ML20248J413
| ML20248J413 | |
| Person / Time | |
|---|---|
| Issue date: | 11/08/1990 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1822, NUDOCS 9011160166 | |
| Download: ML20248J413 (134) | |
Text
,
b S\\A W l 5 2 2 m
l OFFICIAL TRANSCRIPT OF PROCEEDINGS
/
. 'n ).
['
N 1),-
L l
Agency:
N clear Regulatory Commission l
Advir,ory Committee on Reactor Safeguards
Title:
367th ACRS General Meeting L.
Docket No.
g..
- g.,
fa,,
i Bethesda, Maryland.
IOCATION:
4 i ' s ( '.
Thursday, November-8, 1990 PAGES:
1 - 95 orm u;!
Y r
e u.
w.::::=w. '.,
4 L
Au lS m, g%ywece dopy-1eri forthe L'fe of the kate 1-ANN RILEY & ASSOCINFES, LTD.
p 1612 K St. N.W, Suite 300
~
Washington, D.C. 20006 (202) 293-3950
l'
'O 3'
4 PUBLIC NOTICE BY THE 5
UNITED STATES NUCLEAR REGULATORY COMMISSION'S 6
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 7
8 DATE:
November 9, 1990 9
10 11 12 13 The contents of this transcript of the 14 proceedings of the United States Nuclear Regulatory 15 Commission's Adviscry Committee on Reactor Safeguards, 16 (data)
November 8, 1990 17 as reported herein, are~a record of the discussions recorded at 18 the meeting held on the above date.
19 This transcript has not been reviewed, corrected 20 or edited, and it may contain inaccuracies.
21 22 23' 24 25 1
1 1
UNITEv STATES OF AMERICA 2
NUCLEAR REGULATORY COMMISSION 3
4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5
367TH ACRS GENERAL MEETING
=
r 6
7 Nuclear Regulatory Commission 8
Room P-110 r
T 9
7920 Norfolk Avenue
)
10 Bethesda, Maryland 11 12-Thursday, Novenber 8, 1990 13 14 The above-entitled proceedings commence' at 8:30 15 o' clock a.m., pursuant to notice, Carlyle Michelson, 16 Committee Chairman, presiding.
17 PRESENT FOR THE ACRS SUBCOMMITTEE:
l'8 James.
C.
Carroll, Member
=
19 Ivan Catton, Member in 20 William Kerr, Member 21 Harold W.' Lewis, Member 22 Paul G.
Shewmon, Member
'23 Chester P.
Siess, Member David A. Ward, Member 9
24 25 J.
Ernest Wilkins, Jr., Mcmber i-i..m-
2-1 PARTICIPANTS:
O 2
3 R.
Fraley S. Duraiswamy
~
4 S. Mays T. Murley 5
W. Russell F. Gillespie 6
R.
Nease G.
Imbro 7
B. Grimes 8
l 9
10 s
11 12 13 14 15
-16 17 18 19 20 21 22 23 24 25
3 1
PROCEEDINGS 2-
[8:30 a.m.]
3 MR. MICHELSON:
The meeting will now come to This is the first day of the 367th Meeting of the 4
order.
5 Advisory Committee on Reactor Safeguards.
During today's 6
meeting the Committee will discuss and/or hear reports on 7
the following:
Regulatory Impact Survey; Level of Design 8
Detail for Standardized Nuclear Power Plants; we will meet 9
with the NRC Commissioners at One White Flint; and, we will
'10 discuss future ACRS activities.
11 Topics-for tomorrow's discussion are listed on the 12 schedule posted on the bulletin board at the rear of the 13 meeting room.
This meeting is being conducted in accordance with the provisions of the Federal Advisory 14 15 Committee Act.
Mr. Raymond F.
Fraley is the designated 16 Federal official for the initial portion of the meeting.
We r
17 have received no written statements or requests for time to 18
.make oral statements from members of the public regarding 19 today's sessions.
20 A transcript of portions of the meeting is being 21 hept, and it is requested that each speaker use one of the 22 microphones, identify himself or herself, and speak with 23 sufficient clarity and volume so that he or she can be 24 readily heard.
25 First of all, I would like to intraduce our new
4 1
ACRS fellow, Steve-Mays.
He is a U.S.
Navy Academy
\\
i#
2L Graduate, and'was in the Nuclear Power Training program, and
~
3 also went-to the University of Idaho and has an M.E.
4 electrical engineering degree ~
He served in the Navy from 5
1974 to 1979.
He worked as a Senior Engineer at EG&E from 6
1979 to 1982, was a Senior Consultant at Tunare from 1982 to 7
1989 and1for a short time before he came nere became an 8
independent consultant.
Steve, we certainly welcome you and 9
are looking forward to your activities.
10-I have no particular items of interest other than 11 to point out to you that there is some interest in the
-12 Federal Appeals Court Decision.
You have a handout, number n.)
13.
three, which gives you some background. I understand that a 14 little later'on this morning, about 10:45, we are going to 15 have Harvey Malsch give us a rundown on what this really
.16 means as opposed to what some of the trade literature seems 17 to have to say.
So, we will look forward to hearing from 18 Mr. Malsch at that time.
19-I have no other particular items of interest this 20 morning.
Do any members have any items they would like to 21 bring up?
22.
MR. SHEWMON:
Larry has officially resigned, or is 23 that a rumor?
(}
24 MR. MICHELSON:
Excuse me.
Yes, we have two
'v 25 absences this morning; Larry Minnick and Charlie Wylie are
5 l'
both out for health reasons.
The exact status on Larry is O.
2 not official yet.
We are waiting to get an official status.
~5 3
Ray will discuss it later when we have a session on the 2
4 nomination of new members.
5
.Are there any other items?
6 MR. KERR:
I talked to Charlie Wylie yesterday or 7
the day before, and he said he was getting along very well
~h 8
and is out of the hospital.
9 MR. MICHELSON:
Yes, I chatted with him also.
He 10 seems to be ready to go, except he has a little waiting 11 period yet.
Are there any other items?
12 (No response.]
()
13 MR. MICHELSON:
Seeing none, if the staff is ready 14 we would like to proceed with our first agenda item a little 15 ahead of schedule, the first agenda item being the 16 Regulatory Impact Survey.
We have some distinguished j
17 members of the staff here this morning to give us the top 18 down view of this thing.
19 MR. CARROLL:
At least the usual suspects.
20
[ Laughter.)
21 MR. MICHELSON:
Mr. Murley, did you want to be the 22 lead on this?
23 MR. MURLEY:
Yes, I will, Mr. Chairman.
I would 24 like to kind of make an announcement of another item of 2E business if I could first.
It will take about five minutes.
6 1
I-have recently received approval from the Commission and-
.2 from the EDO to make some organizational changes.
They are L
3-relatively minor, but I think they are minor changes but 4
will be important in the way we do business for reviewing 5
advanced reactors.
6 There will be a new division, and I didn't bring 7
the announcement with me, but it will be like the Division 8
of' Advance' Reactor Review.
Its sole purpose will be to look 9
at the new advance designs that we have under review.
It 10 will be headed by Mr. Dennis Crutchfield as the Division 11 -
Director, and Bill Travers will be the Deputy Director.
12 There will be four branches in the-Division; one branch is
(
13 Charlie Miller's branch that you deal with.
It will be 14 primarily associated with the evolutionary light water 15 reactors, the passive light water reactors, and the EPRI 16 requirements document.
There will be a new branch that will review the 17 18 more further advance designs lik PIUS, like the modular
=
19 HTGR, and like the liquid metal. reactors.
That will be Bob 20 Pearson, who will be moving over from another projects 21 branch.
In addition, we will have the Plant Life Extension, 22 Plant License Renewal Branch under John Craig will remain 23 pretty much as it is.
Then there's a branch dealing with 1
24 non-power reactors, test reactors, university reactors, and 25 decommissioning under Cy Weiss.
That will remain pretty m
E
=
==m m
i 7
1 much as it is.
f;.
i a
\\~/ '
'2 So, we hope that by focusing all these licensing 3
activities in one division that we be more efficient, and 4
this will be announced I think relatively soon.
We hope
.5 after we can make all the arrangements, it will actually be 6
in place probably within a month or so.
We have to notify 7
the Union and do all those things.
8 MR. SIESS:
On the DOE advance-reactors, this 9
moves them from research into NRR?
10 MR. MURLEY:
Yes.
11 MR. SIESS:
Are they going to change the name of 12 their Advance Reactors and Generic Issues Branch?
l l
r N.
'; vJ-13 MR. MURLEY:
They will still be doing research 14 work in support of this.
I don't know that they contemplate 15 making any changes because there is going to be, I think, 16 maybe only a couple people actually change.
Their research 17 is still going to be very, very-closely-involved with-us.
18 MR. SIESS:
But they won't be reviewing.
19 MR. MURLEY:
They won't be reviewing or licensing 20 or things like that, we will be responsible for putting out 21 the SER's and that sort of thing.
22 Thank you, Mr. Chairman.
With that, we will move 23 into our discussion today of the Regulatory Impact Survey.
- (s 24 Back in April of this year Burt Davis and I briefed the i
25 Committee on the survey that had been done about a year ago
L L
8 1
now, and that was outlined in draft NUREG 1395.
Since that 2
time there have'been two reports that are part of this 27 3
survey really, SECY 90-205 issues in June of 1990 was a 4
brief survey of all licensees on the management time spent 5
responding to outside inspections and audits; that is, NRC 6
inspections but also audits from INPO and other 7
organizations.
8 In July, 1990, we issued SECY 92-50 which was a 9
survey of the NRC staff on its thoughts with regard to 10 regulatory impact.
I do not propose today to go over those.
11 I think we will want to focus primarily on what we are doing 12 about the impact and what proposals we intend to make.
()
13
[ Slide.)
14 By way of background, I think I would like to 15 repeat why we did this survey, and it goes back really to 16 about mid-1980's.
In 1985 you recall the Davis-Besse event 17 of loss of all feedwater caused us to really ask whether we 18 were doing the right thing in our regulatory program, 19 because that was quite a close call with regard to core 20 damage.
We went through a kind of introspective lessons 21 learned effort where we decided that the staff had a lot of 22 information about Davis-Besse but we weren't pulling it 23 together very well and we weren't drawing the conclusions
(
that we needed to draw about that plant and others that we 24 25 did not think were operating well.
I
9 1-As a result, we changed our inspection program'to
(
)
\\~
2-be more diagnostic nature, more team inspections.
We I
3 changed our SALP process to look more carefully and 4
critically at the way plant operations were being managed.
5 There have been changes over the last few years with regard 6-to the way we are approaching looking at operations of 7
reactors.
We think it's been successful, in the sense that 8
the performance indicators that we look at all look in the 9
r'.ght direction.
10 We have briefed the Committee on one particular 11
. program, the accident sequence precursor program, and we 12 have just recently received the results from 1989 look at A
i ).
13 precursors.
What I wanted to show the Committee here is the l
'14 cumulative conditional probability of core damage.
15
[ Slide.)
16 For the calendar years 1985, 1986, 1987, 1988 and 17 1989, this is the year I am talking about, we have plotted 18 the cumulative conditional probability of core damage for 19 all the reactors and all of the sequences that were looked 20 at in this program.
Without belaboring the results, what
'21 you see in 1985 are some sequences that had conditional core 22 damage probability in the range of ten to the minus one to 23 ten to the minus two.
These were the Davis-Besse event that eN 24 I
mentioned and, also, the Rancho Seco overcooling event in
'Y 25 1985.
10 1
Since that time the trend has been steadily (u,\\
's-2 downward, and 1-think perhaps the most significant thing is 3
not only
-- two parts -- not only is the absolute value of 4
the conditional core damage probability going down but we 5
are not seeing any of these high conditional core damage 6-precursors.
What we are seeing is a fairly large number of 7
smaller events.
This does not prove, of course, anything.
8 It's an indicator.
I think it's the one indicator that I 9
personally pay most attention to because it's based on 10 actual data looking backwards in time.
11 It's clearly in the right direction.
Insofar as l ---
12 we can use indicators like this to tell us where we are (m) 13 going, I-think this one is quite useful to us.
14 MR. KERR:
What confidence do you have that these 15 indicators are related to risk?
I have high confidence that they are 17 related to risk.
Are they related to total risk, do they l
l 18 indicate and show all of the possible sequences that could l
19 lead to core damage, I think the answer is no.
It does not l
20 really look at external events, for example, that may have 21:
low frequency but have high consequences.
They don't have a 22 methodology say, for analyzing what the Loma Preta l
23 earthtflake, what that contributed to potential risk for 1
/T 24
- example, b
25 What I worry about a little bit is that if the
11 1
methodology that they use, which-is fundamentally PRA i
x' methodology is-not complete enough to pick up a precursor, 3
it won't really find it:
So, what we don't see in this 4
precursor study for example is much discussion of 5
intersystem LOCA precursors.
Yet, when we go out and look 6
separately -- we have a separate program to look for that --
7 we do find it, these precursors.
E The' reason is subtle.
I think you need to get J
9 briefed on the intersystem LOCA program separately, but the 10-reasons are that some of the classical PRA methodology just
'll has not really focused on the way that human errors can i
12 introduce certain sequences.
I guess that's a long-winded (G,_)
way Bill of saying that it's not complete, we know it's not 13 l
14 complete.
Yet it is the best indicator that we have, I 15 think, of operating experience and tha indicators of risk.
16 MR. LEWIS:
Tom, I think that you said it 17 reasonably well.
In a purely statistical sense, any 18 accident consists of a precursor and consequence 19 developments.
A program like this will pick up those 20 sequences which begin with the precursor which has a 21 probability like ten to the minus two, because you have a l
22 few years data that will not pick up the things that start 23 out with ten to the minus three and then end up with a ten
(~N 24 percent chance of going all the way.
\\..)
25 You mentioned external events, which are in that
12 1
category and certainly won't pick up the things which are 2
all at once things like a catastrophic earthquake.
But for 3
the ones that hsve a reasonably high probability in the 4
initial stages for the precursor this should be a better way 5
of doing it than calculation because you are studying what 6
'really happens.
I think that's a very good thing to do.
7 The only question I have is, has anything else 8
changed in those few years?
Are people now still
-9 calculating the conditional probabilities the same way they 10 did four years ago, because in those curves as each 11 precursor comes along somebody calculates the conditional 12 probability of core melt.
Are they doing it the same way
()
13 across those four things?
14 MR. MURLEY:
Generally the methodology has stayed 15 the_same but they have made some changes.
For example, 16 success criteria have changed over those-years.
There have 17 been some minor changes.
I don't think -- I am quite sure 18 it would not affect the trend, but it might change some of 19 the --
20 MR. LEWIS:
If anything changes then it would 21 affect the trend.
For example, 1150 has been going on 22 during these years.
If they have taken the lessons of 1150
'23 to heart in calculating what happened last year and clearly 24 did not in calculating what happened four years ago, then 25 that does affect the trend.
I just wor.dered to what extent i
13 1
that is something I should cross out of my mind.
p_
A-2 MR. MURLEY:
I don't have a good answer for that.
3 I think-the question'that you are both kind of raising is 4
that these -- I don't want to emphasize these as absolute 5
_ values.
They were done generally consistently over the 6
years.
They are looking at data in the same way but, of 7
course, there are changes.
I think I will still take some 8
comfort _in the trend that plants are being operated better.
9 I think that's the main lesson that I wanted to leave.
10 MR. CARROLL:
You also have a number of plants 11 start up through this period where you might expect some 12 problems with'that also.
l
-f x
(
)
13 MR. MURLEY:
That's right.
14 MR. CARROLL:
In fact, the trend has been 15 decreasing.
16 MR. MURLEY:
That's correct.
In the outer-years, l:
17 there.are several more plants operating than in the inner
~
l 18 years.
I hope you didn't take my comments as 20 negative.
I think it's a wonderful thing to do, and it's i
21 even better to do it right.
l 22 MR. MURLEY:
Yes, I understand that too.
Let me l
23 move on then to why we did the survey.
That is, we know I'
(Jy 24 that the changing reguJTtory focus on operations and in 25 particular things like increased team inspections has had an
14 1
impact on licensee's and the way they operate their plants.
2 We have become more intrusive at plants, intrusive in 3
operations.
4 It was-at the Regulatory Information Conference in
'5 April of 1989 that I began to hear kind of in a systematic 6
way at least, a large number of complaints about that maybe 7
we were having a negative impact on safety by the way_we 8
were regulating and inspecting and that sort of thing.
Most 9
of the information that we heard was highly anecdotal-; that 10 is, somebody would come back with a story did you hear about 11 this and did you hear about that.
From the times that I or 12 my staff would look into these things we always found that 13 the story was not exactly the way as portrayed.
14 I think-it was more because we wanted to get a 15 systematic view of the impact that our regulations were 16
-having out there; that we undertook this survey about a year 17 ago.
We have briefed you on the results of it.
It was not 18 meant to be an even-handed or balanced result.
We wanted to 19-hear directly what was on the minds of the licensees.
We 20 told ourselves not to get into a debate with the licensee, 21 and even though'there were times when I was hearing stuff I 22 knew myself was absolutely wrong, we did not want to stop 23 and get into a battle or turn it into a debate at the time.
24 So, the survey that you have, NUREG 13-95 is 25 really as we heard it, warts and all.
We then took that
15 1
and from it drew three major conclusions of where.we do need p_,
\\--
2 to improve.. They are the following.
We are going to talk 3
about each one of these in detail today and Bill Russell and 4
Frank Gillespie will do that.
The first one is, we do not 5
do a good job of considering the cumulative effect of NRC 6
generic requirements.
This is probably not a new theme 7
actually, but we thought we had made some major corrections 8
back in the early 1980's when we. installed the CRGR and we 9
have a-systematic process for going through each of our..new 10 requirements.
11 We do that today.
In fact, this was probably the 12 biggest misconception that utilities have.is that, they-felt L. :,7
\\
(,J 13 we weren't following our bacKfit rules and that all these m
,14 generic letters that come out were kind of illegal products L
L 15 of the staff.
They did not know that we go through this 16 formal process of a regulatory analysis and backfit 17 analysis.
I think partly they had been led to believe the 18 backfit rules has no backfits ever, under any circumstances.
19 MR. KERR:
Tom, I think it does not follow that 20 because you go through the formal backfit. rules you are not 21 following the staff's inclinations.
There is n lot of 22 flexibility in the way one does backfit analyses.
I am sure 23 you know, because you have stood the process.
I don't think
(~]
24 the licensees, because they felt that individual staff input Q,/
25 was influencing what was going on, necessarily thought there
l:
16 1
was no process.
They just wondered how much the process r~'y
?
x-_ l 2'
influenced the opinions of the staff.
That's an
'3 alternative.
-4 MR. MURLEY:
It could be.
That's not what I heard 5
when I was out there.
They did not really understand that 6
we were going'through any process at all.
Maybe as one 7
moved up in the organizations they understood that subtlety 8'
but'I didn't hear it.
9 Nonetheless, as we considered what we were doing, 10 we concluded that there is an area that we don't do very 11 well.
That is, we don't consider -- we consider each issue 12 in isolation.
We don't consider what happens when we lay A.
(
)
13 three, four, five or ten of these new requirements on
. x,)
' 14..
licensees.
.So, we have a proposal on how to address that 15 issue, and Bill Russell will talk about it in a minute.
16 We did not do a good job of scheduling and control i
17 of inspections, particularly team inspections.
There were 18-many stories there which we took to be true, and I am sure 19 they were true, of sometimes eight to ten to 12 team 20 inspections in a year at a plant.
They are very disruptive, 21 and we concluded that clearly we had to do a better job of-22 controlling that and we have already implemented that.
23 Frank will talk about that.
<"'s 24 MR. WILKINS:
Tom, did I understand you to say b.
25 ca' er that these team inspections are not just the NRR
4 17
' l inspections or the regional inspections, these are INPO, 2
NUMARC or whatever.
3 MR. MURLEi:
That's right, although we can't 7
4 control the outside inspections.
What we can do and will do 5
is' schedule ours so that they don't interfere and that we 6
take into account the impact of these outside audits.
We E
7 are only controlling our own, that is, both regional and NRR 8
team inspections.
9 MR. SHEWMON:
That eight to ten, is that just NRR
~
10 teams or is that everybody's teams including INPO?
11-MR. MURLEY:
The ones that I recall included INPO.
1 12 Finally, and perhaps the toughest one is what we call staff-4 a
h 13 professionalism, management expectations, training and 14 oversight of inspectors, both headquarters and regional.
We 15 have a program to go over that, and Frank will do that in a 16 second.
Let me turn first then to Bill Russell, who will 17-talk about the first item.
R 18 MR. RUSSELL:
I'm going to talk about how we are 19 proposing to better and more effectively manage cumulative 20 effects.
Tom had indicated that through the CRGR process
]"
21 and our internal reviews, we essentially look at issues one 22 at a time.
=
23
[ Slide.]
24 Because of plant-specific differences, it is very 25 difficult to consider cumulative impacts other than doing it i
ir
'l 18
.1 for a theoretical facility.
We essentially concluded that l
'2 that would be' extremely difficult.
We however concluded s-3 that Tna need to communicate better the process that we 4
followed, articulate with each requirement that is issued, the reason for issuing it, and summarizing whether it is 6
something which we conclude is an enhancement pursuant to 7
the backfit rule, whether it is being imposed because it is 8
required to comply with existing rules and regulations or, 9
in fact, what is the basis for proceeding.
10 We have started doing that, and we have been doing
.11 that for over a year now.
We are including specific I
.12 language in each generic communication that is issued that-
)
13 describes how that particular requirement comports with tha 14 backfit rule.
'15 MR. KERR:
I guess I'm not quite sure that I 16 understand your opening statement which was that it is very 17' difficult to consider the cumulative effect of these rules 18 on a plant.- Of course it's difficult, but if it's extremely
'19 important'how can you avoid --
20 MR. RUSSELL:
I submit that it is difficult to do 21 it generically, and we are proposing to do it on a plant-22 specific basis.
I will be coming to that point later.
I am 23 just describing that because of the differences in plants,
("'y 24 if you do it generically through some kind of generic O
25 analysis we are not really assessing the effectiveness of
19 1
the cumulative impact'of those requirements on a plant-
[,\\
~\\/
2 specific basis.
That's the only point I wanted to make.
3 MR. KERR:
I misunderstood you.
I withdraw my 4
_ comment.
5 MR. RUSSELL:
We are involved in other activities 6
associated with the front end of the process;-that is, the t
7 development of regulatory analysis and review.
We have had 8'
workshops that have discussed the backfitting process.
We 9
have issued guidelines' on backfitting, and we are in the
'10 -
' process of revising the value impact handbook for doing 11 regulatory analysis.
12 The most important concept I think is how to~1ook
}n.)
13 at, on a plant-specific basis, the impact of the regulatory 14 requirements on a plant.
.15 -
[ Slide.)
16 The point I would like to start with is that this 17 is really not a'new concept.
That is, considering h
L 18 cumulative impact on a plant-specific basis started I think l
19 really with the systematic evaluation program with the 20 integrated assessment process.
That was essentially holding l
21 until the end of the review those potential areas for which 22 modifications of facility were needed and-evaluating those 23 using tools such as risk insights and others.
/~N 24 From that program the integrated safety assessment 25 program was developed which was also -- that was a voluntary l
20
. ;~s 1
program.
There were not many volunteers for the program 3-')
2 that had not already been through the SEP program.
It did j
3 require the development and use of a PRA, and it had other 4
elements for new facilities that had not reviewed some of 5'
the external events gone though the SEP process -- I think 6
detracted from the participation.
So, there were only two 7-facilities really that participated, Millstone I and Haddam 8
Neck, both of which had been en SEP.
9 We also have a policy statement on integrated or-10 living schedules which would al]ow licensees to propose 11 schedules for implementation that required condition of a 12 license and other regulatory interface which had the
) p).
js,
13
' potential for incorporating licensee desired enhancements 141 into a regulatory process, and was also not well accepted.
15 Based upon those early starts -- we still think 16 the concept is good and we are trying to address a program 17
'that would eliminate some of those reasons why utilities had.
18 not earlier participated.
We believe that the program --
19 and let me describe broadly what we are considering by way 20 of a time flow.
21 Approximately one year prior to an outage, we 22 would propose that a utility would submit a prioritization 23 of all regulatory requirements that are outstanding at that
~'N 24 time considering both safety significance and impact on the
[d 25 facility, and would identify those items which would be
i 21 1
proposed to be implemented in the next outage.
That sould
'd 2
he submitted to the NRC for essentially a negative consent; 3
that is, a review ~rocess that, unless the NRC objected-and 4-wanted to modify the schedule, would be accepted.
5 That would freeze the items --
6 MR. SIESS:
How much time would they have?
Right 7
now it takes --
r 8
MR. RUSSELL:
On the order of 60 to 90 days.
If i
9 they didonot near otherwise from the NRC, they would 10 proceed.
11 MR. SIESS:
That sounds like almost automatic 12 approval.. The NRC can't do anything unless they hear.
A)
C
.13 MR. RUSSELL:
I-understand that there are going to l.
14 be some difficulties with this, and I can talk about the 15 approach.
What I wanted to do was describe the concept 16 first.
Absent negative consent or a negative-finding by the 17 staff, that-schedule would be implemented for the next l
18 outage and for the next nine months there would be no 19 changes and no additional requirements would be imposed, 1
20 which would be enhancements.
21 Clearly, if we found a situation where a licensee 22 failed to comply with the requirement where something was 23 needed to be imposed on a faster schedule to meet a 24 reasonable assurance standard for adequate protection that l
25 would be imposed, that group of items which are subject to
I 22 1
regulatory analysis, cost benefit which are enhancements t, sg.
5 2
which improve an already adequate level of safety..
i 3
The process would then repeat itself at the next-i 4
outage.
There are some rather interesting' aspects to this 5
. process.
Those things which, on a plant-specific basis do 6
not have much safety payoff, could conceivably be at the 7
bottom of the list and not be implemented during that outage 8
or subsequent outages.
There is the potential that things 9
which got through on a generic basis, the regulatory process 10 as a potential backfit, when it's looked at on a plant-
'll specific basis is not as important and would be at the i
12 bottom of the list.and may not be implemented.
I 13
'It also would allow a utility to-consider other t
lI
-14 things that are going on, resource impacts, what else is l-15 scheduled during the outage that may be on his list that he 16 would like to do that is not a regulatory requirement, and l
17 would. allow them to identify those issues such that they l
L 18 scope of. work to be done during a particular outage would be i
19 smaller and would be able to be managed more effectively.
l 20 MR. MICHELSON:
How do you control the situation l-21 wherein a licensee may not be willing to put forth a whole 22 lot of effort on a given outage and, therefore, he is 23 slipping a number of items on out to the second or third l
/~N 24 outage into the future?
How do you prevent that sort of 25 thing?
l
g -
23
..j 1
MR. RUSSELL:
We are proposing a process of
-l
).
L's '
2 development that would involve a pilot' project with a 3
facility in each region dealing with the staff over the 4
course of about a year to develop the criteria for doing the 5
prioritization, assessing the. impacts generically, providing 6'
guidelines that project manager's would use during this 7
negative consent period.
If we found that there were a 8
significant number of items being deferred and there was not i
9
- a. rational basis on either safety or impact for deferring 10
.those,-that would require the NRC to not concur and take an 11 action to say we disagree with that approach and you should 12 implement it on a particular schedule.
w f
f 13 The important part I think is-that one, we-are 14 going to be looking on a plant-specific basis and L
i i
[
15
'prioritizing based upon safety.and impact and,-two, that the 16 process would be done significantly in advance of an outage l
l 17-such as the planning and engineering can be done and the 18 outage can be managed more effectively.
I hope it's
[
'19 -
understood that we did do a review and found that the bulk 20 of NRC requirements that are imposed through generic 21 communications are indeed implemented during outages.
22 That's where the impact occurs.
I
.23 It does not occur during the period that the plant l
I
's 24 is on-line.
Most of these require outages to implement, and
-(h 5,
25 that's where there is a significant impact.
That is also
h 24 1
where the bulk of the utility work.is accomplished, whether 2
it *.s a modification to enhance the performance of the 3
station or something else that the utility desires to do-4 during that period.
5 This is intended to integrate those requirements 6
on a plant-specific bases which we have required,_look at the safety priority and schedule, and freeze that in time 8
prior to an outage.
9
- MR. SIESS:
Bill, how does this differ from ISAP?
[
10 MR. RUSSELL:
I guess the principal difference, 11 ISAP also included reviews'of technical issues which were 12 not on the agenda for that licensee.
They had not been
()
13 imposed through a process, a seismic re-review, some of the 14 external events.
There were some 27 issues that came out of
{
15 SEP that were going to be addressed during the ISAP process.
16 That was one.
There was new work they would have to address 17 in order to buy into that program.
18 The second thing was that we wanted a PRA to be 19 performed.
This was clearly before you had the IPE's with 20 present utility commitments with PRA.
21 MR. SIESS:
This does require a PRA.
22 MR. RUSSELL:
It requires a safety prioritization 23 through the IPE process.
Utilities are, indeed, already 24 generating and PRA and we believe that would be the most 25 logical way of developing safety priority on a plant-1 l
i i
iii,,,
in
=
=m
---um----mm-
25 1
specific buses, would be to fact the new requirements into 2
the IPE's.
3 MR. SIESS:
How much industry input have you had 4'
in developing this?
5 MR. RUSSELL:
Right now it is a concept that we 6
have described to the Commission.
The Commission has not b
7 yet acted on the staff proposal.
We have recommended that
_8 it be issued.for public comment and that after that comment 9
is received, the Commission then act on the staff proposal.
L 10 We will have a: final report to them that addresses those
~11
-comments.
Then we would develop the program on a pilot 12 basis over the course of approximately one year.
()
13
'MR, SIESS:
It just seems a-little strange to me
.14 that the ISAP program which, for reasons I don't yet
.15 understand, is met with very, very little enthusiasm and 16 this looks a lot like it -- that the' industry has never come
~17 forth with a proposal of their own to manage these things 18
'that obviously give them a problem and that they complain 19 about.
20-To your knowledge, has industry ever proposed a 21 way of handling this or does it always have to come.from 22 NRC?
23 MR. RUSSELL:
I am not aware of an industry 24 proposal.
We have had discussions with some utilities.
I 25 have had discussions personally with representatives of
I 26 1
Commonwealth Edison as to whether this approach on the part
.,~'M i
' ~'
'2 of the staff, limiting it to NRC regulatory requirements and i
3 prioritizing those and freezing them at a point-in time 4
prior to an outage and reaching agreement on what is to be l
5 done, would be responsive to the issues that came out of the 6.
impact survey.
At least in that one case it was felt that L
7-it was responsive and that this would be a reasonable 8
approach.
That's a sample of one.
L l
9 We have not had significant comment back yet.
In 10 fact, we have not had commission approval to issue the staff L
i 11 paper for public comment.
12 MR. MURLEY:
Let me partially answer that, Chet.
I e%Q 13 The-industry was not uniform on their response to ISAP, if
-l 14 you recall.
Northeast Utilities was very enthusiastic about.
15 it, and they kind of on their own initiated a program 16 whereby they used their PRA to prioritize work, and they 17 still use it to this day for a dialogue with the staff on 18 what's important and what's not important.
19 I think partly the reason for that was the fact 20-that they were -- let me say it -- they were enlightened in 21 the use of PRA and the techniques, and they saw the tenefitc 22 of ISAP.
Now, perhaps the time -- this has been several 23
- years now, five years since we first floated that.
Now that
()
24 each utility is going to have no do an IPE type of analysis,
'g/
25 perhaps will meet with more success this time.
I don't know
27 1
Bill if you mentioned the notion of pilot program 7 but you 2
might want to talk about that.
3 MR. RUSSELL:
Yes, I did.
4 MR. SIESS:
I think both of you have used IPE 5
essentially as synonymous with PRA.
It seems to me that the 6
new issue comes up and you presumably can plug it into a 7
living PRA.
I am pretty sure that Northeast Nuclear can do 8
that.
I don't see how you could do that with something that 9
isn't a PRA, where it's a new issue that hadn't been looked 10 at before, how you plug it in.
It seems to me that you 11 start over on it and decide what to do about it.
12 MR. RUSSELL:
That is probably the biggest issue, 13 the IPE process does not require a living PRA.
It's a one 14 time look for severe accident outliers.
It would not 15 require a utility to maincain the PRA current with time or 16 to factor new issues into the PRA as they evolve once it's 17 complete.
All utilities are, in fact, doing essentially a 18 level one PRA for internal events.
We think that makes a 15 lot of sense and that there are benefits to a program like 20 this that would have them prioritize new regulatory 21 requirements on a plant-npocific basis if th&t makes sense, 22 and they could use the PRA to prioritize their own issues.
23 That's why the prog:am is characterized as being voluntary.
24 If they choose not to, we would fall back to 25 here's a now requirement and implement it within two
F-28 j
1 outages.
7
-s/
2 MR. SIESS:
You may remember that the industry 3
itself -- I guess it was EPRI -- made a survey of utilities 4
that had PRA's and what they thought about them and how they 5
were using them.
Some of them were awfully enthusiastic and 6
others apparently don't find much interest there.
7 I spoke to one utility with a newly licensed plant 8
and asked them flat out why aren't you interested in ISAP.
9 The answer was, we don't have that many requirements.
I 10 told them to wait a while, and maybe by now they have more 11 interest, I don't know.
12 MR. WARD:
I guess I have the same reaccion that
-,s l q) 13 Chet did to this, similar to ISAP.
I hope you plan to 14 interact with the industry soon or get their feedback.
l 1
l 15 MR. RUSSELL:
We are just waiting on the
!=
16 Commission's approval to do that, and we do see a need to j
17 interact with them on this issue and get their comments, and 18 then work it through a pilot basis with a number of 19 utilities.
20 MR. WARD:
One difference it seems to me is, this 21 program is very frankly a year at a time, where you promise 22 them a respite from new requirements for nine months or 23 something.
That may be more practical.
ISAP was maybe a
,/~T 24 little too idealistic and looked as if you were going to V
25-write things off forever in that.
l E
i 29 1
MR. RUSSELL:
I used a year as an example. It may 2
be 18 months for a utility that's on an 18 month fuel cycle.
3 We are basically talking about cycle-specific and freezing 4
it at some point in time prior to the outage, and then not 5
adding on to that list unless it trips the threshold of 6
being required for adequate protection.
7 MR. SIESS:
Do you have any feeling for how often 8
that has happened, that the ill you are trying to correct 9
has occurred where the utility has developed their wholo 10 plan for the next outage which I assume they have quite a 11 few people working on, and then NRC comes in and says here's 12 a new requirement and you have to implement it at that l
g I.
13 outage.
How often does that happen?
14 MR. RUSSELL:
That came up relatively frequently 1
15 during the discussions.
I think it is a relatively frequent 16 occurrence.
We often will issue a requirement and have it 17 be done within six months if you are in an outage or the 18 next outage, if it occurs more than six months after that.
j 1
l 19 The impact on planning for an outage cc.n be quite severe, 20 depending upon what the issue is.
21 We went through and we did look at all the 22 requirements that are currently out on the street and what 23 the impact was in terms of estimated costs by the NRC.
p 24 These are not utility costs.
We found that steady-state,
'N 25 those things which could be implemented during operation
30 I
were less than $100,000.00 a month, what I will call
')
i
'/
2 regulatory dollars.
3 The impact on outages was in the few milliens to 4
be implemented during the course of an outage.
We were 5
already impacting two outages out.
On a theoretical 6
facility there is a fairly significant impact on average.
7 There may be some that only have a few items that are 8
relatively easy to do and they would choose not to 9
participate, because they just don't have very many things 10 to implement.
Other plants may have quite a large backlog 11 and would want to participate in such a program to make it 12 manageable and to prioritize the value for that which they n
I, )
13 are expending.
14 If there are no other questions, the next issue is 15 16 MR. CARROLL:
I guess I do have one, Bill.
Over 17 the years thinking about the various kinds of schedule 18 agreements between licensees and the NRC, I have always had 19 the problem that a utility has a finite amount of resources 20 and money.
I just have never been able to figure out how 21 the utility and the NRC are going to agree on how that money 22 is to be spent.
23 In a lot of cases for almost every outage there's r'~'s 24 a big long list of plant enhancemente that have absolutely U
25 nothing, directly at least, to do with the NRC and its
31 1
regulations.
Many of them are big ticket items both in f
)
k/
2 terms of outage time and in terms of dollars.
3 What is your insights into how you wou2d view a 4
utility that came in and said I have this list of NRC things 5
but I don't think any of them are as important to me in 6
effectively running this plant as spending money on these 7
plant enhancements in the next three outages.
So, I don't 8
think I should really spend my resources on NRC things.
How 9
do you get into that?
10 MR. RUSSELL:
That's exactly the process and 11 dialogue that we propose to get into.
It could be cost or 12 it could be impact in an area.
For example, if there is too
,,()
13 much work scheduled in the containment you could only manage 14 so many people effectively in a containment at one time.
15 The issue may be some work gets deferred that is an 16 enhancement.
Again, we are talking about --
17 MR. CARROLL:
But you can always say to them, I l
18 want to get my NRC work done so just have a longer outage l
19 and then you can control the work.
20 MR. RUSSELL:
That's possible.
That's not the 21 intent.
That's why I say we have to develop through a pilot 22 program, reasonable guidelines of how you make those 23 tradeoffs.
You put it to the utility to propose first and
(~}
24 unless the NRC objects that is what becomes locked in for
\\_/
25 the next outage.
These, again, are issues which are
32 1
enhancements to safety under the backfit process.
They are 2
not issues which are required for adequate protection or for 3
compliance with existing rules and regulations.
~
4 We are talking about more effectively managing 5
backfits so that we have that time flexibility.
We think 6
that's a reasonable process.
There may be some that a 7
plant-specific PRA would show are not cost-beneficial on a 8
plant-specific basis and would be low on a priority list and 9
may not be done at all.
That's the concept.
[
10 We need to work on it, we need to work with 11 industry, we need to develop reasonable guidelines as to how 12 that is done, and work it through a pilot program.
We are proposing to do that with one facility in each of five 13 14 regions.
15 MR. CARROLL:
To all of that, I say good luck.
16 That's a tough problem.
17 MR. MURLEY:
I think I should mention -- I think 18 it's an approach to a disciplined way to do it.
Bill and 19 his staff, I think, have really come up with an innovative approach.
20 21 If it's not clear, I should mention that we are w
22 only talking about safety enhancements here.
We are not 23 talking about issues that are needed for adequate 24 protection.
Those are going to have to be done, period.
25 MR. SIESS:
When, at the next outage?
Something
33 1
needed for adequate protection is going to be done when?
2 MR. MURLEY:
That is usually specified in the rule 3
itself or in the staff guidance document.
I don't think I 4
have a flat answer for that, Chet.
5 MR. GILLESPIE:
Chet, let me just amplify one 6
other thing.
Having presided over the death of ISAP and 7
also still muddling around with a draft policy on --
8 MR. SIESS:
Doesn't Northeast Nuclear still have 9
ISAP?
10' MR. GILLESPIE:
Yes.
One of the things that was 11 found defensive besides what you said, and I got exactly the 12 response you did when I met with Hal Tucker from Duke.
His 13 comment was, I don't have so much on my plate that I need to 14 be involved with the NRC.
Also, the way ISAP was formulated 15 there was kind of an intent -- this is probably the 16 idealistic part of it that weighed on it too heavily -- the 17 process involved in coming up with the list of things to be 18 done would in fact be part of the license.
19 A comment that I got from many utilities -- and I 20 met with the utilities that we sent out a generic letter and 21-se*.'eral came in -- one of their comments was that we don't 22 want the NRC involved in our internal management processes 23 and how we make decisions.
If there was an overriding 24 comment, it wasn't they didn't have so much on their plate, 25 it was that they didn't want us involved.
34 4
1 The idealistic view we had when we proposed that
\\
\\ -
2 was we will get a neat process.and we will all agree on it, 3
and then we will sit back and kind of turn the crank and 4
make sure it's going okay and monitor from that point 5
forward.
We were removing some management prerogatives or 6
getting involved with them at the time.
There were some 7
real strong feelings on the part of soms of the utilities.
8 This, as Bill described, attempts te not take that step.
9 The management prerogative par.t, t.nere was a strong feeling 10 by many utilities on that p9rt.
11 Let me switch now to --
12 MR. CARROLL:
Doesn't making judgments about what
/(j 13 I want to do from a liability point of view or a personnel 14 safety point of view that has nothing to do with the NRC 15 directly of making judgments about spending money on those 16 kind of things versus NRC safety enhancements really get you 17 into that arena pretty much?
18 MR. GILLESPIE:
It does, but as I think Bill 19 described, on a limited basis.
That, in fact, would be part 20 of the criteria we would want to come up with on an
- 21 intritive basis with the pilot plants is, how do we consider 22 those things but not get ourselves involved in them on a 23 continuum.
What we did with ISAP is, we ended up getting
('N 24 involved on a continuous basis with that whole broad 25 spectrum of everything to be done in order to understand the
35 E
1 process for making those tradeoffs.
2 We are trying to take the lessons learned from the
~
3 failures of the past.
I don't particularly want to
~
personally get involved with another one.
We have put a lot 4
5 of time in an ISAP.
We put a lot of time in a living 6
schedule.
I think it's going to take -- Chet, you hit it --
j 7
utility feedback to say at an early stage this ir a viable 8
thing to do.
9 MR. CARROLL:
The other caution I would have is 10 that you could very well got used by the nuclear side of the T
11 utility as a way to get more budget money out of the --
12 MR. GILLESPIE:
If that's enhanced safety, I have
()
13 been used for worst things.
That is, again, what I think 1
14 Bill said.
The pilot program, if all of a sudden there's a 15 30 percent cut back in O&M funds for utilities, how do you 16 consider that.
Obviously, that is something that is going 17 to be very, very difficult to come up with a criteria for.
18 Again, we have taken on two impossible tasks in the past and 19 told that they were impossible.
We are going to try it 20 again.
21' You don't quit trying, I guess is what it comes 22 down to.
How big is that part of resources; is it 23 artificially small because of internal manipulations, 24 whatever prescures from outside the utility.
That is the 25 difficult part.
Maybe when you focus in on an outage by
36 1
outage basis and don't to solve a problem forever it may be
[,_>\\
\\-
2 doable.
Establishing the criteria is going to be extremely, 1
3 extremely difficu]t.
4 (Slide.)
f 5
Let me switch now to scheduling and control of j
l 6
inspections.
I think if you weighed the comments in the 7
NUREG at each level there was the great mass of them seemed I
O to deal with the inspection program and the impact of
~
9 inspections, in particular team inspections.
We made an l
10 effort and I think quite successfully as the data is 11 starting to bear us out, and we are continuing to refine it l
12 to inspection manual chapter 2515 which addresses reactor ly( )
13 inspection program, to not change the substance of the l
l 14 inspection program about two and one-half years ago, but 15 witt* the incorporation of INE into NRR, I will say to i
16 translate part of the decision and planning process from l
17 what was INE headquarters to the regions.
l l
18 Regional administrators say we did this anyway and 19=
I think in many cases they did, but this was to formalize it 20 and say we believe in doing it this way.
As a resul' we 21 came up with our initially one-third, one-third, one-third 22 program where we have approximately one-third of the 23 inspection services dedicated -- I will tell you the truth I
('~S 24 actually, over the last two years it has come out to be 1
\\,!
25 about 45 percent of the inspection program -- going to what
37 I
we will call the core program.
The core program, the way
~s 2
it's defined in the overall programmatic documen':s is that 3
minimum level we want to at every plant in the country 4
independent of the plant's performance.
5 It le that level of program that we would consider 6
to be insightful enough to tell us if we need to do more or 7
that our general appraisal of the facility and its 8
performance is correct and is a good performer.
That's one-9 third of the program.
10 In general when I look at the demand power loading 11 of the facilities-and how the resources are being spent, we 12 span from anywhere from slightly less than 2,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> for a
/~N
(
)
13 12 month period at our better performer per unit to over 14 10,000 units that would be obvious, Browns Ferry, which 15 happens to be right up there at the top of the list, right 16 up there at 10,000 for unique reasons.
There really is a 17 spectrum.
It's not like there's a few at the bottom and a 18 few at the top.
It spreads right across and builds.
L 19 We have approximately 25 percent of the plants, 20 when you look at the hours at those facilities, there are l
l 21 only slightly more than just the residents assigned.
When I l'
22 look at the real data I know that our better performers, we 23 are in fact doing less than at what would be considered as l
g 24 our performers who aren't as good.
{J 25 MR. CARROLL:
How do I interpret what you are
r 38 1
saying?
You have deleted from it things like team
~~
\\
\\
\\-
2 inspections for maintenance and that sort of thing.
3 MR. GILLESPIE:
Yes.
4 MR. CARROLL:
You are just talking about the basic 5
inspection.
6 MR. GILLESPIE:
The basic inspection program is, 7
in fact, pretty much what is being carried out at about 8
one-fourth or one-fifth of our facilities.
It's a one-9 third, one-third, one-third program.
There is the second 10 third.
The second third we have, in fact, not used up one-11 third of the resources.
That's special temporary 12 instructions for safety issues.
rN l- (_)
13 In fact, if you lookqd at the statistics from the 14 old INE days and the statistics today, you will see that we 15 have significantly clamped down on any kind of-proliferation 16-of special type inspections.
What has come out of it are 17 the ones that have been highly visible, the maintenance team 18 inspection teams fall into this category, the emergency.
l 19 operating procedure inspections fall in this category, where 20 there was clear decision to do this on a national basis.
21 When we make a decision to do them on a national 22 basis in that third of the program, they are done 23 everyplace, and that is additive to the core.
We are now i
N 24 working on what I will call the next set of inspections, and (O
25 right now that is electrical system inspections.
39 Now there's another third, whicn is --
1 2
MR. KERR:
When you inspect, you inspect to see if 3
regulations have been followed or you go beyond that?
~
4 MR. GILLESPIE:
We are inspecting for safety and
~
5 not only compliance.
But you can't lose sight of I
6 compliance.
E 7
MR. KERR:
For example, I don't know how you 8
inspect emergency operating procedures because I don't know
[
9 of any regulations that deal with those other than you have j
10 to have them.
That's another subject, and I don't want --
11 MR. GILLESPIE:
The how of doing it?
12 MR. KERR:
What do you inspect against and what
()
13 are your criteria?
Don't try to undertake that here.
14 MR. GILLESPIE:
I won't.
That's the second third.
15 The last third is -- and I don't differentiate on 16 discretionary resources.
Whlie we call them discretionary, 17 they are really reactive resources which are the regional
=
18 administrators reaction every third.
We make very stringent 19 efforts to try to preserve that third for each region.
That 20 third we fully expect to be distributed through the 21 facilities.
22 That is SSFI's, SSOMI's, OSTI's, PRA inspections, 23-follow up to maintenance, it is what the regional 24 administrator feels he needs at which facilities he feels he 25 needs it at.
Included in that though to a large extent is i
_ ____..-i..----
40 1
headquarters inspections.
Headquarters inspections have 2
significantly declineid.
I think for the most part you will 3
find many team inspections and the responsibility for major 4
team inspections having been fully translated to the 5
regions.
6 In many cases you will find the headquarters 7
people participating, even if they are participating as a 8
team leader, it is always with the -- not only the regional 9
concurrence but the region's desire.
That's how we have 10 worked for about the last year, and the intent was really to 11 focus that extra one-third and those team inspections within 12 the regions.
13 MR. WARD:
Frank, I didn't understand what you 14 said.
You said this third was regional discretionary, but 15 also headquarters inspection?
16 MR. GILLESPIE:
Headquarters inspections have 17 really been translated to regional inspections.
The 18 headquarters group has grown into really a developmental 19 group.
They have led the effort on the ISLOCA audits, they 20 are looking at shutdown risk.
In the presentation after 21 this on level of detail, their expertise in design has been 22 brougnt to bear on that.
23 The headquarters effort is basically 24 indistinguishable from the regional effort now.
We don't go 25 out and do lots of unilateral headquarters inspections.
I
A l
l l
41 1
am only saying that because it came up earlier when Tom was 2
talking, someone said does that include headquarters 3
inspections.
Basically in that extra one-third the answer 4
is yes.
5 MR. MURLEY:
Don't forget vendor inspections, 6
which tend to be reactive.
There's also, we tend to lump 7
RER's in this and tend to lump some recall exam work in this 8
too.
9 MR. GILLESPIE:
Go ahead, Bill.
10 MR. KERR:
If, perish the thought, the regional 11 administrator did not spend his discretionary third, is he 12 docked for that?
13 MR. GILLESPIE:
I will tell you, the regions 14 continually come in with letters saying we don't give them 45 enough resources.
I guess that could happen.
It hasn't 16 happened yet.
17 MR. KERR:
That's enough.
18 MR. GILLESPIE:
The basic program is one-third, 19 one-third, one-third.
The middle third hasn't come up that 20 high, because the core program is basically consuming about 21 45 percent.
The national kind of program really does 22 consume about 45 to 50 percent of the program.
23 MR. CARROLL:
A very good performer, if he just 24 happened to be -- let's say after a year happened to be 25 getting some of these special inspections.ike maintenance
42 1
team inspections --
s' 2
MR. GILLESPIE:
He would have a peak number.
We 3
look at that, as a matter of fact.
4 MR. CARROLL:
He could have one heck of a big 5-number compared to an average performer in the same region.
6 MR. GILLESPIE:
Yes.
We ask that.
We now look at 7
that data in headquarters and do a comparative within 8
regions between plants between good performers and bad 9
performers in and across regions.
When we see a peak in a 10 number of hours and we understand the general impression of 11 that plant and look at his SALP score and see a peak in 12 those areas, we pick up the phone and say hey, how come?
/~N
(
)
13 Usually what we will find is, there has been a maintenance 14 team inspection or there's been an RER, or there's been --
15 we ask the question.
It's not just done without being 16 noticed.
17 We systematically ask the question.
18 MR. CARROLL:
You do now have a system where you 19 are trying to track these different categories?
20 MR. GILLESPIE:
Now that I have said we have --
21 what the philosophical structure is -- what we realized at 22 the time was as you translate to the regions this scheduling 23 responsibility, you do need a skeleton to hang it on.
(S 24 That's where the master inspection program had its roots,
. \\.J 25 was to provide a skeleton and visibility.
Visibility, so
43 1
that people in headquarters could see it without calling a
[
)
\\
~/
2 region and bugging them and it was a structure.
It was a 3
way to structure the philosophy that we had written in the 4
manuals.
5 Within this -- and Tom is right.
Tom made a trip 6
to Salem and Oyster Creek or Salem, and he came back and he 7
said they got hit with six inspections or something.
In 8
five minutes I went down the hall and pulled it up on the 9
screen, and he was absolutely correct.
They did get hit 10 with six inspections.
So, we have started looking at that 11 and we do have a system which will now display it.
12 The regions are in the process -- they aren't all
,()
13 there -- they are in the process of putting it in place and 1
1 14 maintaining it.
One of the things that we are proposing --
15 this gets me from sheer numbers to scheduling -- is to make 16 those team inspections visible early when they come up on 17 the planning horizon.
We have the-structure.
The structure i.
f 18 can give us reports.
We can review the reports and have, in 19 fact, started doing that, to see how the inspections are 20 scheduled.
21 Now what we are proposing is to actually publish 22 that schedule, publish it and let the utilities at least 23 have an opportunity to then respond at the earliest point 24 when they are scheduled.
This would also tie back into
(}
LJ 25 program itself.
Another programmatic change was quarterly
44 1
meetings within each region.
There are quarterly meetings
-,-s
)
2 within each region where all the facilities within the 3
regions are reviewed, and the allocation of resour:es is 4
reviewed for all the facilities against each other to ask 5
the question if we do this here we have to not do this 6
here.
7 What it is, it's more than just a wish list.
It 8
focuses the regions on they have "x" number of inspectors,
's and if you are going to do something someplace else you have 10 to cancer it out.
Now we have a ntructure, a philosophy, a 11 way to display it and quarterly updates.
We are looking at 12 it in headquarters to see that it is, in fact, being n(,)
13 implemented.
I talked to Tim Martin in Region I, and he 14 feels that he will be fully implemented by about the end of 15 February.
In Region II, III and IV, and I haven't talked to 16 Region V in the last couple of weeks, they are well on the 17 way.
18 Scheduling.
What we found was -- what Tom found 19 when he went out there and was absolutely right -- is, 20 without an integrated display of what was going on, a site-21 specific display we had at a facility a maintenance team 22 inspection planned by one branch and conductcd, an EOP 23 inspection planned and conducted -- this was all within a 24 six month period -- a recall inspection.
There was a major 25 EP exercise which we sent a team to observe.
There was a 1
45 1
major INPO visit.
Part of our site-specific plan has major i
i 2
licensee evolutions like outages and major INPO visits on 3
it, so we can schedule around it.
4 MR. CARROLL:
Your regional people or certainly 5
the residents of this hypothetical plant knew that --
6 MR. GILLESPIE:
It wasn't hypothetical.
I 7
MR. CARROLL:
Kne.
?. hat was going on.
It seems 8
they don't have the right attitude towards life if they were 9
not complaining about it or trying to do something about it.
10 MR. GILLESPIE:
It wasn't a complaint for them, it 11 was part of their day-to-day business.
It's how we had done 12 business, it was a calibration point from a regulatory
("'y
( j 13 impact survey that said we were on the right track and now 14 we had to implement it.
I will be honest.
Traditionally, 15 engineers view keeping data systems and putting in schedules 16 as a pain.
We are putting a new burden on them.
Now the 17 need for that burden, the need for having an integrated site 18 plan on what we intend to do needs to be at least quarterly 19 updated and have it fairly reflect what you intend to do --
20 it has been brought home by the survey.
21 Our proposal in this area was two-fold.
One was 22 to basically put a speed limit -- which I think everyone 23 read in either Energy Daily or Nucleonics Weekly raised much
(~~').
24 controversy as to what we were doing.
One was to put a i
/
25 speed limit on the number of inspections per SALP cycle team
46 1
inspections, and to ensure that those schedules were 2
somewhat equitably distributed to the SALP cycle so that in 3
six months you weren't getting -- and we proposed four --
4 that you weren't getting all four.
That was to help with 5
the scheduling part of it.
6 The other piece was to publish the schedule so 7
that everyone has a chance -- every utility would have a 8
chance to come in and lobby its region on why this couldn't 9
be done.
10 MR. CARROLL:
The fact that you had to do all of 11 that suggests to me that you have a bunch of people out 12 there in the regions that require some re-education as to F'
()
13 what an effective regulatory program is.
They don't 14 understand the concept that too much regulation can really 15 be counterproductive to safety.
16 MR. GILLESPIE:
I think it was --
0
'17 MR. KERR:
I would put the question in a slightly 18 different context.
Is there somewhere in the staff, a group 19 or person who worries about things like this?
It seems to r:
20 me that in an organization this large --
21 MR. GILLESPIE:
I do.
22 MR. KERR:
It would not have taken a survey --
23 MR. GILLESPIE:
It didn't.
Believe me, the master 24 inspection program started -- it took us two years from when q
25 we first rewrote the manual chapter itself on the
47 philosophy, it has taken us LWo years to develop a tool or 1
2 skeleton to take all this input and display it simply as a 3
site plan.
4 MR. KERR:
What I was concerned about was that 5
sort of got the impression that nobody ever thought about 6
this before until the survey was taken.
7 MR. GILLESPIE:
No.
8 MR. KERR:
I feel better if --
9 MR. GILLESPIE:
This has been about -- like I
(
10 said, it's been two and one-half years since Cim Sniezik 11 said I want it one-third, one-third, one-third, and we kind 12 of wrote the philosophy down of what we intended.
Almost in
()
13 parallel with that, i hen started the development of the 14 structure or skeleto.
. the information system.
I will be y
15 honest with you, we had some false starts on-how to do it.
16 Sometimes we over complicated it and in other cases it 17 looked like we were under complicating it.
18 We seem to have something that is in place in 19 several regions now and is working fairly well.
Some of the 20 comments back from the regions are causing us to -- we had 21 something that spit out site-specific inspection plans, but 22 what it didn't give -- for example for and engineering 23 section chief --the section chief tends to be the key to the 24 whole program.
It didn't give him a list of all the people 25 in the section and what their schedules were.
What he iimim
.is i
nsiisi uni _msasei neiimm-um-
_ _ _ i i_mn i ---
48 1
had to do was get a list of what facility is person was 2
. going to.
3 So, we are really trying to human factor the system so that our first line supervisors will have an 4
5 easier job both scheduling their people so it will be 6
convenient -- they are trying to do a good job.
It will 7
also then roll up for the projects group so that they get --
8 they are interested in it by facility, and they get a site-9 specific plan without generating lots of paper.
10 The other problem we had is in regions that are 11 highly reactive, is lots of things would be cancelled and 12 rescheduled.
When a section chief would cancer and
}
()
13 reschedule something it may still stay on the list every 7
14 place.-
That's where the idea of a quarterly meeting to look across what is being done -- in many cases now when an 15 16 optional procedure, an optional and sensitive falls into the 17 prerogative of the region to have chosen to do it -- doesn't 18 get done, headquarters isn't beating on them for doing it.
19 They are making that safety decision.
I didn't get to do 20 this; do I want to reschedule it or is something more 21 important at a different facility.
22 MR. MURLEY:
Frank, one of Jay's comments about 23 the regional views.
I think we in headquarters have to 24 share -- a number of these team inspections originated in 25 headquarters.
The fact is, when a division director in
-m 49 1
headquarters says I want to go to that plant and do this 2
inspection, the regions didn't feel they could turn him 3
down.
It's a matter -- Bill is right -- it didn't take the 4
survey to tell us this, but it did kind of focus our 5
attention on it.
6 All we are saying is that we now are going to have 7
a discipline system for controlling these activities.
We 8
are going to have to move on, Frank.
9 MR. GILLESPIE:
The third area, inspector J
10 professionalism.
This stems -- again, this is not something 11 we didn't suspect.
This did cause us to focus on it a whole 12 lot more.
The workforce of inspectors that we have out
()
13 there is changing.
We are bringing many, many more junior 14 people in.
I believe everyone has heard stories at one trip 15 or another at one facility or another, at one region or 16 another on how soon resident inspectors before they were 17 long term NRC employees, in some cases we had residents going out with less than a year experience with the Agency.
18 19 MR. CARROLL:
Did I see in the little weekly
(
20 newsletter somebody was hired in as a senior resident r
21 inspector?
That surprised me.
22 MR. GILLESPIE:
I didn't see that newsletter.
23 MR. CARROLL:
That's happened?
24 MR. GILLESPIE:
I don't want to say it hasn't 25 happened.
If you read it in a regional newsletter, it
i:
50 1
highly likely happened.
It would depend on the individual
,q I
2 and is he a rehire.
Was he with us before and left for 3
three or four years.
4 MR. CARROLL:
Yes.
5 MR. GILLESPIE:
Did he go to DOE for a couple of 6
years and decide to come back?
7 MR. CARROLL:
That's okay.
8 MR. GILLESPIE:
It's difficult to deal with 9.
anecdotal information because it always seems when you dig 10 into it there's a little more there.
11 In looking at our expectations out in the realons, 12 we have not really stepped back and recognized -- I will say O) 13 the process and system that we have created with putting
(
14 residents at sites and having section chiefs back in the 15 region supervising.
Basically, a supervisor who is trying 16 to deal with a. staff that is geographically very, very 17 dispersed.
We recognize that we had two things.
18 We had somewhat of a less experienced possibly 19 staff out there, and we had a very difficult supervision 20 process.
One of the things we want to do is both relook at 21 what we expect of our section chiefs -- this may include 22 needing to put more resources in the region to provide 23 adequate supervision.
We want to look at the interfaces
['T 24 between-our regional managers and the utilities.
If you
%-)
25 notice in reading the report, all the comments seem to go to
51 1
the residents of the regional administrator.
You may ask, 2
what happened to all of those people in the middle.
_\\
3 What are our expectations of section chiefs, 4
branch chiefs, division directors when they go to a 5
facility; do we expect them to take 20 minutes or one-half 6
hour to meet with what might be their counterpart and say 7
how is the operation under me working for you.
Also then, 8
if they do get some feedback, to get back to that person.
I 9
mean, to be credible and not just get around the concept of i
10 retribution for credible and factual feedback on how the system is working.
11 12 What is the section chief expected to do relative 13 to supervising this dispersed group of people that work for 14 him?
Generally there is one engineer back in the region 15 that works with the section chief, and he could have four to 16 six units with maybe eight to 12 residents -- very senior 17 residents, resident inspectors working for them.
How does 18 he ensure that they are carrying out the program as it is 19 intended to be carried out and is it intended to be carried 20 out without ratchoting by the resident inspector.
He should 21 be calling back.
p 22 How do we get that feedback, how do we know to get 23 that feedback.
We are looking at the basically -- the 24 position descriptions.
Have we asked these people in the 25 regions to do the right things for the roles that we have
52 1
given them?
Has it changed as we have changed the program?
2 Also, look at their performance elements and standards.
3 Generally, people will perform to those things they are 4
getting graded on.
5 If these aspects of interaction with the licensee 6
and interaction with the resident aren't in there, which 7
generally they aren't, then we are not causing these people 8
to do really what we want them to do.
We haven't 9
institutionalized the process that we are really desirous i
10 of.
An example would be that my guys and I have been going i
11 out to some units, and one of the questions that we always 12 ask is, when your section chief comes out what does he do?
()
13 Does he spend time with you, does he come out and spend a 14 couple of days seeing how you are doing the job and attend i
15 your exits?
Or, is he focusing on technical problems and 16 the compliance problem, or the last reply from the licensee?
y 17 We have a mixed bag.
Some section chiefs went out I
18 and they were on compliance, they were on the last response, 19 they were on dealing with the licensee.
They were very 20-supportive of everything their resident did.
In some cases 21 we do have section chiefs who are going out and spending one 22 and two days with their resident, so you get a mixed bag.
A 23 mixed bag gives you mixed comments, but it's also something 24 we need to correct and institutionalize it so that we cut 25 down that fraction that is happening.
53 p
1 That's in the structural role.
In the training 2
role, because we are getting people who are fresher off the 3
street, the inspection force or number of gray hairs in the 4
inspection force is significantly down.
We are looking at a 5
more balanced training effort.
While we feel comfortable 6
with the technical training we are giving the inspectors --
7 in fact for these newer people, we have instituted with TTC 8
extra training courses that they will take before they get 9
to 13ght now what is called the full series technology 10 courses, to give them a better chance of doing well in the
'll full series courses.
12' While we are fairly pleased with the technical 13 training, we are relooking at have we left as optional the 14 professional training, the interpersonal aspects?
How to be 15 a team leader, we are asking the question and our intention 16 would be to establish team leader positions in the regions 17 where you select people that have the attributes as team 18 leaders, who have the ability and personal skills to control 19 the diverse group of professionals that they are out with to 20 control --
21 MR. LEWIS:
You just tickled something in me with 22 what you just said.
You used the word " professional" and 23 then you said interpersonal.
Is that your definition of 24 professionalism?
25 MR. GILLESPIE:
Professionalism has those aspects.
1
54 i
1 There's a technical aspect and then there's the softer side, 2
the human -- how you carry out your business and how you 3
carry out your business, how you come across to people.
4 MR. LEWIS:
I see.
5 MR. GILLESPIE:
It's both.
What I am saying is --
.S 6
MR. LEWIS:
That's very interesting.
7 Professionalism to me means something entirely different.
8 So,-now I understand some of your viewgraphs that I didn't 9
understand before.
t 10 MR. GILLESPIE:
What I am saying is, we feel i
11 fairly comfortable with the technical aspects of what we 12 have people out there doing.
13 MR. LEWIS:
My problem is that some of your i
14-viewgraphs which use the word " professionalism" that went on 15 to deal.with things that didn't mean professionalism to me.
16 But with the construction that you put on it, I understand 17 it.
Professionalism, to me, is very simply that you know 18 your job extremely well and are capable of doing it 19 extremely well.
You are a professional.
20 Technically, of course, it means only that you 21 make money out of it.
That isn't what we normally mean by 22 professionalism.
23 MR. GILLESPIE:
Let me agree.
There are two i
24 aspects of the job.
I think our people do know their job 25 extremely well, and I have a feeling that they are doing
L 55 1
exactly what we have written down and have asked them to do 2
which is why we have to go back and examine:what we have 3
asked them to do and make sure we have asked them to do the 4
right thing.
Now, if we change what we have asked them to 5
do we have to provide them with the background now to do it P
([
6
'or ensure that they have it.
7 MR. LEWIS:
This is an important' issue, just what 8
you expect from people.
For example, I don't know -- you 9
are too young to remember Charlie Chaplin's movie Modern-t:
10-Times but I'm not.
In that movie Charlie Chaplin played a
>z l
11 guy on a production line who did a vety good job of
}
N 12 tightening something in the production line.
I am sure he 7a y
ll 13' did it very well, but I wouldn't have called him a 14 professional even though he did make money at it, but not j-15 very much'because he wasn't allowed to.
f
'.16 That isn't what I mean by professionalism.
- Yet, g
17 you can train somebody to do that very well.
j[
18 MR. SHEWMON:
A professional usually has to
_' ;=
-19 interact with people.
In fact, I can't think of any E
E 20 professional which is not involved in interaction with
.a_
21 people.
So, there's partly a matter of how effective they j[ i 22 are there, there's partly a matter of the integrity.
The L
23 integrity may not correlate exactly with expertise as you 24
-defined it earlier.
r 25 MR. SIESS:
Again, you are using the right words, i
M m
56 1-effectiveness, integrity, et cetera, none of which are 2
connoted by the word professional.
You have to find a 3
better word for it.
It doesn't mean what you are making it 4
mean, and you just can't do that.
You are confusing people.
5 MR. LEWIS:
That's what was troubling me.
I Vill 6
argue with you later, Paul.
I think there is such a thing 7
as a professional mathematician.
8 MR. SIESS:
I am a professional engineer, but that 9
means I'm licensed by the State of Illinois and doesn't mean 10 anything else.
It doesn't say.anything about my 11 interpersonal relationships which are lousy.
12 AR. GILLESPIE:
That's not true, Chet.
We have
(
13-been going back and forth since the late 1970's.
14 MR.-KERR:
He doesn't really think that.
15 MR. LEWIS:
I apologize to the audience for having 16 brought up the subject.
17' MR. GILLESPIE:
I think it's our view of a 18 professional inspector does include interpersonal 19 relationships, interpersonal skills, how he carries out his 20 business.
Because the State of Illinois --
21 MR. SIESS:
Even an amateur inspector though would 22-need those things.
23 MR. GILLESPIE:
We don't have amateur inspectors.
24 MR. SIESS:
You could argue that a professional 25 inspector by one definition of professional wouldn't need l
57 1
all those detailed guidelines you write out for him.
If 2
he's a_professiona? he goes out and does the job that his 3
professional' expertise tells him should be done.
You have 4-written instructions there that you could take a recent 5
graduate out and have him do it or maybe even a non-6 professional.
7 MR. GILLESPIE:
I think it's been shown that our 8
recent graduates can't take our written instructions and
-9 just do it.
10 In summary, we are going to try to hit two 11 aspects.- What we expect of the people including the whole 12 supervisory chain, and then after looking at what we expect h
13 of them, have we given them the necessary tools to carry 14 that out.
If it makes my viewgraphs clearer, in 15 professionalism for inspectors we have included the need for 16 interpersonal skills, the skills to lead a meeting, the 17 skills to cope with a meeting under very, very adversarial 18 circumstances and to control a' team who is at the meeting 19 with you.
20~
In some cases that's the ability to say we need a 21 five minute break to caucus and walk out of tho meeting and 22 take a five minute break.
Much of this is being done right 23 now, but it's on a voluntary basis.
We are looking at the 24 possibility of making some of these things mandatory.
The 25 example would be the IIT team leader training for team
58 1
. leaders.which involves. video taping people in adversarial
, _, 'l-
?' /
2 type of meetings and then going through it and critiquing 3
it.
I mean,.actually trying to improve someone's skills and 4
not just a lecture kind of basis.
5 There has been a lot of success with that with the 6
IIT team leaders, which speaks both to the training and to 7
the qualifications of the people.
8 MR..KERR:
Frank, while you are talking about 9
interpersonal relations which I personally think are 10.
Important, you didn't say anything about attitude.
I don't 11 think this is typical of people at the levels of management 12 that we are talking about today.
At the inspector level, my
?N j, j 13-experience would indicate that there is a significant 14 fraction of inspectors who are convinced that were it not 15 for their presence and activities licensees would do all 16 sorts of criminal activities; and that, only their energetic L
17 surveillance prevents things from coming completely apart.
18 They really believe that.
I think that's -- I l
19-personally think that's unfortunate, and if I were in your 20 position I would want to try to convince people that that's
[
21-not necessarily the case.
It's clear to the licensee when i
22 this attitude -- it is irritating and it's 23 counterproductive.
Furthermore, it's nos so.
("')
'24 MR. GILLESPIE:
Part of -- on the viewgraph that
' (s,/
25 you have titled enhanced training -- something that we are l
t
59-1 going to add that is on there, refine fundamentals of 7,_T-
- k. l 2
inspection course we give them in the beginning.
Something 3'
-that we really need to add to attack that, and that's the 4
add refresher.
What we are realizing now is that not only 5
do you say that the first time in that first three day f
6 course before someone goes out, it's the need to continually 7'
drive it home.
8 Independent of how long a guy has been out there 9
and he comes in and says that I have heard this before, what 10 we are proposing now is that we will continually drive it 11
.home and formalize the need to go through it over'and over 12 again to reinforce it.
We do recognize as part of the 4
/"N
(,)
13 training is also what might be more education.
We 14 continually have to keep the education current and drive it 15.
home.
16
.If we are successful -- and it's a cultural change 17 now also that we are going on and it's long term.
It's not
.18-something that is going to happen, I don't think, in just 19
.six months.
If we are successful in orienting the section 20 chief and the first line supervisors then I think we can be
'21 successful, because I think a lot of inspectors react to the 22 expectations of their supervisors.
23 I think there are very few inspectors out there, g)T particularly the newer people that are coming in, they are 24
\\_
t 25 taught what is inspected.
The biggest influence on them is
60 1-the influence of their first line supervisor, and probably
~
2 the person above him is the person they interface with the 3
most.
It's a dual item.
The other thing we have recognized 4
is the role of the senior resider.t has to be more clearly 5
defined.
A senior resident at a multi-unit site, for 6
example at Millstone which has three units -- there are four 7
residents there.
8 The= senior resident right now is a pseudo-9 supervisor.
The residents'-- he writes their appraisal, yet 10 all the residents really work for the section chief.
We are 11 recognizing more and more that the senior resident's 12 leadership role has to also be addressed.
When we look at 13 what we expect of him, we have to also look at his 14 supervisory role.
15 This is going to be a multi-faceted kind of review 16 and chtige.
Simple to do the paper but difficult to change 17 the attitude.
I can go in and we can review and rewrite 18 elements and standards and review and rewrite position 19 descriptions, but we are looking at years to change attitude 20 and culture.
But you have to start someplace, and we have 21 had it emphasized to us the need to change.
This is very 22 much compounded by the junior level people that -- I know 23 the regions are hiring in intern and entry level people just 24 as we are here.
A proportion of entry level -- I mean real 25 recent graduates -- is I think significantly higher this
61 1
year than it was ten years ago.
2 Yet, those of us who were in the program and in
-3 the regions ten years ago you tend to think that it's the 4
same people.
Where a lot of the people we had in the 5
regions were four to six years of Navy experience and coming 6
in, and as critical as people are of those who drive 7
submarines -- it was four to six years of fairly intensive 8
experience and education.
You came with that, and you have 9.
been kicked around for six years.
That's different than 10 someone who has no experience or one year experience and 11 comes directly to us.
12 The way it is looking now'is, we are going to be h
13 approaching 25 to 30 percent of our people being'new people 14 right out of school.
I have to look at the mix of people I' 15 have coming in the pipeline now and adjust my whole training 16 concept and have multiple levels.
I need one for the 17-experienced guy that I'm still going after to bring in, but 18 then I have an obligation to the junior person to get him 19 along and get him to that same level.
20-That's an obligation we are now taking on.
21 Personnel is writing a paper to the Commission on this which 22 should hit the Commission in a couple of weeks or maybe a 23 month.
24 MR. CARROLL:
In thinking about how to change the 25 culture and what kind of training to give people and so
62 j~3 1
forth, have you involved the Human Factors people?
\\/
2' MR. GILLESPIE:
We have started, yes.
In fact it starting 3
was Jack Rowe's people who gave us -- we ar e 4
hopefully with an overview and just call it a white paper of.
5 philosophy and what is intended and expected.
Jack gave us
.6.
it must have been about 20 page paper on it.
That's
-- gee, 7
going to be our starting point.
Yes, we are going to 8
iterate back and forth.
9' We just realize now that we have to do it.
We 10 don't have an obvious expert at it.
-11
~MR.
CARROLL:
Have you asked the Human Factors 12 people'whether they are --
eq q )-
13 MR. GILLESPIE: 'We have.
What you are saying is, 14 it's an expert in taking the relatively novice person and 15 trying to as expeditiously as possible, turn him into a 16 journeyman level person.
At the same time, taking the 17 people who may have developed attitudes or not quite exactly 18 what we wanted and re-educating and starting them over.
We 19 want to start'with a higher level document, and we are going l-20
.to be working on that higher level document first.
21 The Human Factors are from the beginning, deeply 22 involved in it.
23 MR. WILKINS:
Let me comment on Bill Kerr's b(\\
24 observation.
You don't want to throw the baby out with the 25 bath water.
l l
63 MR. GILLESPIE:
Right.
O
.1 2
MR. WILKINS:
Every human being has a need to be 3
needed.
The inspectors need to feel that what they are 4
doing is-important and relevant.
Use Hal Lewis' Charlie 5.
Chaplin example, and ask 'him what is his job.
Is his job 6
tightening bolts or is his job building an airplane.
You 7
would rather he say building an airplane than tightening 8
bolts.
There's another example-and that's the Hanes 9
television commercial for Hanes.
This lady says it doesn't-10 say Hanes until I say it says Hanes.
11 You really do want your inspectors to have that 12 kind of gun ho at it.
The problem is to get it without
()
13-having them feel the adversarial attitude.
14 MR. GILLESPIE:
It's to get it and do it smoothly.
15 MR. WILKINS:
Yes.
And that's the interpersonal 16 relationship kind of thing.
17 MR. GILLESPIE:
Right. I think they have it.
I 18 think the inspectors have been enthusiastically carrying out 19 the program that we have asked them to carry out.
What we 20 are saying now is that we need to step back and look at what 21 we have asked them to do.
I don't think the technical 22 substance is different.
Like when we reoriented the 23 program, we didn't go rewriting all the inspection 24 procedures.
What we did was, we said was that we trust the 25 regions in the decisional part of it.
I
}
i l
64 f
1 I don't want to throw anything out.
I think what L
,2,
\\ -}'
JJ t
l 2
we are recognizing is, we are tweaking the structure.
We l
3.
~ want them to be just as enthusiastic.
We are not trying to i
4 dampen their enthusiasm.
We would'like them to express it.
5 MR. MURLEY:
Can I interrupt?
That point I am 4
-6 glad you made it.
You made it better than we have.
To'me, 17.
that is the challenge that we have in management, is to 8
instill that proper mix in the new people that we are hiring 1
l 9
and training and sending out there.
They do have to be
'l
-10
'important in what they are doing.because it is an important t
11-job what they are doing.
j 12 Yet, we don't want them to develop this
!q 13 adversarial nature or view that Bill Kerr mentioned, that.
j 14 there probably are some inspectors out there that believe 15 licensees are basically untrustworthy.
We don't think.that, j
16 and we have-to develop this program and this training that i
17 allows inspectors to walk the proper line.
l
'18 -
MR. GILLESPIE:
Again, it's a mix.
We have some 19 real good examples out there of where it's been carried --
j i
~20 the program has been carried out just superbly.
I think we
-21 are looking at an increment.
22 MR. MURLEY:
That concludes our presentation.
23 There is one point that I would like Bill Russell to address
('y 24.
so that we don't leave it hanging. I think Bill Kerr asked l
\\~J 25 how do we do an EOP inspection, what criteria do we use.
m 65 1
MR. RUSSELL:
Let me start cut that the basic 2
requirement. flows from TMI requirements that were imposed 3
ofter the TMI accident.
The framework was that utilities E
4 were required to develop emergency procedure guidelines
(
5 programmatically that would describe everything from'the f
6 content of the procedures to how they train pecple on the 1
7 procedures, et cetera.
i 8
That was then used to develop the inspection 9
procedures.
The inspection procedures looked at such things
[
10' as differences between the approved generic technical 11 guidelines for a facility which were based upon an analysis 12 of accidents and transients, the symptom-based approach and
()
13 whether the licensee had adequately identified that 14 difference on a plant-specific basis to how validation of
.15 the procedures was conducted on the simulator with respect 16 to Human ~ Factors elements, the procedure itself, to the 17 actual performance of individuals in carrying out the 18 procedures which was done on a simulator.
19 It was a combination of a technical review and a 20 performance review of observing people performing using P
21 those procedures.
That flowed from the TMI requirements 22 which were imposed with 0737 Supplement 1 and were confirmed 23 by orders.
That's the regulatory base for those procedures 24 and the process that was followed.
Your observation that 25 there was not a specific regulation other than saying you O h
66 j
'1 have to have procedures for the conduct of operation of a i
)
i'- '
2 facility, even the particular procedures you have to have a 3
regulatory guide.
Your observation was correct, and we 4
relied on.the programmatic requirements from 0737 Supplement 5
1 which were TMI requirements.
6 MR. MURLEY:
Mr. Chairman, that concludes our 7
presentation.
We will be glad to answer any other questions 8
that the committee may have.
9 MR. MICHELSON:
Are there any other questions?
10 MR. KERR:
Tom, I'm not sure it's related to this 11 discussion but it is something that I have observed 12 recently, and if it isn't you can tell me so.
In a recent I
n
( J-13 discussion of the review of advanced plants or semi-advanced 14 plants, the review of.non-safety systems is still making use L
'15 of the very early regulation that says the only thing you 1
16 have to worry about -- maybe I am oversimplifying about non-l 17-safety systems --is that their failure shall not disable a 18 safety system.
19 On the other hand, we now have performance 20 indicators which are presumably related to plant safety 21 which, among other things, count the number of automatic 22 scrams.
We certainly have a good bit of experience 23 indicating that the performance of non-safety systems can
("]
24 increase the number of scrams and indeed it is responsible
\\~J 25 for a significant fraction of automatic scrams.
67 1
It seems to me that there is a disconnect-between 2
the review criteria that are being used and the criteria 3.
that are being used to judge the performance of operating 4
plants.
Do you understand what I am talking about?
5 MR. MURLEY:
Yes.
We intend to take a broader 6
view of safety and risk in our reviews of future plants to 7
the point where, if a system can affect the bottom line risk 8
numbers or the reliability of a safety system and we intend 9
to take a look at it.
However, we have to also be careful 10 that we-have to stay within our regulations too or at least 11 follow our regulations and our standard review plan.
12
-It may mean that when we find the systems that do I k 13' affect safety and risk and we don't have a good regulatory 14 basis for requiring something, then we will certainly have-to flag it to the Commission and tell them that we are going 16 beyond our regulations.
In some cases we may even have to 17 change our regulations or standard review plan.
18 I guess the answer is that we don't intend to 19 limit ourselves arbitrarily, if that was your major point.
20 MR. KERR:
It just seems to me that there is a 21 contradiction, and I don't know which is the better 22 procedure.
Maybe, indeed, there is no significant risk that 23 can be caused by malfunction of non-safety systems as long 24 as they don't disable safety systems.
On the other hand, as 25 a measure of risk of operating plants one is now using -- I
68 1
think, if I understand these operating parameter tabulations O
2 that we Lee now -- one is using the number of automatic 3
scrams and this is clearly not all together caused by the 4
. malfunction of non-safety systems but a significant fraction 5
of them are.
6 MR. RUSSELL:
If I could comment with an example.
7 I think-the example is the performance of the feedwater 8
systems which have historically been non-safety systems.
9 Yet, we find that they are frequent transient initiators and 10 can cause automatic scrams or upsets that, if you had some 11 other failure that occurred other than the simple loss of 12 feedwater transient you could have a more complicated event; 13 that is,' things don't happen one at a time.
14-Clearly, there is a desire to reduce the number of
'15 challenges to-the safety systems-to make them overall the 16 facility more reliable.
In that context we would be looking 17 at various potential transient-initiators.
It is in that.
18 context that we look at the failures that have occurred and 19 has the licensee addressed those failures, not from the 20-standpoint of using automatic scrams as an absolute 21 indicator but rather going beyond that seeing what caused it 22 and what action was taken by the licensee to address it.
23 MR. KERR:
My point Bill was, in the review of the 24 design no attention is given to the reliability of these 25 non-safety systems.
Yet, it is the lack of reliability that l
69
-1 leads to these aatomatic scrams.
2 MR. RUSSELL:
That was historically the process up
.3 until probably three or four years ago when we started 4
paying attention to what were the transient initiators and 5
started observing this from operating experience.
Dr.
6 Murley's point was that we expect to do more that, 7
particularly with systems for advance reactors to look at 8
.the reliability of those systems and precluding transients 9
Hso that you don't challenge the safety systems as 10 frequently.
11 MR. KERR:
My point was, and tell me if I'm wrong, 12' in the review of plant designs that consideration or that l
h 13 possibilityJis given no consideration. 'There isn't any look 14 at the potential reliability of these systems, a malfunction 15 otz which is likely to cause additional scrams.
16 MR. RUSSELL:
That has, in the-past, been the 17 case.
We believe with the performance of PRA's is a part of 18 a design basis which is required for new facilities; that wo 19 will have some idea of the sensitivity of various system 20 failures to transient initiators, their importance and 21 reliability.
We have not yet defined how to handle these 22 systems that are important to safety but not safety related 23 in the course of the review.
24 That was a major issue that our staff just 25 discussed this last week on Monday, as to how to handle that
70 1
conceptually in a review process.
What is intended by way of reliability assurance of these non-safety systems.
The 2.
best example that I can think of is passive plants that will 3
have on-site DC power but it won't be safety-related.
- Yet, 4
clearly, the ability to get to cold shutdown is going to be 5
dependent upon pumps and AC power and other things.
6' To what extent did we review the designs of the 7
8 on-site power systems, the off-site power systems, et 9
cetera, they clearly play some role in transient initiation.
10 Their reliability'is important to safety.
Yet, they are 11 proposed to be not safety-related.
That particular issue that you are identifying, we have to come to grips with how 12 f) 13-to do those reviews.
14 MR. KERR:
You are aware of the problem.
15 MR. RUSSELL:
We have not done that yet.
16 MR. KERR:
You are aware of the problem but don't 17 yet have a solution.
18 MR. RUSSELL:
That's correct.
19 MR. KERR:
Thank you.
20 MR. MICHELSON:
When you look at that problem you 21 need to look very carefully at certain kinds of systems like 22 instrumentation and control error systems, wherein the 23 concern might not be the reliability in the sense of losing air pressure totally but rather in the sense of losing air 24 25 pressure partially.
PRA's aren't much help in telling you
71 1
what is happening.then.
It takes some kind of deterministic j
\\
A-2
-examination on a plant-by-plant basis.
1 3
There is a number of LER's coming out from time to 4
time lately on these partial loss of air pressures and the 5
funny things they-are doing to safety-related equipment.
6 MR. RUSSELL:
Yes.
Probably the-area that.is i
I 7
going to be a significant review effort relates to the 8
control room itself.
They are proposing to use digital 9
systems with displays at a work station with the entire man-10 machine _ interface going through digital control.
If you 11 have a safety-related shutdown system you can do validation 12 verification, and you can ensure that that's reliable.
What i.
)
13 reliability do you'want to have-associated with the controls 14 that he is using to manipulate the rest of the plant and 15 what' kind of. review process should we apply to that.
16_
Clearly, current control room probably 80_ percent 17 or more-of the controls and things that are used are not 18 safety-related.
Yet, we have some understanding 19 historically of what has been their reliability.
We don't 20
.know what that is going to be when you go to an all digital 21 system from the standpoint of software.
Air systems, 22 support systems and their importance to transient initiation 23 are also areas that we are going to have to look at.
24 You notice that the issue that identified a 25 significant portion of the risk from the precursor study was
72 1
indeed that over cooling transient that occurred on 26
)
>: > ~
2 December 1985.
I remember it well.
I was acting office 3
director that day.
The day after Christmas nobody was 4
around.
It was quite an interesting event.
5 MR. CARROLL:
Speaking of events that I guess have 6
some relationship to this whole regulatory picture, this may 7
just be an analogous data point.
I was very concerned to 8
learn.that Vogtle had consistently been receiving number one 9
SALP ratings in the area of maintenance and had this long 10 history of problems with those diesel generator temperature 11-sensors.
It could have been really bad news.
l 12 Have you guys given any thought to what that data-1 l
f
( j.
13 point all means in.the overall picture?
l 14 MR. MURLEY:
No, not really, Jay.
We are, of 15 course, aware of it.
My guess is that when they were going 16 through the SALP review they probably did not have this 17 compilation of history on the emergency diesel temperature 18 sensor in front of them.
I wish I could promise you that we 19 always take into account every piece of information on a 20 plant when we do our SALP, but as'a practical matter we 21-don't always do it.
22 MR. CARROLL:
But a number one performer on 23 maintenance and SALP, in my view, i.t9t have had a good I
f' N 24 trending system for maintenance problems that were b
25 occurring.
It seems to me that the inspectors would have l
a 73 1
been looking at that-program and saying gee, 70 failures of f '/
y K-2 these sensors or whatever the number was.
3 MR. MURLEY:
That's a very good point.
I am sure j
4 the region and my staff are looking at that.
5 MR. MICHELSON:
Are there any other questions?
6
[No response.)
7 MR. MICHELSON:
If, we will take.a break until E
~8-10:45, at which time Marty Malsch will be here to tell us 9
about that Court decision.
10 (Brief recess.]
11-MR. MICHELSON:
Gentlemen, if we can get started 12 again.
Before we get to the subject of. level of design jw) 13
' detail, we had a little time left in our schedule since 14 there isn't a NUMARC participation on that subject.
I 15 wanted to have Marty Malsch fill us.in on this Court-16 decision so'we will be'more fully aware of what the 11 7
. implications-are if any.
18 MR. MALSCH :
I will be very brief.
On November,.
19 the U.S~.
Court of Appeals in the D.C.
Circuit decided to 20 petition for review which had been filed by a number of groups including.the Information and Resource Service of the 21-22 Commission's regulations in 10 CFR Part 52.
^
23 The challenge which had been brought had been 24 fairly broad and it focused on many parts of the Part 52 25 rule, but focusing specifically on the Commission's
74 1-authority to issue a combined license, the scope of review 2
and hearing prior to operation under a combined license, and 3
the Commission's authority to issue design certifications-4 under the Atomic Energy Act.
5 By in large, the rule was upheld as against these 6
broad challenges.
Let me read you or.e part of the 7
conclusion in the court's opinion which is by Judge Wahl, 8-the Chief Judge.
In responding to the industry's changing 9
knowledge and the public's changing needs, the NRC has 10 promulgated' bold and creative new regulations.
The 11 Commission's creativity, however, is constrained by 12 statutory authority._ Then they go on to describe, in one 13 respect thef find that the regulations fall short.
14' That is, in the regulation's failure to provide 15 for an opportunity for hearing and a Commission finding on 16 new information which has arisen since issuance of site 17 permits, design certifications, or combined licenses.
18' Under Part-52 we had confined the scope of review 19
'and hearing prior to operation to issues of conformance with 20 acceptance criteria in the combined license, and had said 21 that new information which pertains to the actual terms of 22 the license itself or the design or early site permit would 23 need to be resolved separately in a either petition to 24 modify the site permit or certify design, or a request for 25 enforcement action under 10 CFR 2.206.
j l
75 1
The Court said that no, there needed to be a 2
finding and opportunity for hearing under 18980 Atomic 3
Energy Act on new information, significant new information
~
which had arisen since the earlier commissica 4
5 determinations.
All other aspects of the Rule were upheld 6
including the concept of issuing a combined license and the L
7 concept of a design certification.
8 MR. CARROLL:
What definition did the Court agree 9
to with respect to the term new information; what does that-
{
g 10 mean?
11
,MR. MALSCH:
The phrase they used was significant 12 new information.
They left it to the Commission I think, to 13 define exactly what that could mean.
We are in the process 14 of studying the Commission's opinion further.
I don't think 2
15-it has much bearing at a level of detail question which you 16 all are considering this morning.
17.
We are still: studying the opinion and we are 18 considering what the options might be.
Three obvious 19 options are to proceed to implement the Court's opinion by 20 revisions to Part 52 in a rulemaking proceeding;, seek 21 further judicial review of the decision by the D.C. Circuit;
[
22 or, perhaps, pursue legislation.
We are considering all the 23 options at the current time and have made no decisions at all about what path to pursue.
24 25 We have some time.
There is no time urgency in r
z.
76' 1
terms of considering the options.
2 MR. WILKINS:
Any of those options will take a lot 3
of time.
4 MR. MALSCH:
Rulemaking would not take a lot of 5
time.
The have engaged in conformance rulemaking in a 6
relatively short period of time.
A good example would be t
7 the backfit rule which we conformed in a relatively short 8
period of time.
Legislation would be --- could conceivably 9
take a good deal'of time.
Further judicial review would 10 take probably a year or so for final resolution.
Lil MR. WILKINS:.
You have to get all the way to the 12 Supreme Court, don't you?
[
13' MR. MALSCH:
If it went all the way to the Supreme 14 Court -- if we stopped short at.the Court of Appeals or 15 petitioned for Supreme' Court review and the petition was 16 denied, it might take a shorter period of. time.
17 MR. MICHELSON:
Are there further questions?
18 MR. CARROLL:
Is it fair to ask about another 19 legal issue?
20 MR. MALSCH:
Sure.
'21 MR. CARROLL:
What's going on with the NUMARC 22 appeal with the Appeal Court decision with regard to INPO-23 certified training.
24 MR. MALSCH:
The public citizen case.
There's a 25 petition for review -- for Supreme Court review -- pending
E.
77 1
now,.I believe, which has been filed.
'2 MR. CARROLL: 'By who?
3 MR. MALSCH:
I guess it's by NUMARC.
The
~
4 government:has responded, and we are waiting for the court
- 5' to decide the petition, whether to grant review or not.
L 6
MR. CARROLL:
Your response was what?
7 MR. MALSCH:
Our response was to say that the g
8 decision'was wrong, it was incorrect, but then to explain 9
that the government itself had not petitioned for Supreme L
10 Court review because we thought that we could minimize the 11 damage by. issuing regulations in accordance with the 12 statutes which would be non-proscriptive in nature.
()
13-In fact, we have been working on such regulations 14 for some time-and are prepared to move forward.
15 MR. CARROLL:
Knowing that, why did NUMARC 16 petition?
17-MR. MALSCH:
I think you have -- they would be the 18 best ones to answer that question.
t
=
19 MR. MICHELSON:
Are there any other questions?
20 (No response.)
121 MR. MICHELSON:
Seeing none, thank you Marty.
We 22 will proceed with the discussion of level of design detail.
23 Before we start on that, I just wanted to add that we did 24 hold a Subcommittee meeting on this subject.
I chaired the 25 Subcommittee since Charlie Wylie was not available for it.
78 1
The purpose of the Subcommittee meeting originally
^<s' 2
was to review a proposed SECY dealing with the subject, but 3
'the proposed SECY is not yet available.
We spent, I think, i.
l-4 the time very usefully in discussing a. number of associated 5
questions concerning this matter.
We are now going to have 6
to reschedule a Subcommittee meeting for the 5th of December 7
in which to hopefully review the SECY, since it has to be in 8
to'the Commission we.11 before that because the Commission is l
9' having a-briefing on it on the 7th of December.-
10 We will revisit the issue for an hour or so at a 11 Subcommittee meeting just.before our full Committee meeting,.
12 and we would very likely write a letter on this subject at
'13 that time.
Ll 14 With those few remarks, I have askedLthe staff to p
L i
'15 come in and give us a very abbreviated briefing, keeping in L
'16 mind that once you receive this SECY in front of you and L have reaf
't that you will probably have a lot more 18 questions.
Right now'the staff will give us an. idea of l
19 where they are at.
20 MR. GILLESPIE:
I get to introduce the next staff
(
21 members.
Having run the Research Information conference in f
22 many years past and the regulatory information, everyone 23 goes who is going to introduce the first guy who introduced i
/~}
24 overyone who is important.
The unimportant guy always gets
\\.j '
25 up first and says I would like to introduce so and so.
l l
79 1
Let me.give you the status on the SECY.
We>
2 incorporated kind of the last round of comments which we got 3
from Research yesterday, and it is now back up in-EDO's 4-office.
I think the EDO needs some time to read it.
OGC 5
has given us a no legal objection to the current version.
6 As'long as we don't change it a lot I think it will stay.
7 Since Rebecca and_ Gene talked to you the last' 8
time, let me just summarize and Rebecca will go through and Gene in more detail.
I think theirt last presentation pretty 9
L 10 fairly represented the paper before, and the change from the 11 Subcommittee to now has generally I would say been more 12
' editorial and clarification in nature.
But when Research
{
h 13 realized we were recommending a regulatory guide, basically 14 a format and content guide of what you have'to do and how 15 much you-have to do, they started reading it in infinite 11 6 detail.
17 So, their comments generally related to what Gene 18 is going to cover, the graded approach.
We ended up 19 yesterday with the last set of comments, making-sure that we consistently referred to a graded approach and the maximum 20 21 level of detail that we would want for a particular system 22 would be what was is currently referred to as level two.
It
=
23 softened how many systems we need level two for, to the 24 point of saying that's part of developing the guide.
We are 25 not trying to write it right now.
What we feel is that we
l 80 1
have a very good outline.
2 Actually Rebecca, Tom and I went through the 3
research comments yesterday and I don't think anyone else 4
has'really had a chance to-see them.
We have made that 5
' consistent change through it.
We-really are looking at a 6
graded regulatory guide which will cover many components 7
which are currently not done in an application.
8 MR. MICHELSON:
As I understand it, the SECY will 9
just have some kind of an outline of what will be in the 10 guide?
11 MR. GILLESPIE:
I think if you looked at it --
12 Rebecca has a copy -- I would say it's three-quarters of an
( )
'13 inch thick and one-half inch of that is probably a very 14 detailed outline of what would be in the guide.
That is 15 some of the stuff that Gene will go over.
16 They have done a lot of work right down on a 17 system by system basis, evaluating.how much information 18
~ system by system.
That was research's concern, and that was 19 ~
a consistency change that when the guide was written it 20 would indeed go system by system on how much detail was
-21 needed.
22 MR. MICHELSON:
If the gu.'de is indeed that thick s
23
-- if the SECY is that thick, then we probably better plan 24 on a little more than an hour of Subcommittee meeting in
)
25 December.
l
81 1
1 MR. GILLESPIE:
I think Rebecca could show -- a 4
2
. lot of ic is tables.
3' PR.- MICHELSON:
They raise questions.
4 MR. GILLESPIE:
Yes.
5 MR. MICHELSON:
This is the final shot by the 6
Committee on this.
It will not go out for public comment-7 and is going to be endorsed and proceeded with by the 8
Commission; is that correct?
9 MR. GILLESPIE:
We are asking for the Commission 10-to direct the staff to write a guide.
11:
MR. MICHELSON:
The guide would eventually go out 12 then for public comment?
hI 113 MR. GILLESPIE:
That would go out for public 14 comment, yes.
15' MR. MICHELSON:
That would be a few months off.
16 MR. GILLESPIE:
When the expanded outline is over 17' one-half inch thick, I'm afraid -- I am not even going to 181 guess.
I would like you to ask Gene, and Brian can chime'in 19 on what'their estimate of how long it would be.
20 MR. MICHELSON:
It escapes me just a little bit of 21 course, the immediacy of the need for a firm decision on 22 what has to be included so that we can get on with reviewing 23 the ABWR for instance.
That review is scheduled -- can't 24-wait another year for a guide.
25 MR. GILLESPIE:
No, the review will not wait.
We m.
_---__m.-______.________.._._._.__.____s
82 1
will do it by what I often refer to as the brut force
'2 method.
.3
.7h. MICHELSON:
Yet, the reviewer won't know what 4-is required of the Part 52 relative to content of the 5'
application even.
6 MR. GILLESPIE:
There's enough information in this 7
paper that they will --
8 MR. MICHELSON:
There will be.
9 MR. GILLESPIE:
Yes.
We are.trying to deal with a 10 black and white, do they know or don't they know.
11 MR. MICHELSON:
We will see.
12 MR. GILLESPIE:
Within the paper and I think l
13 within vaat Gene is going to go-over,-there is sufficient
-14 detail there that it's not going to be ambiguous.
15 MR. MICHELSON:
Let me mention to the Committee 16
.that for our last Subcommittee meeting NUMARC elected not to 17 participate and they-have not asked'for any time today.
I 18 don't know if they will for our December Subcommittee 19 meeting or not.
70 MR. CARROLL:
Frank, we heard from Combustion for 21 example, thel'_ characterization of their application as 22 being a level three plus.
We hinted to them that perhaps 23 they ought to be thinking more like a level two.
We haven't 24 seen anything yet, and we don't know what --
25 MR. GILLESPIE:
Let me allow Rebecca and Gene and i
83 1
particularly Gene to go through all the work they did.
02 There is significant detail in the enclosure of this paper 3
that will tell them what they will need.
]
4 MR. CARROLL:
They will have that when?
5 MR. GILLESPIE:
Ten days after it goes to the 6
Commission.
So, Commission meeting on the 7th and a week 7
before that we will get it to the Commission right before 8
Christmas.
9 MR. CARROLL:
All right.
A Christmas present.
]
10 MR. GILLESPIE:
Christmas present.
I will say 11 that I think in general it's going to require significant 12 more work on the part of both GE and Combustion.
()
13 MR. MICHELSON:
Can you give us a rough idea of 14 when we might see the SECY itself in view of the December 15 5th planned Subcommittee meeting?
16 MR, GILLESPIE:
Let's see.
The EDO probably needs 17 a week to -- I. hope it's closer than a week.
I would be 18 pretty confident of not longer than a week.
I would say not 19 longer than a week.
We feel that we have the -- this 20 package has been so refined, it's just no more refinement 21 left.
You have to give the EDO a week to read it.
We have 22 to print it out -- two weeks from now.
23 MR. MICHELSON:
We have to see it ten days before 24 our Subcommittee meeting if we expect the members to get it 25 and see it and talk about it.
84 l
1 HR. GILLESPIE:
I think we can get it to you ten 2
days before.
3 MR. MICHELSON:
We have to work both ways toward 4
that middle.
5 MR. GILLESPIE:
We can make ten days before.
6 MR. MICHELSON:
This is not even our usual 7
practice.
We usually require it a month before the 8
deadline.
9 MS. NEASE:
Good morning.
My name is Rebecca 10 Nease.
I am with the Policy Development and Technical 11 Support Branch at NRR.
With me today is Gene Imbro, and he 12 will be presenting part of this, There are a few other 13 staff members who have been iiivolved in the recent push to 14 develop our position in Part 52, and they are available to 15 answer questions.
16
[ Slide.)
17 This morning we are going to be talking about the 18 design certification under Part 52.
Just to refresh your 19
-memory on what has gone on so far, on July lith of this year 20 we issued SECY 9-241; July 18th we had a commission meeting 21 on that paper; on July 26th Brian Grimes and Gene Imbro were 22 tasked to study the design processes under Part 50 and 52 23 and to determine the design products expected at design 24 certification.
25 On August 9th we met with you.
On August 14th you 1
85 l
l 1
wrote a letter recommending level two.
An SRM initiating 2
this current SECY paper was issued on August 22nd of this 3
year.
4 In SECY 90-241 we introduced a number of concepts 5
associated with Part 52.
We talked about the contents of 6
the application, tier one and tier two, we identified three 7
bodies of design information:
that in the application 1
8 certified which we are calling tier one; that in the 9
application and not certified which is tier two; and,
~~
10 information available for audit.
We also presented four i
11 level of detail and the resulting standardization from each.
12 (Slide.]
()
13 In SECY 90-241 we varied the content of the 14 application and the content of the design certification to 15 achieve these four levels.
The four levels range from 16 identical, physical, functional and performance 17 characteristics down to functionally identical similar 18 principal features.
19 (Slide.)
20 Staff requirements memorandum issued in August, 21 1990 asked these seven questions.
In addition to responding 22 to these questions, we developed recommendations to the 23 Commission on how to implement the requirements of Part 52.
24 (Slide.)
25 Regarding the level of detail, the staff proposes
86 1
to use a level of detail to be developed on a graded 2
approach based on that wnich is necessary for the staff to 3
make a safety determination.
The application will contain
~
4 design information similar to an FSAR, less the as-built and 5
site information, and it will be organized into two parts; 6
tier one and tier two.
Again, tier one is the certified portion of the application and tier two is the uncertified 7
8 portion.
9 Information normally contained in procurement and construction and installation specifications will be 10
-11 available for audit to support the staff safety 12 determination.
It will be of sufficient detail to conform
()
13 translation of the safety critoria into design products.
14 Together, the three bars of design information, tier one, 15 tier two and information available for audit will be 16 developed on a graded approach.
The graded approach will 17 recommend greater than level two for some systems in the 18 nuclear island and will recommend level two for the balance 19 of the systems in nuclear island, and for the turbine 20 island.
21 Level four will be required for certification for 22 the site related features, and we expect the site related 23 systems to be developed to a level two at COL.
It is 24 important to remember that this is input for a Reg. Guide concerning a level of detail to be developed on groups of 25 m
i i imu==
--i----mm mmm mmmm m
87 1
systems.
The level of detail for individual systems may be s
T x-[
2 less than that shown here for the entire group.
3 (Slide.)
l 4
Concerning flexibility, we had to consider what is j
5 necessary for the staff safety determination while allowing 6
flexibility needed for the design to incorporate the vendor 7
specifics.
There are three ways for changing tier one which 8
is a certified portion of the application.
Through a l
9 rulemaking to amend the certification and exemption or 10 waiver.
i.
11 (Slide.]
12 concerning changes to tier two information,
(/~'S) 1 13 between design certification and COL, the staff is proposing i
i l
14 to use the same change process that we are using for tier 15 one.
That would be an amendment rulemaking, waiver or an
]
16 exemption.
Between COL and operation, the staff is 17 proposing a change process paralleling that of Section 18 50.59.
Following authorization to operate 50.59 will apply, i
i 19 MR. WILKINS:
Is there any significance to the-l 20' distinction between Section 50.59 will apply and provisions 21 para 11 cling Section 50.59?
22 MS. NEASE:
Yes.
Part 52 doesn't allow the 23 application of 50.59 until you have permission to operate.
24 What we are going to do is, condition the COL and provide (v')
i 25 the same provisions that are provided in 50.59.
The
88 1
information available for audit can be changed as long as 2
the provisions in tier one and tier two are not violated.
3 The cost of redesign will control to some extent changes 4
made to this body of information.
5 (Slide.]
6 Our current paper recommends that the Commission 7
support the staff proposal on the graded approach to level 8
of detail, the contents of the application and the design 9
certification, and the processes governing changes to that j
10 in the application as certified, tier one, that in the 11 application in not.ertified, tier two, and information 12 available for audit.
()
13 The paper also recommends that the Commission 14 authorize the staff to develop a Reg. guide to incorporate 15 these staff proposals.
16 MR. MICHELSON:
Are there any questions?
17 MR. MICHELSON:
Are there questions?
18 (No response.)
19 MR. MICHELSON:
I see none.
I see that you are c
20 staying very well on the schedule.
This will all be
~
discussed in much greater detail once we see the SECY and 21 22 get all the information.
This was just to know where we are headed, and you did a very fine job of showing where you are 23 24 headed.
25 MR. GRIMES:
We would like Gene Imbro to say a few L
s 89 1
words about how the tables were developed.
I 2
MR. MICHELSON:
All right.
3 KR. IMBRO:
My name is Gene Imbro.
I am with NRR 4
in the Special Inspection Branch.
Just to give you a quick 5
rundown on how we approached developing what we felt was an 6
adequate level of detail to enable the staff to make a L
7 safety judgment.
We thought the best thing to do was to 8
focus on the engineering products that needed to be 9
developed at the time of design certification.
sj 10 (Slide.)
11 In doing this, we developed four -- we went 12 through the design process and looked at the four phases of
()
13 design process; conceptual, preliminary, detailed and final, 14 and divided the systems for simplicity into fotr major 15 Proups, as Rebecca pointed out.
We divided the systems into a nuclear island which we considered to be the primary i
16 17 coolant system, reactivity control systems, and other 18 important systems.
19 Then we took the balance of the nuclear island and E3 20 mcde that a separate group, and that would be actually the 21 ECCS systems; CVCS, component cooling water aux feedwater.
We grouped the systems in the turbine building into another 22 23 group called turbine island.
Then one final group remained, E
24 and that was site-specific.
That would be systems like 25 central service water, circ water, ultimate heat sink.
E
90 1
After we defined the design products that were 2
necessary -- at least we felt were necessary -- we then 3
tried to parcel them up into what information would need to 4
be in tier one to assure the level of safety was maintained, 5
what information then would be in tier two and how far the 6
design had to proceed in order to demonstrate to us that the 7
commitments in tier one and tier two were met.
This final 8
group of information is what Rebecca referred to as the 9
information available for audit.
10
[ Slide.)
11 As Rebecca alluded to, we basically came out with 1
12 the following conclusions or recommendations I guess at this
()
13 point.
For the nuclear island we are recommending greater 14 than level two.
What greater than level two means is that 15 not only do you have the functional and performance data 16 specified for the system, but you also have some vendor-17 specific information, particularly confined to the reactor 18 coolant system.
In other words, you need the geometry of 19 the core to do accident analyses.
You might need a detailed 20 control room layout, and you might need additional 21 information on the reactor protective system and engineer 22 safety features actuation systems down to maybe some 23 component level.
24 Also, in the greater than level two we felt that 25 since some of the advance reactor designs are using
91 1
innovative concepts such as multi-plexing, fiber optics and l
\\
\\/
2 other computer-related control systems, we felt that where
~
3 these innovative concepts were being used that it would 4
probably be prudent to have prototyping done.
5 To balance a nuclear island and turbine island we 6
pretty much treated the same in terms of level of detail, 7
and level two basically we are defining as mentioned in the 8
.SECY paper.
You have functional and performance data 9
defined, and you have some physical data available, and you 10 have equipment locations specified, have general 11 arrangements specified in a fair amount of detail which I 12 can get into a little bit.
I,,)
13 Level four we considered to be -- at the site-14 specific we considered to be level four primarily because 15 you don't know what information you have.
You are trying to 16 design generally so that you might not know the 17 configuration of the -- or whether or not you need cooling 16 towers if you have a river or lake or whatever.
A design L
19 certification needed to remain very flexible and that's 20 basically the level of conceptual design and maybe some 21 general performance data.
l l
22 MR. MICHELSON:
Gene, we will have to get ready 1
23 for another meeting here.
Just flip through the others N
24 quickly.
We will hear it in detail the next time.
U(~
25 MR. IMBRO:
Some of the examples of tier one
l 92 1
information quickly, there are two slides with this type of s
I i
l 2
information.
It kind of breaks down into three categories.
3 We felt that tier one needed to contain at least the design 4
bases and criteria, and also programmatic documents such as 5
diversity philosophies and that type of thing, things that 6
would be generically applicable across the design.
7 Also some engineering data needed to be contained 8
in tier one --
9 MR. SHEWMON:
Level one would show you all the 10 separations of trains.
11 MR. IMBRO:
Tier -- you are getting levels and 12 tiers confused mcybe.
A
!. )
13 MR. SHEWMON:
Level one and tier one, you are 14 right.
15 MR. IMBRO:
For the tier one information -- yes, 16 we would really want to have that information available.
In 17 other words, you have to know pretty much --
18 MR. SHEWMON:
You answered my question.
i 19 MR. IMBRO:
Also needed to know some physical 20 considerations such as general arrangements, routing of 21 piping and I guess that's the answer to your question, 22 location of components.
23 (Slide.]
(
24 What we wound up doing was -- we surveyed several 25 architect / engineers and between the information that we i
l i
93 1
gathered from them and information that we got from our s
\\ms/
2 consultants, we came up with this card which is a 3
representation of design finality versus engineering hours.
4 For this we feel that -- of course, this is an integrated 5
curve over the total plant and all the groups and systems.
6 We felt that at the time of design certification 7
that something on the order of 50 percent of the engineering 8
hours expended would be appropriate, and that would give you 9
a design finality roughly of about 85 percent which
~
10 basically means that at the time you spend 50 percent of 11 your engineering hours you are not going to change the 12 design a whole lot after that.
The changes would be minor.
13 (Slide.]
l i
14 On a kind of system-specific level, this curve i
15 really depicts what the design level of engineering effort 16 would be for the different grades of systems that we
[
l 17 specified.
So for the reactor coolant system you would be l
l 18 up there somewhere of 70 or 80 percent.
For the site-19 specific systems you might be down somewhere between 10 and l
20 20 percent of engineering hours expended.
l l
1 l
21 (Slide.)
22 This basically depicts the fact -- it's sort of 23 like the other curve -- that is pretty much what I just
(~T 24 said.
The primary system and containment, you have quite a L) 25 bit of the engineering completed, less for the nuclear i
94 1
island and turbine island and still less then for the site-i
\\
.V 2
specific systems.
3
[ Slide.]
4 This simply depicts at the time of COL issuance 5
you would bring the site-specific systems up to the level at 6
which the balance of nuclear island and turbine island were 7
at the time of design certification.
8 MR. SHEWMON:
Can you tell me what, if any, 9
relationship there is between tier one and level one?
10 MR. IMBRO:
They are really different.
Tier one 11 is the --
12 MR. SHEWMON:
I know that, but I don't have any
)
(
13 idea of what they are.
14 MR. IMBRO:
Tier one is the information that would 15 be certified in the design certification rule that would i
16 basically be unchangeable except by waiver or exemption or 17 other legal means.
Level one is really referring to a stage 18 of design completion, where a level one system would be 19 defined to the detail where you would actually have 20 dimension drawings and be able to go in and build from 21 drawings like level one drawings.
22 That concludes my presentation.
23 MR. MICHELSON:
Are there any questions?
/^T 24 (No response.]
'Y 25 MR. MICHELSON :
We will hear about this again next
95 1
month when we have to write a letter, so we don't have to
,m
(
'I
?
2 understand it all yet.
I see no questions.
Thank you very 3
much for staying well within the schedule since I stole part 4
of your time anyway.
Thank you again.
5 The next item on the agenda is the preparation for 6
our meeting with the NRC Commissioners.
7 (Whereupon, at 11:26 a.m.,
the transcribed portion 8
of the meeting concluded.]
9 10 11 12 r~}
(_/
14 15 16 l
L 17 l'
18 19 l
20 l
21 l
l 22 l
23 O
L 25 l
l
I
(
REPORTER'S CBRTIFICATE This is to certify that the attached proceed-ings before the United States Nuclear Regulatory Commission in the matter of:
367th ACRS General Meeting NAME OF PROCEEDING DOCKET NUMBER:
PLACE OF PROCEEDING: Bethesda, Maryland were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and thereafter reduced to typewriting by me or under the direction of the court report-ing company, and that the transcript is a true and accurate record of the foregoing proceedings.
R Official Reporter At' Riley & Associates, Ltd.
'l
. i:, i ;
l tL
!t;l
- l
!!ii
!t!!-
!6:
!.!!ij!
f.
~>
a >-
O :m e
l i
l ye p ess rse l
l uul i MR Y
TW.G E
F VRUS 0
T 99 C
1 A
P 8
O M
re I
b YR m
O ev T
o A
N LU G
ER O
,, lel1:lL1l\\l1!
.,l
- li l,;!
.i:i!
4
{ i !;i!,
~
O O
G
~
REGUI ATORYIMPACT MILESTONES
- LICENSEE SURVEYS FALL 1989
- GL 90-01 JANUARY 12,1990
- STAFF SURVEYS SPRING 1990
- SECY 90-080 MARCH 9,1990
- SECY 90-205 JUNE 7,1990
- SECY 90-250 JULY 16,1990 s.
g g
~
~
O 0
9 REGULATORY IMPACT LICENSEE SURVEYS CONDUCTED SEPTEMBER 25 - DECEMBER 1,1989 BY SENIOR NRC MANAGEMENT TEAMS
- 5 TO 6 MEMBERS PER TEAM 13 LICENSEES VISITED
- AT LEAST 2 LICENSEES PER REGION APPROX. 5 DISCUSSION GROUP SESSIONS PER VISIT
- OPERATORS
- CORPORATE PERSONNEL
- ENGINEERS
- MANAGERS
- SENIOR MANAGEMENT J
O-O-
D' REGUI.ATORY IMPACT SURVEY Comment Topic Areas i
l
- 1. Requirements and Perceived Requirements l
l
- 2. NRC Licensing Activities i
l
- 3. NRC Inspection Activities i
I
- 4. Performance Evaluations I
l S. Multiple Oversight Organizations 4
i
(
4 l
't
.7.
~
~
O O
O REGULATORY. IMPACT SURVEY (cont.)
Comment Topic Areas (cont.)
- 6. Operator Licensing
- 7. Enforcement and investigations
- 8. Reporting Events
- 9. Communications
- 10. Qualification of NRC Personnel
o
~
O O
O i
LICENSEE SURVEY - PRINCIPAL THEMES l
e Licensees Acquiesce to NRC Requests i
i e
NRC Dominates Licensee Resources l
4
[
t I
I i
)
r O
O O
REGULATORYIMPACT i
MAJOR AREAS FOR IMPROVEMENT
)
i
- 1. CONSIDERATION OF THE CUMULATIVE EFFECT OF NRC GENERIC REQUIREMENTS i
- 2. SCHEDULING AND CONTROL OF J
INSPECTIONS, ESPECIALLY TEAM j
INSPECTIONS j
j
- 3. MANAGEMENT EXPECTATIONS, TRAINING j
AND OVERSIGHT OF INSPECTORS
i.
O i
O O
i REGULATORYIMPACT l
- 1. CUMULATIVE EFFECTS OF NRC GENERIC REQUIREMENTS
- ADDED BACKFIT DISCUSSION TO GENERIC l
CORRESPONDENCE
-NUREG 1409 BACKFITTING GUIDELINES I
- REVISING VALUE IMPACT HANDBOOK
- INTEGRATED REGULATORY REQUIREMENT IMPLEMENTATION SCHEDULE (IRRIS) l t
?
o
~
O O
.O' l
REGULATORY. lMPACT i
i i
IRRIS t
t i
_ VOLUNTARY LICENSEE PARTICIPATION l
{
l
- APPLIES TO UNIMPLEMENTED SAFETY ENHANCEMENTS, NOT l
TO ADEQUATE PROTECTION OR COMPLIANCE REQUIREMENTS f
4
- PRIORITIZED AND SCHEDULED BASED ON SAFETY AND IMPACT l
i
- FROZEN IN ADVANCE OF OUTAGE l
i E
i 4
4
e O
O REGUALTORY IMPACT
- 2. SCHEDULING AND CONTROL OF INSPECTIONS
- RESTRUCTURING OF INSPECTION PROGRAM
- EMPHASIS ON INSPECTION PLANNING (MIPS)
- POLICY FOR COORDINATING SITE VISITS
- ADDITIONAL PROGRAMMATIC IMPROVEMENTS
O O
OT i
REGULATORYIMPACT I
PROGRAMMATIC IMPROVEMENTS
- FOSTER BE' ITER TEAM AND ROUTINE INSPECTION Pl.ANNING l
- NEED ESTABLISHED BASED ON LICENSEE PERFORMANCE OR GENERIC SAFETY i
- ANNOUNCE TEAM INSPECTIONS
- CONTROLS ON NUMBER OF TEAM j
INSPECTIONS ( < 4 PER SALP CYCLE) l L
i
~
O O
O'
~
1 1
REGULATORY IMPACT i
PROGRAMMATIC IMPROVEMENTS i
oBETTER SCHEDULING OF NRC SITE ACTIVITIES UNIT SPECIFIC PLANS MAINTAINED k
BY REGIONS j
- UPDATED QUARTERLY 1
i
- CONSIDER ALL MAJOR SITE ACTIVITIES I
- TO BE PUBLISHED PERIODICALLY f
i 4
i i
~
~
O O
O l
^
REGULATORY IMPACT
- 3. MANAGEMENT EXPECTATIONS, TRAINING AND OVERSIGHT i
- EMPHASIS ON PROFESSIONALISM c
SPECIALIZED TRAINING FOR STAFF L
- TRAINING ADVISORY GROUP l
l
- TEAM LEADER POSITIONS 1
i
- ENHANCE TRAINING i
t i
l i
l I
i
~
O O
O REGULATORY IMPACT-i ENHANCED TRAINING i
- REFINE FUNDAMENTALS & ADD REFRESHER L
i i
- ROLE / CONDUCT OF INSPECTOR n
L i.
- INTERPERSONAL SKILLS TRAINING l
l
- COMMUNICATIONS, INTERVIEWING,
)
i MEETINGS i
i i
- PROBLEM SOLVING AND DECISION MAKING l
I l
u a
a
.:~--c s
O 0: ~
~
~
D REGULATORYIMPACT PROFESSIONALISM MANAGERS AND SUPERVISORS
- . COMMUNICATE EXPECTATIONS
- SOLICIT FEEDBACK-
.* OVERSIGHT / ACTION INSPECTORS
- PROFESSIONALISM IN ALL ACTIVITIES
...i..
m i e
i
' =
a i
=
REGULATORY IMPACT a
OTHER ACTIONS
- NRC LICENSING ACTIVITIES t,
- NRC INSPECTION ACTIVITIES
- PERFORMANCE EVALUATIONS i
- MULTIPLE OVERSIGHT ORGANIZATIONS
- OPERATOR LICENSING
- ENFORCEMENT i
- EVENT REPORTING l
I i
1
-,..x.-..
.._,....a,
.~. -.
a REGULATORYIMPACT CONCLUSIONS
- RECOGNIZED AREAS FOR IMPROVEMENT
- INITIATED CORRECTIVE ACTIONS
- SOLICIT & EVALUATE PUBLIC COMMENTS
- ADJUST ACTION PLANS
- FINALIZE NUREG 1395
O.
- > : I L. X:ih L:O M J ' r
- w-b
- f e
0 d 72; L2l STANDARDIZATION AND PART52 LICENSING NOVEMBER 8,1990 R. NEASE, TECHNICAL ASSISTANT POLICYDEVELOPMENT, NRR 301-492-1281 G. IMBRO, SECTION CHIEF SPECIAL PROJECTS BRANCH, NRR 301-492-0954 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS q
~-
OL O'
- O LEVEL OF DESIGN DETAll nfAIOR ntILESTONES
- SECY 90-241-JULY 11,1990
- COMMISSION MEETING JULY 18,1990 j
1
- GRIMES /IMBRO BEGIN JULY 26,1990
- ACRS AUGUST 9,1990
- ACRS LETTER AUGUST 14,1990 i
- SRM AUGUST 22,1990 l
s 4
)
,=
y 9
g7-}.
I
~
LEVEL OFDESIGN DETAll l
1 SECY90-241
- CONTENTS OF THE APPLICATION TIER 1 & TIER 2
- CERTIFICATION -TIER 1 MATERIAL AVAILABLE FOR AUDIT
- LEVELS 1,2,3, & 4
_m
~
d~
e e:
LEVEL OF DESIGN DETAll FOUR LEVELS
- 1. IDENTICAL PHYSICAL, FUNCTIONAL &
PERFORMANCE CHARACTERISTICS
- 2.-PHYSICALLYSIMILAR/ IDENTICAL FUNCTIONAL & PERFORMANCE CHARACTERISTICS
- 3. IDENTICAL FUNCTIONAL &
PERFORMANCE CHARACTERISTICS
- 4. FUNCTIONALLY IDENTICAL / SIMILAR PRINCIPAL FEATURES
e e
DT
~~
LEVEL OF DESIGN DETAIL:
STAFFREQUIREMENTS
- 1. SSC WHERE LEVEL 1 OR 2IS FEASIBLE AND PRACTICALTO ACHIEVE
- 2. SSC WHERE LEVEL 1 IS NOT NEEDED FOR STANDARDIZATION
- 3. APPROACH PROVIDING FLEXIBILITY
- 4. ADVANTAGES / DISADVANTAGES OF TWO TIERS
- 5. ANALYSIS OF PUBLIC COMMENTS
- 8. STANDARDIZATION PORTION OF THE REVIEW
- 7. LESSONS LEARNED
,+
o.
o; o
LEVEL OF DESIGN DETAll u
STAFFPROPOSAL-DETAIL
- LEVEL OF DESIGN DETAll l
- GRADED APPROACH BASED ON SAFETY
- INFORMATION IN CERTAIN PROCUREMENT & C&l SPECS
- APPLICATION
- FSAR MINUS AS-BUILT & SITE INFORMATION
- ORGANIZED INTO TWO PARTS / TIERS l
1
- AVAILABLE FOR AUDIT f
- SUFFICIENT INFORMATION TO CONFIRM TRANSLATION OF SAFETYCRITERIA INTO DESIGN l
l
LEVEL OF DESIGN DETAll STAFFPROPOSAL-DETAIL
- GRADED APPROACH BASED ON SAFETY
- > LEVEL 2 FOR CERTAIN NUCLEAR ISLAND FEATURES
- LEVEL 2 FOR OTHER NUCLEAR ISLAND FEATURES
- LEVEL 2 FOR KEY TURBINE ISLAND FEATURES
- LEVEL 4 AT CERTIFICATION AND LEVEL 2 AT COL FOR SITE SPECIFIC FEATURES
x 3
o o
g
=
LEVEL OF DESIGN DETAIL.
STAFFPROPOSAL-FLEXIBILITY
- CERTIFIED PORTION OF THE DESIGN / TIER 1 l
- RULEMAKING TO AMEND CERTIFICATION
- EXEMPTION PER SECTION 52.63
- - WAIVER PER SECTION 2.758 l
l i
j
. S
=: }
o Lo q
LEVEL OF DESIGN DETAIL-STAFFPROPOSAL-FLEXIBIUTY
- IN APPLICATION BUT NOT CERTIFIED / TIER 2
- BETWEEN DESIGN CERTIFICATION AND COL l
l AMENDMENT RULEMAKING, EXEMPTION, WAIVER
- BETWEEN COL AND AUTHORIZATION TO OPERATE l
PROVISIONS PARALLELING SECTION 50.59
- FOLLOWING AUTHORIZATION TO OPERATE SECTION 50.59 3
s 4
O
~
IOi
~
!O!
~
LEVEL OF DESIGN DETAll STAFFPROPOSAL-FLEXIBlUTY
-INFORMATION AVAILABLE FOR AUDIT
- SAFETY a
- TIER 1 & 2 -
i
- COST OF REDESIGN I
l s
t
- l
=
=
O IO IQ.
, +,
~
g LEVEL OF DESIGN DETAILL
-l RECOMMENDATIONS
- SUPPORT STAFF PROPOSAL ON :
GRADED APPROACH TO DESIGN FINALITY q
- CONTENT OF THE APPLICATION AND CERTIFICATION j
- CHANGE PROCESS FOR MATERIALIN APPLICATION, CERTIFICATION AND HELD FOR AUDIT
- AUTHORIZE DEVELOPMENT OF REG. GUIDE i
g g
LEVEL OF DESIGN DETAIL APPROACH
= DEFINE THE DESIGN PRODUCTS DEVELOPED AT EACH PHASE THE DESIGN PROCESS
- DEFINE THE ENGINEERING PRODUCTS WHICH MUST
~
BE COMPLETED AT THE TIME OF DESIGN CERTIFICATION
- DEFINE THE INFORMATION TO BE CONTAINED IN :
= TIER 1
= TIER 2
= APPLICANT'S FILES AND AVAILABLE FOR AUDIT l
~
~
e o
o
~
LFVEL C'F DESIGN DETAIL RECOMMENDED ' EVEL OF DETAIL GREATER THAN LEVEL 2
- NUCLEAR ISLAND
. BALANCE OF NUCLEAR ISLAND LEVEL 2
- TURBlNE ISLAND LEVEL 2 LEVEL 4 AT DESIGN CERTIFICATION
- SITE SPECIFIC
' LEVEL 2 AT COMBINED O/L
~
g g-LEVEL OF DESIGN DETAIL EXAMPLES OFL TIER I INFORMATION y
- DESIGN. BASES AND CRITERIA i
- SIMPLIFIED PIPING AND INSTRUMENTATION DIAGRAMS I
l
- SIMPLIFIED ELECTRICAL SINGLE-LINE DIAGRAMS CLASS 1E AC/DC l
- GENERAL ARRANGEMENT DRAWINGS
- LOCAT_ IONS OF EQUIPMENT (NOTE 1) j
= ROUTING OF ALL PIPING > 6 INCHES (NOTE 1)
- ROUTING OF HIGH-ENERGY PIPING > 2 1/2 INCHES '(NOTE 1)
- ROUTING OF HVAC DUCTS (NOTE 1) j l
NOTE 1: WITHIN ' DEFINED ENVELOPE i
r l
4
,l_.
,I..
g, ffs
.s
,p.e.c
,,,,j,a r
E Ac
^
~
1
~
~ _
i I EVFL C'F DFSIGN DETAIL l
EXAMPLES OF TIER 1 INFORMATION l
ROUTING OF. CABLE TRAY SYSTEMS (NOTE 1)
EQUIPMENT PERFORMANCE REQUIREMENTS STRUCTURES (EXCEPT FOR LOCALIZED -DESIGN INFLUENCED BY VENDOR' SPECIFIC-' DATA)
SYSTEM FUNCTIONAL AND. PERFORMANCE REQUIREMENTS (EXCEPT FOR VARIATIONS RESULTING FROM' VENDOR SPECIFIC DATA) l ROUTING 'OF ASME CLASS 1N PIPING > 1 INCH (NOTE 1) e MAN-MACHINE FUNCTIONAL-INTERFACES DATA NETWORK' DESCRIPTIONS (SYSTEM ARCHITECTURE)
PROGRAMMATIC DOCUMENTS SOFTWARE. VERIFICATION AND VAllDATION i
SEISMIC QUALIFICATION PLAN ENVIRONMENTAL QUALIFICATION PLAN
- SETPOINT TOLERANCE METHODOLOGY NOTE 1: WITHIN-DEFINED ENVELOPE
b a
Ye A,..y-.
w.
.,,.., '*.~
+,...,J 4.
...,s e
D 4 -
- 9;j
~
- g.
DESIGN FINALITY vs. ENGINEERING HOURS i
i l
1-P E.
? - - - -- y;
' R 3og I
E N
- ~ - -
~
T 80 D
E S
~ - -
~
l 80 l
e l
N
- - ~ ~ - - - - -
- - - - - - ~
~ ~--
F 40 I
N A
~~--
~ ~ - ~ ~ ~ ~
g-20 8.
T y.
O
- 20..,
40 60
- 80 100 PERCENT ENGINEERING HOURS d'd
O Y
Q
~$
~
i DESIGN FINALITY vs._ ENGINEERING HOURS y
P E
R
- - --- Levn-S - - - - -
7 C
100
.E N
(gyet,
T 80 l
D E
l S
l 60 j
G
'N l
-F 40 t
tevet a g
N A
-=
I tevet 4 T
~
Y
/
O O
20 40 60 80 10 0 PERCENT ENGINEERING HOURS
~.
i i
i T
b
g
~
O O
C.
': ~
+l DESIGNLFINALITY vs. ENGINEERING HOURS AT DESIGN CERTIFICATION P
E
/
R SYSTEM C
100 E
satance AwD N
0F_____
. cow 1RINMENT ~
muctuan T
AND 80 tuneswe 0
ISLAND E.
=
S 60 l
G y
40 site -
F sesceFic i'
I
__._ __ sysismo..
N
/
~
A 20 L
?
l T
/
B/
/
Y
'O 20 40 60 80 100 PERCENT ENGINEERING HOURS n
i
- N
-4
-mw'--
n m :;
i., m :: : x-y ; i _
.[... _ _ ['-;
L __ _
.~
T.f s k ' i 3 m
C)l;;
V.
ps 7
Q__
~ ~ '
S DESIGN FINALITY vs ENGINEERING HOURS AT COMBINED. OPERATING LICENSE P
E R
,,,,,,y C
site sysrtu E
-spEctr8C
,yy p
/
smaus e,,,,,,,,,,
T n
400 E
/
E - -- _ _.-
8
.80 G
N 60 tance i-,
or p
""C'**"
I 40 AN A
/
satano L
20 I
/////////.
/ /.
//////l h'O //////////
- ///
//////
20 40 60' 80 100
.. PERCENT ENGINEERING HOURS
~
..