ML20245H785
| ML20245H785 | |
| Person / Time | |
|---|---|
| Issue date: | 08/10/1989 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1750, NUDOCS 8908170282 | |
| Download: ML20245H785 (166) | |
Text
,
0R!GlNjz
~
O UNITED STJsTES NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR :
SAFEGUARDS (ACRS) 352nd MEETING, DAY ONE mb PAGES:
1 through 124 PLACE:
Bethesda, Maryland
[.
DATE:
August 10, 198 I
I ri n
u
...v.M,.w.g
.O.7 L. g. 0.p 7..~lOd g.
- u.. 1 O
HERITAGE REPORTING CORPORATION OB%elReporters 1220 L Street, N.W., Suke 640 l
Wuhington, D.C. 20005
- c o s 1 7 0
- 5 2 5 05!O (202) 6M
- DR ACRI
)
T - - 17 "'O
'l" L_ __ ___-.-
C M 'i 1
PUBLIC NOTICE BY THE 2
UNITED STATES NUCLEAR REGULATORY COMMISSION'S 3'
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS
4 5
6 7
The contents of this. stenographic transcript of the j
8 proceedings of the United States Nuclear. Regulatory 9 -'
Commission's= Advisory Committee on Reactor Safeguards (ACRS),
10 as reported'herein, is.an uncorrected record of the discussions c
11 recorded at the meeting held on the above date.
12
'No member of the ACRS Staff and no participant at 13
.this meeting accepts any responsibility for errors or
. ()
14 inaccuracies of statement or data contained in this transcript.
15 16 17 18 19 20 21-22 23
'24 25-O Heritage Reporting Corporation (202) 628-4888 e
1
,y g_).
U.S. NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR :
SAFEGUARDS (ACRS) 352nd MEETING, DAY ONE
- Thursday, August 10, 1989 Room P-110 7920 Norfolk Avenue Bethesda, Maryland The meeting came to order, pursuant to notice, at 8:30 a.m.
BEFORE:
FORREST REMICK (Chairman)
Tk-}
Chairman, ACRS Associate Vice-President for Research Professor of Nuclear Engineering The Pennsylvania State University University Park, Pennsylvania MR. CARLYLE MICHELSON (Vice Chairman)
Retired Principal Nuclear Engineer Tennessee Valley Authority Knoxville, Tennessee,'and Retired Director, Office for Analysis and Evaluation of Operational Data U.S.
Nuclear Regulatory Commission Washington, D.C.
ACRS MEMBERS PRESENT:
DR. CHESTER P.
SIESS Professor Emeritus of Civil Engineering University of Illinois Urbana, Illinois O-IIeritage Reporting Corporation (202) 628-4888
2 t'
\\
ACRS MEMBERS PRESENT:
DR. PAUL G.
SHEWMON Professor, Metallurgical Engineering Department Ohio State University Columbus, Ohio DR. WILLIAM KERR Professor of Nuclear Engineering Director, Office of Energy Research University of Michigan Ann Arbor, Michigan MR. CHARLES J. WYLIE Retired Chief Engineer Electrical Division Duke Power Company Charlotte, North Carciina MR. DAVID A. WARD Research Manager on Special Assignment E.I.
du Pont de Nemours & Company Savannah River Laboratory Aiken, South Carolina DR. HAROLD W. LEWIS
()
Professor of Physics Department of Phyoics University of California Santa Barbara, California MR. JAMES CARROLL Retired Manager, Nuclear Operations Support Pacific Gas & Electric Company San Francisco, California DR. IVAN CATTON Professor of Engineering Department of Mechanical, Aerospace & Nuclear Engineering School of Engineering and Applied Science University of California Los Angeles, California ACRS COGNIZANT STAFF MEMBER:
RAYMOND FRALEY, Executive Director, ACRS Heritage Reporting Corporation l
(202) 628-4888 L_=--_--____-.--_
2A t'3
(_j NRC' STAFF PRESENTERS:
MARK REINHART Technical Specifications Branch Office of Nuclear Reactor Regulation ROBERT DENNIG, AEOD LARRY BELL, Engineer, AEOD JOEL PAGE, Task Manager on Generic Issue 79 i
Office of Research i
ALSO PRESENT:
ERNEST ROSSI Director, l
Division of' Operational Events Assessment Office of' Nuclear Reactor Regulation JOSE CALVO, Chief Technical Specifications Branch Office of Nuclear Reactor Regulation DAVID FISHER, Chief Special Projects Branch Office of Nuclear Reactor Regulation
[}
\\
MILLARD WOHL TOM NOVAK, Division Head, AEOD ROBERT BAER RICHARD JOHNSON WARREN MINNERS Heritage Reporting Corporation l
(202) 628-4888 i
3 V) f 1
P R Q
,C_ E E D_ I NGS 2
MR. REMICK:
The meeting will now come to order.
3 This is the-first day of the 352nd meeting of the 4
Advisory Committee on Reactor Safeguards.
5 I am Forrest Remick, Chairman of the committee.
6 During today's meeting, the committee will hear 7
about and discuss the following topics:
nuclear power plant 8
technical specifications; progress and activities to reduce 9
scrams in nuclear power plants; we will prepare for our 10 meeting with the NRC Commissioners; and we will meet with 11 the Commissioners this afternoon at One White Flint North; 12 Discuss Generic Issue 79, unanalyzed reactor vessel thermal 13 stress during natural convection cooldown; and we will
(}
14 discuss future ACRS activities.
15 Topics for consideration on Friday and Saturday 16 are listed on the schedule posted on the bulletin board 17 outside the meeting room.
18 This meeting is being conducted in accordance with 19 the provisions of the Federal Advisory Committee Act and the 20 Government in the Sunshine Act.
21 Raymond F.
Fraley is the Designated Federal 22 Official for the initial portion of the meeting.
23 We received no written statements, nor requests to 24 make oral statements from members of the public regarding 25 today's sessions.
Heritage Reporting Corporation (202) 628-4888 l
l
4
.Q
(_j l'
A transcript of portions of the meeting is being 2
kept.
It is requested that each speaker use one of the 3
microphones, identify himself or herself, and speak with 4
sufficient clarity and volume so that he or she can be 5
readily heard.
6 We will begin with several items of current 7
interest.
8 Last month we announced that Vic Stello had been 9
nominated to become Assistant Secretary for Defense Programs 10 at Department of Energy.
He has been detailed to DOE on the 11 27th of July, and James Taylor has been named Acting EDO.
12 David Williams has been nominated to the new 13 position of Inspector General.
()
14 I have asked three of our members, Chet Siess, Jay 15 Carroll and Ivan Catton to serve as our nominating committee 16 for new officers of next year, and I have asked Chet if he 17 would serve as Chairman, and he has agreed.
18 The second draft of NUREG 1150 was published for 19 peer review and the Commission has approved the NRC staff's 20 interim use of the document methods and results, while the 21 peer review panel is underway, or the process is underway.
22 Also, the staff has requested utilities to inspect 23 certain types of swing check valves to see if they contain 24 stainless steel bolts that are susceptible to stress 25 corr'osion cracking.
Apparently there have been failures Heritage Reporting Corporation (202) 628-4888
I:
L:
5
/ml).
1 reported at Diablo Canyon, D.C.
Cook, and Fitzpatrick.
m 2
The Commission has ordered the staff to move 3
forward promptly with the emergency response data system L
1 4
voluntary initiative, and at the same time to develop in 5
parallel a rule which would require that.
Also, in that 6
staff requirements memorandum, the Commission asked the 7
staff to report back on any utilities who do not volunteer.
l 8
(Laughter)
I 9
MR. REMICK:
I thought you would appreciate that.
10 I would like to remind you again of the fact that 11 we do meet with the Commission today.
We have at least five 12 topics.
Time has been set aside later this morning to 13 discuss those.
(])
14 Two changes in the schedule, though, that appears 15 in your three-ring binder:
to allow sufficient time to get 16 out there and get settled before the scheduled meeting, the 17 buc will leave at 1:30 outside our building here rather than 18 1:40 as indicated on the agenda, and we will return at 19 approximately 3:45 from the basement of the building rather 20 than 4:00 O' clock.
21 You should have received a note of the fact that 22 the back door exit from our conference room here has now 23 been alarmed.
To exit from here to the other parts of the 24 building requires a security card which can be obtained I 25 believe from Mrs. Lee.
Coming into this conference room you
)
Heritage Reporting Corporation (202) 628-4888 i
1 J
6
(
1 do not need to use the card.
2 I am also asked to caution you that the door 3
should not remain open more than ten seconds, or the alarm 4
goes off.
5 Also, there are several subcommittee revisions, I 6
am told, in structural engineering.
Chet Siess is Chairman.
7 Paul.Shewmon is on that committee, and Carlyle Michelson.
8 Is that correct?
9' MR. MICHELSON:
I don't think so.
10 MR. REMICK:
Charlie Wylie, excuse me.
My "W"
11 looks like an "M."
12 Structural engineering has proposed that the three 13 subcommittee members would be Chet Siess, Paul Shewmon and
()
14 Charlie Wylie.
15 Rather chan cause such confusion early in the 16 morning, let me drop the other proposed changes.
17 There is a very good chance, if we are diligent 18 and dedicated, like we always are, that we could finish up 19 Friday evening rather than Saturday, as I had earlier 20 announced.
It might be that we will have to go until 6:30 21 or 7:00 0' clock Friday evening.
But looking at the 22 schedule, it is entirely possible that we could finish up 23 Friday evening.
So I mention that for you to make plans.
24 If we are diligent.
25 Any other items of current interest?
Heritage Reporting Corporation (202) 628-4888
7 1
(No response) 2 MR. REMICK:
If not, let's move then to our first 3
agenda topic.
That is nuclear power plant technical 4
specifications.
And our subcommittee chairman is Jay
'5 Carroll.
And I see a familiar face, Dave Fisher.
We 6
welcome Dave back this morrIing.
7 MR. ROSSI:
I'm Ernie Rossi, Director of the 1
8 Division of Operational Events Assessment in the Office of 9
Nuclear Reactor Regulation..And we are very pleased to be 10 here today to give you a briefing on our technical 11 specifications improvement program.
12 As I am sure you know, we consider this within NRR 13 to be a very important program.
We think that this program
()
14 will significantly improve operational safety.
15 The presentation today is going to be given by 16 Mark Reinhart, a member of the Technical Specifications 17 Branch in NRR.
And I believe Mark is going to tell you who 18 the other members of the staff are that are here today.
19 MR. REINHART:
Good morning.
My name is Mark 20 Reinhart.
I am with the Technical Specifications Branch.
21 We are here to brief the ACRS on the technical 22 specifications improvement program.
23 In attendance, as he introduced himself, is Mr.
24 Ernie Rossi, the Director of the Division of Operational 25 Events Assessment.
Heritage Reporting Corporation (202) 628-4888
8 r~5 m
l
(_/
1 Also with us is Mr. Jose Calvo, the Chief of the 2
Technical Specifications Branch and Mr. David Fisher, Chief 3
of the Special Projects Branch.
We also have a number of 4
members of our staff with us, too.
5 OTSB has previously discussed with ACRS the 6
interim Commission's policy statement on technical 7
specifications improvement in July 1986 and November 1987.
8 Today, at your request, we return to provide an 9
update status on the progress of that program.
We do intend 10 to come back again with the final policy statement.
11 The goal of our program is to improve operational 12 safety.
We have three aspects.
Based on the CommLssion's 13 policy statement, the interim policy statement of February,
()
14 1987, the three aspects are to develop new standard 15 technical specifications, at the same time have a parallel 16 program for line-item improvements to the standard tech 17 specs, and some other supporting activities which I will 18 address as I go along.
19 The goal to improve operational safety has a 20 number of aspects.
We want to focus attention on the most 21 safety-significant requirements, or to reduce action 22 statement-induced challenges to safety systems.
We want the 23 technical specifications to be more operator-oriented, to 24 reduce their size and complexity, to improve the bases, 25 particularly to capture the analysis performed to show why
(
Heritage Reporting Corporation (202) 628-4888
9
(
1 an LCO is there, why a given allowed outage time is there, 2
why a given surveillance test is required and why that 3
interval is appropriate.
- 4 We want to apply human factors to the writing of 5
the spec and allow more efficient use of both the industry's 6
and the NRC's resources.
7 We feel we have made significant progress in those 8
directions.
9 Based on the Commission's interim policy 10 statement, industry proposed to the Staff their idea of 31 which requirements could be relocated to licensing control 12 documents and which requirements must.be maintained as 13 requirements, as LCOs in the technical specifications.
(])
14 Based on their submittals, the staff made a 15 decision and published that decision in May of 1988.
That 16 publication became known as the Split Report.
It split 17 which requirements were to stay in tech specs and which were 18 to be relocated.
19 Based on that Split Report, the industry was to 20 submit new proposed standard technical specifications.
21 While industry was developing that, the staff 22 developed a process and an ambitious schedule to review 23-industry submittal and issue an SER within ten months of the i
'24 submittal.
25 The submittal dates were to be in March of 1989.
Heritage Reporting Corporation (202) 628-4888
10
()
1 And industry was fairly close.
All four came in between 2
March and June of 1989.
Based on our ten-month schedule we 3
anticipate SERs in the February-April 1990 time frame.
4 The first step of our process was an acceptance 5
review.
We would look at the proposals, determine if there 6
was sufficient information there to conduct a review.
We.
l 7
determined that there was.
We accepted all four submittals.
I 8
However, we did notice some deviations from the anticipated i
9 results.
10 There were a number of proposed changes to 11 technical requirements in excess of what we anticipated or 12 planned for.
The justification for those proposed technical 13 changes was, in a number of cases, incomplete.
()
14 Based on an ongoing audit by National Lab. and our 15 own staff review, the bases are not as comprehensive as 16 anticipated by the Commission's interim policy statement and 17 the owners groups were not as consistent among themselves as 18 we felt they should have been in this submittal.
19 So what that did to us, it kind of put a 20 perturbation in our review schedule.
We are really not able 21 today to fully assess what that is going to do.
We do know 22 that the time involved to get out the first markup has been 23 twice that which we initially planned.
24 However, both NUMARC and the collective owners 25 groups have come back and expressed to the staff that their l
()
Heritage Reporting Corporation (202) 628-4888
=
7_.
l 11 l
L ()
1 desire is that the review of the proposed technical changes 2
take precedence over the schedule.
3 So in trying to cooperate with the owners groups, l
4 the staff has made some adjustments to accommodate these 5
difficulties, while moving ahead, we do feel, at least right 6
now, barring any unforeseen occurrence, that we will finish 7
our review within the time frame of our original schedule.
\\
8 MR. REMICK:
I'm not sure I understand what you l
9 mean, that the technical review should take precedence over 10 the schedule.
11 In other words, they want the technical review to 12 be completed regardless of the time it takes?
13 MR. REINHART:
Yes, sir.
(~}
14 MR. REMICK:
But what would be the alternative?
15 MR. ROSSI:
Let me say a couple of words about 16 that.
17 Originally, what we had planned on doing at this 18 point in the tech spec improvement program was approving 19 primarily the tech specs which would improve the bases and 20 the human factors without changing surveillance intervals or 21 action times, and that kind of thing.
22 So it was primarily going to be directed at 23 approving the removal of certain items from the tech specs 24 and placing those items in other documents that are solely 25 under the control of the licensee.
r Heritage Reporting Corporation (202) 628-4888
12 1
What the owners groups have done is they have 2
asked that we also review in parallel some actual technical 3
changes to things like out of service times, and that sort 4
of thing.
5 So what they are telling us is they would like us 6
to review those changes in out of service times and action 7
statements, even if it means it takes a little bit longer
~
8 than we had anticipated in our schedule to just do the Split 9
Report.
10 MR. REMICK:
Thank you.
11 MR. REINHART:
Based on the new standard technical 12 specifications, implementation would be through plant-13 specific conversion of technical specifications to the new
()
14 standard.
15 We plan to have five lead plants convert to the 16 standard.
17 I want to point out, if we look at our submittal 18 dates, those five lead plants are scheduled to come in in 19 the September-October 1989 time frame.
We have a nine-month 20 review process anticipated to have a license amendment out 21 in June and July of 1990.
There is an overlap there.
22 The lead plant proposal would come in before the 23 staff has approved the new standard.
What we are trying to 24 accomplish there is get a feeling for how the standard is 25 actually implemented at a plant.
And we hope to learn apme O
Heritage Reporting Corporation (202) 628-4888
13
()
I things from'that, probably iterate back and forth before we 2
approve their new STS and ultimately the lead plant.
3 The five lead plants consist of one lead plant 4
from each of the PWR owners groups and two lead plants from 5
the BWR owners group.
6 Westinghouse would be North Ana Units 1 and 2.
7 General Electric would be Hatch 2 at one site, Grand Gulf 1 8
at another site.
I 9
Babcock and Wilcox is Crystal River 3.
Combustion 10 Engineering is San Onofre 2 and 3.
11 Following the lead plants, based on our 12 discussions with industry, we anticipate an additional 70 to 13 80 plants to be submitted for conversion between 1990 and 14 1995.
{)
15 The staff wants to process those amendment 16 requests in a timely manner and to efficiently utilize our 17 resources to do that we are developing a certification 18 process by which the licensee would certify that he has 19 adopted the new STS and that he has identified all 20 deviations and justified them.
21 MR. WYLIE:
Was there any significance to the 22 picking of those lead plants?
23 MR. REINHART:
Yes, there was.
It was really the 24 owners groups' choice.
I think it was a combination of what 25 the owners groups thought would be representative of a
()
Heritage Reporting Corporation (202) 628-4888
14 1
typical plant.
But that was modified by what was available 2
among the plant schedules, too.
3 Do you have anything to add to that?
4 On the certification process, the staff would 5
audit the industry certification of the portions of the STS 6
that they have adopted in full, and then do a complete 7
technical review of all deviations from the new STS and 8
their justification.
9 Our nine-month review process would be adjusted 10 somewhat depending on how close or not how close a given 11 licensee stuck to the new standard.
12 MR. REMICK:
You may have mentioned this earlier 13 and I may have missed it.
Will there be one SER for a group
()
14 of plants, one SER for each plant?
15 MR. REINHART:
There will be one SER, first, for 16 each new STS and one SER for each plant when they convert.
17 It would be a normal licensing amendment process.
18 MR. REMICK:
Opportunity of hearing?
19 MR. REINHART:
We have discussed that.
And 20 probably that's the way it would go.
There would be a 21 notice of opportunity for hearing posted.
22 The Commission asked us, when we talked with them, 23 to address the final policy statement in this context.
We 24 are scheduled to go to the Commission in September of 1989 25 with a final policy statement.
However, as we expressed to
(
Heritage Reporting Corporation (202) 628-4888
s 15 h)
(_
I them, and the situation still is the same today, based on 2
our review and what we are learning from the review, we may 3
consider going back to the Commission and asking.an 4
extension on that time period so that.we can factor what we 5
are learning into a final policy statement.
6 Particularly, we are asked about the adequacy of 7
the criteria, and our experience with it so far.
8' The acceptability is still under review.
There 9
are several issues and interpretations that we need to make 10 on the application.
And these need to be factored into the 11
' final policy statement.
12 The primary issue is addressed here in italics.
13 It is the need to clarify the need of the criteria to (a~)
14 capture active design features and operating restrictions 15 that keep a plant within its design bases.
16 A couple of examples of those would be the 17 pressure-temperature limits, or high pressure-low pressure 18 system valves and interlocks installed to prevent what is 1
19 commonly called an event V intersystem LOCA situation.
20 We are still learning from the policy statement as 21 we are looking at the proposed new STS.
We are seeing that 22 industry proposed to relocate some LCOs that were included 23 to be retained in the Split Report.
But we need to have 24 some discussions with them to consider their reasons and see 25 what needs to be adjusted in those cases.
Is there a valid Heritage Reporting Corporation (202) 628-4888
L I
16 f')
(_j 1
reason, or do they truly in fact need to stay in the-2 technical specifications.
3 Our priority over the next several months has to 4
be, though, the review of the new STS.
5 MR. WYLIE:
Let me ask a question.
Maybe you l
6 haven't got that far yet.
But can you assess whether or not 7
they are eliminating LCOs, adding LCOs?
8 MR. REINHART:
There are a couple LCOs that they 9
have eliminated, or proposed to eliminate.
There may be one 10 or two, I'd say, requirements that they have added.
I can't 11 think off the top of my head of an LCO that they have added.
12 MR. WYLIE:
Well, there is some discussion about 13 that some plants do not have adequate LCO techs that prevent 14 taking certain pieces of equipment out of service.
Are you
(}
15 going to look at that as you go through these reviews?
16 MR. REINHART:
If I understand your question you 17 are asking would we be looking at individual plants?
18 MR. WYLIE:
Well, I assume you are going to look 19 at it, since they are standard techs, you are going to do it
.20 on a generic basis more or less.
21 MR. REINHART:
Our basis right now has been the 22 standard.
So a custom plant out there is not really in the 23 realm of the area we are looking at.
If that custom plant 24 volunteers to adopt a new STS, and make a conversion, we 25 would assess that at the time of that licensee's conversion.
1
()
Heritage Reporting Corporation (202) 628-4888
17 (O
_)
1 Jose?
2 MR. CALVO:
Maybe I can add something to this.
My 3
name is Jose Calvo.
4 I think your question, would there be any 5
additional LCOs there, which.was not originally in.
- Yes, 6
the technical staff, as they are going through the review of 7
this, they are considering those things in.there.
Keep in 8
mind that when we had the assistant reference standard tech 9
specs, when you bring a new plant into it, you modify that 10 reference, so you create a new base for the next spec who 11 comes up to that plant.
12 Sometimes as we learn through interpretations and 13 questions that we get from the regionals, as we go through
(])
14 this, the technical staff brings some other things up, which 15 would have to be discussed between us, the staff and the 16 owners, to reach some kind of amicable resolution of those 17 things.
18 MR. WYLIE:
I got a feeling, in the answers I got 19 originally, that so far the owners groups have not 20 identified additional LCOs.
Is that correct?
21 MR. CALVO:
I believe they have identified 22 additional LCOs.
If I remember, they added about two or 23 three LCOs.
The maintenance team isolation valve, there 24 were some LCOs in there on the General Electric Plants they 25 had not considered before, and they had proposed on those Heritage Reporting Corporation (202) 628-4888 1
10
)
1 LCOs.
So there are some requirements there that are 2
important to safety.
Both the staff and the owners groups 3
will bring them forward.
And we are going to have to j
4 resolve the difference or come up with some kind of 5
resolution of those.
We have been on some of those.
That 6
is correct.
7 the biggest problem we are having these days is I 8
guess the eagerness of maybe the owners groups maybe in 9
taking things out of the standard tech speus, and the staff 10 to put them back in.
So we are going through that one.
But 11 everything so far has been worked up in a fairly cooperative 12 manner.
So we are proceeding along on that basis.
13 MR. ROSSI:
I was going to suggest that Mark, you
()
14 may give them a feel from the Split Report of about what 15 percentage of things are being taken out of the technical 16 specifications and put into licensee-controlled documents, 17 because that will give them a feel as to how many things we 18 are removing from the tech specs.
19 MR. REINHART:
The split is about 60-40.
60 20 percent of the LCOs in the existing standard would stay in 21 the new standard.
About 40 percent would be relocated to 22 licensee-controlled documents.
23 Yes, sir.
24 MR. SHEWMON:
One comment on Charlie's thing and 25 then a question.
Heritage Reporting Corporation (202) 628-4888
19 TN (s)-
1 Charlie, you might inquire what they are doing on 2
oscillatory power conditions in PWRs at 30 percent power, 3
was one example of where it might change.
4 But I'm trying to catch up with you.
There have 5
been standard tech specs that vendors have had for some-6 years, I had thought.
And now you are talking about 7
something else which has the same name which started more 8
recently.
9 Is this a matter of getting everybody to work with 10 those standard tech specs that were in existence before 1987 11 or where am I lost?
12 MR. REINHART:
Okay.
.There were sets of standard 13 technical specifications.
This is really an update of those
(])
14 existing standards.
But it is such a big job that we wanted 15 to go back.
There were a number of studies done both in 16 industry, both.at the NRC, that pointed to areas where the 17 technical specifications needed to be improved.
18 So in updating the existing standards, we are 19 trying to incorporate the results of those studies and to 20 produce a new standard.
21 MR. SHEWMON:
Okay.
So there's custom tech specs, 22 there's old standard tech specs and there's new standard 23 tech specs.
All three sets are floating around.
Is that 24 right?
25 MR. REINHART:
There is custom tech specs, there
?
(
Heritage Reporting Corporation 1
(202) 628-4888 b
7 1
20
(
1
.is standards or old standards and there is proposed new 2
standards that will be issued.
Yes, sir.
I 3
MR. CALVO:
You are correct, we have a mixed bag 4
of standard tech specs.
The older plants tend toward ESO.
5 There was very little-in those specs.
All those are common 6
to adopting this new base.
This is the new base, l
7 presumably, if everybody will come, we will have a new base, 8
a point of departure for all of them.
9 MR. SHEWMON:
And whereas before the utilities 10 could or could not update to the standard tech specs if they 11 had an old plant, now their arm will be twisted stronger and 12 over the next five years they will adopt the new standard.
13 1s that right?
()
14 MR. CALVO:
We aren't going to be twisting any 15 arms.
I hope we can sell them into the idea that they can 16 improve the safety and it would be good for the staff and 17 good for them.
It is a voluntary program.
18 MR. SHEWMON:
You expect about 98 percent 19 compliance with your voluntary program, or what do you hope 20 for?
21 MR. CALVO:
Right now, we have about 116 plants up 22 there.
Out of 116 plants, you would be talking about 23 approximately, when you combine dual plants, about 80 24 technical specifications conversions.
Right now there are 25 about 20 to 25 committed to do it.
And they are learning Heritage Reporting Corporation (202) 628-4888
21
()
I how well this foundation is going to be established.
It is 2
something that the staff is going to be hardnosed about it 3
and say well, we are not going to change this, we are 4
looking for that kind of inflexibility.
Everybody is going 5
to gain from this.
The staff is going to gain, the 6
utilities are going to gain.
Hopefully at the end, all of 7
them will convert.
But keep in mind that some of the 8
customized plants today they had committed to do it.
We i
9 have Hatch 1 that is committed to do it, we have Policy that 10 is committed to do it.
And they way they are going about it 11 I think is the way that it can be done.
12 MR. MICHELSON:
Question.
13 MR. REINHART:
Yes, sir.
14 MR. MICHELSON:
Did I understand you correctly to 15 say that the number of LCOs in the proposed tech spec will 16 be somewhat reduced?
17 MR. ROSSI:
Significantly reduced.
18 MR. MICHELSON:
Significantly reduced.
19 MR. ROSSI:
By about 40 percent, about 40 percent 20 of the things will come out and go in.
21 MR. MICHELSON:
I wanted that for a clarification 22 before my question.
23 As I recall, the LER reporting rule requires among 24 other things that if you are under an LCO, that also prompts 25 an LER.
()
Heritage Reporting Corporation (202) 628-4888
f I
I 22
/N
()
1 Is that correct?
2 MR. ROSSI:
I think the current LER rule does not 3
require you to report just because you enter an LCO.
What 4
it requires you to report is if you have actually violated 5
the tech spec.
So if the LCO runs out and you are in l
6 violation then you have to report it.
l 7
MR. MICHELSON:
That's what I meant.
8 Now, under the proposed tech spec change, when you 9
remove an LCO and put it over into some other document, 10 whatever that document might be, and you enter into such an 11 LCO and exceed it, are you still assured that you will get 12 an LER?
13 MR. ROSSI:
No, I do not believe you would.
,/ }
14 But recognize that the things that we are removing
-15 from the technical specifications under this program are 16 those items that are considered not to have as much safety 17 significance as the ones that remain in.
18 MR. MICHELSON:
Well, it is all in the eyes of the 19 beholder.
For instance, fire protection being removed.
20 What little information we are still getting on fire 21 protection will now be lost.
And there is a number of other 22 things I can point out.
23 Also, the present tech specs in many cases require 24 special reports for certain kinds of failures, even though 25 they are single component failures in some cases.
()
Heritage Reporting Corporation (202) 628-4888
p 23
)
1 Those kind I suspect will be moved somewhere else 2
or lost also.
It looks like the data base is going to 3
shrink quite rapidly.
I've read some of these special 4
reports that are required, and-they are quite interesting.
5 They don't call them LERs.
They just' call them special 6
reports.
And they are some of the most interesting material l
l 7
I have read, in fact.
8 MR. ROSSI:
Well, I'm not sure.
Somebody else may 9
be able to address the special reports.
But let me point 10 out that the things we are removing from the tech specs are 11 those that we consider to have lesser safety significance so 12 they can be removed from the tech specs, and in addition to 13 the reporting under LER and 50-72s we have the resident
()
14 inspectors that are at the sites essentially all the time, 15 and we get many reports directly from the regions in region 16 daily reports of things that are not reportable.
17 MR. MICHELSON:
But that doesn't enter into any 18 kind of an organized view that people that do PRAs and other 19 things later can go through and peruse in some organized 20 fashion.
They are lost in the regional office inspection 21 reports.
22 MR. ROSSI:
I would not really say that, becauce 23 the region daily reports come in and are thoroughly reviewed 24 every day both in AEOD and in NRR and we take actions on 25 them, and the important ones get briefed at management Um Heritage Reporting Corporation (202) 628-4888 i
24
)
1 meetings.
2 So I think that the likelihood of losing 3
information of safety significance is very low.
4 MR. REINHART:
We might add to that, the items to 5
be relocated are still going to be under controls that the 6
staff will approve.
7 For instance, fire protection goes to a fire 8
protection program.
9 MR. MICHELSON:
But that wasn't my point.
My 10 point was how does the agency as a whole continue and how 11 does the industry as a whole continue to get the information 12 if it is no longer required to be reported?
And as to 13 safety, if I believe another part of the staff, they tell us
(-}
14 that fire is probably the most significant contributor to v
15 risk, depending of course on whose risk analysis you 16 believe.
17 So I can't really understand that it is not 18 important to safety, because you've got other people on your 19 staff who tell you it is perhaps the most important 20 contributor.
I 21 MR. ROSSI:
Well, certainly, different staff 22 members have different views on the importance of thing.
23 Dut I still believe that what we are doing will not in any 24 way reduce the amount of information that we have.
And of j
l 25 course we also look at the 50-72 and LER rules periodically j
)
()
Heritage Reporting Corporation (202) 628-4888
)
25
/
(
I to see that they are adequate.
2 I don't really believe that this is a significant 3-problem.
I mean, it may be that we lose some information 4
once in a while but I believe we have enough overview of the 5
plants that anything that is safety significant we will find 6
out about and do the appropriate things.
7 MR. CARROLL:
I think you are having a 8
communication problem.
9 What Carl is saying is that very often when you 10 want to look at a safety issue you do a search of LERs.
11 MR. ROSSI:
He is correct.
You will ont get some 12 of these items in searches of LERs.
That his correct.
13 MR. CARROLL:
How will you get them?
14 MR. MICHELSON:
How much do you lose I guess is 15 the other way of asking the question.
Is this 5 percent of 16 the information we now get or 25 or 50, or do you have any 17 feel yet, or have you looked into the loss of information?
18 Because it may be that we need to start thinking of some 19 other program to pick up that information if it is a 20 significant loss.
21 MR. ROSSI:
We do have other tracking systems that 22 pick up the event briefing items and that kind of thing.
I 23 mean, there are other ways that we can search for 24 information other than just the LERs.
25 I think we will get less reports because I believe
()
Heritage Reporting Corporation (202) 628-4888
f 26 1
fewerLthings will be in the tech specs, there will be fewer 2
reports.
But it is my opinion that that will not degrade 3
our safety oversight of the plants.
4 MR. MICHELSON:
AEOD has also looked at this and 5
agreed on that point or do you know?
6 MR. FISHER:
We have consulted with AEOD and with 7
the Performance Evaluation Branches, prior to going out with 8
the policy statement and subsequent to that, and they have 9
indicated that they did not believe this would have a 10 significant impact on their data collection.
l 11 MR. REINHART:
There are other parts to the l
12 program besides the development of new standard technical i
13 specifications.
()
14 The next major part is our program for line item.
15.
improvements.
The line item improvements are ongoing.
16 They will continue e1en after the STS are issued.
And they l
17 provide changes, impressments to technical specifications 18 that are immediately available to licensees.
19 Right now they would be immediately available in 20 advance of the new STS.
They will be factored into the new l
21 STS so a licensee that adopts the new STS will not have to l
22 submit a number of license amendment requests to adopt each 1
<3 of these changes individually.
24 So far, we have eight topical reports and five 25 generic letters that have been approved.
The eight topicals Heritage Reporting Corporation (202) 628-4888 l
l
=~2 l
27 1
allow extensions for specified reactor protection system and
)
l 2
engineering safety feature actuation system allowed outage 1
3 times and surveillance test intervals.
j 1
4 We have two additional topical reports that are l
l 5
almost through review, should be out either late August,
{
6 early September, that would bring that eight topicals to l
7 ten.
8-We have five generic letters that have been 4
J 9
approved.
They make improvements to Sections 3 and 4 of the 10 tech spec, the general requirements section. They approve 11 the relocation of organizational charts to licensee-12 controlled documents.
They approve the relocation of fire 13 protection system requirements to a staff-approved fire
()
14 protection program.
15 MR. MICHELSON:
When you move it to that program, 16 will it have any reporting requirements or will it just be 17 kept at the plant?
18 MR. ROSSI:
I don't remember the details of the 19 reporting on the fire protection system requirements..But I 20 think what we did is we decided that the current 50-72 and 21 50-73 rule words were significant to cover the reports that 22 we felt we needed on fire protection.
23 MR. MICHELSON:
Inadvertent actuation which we 24 kind of track now and have found very interesting, none of 25 that would be reported under the new tech specs, as I see Heritage Reporting Corporation (202) 628-4888
28
)
' 11 it.
If you have a major fire, it would be reported.
2 MR. ROSSI:
I.think if you had a problem from an 3
inadvertent actuation that that could very well be reported 4
also.if it interferes in any way with the operation of the 5
plant.
6' MR. MICHELSON:
Only if it involves both trains of 7
equipment and so forth.
8 MR. ROSSI:
Yes, it would have to meet the level 9
that is in 50-72 and 50-73 to be reportable.
But again, you 10 know, if it is significant enough it will be reported; if it 11 doesn't rise to that level, it won't be reported.
But 12 again, I go back to the fact that we also get continual 13 reports every day from the regions, in region daily reports,
()
14 and those get just.as thoroughly reviewed within NRR --
15 MR. MICHELSON:
I don't want to go into that 16 argument again.
It's the data bases.that other people like 17 PRA specialists use and so forth when they try to decide 18 whether or not there is a problem in some area.
They don't 19 go back and look at all your daily reports that were 20 reviewed because they are lost as far as any organized 21 searching.
They go to where you can search in an organized 22 fashion.
23 MR. ROSSI:
I understand your point.
24 MR. REINHART:
The last two generic letters we 25 have listed here, the relocation of cycle-specific parameter Os Heritage Reporting Corporation (202) 628-4888
L l
29 l
/
l 1
limits, the LCO remains in the spec.
The number itself goes 2
to a cooperating limits report.
And we have also approved'a 3
relocation of radiological effluent requirements, or (RETS),
4 to a licensee-controlled document.
5 Two other generic letters that are soon to be 6
issued address the elimination of the 3.25 factor for outage 7
surveillance intervals and the allowance in the design B
features section of allowing water rods instead of zire rods 9
in fuel assemblies.
10 The final portion of the program, which we have 11 entitled "Other Support Activities," the first item 12 addresses a guidance document for 10 CFR 50-59, safety l
13 evaluations by the licensee.
^
14 Both industry and the staff recognized a need for 15 improvement in this area.
Industry requested to take the 16 lead in that area.
The staff agreed.
And industry produced 17 a document, NSAC 125, in June 1989, which provided guidance 18 in the evaluations.
19 Staff participated in workshops in June in Palo 20 Alto and Baltimore.
We have a workshop scheduled for August 1
21 16 at lleadquarters for the staf f.
And we are allowing a 22 six-month trial period with the use of this document in i
23
' anticipation of issuing a regulatory guide.
24 The second area on reduced testing, the staff has 25 completed its second drait review of its report on reduced
()
Heritage Reporting Corporation (202) 628-4888
30
()
1 testing.
We expect to go to CRGR this Summer or Fall with 2
that final report.
3 And the recommendations we intend to implement in 4
two ways, all of them through the new STS, as it is f
developed, and some of the more safety significant items we 6
intend to address through a generic letter in advance of the 7
new STS.
8 The third item, risk-based technical 9
specifications, the staff ln cooperation with the industry 10 is looking at the use of risk assessment in developing 11 technical specifications.
A number of other countries are 12 involved in doing the same thing.
We are working toward a 13 prototype configuration management system or a pilot program 14 at a plant that would use that plant specific PRA, not at 15 this time to replace, but to parallel their original tech 16 specs and try to develop this program.
17 We feel that the technical specification 18 improvement program is moving in the right direction to 19 improve operational safety and that it will result in more 20 reliable and efficient plant operation.
21 We feel that we have made and will continue to 22 make significant progress in each of these three areas.
j 23 And again, the three areas are, development of our 24 new technical specifications, parallel program for line item 25 improvements and other supporting activities.
()
Heritage Reporting Corporation (202) 628-4888
31
(
1 That conclude the portion of the technical 2
specification improvement program.
At your request, we also 3
intend to address a letter written to hlan Weise.
4 MR. CARROLL:
You will find that in Tab 2 of the 5
binder.
6 MR. REINHART:
On September 7, 1988, an anonymous 7
letter was written to Alan Weise, then of the Union of 8
Concerned Scientists, and a copy was sent to Dr. Kerr.
9 The staff reviewed the letter.
It was broad and 10 without specifics in a lot of area.
But it did indicate 11 some valid areas where we felt we could make improvement in 12 the technical specification program.
13 Thosc areas were generally already recognized and
()
14 being addressed by the staff in the tech spec program.
15 There was a Commission paper, SECY 86-10, dated January 113, 16 1986, recommendations for improving technical 17 specifications, that appeared to include most of the 18 specifics that we could find in the letter, were almost 19 paralleled in the SECY paper.
20 The conclusions of the SECY paper were that there 21 were no acute safety concerns but that we could realize an 22 increase in safety through improving the tech specs; we 23 could increase our attention to 50-59 evaluations, and that 24 we could continue to work on the use of risk assessment.
25 All of those areas, we feel we are doing in the 1
Heritage Reporting Corporation (202) 628-4888
_ _ _ _ _ _ _ _ _ _ = _
32
!( )
1 technical specification improvement program.
Particularly 2
our improved bases are designed to provide the specifics of 3
the analyses that show why LCRs or LCOs, why we have the 4
surveillance we have, the intervals' involved and why we 5
have the allowed outage times we have with those LCOs.
6 Are there any questions?
7 MR. CARROLL:
Since it was an anonymous letter and 8
since Bill Kerr got a copy of it, I guess there is nothing 9
ACRS-needs to do about the letter other than be informed?
10 MR. REMICK:
I would agree.
11 MR. REINHART:
Yes, sir.
12 MR. SHEWMON:
To go back to the general. revision 13 of tech specs program, is it your belief that this will
)
materially decrease the number of LERs?
You have said you 14 15 didn't think it was going to reduce the safety of the plant.
16 MR. REINHART:
Yes, sir.
17 MR. SHEWMON:
But one of the questions one hears 18 about periodically is the number of LERs which are make-work 19 and don't...
Jessarily represent significant risks or even 20 precursors to that.
21 MR. REINHART:
I would certainly hope that it 22 would.
23 Again, remember, the one goal is to focus 24 attention on the most safety significant requirements.
And 25 if we can home in on those, not totally ignoring other
()
Heritage Reporting Corporation (202) 628-4888
J 33 1
requirements, but really focus the operator, the licensee, 2
the regulators' attention on the cardinal safety issues, we' j
3 hope we would do that.
4 I think what we have said, if we are relocating 40 5
percent of the LCOs, that would reduce LERs involved with 6
those LCOs.
7 MR. SHEWMON:
Those 40 percent would represent 8
then the things that would generate LERs which you don't 9
think represent risk addition?
10 MR. ROSSI:
I think we need to be a little 11 cautious.
Because LERs are generated for many reasons.
1 12 Like all reactor trips generate an LER.
All actuation of ES 13
-systems generate an LER.
Failures of-safety functions.
['s)
'14 generate an LER.
All of those kinds of things.
And those u
15 will continue to generate LERs.
16 Now, there are some LERs that are generated by 17 actual, and I think it takes an actual violation of the tech 18 specs to generate an LER.
So the only LERs that will not 19 now be generated are the ones that are currently generated 20 by a violation of a tech spec with one of these systems that 21 we are going to remove.
22 MR. SHEWMON:
What was the violation of a tech 23 spec under the old tech specs won't be a violation --
24 MR. ROSSI:
That won't be a violation of the tech 25 spec.
)
Heritage Reporting Corporation (202) 628-4888
l k
34 I
(~'N
(,)
1 MR. SHEWMON:
But it will be an LCO.
)
2 MR. ROSSI:
Well, yes, it will be in a procedure 3
or someplace like that.
And even some procedures violations 4
from time to time may get reported under certain things.
I l
don't think you are goi g to see -- I mean, 5
So I 6
you are certainly not going to see a 40 percent reduction in 1
7 LERs.
You are going to see a 40 percent reduction in some 8
subset of the LERs which I suspect is a rather small subset.
9 MR. SHEWMON:
I'm not arguing for writing more 10 LERs.
I don't think that safety is proportional to the 11 number generated.
I'm rather hoping it will go the other 12 way.
13 MR. ROSSI:
I'm not arguing, I'm just telling you,
)
14 that I don't think it is going to be a big reduction in the 15 number.
I'm not arguing whether the number is good or bad.
16 MR. SHEWMON:
What's your carrot for the industry 17 if it's not to eliminate some of these things that they 18 don't think have much to do with safety?
19 MR. ROSSI:
Well, what the tech spec improvement 20 program will do is it will now bring the items that are 21 removed from the tech specs into more control of the 22 licensee.
They will be able to make modifications to those 23 under the 50-59 process without coming to the NRC for prior 24 approval.
And it also will allow the operators to 25 concentrate on the more important things to safety that
)
Heritage Reporting Corporation (202) 628-4888
_I
35
()
1 remain in the tech specs.
So we feel that it will better 2
focus their atte ion.
3 MR. CAhKOLL:
Plus you won't get cited for an LCO 4
v'.Jlation that isn't there.
5 MR. ROSSI:
Yes.
But you can still be cited for a 6
procedure violation.
I believe you can still be cited for a 7
procedure violation.
And these things will be still be in 8
procedures and other places.
So I think as far as 9
violations go, it is not clear to me that that is going to 10 change a lot of thing.
11 MR. CATTON:
Maybe you could help me a little bit.
j 12 I'm sort of confused by what you are saying.
13 On the one hand you said you were going to do
()
14 risk-based tech specs.
And fire is by many estinates half 15 of the core melt probability.
And yet you are j,)ng to 16 relocate the fire protection system requirements.
There is 17 going to be no way of maintaining a computer data base on 18 the fire systems.
I don't understand.
19 MR. ROSSI:
Well, I don't know the extent of the 20 risk that is associated with the actual failures of the fire 21 systems..
Certainly significant failures I believe will i
22 still be reported.
And I don't know, you know, we haven't 23 done a study of the effecta on reporting of these.
24 MR. CATTON:
You indicated that you were going to 25 develop risk-based tech specs.
If you are going to develop
(
Heritage Reporting Corporation (202) 628-4888
s i
36 1
risk-based tech specs, it seems to me somehow that would 2
include the major contributors to risk of which fire is one.
3 MR. ROSSI:
Well, the risk-based tech spec program 4
is a longer term program than the tech spec improvement l
5 program.
The new standard tech specs are different than 6
this risk-based approach.
The risk-based approach is a 7
longer term thing.
And I mean, by longer term, I mean many 8
years, where instead of trying to control operability of i
9 equipment through LCOs and the way it's done now with tech 10 specs, you would have an online computer that keeps track of-11 what equipment in its totality within the plant is out of 12 service at any given point in time and give you a measure of 13 pisk and you would control that risk rather than the 14 Individual LCOs.
15 MR. CATTON:
Does that mean at some point in the 16 future you might bring fire back under the tech specs?
17 MR. FISHER:
Dr. Catton, the fire protection 18 requirements, the detailed fire protection requirements will 19 be taken out of the tech specs as part of the new STS 20 development effort.
But there will still be administrative 21 controls that stay in tech specs.
And the program itself 22 will be controlled.
23 Now, while we go through the risk-based tech spec 24 effort, we may identify certain aspects of the fire 25 protection program that would be of importance to the staff (f
Heritage Reporting Corporation (202) 628-4888
37 1
and could find their way into the administrative ccetrol j
2 section of tech specs.
3 MR. CATTON:.My concern is the same as Carl's.
4 You are going to lose the information from the computer data l
5 base which means nobody is going to go dig it out, if you 6
take it away.
7 So if you are going to go to something that is 8
risk-based in the future you are not going to have that 9
information.
At least you are not going to have it in a 10 usable form.
11 MR. FISHER:
I believe that that data will not be 12 lost.
I believe that the collection of that data is part of 13 the licensee's fire protection programs and it may not be
)
reported to the NRC in the form of an LER but I believe it 14 15 is either available or it perhaps may be submitted to the 16 staff.
I'm not certain of whether it is just sitting there 17 by the --
18 MR. CATTON:
Paper on somebody's desk doesn't get 19 into a data base.
Somebody has to put it in there.
And the 20 LERs do that for you.
21 MR. CALVO:
About a week ago, I was in San i
22 Francisco with three utilities talking about this same I
23 subject.
Because there are a lot of other things going on-1 24 in that plant that are not on the tech specs.
A lot of i
25 other things to lo with the balance of the plant, and it has
()
Heritage Reporting Corporation (202) 628-4888
)
l 1
38 1
some impact on the risk, or how do you treat things on the 2
nuclear steam supply systems.
3 So all that part is being considered.
How much of 4
that stuff had to be some kind of way, got to be organized 5
in such a manner to be factored into these PRAs for the 6
risk-base, I don't know.
We are looking at that right now.
7 It is not only the fire part you are talking about.
A lot 8
of information up there is very important from the 9
standpoint of impacting that plant, which often happens, it 10 is not in the standard tech specs.
Most of the precursors 11 that you have today, those come from the balance of the 12 plant.
Those are not in the standard tech specs.
How are 13 you going to factor that into the risk base?
We are j
{}
14 currently looking into those programs right now.
And this 15 pilot program is going to tell us how we are going to deal 16 with these things.
But again, we can't collect that one, 17 what we have on the' standard tech specs.
We plan to do one 18 job now, and then we will look into that one, and I don't l
19 know what the future is going to bring and what else has to 20 be put into these computers to assess what is going on in 21 that plant.
l 22 MR. ROSSI:
I don't think we ought to put things 23 into the tech specs just to have a reporting requirement.
24 That is the wrong thing to do.
25 If we have a problem with collecting information h
(
Heritage Reporting Corporation I
(202) 628-4888
.3 m __ _ _ _ _ _ - - - - __
l-L
- E 39 1-for getting statistics on failures of important equipment, 2
that should be addressed directly through the LER rule, and l
3 that is something that AEOD looks at from time to time as to 4
whether the LER rule is or is not adequate.
5 I would also like to point out that even things 6
that are in the tech specs, that as far as I know, single 7
component failures of safety-related equipment are not 8
reportable under the LER rule.
9 MR. MICHELSON:
Except for these special reporting 10 requirements in some tech specs.
11 MR. ROSSI:
There may be some.
But in general, 12 single component failurej are not reportable.
13 MR. MICHELSON:
But most of them of importance are
/)
14 reportable through the NPRDS system.
Fire protection is 15 not.
16 MR. ROSSI:
Well, there are other reporting 17 systems.
There is the NPRDS system which picks up 18 components.
19 MR. MICHELSON:
Not fire protection.
20 MR. ROSSI:
That's true.
We do have other 21 searchable data bases.
We have information notices and 22 bulletins and that kind of thing that you can search through 23 and get some information.
24 But I think the main point I would like to make is 25 that it we have a reporting problem on collecting i
i Heritage Reporting Corporation (202) 628-4888 1
4
]
40 1
information for-use in PRAs, the way to get that stuff is l
2 not by keeping it in the tech specs, the way to get it is 3
rou go directly to the rules on reporting and fix them or 4
you fix the NPRDS or whatever.
5 MR. MICHELSON:
That is why I would expect to see 6
some kind of report from AEOD in conjunction with the 7
improved tech spec program, indicating they have looked 8
'through to see what they have lost and whether or not 9
actions are required to gain something.
And I haven't seen 10 any such correspondence.
But perhaps you have, and you can 11 send it to us, and then we can kind of look at it, for 12 assurance that they really are convinced that we still have 13 an adequate data base.
There are some members of the
(}
14 committee even that aren't convinced that single failures 15 are being adequately documented simply because they are-lost 16 except to NPRDS.
Now, in some areas, like fire protection, 17 it is even lost there.
I 18 MR. ROSSI:
You are absolutely right.
And I don't 19 know that AEOD has done a detailed study on the effect of 20 the tech spec improvement program on reporting but I do know
{
21 that AEOD continually looks at the reporting and they are 22 looking at it in the context of performance indicator 23 program to see whether they are collecting single failure i
24 data in the proper way.
25 MR. MICHELSON:
That's another question.
I O,
Heritage Reporting Corporation (202) 628-4888 l
u__________._____
i 41
()
1 MR. ROSSI:
Let me say one other thing about the 2
removal of the details of the fire protection program.
3 That particular item has already been implemented 4
in the form of a generic letter to the industry and it does 5
not have to wait for approval of our new standard tech 6
specs.
Removal of the details of the fire protection system 7
from tech specs is what we call a line item improvement. We 8
have issued.a generic letter already giving guidance on 9
exactly hov-to submit an amendment to do that.
And any 10 licensee out there who wants to remove the details of the 11 fire protection system requirements from their existing tech 12 specs can come in with an amendment request and if it 13 conforms to the guidance in our generic letter they can
()
14 remove it.today.
And many have, I think some have done it.
15 I don't know exactly how many.
16 MR. MICHELSON:
What's the number on that letter?
17 MR. ROSSI:
Do you know the number?
18 MR. REINHART:
The fire protection letter?
We 19 have it here.
20 MR. MICHELSON:
It's another one of those I guess 21 we never saw.
22 MR. REINHART:
88-12 is fire protection.
Generic 23 letter 88-12.
24 MR. MICHELSON:
Okay.
Yes, generic letter 88-12.
And again,
(
Heritage Reporting Corporation (202) 628-4888 i
i
('
l#
.]
42-1
()-
1-that takes care of the fire protection.
But it is also a 2
good example'I think of one of our line item improvements l
l 3
where we have gone out with a generic letter to the industry l
4 that says you can improve your tech specs by doing what is
[
5 in this generic letter and here is the standard way we would 6
like you to submit the amendment to'get it approved.
And it 7
is available without adopting our new standard tech specs.
8 They can come in for an amendment in that form and we will
-I 9
approve it.
10 MR. MICHELSON:
The committee has been somewhst 11 interested in how changes to regulatory requirements are 12 coordinated or integrated or whatever.
And I would expect 13 as a part of any integration package on an iter, that removes 14 material from the information-base that there be a 15 comparable document indicating why it is okay to remove l
16 this, or what other alternate actions are needed to bring 17 that information in by another route, I admir. that it i
18 doesn't have to be in the tech specs.
That I would not 19 argue with.
But I think we Imve to make sure we are not 20 losing information that we really do need.
21 MR. ROSS!:
I understand.
22 MR. MICHELSON:
And somebody should give us that 23 kind of a docuwent that shows they have looked at it and 24 they are satisfied, if you really are integrating these 25 changes with other parts of the agency.
()
Heritage Reporting Corporation (202) 628-4880
I 43
/~N i
I) 1 MR. FISHER:
I can't point to a document, Mr.
)
s 2
Michelson, but we did have extended discussion or I'did have 3
-extended discussions with Jack Heltamus and Fred Hebden and
-4 they did concur in the issuance of the policy statement.
5 MR. MICHELSON:
I would have thought that's a 6
fairly significant concurrence that would have been 7
documented by some kind of letter.
8 MR. FISHER:
Well, the concurrence is documented 9
but 1 don't know to what extent they analyzed it before they 16 gave as their concurrence.
11 MR. CARROLL:
I would like to go back to the 12 future risk-based tech specs, to get a little better feel 13 for the direction you are going there.
I guess I heard
(])
14 about a computer sitting in the control room keeping track 15 of the equipment that is out of service, as an alternative 16 to LCOs.
Can you amplify a bit on that?
17 MR. REINHART:
Yes, sir.
I have a slide that can 18 show we are to date on that program.
19 I really need to point out, as has been discussed, 20 we are in an eerly, early pilot program etage.
The goal is 21 to develop a real time risk-based configuration management 22 system.
23 And what the system will do we hope, today, I will 24 contrast today versus what we hope it will do.
Today we 25 have some LCOs that say if this happens, do that.
And that Heritage Reporting Corporation b
(202) 628-4888
_m____m__________ _ _ _ _ _. _ _. _ _ _ _ _ _ _ _. _ _ _ _ _ _ _ _. _ _ _ _. _ _ _ _ _ _ _. _
44 O(_)
1 statement on the page does not take into account the plant 2
conditions right now.
3 Now, to some extent we have an operator thinking 4
about what to do.
But he is still constrained somewhat by 5
the written word.
So we want the system to provide him 6
assistance to say no, this isn't a good time to do that 7
because of other equipment out of service, the risk is very 8
high in the plant.
Or, it might say, this is a very good 9
time to do that.
Don't wait until tomorrow.
Do it now.
10 So that is what we are looking for in this risk-11 based configuration management system.
12 We have a working group established which includes 13 individuals from the Nuclear Regulatory Commission; our
(' )
14 contractor, Science Applications International Corporation, 15 SAIC as it is commonly known; Southern California Edison; 16 Philadelphia Electric Company; Pacific Gas and Electric 17 Company.
18 I might mention also that in England at Haitian B 19 there is a configuration management system.
In effect we 20 are trying to interface with them and learn what they are 21 learning.
22 So our plans are to work together, take what 23 information we can, resolve our technical and institutional 24 type issues, identify what it is going to cost, and get some 25 guidance on software and a prototype demonstration to put
)
Heritage Reporting Corporation (202) 628-4888
45 1
this together in a manner that it is really going to help 2
the operator that it is not going to be something that he is l
3
' going to be diverted from important things he has to do but.
4 give him assistance on what._he is trying to do.
5 MR. CATTON:
Do you envision a big computer code 6
that simulates the plant in order to do these things?
7 MR. REINHART:
I'm.not sure I can address the 8
code.
Millard?
9 MB. WOHL:
The PRA will basically be the guts of 10 the system.
We have some things in mind to bring more 11 balance'to plant interactions.
But things like the PRA El:2 being done before the IPE program right now, the British 13-have a PRA base code in their essential system status
/()
14 monitor.
It's called ESSM.
There is a CRT in the control 15 room.
Dave Fisher has observed it.
He can tell you a lot 16 about it.
It is not totally automated.
And when a 17 configuration change is made, the operator is given a few
.18 choices of actions that he might do, instead of having to 19 make his own estimate.
There are several plant reliability 20 categories that are outlined as part of the code's structure 21 and he is told what category he is in, and based upon the 22 selection of instructions as to what actions might be useful 23 to take.
All the calculations for this system are done in 24 less than three minutes.
25 MR. CATTON:
It sounds to me like a look-up table Heritage Reporting Corporation (202) 628-4888
l 46
()
1 o'f some kind.
2 MR. WOHL:
Well, it is an electronic look-up.
3 MR. CATTON:
I understand.
4 MR. WOHL:
It is much harder to have a catalog of 5
pre-calculated --
6 MR. CATTON:
No, no, no.
I understand that it is 7
a computer table look-up.
8 MR. WOHL:
We have sent, Jose has sent a letter to 9
Mr. Pollock, to ask for their software, so we can look.at 10 that and see what they can do with it in' addition to other 11 approaches.
But basically the PRAs will serve as the basic 12 structure.
13 MR. CALVO:
We are not quite there.
We are in the 14 very initial program.
We will be very happy to come back 15 in the future and just give you a little more insight on how 16 this will be done.
We are hoping to give a presentation to-17 the Commission somewhere in the early part of next year, and 18 by that time I think we will have something more concrete.
19 San Onofre has volunteered to implement this program, to 20 compare what they have, and what you will have done if this 21 program will be in the plant.
We are going to learn a lot 22 during the next months we hope.
And maybe by that time we 23 will be in a better position to get into some of the details 24 and foundations of the program.
l 25 Right now, we are trying all kinds of things and
()
Heritage Reporting Corporation (202) 628-4888
47 rm()
I we are getting involved with how to collect data, other than 2
what is in the tech spec, that has an impact on the risk.
3 And those problems have not been resolved.
4 Also, together with those we have the Brookhaven 5
National Laboratory who is also helping us on this matter.
6 So we are right in the middle, a lot of things going on, but 7
we have not quite yet come up with the direction that we are 8
going to follow.
9 MR. CARROLL:
I'm sure the committee would be 10 interested when you are ready to talk about this in more 11 detail.
12 MR. ROSSI:
I just want to reiterate that this 13 program is a longer term one and different from our approval
-)
of the new standard tech specs that they talked about 14 15 earlier.
16 MR. REMICK:
Anything else?
Jay?
17 MR. CARROLL:
Not that I can think of.
18 MR. REMICK:
Is the staff finished with their 19 presentation?
20 MR. REINHART:
That concludes our presentation, if 21 there's no more questions.
22 MR. REMICK:
Any other questions from other 23 committee members?
24 (No response) 25 MR. REMICK:
If not, we thank the staff very much
()
Helftage Reporting Corporation (202) 628-4888 l
48 1
and we would like to be briefed at some future time on the 2
risk-based tech specs.
3 MR. REMICK:
The next topic scheduled after the 4
break is the nuclear power plant operating experience.
Do
'1 5
we need staff here for that?
Do we expect staff here?
l 6
MR. LEWIS:
Yes, we do expect staff.
In fact, the 7
whole thing is a staff briefing.
8 MR. REMICK:
All right. I suggest then that we 1
9 take our break, mid-morning break at this time, returning at 10 10:00 0' clock, and if the staff is not here, we.can take up l
11 our pre-meeting discussion for the meeting with the 12 Commissioners at that time.
13 (Whereupon, a brief recess was taken.)
()
14 MR. REMICK:
Gentlemen, may we reconvene?
And 15 although all the staff are not here, I believe we have a 16 sufficient number to get started.
So we will proceed with 17 the next item on the agenda, which is nuclear power plant 18 operating experience.
Hal Lewis is our subcommittee 19 chairman, so I will turn the meeting over to him.
20 MR. LEWIS:
This is purely informational.
There 21 is no proposed action.
The staff has been working to 22 analyze the trends and patterns on unplanned scrams in the 23 plants, and they are simply here to bring us up to date on l
24 what they have discovered.
I L
25 I really don't want to say any more than that, and l
(
Heritage Reporting Corporation (202) 628-4888
g,k+>$h+th*
l 40
((///
<dde, I
, MAGE EVALUATION 34, s
,.o v,,.
W V: n: B^?2
==
l,l hb I.8 1.25 1.4
_1.6_
4 150mm 4
6" h
l
/ fb s p,o sp
,,v 3 u
,6 @".s 9,o y
O 9
kb
'K
~
<w A
fjbyg^ < '~~}['
~
~
~
~~
-~~
~~
a
'49' s.
([
-1!
-just. turn it over.to the staff.
,It's all'yours.
2
- 2 -
MR. DENNIG:
Thank you,.Dr. Lewis.
'OG 3:
Tom Novak,!the Division Head, DSP of'AEOD-was-G 4
1 going to be here to sort of give some introductory remarks.
5-I willLjust:go ahead and do that.on his behalf'.
And then I-6 have affewfbrief remarks to put what.Mr.. Larry Bell, one of-7 our engineers, will be' telling you later in the technical.
8 part-of the briefing.
9 Just by'way of putting things in' context, and Dr.
11 0 ? 1Michelson will hearken back to some early days in AEOD, what' 11i we are-going ~to talk'to you about today, as Dr. Lewis 12 mentioned, is trends'and patterns analysis of a particular
'13 topic,. unplanned reactor scrams.
[a') :
14 The origin of this work goes all the way back to 8
15, really the rule change in LER reporting in 50-73 wherein for 16 the first time really we began getting systematic and fairly.
17 consistent information on the topic of reactor scrams, and a L18. onumber of aroac wo have looked'at from the 50-73 data.
19 We have done other kinds of trends and patterns 20.
analysis, for example, in the broader topic of engineered 21 safety feature actuation in general and some work in tech N
227 specs, which as you may know, are the two largest categories 23-or types of information that were brought in by the change 24 in reporting requirements.
25 The particular work, the center of our discussion Heritage Reporting Corporation (202) 628-4888
=_1x_.
1"WM y 4 l.hMyjj m
50 z.._
fry
, RJ 1
today,,is_a7NUREG' document'that'we have produced on the
'2_
. subject of unplanned reactor scrams.- -That document, like gg 3L other major AEOD studies, case' studies, has received
.w
(
14
. extensive' peer review by the staff,and.the industry and in u
5
'particular we want to point out that the owners groups were 6'
involved in that peer review.
7 And_in that peer review process, which in 8
.accordance with AEOD procedures was purely for the purposes 9-of obtaining; checks on the factual' content of the report and 10 not onJany of the, conclusions or any recommendations that V
11.
might:be'in there, we did find, coming back from the 12 industry and the: owners groups, that this effort was found 13.
to be useful, Land continues to be so as a compilation that n
(
] ]{-
.14) does go across the. industry at a level of' detail that-is not L
15-being captured in other places, and the study continues to b
16' be used today as a cross industry reference by people for
'17 '. example like INPO.
-18 Very briefly, just some history of the origin of 19 the study and the documentation.
20 As I mentioned, the report was put out for peer l
21 review.
That occurred back in June of 1988.
Peer review 22
-comments were finally received from all parties and 23 incorporated.
And let me say that the peer review comments b )-
f 124' were fairly minor, but they were slow in coming in.
7 25 Completed that in December of 1988.
Finally had the thing b
Heritage Reporting Corporation L
(202) 628-4888 f
K --
t_A__-----'__-x-
1 51
\\m- [
1 in NUREG' shape and ready for publication in March.
2 Now, by that time, since the report itself really 3
focuses on a period of time that ends in 1987, we felt that 4
it was necessary to update that and to pair the NUREG with a 5
companion addendum document which took the major highlights 6
refinings into 1988 and in fact, we got a little bit, first 7
quarter of 1989.
8 So we went ahead and did that, and had that ready 9
in May.
And so this addendum that is pointed out in the 10 slide is issued in conjunction with the NUREG 1275 Volume 5.
11 The last thing that I want to talk to you about 12 before Mr. Bell comes up to give you some technical 13 highlights from the report, is what really we were trying to
()
14 do.very simply in the spirit of the AEOD charter, was to 15 provide feedback on operational experience in this 16 particular area.
17 As I mentioned, we are one of the fea places that 18 has the charter really to look at all plants all the time.
19 The owners groups have their own data bases and focus on 20 their own plants and their own issues.
But for the most 21 part they will not get into what is happening at the other 22 guy's shop.
And we have it all in one place.
23 Naturally, we wanted to track down causes of 24 unplanned scrams.
And in this particular effort, in this 25 particular study, we really had something to work against in
(
Heritage Reporting Corporation (202) 628-4888
_________J
i-I l
l 52 o
' (,)
1 the way of'what industry was doing. We had an objective of 2
Laking the programs that were in place or being put in place 3
by owners groups in concert with NSSS designers, in the way 4
of a scram reduction program, look at what they were 5
focusing on, look at what their modules were, what areas 6
they were hitting on, and to see if those were the areas 7
that were indicated by operational experience.
So we took 8
our objective look.
We looked at what they were doing and 9
did a matchup to see if they were working in the right 10 areas, to the right level of emphasis.
11 And with that, I would like'to turn it over to 12 Larry Bell, who will walk you through some of the highlights 13 really of what is in the report.
There is quite a bit of
{}
14 material there, and I would hope that what our briefing will 15 do is pique your interest and get you to look into some of 16 the details that are captured in the report.
17 MR. CARROLL:
At some point in the presentation I 18 guess I am interested in the issue of how important from a 19 risk point of view is scram reduction?
I know we all talk 20 about it because it's an easy thing to measure.
It 21 certainly has been a precursor to the one serious accident 22 we have had.
But overall, is it really important to risk 23 that the NRC be this worried about it or is it an economic 24 issue for the utilities?
25 MR. DENNIG:
Let me very quickly take a stab at
()
Heritage Reporting Corporation
~
(202) 628-4888
r 53 g
\\_4 1
that.
2 In places where there is a quantification of risk 3
I think the general result as far as things that dominate is 4
not in transient initiators.
And in terms of reducing risk, 5
in general, reducing the rate of transient initiators, 6
things that cause scrams, is not a dominant issue.
So I 7
think purely from a' risk analysis standpoint, my 8
recollection is it is not an area that you go after first 9
purely for risk reduction purposes.
10 Now, having'said that, in doing this work, and in 11 particular at the level that scrams, the number of scrams 12 per plant is reached today, after some reduction, keeping a 13 low scram rate in place requires a level of attention to
()
14 detail and diligence at the plant in a range of activities, 15 whether it be operations or maintenance, that we feel is 16 beneficial across the board.
17 So as a goal, as a thing to focus on, we think 18 scram reduction has benefits spread out through the plant.
19 MR. KERR:
To what point do you think the scram I
i 20 rate should be reduced?
21 MR. DENNIG:
The words as low as reasonable 22 achievable come to mind.
23 The industry has set a goal for itself that is a 24 rather ambitious goal.
l 25 MR. KERR:
But the industry setting a goal may be Heritage Reporting Corporation l
(202) 628-4888 l
l
54 1
based on things other than safety, which it would-seem to me
~
2 would influence the goal'of NRC more than it might industry.
3
_ Industry can have a variety of_ reasons for wanting.to reduce 4
5-MR. DENNIG:
As you know, we have not set a 6
specific target or goal.
7 MR. KERR:
Well, you have said that the present 8
rate is too high, otherwise you wouldn't be trying to reduce 9
it.
I mean, implicitly, if you are trying to reduce it, the 10 implication that the existing rate is too high.
Or am I 11 missing something?
12 MR. LFWIS:
I think you may be beating on the 13 wrong guys.
Of course I agree with you.
But I think you
(]}
14 may be. beating on the wrong guys because these, guys simply 15 studied what the trend has been, if I understand correctly, 16 and trying to find the reasons for it.
17 MR. KERR:
I'm not trying to beat on anybody.
I'm 18 trying to understand if they have such a goal.
19 MR. LEWIS:
Forgive me.
The word " beat" may have 20 been badly chosen.
But 1 think the question you ask is a 21 deep and important one, and these guys I think are just 22 studying what the experience is and trying to find what the i
23 causes are.
Am I wrong about that?
I 24 MR. DENNIG:
No, Dr. Lewis, you are not wrong in 25 that.
I 1
i Heritage Reporting Corporation (202) 628-4888 l
.(.-
h:
55 m
. $_)
1 MR. WARD:
Yes, but why are they studying it if 2
there's not --
3 MR. LEWIS:
That's a good question.
4 MR. WARD:
-- an important safety or risk 5.
implication?
I think that is the underlying reason for 6
Bill's question.
7 MR. KERR:
I guess the answer is you don't have a 8
goal at this point.
9 MR. DENNIG:
No.
We in AEOD would not I think be 10 in the business of setting such a goal.
11 MR. KERR:
You do occasionally talk to other parts 12 of the Commission staff who are involved in risk reduction, 13 I assume.
So you don't know whether such a goal exists
()
14 within the Commission or not?
15 MR. DENNIG:
I'm fairly certain that there is no 16 such goal, per se, set within the Commission.
17 MR. KERR:
Okay.
18 MR. CARROLL:
INPO of course has set some targets 19 by years in the future.
20 MR. LEWIS:
But what is the specific answer to Mr.
21 Ward's question?
Why did you do this?
Who put you up to 22 it?
23 MR. DENNIG:
Harkening back to my earlier lead-in, 24 one fundamental reason, that goes back to the design of the 25 trends and patterns program, its origins in the change in
(
Heritage Reporting Corporation (202) 628-4888
56
(
lV) r 1
-the reporting requirements, with the two-fold path, if you 2
will, of examination of individually significant events and 3
then a collective examination of those events that per se 4
wouldn't merit a case study or generate a case study.
5 So that is one kind of a programmatic 6-justification for doing it.
7 Another justification for doing it is the intense 8
interest that wan focused on this particular topic across 9
the agency and within the industry and to provide an 10 objective point of view to keep the staff informed on what 11 was going on vis-a-vis what the industry was doing, and to 12 bring some objectivity to the area.
13 Another motivation for doing the work is that
()
14 because of our unique position in looking across all tre 15 plants and our interest in the entire industry, and our 16 ability to get down to a fairly fine level of detail, we 17 were able to feed back to the industry, to the owners 18 groups, some useful information as far as the areas that 19 they ought to be focusing in, the things that were coming to 20 our attention, the messages we were getting from the 21 information.
22 And we hope, and it is our intention and our hope, 23 that that process indeed does make things better.
And so we 24 are not just studying these things just for the sake of 25 studying them.
But at what level you cut these kinds of
(
Heritage Reporting Corporation (202) 628-4888 l
I i
1 J
.i r
57
? s.
l,)
1 things off because for some reason the frequency has. dropped g
2 below the regulatory concern threshold, that is a difficult 3
issue.
4 MR. CARROLL:
But I'm not sure that what you said 5
has really gone across the agency.
There are an awful lot 6
of people that are still pushing scram reduction as a very 7
important thing from the point of~ view of risk.
8 An example of this is that it was the whole 9
justification in the maintenance rule for branching off into 10 the balance of plant.
I mean, these are precursors that are 11 extremely important to public health and safety.
And the 12 answer you gave me earlier was that they are not.
So I 13 don't think people understand that point.
{}
Does your report state how important these are to 14 15 the overall risk point of view?
16 MR. DENNIG:
This particular analysis does not 17 treat that context.
We start more or less with a given.
18 MR. BELL:
I notice Tom Novak has come in, if you 19 would like to say something before I get started.
20 MR. NOVAK:
No, no.
You guys are doing fine.
21 MR. DELL:
What I intend to do is just to discuss 22 the approaches used to collect and to evaluate the data for 23 the mature plants through 1988. I will try and provide an 24 overview of all the unplanned scrams at mature plants and 25 proceed from there into a discussion, and finally into a
()
Heritage Reporting Corporation (202) 628-4888
L L
1-58 i_/s 1
discussion of the mature plants by NSSS vendor.
I 2
I would finally like to get into the conclusions 3
of the report.
'4 Data for approximately 2,300 scrams has been 5
extracted from LERs since 1988.
Once extracted, the data is 6
entered into a data base and once in the data base we have 7
the capability of cutting that data by the particular 8
variables that we used ao input to the program.
9 A key function, a key objective, I should say, of 10
.this particular study, was to take a look at what the scram 11 data was saying and' compare that data with owners groups' 12 scram reduction programs.
13 The types of data that we extract from the LERs f) 14 are listed here on the transparency.
This one may need a 15 little explanation.
It is really manual or automated 16 scrams.
The plant's data says we are talking about whether 17 the plant was starting up, shutting down, or in a power 18 reduction mode.
And one that is not here and not on the 19 next slide either is the initiating system for the scram.
20 MR. WARD:
The initiating system, Larry, or the 21 signal?
22 MR. BELL:
No, the system.
What we did, we took a 23 look at that system that caused the scran as a result of a 24 component failure or human error.
The signals, we captured 25 data associated with the signals, but we did not
(
lieritage Reporting Corporation (202) 628-4888
i 59
. g(_)
1 particularly use it for any detailed analysis.
We just 2
wanted to know what signal was there, and it's a part of the 3
data base.
But what we were really after was those systems 4
where something happened that initiated the sequence of j
i 5
events leading to the scram.
6 The causes we looked at are. fairly high level 7
causes.
What we were trying to do was to categorize the 8
scrams at a point that we could get the size of the number 9
of scrams that we were coing to look at down to a reasonable 10 size so we could do additional analysis.
11 We took a look at the people who were responsible 12 for the personnel errors.
And what I am talking about there 13 are licensed operators, unlicensed operators, technicians,
(}
14 even the janitors that wander through the plant and bump 15-something and cause a trip.
16 The cause of the human errors we capture.
And 17 that is basically things like did the operators or the 18 maintenance people follow procedures when they were doing 19 things, were they adequately trained, were they paying 20 attention to what they were doing, or was it just a 21 cognitive error that occurred while they were doing 22 something?
23 We have an extensive component list that we 24 capture and we capture this also at a fairly high level and 25 we get into the details of it when we see a particular trend
()
Heritage Reporting Corporation (202) 628-4888
60
,mV 1
develop.
2 This diagram-is an example of our analysis method.
3 Basically, what you are seeing from this point to 4
this point is the extraction and input of data into the data 5
base and the manipulation of that data to try and identify 6
trends.
7 MR. MICHELSON:
Could I get a clarification?
8 MR. BELL:
Sure.
9 MR. MICHELSON:
Scram is not always caused, not 10 always a direct consequence of some certain initiating 11 event.
There may be an initiating event which goes down 12 through a few occurrences and then along the way the reactor 13 scrams.
(~))
14 How do you pick up those?
Is that a part of this 15 data base or are you just looking for shall we say direct 16 scrams from certain direct initiating causes?
17 MR. BELL:
We are looking basically for that 18 system that is causing the whole sequence of events that 19 lead up to the scram.
20 MR. MICHELSON:
Maybe a water pipe, a small pipe 21 ruptured and ran water on the floor and the water ran across 22 and it got into some boards and it caused things, and 23 finally along the way, the reactor scrams.
How do you pick 24 that up then?
As a pipe rupture?
25 MR. BELL:
We would pick it up as the initiating l
()
Heritage Reporting Corporation (202) 628-4888 I
L_______--__ _ --
i
61 4
1-system would'be that pipe that broke that caused the water 2
to --
3 MR. MICHELSON:
Okay.
You go all the way back to 4
the root cause of having experienced the scram?
5 MR. BELL:
Right.
6 MR. MICHELSON:
Okay.
7 MR. NOVAK:
Every; reactor trip is in'our data 8
base.
9 MR. MICHELSON:
Yes.
Because most every one, j
10 hopefully, most times you do, you want a scram along the 11 way.
But then you do go.back to root cause of the scram.
.R.
BELL:
Right.
12 M
13 MR. NOVAK:
Yes.
And arguably you read the LER (v~)
14 and if something was upstream of what they told us, we'd 15 never know it.
But we look for what we consider to be the 16 initiating event in the sequence of things that happened 17 which somewheres along the line a reactor trip occurred.
18 MR. MICHELSON:
Okay.
Thank you.
19 MR. KERR:
I thought an earlier comment indicated 20 that you were looking only at unplanned scrams.
That's not 21 the case, then?
22 MR. BELL:
Yes, that is the case.
We are looking 23 at unplanned scrams.
l 24 MR. KERR:
I thought Mr. Novak said every scram.
25 MR. BELL:
Every scram that wasn't planned, non-Haritage Reporting Corporation (202) 628-4888 1
62
!3
(/
1 preplanned scram.
2 MR. KERR:
Okay.
3 MR. CARROLL:
That would include manual scrams 4
that are not preplanned?
5 MR. BELL:
Right.
That includes manual scrams 6
that are not preplanned.
We'll get into that.
7 I want to emphasize that from here to here what we 8
are doing is inputting data, manipulating data and taking a 9
look at a particular system based on initiating systems.
In 10 this particular case we looked at an electrical distribution 11 system as the problem.
And these are the systems that were 12 involved in scrams and we get down to the details of looking i
13 at whether it was a capacitor or resistor that failed that
()
14 caused the whole thing to go back to cause the scram.
15 MR. MICHELSON:
You oftentimes do not know that 16 level of detail from reading an LER.
How do you get it?
17 MR. BELL:
We do, what we are doing is, we could i
19 never de this, if we tried to do that in a data base.
Okay?
19 So what we do is, we capture the information by system and 20 then when we see a trend, we go back and we revisit that 21 LER, plucking out details and trying to figure out what 22 really went wrong.
In recent years, the LERs do get down to % g 23 the detail where you can pick it out.
J 24 MR. MICHELSON:
I'm surprised you have to read the 25 LER.
I thought this would be one of the branches in the Heritaga Reporting Corporation (202) 628-4888
0 p.,
"' J 1
sequence' coding and search system.
2 MR. DENNIG:
When we first started doing this 3
scram trend analysis, we visited sequence coding and search 4
system and tried to use the sequential logic that is 5
embedded there.
6 There are some technical difficulties with the way 7
the data base is arranged as far as. tracking back from a 8
scram to a cause back up a chain, and knowing for example 9
how far back up that chain to go.
I'll give you an example.
10 In sequence coding, because of the way that it is 11 structured, as you know, Carl, there is a cause on each 12 line.
And quite often, and with some logic, the first step 13 in a sequence has a cause of unknown, in the first position
()
14 on that first step.
15 So in trying to do an automated analysis, at a 16 very fine level, backing up, you are left with a lot of 17 resultants, a lot of unknowns.
And so we couldn't get our 18 hands around it, via sequence coding.
19 And at that time, also, there was some detailed 20 information about the signal that caused the scram, which we 21 wanted to pick up.
22 MR. MICHELSON:
I'm surprised though, if it is in 23 the LER, which I gather it is, that it wasn't in the coding.
24 That's the part that bothered me.
25 MR. DENNIG:
Well, it is in the coding, Carl. It l
IIeritage Reporting Corporation (202) 628-4888 1'
'64
/'\\T f
1 is just a matter of trying to, for this particular topic, to 2
back up to the root cause and link it to the scram with the 3
structure that was in place in sequence coding.
It wasn't a 4
trivial matter.
5 MR. MICHELSON:
It wasn't one of the things that 6
one was supposed to be able to do.
7 MR. DENNIG:
Well, we took a shot at it.
And 8
there would be ways to fix it.
And there has been' feedback 9
from this process into sequence coding.
We've added some 10 information.
But as I say, when we began the effort, which 11 was a number of years ago, because of that and because we 12 wanted other specialized, information specialized to scrams, 13 like you said, a signal was one of them, power level
()
14 scrammed from, and some other things, that the efficient'way 15 to do it was just to do a D-base file and proceed that way.
16 MR. MICHELSON:
Okay.
17 MR. BELL:
What I would like to do is use the next 18 series of about three slides to demonstrate examples of how 19 we go and track particular trends that we see, 20 I'm going to look at examples of systems, of plant 21 activities in progress, and causes.
22 And if you take a look at this, this happens to be 23 the Westinghouse plants, and if you look across, you can see 24 what is happening with the feedwater initiated scrams.
25 To this point, I don't necessarily want to get Heritage Reporting Corporation (202) 628-4888
n 65 L
1
]
1 into any more of the details here. 'What we did is, for the
-2 systems, we used the NPRDS data base to bound the systems 3
for purposes of placing the blame on a particular system, i
4 What we mean by activities here is basically what 5
was going on --
6 MR. WARD:
Larry, would you mind going back to 7
that previous slide a minute?
8 MR. BELL:
Sure.
9 MR. WARD:
I'm trying to sort this out by which 10 scrams are unnecessary, spurious scrams, and which are 11 necessary scrams.
And the main feedwater scram, are those 12 all cases where the feedwater system tripped and the scram 13 was necessary?
14 MR. BELL:
Yes.
Yes, in most cases that is what O.
15 is happening.
Some transient occurs in feedwater that leads 16 directly to the scram.
17 MR. WARD:
Okay.
Well, which of those up there 18 are what we call strictly spurious scrams where a scram was 19 not necessary to protect the reactor, to protect any system?
20 MR. CARROLL:
A technician screws up on doing a 21 surveillance test.
22 MR. BELL:
That could happen in any number of the 23 systems.
But you will probably see it more or less in the 24 reactor protection system where they put the plant in a 25 position where you only have one channel left and they screw O-Heritage Reporting Corporation (202) 628-4888 l
l 66
- r s?
1 up and the plant trips.
2 MR. WARD:
Okay.
So the rest of them are what I'd L
3 call real scrams, necessary scrams.
4 MR. BELL:
Right.
i T
5 MR. WARD:
And those are the spurious scrams, 1
6 then.
7 MR. BELL:
Right.
8 MR. CARROLL:
I guess there is also another l
9 problem there in my mind.
10 I think you almost have to distinguish between 11 plants, Westinghouse plants with motor-driven.feedwater 12 pumps and those with turbine-driven feedwater pumps to paint 13 a good picture at'least for me.
()
14 MR. BELL:
We did that in the report.- We did go l
15 into a discussion of motor-driven and turbine-driven pumps.
16 MR. MICHELSON:
Was 1989 a particularly bad year?
17 MR. BELL:
This is only first quarter data.
18 MR. MICHELSON:
But it already exceeds all of last 19 year, like on main feedwater, and if I multiply it by four.
20 MR. CARROLL:
No, it's per thousand.
21 MR. MICHELSON:
Oh, this is a rate.
I'm sorry.
22 Okay.
That's just the rate.
Then it's not significant.
23 MR. NOVAK:
Carl, just to answer your question, 24 maybe it would help to go back to that earlier discussion.
25 If they were absolutes, we'd say hey, these plants are 1
(
Heritage Reporting Corporation (202) 628-4888
)
t 1
l 1
___J
67 l
ra fN -)
1 operating with a lot more trips in feedwater.
But it'is
~
2
.just the rate.
And we are watching it.
And we've got some 3
observations.
And we'll be glad to share them with you at 4
the end of the presentation on what we see 1989 turning out 5
to be, o
6 MR. KERR:
Maybe you have previously covered it, 7
but did you pull out those scrams that were caused by 8
technical spec required surveillance?
9 MR. BELL:
That's included in the table that's 10 coming up.
11 MR. KERR:
Okay.
12 MR. BELL:
That kind would be tied up in here.
13 These are testing, surveillance testing, in all probability,
()
14 related scrams.
Some testing or certain surveillance was in 15 progress that had a direct impact on the scram.
16 And we can cut these things by systems and 17 activities to take a look at which systems they were working 18 in and so..rth.
19 MR. KERR:
Now, are these scrams at power or 20 scrams at any state?
21 MR. BELL:
These are scrams usually -- Let me give 22 you our definition of what a scram is.
23 We call acrams, any scam that occurs with RPS 24 actuation in rod motion.
So the plant is --
25 MR. KERR:
For example, a scram that would occur Heritage Reporting Corporation (202) 628-4888
f l
68
(.
fn(-
1 at Brown's Ferry --
T 2
MR. BELL:
No, that's not in here.
No, that's not 3
in here.
4 MR. KERR:
So it is a scram in an operational 5
plant..
6 MR. BELL:
Right.
7 MR. KERR:
But it could be at very low power.
8 MR. BELL:
That's right, low power or at full 9
power.
10 MR. DENNIG:
The lion's share are split at the two 11 ends of the power spectrum.
They are at 90 percent power 12 and above, is the biggest chunk and then the next biggest 13 chunk is down in the 20 percent below' range for startup
()
14 difficulties.
15 MR. REMICK:
Another question, on the previous 16 slide.
You need not go back.
But I assume if there was a-17 personnel error associated with main feedwater, that would 18 be reported under main feedwater, not the other categories?
19 MR. BELL:
Right.
We can identify by system the 20 causes for all the scrams associated with that system.
They 21 basically fall into four categories:
the equipment-related 22 problems or failures; personnel errors or human errors; and l
23 we have a special category we throw in for the PWRs that 24 just takes a look at steam generator-level problems, usually i
25 in the startup phase of manually starting a plant up;
{
Heritage Reporting Corporation l
(202) 628-4888 l
l l
69
(
1 procedural problems; and very rarely we see some that are 2
known as -- yes.
3 MR. LEWIS:
Testing is rather routine and trouble 1
4 sheoting would be where you know something is wrong and you I
5 may still be following a routine but it is not regular 6
testing?
7 MR. BELL:
Right. You have a problem and you are I
8 trying to identify what the problem is, and basically trying 9
to fix it while you still have power.
10 Just briefly, this is the slide covering the high 11 level of causes that we look at to try and categorize what 12 is causing the scrams, and basically en route.to additional 13 analysis.
()
14 MR. CARROLL:
You probably have some difficulty 15 separating out or distinguishing human error and steam 16 generator-level.
An awful lot of steam generator-level 17 trips or high-level trips as a result of human error in some 18 form, but you often also have some malfunctioning level 19 controller or something else involved.
20 MR. BELL:
You're right about that.
21 MR. CARROLL:
They are not clean and crisp
'22 categories I guess is my point.
23 MR. BELL:
Yes.
that is precisely why we chose to 24 do it that way.
We could either roll it back up if we chose 25 to do so for a particular evaluation, or we could look at it
(
IIeritage Repor t.ing Corp' oration (202) 628-4888
70 AV 1
in both ways.
2 This table is basically an example of looking at j
3 the equipment-related failures of the feedwater eystem.
4 What we are doing here is we are going through and 5
we can say well, this is the level of feedwater scram trip 6
-rates.
This is the contribution due to equipment problems.
l 7
And that is an error right here.
That should be a.07.
And 8
then we can take a look at the feedwater regulating valve 9
problems.
And we found that the main feed pump problems and 10 basically steam generator-level problems were the primary 11 sources of main feedwater system-initiated scrams.
12 This table is available for all the NSSS vendors, 13 for those systems we found to be the major contributors to
()
14 scrams.
15 I would also like to point out that 1275 has the 16 individual plant problems defined in the appendix, and we go 17 through and look at the particular LER that it came in, the 18 cause, the root cause of the scrams, and so forth.
19 MR. CARROLL:
Is the picture about the same for 20 this slide in 1988 and 1989?
21 MR. BELL:
Well, what we are seeing for 1988 at 22 least is, as you can see here, it is lower.
The feedwater 23 reg. valve in the main feed pump seems to be going down a 1
24 little bit.
1989 we haven't looked at in enough detail yet.
I 25 This is kind of a simple-minded way of looking at i
O(,/
Heritage Reporting Corporation (202) 628-4888
71 li ' (
1 what is going on with the overall trends of plants that we-2 were looking at.
1 3
What I would like to kind of talk about briefly is 4
these manual scram rates are staying fairly constant.
They L
5 stayed fairly constant from this period to this period.
It 6
dropped a little bit from here to here.
And it is staying 7
fairly constant even with the reporting requirements that 8
the PIs require no manual scram.
9 Yes.
10 MR. SHEWMON:
Do you find a manual scram as 11 opposed --'I guess I know what an automatic is.
This means 12 the operator pushes a button when he sees it is going to go?
13 MR. BELL:
Right.
14 MR. EHEWMON:
But it says nothing about precursor 15 or what. caused it?
It's just how it happened to occur?
16 MR. BELL:
Well, he can see he is going to have a 17 problem and the plant is probably going to scram anyway.
i 18 And if it is his opinion that it is safer to trip the plant 19 at that point, he goes ahead end does it.
20 We don't treat our analysis of those any 21 differently.
We go back and take a look at the root cause 22 for the scram.
23 MR. CARROLL:
It is not always that it is going to 24 trip.
25 MR. BELL:
It's his opinion.
[')
Heritage Reporting Corporation
's (202) 628-4888
1 i
72
(~5 i
(/
1 MR. CARROLL:
For example, a serious' condenser 2
tube leak that he doesn't think he can reduce power quick 3
enough to avoid salting up his steam generator, as he may 4
red-handle it manually.
5 MR. REMICK:
I assume this is for all mature 6
reactors.
7 MR. BELL:
Right.
8 MR. REMICK:
Not just Westinghouse.
9 MR. BELL:
This is all mature reactors here.
I'll 10 give you a feel for what's happening here.
11 MR. CARROLL:
Why don't you tell us what " mature" 12 means before you do that?
13 MR. BELL:
Okay.
A mature plant by our definition
)
14 is any plant that is operating with experience grade of 24 15 months after OL issuance.
16 MR. CARROLL:
After OL issuance?
17 MR. BELL:
Right.
18 MR. CARROLL:
Okay.
19 MR. BELL:
Just to give you a feel for what is 20 happening here, the delta between here and number is scrams 21 is about 46 scrams, from here to here is about 50 scams and 22 from here to here is about 45 scrams overall.
23 What I am trying to show here is just really the 24 overall scram trend for all the mature plants.
And as you 25 can see, you've got a fairly significant scram reduction l
O' Heritage Reporting Corporation (202) 628-4888
73 i
f's l
(m) 1 from 1984 down to 1988.
2:
Seme interesting things pop out though.
In 1985 3
and 1986, mature plants improvements can be -- I'm sorry.
4.
1985 and 1987, this reduction and this reduction can be 5
basically attributed to the Westinghouse fredwater system 6.
performance.
The 1986 improvements are basically tied to the GE 7'
l p
B plants.
9 MR. CARROLL:
But what happened, for example, in 1
10 1985 and 1987, with the Westinghouse plants?
11 MR. BELL:
Okay.
Westinghouse plants 12 significantly reduced problems with feed reg valves and main 13 feed pumps, main feed pumps, and they reduced the number of
()
14 personnel there associated with scrams caused by that 15 system.
16 MR. CARROLL:
Physically, what was done?
How did 17 this happen?
18 MR. BELL:
Physically what was happening was the 19 owners groups had been making information available, and 20 Westinghouse very well knew that the feedwater system was 21 driving their scram rate.
So what they did, they identified 22 it as a problem, they made this information known to all 23 members of their users groups. Individual plants went in and 24 started paying attention to how they, for example, changed 25 the lube oil in the main feed pump, how frequently they did Heritage Reporting Corporation (202) 628-4888
l t
74
- (~%
(JL 1
'that., how~ frequently they paid attention to how they had air 2
lines going to valve operators mounted.
-Of. course a number 3
of them were due to vibration breaking.
That is what they 4
effectively did.
5 MR. CARROLL:
How common was it to for example 6
junk the Westinghouse feed pump. turbine Governor system and 7
put a Lovejoy system on or that kind of~ thing, major 8
modifications?
9 MR. BELL:
I'm not aware of any major 10 modifications.
11 MR. CARROLL:
I know a lot of Lovejoy systems were 12 sold in that time period, because I bought a couple of them.
13 MR. DENNIG:
In the history of things so far, the
(}
14 major hardware mods have not played a big role in the scram 15 reduction.
They have really focused on people things, for 16 the most part.
.17 As Larry indicated, a lot of things he is talking 18 about are better maintenance.
Some folks used to do 19 maintenance on feed reg valves.
Number one, it wasn't 20 scheduled.
And number two, when they did it, they didn't 21 use a procedure.
So people started doing those things.
22 Another thing in the Westinghouse case that they 23 did early on was they pulled a lot of experience from the 24 best operators in terms of bringing a plant up in manual to 25 auto transition, and they held workshops and fed that back.
Heritage Reporting Corporation (202) 628-4888 u_
75 1
And that had an impact on the low end.
2 MR. BELL:
1988, I guess, the reductions we noted 3
there were a bit unusual.
They were really driven by CE and 4
General Electric plants.
5 CE had been running along at a fairly constant, 6
slightly decreasing scram rate.
Then all of a sudden in 7
1988, they did the same thing effectively that Westinghouse 8
did in 1985.
They got hot and fixed their feedwater 9-problems.
10 And what I see is also unusual about CE plants in 11 1988, they had absolutely no turbine-initiated scram.
And 12 so far in 1988 they have already had some.
So we wouldn't 13 expect that to continue.
14 What I would like to do now is get into the O
15 individual NSSS vendors, what their experience has been.
16 As I mentioned before, the dominant system at 17 Westinghouse plants has been the feedwater system.
1985 18 marked a real significant dropoff in the main feedwater 19 system-initiated scrams.
They sat there for a while through 20 1986 and no significant change was noted, and then again in 23 1987, they had another big feedwater system fix or reduction 22 in scrams caused by that system.
23 In 1988 again they are staying fairly constant.
24 They are at a level right now that they are going to have to' 25 probably do modifications, or it is getting harder, that is l
l
(-}
Heritage Reporting Corporation (202) 628-4888 L
.76 I
what we feel, to get it down any lower than they are now.
2 MR. CARROLL:
Does main turbine include the 3
generator?
I guess it doesn't, does it?
4-MR. BELL:
No.
In our look at the world, it only 5
includes the turbine.
The generator-is considered 6
separately.
7 MR. CARROLL:
Okay.
16 8
MR. BELL:
In fact, I will show you.
Here is the 9
main generator pulled out from the main turbine.
10 MR. SHEWMON:
In your report, do you compare the 11 scram rate, or can you on this basis, with what happens 12 overseas in some of the better run plants, either in Japan 13 or in France or Germany?
()
14 MR. BELL:
Are you asking me did we do that, or 15 can it be done?
16 MR. SHEWMON:
Did you do it?
17 MR. BELL:
No, we didn't.
This was only a U.S.
18 experience.
19 MR. SHEWMON:
It might be interesting to see how 20 this profile compares with that profile to see whether or 21 not they pretty well match up or there aro some real 22 differences.
23 MR. DELL:
What I would like to point out here and 24 move on from this table is really that the main feedwater 25 reactor protection and the turbine systems are the primary Heritage Reporting Corporation (202) 628-4888
_ 1 77
(~}
1 initiators of scrams at Westinghouse plants.
2 One could take a look at feedwater and see the j
l 3
rapid, the increasing decline in the scram rate.
In a 4
similar fashion, there is some decline to be noted for the 5
reactor protection system, but the turbine system, that 6
system doesn't appear to have a particular trend associated 7
with'it.
8 Changes are due primarily to problems associated S
with personnel errors.
As the number of those go up, the 10 number of scrams go up.
1 11 MR. CARROLL:
Transmission problems, main 12 transformer problems, that sort of thing, are under other?
13 MR. BELL:
No, here.
In the electrical 14 distribution system.
7,,U 15 MR. CARROLL:
Oh, distribution.
I thought it was 16 transmission.
17 MR. BELL:
Well, what we did was we took a look at 18 where the power comes into the plant, and if a problem 19 occurred out on the line someplace, it reflected itself back 20 through that electrical distribution system.
That's how we 21 looked at it.
22 MR. CARROLL:
Okay.
So a transmission problem 23 would be in the category of electrical distribution.
24 MR. BELL:
Right.
Right.
25 MR. WYLIE:
But the main stepup transformer is in l
)
Heritage Reporting Corporation (202) 628-4888
j 78 l
L o.
l U
1
-- ?
l 2
MR. BELL:
Right, that's there.
3 MR. CARROLL:
That's not the usual industry 4
definition.
j l
5 MR. BELL:
Yes.
6 General Electric plants, kind of what they did was 7
they had an increase initially from 1984 to 1985 and then 8
from 1985 onward they began a decline.
9 Unlike the PWRs, over most systems, I would say in 10 general, the scram trend was spread out over most of the 11 systems, most of the initiating systems that we looked at.
12 The feedwater system stayed fairly constant until 1987.
And 13 in 1988 it went down.
But that was primarily due to
()
14 unforeseen occurrence of five GE plants not scramming at 15 all.
16 In effect, if those plants had not scrammed, the 17 scram rate for 1988 would have been the same as 1987.
18 You can see here what I am talking about.
The 19 scram rate for BWR's main feedwater system is effectively 20 the same, annually from 1984.
Here, this is the result of 21 five plants not tripping in 1988.
I don't think there is 22 anything else.
There's nothing else, no.
23 Combustion Engineering plants were going right 24 along with small reductions in their scram rate annually 25 until 1988 and then all of a sudden they fixed their Heritage Reporting Corporation (202) 628-4888
79
)
i feedwater problems, apparently it seems that they have fixed 2
their feedwater problems.
They got a reduction of some 60 3
percent in just feedwater system-initiated scrams from those 4
plants.
5-MR. CARROLL:
Do you know what they did to fix the 6
problem?
7 MR. BELL:
I talked to some people in June from CE 8
about what happened.
And basically what they tell me is the 9
individual plants began to look at what owners group were 10 saying, as identifying as being problems for them, and they 11 went back and talked and addreesed problems associated with 12 feed reg valves and the control system for their feedwater 13 system.
(}
14 MR. CARROLL:
It tended to be maintenance and 15 training and that sort of thing?
16 MR. BELL:
Right.
It tended to be --
17 MR. CARROLL:
As cpposed to physically replacing 18 equipment with improved equipment?
19 MR. BELL:
Right.
It tended to be hardware 20 problems that they could fix.
21 MR. NOVAK:
Just, if you looked at the -- The 22 owners groups are just people that we try to talk to 23 generically and it has been my experience that the CE owners 24 groups feel that they are the least homogeneous group.
I 25 mean, the boilers are the same, Westinghouse plants seem to
()
Heritage Reporting Corporation I
(202) 628-4888
}
r 80 1-even believe they are more alike, and certainly the asW.
2 And if you looked at the spectrum of CE quote 3
" designs" from Maine Yankee, which is a Westinghouse design 4
to Fort Calhoun to Palo Verde, there is a wide variation.
5 And so collegially, they don't normally do things together 6-very often.
They are doing it more now.
And I think it was 7
a reaction to saying that what the Westinghouse owners 8
groups are saying and what other people are saying really 9
could apply to us and perhaps wa ought to go back and do it.
10 Dut it is rare to see a problem that would be 11 generically' applied from say Fort Calhoun to Palo Verde.
.I 12 mean, you just, I think that is probably part of the problem 13 there.
(( }
14 But they did get together, and I think that was 15 basically the part of the story.
16 MR. BELL:
Okay.
As I mentioned before, this one, 17 no turbine system scrams.
That is probably a little 18 unusual.
In fact, we have seen some already in 1988.
19 There are so B&W plants in the population that any 20 upset in any particular plant can cause a problem to be 21 reflected across all their scram rates.
l 22 Basically, it's feedwater system at B&W plants 23 that cause their problems, with some contribution to the j
24 turbine.
But usually, for example, what happened here, a 25 couple of plants had feedwater problems and one plant had
)
Heritage Reporting Corporation (202) 628-4888
81
. (D is" 1
turbine problems.
They addressed-them, and the scram rate 2
started to decline.
Effectively over the last couple of l
1 3
years they are about the same.
And if they want to get any
]
I 4
more improvements, they are going to have to address it f
l 5
through the feedwater system.
6 What appears to have happened, because remembering 7
the primary objective of this study, was to take a look.at-8 what was happening and what was causing and what was driving 9
the industry scram trend, and then bounce that off what the 10 industry was doing themselves.
11 The industry was aware that the scram reductions 12 were giving them all kinds of problems, as we talked about 13 previously, and they set about putting into place programs t( )
14 to address those problems.
They apparently succeeded 15 because since 1984 they have reduced their scram rate by 50 16 percent.
They have effectively cut it in half.
17 When we compared what we found in our analysis 18 with what the particular owners groups were doing, we found 19 that they were in general addressing the same systems that 20 we found to be the problem and their projections for the 21 future systems that they were going to look at were pretty 22 much lined up.
23 We find that additional attention still needs to 24 be paid to the feedwater system because it continues to be 25 dominant initiator of all unplanned scrams at the plants.
l Heritage Reporting Corporation (202) 628-4888
82
(^)S 1
And what we feel finally is that where scrams are currently 2
is probably where they are going to stay for a while.
They 3
are effectively stabilized now, and it will be very 4
difficult for them to reduce them any further without 5
serious expenditures of resources.
6 Unless there are questions, I don't have any more.
7 MR. REMICK:
Question.
8 MR. MICHELSON:
Since some people report scram 9
rates in terms of so many per year, what is kind of a good 10 magic number to convert thousand hours to a year?
11 Critical hours.
I just wonder what kind of 12 average.
13 MR. DENNIG:
It's sort of a rule of thumb that 14 we've played with the accounting system and the goal that 15 INPO uses is a little different basis than what we're using.
16 They don't use manual scams in there.
Their goal is roughly 17 down around the.25 per 1,000 to.3 per 1,000.
18 MR. MICHELSON:
Well, yes, but I'm just trying to 19 convert yours, when I read foreign data and so forth, --
20 MR. CARROLL:
-- 65 percent capacity.
21 MR. MICHELSON:
That's what I'm getting at, is 22 what kind of an average capacity factor do I use.
23 MR. DENNIG:
I think you got the right number.
24 MR. MICHELSON:
You just use 65 percent then?
25 MR. DENNIG:
I think that's what it is.
1
)
(~]
Heritage Reporting Corporation v
(202) 628-4888 1
83
(
1-MR. MICHELSON:
Okay.
v 2
MR. DENNIG:
The industry average is'65 or 67.
-3 MR. MICHELSON:
All right.
4 MR. DENNIG:
But that's a pretty good figure.
5 MR. REMICK:
Any other questions, comments?
6 (No response) 7 MR. REMICK:
Very good.
Thank you, Larry.
We 8
thank the staff for the presentation.
Anything else on 9
that, Mr. Subconmtittee Chairman?
10 MR. LEWIS:
Nothing at all.
11 MR. REMICK:
All right.
Okay.
I suggest, then, 12 we turn to the next item on the agenda, which is_to prepare 13 for our discussion with the Commissioners later today.
'14 (Whereupon, at 11:04 a.m. the portion of the
.i 15 meeting to be transcribed recessed, to reconvene at 4:40 16 p.m. the same day, Thursday, August 10, 1989.)
17 18 19 20 21 22 23 24 25 Heritage Reporting Corporation (202) 628-4888
84 7V 1
AETERHQQH SEEE1QH 2
-(4:40 p.m.)
3 MR. REMICK:
Could we reconvene?
The next topic 4
for discussion is the generic issue 79, unanalyzed reactor 5
vessel thermal stress during natural convection cooldown.
6 And Dr. Shewmon is our subcommittee chairman.
7-Paul, I turn the matter over to you.
8 MR. SHEWMON:
This is in Tab 6 of your meeting 9
notebook.
Back in 1980 there was an event at St. Lucie 10 which got people to thinking about the stresses that could 11 Edevelop around the head and the closer flanges up there if 12 the vessel cooled down as it does in various cases and the 13 head, due to lack of circulation in this area, did not cool
{}
14 as fast.
15 So we've got the cylinder shrinking and the top 16 not shrinking.
And this then became a generic safety issue 17 number 79.
18 The staff has got an analysis on this and now 19 feels that they have recommendations for its resolution.
We 20 have not had a subcommittee meeting on it.
I think we would 21 agree that the problem is of such a scale that we can 22 consider it resolved.
But I thought it would be interesting 23 to have a presentation on what the problem was and what they 24 have done.
25 So why don't you proceed.
Your name, sir?
l
()
Heritage Reporting Corporation (202) 628-4888
(
l
85
(')
1 MR. PAGE:
My name is Joel Page.
I am in the V
2 Office of Research.
I am the Task Manager on the issue.
3-I think Mr. Shewmon pretty well gave a good 4
introduction as to what the issue was about.
5 I have prepared a few slides.
I believe you have 6
copies of the slides.
I am going to skip a few of the 7'
initial slides, because I will merely be reading from them 8
anyway, to just speed up the initial part of the 9
introduction.
10 I have broken the issue down into four parts:
the 11 background, what I consider to be the important analysis 12 assumptions, our results of the analyses, and what we feel 13 are our recommendations, to close out the issue.
14 Now, this is a picture of a typical PWR reactor 0y 15 vessel.
I have circled the area that'is of concern in 16 generic issue 79.
17 Okay.
As Mr. Shewmon pointed out, in 1980, June 18 of 1980, St. Lucie 1 experienced a natural circulation 19 cooldown.
In that specific event, there was a loss of 20 component cooling water to the reactor coolant pump seals.
21 And after approximately eight to nine minutes, the pumps 22 were shut down and there became a concern about the 23 integrity of the seals.
In the event they began to notice 24 fluctuations of the pressurizer.
And they finally realized 25 they had a void in the reactor vessel head.
That is, they
(~J)
Heritage Reporting Corporation L
(202) 628-4888
86 l
/~i
'U 1
had a steam bubble.
2 After the event was over, the NRC of course was 3
looking at the thing very closely,.and they subsequently 4
issued generic letter 81-21 in May of 1981.
5 At the time there was really no concern for the 6
vessel plan itself.
It really was stressing operator 7
training, procedural concerns, how you handle the event, how 8
you recognize whether you have a vessel, a void in the head
)
9 of the vessel.
10 Other activities at the NRC included an AEOD case 11 study, C101.
That was issued.in September of 1982.
12 Now, the case study was St. Lucie specific.
And 13 they had numerous recommendations.
()
14 In response to this generic letter, B&W of course 15 was looking over how they were going to respond to it, and 16 they began to do some back of the envelope calculations.
17 And as a result of that, they began to see some stress 18 levels that they felt made them a little uncomfortable.
And 19 so they issued a notification to is in March of 1983.
They 20 said they felt it was a concern but it was certainly ont a 21 near term concern.
22 In J21y of 1983, this was identified specifically 23 as a potential generic issue and that same month it was 24 prioritized as medium.
25 Subsequently, in March of 1984, there was a for mal Heritage Reporting Corporation (202) 628-4888
i l-l 87
/3
\\/
1 meeting with the B&W owners group.
2 And in May of 1984, the ACRS reviewed and approved 3
the prioritization of the issue.
4 In October of 1984, B&W issued a-formal report, 5
which is actually quite thick, a lot of calculations.
The L
6 NRC took the report, looked it over, contracted with the 1
7 Brookhaven National Laboratory in June of 1985.
8 Between June of.1985 and April of 1988, there were 9
numerous interactions between Brookhaven, the staff and the 10 B&W owners group.
And finally, in June of 1989, we were 11 able to get a final Leport from Brookhaven regarding their 12 review of the issue.
13 MR. CARROLL:
I'm missing something.
Isn't St.
()
14 Lucie a combustion plant?
15 MR. PAGE:
Yes, sir, it is.
16 MR. CARROLL:
And h>w does this, how did the B&W 17 owners group get into this thing?
18 MR. PAGE:
Everyone was responding to the generic 19 letter, all PWR licensees.
It was addressed to all PWR 20 licensees, not just combustion.
The concern is that the 21 separation -- If you want, in a few minutes, I'll put up a 22 more complicated slide.
There seems to be a separation 23 between the dome and the lower part of the vessel, and the 24 stagnation in that region would be a concern for any PWR.
25 MR. BAER:
Let me try and give you a little more Heritage Reporting Corporation (202) 628-4888 l
i 88
.n kl 1
direct answer.
1 2
As far as we can understand it, B&W, when they 3
were responding to the generic letter, which as Joel said, 4
was primarily system and procedure oriented, they apparently 5
did some initial calculations of their vessel stresses.
6 And they, and only they among all the PWR NSSS 7
suppliers, wrote us a notification that_their initial 8
calculations indicated that they may have some high stresses 9
under these conditions.
10 MR. CARROLL:
But as it turns out it is at least 11 potentially a problem for all PWRs.
12 MR. BAER:
Yes.
It would be potentially the same
'13 problem, but we have only analyzed the B&W vessel, because
()
14 they are the ones that supplied the information.
15 MR. CARROLL:
When you say B&W vessel, you mean 16 the one for a B&W NSSS as opposed to B&W vessels in general, 17 which have been used in other --
18 MR. BAER:
Right.
And specifically for the 177 19 plant, fuel assembly plants, which all the ones that are 20 operating have identical vessels.
i l
21 MR. CARROLL:
Okay.
22 MR. PAGE:
Okay.
This slide here covers what I 23 considered, because there were numerous parameters of 24 course, in the evaluation, I believe that these were very 25 important assumptions that were made during the ant. lysis.
Heritage Reporting Corporation (202) 628-4888
89 (m_)-
'l-The B&W owners group assumed, very conservatively, 2
that the reactor vessel head had a constant supply of 600-3 degree heat.
In our confirmatory analysis, we made the same 4
assumption.
5 B&W used a 50-degree per hour cooldown rate.
We 6
decided to use a 100-degree per hour cooldown rate since 7
that was the tech spec limit.
8 As far as effective stud lengtha, which would be, 9
on the previous pictures of course, the reactor vessel 10 studs, B&W used 40.25 inches of-an effective stud length.
11 We used 33.5.
The difference in those two basically is they 12 took credit for half of the engaged threads as still being 13 part of the working length of the stud.
We didn't.
We used (O")
14 only the free length of the studs.
15 As far as flange interface conditions, we used two 16 bounding conditions just like the B&W owners group used.
We 17 used one with infinite friction, basically.
That was where 18 the flanges were basically locked together.
And we used one 19 where they were absolutely frictionless.
And we felt they 20 were bounding the first one, where you have infinite 21 friction would be the worst case for the shell itself.
The 22 second one would be the worst case for the studs.
23 As far as void conditions, and I will put a slide 24 up in just a second to show these void conditions, there 25 were three that they used in combination with the frictional
()
Heritage Reporting Corporation (202) 628-4888
i L
90 L
e)
(_
1 coefficients.
And one was the void to the top of the 2
The second one was a void to right where the flange 3
interfaces.
And the third one was a. void below the vessel 4
5 When Brookhaven took their report and reviewed it 6
in high detail,a or in great detail, excuse me, they felt 7
that it had been done very well.
And the only part that 8
seemed of concern was the middle one, the void to the flange 9
interface.
There seemed to be a concern there.
10 The shell stresses for all the cases seemed low 11 enough to really not warrant an independent analysis.
12 However, they felt that that middle one brought up some 13 interesting questions, particularly for the studs.
(}
14 As far as fatigue concerns, B&W considered 20 15 cycles, we considered 40.
16 MR. SIESS:
Would you tell me what the effective 17 stud length means, or how it comes into the problems?
- 18 Effective for what?
19 MR. PAGE:
Well, it's the working length of a 20 stud.
If you can picture a stud, it's threaded into the 21 flange.
So you have so many threads that are engaged here 22 next down into what I call a working region.
And then you 23 have a washer, and then the nut is on top.
24 Now, the effective length means when you prestress 25 the stud, or bend it, or anything like that, you are saying
()
Heritage Reporting Corporation (202) 628-4888
91 f)/
\\-
1 how much of it is actually doing the work?
Now, in a 2
reactor vessel stud, you have very close tolerances.
So 3
when you start to. stretch it, we'll talk prestress right 4
now.
In effect, all the stretching is really taking place 5
away from the threads or in the very first few threads is 6
what is happening.
Your stress concentrations are going to 7
be much higher there, as opposed to all the way up the 8
thread.
9 MR. SIESS:
Is there an unthreaded region which is 10 33-1/2 inches long?-
11 MR. PAGE:
This is the unengaged -- It is mostly 12 unthreaded.
13 MR. SIESS:
Okay.
It is mostly unthreaded and all
()
14 unengaged?
15 MR. PAGE:
All unengaged.
that is correct.
16 MR. CARROLL:
The part going through the stud hole 17 in the upper head, basically.
18 MR. PAGE:
It is totally unsupported.
19 MR. SIESS:
So there is a counter bore in the i
20 pressure vessel and that part isn't threaded, it is only 21 threaded at the bottom of that counter bored hole?
22 MR. PAGE:
There is a slight counter bore.
But 23 basically, yes, in the lower flange, which is in the vessel 24 proper, you do thread in there.
then you have your flange 25 cover which has a hole in it bigger than the stud.
Heritage Reporting Corporation (202) 628-4888
_ _ _ _ _ _ _ _ - _ _ = _ _ _
92
'l '
1 So that part is not really engaged.
You are only 2
compressing the flange.
3 MR. JOHNSON:
May I say something?
I'm Richard 4.
Johnson, a colleague of Joel Page.
1 5
It's very simple.
The longer, the more flexible.
{
6 And it.gets, the bending stresses then cre a function of 7
beam length, as it were, although it is a vertical Leam.
8 The stubbler.it is, for a given deflection, the higher.the 9
stress will be.
So it becomes very important in the stress 10 analysis that way.
11 MR. SIESS:
Thank you.
12 MR. PAGE:
Let me expand a little bit more on what 13 Dick said.
It also becomes very important in pre-stress.
14 It may not be obvious.
But when they pre-stress the stud on g-V 15 a reactor vessel they actually set deflection.
So it is 16 delta length over original length.
And the original length 17 in our case would be the 30 inches.
18 MR. LEWIS:
Could you just-indulge me for one 19 second?
You said something about the stress being 20 concentrated in the engaged part of this long stud.
And I'm 21 missing the point.
22 MR. PAGE:
No, no, no.
I'm saying that you would, 23 you don't share the load through all the threads.
24 MR. LEWIS:
Yes, I understand that.
25 MR. PAGE:
Okay.
What happens is that it very
()
Heritage Reporting Corporation (202) 628-4888
r e,
93
()
11l quickly'goes away.
As it stretches, you find'that almost 2
all the? loading.that is in the threaded portion is in:the 3
very first1few threads.
4 MR. LEWIS-No, I understand that.
'5 MR. PAGE' So we decided to be conservative and 6-say.we're not giving any credit --
P j7; MR.. LEWIS:
I just'want to be'sure.I didn't 8
. misunderstanding something earlier..So the rest ofL the t-9 tension, though, is uniformly distributed through the 10
.. length?
1
'll MR. PAGE:
The unsupported,.yes.
12
.MR. LEWIS:
The unsupported length.
Okay.
I 13 misunderstood, then.
14 MR.'PAGE:
But it becomes important not only for-j
[
15' bending,.which of course it is very important, when you 16 bend, if you have~a shorter beam, but also for pre-stress 17
-because of the fact-that you are setting a length:as opposed 18 to a stress.
19 Olkay, now, this right here is a' diagram of the.
20 vessel flange.
And it shows the three transient conditions 21 that were considered by the B&W owners group.
The void.
22
-level of course for condition 1 would be at the top of the 23 flange.
Transient condition 2, the void level would be at 24 the flange interface.
And transient condition 3 would be 25 below the. interface.
O ueriteee negerties cergeretioe (202) 628-4888 2______-___.--__-__
1 94 4
4
/^
is)' -
1-MR. CARROLL:- Does it matter how much below?
j 2
MR. PAGE:
I don't recall exactly how far'below, i
3 There is another, there is a support plate right here for 4
the lower internals.
This would be actually steam, to right l
5 here.
6 MR. CARROLL:
That's the ledge, okay.
7 MR. PAGE:
There is a ledge here for an-upper core 8
support plate as I recall.
9 MR. CARROLL:
Gotcha.
10 MR. PAGE:
This slide here shows the transient 11 condition as we saw it in terms of head temperature, hot leg
.12 temperature, cold leg temperature, and pressure, over a 13 period of 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br />.
14 This of course shows 100 degrees per hour.
15 This is the whole time here.
It's a shame we 16 didn't have overlaid here the metal temperature.
But there 17 are, if you have the report which I included in the package 18 to Dr. Shewmon, it does show the stress time, or the 19 pressure, temperature contours in the vessel flange.
20 Basically this chows you the fluid temperatures.
21 MR. DAER:
To make this clear, with the 22 assumptions used in the analysis, the head was held at 600 23 degrees constantly.
24 MR. CARROLL:
What happened at three hours?
25 MR. PAGE:
At three hours?
(~'
Heritage Reporting Corporation
's (202) 628-4888 w____
1
-i 95 o
7 6
(/
1 MR. CARROLL:
Why is pressure constant?
2 MR. PAGE:
Well, this is a hold point.
3 MR. SHEWMON:
They stopped cooling down.
4.
MR. PAGE:
Yes.
5 MR. SHEWMON:
for some' reason.
6 MR. PAGE:
Yes.
7 MR. CARROLL:
What is that, they went on RHR or 8
what?
9 MR. PAGE:
No, no.
You go on RHR right here.
10 This is where you begin to cool off again.
This is a hold il point at which you keep the hot leg and the cold. leg 12 basically. steady.
And then at this time of course there is 13 a lagging of the temperature, meaning that the temperature
()
14 of the metal is of course lagging the temperature of the 15 fluid.
So this lets it start to stabilize.
16 MR. SHEWMON:
Why did you assume this history 17 instead of any other?
You've cooled her down for three 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br />, then you go into a hold.
Why?
19 MR. PAGE:
Well, it's a very complicated thermal 20 hydraulics aspect.
And Brookhaven did it.
All I can say is 21 that this is the way they did it and they do go into this 22 long hold point.
And I'm saying that effectively, they said 23 it really wasn't that important, whetner this hold point was 24 two or three hours or ten or 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br />.
They said that once 25 the metal started cooling down, it would effectively conduct l
)
Heritage Reporting Corporation (202) 628-4888
n...
96 f~)-
\\/
1 heat from the head down into the vessel proper.
I really 1
2 don't understand all the thermal hydraulics aspects of it.
3 But it appears to be a hold point and there is no question 4
about it.
5 MR. BAER:
Well, I think once they got down to a 6
250-degree cold leg temperature, you couldn't realistically 7
assume you were going down at anything like 100 degrees an 8
hour.
9 MR. CARROLL:
Well, in principle, at that point, 10 you could have put RHR on and cooled it down some more.
11 MR. BAER:
More, yes.
They could have put that 12 end curve starting there.
13 MR. PAGE:
In general, though, I think it is
()
14 considered a better idea to hold for a while and let the 15 metal temperatures catch up and equalize.
16 MR. CARROLL:
I'm sure that's right.
17 MR. PAGE:
I'm not so sure it really is that easy 18 to cool down that fast.
19 This slide shows a few of oug analysis results.
20 The top block shows a comparison of the maximum triaxial 21 stresses for the shell.
22 Now, triaxial stresses are of course a combination p
23 of the maximum hoop stress, the maximum axial stress and the 24 maximum radial stress.
25 The B&W owners group calculated a maximum of Heritage Reporting Corporation (202) 628-4888
I 97
-~k )'
f 1
66,000 psi.
Brookhaven calculated 80,000 psi.
The code l
~-
2 allowable is 128,000 psi.
So we felt we had a pretty good 3
margin on that.
4 MR. KERR:
Can't stress analysts come closer 5
together than-that?
6 MR. PAGE:
pardon me?
7 MR. KERR:
I say, can't stress analysts come 8
closer than-that, or is that just because they used 9
different bolt lengths?
10 MR. BAER:
Probably more of 100 degrees an hour 11 versus 50 degrees an hour.
12 MR. KERR:
Okay.
13 MR. BAER:
You see a bigger difference in the
()
14 studs.
15 MR. CARROLL:
Even if they can't get closer 16 together, Bill, they can sure get significant figures.
17 MR. PAGE:
I think what stood out to me the most i
18 was the fact that in the worst of the worst of the worst, 19 you still show some margin.
And in most cases, you showed 20 substantial margin.
And I think if you look for that in 21 most of the slides, I think that is what jumped out at me.
22 MR. CARROLL:
Did you do anything, though, to see 23 whether Brookhaven and BWOG really would come together if 24 they used the same assumptions?
Did you ask them to do the 25 same problem?
Heritage Reporting Corporation (202) 628-4888
y1
.a
\\
b 98~
.:q A_/)
11 MR. PAGE:
Well, they were< fairly-close..
Some of 2.
the modeling details'of course, the thermal hydraulics I 5.
f
'3-thoughtc'was probably.the;part that was the'most complicated; 4-and if you would like, I will -- Let!me put up'another slide.
q 5:
here that I really hadn't planned to use.
6 MR. BAER:
I think the answer was that we didn't 7
ask BNL to go and' find duplicate B&W --
8 MR. PAGE:
We really.weren't'asking for a.
TJ
.duplicative effort.
We were asking for them to do.something-p 101 on their own with the copy of course they had of the B&W
- 11 owners group calculations.
However, we didn't want them to
' 12 just copy it.
.And there are -just.an incredible number of
~
13 heat transfer coefficients, boundaryclayer coefficients.
I-14 mean, it is just so incredibly. complicated.
To.tell you the
- 15 truth, I'm surprised they'were that close.
16 MR. KERR:' And it only.took them three years to do 17 it.
18 MR. PAGE:
Well, now,.one-of the reasons they took
- 19 three years to do it is because it was a medium priority
- 20 issue.
And we really had a lot of changeover personnel.
I 21.
mean, we really went through, I went through three different
-22; contacts at the B&W owners group, I went through four 23 different contacts at NRR just in a one-year period.
It's a 24 very complex problem of just getting questions to 3&W, 25 getting answers back.
I can't blame it all on Brookhaven.
()
Heritage Reporting Corporation (202) 628-4888
l 99 m
'kl 1
And we lost our primary stress analyst.
2 MR. FERR:
I wasn't blaming it on them.
I thought 3
after your telling me how complicated it was, it was 4
remarkable that they got it in three years.
5 MR. PAGE:
I'm sorry.
Actually, they would have 6
done it much quicker.
I think that the problems were not-7 all at Brookhaven.
8 MR. KERR:
Okay.
9 MR. LEWIS:
Are we willing-to continue my.
10 education just for a second, just for my information?
11 The maximum triaxial stress means that they 12 resolved the stress into its three principal directions, and 13 that is the maximum of the three?
Is that what that means?
(( )
14 MR. PAGE:
that's right.
15 MR. BAER:
Isn't it a vector sum?
16 MR. LEWIS:
What does that mean?
17 MR. PAGE:
I thought you just described it.
It is 18 the maximum stress.
You take the three principal stress 19 directions.
20 MR. LEWIS:
That's what I said.
And that's what 21 it is.
22 MR. PAGE:
That's right.
23 MR. LEWIS:
And is it that that the code provides 24 as the limiting condition?
That is, the code provides that 25 you resolve the stress into its three principal directions
(
Heritage Reporting Corporation (202) 628-4888
100
. /'l
\\_/ -
1 and pick the largest and that_is what is limited?
2 MR. PAGE:
It also breaks it down into primary and 3
secondary stresses, and other aspects of it.
There is a 4
slide that covers briefly the stresses we discuss here.
I 5
hadn't planned to put it up, just to save time.
But there 6
is one in there that discusses it.
7 MR. REMICK:
Anytime there is an opportunity to 1
8 have you educated, Dr. Lewis, we are in favor of it.
9 MR. PAGE:
What is the source of the stress.
10 MR. LEWIS:
But it says it's the sum of the three.
11 MR. PAGE:
Yes, it's the sum of the three.
12 MR. LEWIS:
But radial-axial hoop, it says.
13 At every point there is a stress tensor.
It does
()
14 have three directions.
And this is the sum of the three.
15 MR. PAGE:
The sum of the three.
16 MR. LEWIS:
Not the largest, okay.
17 MR. PAGE:
For all the loading conditions, in the-18 case of the triaxial.
I was going to say you have to worry 19 about primaries and secondaries because you have different 20 things to judge them against.
21 MR. SHEWMON:
Can we go to your,-three slides over?
22 MR. PAGE:
Okay.
Did you say to go back?
23 MR. SHEWMON:
No, go on to the slide we are 24 talking about.
25 MR. REMICK:
The one that says stress definitions.
O Heritage Reporting Corporation (202) 628-4888 L----_-_-_
101
(}
1 It's.two ahead of where you are, I believe.
2 MR. LEWIS:
Just tell me, Paul, if I look at that, 3
can I stop asking questions, it will all be in here?
4 MR, SHEWMON:
That is called a stress intensity, 5
not what the rest of the world calls a stress intensity?
6 MR. PAGE:
I would call it a stress intensity.
7 MR. SHEWMON:
It's certainly not K-1.
8 MR. PAGE:
No.
No, they're not the same.
I,think 9
K-1 is a stress intensity factor.
Yes.
10 MR. SHEWMON:
Now, that has no particular 11 relevance to its yield criteria.
12 MR. LEWIS:
That's what is troubling me.
13 MR. PAGE:
No, stress intensities are primarily 14 aimed at fatigue.
And there is a stress intensity factor O-15 which is related to it, somewhat related to it, which is 16 used for fracture toughness.
17 MR. SHEWMON:
That's what I meant when I said 18 that's not what the rest of the world calls a stress 19 intensity.
Okay.
You say it comes into fatigue.
I don't 20 see why the code has a particular interest in that it is not 21 a property that you have included.
l 22 MR. PAGE:
I didn't hear you.
23 MR. SHEWMON:
You can't look up in any table the l
1 24 triaxial stress resistance of any metal.
25 MR. LEWIS:
Thank you.
Thank you.
Because that O
Heritage Reporting Corporation (202) 628-4888
)
I L
l
102
-(>
'l is what was troubling me.
2 MR. SHEWMON:
Yes.
3 MR. PAGE:
You're talking about the maximum stress 4
at a point for a set of loading conditions.
And it would 5
just be a point.
It wouldn't be a membrane type stress 6
which is across the entire section.
7 MR. SHEMMON:
It's not even a stress, it's the sum 8
of three numbers each of which is an orthagonal stress.
9 MR. PAGE:
That's right.
But it would give you 10 stress at a point.
11 MR. SHEWMON:
No.
Stress is measured across a 12 plane.
13 MR. PAGE:
You can have peak stresses at certain
[
14 points.
s 15 MR. SHEWMON:
You can sum them if the code finds 16 it convenient, but physically I don't see what it means.
17 MR. LEWIS:
That's what is troubling me.
The 18 yield criteria usually depend on the differences between the 19 major stresses.
And there are three that apply.
And I 20 don't understand the term radial axial, because what you are 21 supposed to do is take the stress ellipsoid at any given 22 point and find its axes, wherever they are.
And the term 23
" hoop" applies to something in the large, not the stress 24 ellipsoid.
25 MR. SHEWMON:
No, if those are three cylindrical, Heritage Reporting Corporation (202) 628-4888 L__ __n__ _ _
103-(A) 1 and three circumferential --
2 MR. LEWIS:
I understand that.
But there is 3
nothing that guarantees that the stress tensor will resolve 4
orthagonally on those axes.
5 MR. SHEWMON:
They may'not be the principal axes.
6 MR. LEWIS:
That's what I'm saying.
7 MR. SHEWMON:
Okay.
8 MR.~PAGE:
Well, these are calculated stresses, of 9
course.
So they do come out that way.
In other words, 10 these are mathematically arrived at.
11 I don't know how much more -- Do we want to dwell 12 on this particular thing or go back to the --
l 13 MR. REMICK:
Go ahead.
.( }
14 MR. PAGE:
Okay.
Let me back up then.
15 MR. REMICK:
You can go back to the other membrane 16 stresses.
17 MR. PAGE:
Okay.
We found that the maximum 18 average membrane stresses in the studs, these lower two l
19 boxes are on the studs.
B&W had 38, almost 39 ksi.
20 Brookhaven was 49 ksi.
And the code allowable to 71, or
'21 almost 72 KSI.
The maximum membrane plus bending stress in 22 the studs was e. bout 69,000 for B&W, about 104,000 for 23 Brookhaven.
This again points out the difference in the 24 effective length of the studs.
And the code allowable is25-107.7 psi.
)
Heritage Reporting Corporation (202) 528-4888
)
)
__o
I 104
'h l._)
1 MR. CARROLL:
In each of these cases, we are 2
dealing with whatever the worst case in terms of friction 3
is.
4 MR. PAGE:
In this case of course.for the studs, 5
these lower two boxes would be frictionless.
So'this would 6
allow the flanges to just slide on each other.
And of 7
course in our case you've got a shorter stud, so you are 8
bending it, you see a lot more bending.
9 MR. CARROLL:
Okay.
10 MR. LEWIS:
At the risk of prolonging this, can 1 11 ask Paul one question?
12 The sum of the three stresses is the thing which 13 is in variant under rotation, and therefore you can
(}
14 calculate it along'the radial and so forth, and it will be 15 the same as if you had resolved it, so it saves you'from 16 having to resolve it even if it has nothing to do with the 17 yield properties of the material.
18 Is that a fair statement?
19 MR. SHEWMON:
I think so.
20 MR. LEWIS:
Thank you.
21 ME. KERR:
It's fair.
It may not be accurate, but 22 it's fair.
23 (Laughter) 24 MR. LEWIS:
Actually, I don't care about being 25 fair.
But I hope it's accurate.
Heritage Reporting Corporation (202) 628-4808
L V
i 105 1
(
I'T
. (_,/
1 MR. PAGE:
Okay.
In these.two boxes I have l
2 charted the fatigue evaluations that were done.
One for the 3
shell, the upper box, and the lower box is on the studs.
4 B&W of course did 20 cycles.
I went ahead of 5
course and just doubled the number, for 40 cycles, and 6
compared it to what we got.
7-This value, cumulative usage factor, has to do 8
with the number of stress cycles between, well, it's 9
membrane plus bending range.
So you are ranging between the 10 prestress value and the membrane plus bending due to this 11 transient.
12 MR. SHEWMON:
What is the second column?
13 MR. PAGE:
Second column?
-( )
14 MR. SHEWMON:
Second column is per one cycle?
15 MR. PAGE:
I'm sorry.
I'm not sure which cycle i
16 you mean.
This one?
17 MR. SHEWMON:
Upper cycles is the first column.
18 That's the second column.
Now is that for once?
19 MR. PAGE:
No, this is for 20 cycles.
20 MR. CARROLL:
What does CUF mean?
21 MR. PAGE:
Cumulative usage factor.
22 MR. SHEWMON:
What is the third one there?
If the 23 second one said 20 cycles, what is the third one for?
24 MR. PAGE:
This is for everything, all calculated 25 transients.
See, B&W had a number for non-natural
(
Heritage Reporting Corporation (202) 628-4888
i a -
106 1
c'irculation cooldown transients.
So we took their value and 2
we added our value to theirs.
And this was the difference 3
in-the two'.
Excuse me.
This was the difference in the two.
4 As you can see, the cumulative usage on the vessel 5
is still considered quite low, so we considered fatigue to 6
be basically a "no never mind" as far as this transient.
I 7
can't imagine anyone experiencing 20 or-40 actual cycles.
8 We of course like to keep the margins in there.
9 I would imagine they would probably be shut down 10 for a long time.
But in the bottom one here, it is the same 11 subject on the studo.
Of course, 20 cycles was what they 12-calculated. I merely doubled the number.
And that is why 13 there are slight differences in the total.
But basically as
'(])
14 you can see we almost came to grips.
There's a lot of 15 calculational noise when you get down to such small numbers.
16 We basically felt that from a fatigue standpoint 17 this really was not a problem.
18 Okay.
Now, the last part of our analysis dealt 19 with fracture toughness.
In looking at the curves that we 20 had in Section 3 of ASME, we realized that for the given 21 materials, that really the only concern could be below a 22 level of 250 degrees F, because above a value of 250 metal 23 temperature, there really was no fracture toughness concern.
24 Dasically, you are off the curve.
25 There was a heat conduction calcu.lation done as a
(
Heritage Reporting Corporation (202) 628-4888
107
<~s.
I) 1 backup by one of the staff.
It was very conservative.
It 2
did not assume any heat transfer from the steam.
And so it 3
also, let's see, I think as I recall, at about the 11-hour 4
point in the analysis, which was about the worst for the 5
stresses, the dome temperature would have in fact reduced to 6
350 degrees when the lower vessel temperature was 250 1
7 degrees.
8 So it tended to indicate that just through sheer 1
l 9
conduction, you really would never be able to keep that dome 10 at 600 degrees over that kind of a transient.
And so it 11 indicated that we had an adequate margin in the vessel 12 itself that you really shouldn't worry about.the vessel 13 tearing in such an event.
14 The last part of our analysis dealt with the same 15 subject, only for the studs.
And in this case, another 16 member of our staff provided two different types of 17 calculations.
Let me see here.
I made a couple of notes on 18 my slide here.
I'll try not to state this incorrectly.
19 The factor K(sub)1 there is a stress intensity f
20 factor, not the same as the triaxial stress that we were 21 talking about.
He made these two separate calculations.
22 And basically they were quite comparable to one another, the 23 final product, the final answer was very comparable.
And 24 then when compared to the acceptance curves, it was 25 determined that K(sub)1 was always greater than K(sub)1C,
()
Heritage Reporting Corporation (202) 628-4888
h i:
i 108
' /^)
's/.
1 certainly at 150 degrees'F and probably much lower than 2
that.
3 So we felt that even in the studs there's more i'
4 than adequate margin certainly for any one time event.
We 5
figured that if there is any real event, people are going to 6
be taking their vessel apart, and if there is any problem 7
with those studs you are probably going to find it the 8
minute you start to take them off.
But it won't be that 9
they tore.
10 My last slide has to do with the recommendations 11 we have after going through this long evaluation.
12 Since we did evaluate the 177 B&W vessel only and 13 we did it for 100 degrees per hour, we decided that if any
(' )
14 B&W licensee with this particular vecsel exceeded that 15 criteria of 100 degrees per hour, that he should confirm 16 that he has not violated any regulatory design or fracture 17 toughness criteria.
10 However, as I pointed out earlier, this is a 19 terribly complicated thermal hydraulics problem and so we i
20 felt -- oh, my slide is wrong, sorry, there on that lower l
21 bullet.
That should say "CE,"
not.
"GE."
For any 22 Westinghouse vessel or Combustion Engineering vessel or any 23 non-177 fuel assembly B&W vessel.
And what we had in mind 24 there when we said non-177 was Bellafonte, which may indeed 25 come online.
_ )
Heritage Reporting Corporation (202) 628-4000
l' 109 l-
~ 1 We feel that if they experience any natural-l
('] -
2 circulation cooldown that is reportable in accordance with 3
10 CFR 50, then we feel they should supply the same' 4
confirmation.
5 MR. CARROLL:
Doesn't that almost in effect say 6
that if you are a smart utility operator you better have the 7
analysis in.your back pocket?
8 MR. PAGE:
I wouldn't leave it alone.' I think I 9
would start to evaluate it early.
Some people in fact have 10 instrumented their vessels, to try to chart some of these.
11 that is one of the real problems is actually getting. metal 12 temperatures, head temperatures.
13 MR. CARROLL:
I know all about it.
What happens?
14 How, if I do have a natural circulation cooldown on my f-15 plant,.you are not going to let me start up until you get 16 the analysia?
17 MR. PAGE:
No, no.
A reportable natural 18 circulation cooldown.
10 CFR 50 very clearly delineates 19 what is reportable, what kind of an event.
20 11 it is mild enough to where it does not fall 21 into the reportable category.
22 MR. BAER:
I think the C&W 177 vescel is okay for 23 100 degreen an hour for several hours.
24 MR. PAGE:
It's a little more complicated than 25 that.
But that is one of the things.
That is one of the
/~'
Heritage Reporting Corporation
\\-
(202) 628-4888
h 110
<~s k_)'
1 criteria.
2 MR. BAER:
So if someone had an event where they 3
cooled down more than 100 degrees under natural circulation, 4
that's when we'd go ahead and 1,ook at it.
-But that's our 5
recommendation.
6 MR. CARROLLs I haven't had a chance to read _your 7
letter.
But one of the historic controversies is what does 8
100 degrees an hour mean?
9 MR. BAER:
More than 100 degrees total.
10 MR. PAGE:
That's averaged.
11 MR. CARROLL:
Averaged?
12 MR. PAGE:
Averaged in incremental time.
13 MR. CARROLL:
Over the transient?
(~')
14 MR. PAGE:
I'm not sure, I don't know if it is in 15 exact hour increments.
But there is a technique by which 16 they average it over time.
It is a step change as opposed 17 to every minute.
18 MR. BAER:
I've been told several times by 19 analysts on the stand that 100 degrecs an hour is based on a 20 step change of 100 degrees 21 MR. CARROLL:
Are you following me, Bob?
I've had 22 inspectors tell me I've got a tech spec limit of 100 degrees 23 an hour on a normal cooldown for example, and he says 24 gotcha, because for one second you were going 105 degrees on 25 an instantaneous basis.
On the average, you were well below
(
Heritage Reporting Corporation (202) 628-4888 l
l:
111
-Q 1/
1 the 100 degrees.
2' MR. BAER:
All-I can say, in the past, people like 3
Jim Knight, when he was at the agency, and I think Bob 4
Bosnick, told me that that 100 degree per hour rate was 5
really based on a calculation where someone did a step i
6 decrease of 100 degrees on the inner surface and looked at 7
it for an hour.
And the 100-degree step is obviously worse 8
than the ramp.
9 MR. CARROLL:
So you are telling me instantaneous 10 100 degrees and then once an hour.
11 MR. BAER:
Once an hour.
12 MR. CARROLL:
Okay.
And you used that temperature 13 on the inner surface?
()
14 MR. BAER:
No, that's fluid temperature.
15 MR. PAGE:
That's fluid temperature.
16 MR. BAER:
Well, I'm not sure of the analysis.
I 17 suspect the analysis may have used it on this deal.
May 18 have used it.
I'm not sure about that.
19 MR. CARROLL:
I haven't really had a chance to l
20 read the letter, and if the letter in effect tells me for i
21 Westinghouse and CE plants, if I exceed 100 degrees an hour, 22 I've got to do this kind of analysis, I'm trying to fina out 23 If they've made it clear what greater than 100 degrees an 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> means.
25 MR. PAGE:
I believe that that is already defined Heritage Reporting Corporation (202) 628-4888
I' 112 1
for the plant.
And I do believe it is what Bob said.
It is 2
a step change, it is not instantaneous change.
3 MR. CARROLL - Okay.
4 MR. PAGE:
I'm basically finished.
5 But just as a closing slide, I wanted to throw l
l 6
this up.
I kind of had this as a backup slide.
I guess it-7 could give everyone a little bit of appreciation.
I don't 8
know how much you know'about vessels.
But this gives you 9
some idaa of just how complicated the analysis can be, and 10 why people should be reluctant to hurriedly coo] down.
11 Basically, your water comes in, down the outside 12 of the core barrel, up through the core, and then out the 13 hot leg.
()
14 The problem is this upper support plate seems to 15 isolate the head region.
16 Now, although it is generally felt by most of the 17 licensees, certainly B&W felt that they had 7 to 9 percent 18 flow.
Under any flow condition they felt 7 to 9 percent of 19 that flow would Indeed be coming up through these control 20 rod guide tubes.
They just can't prove it.
They don't have 21 actual measured data to show that.
22 We tended to agree with them that it wouldn't be 23 quite as stagnant as the analysis, of course, shows, which 24 was a very conservative bounding analysis.
25 MR. MICHELSON:
llow would there be flow of that O
V Heritage Reporting Corpora t. ion (202) 628-4888
113
[]
1 magnitude to the upper head?
\\_/
2 MR. PAGE:
How would there?
't MR. MICHELSON:
What's driving it?
4 MR. PAGE:
The flow coming up through the core.
5 MR. MICHELSON:
Tho natural circulation flow.
6 What's driving it up.to the upper head?
7 MR. PAGE:
Just the flow itself.
Whatever is 8
coming into this chamber here that's going out here, some of 9
it is going through these. control rod guide tubes up into v
'j 10 the head.
11 MR. MICHELSON:
Well, the head is dead end of l
12 course.
What makes it circulate in back?
If it goes in, it 13 is going to come out somewhere.
And the only somewhere is 14 there is a gap around the edge there.
(~)
15 MR. PAGE:
I see what you're saying.
It's 1
16 basically gurgling up into the head.
Now, whether you get a 17 bubble or not depends on how fast you're cooling down.
18 MR. MICHELSON:
If 10 percent is going in, 10 19 percent is also coming out, if you are talking flow.
20 MR. BAER:
I think one of the reasons B&W in their l
21 analysis of the owners group assumed 600 degrees constant, 22 but one that's hard to prove, and Joe, when you looked at 23 some data taught by CE, they showed a fair amount of 24 stagnation, didn't their head stay very hot for a long time?
25 MR. PAGE:
In the initial part of the transient, 1
l
[]
Heritage Reporting Corporation x
(202) 622-4888
.m 114
/"h.
3%/
1 yes.
2 MR. MICHELSON:
There is some data for some of 3
those natural circulation shutdowns, inadvertent ones of' 4
course, which indicate indeed in some. cases they stayed very S
not.
6 The fact is, it became the pressurizer.
It formed 7
the bubble and became the pressurizer of the system.
And it-8 stayed there for a long period of time.
That's why I'm 9
puzzled about where this flow is coming and why it is 10 flowing and what's driving it.
11 MR. PAGE:
Well, as I recall, it was somewhat of a 12 gurgling type thing.
They did feel though that there was 13 some flow getting to the. upper head.
j
)
14 MR. SHEWMON:
Water won't flow uphill.
15 MR. REMICK:
Did they take credit for it in their 16 calculations?
17 MR. PAGE:
No.
18 MR. REMICK:
No, that's what I thought.
19 MR. CARROLL:
Now, there's the cold leg coming in.
[
20 I want to point that out.
The arrow.
And there is leakage, l
21 Carl, up at the upper support --
22 MR. MICHELSON:
Yes, there has to be, that's what 23 cools it --
I 24 MR. CARROLL:
And that's where the water flows 25 into the upper head area, and it flows back down the guide i
Heritage Reporting Corporation (202) 628-4888
115 f~X
- .,.)
1.
tubes.
\\
L 2
MR. MICHELSON:
That may be.
i.
3 MR. CARROLL:
And out into the space above the 4
core.
5 MR. MICHELSON:
I thought it said it came up the 6
guide tubes.
Maybe I misunderstood.
7 MR. DAER:
-- very low.
8 MR. CARROLL:
I understand.
9 MR. BAER:
-- just about shuts off.
10 MR. CARROLL:
Well, and that is what they are 11' saying the assumption is.
i 12 MR. SHEWMON:
You at least see why they assumed, 13 just stayed at 600.
t( )
14 MR. PAGE:
It was the easier thing to do, for 15 sure.
16 MR. SHEWMON:
Let me go to a different question, 17 if we can go back to the last slide in your handout for a 18
' minute.
19 MR. PAGE:
Okay.
20 MR. SHEWMON:
They do the calculation to five 21 figures.
22 Could you tell me, you said require licensee to 23 confirm regulatory design under these conditions.
Where is 24 that spelled out and where would the licensee find that, and
~25 how would they agree to then have to do that?
(
Heritage Reporting Corporation (202) 628-4888
l 116
(")
1 MR. PAGE:
Well, now, this recommendation, maybe I
]
2 didn't fill this out clear enough.
3 My recommendation is going to be to'NRR, from the l
4 Office-of'Research.
We in Research don't directly require 1
5 licensees, as I understand it, to do anything.
6 MR. SHEWMON:
Okay.
And do we have a draft of 7
what you are going to send to NRR?
8 MR. PAGE:
I thought I did supply that.
9 MR. SHEWMON:
Here is Beckjord, Murley Research 10 Recommendation.
11 MR. PAGE:
Yes.
12 MR. SHEWMON:
Can you show me,'
Bob, in that, where 13 it says just how they are supposed to do the calculation?
f( )
14 MR. PAGE:
Oh, no.
We don't go into that kind of 15 detail as to how they are going to do that.
16 In other words, I would imagine that the 17 individual event will help, NRR will have to make that 18 determination.
They would probably seek our help if such an 19 event happened.
But I don't think you could clearly at this 20 point state precise guidance on what it is you want.
21 MR. KERR:
Are you sure that NRR will know what 22 you mean when you tell them to require the licensee to l
23 confirm, then, if you don't give them any details about how 24 they confirm it?
25 MR. BAER:
The wording may change slightly.
I
(~\\
\\J Heritage Reporting Corporation (202) 628-4888
l 117
(~T 1
think we have general agreement with NRR.
s_J 2
What we are trying to do frankly is to say we've 3
only analyzed one situation, a 177 B&W vessel for 100 4
degrees an hour.
And I want to make'it clear that we're not 5
sure that that can be completely extrapolated to all other l
6 vessels for all other conditions.
7 MR. SHEWMON:
Well, you say you think it's a "no 8
never mind," which is clear enough, but here you are making 9
a recommendation that in spite of this we think they ought 10 to be required to analyze something if they go over 100.
11 That is because you did your calculations for 100?
12 MR. PAGE:
That's correct.
13 MR. CARROLL:
I'm a little confused.
None of the f-stuff we have here is the letter that is going to go to 14 15 licensees.
16 MR. BAER:
No letter to licensees.
17 MR. CARROLL:
Well, ow does a licensee know all of 18 this?
NRR is going to write that letter?
19 MR. MINNERS:
Why does a licensee have to know?
20 MR. SHEWMON:
If there is a requirement on the i
21 licensee, the licensee has a right to know about these 22 things.
23 MR. MINNERS:
But this is when you are operating 24
-- That's what the operators do.
We want to ask every l
25 licensee who is not doing it to do it.
l C~) -
Heritage Reporting Corporation (202) 628'-4888
-mm--m_______. _ _ _ _ _ _ _ _ _ _. _ _
118
( s)
(~
1 MR. CARROLL:
I guess if I was a licensee, Warren, 2
1 would want to know that this sword was hanging over my 3
head, and I may want to do a bounding analysis in advance of 4
what I think the worst case is so you don't keep me shut.
5 down for three months while 1 get the analysis done if I 6
ever have this kind of event.
7 MR. SHEWMON:
Do you put out an information notice 8
or something 31%e that?
9.
This thing is resolved when Research writes NRR a 10 letter and says we think it is resolved?
Is that right?
11 MR. MINNERS:
If there is no action from 12 licensees.
13 MR. SHEWMON:
NRR doesn't have to act on it.
(}
14 MR. BAER:
It's a pretty rare event.
15 MR. SHEWMON:
I'm trying to get how a generic 16 issue gets resolved and what tracks get left, is what I'm 17 trying to find out now.
18 MR. KERR:
So you tell NRR what you recommend and 19 NRR then does nothing because nobody is supposed to operate 20 above 100 degrees F.
Is that right?
And only somebody 21 operates above that, does NRR then have to decide what to 22 tell them?
23 MR. JOHNSON:
Correct.
24 MR. PAGE:
This is even more specific than that.
25 This is a natural circulation which also violates that C(
Heritage Reporting Corporation (202) 628-4888
119 sg k_
1 criteria.
In other words,~the 100 degrees per hour I 2
believe is fairly standard for most plants under any --
3 MR. KERR:
No, I was.just wondering at what point 4
somebody has to do something and it is.only when somebody 5
violates these criteria then NRR has to decide what to tell 6
them to do.
7 MR. PAGE:
That is our recommendation to NRR.
8 MR. SHEWMON:
Any other questions?
9 MR. CARROLL:
Well, I still don't feel very 10 comfortable with how this thing is being resolved, I guess.
11 MR. PAGE:
You're more sympathetic to licensees 12 than we are.
13 MR. BAER:
Jay, let me take an example without
()
14 natural circulation.
15 Supposing somebody had an event and the 16 atmospheric, they had a trip and the atmospheric dump valves 17 just popped open and they weren't able to reclose them.
And 18 the reactor coolant pumps were running fine.
But they 19 cooled down, you know, 300 degrees in an hour.
20 MR. CARROLL:
Fast.
Yes.
They've got analysis 21 for that.
22 MR. BAER:
You say they have already done a 23 bounding analysis.
Okay.
That I wasn't aware of.
24 MR. SHEWMON:
Analysis if stress is in the head 25 bolts, head studs?
i Heritage Reporting Corporation (202) 628-4888
120
[)
1 MR. CARROLL:
This case, the case we are
%s 2-interested in is the very unusual case where you are cooling 3
down on natural circulation.
That's not going to happen to 4
anybody very often.
5 MR. DAER:
That's right.
I think it would take 6
some other abnormal condition, uncontrolled.
7 MR. CARROLL:
I guess the least I would expect is 8
that you tip off the Westinghouse owners group and the CE 9
owners group as to the fact that you have a. concern here and 10 let them figure out what they want to do with it.
11 They may fe'el they have enough analysis on the 12 various vessels that they employ that it's not a. problem.
13 They may feel that they want to do some kind of a bounding r
14 case generic analysis so that if this ever happens to one of
+
\\~
15 their members, they have something to throw on the table so 16 they are not shut down, pending an analysis.
17 MR. SilEWMON:
Is there any reason why the thermal 18 natural convection would set up a different stress state 19 than other means of cooling this situation you talked about?
20 MR. MICIIELSON:
Channel flow for one thing.
21 MR. SilEWMON:
We talked earlier about it's not 22 where you get any circulation up in there anyway.
23 MR. CARROLL:
You do when you have pumps running, 24 see.
You've got a lot of pressure drop across the core to 25 force water in there.
()
IIeritage Reporting Corporation (202) 628-4888
121 1,O
( /'
1 MR. MICHELSON:
But you don't have pumps running.
2 MR. SHEWMON:
But this 7 percent which may get up 3
there will drive circulation.
4 MR. CARROLL:. I think the number that they are 5
quoting, it is 7 percent when you have full delta P across 6
the core, when the pumps are running'.
7 MR. MICHELSON:
No, I thought this was the natural 8
circulation.
9 MR. SHEWMON:
Let's talk about what happens when 10 the pumps are running.
Then there is circulation up in 11 there.
12 MR. MICHELSON:
1-2 percent bypass.
13 MR. SHEWMON:
See, my problem when I first heard
(}
14 of this was to come at it from the other end and say that 15 head is only half as thick as the walls.
16 When you've got the pumps going now and it cools 17 down, you are going to cool the head faster than the sides.
10 But that was, I just didn't understand these situation.
19 Any other questions?
20 MR. MICHELSON:
Yes, one question.
Did you look 21 at the Westinghouse upper head injection plants?
22 MR. PAGE:
No.
23 MR. MICHELSON:
Because that is a somewhat 1
24 different arrangement you know, in the upper head area.
Any 25 reason why you wouldn't look at it?
I
)
Heritage Reporting Corporation (202) 628-4888 l
g g--
m u
'122 1
2 1;
lMR. MINNERS:. We'didn't look at any other plants
- 2 exceptlthe one design.
4 3
MR. MICHELSON:
Just thatEone.
But'it would be
'4 perhaps,1I don't.know, you'd have to look at-it in detail'.
N 5
But~it is a?somewhat different barrier arrangement'on;the
+6
' upper ' head injection.
7 MR..MINNERS:
I understand what Mr. Carroll ~is 8
saying. A'nd I guess.if the licensees think it's worth their
.9 resources to get. ready for something that.might happen, that-10 might be-worthwhile.
11 I guess the staff took the point of view it's a 12 very rare event, and when it happens, well, then, we'll
-13 react to it.
]}
14 MR. MICHELSON:
How many have we~had?
15 MR. PAGE:
We're only aware of one that actually 16-resulted in a full shutdown.
Now, there have been other 17
. losses of power,'but they usually get the return of power-18 during a standby condition.
. 19 MR. DAER:
You're talking about two different 20 things.
21 MR. PAGE:
I'm sorry.
22 MR. MICHELSON:
I'm talking about the 100 degrees 23 per hour.
24 MR. PAGE:
100 degrees per hour with natural
' 25 circulation?
(
Heritage Reporting Corporation (202) 628-4888 i
2______--______
t
..:n 123
.1 MR.IMICHELSON:
Yes..EI'm' talking about the cases
., 2 like that.
We've had couple of --
J3
'MR..PAGE:
There's only one..
'4-MR. MICHELSON:
And whichfone-was that?-
5' MR. PAGE ' St. Lucie. ;And that was notL100 g
61. degrees per hour.
'St.
Lucie was.approximately 60 degrees 7:
per. hour,' bulk reactor coolant system.
81 MR. SHEWMON:
Mr. Chairman, Ifthink'we are Ac 9
approaching.the end of this.
10
.MR.
REMICK:
I'm glad to hear that.
Lil
.(Laughter) 12 MR. SHEWMON:
There.is a letter which we have j
13 ready which is basically-one sentence, saying we heard about-(]}
14 it,.we have no objections to the proposed action.
11 5 MR. CARROLL:
However, we think the concerncon 16
. Westinghouse and CE plants ought to be-shared.with the 17 owners group or.something to that effect.
.i 18 MR. REMICK:
Okay.
Let's not read the letter at i
19-this time.
20 Why don't you two get.together and propose a
{
21-letter for tomorrow.
22 MR. CARROLL:
Okay.
]
l 23 MR. REMICK:
We thank the staff very much.
24-Gentlemen, if you would bear with me, I would like 25 to take up one other matter yet today before we adjourn, and i
. ( ).
Heritage Reporting Corporation (202) 628-4888 i
t'
!q 124 f%.
1
'that is the subcommittee report from Mr. Michelson'.
i ),.
s 2
'(Whereupon, at 5:35'p.m. the portion of the
- 3 meeting.to be transcribed adjourned, to reconvene at 8:30 4
a.m.
the following day, Friday, August 11, 1989, at the same 4
-5' location.)
6.
7' 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25-Heritage Reporting Corporation (202) 628-4888 j
t i
a
G l1
)
' ()/.-'
1 CERTIFICATE j
l g
'-2 1
I 3
This isjto-certify that:the attached proceedings before the!
4 United States Nuclear Regulatory Commission in the matter j
l i
5.
of:
i 6
'Name:
352nd Meeting of the Advisory Committee on Reactor Safeguards
]
e 7 '-
8 Docket Number:
- 1.
9 Place: Bethesda, Maryland 10 Date:
August 10, 1989 11 were' held as herein appears, and that this is the original 12 transcript thereof for the file of the United States Nuclear 13 Regulatory ~ Commission taken stenographically by me and,.
14 thereafter reduced to typewriting by me or under the 15'
' direction of the court reporting company, and that the 16 transcript is a true and accurate record of the foregoing 17 proceedings.
1B In/k MA i
f%W 19
(. Signature typed) :
Irwin L. Coffenb>r 20 Official Reporter 21 Heritage Reporting Corporation 22 23 24 25 t
Heritage Reporting Corporation (202) 628-4888
L E
N S
W S
S O
N EG D
O V NL T
I I
O P
9 RR M
O 7 OUC U
T D
S EC S
N A
UAS O
S ES S
I S RET S
I O
S S
I RC Y
T N
R C
TE L
L O
A WS V
N T
U I
I S
RP N
E L
D A
A E
T R
D O
N NDA C
U N
N E EM S
E O
A I
M G
Z RL R
T S
YEA G
R Y
M L HR K
O L
O C
P A
C A TU A
M N
E N
T B
A R
I A
A N
N O
U o
o o
e
J'
\\.
/
O I? Q @
5 3 E5v
_ _. s s.
- ~
s s
/
\\
/
\\
l
\\
u c4 f*
/
i r Ji
~?
~
d' h}
2 O
')
[
I PWR Reactor Vessel k
d 4
8/
9 1
8 1
/
0 5
1 C
3
/
8 D
D
/
1 N
O 3
2 U
E A
O 0
1 Y
N 8
8 R
/
O G
6 R
D O
I K
E U
T T
T A
C T
S C
A I
1 E
B L
E F
I E
S T
I C
A O
C I
C U
R N
L E
D W
N O
T E
E S
G A
B o
o o
o O
3 8
/7
-4 8/
E 5
U
)
S D
S Y
I E
T U
C I
I R
N R
O I
E I
T N
R N
E P
3 4
G 8
8 F
O
/
/
O C
L
(
A 7
3 L
O IT A
D N
V N
E M
G O
U T
U O
R W
P I
O D
O P
B P
R E
A G
S M
H
/
A T
W K
D I
E C
D E
W I
A E
Z V
G E
I I
B F
T N
R I
I T
R IT S
N O
E R
I E
R E
C D
I P
M A
o o
o o
O l
i
)
D 8
E 8
U
/
5 4
N 8
I 4
/
T 8
6 N
/
G 9
0 O
8 O
1 W
/
C D
B 6
E
(
O T
H T
C D
R A
T
.T I
N O
R W
R U
P T
O E
N G
P O
R O
N E
R C
I R
G G
T O
L E
L K
W N
E N
C B
B M
B A
B o
o o
o O
TN T
N G
E G
D O
A O
S W
N W
T S
E B
S B
S N
+ E P
L r
D E
D N
L O
EN O
N C
/
B F
F NT E L I-Y EL U
E L I
T O
0 I C I
P O
0 5
FI VS A
V M N E
O 0
3 NR E E M
G 1
3 I F RR A
R A
S U
S S
A T
E S
N G
W O
I A
S N
O S
T S
P A
LE Y
S E
O L
ET L
G N
+L TE F E BN A
O O
N G
C r
C E
O ON OA O
N W
h TI TA T F T
A B
F
/
T L
R I
F 5
T O
2 0
I I
I F
I T
I O
0 0
G 5
4 I
ATR E
O C
S P
E D
A M
DR N
U F
N I
AU W
T R
O I
ET O
S E
T) G T
I HA D
E N N D
R L
V H O
N I
G RE O
C E A
E E NR DE A T S R F L AF I E O
O E
C F
L R
R E
F V (S
.,e~,,
i
\\,).
\\
\\
\\
.m..
Transient Condition 1 kN il l '
i i !4e i ail l
l 11I I
I a lll t i f i I(l 11 1 4A i e i i
Flange Transient Conditior.2 lll Interface lll l lil i I Transient Condition 3 Ill l ll1 l l
h \\, \\,,_\\ i V.
.O 1
l Figure 6. Void Level Locations O
l
- l-i
NEAo
)
g_-
50 0 l
!\\
mk Leo j. \\
O
\\
T rLE(
Ho R 500
\\
% co tec
~
)
200
\\
d_REstuRE
_. 3
\\
\\.
/00
\\
1 o
2_
4 6
7 8
9
/o
//
/ 2.
/a I4
/ E.
Tine (Hourc)
O I-Figure 3. Cooldown Transient Analyzed 4
___-_.___m._._________.__-__..._____--__._-..__.________________-__.______-__--_.._._..___..___-___.._.m._
________._______.__J
U S
D TS U
T L
E E
S E
L E
L L
L L
S B
L B
S A
is E
B i
E A
s E
p S
A W
H W
p V
S S
W O
O 0
E O
0 R
LL L
L 0
V L
0 O
A L
E 2,
L A
R A
8, T
S 8
1 C
E S
E 2
O D
1 T
E 7
A D
E V
C D
E O
A O
R C
O S
R C
E C
O R
T T
L S
C S
U A
E S
E S
R E
R S
T E
S R
R i
T i
G O
s S
S S
s S
p N
p I
I S
E E
D 0
5 R
L 2
N L
1 N
L Y
A N
0 E
N T
N 8,
L S
B R
B B
B A
0 B
9 L
8 N
M 4
S A
E U
A IX M
L A
P I
R E
T G
E A
N M
R A
i U
E s
R M
s V
p B
i M
p A
G 7
I G
E G
X O
9 A
O 0
M W
7 M
O M
W 1
U B
W B
4, M
M B
8 3
U 6
IX 6
A M
I M
XA M
O
q.
!lIl
.)
I
.- O
~
L LN N
B B
F F
U 8
U C
0 C
3 5
5 L
L A
0 A
7
~
T T
O O
T T
)
L L
D L
N 0
S N
8 E
L B
6 D
B 4
U E
1 U
5 F
N H
F 0
T I
S U
S U
0 T
C C
L L
N E
E O
S S
S G
S G
C E
O E
O V
W V
W
(
B B
S R
R 8
3 T
O F
6 9
O F
6 4
O T
U 2
5 L
T U
3 3
C C
U C
C 0
0 A
7 7
A S
E L
0 0
'E L
0 0
E R
A R
A R
T T
O O
T T
S E
E I
U U
S G
G Y
G G
I I
L T
O T
O A
A W
5 0
A W
F 2
5 F
0 6
N B
D 0
0 2
5 A
F 0
0 F
0 0
U 0
0 U
0 0
C C
S S
E E
L L
C C
Y 0
0 Y
0 0
C 2
4 C
2 4
.o O
N N
O e
- .I ll ll l
1l ll i
S D
A E
O R
Y L
U R
S A
Y L S
O D A
S I
S TL M
B S
M O E OE I
C A SR R TDR R R L
PNEE P
U A
N F ESH S
- Y OO LNUT I
S A OA SS B T
E E'
R D NPC&
R R D OOP OM I
G O
.L T E PR A
S S
S OTY O
S C )P C N
U RNB H
I D
TS O N YA A PE S
RSO A O
RC O C D D OEH H IT A
L S
E A R
,C I
M S M S O O UL ET LE NL E
N R O A R
I R A ESAM I
R PI C T
CL I
F TI S
A H YX F
O E
DCN C
T RAE E
A
,R D
EE A Y
I CNN
.FDLU GSH R
S A
C A NOA ONAS AI S
RDE M T T H O D I
S I
C C MC E
I ELM A
E VO R S E E UE R
R A S&
PDSM S S (R P I
TS E
G L )Y A
N N
T I
A I
R D
XI AS B
N N
I M
E R E B
T T E
M N
I E
S G
S A
E R
R E
T V
S A
(
O
F
)
0 5
D S
2 P
E N
M U
O RE I
N F
T OT I
L A
F T
L 0
L E
E 5
U FS N
H C
0 2
R U
O S
L 5
N W
A O
C 3
I V
O C
CG
(
R L
P R
E NE L
S EEM O
T I
I L
S Y
TTTH U
S L
CA S
E E
N UVE T
S N
O D RME A
E H
N EOLU S
R G
N O
D Z
Q N
ZE U
R C
S O
E OVOD I
T C
T CRNA S
N A
Y E
O E
L R
C H
A UT N
C o
o A
AR O
F
)
D F
E 0
U S
5 E
N D
1 V
U R
E I
1 T
T S
K U
V N
C O
O V
R B
C R
O E
A S
F M
N
(
S R
I S
A O
G S
S N
R O
T S
O O
T A
L E
I T.
A M
U N
T H
A D
c E
L E
T S
1 G
U R
K A
E U
C A
U R
O L
P Q
T A
S C
M E
I E
O D
1 S
R 2
C K
A Y
U L
T A
C o
o o
o A
N R
O A
F l
E A NRI GUR E
T E I
E S C T S O E I
A R N ND R C E
F C M Y S
I R R S LR O
D OE I
EFT E
N RNA S D
I O L S H E U C UE G E E
Q GRUC M
S EOETOX A
N RTRS TE S
O I
TA D
R H
N
/
E F
0 9
E A )5 M
LF 0
B 0
M 0
A7 2 O
1 T7 S
C R1 I
E O
P NE R
W EOT RNN B
O WF A
E &A F
L G B L 7
E 7
.R B
WO 1
R RC e
O OC F
F N (i o
o l
lllllllllllll lllti l
SD M
R A
A R
U C
G S
EF D
A E
S N
E N
HR T O A
T L
R C PN 9
O A 8
F UE 9
E C
1 GR R N N N O E 0
o FO 1
TI I
CR T
E E AE S
IRE E P U
BT R X G
T E
U DI R
OM E
A EM TA A OC W
Y T
R H
O G
I S
L IV D
S.
A U
O
GN ID 1
S U
A LC A
S T
N I
I A
S D
Y S
Y R
P R
T U
L E
S A
L U
O S
R N
3 D
M
- 7. M N
G A
0 A A
I Y
5 S
R D
R R
N G N
T N C O O A
S R S E
DR F
U T D EP F
D SS A
N O *1 E AN T
I AN BR S
M PN SE Y
RF NL YA W
P LP E
TS A N AD IV N
SU NN E
E D
AA R
M N
LS R S M
E AD E PU O
RN E
R E E PO C
T VR E R E
F I SR A
FO S
O N E R
T E N O
E R T
V N A XW A
OP EO F
O l
5 EM 4 U 8 5
L9 K
E CD W
O1 VY M
AN E
5R I
BA V
7A U
D Y E
2 U
L E
1 E R R
T N
SW R
R G
O F
T A
E O
E VM EUE E
P R
J DV P
E U Y C
I 5U N NE F
R N R 7
A 2D EI R
O L
OM I
N RO R N
A T
1 ET N
M".
GE P
MU9 E
O I
DE F
OED X EP T
I 5
US8 RD E T R U
E E DE19 L
H M NC UA GT O O
T U EN H I
N MF S
F N D TSF E
O L DE C I
FN AN F R
O DI R O A RA A NV A RA EM ER T D
S O5 DP S
E T 7
EXH Y
PT T N A1 E
IT2 TEG R
O C
E T
E MU O
TR R L
C G L
T APE MM BR MR R S
REH OO UU O C H I
DRT CC PN C "S T H
8 8
9 9
8 8
8 8
/
/
/
/
4 3
9 0
2 2
0 3
/
/
/
/
6 2
3 5
O 0
0 0
1
S EO E
ST C
U S N
AE E
S CVS I
R M
E T M I
E A
H A A P
R TI R X
C T C S
E S
NI E
ENS L
D EI D V
A E
WYE IT N
N T RN O
N E TN C
I E
T A
B S A J
A L
U L P
P DP R
N HNN OB I
O ES U
ND T
O A F
R FR O
ON O I
O OC S
T A Y E
AS C P
KS S
LM N E
CR U
E AE AO A
R R
BT C
HS E EE H
T DR FR T
EEF EG E
TNE DN N
ANH U AT I
I I
V D M
L L E OR A
VN U P G E
EU D OE O
OFE TR T
TOR O
88 9
1 H
GU O
W RH E
T I
S S
S
.V D
S OR N
N E
E Y
V R
B T
O S
T T
N N
S A
A N
H L
L O
P P
C IS A
E E
U O
R R
L R
U U
C P
T T
N P
A A
O A
M M
C 0
\\l l1
~
=
S S
E A
P C
R R
R G
(
U O
S O
R M ).
S P
AN E
Y N
RO S
F O
CI A
I I
T T
T S O B
N C
H R M A
E U
C O
T D
L D
FD A
A 4O D
O A
E I
I oO 8R A
T E
M T
R 9
R L
1 P
EO O
D D
A T
E E
R P
CR N
T R
C A
NT R
I N
D O
O SS I
M SG S O E
S S
PN SC R
D M R
UI REH E
N O
O D A
F N
LT T
A R RI N
I LW E
TC D
GF L
US E
S S AS A
C D T
RV F N T
I O O A
EE S
E W T D
SN V
N D I
I E
AN E
W W A
M BA R
O E U A
AL E
I S
TP R
SV EC C
AN E
DU L
NR
. o 7l L Ll l
E SA M
B A
S R
ATM C
S AA F
DR O
C E
S S
M N
I R D T
E E T
N A
I oTN D
L E
E S
AA R
V S
P L.
R E
E D P U
L R
G N N C
R O
C E
M A U O
W A
U P
A S
R L
E R
E S R M
E T
M O
R T
N N
D O A
B A
A P
C A
I G
N F N
M D
R R
S T
Y I
C O
F S
T S
E T
U M
S T
O T
V M
R N
N A
E C
E N
I A
T A
R R
M A
P A
T R
I L
E C
I E
Y L
C C
P L
S T
R T
P A
S
)2 F
O o
1
(
I
~
R E
O S
R A
R B
E S
A N
TM A
AA M
D R U
C H
S S N
N I
R D D
E E EV oTN L
TN O
R AA V
O D
PL N
R E
I D P L
R V
N N E
E L
A U N
N O
N A
V R
O M
N SD O S
I U
T F
R N
E H
N E
P F
E R
F O
N T
E O
E O
S E
S P
U P
U M
A Y
A O
C T
C C
)2 F
O o
2(
1f1'I!
I I
O' LOp R
T N,
L I
O S
C E
S 1
N L O R
AI E
CT M
U E
I R
R B S
I O
TI l
U I
EFE CR I
ET A
I Y RSR LS C
A UNU EI LAI I
L D
L
~
TS EARA RM T
RFTF
^
O M
E N
+
V PT E
\\
T E
P S S
M I
UY I
I Y
P SS S
IU Q
D R
E E
T R
M N
RR A
TO OO E
I HP H
TP 1
I DCFET D I
C OU T
OASRA K RTNU L S
I U A W
RN AL E
S S
OORA I
R CCTFIN B S
SST NTN ORE E
Y I
OS R L
IT C
R P I
O U AV NLR A TS I
I NAT F
S SI 0
CR 2
OCEDME A E I
C R G NO D 1
PW R TA UDIO AO O CTLONDCP E
OLORA-PEVGR
~ -
R E
W O
1 I
P CD O
,i -
l
. O.
H 9
1 0.0.0.0.0 0.0 0.0 941 9331 373 T
8 I
9 0000000000 W
1 S
S 8
40462421 21 8
1 1 0.0 0.0 0.0.0.0 T
R 9
0000000000 H
1 N
L A
A L
C 7
095044 331 I
8 1 0. 0 1 0. 0
- 0. 0 0 Ps T
9 000000* 000 S
R I
1 E S C
RS 0
6 1 302255 32 UN 0
8 21 1 1 0. 0. 0
- 0. 0 0
9 T E 0000000* 00 S
1 1
AS R
R E
H MU P
5 8465564324 L
.O S
8 1 1 1 0.0.0.0.0.0 0 A
T O M
9 0000000000 C
A H A
T 1
I G
R I
C R
S MIN S
4 4661 822634 C
8 31 1 1 0.0 0.0.0 0 0
9 0000000000 0
ET 1
0 TS 1
SE R
9 E
YW 8
P 9 S
N S1 O
E MF G
V T
A O IT N
R C RIR N
RR E
E O D A
T L
ST I
E AD O
R A O N
A WRLI RO OMT
- 0. U E
1 A
I DPABER C AA 0
T ERCRNL RES
<Q I
F I
E 1
I T
NCCNNTECND AR I
NHAI N RF S
AE I
I I Y
A EL AA OTE A O
. O S
MREMMCORMC A
(l[l1
- Il!
l l
l l
Tll uJ e.
u O
s 9
565 44 MS 8
1 0.2.0.0 0.0 S
S 9
0000000 A S R
1 R N H
C L
8 1 1 7 24 SE A
98 1 1 1
0 0. 0 000* 000 S
C 1
F U I
O T
O I
7 850 25 R
EH C
9 2.0.0 0.0 8
01 MG 0000000 S
1 0
I 0
R TN 0
H I
1 6
5051 48 1
TT R
9 2.2 0 0.0.0 L
8 1
0000000 A
AS E
1 E
P C
S I
SW E
T T
5 65621 69 I
dE H
A 9
- 0. 2. 2. 0. 0. 0. 0 R
8 0000000 C
R T R
1 0
G I
S 0
OW M
4 801 2694 0
1 8
1 R
A 9
3 3.0.0.0.0 0000000 R
S R
P T C
E9 1
P8 NN S
9 S1 I A MF S L G
AO E P N
R R I
I C
TE T
S E lV R E
- 0. A 1
NAO DHN 0
TT SE OTITESA
< Q U
CA I
I S AN EN T
GT Y LE E T AM I
I V
NSD RIFBT T S I
B TN AI EUN A R IT SAE LD OI RI C
ERTANR A F
o A
TTSCUTM Ll-l l
s SSS N
- 9 39 6
E 8
21 00 S
9 0000 S
MU 1
S AO R
8 72 5
RH H
8 21 0
9 00* 0 S
CG L
1 A
R S N C
H DI I
7 1 31 2 E T T
8 31 00 L
I 9
A 0000 S
R 1
C N E C
I N
T OA W 0
6 51 25 I
8 4200 R
00 LH 9
C 1
0000 1
0 PT R
0 NI E
0 UW P
5 5433 1
8 4200 S
9 R
0000 FS M
E9 1
OT A
9 P8 N
R 4
757 S1 E A C
8 620 MF S
S 9
000*
AO UL R R 1
P A
C E C E S T R
R O
- 0. AR 1
U TR E
0 T
N R R
U A
E ELU
:nU tnalP erutaV O
~~
'=u
~
W)
T f
O(
A TT S
S EA P
T R
N U E E
DC N
S M
M M
I EN E
E E
E CA S
V T
A C
NM A
O S
R N
AR E
R Y
C E
MO R
P S
S I
RF C
M E
Y R
OR E
I N
N E
FE D
M A
I P
R P B
Y O
R T
X EM B
R U
U P
E E
N F
T O
TT E
D N
H 9
NS T
3 AY VS E
I T
O N 0
LS R T T
A I
I N
L M
W A
PR D A U
L 2
DE 6L S
D DE 5
E T 8
P E
N T
P 0
VA 9
R A
A E
E OW NG S
R R
1 R
8 RD T
E E
I U
6 P E SM N
T P
0 T
ME NO E
.A O
I F OR MSTWE A
9 7N
.S I F E
0 9A C N V NDC T M
7 8I T
M U O O AEN N RPFLEA A 1
L I
3 D
DT P
MP 8
A E RB I
8 VS AOG8 I
8OT MR 8D MF E 5
9 S
R E
9RN AT 8N 1
9 AEE V T
PA RN N
1 I
CPFI 1
A NML CO I I P SC N E R
C I
O
m MA E
R TE S
5 A
9 C
AN 8
S F
1 S
RI T
O W
B N
R E
M M
MR O
T O
E A U A
T A
R T
R T L
A R
F S
CN Y
SI E
I P
T M
T S
L A S
N A
A R
LM I
I R
R E
AE S
Y C
T RH S
M A
ET S
R N
A M
A W
V N OE 4
M E
R D
OI 1
0 I
T C
E ES S
R S
S E
HT P
Y L
F TN U
9 S
L E
4 N
N M
R A
O 0
M I
I E
E R
A7 H
8 SE 4
T T
E M9 EV N G
7 S
A V
0 YS T
GOIO 1
N SM W
O N "r NRT N
I 5
RA D
E Al APU E
HMB T
7 ER E%
H C
M CI T
0 TC E 0 F E L R I
I S
AS F 5 N
L NV L T I
E 0
WD NY 1
I L A N W
DE A B E
G O A
I S P M M O R
G 1
I C
EN MD N
S EN E
A DM 8OM S
S F A 5C 8
H NI 8 TE E
NL U
OE 9
T 9
C T
C T E S I P D
6 1
A AN 1
R MU N E O8 EA N DS U Y 9
R N1 SR I
I O
~
l
98 941 9331 9
1 00000000000. 0. 0. 0. 0. 0. 0 373 1
8 404 62421 8
s 9
1 1 0. 0. 0 21 0 0 0 0 0 0.0. 0. 0. 0 S
s1 00000 r
S u
S o
N H
H lac7 09 50441 i
H t 8 1 0.01 331 T
C1 00000000. 0. 0. 0. 0. 0 ir9
<000 I
0 W
0 0,
S 1
T r6 N
e8 213022551 1
32 p9 1 1 0.0.0. 0.0. 0 0000000000 A
s1 L
m P
a O
r E
c S
R 5
U 8465564324 8
T 9
1 1 1 00000000000. 0. 0. 0. 0. 0. 0 1
AM TA 48 4661 S
9 31 822634 001 1 0. 0. 0. 0. 0. 0 M
1 E
00000000 TSY S
G N
IT no A
i 9
t t
I r u e
8 ob 9
T ri v
i 1
I et r
ri r
t ct t
oD n
f N
a e s et a
o I
wt i
l o
na d rDird o
r ePl b e o ome e
t r n s
r u eR Ca a a
e a
t r
m F
c e
oiTGl s
u r
t e
r o
oS n q
nt t
ict y
a a cinntr r t
e s
e c nd t
i n
e e a a s
MREMMCORMC oh ai n r
S l
t e a o
iF g
M h
A D
R E
S C
N T
S N
M IA D
A E
ST M
E L
T E
V P
S N
R O
E Y
A V
S N
R E I
L O
P N F
R MI E
P IT I B O
T s
U SR T
A SS B
TU E
W CT U8 I
S R
E D
N T
LN D89 E
9 N
FI S1 E
C 3
O E A T N F
RM NI N
OI 0
C 7
R 8
M9 TN MM A
T 4
E1 C
5 T
A A EA M
E 0
U R
V R N
S L
Y R S
OC I
E 3
SH MP RS E
T A R 8
P O N8 6
R RE M N L1 L
0 E 4 I 9 8
CH I
0 A1 6F CE EI 8
WM 8O NC D D E
0 9
E AN EE N
DO 1
E ER OC MI LR E
2 EF G
7 F
TN RR BR 0
9 RX OC AA R
EE NO 1
S T
O P T TE E
MS M F T
HO RE 9H R A 8
T FR 1
O
E N k k k k k N k k o.
oooooooooo 3
E.E.8.U. S.M N N N w m
cocco
.m oooooooooo eZW SEkkkE o o o o e o.k k k k E
'C -
oooo o
t m
b H
~*
N S. N $. N S. e N m m Z
(E 9
C o o e?
C.
C o o o o. o. C. o.
g ooo A
'E A
E W
D o
,5 e
4 h
SSNSkkkkkk ooooooogoo m2W b>
m 0
5H E
8 6
m e
d 3
j.*
o
~
E o
m w
S
?
G
=c o-E" j.5 EB3
{"
"5 5o lj uQW SEE L
AF Egh,O$j f
a M
$ 6 m 6 8 g.y $ W c
.E 22k2 GOO 25d I
N
-X O
E R
N P
I S
O B
P T
T R
A N
E U
Y A
U T
B 8
S 8
NM D
9 L
DS 8
M 8
I 1
AE E
P ER E 9
T N
N T
M3 1
I I
S E
L D
S O
D Y S
WY T
E S C
E S
E R
N S S 7
VR D
9 OO U
EG 8
G GN E
C RT T
C 1
N NIT M
PA A
O 8
A A MR R
IR 2
HI O
I E S
0 CT R
ON M
M cE I
F E
EN TE A
A 7
TI G
R R
E N
6 AL 0
UN C
C I
0 RL 6
DI S
S G
A Y
A N
M EM M
M 3
S B
C E
E E
7 A S 0
RO D
ND T
T N
C R E
AN S
S Y
Y S C N
MA O
2 8
I S 8 S
7A L
R R R9 E
I 8
8 T
0 9S C
O E E
N 1
E E
F T TN S
1 I
7 HG D
R A AI B
U E
R B
8 GN E
PW W%
0 UA T
U M
OH A
D D D0 T
6 O
RC R
E E E
V E EY N
S HL M
O I
F FL A
C E
TL A
RN NE M
T T
A 4A R
P A AA O
I I
8M C
M M MM N
R 91 S S
I o
l O
E 4E??;?????
oooooooooo a
.E
%?????3544 g-oooooooooo O
z
=
W 1
- !E n: 5883B833 U
M u-6666666666 h
=$
'a EE 94~?5EE344 g
oeooeoeooo en -
.J a-e w
=
5 E
GRE9??5$4R H
oooooooooo Is j
E RER49?????
g oooooooooo wH=>w 0
E g
s H
=
8 EU 5o'858 t:
E3a E
~ 3.!32 $ h ! U N
$5 O
nnnvag g
3 5 5 8 = = = 5 51.]
.s
- 22mosa2260 t
MA RR EC N
M TS S
I T
A O
AE N
M R
WH 0
F DT 4
A E
L H
O ED Y
EE B
P T
W F T T
S N
NC D
E E
I F
I S
S O
AF N
S E
R MF L
I N
G O
E C
A 7
N T
HM E
X 2
A E C
TE D
O 0
H T A
T S
QC CA F
O S A
TY L
3 YR A
H NS I
3 BM Y
OE E
W 0
T NA B
TIN A
I ER D
I 6
VC E
D B R
4 R
I K
0 R S S
D M
U C
DM A
A T A
E E E
R O
9 R T R
N N C
1 I
C 0
A Y N
S C
I 7A S
B 89M E
SS I
A H
E E
1 1
B 4
T R T5 D HE T
E 8
0 A T A9 NT 68 R A R1 A
9 MW MO M
1 S
9TE C
A RE R4 ST 1
R CE C9 N YA N
8 I
SF S1 I
SR S
Q
9
& '828tsggg!
R
==desese a
Q lE j!=8888s88 z
=desse
.g
- jnlI is!
5!!!!!ggg EY 48ss8ssgg R
g j'
====esese i
d o
E 353e=8sssa 2
9g eees l
0553E!!g!
l t;
{
w I
I O
EF g
=
E 2
l t
y Z
==35 N3 o
jfpisogg I
a m5g3Mf5~?
8 l
E sa s
e.g.g&bbhbh5 l
~
u N
22 1
L l
l
M S
A T
D N
R E
E C
S N
SE N
SU O
D I
DE8 RIN P
E E
4 N
M E9 DT O
N C1 DN C
UC SN E C R
P E L N
VD TA U
I E S V
AL AIC S
A %
HU WI
'Y H5 L O 0
F S
A H DE R
SE S
EN T
N TT R
E E S
ES F
O R A NM O E D
B U
OR IS F
EA T
oU FM GR B
N N Y I
E A NG O A H
E L
O NR I
I C
OC SR TM T
N I S MP N
TE AN C
U R RN TT HI D T GA AS TIZ S
O S Y
S L ER U R A YS T B I
D P E RC S
MN SR T H E A E
G T AI P A S
S R E U
U G'
E NIN U
ST R
B S
YG G R B
SY I
RN SR RR A
E C N
T L E U U D D E
RD AT N E WH UN NA I R OT FA AR o
I
'