDED RESEARCH IN SPECIFIC AREAS WORTHWHILE k
l j
.,