ML20248J026
| ML20248J026 | |
| Person / Time | |
|---|---|
| Issue date: | 10/03/1989 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1762, NUDOCS 8910130008 | |
| Download: ML20248J026 (172) | |
Text
m2 p
s
!G yy OR
-0 UNITED STATES NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR SAFEGUARDS In the Matter of:
SUBCOMMITTEE ON MECHANICAL COMPONENTS
,ua n. e s.u
{
I TRO4 S
g DELETE B. h1ITE RETURN ORI$INAL 70 B. hMITE,
^CRS " ~" P-5/ r O
Barbara Jo hhite t
I 6 972 8 Dages:
1 through 132 Place:
Bethesda, Maryland Date:
October 3, 1989 n
LngS r ce Copy
,ne.7 air
................;fpp y.. y 9.gq q g.g y Q g......
O HERITAGE REPORTING CORPORATION OficialReporters 1220 L Street, N.W., Suite 600 Washington, D.C. 20005 3:.1 0; 1.,g,-
- ,c,,
(202) 628-4888 jaj_yhi
- .g.
i 1
~()
1 UNITED STATES NUCLEAR REGULATORY COMMISSION 2
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 1
3
)
In the Matter of:
)
4
)
)
5 SUBCOMMITTEE ON HECHANICAL
)
COMPONENTS
)
6
)
7
- Tuesday, October 3, 1989 8
Room P-110 9
7920 Norfolk Avenue Bethesda, Maryland 10 The above-entitled matter came on for hearing, 11 pursuant to notice, at 8:35 a.m.
12 BEFORE:
MR. CARLYLE MICHELSON
()
13 Retired Principal Nuclear Engineer Tennessee Valley Authority 14 Knoxville, Tennessee and Retired Director, Office for 15 Analysis and Evaluation of Operational
- Data, U.S. Nuclear Regulatory Commission 16 Washington, D.C.
17 ACRS MEMBERS PRESENT:
18 DR. IVAN CATTON Professor of Engineering 19 Department of Mechanical Aerospace and Nuclear Engineering 20 School of Engineering and Applied Science University of California 21 Los Angeles, California 22 MR. CHARLES J.
WYLIE Retired Chief Engineer 23 Electrical Division Duke Power Company 24 Charlotte, North Carolina 25
2
()
1 MR. JAMES C.
CARROLL Retired Manager 2
Nuclear Operations Support Department Pacific Gas & Electric 3
San Francisco, California j
l 4
DR. CHESTER P.
SIESS l
Professor Emeritus of Civil Engineering 5
University of Illinois Urbana, Illinois 6
'ACRS COGNIZANT STAFF MEMBER:
7 A.
Igne 1
8 j
NRC STAFF PRESENTERS:
9 i
T.
Sullivan 10 T.
Scarbrough R. Woods 11 O.
Rothberg 12 O
.(_j 13 14 15 16 17 1
18 19 20 1
21 22 I
23 I
I
(:)
24 25
3 O
2 es c88oluos 2
CHAIRMAN MICHELSON:
The meeting will now come to 3
order.
4 This is a meeting of the ACRS Subcommittee on 5
Mechanical Components.
I am Carl Michelson, Chairman.
The 6
other ACRS members in attendance are Ivan Catton, Chester 7
Siess, Jay Carroll, Charlie Wylie.
Al Igne on my right is the 8
cognizant staff member for today's meeting.
9 In today's meeting, we will discuss the ACRS concern 10 on the MOV Generic Letter 89-10.
We will also discuss Generic 11 Letter 89-04, which is guidance on developing the acceptance 12 in-service testing programs, and we will discuss the status of
()
13 check valve action plan, status of MOV action plan, and MOV 14 failure to close against differential pressure.
15 The rules for participation in today's meeting have 16 been announced as a part of notice of this meeting that was 17 published in the Federal Register on September 19, 1989.
18 This meeting is being conducted in accordance with i
19 the provisions of the Federal Advisory Committee Act, and the 20 and the Government and the Sunshine Act.
21 We received no written or oral statements from 22 members of the public.
It is requested that each speaker f
1 1
23 first identify himself or herself and speak with sufficient 24 clarity and volume so that he or she can be readily heard.
[}
25 Do any of the Subcommittee members have any initial l
l
p 4
(}
1 comments at this time?
2 Seeing none, then we will proceed with the meeting.
3 I believe I don't have any introductory comments to-make.
I 4
think the agenda is in front of us here somewhere, and it is 5
self-evident.
There is quite a few items.
The main thing is 6
we do have to cease and desist promptly at twelve o' clock I 7
understand, so we will try to do that.
8 Since seeing there are no other comments, I think 9
Owen Rothberg is going to be first up.
Owen?
10 MR. ROTHBERG:
Well, reviewing things a bit, ACRS 11 sent a letter to the Chairman on May 9th, and listed several 12 concerns about our MOV generic letter.
()
13 Of the four concerns we incorporated, I would say, I 14 guess one way to characterize it would be two and a half in 15 the generic letter.
16 One of the concerns had to do with the fact that the 17 generic letter didn't say when we would--there were, well, the 18 concern that we didn't address was ACRS asked us--I'm a total 19 blank right now.
I'm sorry.
20 CHAIRMAN MICHELSON:
For the Committee members, our 21 letter is in this package.
22 HR. ROTHBERG:
And let's see.
Let me go back to 23 that.
24 One of the, one of the things that they asked us to 25 do was to, what happens if you don't have to test MOVs?
When
_mm m
5 (f
1 do you not have to test HOVs in design basis conditions, and 2
we said well, those tests need not be repeated unless the MOV 3
is replaced, modified, or overhauled to the extent of the 4
licensee conditions, the existing test to be unrepresentative 5
of the MOV in its modified configuration.
6 We also put in words about the alternatives to MOV 7
testing.
That--those two things were things that ACRS asked 8
us to do.
9 One of the things was that they said well, ACRS said 10 well, you should allow them the alternative of not doing 11 design basis testing.
We modified that a little bit and we 12 said we would prefer that people do design basis testing, but
()
13 if they have a good reason for not doing it, we would 14 entertain alternatives, and that I guess is the half that we 15 are speaking of.
16 The one area that we have a disagreement about was 17 the problem of the existing design basis versus the idea of 18 well, if there is a change or a modification to the design 19 basis intended by the letter, and we said that there was none.
20 (Slide) 21 MR. ROTHBERG:
In any case, what happened was the 22 staff sent a letter back to ACRS in July and talked about 23 that.
This last--in preparation for this meeting, Mr.
24 Michelson sent us a memorandum talking about the possible
[}
25 misinterpretation of what ACRS wanted, and so we were, we were l
6
.( )
1 to come prepared to answer five questions, and what I've done 2
here is instead of repeating the questions which are in the 3
memorandum, I've just written down the answers.
It is kind of 4
a wordy slide.
I apologize a little bit for that.
5 The first question that was asked was what is meant 6
by the terms existing approved design basis as used in the 7
requirement of the generic letter, and quote, the existing 8
plant design basis as used in the Requirement E.
What 9
documents might define the design basis was another part of 10 that question.
I 11 Well, the phrases are meant to be the same.
The i
12 design basis is defined in the regulations and the FSAR, in
]
()
13 the Commission orders, license, licensee commitments.
That's 14 the design basis.
It's events, and we mean to change none of I
15 that.
l l
16 CHAIRMAN MICHELSON:
It is not just the FSAR thing?
17 MR. ROTHBERG:
Well, if you recall in the generic j
l 18 letter, it says such as the FSAR.
19 CHAIRMAN MICHELSON:
Yes, but some people are kind 20 of narrowly interpreting that to mean that if I said solely i
21 FSAR, then I do, I did and I don't, and your clarification is 22 that it doesn't have to be, it is not exclusive.
It could be 23 in other forms, could be in letters and could be in orders, or 24 whatever?
25 MR. ROTHBERG:
Yes, it is whatever.
I 1
7 1
CHAIRMAN MICHELSON:
Just a clarification.
()
2 MR. ROTHBERG:
That may be the crux of the problem.
3 CHAIRMAN MICHELSON:
That's not the crux of our 4
problem, but that's the clarification sonte people need at 5
least.
6 MR. ROTHBERG:
Yes.
That's true.
That's not in the 7
letter.
What it says in the letter is a footnote, is the 8
design basis is that documented in documents such as the FSAR.
S
{
9 CHAIRMAN MICHELSON:
Unfortunately, that you should 10 have expanded that to give at least one example of something 11 other than FSAR if that's what you really meant.
12 MR. ROTHBERG:
Perhaps so.
()
13 CHAIRMAN MICHELSON:
It goes without saying.
14 MR. ROTHBERG:
The second question was if the 15 deficiencies in the design basis are found during the review, 16 under Requirement A--I would characterize it as Recommendation 17 A--but what should be the process of the documenting and 18 reporting?
19 Our answer was that, that deficiencies in the, in 20 this--and this gets to the, I think the heart of the 21 matter--that the deficiencies in the design basis scenarios, 22 the events and conditions of the plant, are not within the 23 scope of Generic Letter 89-10.
However--
24 CHAIRMAN MICHELSON:
This is deficiencies found in
[}
25 t?
equipment as a result of your investigation, not on the--
HERITAGE REPORTING CORPORATION -- (202)244-8817
8
()
1 MR. ROTHBERG:
I'm coming to that.
- However, 2
deficiencies in the implementation of the design basis l
3 scenarios, again in the equipment, as applied to the design 4
installation, surveillance and maintenance, are covered by the 5
letter.
That is A, review and documented design basis.
6 CHAIRMAN HICHELSON:
So you found one under A, for 7
instance.
Now what do you do?
8 MR. ROTHBERG:
They have to, they have to do what is 9
in the letter-.
That's the purpose of the letter.
10 CHAIRMAN MICHELSON:
The letter doesn't address what 11 you do when you find a deficiency necessarily.
For instance, 12 do you report it?
The letter doesn't seem to have a reporting
()
13 requirement.
You have put it in your documents.
Once a year 14 somebody looks at it.
If you had not had 89-10 and you found 15 the same deficiency, what would you have done?
16 MR. ROTHBERG:
I don't--well, without 89-10, I don't 17 think they would be able to look for it in the first place.
18 CHAIRMAN MICHELSON:
Clearly if I read my SAR, it 19 says I have got to isolate a pipe break on reactor water 20 cleanup system, and then I discovered gee, my valves can't 21 really do that, what would I then do?
Certainly I think you 22 would report it.
23 MR. SULLIVAN:
I thin't as Owen pointed out in the 24 response to No.
3, there are places where it would be reported
()
25 under 50.72 and 50.73 or even Part, possibly Part 21.
HERITAGE REPORTING CORPORATION -- (202)244-8817
l l
9 i
()
1 CHAIRMAN MICHELSON:
Those still pertain, even 1
1 2
though 89-10 has been written, or does 89-10 supersede those?
l 3
MR. SULLIVAN:
No, 89-10 doesn't supersede those.
4 CHAIRMAN MICHELSON:
If you find a deficiency, you 5
still report like you have always reported, resolved like they 6
were always resolved?
7 MR. SULLIVAN:
Right.
I think if Owen goes on to 8
his third question here--
9 CHAIRMAN MICHELSON:
That's not very clear from the 10 letter unless you can point out the section of the letter that 11 tells the licensee what he is supposed to do.
12 MR. MINNERS:
Well, we had a problem with that.
()
13 Everybody has their opinion, and the CRGR's opinion, we tried 14 to put reminders of licensees' obligations to report things 15 under other parts of the rule.
They said that's a given.
16 Don't put it in there, so I think we have very strong policy 17 direction.
We are not supposed to repeat that.
18 CHAIRMAN MICHELSON:
So if I find in the process of 19 doing the Part A examination the torque switches say were set I
20 wrong, is that a reportable deficiency?
21 MR. SULLIVAN:
I think we discussed this in some of 22 those workshops, and as I remember our discussions, we did 23 consider that that would be reportable if it was determined 24 that the switch settings were far enough out that the valve
(}
25 wouldn't be able to do its function.
10
,}
1 CHAIRMAN MICHELSON:
Okay.
So everybody agrees 2
those are reportable deficiencies?
3 MR. ROTHBERG:
I don't think I do because the 4
reason, the reason is if it is not, if it is just a single 5
field, it is perhaps recordable under the letter, but it might 6
not be reportable as a, let's say an LER or a Part 21 defect, 7
so it might not be reportable.
If it wasn't reportable 8
before, then it wouldn't be under the letter.
9 MR. SULLIVAN:
But it is also possible that both 10 valves, similar valves, redundant trains would be set up the 11 same way.
12 MR. ROTHBERG:
Exactly.
If you had that sort of
(}
13 situation, if you had a single valve that had switch settings 14 out, it wouldn't be reportable under NPRDS because it is'an 15 incipient faiinee.
It wouldn't be reportable under 50.75 or 16 50.73 because it's a single failure, and it didn't, it didn't 17 knock out the train, so it might not be reportable, and the 18 letter does not require that.
The letter says that when you 19 find these things--
20 CHAIRMAN MICHELSON:
The letter is supposed to 21 remain silent on these things since it was already told that 22 you don't repeat these, and if it is already a requirement to 23 report, you don't put it in the letter.
24 MR. ROTHBERG:
If it is not a requirement to report, 25 the letter again doesn't help.
_ ~, _
11
. (])
1 CHAIRMAN MICHELSON:
If I find that I, my switch 2
settings are incorrect, I,
clearly I don't report them under 3
702 or 3 for one valve.
If I find a whole series of them 4
incorrect, I have to report them.
5 MR. ROTHBERG:
That's true.
6 CHAIRMAN HICHELSON:
Let's take the one valve.
Now 7
that I determined that it does need to be reset, you don't 8
think there is any reporting mechanism in this case?
9 MR. SCARBROUGH:
If you find that maybe a design 10 defect where the manufacturer has given you an incorrect valve 11 friction factor that's, on the Part 21 that is a defect in the 12 design--there is a lot of filters there.
There is a lot of
()
13 filters.
14 If you can assume that's an isolated event, you may 15 not have to do an LER, but if you can't assume that's an 16 isolated event and you don't file a LER, you could have 17 problems down the road so that--if you submit an LER and 18 alert.
19 CHAIRMAN MICHELSON:
All of this is irrespective of 20 89-10.
This is the way it has always been.
21 MR. ROTHBERG:
That's right.
I 22 CHAIRMAN MICHELSON:
It wasn't so clear to me when 23 89-10 came out whether all these still pertain.
There are a 24 lot of things in 89-10 that weren't the way you would have
{}
25 necessarily done in the past, but the reporting is still for
l 12' old requirements.
1 2
MR. SCARBROUGH:
Each public meeting, that question 3-came up.
It is emphasized if they have an obligation to.
I 4
report according to the regulations, they must. report it.
5 MR. CARROLL:
My way of saying it, Carl is Part 21, j
6 and 50.79 and 50.73 are regulations.
They are the law.-
This l
I 7
generic letter certainly can't supersede the law.-
8 MR. SCARBROUGH:
That is correct.
9 MR. ROTHBERG:
If you look at the way the generic 10 letter is= worded, except for the part about reporting in.the.
i 11 last part, the letter only contains recommendations.
The last 12
.part about reporting is the only part where we use the word
(])-
13 shall and must.
14 CHAIRMAN MICHELSON:
I don't have any p0oblem with 15 the letter.
Only by the absence of this clarification that I 16 have a problem.
Now that we have got it and everybody and 17 myself understands it, then that's great.
We will proceed now j
18 with the rest of the scope accordingly.
f 19 MR. MINNERS:
If you have got an opinion on that, 1
20 maybe--
21 CHAIRMAN MICHELSON-I don't disagree with you.
I 22 just could not get it out of the letter.
It was unclear and.
23 confusing, and I expect that's why you got the question.
24 MR. Ml'NNERS:
If it is confusing to you, and you
]
'25 think might be confusing to licensees, maybe you should just l
l 1
-_---___-_-_____________-__~-n--,n
,, n n s no, nana
l 13
~
[v) 1 recommend to the Commission and to the staff to change their 2
policy, reminder about reporting requirements being proven, if 3
that's what your opinion is.
4 CHAIRMAN MICHELSON:
Maybe we will proceed.
Then I 5
will tell you what my real problem is.
I haven't come to it 6
yet.
7 MR. SULLIVAN:
As Warren pointed out, there may be 8
some confusion, and this question did come up at our workshop, 9
so I think that's further evidence that there may be some 10 confusion.
We are going to try to clarify or eliminate any 11 confusion by specifically addressing this point in our meeting
]
l 12 summaries on our workshops of last week and the week before.
()
13 CHAIRMAN MICHELSON:
That would be helpful.
14 MR. WOHLD:
It might be worth mentioning one of the 15 confusions on reporting, if the torque switch is found set j
16 extremely low and it is found to, necessary to increase the 17 torque switch setting, a lot of licensees are going by the 18 stroke timing, say it is operable per thd.ASME codes and tech s
19 spec, before the test was run, and it is still stroke time 20 after the test and there was no inoperwSle condition, you 21 know, with the torque switch was set wrong, so the valve may 22 not have worked, but as far as legally, it was operable for 23 the desired situation.
/')
24 MR. SULLIVAN:
We tried to dispel that notion at the I
(/
25 workshops.
We did talk about the significance of the tech i
14
()
1 spec vis-a-vis the generic letter.
2 MR. ROTHBERG:
Section 11 of the generic letter does 3
say that the stroke timing test is insufficient to demonstrate 4
operability of design basis conditions.
That's in there, in I
l 5
the letter in the background section, and a lot of--there was, 6
as Ted pointed out, a lot of discussion about that at the j
l 7
meetings, and some people tried to take a legal view of that, 1
8 and it is just not reality, and that's what we tried to do l
9 now.
1 10 The third question that was asked in your September 11 4th memo was how will the regulatory resolution of such design 12 deficiencies be handled?
l
()
13 Again, we said that deficiencies in the design basis 14 events and conditions are beyond the scope of the generic 15 letter.
16 I went through the regulations, and not being a 17 lawyer, I might, I might have missed it, but I didn't see 18 anything absent in an event or a plant physical condition that I
1 19 mandates a 50.79 or 50.73 response, and there may be guidance 20 in Appendix A or B, as Tom pointed out, maybe something in 10 1
21 CFR Part 21.
I went through those, and again I don't see 22 absent an actual event, anything that requires reporting of a l
23 deficiency in the design basis scenarios, so that seems to be
(}
24 a,
a hole, a defect in the regulations.
25 CHAIRMAN MICHELSON:
But the scenario I think is one
r i
15 1
thing. and the equipment is another thing.
I would agree.
[}
2 What you are saying is if there were important scenarios that 3
should have been and some people did not include, for 4
instance, on the HIPSI steam lines in the early days the fact i
5 that it might break, isolation valves were in there for that J
6 purpose, that wasn't even in the FSAR, in some FSARs as I-7 understand it, but that's a deficiency in the scenario all 8
right.
9 MR. ROTHBERG:
That's what we took to be the meaning 10 of your questions.
11 CHAIRMAN MICHELSON:
No.
I was more concerned with 12 really what it gets down to is the final item, and that is if
(])
13 you find that the torque switches are set wrong, is it all 14 right to leave them that way?
15 MR. ROTHBERG:
That's right.
16 CHAIRMAN MICHELSON:
And I read letter as saying 17 yes, whatever your original design basis was, that's your true 18 adjustment.
That's the part that gives me a hard time.
19 MR. BAER:
No.
20 CHAIRMAN MICHELSON:
We will get to that in a 21 minute.
It doesn't say that.
22 MR. ROTHBERG:
I'm sorry.
I missed that.
Okay.
We 23 will get to that.
24 In any case, it was assumed that licensees, that the 25 forth question was can interim corrections of design basis
16
- (
1 deficiencies be implemented before final resolution?
2 When you wrote design-basis deficiencies,-we took 3
that to mean deficiencies in the design basis scenarios, and 4-we said that--
5 CHAIRMAN MICHELSON:- Deficiencies in the equipment;-
6 if-the torque switch is setting incorrectly, that's a-7 deficiency in the equipment.
8 MR. ROTHBERG:
Again, it was assumed that licensees
. ould expedite those deficiencies as they found them.
I can't 9
w 10 conceive of anybody--
11 CHAIRMAN MICHELSON:
That would mean they would do 12 it before they reached some kind of regulatory approval to
()
13 change it.
14 MR.'ROTHBERG:
I would assume so.
It tells them i
15 they are on a schedule of, five-year schedule.
I can't-l 16 conceive that anybody that found deficiencies in his settings 17 of his motor-operated valves would just wait a-while.
18 CHAIRMAN MICHELSON:
Keep in mind the deficiency is 19 there is no deficiency, if you are trying to meet only the 20 original design basis.
The deficiency is now you recognize 21 that you have got a problem, and then you get into that 22 scenario question again.
l 23 HR. ROTHBERG:
That's right.
Exactly.
24
~MR.
BAER:
Bob Baer, the staff--I would like to
}
25 comment on that.
.--~ --- ----
17
()
1 Despite what Pete said, it seems to me if a licensee I
2 goes back to his design basis scenario, and finds that there 3
is a valve that ld not open or not close because of an 4
undersized actuator, the torque switch is incorrectly set, 5
could not mitigate the accident it was designed to mitigate, 6
the handle, then he has to declare the valve inoperable, and I l
7 don't think there is an inspector--if Pete Wohld was back out 8
in the region and he made a determination that this valve on 9
the ECCS line would not open or close to allow ECCS flow, I 10 don't think Pete would have any problems saying that valve is 11 inoperable, tech specs apply, you have so many days to get it 12 fixed.
You do have to submit an LER if you have to shut down
()
13 because of that, and you can't start up until you get it 14 fixed, and I don't see where that is ambiguous at all, 15 frankly.
16 CHAIRMAN MICHELSON:
Bob, I don't either.
I fully 17 agree with everything you have said, but when I read 89-10, it 18 says I only have to meet my original design basis, whatever it 19 was.
This does not include that scenario.
20 MR. ROTHBERG:
It says existing approved.
It 1
l 21 doesn't say original.
22 HR. BAER:
If the scenario is like a large LOCA, or 23 a small LOCA, it has to meet that scenario.
Now if the
[}
24 licensee ordered a valve originally that only had opened 25 against 200 psi and he now finds it has to open against a
.--~ -__ ----
18
()
1 thousand psi, yes, he has an obligation to report that in my 1
2 mind, how he procured the valve, not what we mean by the j
3 design basis.
We are talking about the underlying events and 4
scenarios.
5 CHAIRMAN MICHELSON:
I fully agree he better report' 6
that.
Now the next question is, though, does he readjust his 7
valve as best he can to meet that new condition, or is it all 8
right to leave it adjusted in the old design basis condition?
9 MR. BAER:
My opinion is the valve is inoperable and 10 he has to follow his tech specs.
If I can't make a claim that 11 is now do something to make it operable, he has to shut down.
12 CHAIRMAN MICHELSON:
You are saying it is inoperable
()
13 on the new design basis, but not on the old one?
14 MR. ROTHBERG:
Why is it a new design basis?
16 CHAIRMAN MICHELSON:
New design basis, some new 16 scenario now that you have to meet that you didn't originally 17 meet.
18 MR. BAER:
No.
Same scenario, but that the valve 19 was misordered or misadjusted.
20 CHAIRMAN MICHELSON:
I am talking about everything 21 is correct, but does not take account of having to isolate, 22 for instance, a pipe break.
Now what do you do?
23 MR. BAER:
No.
There he does not have to, I don't 24 think the licensee--
{}
25 CHAIRMAN MICHELSON:
What we are saying, we still
19
/~%
1 don't have to do anything.
U 2
MR. ROTHBERG:
That's right.
I guess.
Yes.
I 3
think even.
4 CHAIRMAN MICHELSON:
Even though I might be able to 5
alleviate this problem in part by readjusting the torque 6
switch even until such time as I can get a better--
7 MR. ROTHBERG:
If the pipe break was not part, is-8 not part of his design basis, either because it wasn't in the 9
original design basis or because of some Commission letter or 10 because he made a new commitment, that design, that pipe break 11 is not part of his scenario now.
'12 CHAIRMAN HICHELSON:
And you are just.saying yo'1
(}
13 ignore that even though--
14 MR. ROTHBERG:
That's right.
15 CHAIRMAN MICHELSON:
Everybody recognizes that l
16 today's~ basis, it clearly, direct water cleanup, clearly the l
17 isolation valves have to isolate failures in the reactor water 18 cleanup system.
That's what they are there for.
19 MR. ROTHBERG:
The block valves which were not 20 originally--
21 CHAIRMAN MICHELSON:
Reactor water cleanup are those 22 valves in there to isolate pipe breaks in the reactor water 23 cleanup system.
24 HR. ROTHBERG:
I believe there is a generic issue on 25 that right now that essentially talks about whether they can n
c-
20
.j 1
or cannot close, Generic Issue 87.
{}
2 CHAIRMAN MICHELSON:
That talks about their 3
operability, not about what they are in there for.
Now some 4
people did not identify the FSARs, that they were putting them 5
there to accommodate full-blown pipe breaks.
The documents
-6 mention pipe breaks in some cases.
Now that we understand 7
that really that's what they are there for, we do not adjust 8
them to that condition.
We still adjust them to a no pipe 9
break condition.
That's the original design basis.
That always brings into it the 11 question you get back into in history lessons--did he just 12 omit it and forget about it, or did he consciously say I'm not 13 going to have this close under pipe breeak conditions?
14 I think that's two different things.
I think if it 15 was just omitted and he didn't have it in the design basis, if 16 he thought about it, and it would be in the--I think now it 17 ought to be part of the design basis, but if he has got, but 18 if his clear intent was not to isolate the thing, I would say 19 he would be imposing a duty--it all sounds very legalistic.
I 20 would certainly hope that the resident inspectors if they find 21 stuff like that, they get it fixed either by enforcing current 22 requirements or going back and getting something done about it 23 at a higher managenent level.
I can't believe people are 24 going to sit around and say legally you don't have to do this.
25 CHAIRMAN MICHELSON:
The letter is very explicit
-~.~n
_n
r.
- - - - - - - - - ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
a 21
([
1 about this point.
It-says we do not'do anything other than 2
accommodate original design basis.
3 MR. MINNERS:
I don't read the letter that way.
The 4
existing approved design basis--
5 MR. ROTHBERG:
It doesn't say original.
6 CHAIRMAN MICHELSON:
I'm sorry--the existing 7
approved--that might have been original main case.
It is, it 8
is still the original.
9 MR. MINNERS:
It the' guy finds a mistake in his 10 design basis and changes it, okay, that's now the existing 11 approved.
12 CHAIRMAN MICHELSON:
And the question is whether
()
13 these changes in scenarios, for instance, are a mistake or 14 what?
15 MR. MINNERS:
The question was the intent of the 16 letter is not to have people go out and re-evaluate the design 17 basis.
We didn't, that's what we didn't want to do.
You 18 know, if the guy finds out his design basis is wrong for 19 whatever reason, I mean the design basis fairy comes along, 20 his design basis was wrong, I mean then I think I can't see 21 how any responsible person cannot fix that, i
22 CHAIRMAN MICHELSON:
If it is reported and l
l 23 accommodated right away.
j l
24 MR. ROTHBERG:
Regardless that it is beyond the
()
25 scope of the generic letter.
l I
___s
__m
.mm.
22
()
1 CHAIRMAN MICHELSON:
That was 2
MR. ROTHBERG:
That's the point of all these 3
answers.
4 CHAIRMAN MICHELSON:
And I recognize it was the'way 5
it was written.
Then I asked well, how are these situations 6
going to be handled?
And I'm getting clarifications which I 7
gather now the only uncertainty is how do you handle these, 8
these changes in design, in scenarios that might have occurred 9
o'rer the last, ten, fifteen, twenty years and now changed the 10 design basis for a component?
These regulations are a little 11 unclear on how you report those.
12 MR. CARROLL:
Let me ask this.
Is this real, Carl?
()
13 I mean I don't believe people didn't, didn't design cleanup 14 isolation valves for a break outside of containment?
15 CHAIRMAN MICHELSON:
It is very real.
The case of 16 HPCI is even more real.
They didn't design for HPCI steam 17 like break, either.
18 MR. CARROLL:
Let's stick to the cleanup line.
You 19 are telling me that there are plants out there that are not 20 designed--
21 CHAIRMAN HICHELSON:
Yes.
22 MR. CARROLL:
Can you tell me which ones?
23 CHAIRMAN HICHELSON:
I don't think of one name.
I 24 will give them to you afterward.
I think Pete knows the same 25 plants I know.
23 1
MR. CARROLL:
How the hell could they have done
{
2 that?
i 3
MR. MINNERS:
Why don't you tell the staff that and 4
we will get the staff a standing order and say fix HPCI design
]
5 basis?
6 CHAIRMAN MICHELSON:
I think you have sent out a 7
couple of letters in the last two, three years, warning people 8
they may not be prepared for some of those older bulletins.
9 MR. CARROLL:
Your point is it ought to be handled 10 some other way then this particular generic letter.
11 Did this issue come up in your workshops?
12 MR. SCARBROUGH:
Not explicitly RWCU.
What we told 13 them, we have scenarios.
They have to meet the scenarios that
(}
14 they have been approved and required to meet.
We are not 15 telling them to go out and create new scenarios, to pose new 16 scenarios.
That would be handled during their analysis.
If 17 we find in review of their analysis they have left off an RWCU 18 red line break scenario, we would go through 51.09 backfit 19 together, look into that, and require they add that.
Then 20 they would have to add that MOV into their program.
21 CHAIRMAN MICHELSON:
MOV is already in their 22 program.
23 MR. SCARBROUGH:
Under the new--
24 CHAIRMAN MICHELSON:
The break condition has to be q%)
25 added.
24
}
1 MR. SCARBROUGH:
Right, new design condition.
2 CHAIRMAN MICHEbSON:
Same way with the HPCI steam 3
line, RCIC steam lines, auxiliary steam lines.
They must be 4
sure they have this as a part of the design base.
If they 5
don't, I understand they have to report it.
I 6
MR. MINNERS:
If this is a--
7 CHAIRMAN MICHELSON:
Is that correct?
They do have 8
to report these deficiencies?
9 MR. SCARBROUGH:
Depends whether they are considered 10 a deficiency.
If their approved licensing basis lists 11 scenarios that they have to meet, and there is something left 12 off that which is not a mistake, it was just apptoved, NRC
()
13 approved it, they are within their legal right to have those 14 scenarios.
15 CHAIRMAN MICHELSON:
These things ever get reported, 16 then you just forget it?
If I never consider that break to 17 begin with, you are saying I don't have to report it now?
18 MR. SCARBROUGH:
If it was one of the ones that were j
19 to be looked at, and it wasn't on that list and NRC approved l
20 it, they have a legal license.
They don't need to go back and 21 add the scenario.
22 CHAIRMAN MICHELSON:
Really don't have to report it.
l l
23 MR. ROTHBERG:
The generic letter doesn't address 24 it.
25 CHAIRMAN HICHELSON:
The generic letter does not
25
(~ }
1 require it.
2 MR. SCARBROUGH:
No.
3 CHAIRMAN MICHELSON:
You have to go back to your 4
approved design basis, make sure everything is all right for 5
it--not that it is a good design basis; just that everything 6
that is required.
7 MR. SCARBROUGH:
We are doing audits and inspections 8
to the generic letter.
As we come across an MOV that has a 9
delta P pis sure of ten pounds and we know that there is 10 normally an RWCU line break outside that, we ask them why is 11 that, and they say well, we don't have that scenario in our 12 design basis, that will start the process for us going through I
()
13 order backfit becauee it obviously needed to be in there.
14 CHAIRMAN MICHELSON:
Really what you are saying is 15 these changes from the original design basis are going to have 16 to go through an entirely separate process.
They aren't even 17 required to be reported now, so therefore, you are unaware of 18 them?
19 MR. SCARBROUGH:
That is correct.
l l
20 CHAIRMAN MICHELSON:
You have to somehow become 21 aware of these, if you think there is a problem and then go 22 through a, something in the backfit regime and put out a 23 letter.
14 MR. SCARBROUGH:
That's right.
I would assume 25 another NRC program would probably look at something like that l
l
26
()
1 for the older plants like the SEP program, something that to 2
bring plants up to current requirement, looking at what they 3
were required to do.
4 CHAIRMAN MICHELSCN:
I think this whole subject is 5
kind of academic.
These are non-recordable items to begin 6
with because they weren't.
I l
7 DR.. SIESS:
These are all covered by requirements?
)
8 MR. SCARBROUGH:
I'm sorry?
l 9
DR. SIESS:
These are all covered by current 10 requirements?
All of these scenarios of pipe breaks?
11 MR. SCARBROUGH:
I would assume they probably are.
12 CHAIRMAN MICHELSON:
Yes.
()
13 MR. SCARBROUGH:
But there is a moving target.
14 Things are happening.
15 DR. SIESS:
Design basis accidents, Chapter 15 16 accidents?
l 17 CHAIRMAN MICHELSON:
No.
A lot of these are not 18 Chapter 15.
19 DR. SIESS:
Where is it covered in the regulation?
20 CHAIRMAN MICHELSON:
It is covered right now under 1
21 the pipe breaks outside of containment kind of letters, 22 written way back; standard review plan now.
23 DR. SIESS:
Pipe breaks outside of containment
()
24 looked at--for the S and P plant, they would not be the only l
23 ones that pipe break outside of containment was not
--_m
27 Q
-1 backfitted?
()
2 CHAIRMAN MICHELSON:
No.
You reported whatever you 3
had to the NRC, and until they came back and said that wasn't 4
. good enough, nothing more happened, though, I know.
Is that 5
right, Pete?
6 MR. WOHLD:
What I remember is that the service 7
things that came out were everyone had to position--some sort 8
of pipe break.
9 CHAIRMAN MICHELSON:
It didn't treat the question of 10 whether the valves were closed.
It doesn't even ask that 11 question.
Just says if the pipe breaks, here is what happens 12 to the environment and so forth, and here is the more effect
(}
13 the equipment needed for safe shutdown.
Part of those was to 14 isolate because these got horrendous in some cases, if they 15 didn't close the isolation valves, HPCI steam line.
That was 16 a given, that you close the valves, because they were 17 redundant.
18 MR. WOHLD:
I know in the non-safety-related system 19 pipe breaks there was no single failure required, but that 20 raises a question on the valves that are included in the 21 program.
22 DR. SIESS:
Before we get off the scenario for a 23 plant that to which the pipe break outside containment was 24 applicable, somebody just said it was assumed that it would O.
25 isolate.
1
28
([
1 Now if the valves are not capable of isolating, I 2
don't see that that's a deficiency in the scenario.
That's a 3
deficiency in the valve.
4 MR. SCARBROUGH:
That is correct.
They will have to 5
fix that valve.
6 DR. SIESS:
So any plant to which pipe break outside 7
containment applied, valve that can't isolate is a valve 8
deficiency.and it is reportable?
9 MR. ROTHBERG:
That's right.
10 MR. SCARBROUGH:
It is inoperable.
It is inoperable 11 valves.
12 DR. SIESS:
Which plants would this not apply to?
()
13 Was there one time at which--
14 MR. ROTHBERG:
Those that did not have the 15 commitment imposed upon them to isolate pipe breaks outside of 16 containment.
17 DR. SIESS:
Which ones were those pipe break outside 18 of containment rule or whatever it was?
Was it a rule?
19 CHAIRMAN MICHELSON:
Just a letter.
20 DR. SIESS:
A letter; to which plants did it not 21 apply?
Do we know?
22 MR. SULLIVAN:
The letter is the O' Leary and 23 Giambusso letters.
I don't know which ones are first.
I 24 believe it applied to all the plants that were in the 25 licensing stream and licensed at that point in time.
nn
29
])
1 What I'm personally not clear on is whether or not 2
people made the connection between the requirement postulated 3
break and protect environmentally the equipment around it, and 4
the, the other half of it which would be to be able to isolate 5
the break and mitigate consequences.
That wasn't in the 6
letter.
7 DR. SIESS:
But the environment would certainly be 8
different if they couldn't isolate, so they had to make some i
9 assumption about isolation or they couldn't get idea, any idea 10 what the environment was.
You can't isolate.
You have got a 11 flood.
You have got steam all over the place.
Must have made l
l i
12 some assumption there.
If that assumption was incorrect, it
()
13 is not an error in scenario.
It is just sa error.
l 14 MR. CARROLL:
You have got water or steam all over 15 the place just in the closing time of the valve.
16 CHAIRMAN MICHELSON:
That's right.
You didn't close 17 it.
Then that was not analyzed.
What would happen if the 18 valves failed to close?
19 MR. SCARBROUGH:
If they are relying on MOV to close 20 in emergency procedures, to mitigate some accident, then it is I
21 in the program.
I don't see how they could properly move it 22 out.
It has to be--
23 DR. SIESS:
We are talking about the plants that i
I 24 that did not apply to?
What, 40, 50 plants?
l
{}
25 CHAIRMAN MICHELSON:
The later plants; in other l
---_~~_m-~
~mmmmmmmmne amann m A a agna
30. 1 words, what happened to plants coming on after the Giambusso 2 and O' Leary letter? Didn't' check I assume the standard review 3 plan, picked up'the requirements. 4 DR. SIESS: What about the earlier plants to which 5 it didn't apply? l 6 CHAIRMAN MICHELSON: I don't think there were any j 7 earlier plants that-- 8 DR. SIESS: 'That's what I have been trying to find 9 out. I have gotten four answers, none of which I understand. 10 Which plants would the generic letter or whatever it 11 was, outside containment not have applied to? 12 MR. SULLIVAN: There weren't any it didn't apply to. () 13 The basic pipe break criteria applied to all the plants. I 14 think the problem is that the utilities were required by that 15 guidance to postulate breaks and show that the plant could 16 safely shut down, and I don't know how thoroughly this was 17 evaluated in terms of pipe break isolation. 18 CHAIRMAN MICHELSON: They were not cautioned as I 19 recall to determine that you keep the--you are going to count j l 20 out isolation--was capable of isolating. That was not a part 21 of the letter. 22 MR. ROTHBERG: But that letter, that came--it 23 applied to-- 24 DR. SIESS: They had to postulate isolation and if O 25 the valve is not capable of doing it, it is an equipment
m _-_-,,- - -. r y, ,.f, 7c. AM 31 g k: I 1: deficiency. ty y Vl. "- 2 ' CHAIRMAN MICHELSON: I fully agree with you. t k 3 DR..SIESS: And reportable. 4 MR. SULLIVAN: But the letters didn't address p 5 ' isolation. 6 DR. SIESS: Told them to analyze it, didn't it? 4 7 MR. SULLIVAN: It said to, it s aid analyze the 8 consequence and make sure that the plant could be safely shut 9' down. ) 10 DR. SIESS: They have to make some assumption about ~ 11 isolation to analyze the consequences. -12 MR. MINNERS: Not--it didn't explicitly address n(_) 13 " isolation. By implication, obviously you are going to have to -I 14 look at isolation. 15 MR. CARROLL: No. I remember responding to that 16 -letter, and if you had a water line break where you could say 17 the water would go some place and not impact your ability to, 18 to shut down, I don't think, I don't remember, you know, 19 asking the question well, will the isolation valve close? I 20 don't care. 21 MR. MINNERS: I am agreeing with you. I think l 22 that's right. j l 23 CHAIRMAH MICHELSON: They didn't care, but some /~"i 24 cases-- i LJ i i 25 MR. CARROLL: If I had asked the question I probably i _m_ms m m - m_m l
y BJ ' jy, 3.4r 32 ,(~N 1 .would'have, would.have found that yes, the valve closes, but I Q '2', didn't even ask the question because I didn't' care. 3 DR. SIESS: What about a steam line break? 4 MR. CARROLL: I really can't remember, but I don't 5 think I cared there,- either, because I had a way for the steam 6 to get.out'of the area. -There wasn't any equipment in the 7 area that I was concerned about. 8 CHAIRMAN MICHELSON: You didn't look at a boiler? 9. MR. CARROLL: Sure did. 10-CHAIRMAN MICHELSON: You did? You sure worry about 11 the HPCI steam line break. 12 MR. CARROLL: We didn't have a HPCI steam line in [J 13 Humboldt. 14 CHAIRMAN MICHELSON: No. That's true. 15 MR. ROTHBERG: The last question-- 16~ MR. MINNERS: We supposedly have a big thing going 17 on intersystem LOCAs. I don't know whether this is explicitly 18 included in that. That question has been looked at. 19 CHAIRMAN MICHELSON: I think mainly I understand 20 where you are coming from, and what the--I believe I 21 understand what your requirements are. I hope that the letter 22 and the utilities understand as well, and you say you are 23 going to take care of that? i. 24 MR. CARROLL: If Carl and Pete know of a lot of I 25 instances where this is the case, it sounds like we ought to
33 [l 1 'do a generic issue. \\/ 2 MR. MINNERS: Does me, too. 3 MR. CARROLL: But not this one. 4 MR. ROTHBERG: That's right. 5 CHAIRMAN MICHELSON: We already have--it is covered 6 by 89-10. 7 MR. ROTHBERG: No. That's the point. 8 CHAIRMAN MICHELSON: Well, they will find out when 9 they do their 89-10 analysis. My concern was-- 10 MR. ROTHBERG: I don't think that they will. 11 CHAIRMAN MICHELSON: You don't think they will go 12 back and check the ability to isolate? () 13 MR. ROTHBERG: No, I don't think they will check the 14 scenario itself, other than the fact that if they have a 15 letter in their pockets, to, to check that, t' hen they will do 16 it, but if it is not, if that letter doesn't apply to them, 17 then they won't do it. 18 CHAIRMAN MICHELSON: The letter was, the intent of 19 the words that we put in our letter was trying to mean that, 20 you know, and based on today's understanding, would they check 21 the scenario? 22 MR. ROTHBERG: The answer is that they won't until l 23 they have a commitment somewhere or the Commission has imposed r' 24 upon them somewhere the responsibility for checking that. (>} 25 That's the answer to that.
34 1 CHAIR!"AN MIChELSON: We said here licensee be 2 reminded to review the design basir governing the selection of 3 each MOV on the viewpoint of completeness and adequacy in 4 light of current regulatory requirements, make sure their i 1 5 scenarios are complete. 6 MR. ROTHBERG: The current regulatory requirements L 7 that we were talking about were their commitments, or their 8 license, or a letter that was sent to them from the NRC, 9 something hard like that. If it was just well, gee, everybody 10 else is doing it, they don't have any responsibility to do 11 that. 12 DR. SIESS: Is that the sort of vulnerability that 13 we might expect somebody to find in the individual plant ggg 14 examination? 15 MR. MINNERS: No. 16 DR. SIESS: Why not? 17 MR. KIESSEL: I'm Dick Kiessel. I'm with NRR. My 18 understanding of the IPE is you are taking extreme accidents. 19 You are not going back to your original scenario. 20 DR. SIESS: This would not be extreme accident? 21 MR. MINNERS: These could lead to core melts. 22 MR. KIESSEL: But the, my understanding of the IPE 23 program is that you are looking at paths leading to core melt 24 that are outside of the original, outside of the existing 25 design basis. HERITAGE REPORTING CORPORATION -- (202)244-8817
35 L() 1 DR. SIESS: I don't believe that's true, not when 2 you tell-people to use a PRA. You forget about the design 3 basis. You now analyze the plant for anything you can think 4 about that can happen. 5 MR. KIESSEL: That's beyond it. 6 DR. SIESS: There two points to my question.. Is 7 this a vulnerability in the sense we use it in the IPE, that 8 is, something that could contribute to core melt probability? 9 And secondly is a vulnerability something that would 10 be found whether or not it is looked for presumably? 11 MR. MINNERS: If you can't isolate your cleanup 12 line, yes, you have got a problem. () 13 MR. CARROLL: Comes off the bottom of the vessel. 14 You don't get enough ECCS in there. 15 CHAIRMAN MICHELSON: Into areas-- 16 MR. MINNERS: You are hanging on the probability of 17 breaking that line or not breaking that line, detecting the l 18 leak, all those kind of things. 19 MR. CARROLL: Which I think is pretty low.,, 20 CHAIRMAN MICHELSON: Some of these are non-seismic 21 even. Not all water cleanups are even seismically designed. 1 22 There is a lot of other components, resin beds and heat 23 exchangers, a whole bunch of things in there besides just 24 pipes. ] 25 MR. MINNERS: But you know, IPE is not a panacea. l
1 36 l {}L 1 It still depends on the guy being smart enough to pick up the 2 scenarios. They don't just pop. There is no systematic l l 3 method to do this in PRA. The more the guy knows, the better l -4 the PRA, but there is no 100 percent or even probably 90 5 percent assurance that you are going to pick up all this 6 stuff. It is smart to pick up the stuff. 7 CHAIRMAN MICHELSON: If they would pick it up every 8 time it, if they looked at each valve reliability number and 9 said well now, what is this valve going to be doing? What 10 would its failure probability be? If they did it on a 11 valve-by-valve basis, they would pick it up every time. 12 MR. MINNERS: Not true; I don't agree with you, ( ) 13 Carl. I don't agree with that because the guy would look at 14 the valve and there is no assurance that there is a scenario 15 that he might miss. 16 MR. CARROLL: When he would look at it. 17 MR. MINNERS: He would have a better assurance, but 18 that still doesn't guarantee that he is going to get all 19 scenarios that valve has to take care of. 20 CHAIRMAN MICHELSON: That won't assure the 21 scenarios. 22 MR. MINNERS: I know I couldn't assure to myself if 23 I were doing that that I caught all the scenarios. 24 DR. SIESS: That's not the way he does it. He 25 doesn't look at each valve. He looks at the initiators. That
.. t 37 ('g 1 valve is in the fault tree, event tree for that initiator. He - () 2 then looks at it and assigns a reliability to it. 3 MR. MINNERS: You have got to look at more than 4 initiators, because there may be some flip-flops as you go 5 down the event tree, and there may be one branch where that 6 valve is important, and he may not follow that branch. 7 DR. SIESS: He may assign the wrong reliability to 8 it, too. 9 MR. MINNERS: That's possible. 10 DR. SIESS: Unreliability may be close to one rather 11 than close to zero. 12 MR. MINNERS: The other thing they do a lot of times (~ } 13 is they credit their recovery actions, which makes the 14 scenario go away, and when you really look at the recovery 15 actions, you know, they are highly suspect. i 16 CHAIRMAN MICHELSON: I think we will have to move l 17 on. I believe unless the Committee has some other questions 18 that we understand at least what the position is, and I for I 19 one find it interesting with the question of scenarios and how i L 20 you handle them under backfit. I l l 21 DR. SIESS: The design basis versus scenario, I 22 never heard the question. I'm still not quite sure whether 23 the design basis is a scenario or an event. 24 MR. ROTHBERG: Well, okay. An event, we say events 25 and conditions are scenarios. That's our definition. That's I 1
,+ w, i h 38. y 1 :- !my definition'. [2L DR. SIESS: Say that'again. ') 3; MR..ROTHBERG: An event is, a scenario is an.eventi j u 4: or condition'of the plan.t. l l 5 DR. SMITH: Scenario is~a' series-'of' events, i ) 6-MR. ROTHBERG: 'All right. 7 DR. SIESS: The design bases are usually stated as' 8 an initiating' event. 9 MR. MINNERS: We always put loss.of off-site' power, 1 10' with an earthquake-- l 11 DR. SIESS: You don't decide how'the' power is lost. 12 Lost, if you say loss'of off-site power, you say pipe preak 1 (][ 11 3 outside containment. Right? 14-MR. MINNERS: Part of that scenario would be pipe l 15 breaks out of containment with loss of off-site power. The 16-diesel' generators fail and'all the operators get gassed.- I i 17 mean that's the kind, right? 18-DR. SIESS: Is that a design basis event? 19 MR. MINNERS: Yes. It is loss of off-site power, 20 certain amount of operator action. Those are typical limits 21 on design basis accidents that I can think of off the top of 22 my head, in addition to the initiating event. 23 DR. SIESS: But your definition of design basis Ow.' 24 accident-- 25 MR. MINNERS: And single failure--sorry.
39 ( ). 1 DR. SIESS: Design basis events with a single 2 failure would be one of those motor-operated valves? 3 MR. MINNERS: -Yes. 4 DR. SIESS: Not two of tilem? 5 MR. MINNERS: Not two of them. 6 DR. SIESS: So you don't get quantitative on your 7 reliability obviously. Single failure. approach is-- 8 MR. MINNERS: Deterministic, mechanistic. 9 CHAIRMAN MICHELSON: Failing to recognize if one 10 can't close because of flow, it probably can't close, either. 11 MR. ROTHBERG: That's right. 12 DR. SIESS: Which you don't get with single failure. () 13 CHAIRMAN MICHELSON: PRAs take it up with using that 14 flow as a condition for reliability. 15 MR. MINNERS: let's take the HPCI line, got two 16 valves. Okay. With one valve if it doesn't close, it is 17 probably going to partially close, then may enable the second 18 valve even though it is not properly specified for its 19 service, it might close. 20 CHAIRMAN MICHELSON: It would be more difficult to 21 close the second because of the throttling effect of the 22 first. That is another question on this whole testing. 23 What happens if you have got two valves in a series {} 24 and you start putting these terrific turbulences into the 25 input of the next valve? Will it close under those 1 1 1 I
t 40 <^T-1 conditions? i 2 MR. CARROLL; I suspect what happens-- 3 MR. MINNERS: Realize you are going to expand the 4 test series to have moved them to Germany. Be our guest! 5 MR. CARROLL: I would suspect what happens in your 6 two favorite scenarios, Carl, the cleanup line break and the 7 HPCI steam line break, in the PRA world, is that the 8 initiating event, the break of the line, ends up at such a low 9 probability that you don't give a damn or you don't even look 10 at MOV. l 11 CHAIRMAN MICHELSON: I checked into that. 12 Surprisingly enough, they are using about ten to the minus 4 13 for that first break pipe, and I said well, the probability of {} 14 failure to isolate is one, because the valve won't work, you 15 are talking about a real big event, ten to the minus 4. 16 That's what they are using. 17 Now I don't know that ten to the minus 4 is a good 18 number for these pipes, but that's what they are using. Keep 19 asking once in a while. He can ask the next two days also 20 what they are using for pipe breaks and find out. We are 21 going to talk about it tomorrow. It is not ten to the minus 7 22 kind of stuff we are talking about if you believe the PRA 23 people, and I do, and therefore I am using ten to the minus 4 24 and probability of failure of one. 25 DR. SIESS: Why do you believe them? i
41 [ 1 -CHAIRMAN.MICHELSON: And therefore it is a big 2 contributor. The'last sentence of my memo asks the staff a 3 question--have you actually or are you actually aware of any 4 plants in which HPCI, RCIC, or reactor water cleanup valves 5 have not been designed for pipe break flow? That was the last 6 sentence in the letter. 7 MR. MINNERS: The two plants I reviewed, they were 8 design, the design basis in the FSAR said they were supposed 9 to isolate. What the exact-- 10 CHAIRMAN MICHELSON: Isolate pipe break. 11 MR. MINNERS: Yes. What the purchase specification 12 said, I have no idea. -( ) 13 CHAIRMAN MICHELSON: To your knowledge, though, you 14 are not aware of any plants out there in which they have not 15 purchased a valve to isolate for the full conditions of the 16 pipe break? 17 MR. KIESSEL: Dick Kiessel with the Generic 18 Communication Branch--the way our contractor did the review on 19 85-03 was he grouped the plants by basic similarities. In 20 other words, all the BWR ones, BWR twos, the Westinghouse 21 twos, in that way. 22 Once he had grouped them, he then determined similar 23 valves in the respective systems no matter how they were 24 called or how the exact piping layout went. The next step was 25 to compare the design deficiency pressures so that it is my i
l' l: 42 tr~y -1 belief that he would have caught any outlying plants that had j k/. j 2 not had pipe break valves in the HPCI steam line or in the l 3 RCIC steam line, in answer to your question, so that-- ) 1 4 CHAIRMAN MICHELSON: Normally you get a valve l 5 coefficient. You don't get a' delta P number. l 6 MR. KIESSEL: The licensees were required to submit l 7 their design delta Ps. 8 CHAIRMAN MICHELSON: Okay. 9 MR. KIESSEL: And he compared them across the plants 10 within the same grouping of plants, and so therefore, any--and 11 he flagged outliers, and we specifically went back to them and 12 asked them why their pressures were so different from their {} 13 sister plants or from similar plants of the same design 14 vintage, and in those cases, the licensees came back and gave 15 us revised differential pressure numbers. 16 You also had-- 17 CHAIRMAN MICHELSON: Did they go back and revise 18 their valve setting? 19 MR. KIESSEL: They would have had to have gone back, 20 verified the valve would function at that, under those 21 pressure conditions. 22 Similarly, you also had a BWR Owners Group and 23 Westinghouse Owners Group that made formal, which we have 24 formal submittals from, and we matched each licensee against 25 those submittals so that even if the plant was not a member of
I e l 43 () 1 the Owners Group, we were comparing it against the generic 2 work that had been done by the owners groups. I 3 Informally, and one that we did not get our hands 4 on, my understanding was that there was a B&W Owners Group j l' 5. that worked. We did, but we did not use the B&W Owners Group 6 analysis because we never saw it. However, we did again do j 7 this comparison across the, against, across the B&W--I'm l 8 sorry--across the B&W plants to make sure that they were all I 9 coming in with approximately the same design differential I 1 10 pressures for the respective valves in their systems. l 11 CHAIRMAN MICHELSON: You did an end calculation to l 12 see if that was a ree-nable number to begin with, could come () 13 in wrong, of course. I assume.you verified that it was a 14 number that you--uniformly reported it was the right number to 15 begin with. 16 MR. KIESSEL: We did not do independent analysis of 17 every delta P. 18 CHAIRMAN MICHELSON: Doesn't take much to decernine I 19 roughly what that should be. } 20 MR. KIESSEL: Again, Carl, we did not do independent 21 verification of every delta P that was submitted. 22 CHAIRMAN MICHELSON: They are all the same. The 23 outlier you requested, you didn't question the bulk. () 24 MR. KIESSEL: That is correct. 25 CHAIRMAN MICHELSON: The bulk somehow you felt was s .m _.m -m m _ -
44 1 -() 1 correct? 2 MR. KIESSEL: That is correct.- 3 CHAIRMAN MICHELSON: You didn't verify that by an 'l 4 independent calculation? i 5 MR. KIESSEL: That is correct. i 6 CHAIRMAN MICHELSON: Any other questions on this 7 issue? 8 MR. WOHLD: I have got a question. On the, in the 9 workshop, the valves that were stated as included in the 10 generic letter were those valves in safety-related systems. 11 There are a number of valves that are required for accident 12 mitigation that are not in safety-related systems, and I () 13 wonder how that, if there is a problem with the-- 14 MR. ROTHBERG: Are they safety-related valves? Are 15 they classified as safety-related? 16 MR. WOHLD: According to the generic letter here, 17 safety-related is anything that includes the events in Note 2, 18 bottom of the first page here. 19 MR. ROTHBERG: That's right. 20 MR. WOHLD: Okay. 21 MR. ROTHBERG: It does not, that doesn't refer--the 22 only place we refer to safety-related valves and 23 service-related systems is position changeable valves. We l 24 also refer to safety-related valves. l 25 MR. WOHLD: So the operability of MOVs is--
45 f) 1-MR. ROTHBERG: Whether they are safety-related or l 2 not, plus the position changeable plus any other valves that 3 the licensee considers are important to safety. 4 MR. WOHLD: Okay. I didn't understand that. At the 5 workshop I thought you were saying that this generic' letter 6 just covers valves in safety-related systems. 7 MR. ROTHBERG: I think the context of that 8 discussion was having to do with position changeable at the 9 time. You said that at the session that I went to. 10 MR. SCARBROUGH: Owen, talking safety-related 11 systems, this generic letter is intended to make sure that all 12 the valves in the safety-related systems that are used in one () 13 manner or another are operable, and can operate under design .14 basis conditions. 15 We aren't stretching the valves in plant. We 16 recommend that they look in an area to try to determine 17 whether or not there is valves out there that may need to be 18 included in this program, but we are only within 19 safety-related systems, systems that are service-related. 20 MR. MINNERS: Have you got some examples? 21 MR. WOHLD: Well, when I think back on the pipe 22 break outside containment evaluations that I have done in the 1 23 past, a lot of those pipe breaks involved scenarios involving () 24 non-safety-related equipment. 25 CHAIRMAN MICHELSON: Auxiliary boiler is a good
46 (~Y 1 example. Steam line going into vital area on auxiliary Ls] 2 boiler, better isolate it if you break the steam line on that. 3-MR. ROTHBERG: Don't they have to go through 4 containment isolation valve? S' CHAIRMAN MICHELSON: No. It is outside. It is 6 important to safety, but not safety-related. 7 MR. ROTHBERG: The way we recorded the letter was we 8 asked the licensees. We said in addition, that there may be 9 valves that the licensee considers important to safety, and we l 10 left it up to his discretion, 11 MR. MINNERS: But a valve on auxiliary boiler is not 12 used for mitigating an accident or safe shutdown. () 13 CHAIRMAN MICHELSON: It is only there in case the 14 auxiliary steam line breaks in a vital area. 15 MR. MINNERS: That could initiate an accident, but 16 that's not our concern. 17 CHAIRMAN MICHELSON: I know it is this whole 18 question of what scenarios are we really talking about, and 19 whether or not you change them or add them if it is a backfit 20 or how you handle it. Pipe break outside of containment has 21 been a design basis, but not a Chapter 15 design basis. 22 DR. SIESS: You know what is bothering ue I guess is 23 the distinction between design basis accidents which are a i 24 series of postulated events stated in such a way,that they can 25 be used to design the systems specified in the components, et
47 ( )l l' ' cetera, and the kind of scenarios that we are discussing which 2 are really'the kinds of things you do for event trees and 3 fault trees in a PRA to look at severe accidents. Single 4 ' failure criterion is the basis for design.- There is no other 5 usefulness, no other significance, but it is something you can 6 state that a designer can decide he needs this or that and so 1 7 forth. 8 MR. MINNERS: It is a simple-minded design criteria. 9 DR. SIESS: We are looking at reliability problems 10 with valves in terms of scenarios that could lead to, 11 presumably to sovere accidents, so we don't need to be too 12 much concerned with them, not necessarily core melts. We () 13 release stuff outside of containment, also a severe accident, 14 but we are looking at these scenarios and we look at that 15 reliability, but we keep trying to tie it back to design basis 16 accidents or design basis specifications, which it is apples 17 and oranges. They don't relate. 18 MR. MINNERS: Well, as I remember it, correct me if 19 I'm wrong, but I think there was some thoughts in this letter, 20 is to have licensees go out and search for the 21 safety-significant valves, if I can use that, things that you 22 would find from PRAs as being the important valve, and have 23 them fix those. (} 24 I think it is pretty much decided that that was an 25 effort that we didn't want people to spend all their money on.
i 48 f(~} 1 We were kind of going to do it in their IPE anyway, and so we %_/ 2 said, we tried to use the precedents that we had in design 3 basis accidents, even knew what we were talking about, so part 4 of this is the practicality of communicating with the 5 licensee. 6 If we try to develop some new word like 7 safety-significant valve, we just get all tied up in knots and 8 you would have a terrible time explaining to people what that 9 meant. It would take you years to get it defined, so we kind 10 of have to 2ive with the system that we have embedded in, and 11 I don't think this is that bad. I don't think just the design 12 basis events is all that bad. I think it is very well--when (} 13 you do PRAs, you find out that the residual risk outside of 14 the design basis is not all that big, not that terrible. 15 CHAIRMAN MICHELSON: The PRAs didn't account for the 16 operability of components, generic reliability numbers in the 17 PRAs, and so they don't pop out. Clearly HPCI steam line 18 break never comes out of PRA. One of two valves gets closed. 19 MR. MINNERS: That is not the fault of PRA. That's 20 the fault that we have, we have in implementing the design 21 basis. 22 DR. SIESS: Somebody using the wrong reliability 23 figures for valves. 24 MR. MINNERS: But the thing is not to change the 25 PRA. The change is to go back to the plant and get the
V; 49 IL reliability'of-th'e valves when-you want them. 2' ' CHAIRMAN MICHELSON: Under the conditions that you 3. are trying to mitigate at the t'ime. / 4 MR. MINNERS: That is correct. 5 MR..ROTHBERG: .I can recall' discussion with NUREG 6 5 0'. 4 0, one of the things you brought up was-the fact if he- .7 included the reliability. numbers that we were,'we' thought were' 8 realistic rather than what we had been doing previously.- 9 There would be new scenarios, new fault trees built in.' He. 10 didn't consider those, and that was conservative, but you are 11 right. That would come up. 12 MR. WOHLD: I just want to mention, reiterate one -( ) 13 things. As far as I know, it is perfectly acceptable to do a 14 system review and determine what the maximum condition.of 15 valve seating and testing against that condition as opposed to 16 doing a massive paper review of the past document 17 reconstructing the design basis, for instance? 18 MR. SCARBROUGH: Yes. They can do that. They can 19 look at what the conditions are on that valve, worst case for 20 all their approved licensing design bases, and determine where 21 that valve should be set, how it should be set. 22 MR. WOHLD: Right. They don't have to do a paper 23 review in order to get started doing this? I () 24 MR. SCARBROUGH: No, they don't. One other 25 thing--if they find a safety-related valve that is classified
50 {~} 1 as safety-related, somewhere outside of safety-related system, 2 they would have to include that, too. If it is classified as 3 safety-related, performs a safety-related function, then it is 4 in the program, because it is all safety-related valves, but I 5 can't see that being the situation where they classified a 6 valve as safety-related in the balance of plant, but if it is 7 there, they have to include that valve in the program. 8 MR. WOHLD: I can't think of a good MOV example. 9 The air-operated valve example I always think of as the 10 feedwater reg valve in a PWR, and it is required to close to 11 back up the motor-operated isolation valve for a steam or feed j l 12 line break. That's totally non-safety-related valve, and yet () 13 it is taken credit for in the accident scenario. 14 There are not many examples, but tha one I tried to 15 bring up in the workshop was water box break, and I know that 16 that has turned out in the past to cause flooding of 17 safety-related equipment and doors and what not, and perhaps 18 you could take credit for a motor-operated isolation valve in 19 the cire water system or something, isolated break of that 20 nature which would make it safety-related according to this, 21 according to the generic letter. 22 MR. SCARBROUGH: It has to be safety-related 23 according to the definition of design basis event of the 24 plant, through that categorization. Then if it does meet 25 those criteria, then it would be a safety-related valve. l c _ ~ - - - - - - - - - - - - - - - - -. - - - - - - _ _ _ - a n n n na n
51 {) 1 MR. WOHLD: I think you have got a nice definition 2 here, but there are valves, by your definition, in a generic 3 letter, it includes valves outside of quote safety-related 4 systems. 5 MR..ROTHBERG: It might. It could. 6 MR. WOHLD: It could. 7 MR. ROTHBERG: It is systems and components, if you 8 read the footnote. Also add that note, that sentence at the 9 bottom of the page 3, paragraph just before the bottom, 10- "Although this program should address safety-related MOVs and 11 position changeable MOVs as a minimum, NRC envisions that it 12 is part of a good ma$ntenance program. Other.MOVs and balance '(]). 13 of plant should be considered for inclusion in the program 14 commensurate with the licensee's assessment of their 15 .importance to safety." 16 We got a lot of grief from several licensees about 17 that statement, and what we said was we will have to make an 18 evaluation. 19 CHAIRMAN MICHELSON: Is it correct to assume that 20 any valves that would require for mitigation of answering of 21 the Giambusso and O' Leary letters are safety-related somehow? 22 Is that the kind of valves? 23 MR. SULLIVAN: I would--say that again. l 24 CHAIRMAN MICHELSON: In the Giambusso and O' Leary 25 letter, in response to analysis, they did an analysis and then I l l
52 ([ 1 responded that we could accept these kinds of pipe breaks, but 2 in the process of doing the analysis, you found certain valves 3 you had to get closed in order to accept-the consequence of 4 that, the valve that you have to get closed now is considered 5 to be these other safety-significant valves and would be a f 6 part of this 89-10. 7 MR. SULLIVAN: I would say that if their basis'for 8 safe shutdown assumed isolation, those valves would be picked 9 up by Generic Letter 89-10. 10 CHAIRMAN MICHELSON: That would encompass quite a { 11 few of them then. 12 MR. SULLIVAN: If that's how they assumed that they () 13 could get the safe shutdown. 14 CHAIRMAN MICHELSON: Yes, if they had to take credit 15 for them. They wouldn't be called safety-related, but they 16 are in some other category. 17 MR. SULLIVAN: Yes. 18 CHAIRMAN MICHELSON: Not on the safety-related list 19 necessarily. These are out in the water systems, you know, 20 non-qualified water systems, and tanks that had to be isolated 21 before you flood the whole building out. You are allowed to ) 22 use safety-related equipment. It was not just service-related 23 equipment. You could use anything you had that you could () 24 reasonably take credit for, and now I am asking to, I have to 25 now go back under 89-10 and verify the operability of those j
53 ('Y 1 particular components that I assumed credit for? I think it u) I l-2 is unclear from the letter, so you can see my confusion is 3 perhaps not unique. 4 MR. SCARBROUGH: We will take a look at that and try 5 to clarify. 6 CHAIRMAN MICHELSON: We need to get on to the next 7 agenda item, which is the MOV fe.ilure to close against 8 differential pressure, and Mr. Scarbrough? 9 MR. SCARBROUGH: Correct. 10 CHAIRMAN MICHELSON: We changed to so many other 11 faces, I wasn't sure if we changed that one. 12 MR. SCARBROUGH: I guess there are two events that .() 13 you were curious about and what our response to them were. 14 Both of them were failures of motor-operated valves to close 15 under a high differential pressure condition. 16 Catawba event had a long history that started back 17 in March of 1988 and began with one failure of an auxiliary 18 feedwater discharge motor-operated valve that flows through 19 some various testing that proceeded over the entire year. 20 In March of '89, they did some further testing, and 21 they determined that the valve factors that, for several MOVs, 22 were two and a half times greater than what the manufacturer 23 had provided to them, and they are now performing further 24 analysis to determine if that situation exists for other l 25 valves that are similar to this--the Borg-Warner Model 6J219 l 1 l l )
f 54-(^]' 1 valves. \\_ 2 In response to that event, there has been an 3 information notice that has been issued very recently, 8961, 4 August 30th, that was prepared by Steve Tinnon down in Region i 5 2. He is the expert on this, and so if we would like to know 6 more about the details of the events at Catawba, he would be 7 the one to talk to. 8 Millstone was recently. They had inadvertent safety 9 injection of high pressure core cooling. When they went to 10 shut the MOV, it would not shut. They had to shut it 11 manually. 12 Earl Brown talked about that back in June when he () 13 met with you. Two of the actions that we in NRR have 14 specifically performed to, in response to this, the Catawba 15 event--first was a letter issued on August 14th of this year 16 from TAPMAR to NUMARC. That letter--I will give you a 17 copy--requested that NUMARC contact EPRI aird request that they 18 investigate the design qualification of motor-operated valves 19 and specifically cited to the LERs these two events as 20 indicating prompt action in this area. 21 We haven't heard back from NUMARC yet on this, but 22 we planned to contact them in the very near future and find 23 out what their response will be to the letter. 24 We also in that letter indicated that the testing {~) 25 that EPRI could perform would be beneficial in implementing
i'I 55 ])I 1 Generic Letter 89-10. l 2 The'second item that we performed in our NRR was 3 that the workshops where we indicated to the licensees that in 4 response to this event, particularly the application of test 5 data from prototype MOVs should be justified even if you 6 assume those valves were identical, and Catawba assumed they 7 were'the same model numbers for those Borg-Warner valves, and 8 they had vastly different valve friction factors, and so we 9 have indicated that concern at our workshops and requested 10 that they, the licensees, justify the use of that data from 11 one MOV to another. 12 MR. CARROLL: Why was the friction factor vastly () 13 different for the valves? 14 MR. SCARBROUGH: They don't know. They are still 15 looking at that. I think they were quite surprised as to why 16 they were different. It seemed as though the LER that was 17 issued earlier this year walked through a series of tests they 18 performed at their steam plant to try to determine why these 19 valves weren't operating under high DP conditions, and it 20 wasn't until March of this year that they really did a 1 21 signature or trace analysis to determine that the valves were 22 vastly different. I think they were baffled by it. 23 I don't--everyone had always assumed these valves 24 would have valve friction factors that are at least similar, (} 25 and that does not seem to be the case, so this is still an -m_
I 56 f(} 1 ongoing issue. Duke is still working in this area. 2 MR. WOHLD: What was his name? 3 MR. SCARBROUGH: Tingent, T-I-N-G-E-N-T. 4 MR. CARROLL: You don't know whether the valves were i 5 . disassembled and inspected? ) i 6 MR. SCARBROUGH: I don't know. That was basically l 7 all I had on this event. We are well aware of it, and we are 8 'very concerned about it, and we have indicated that concern to l 1 l 9 the licensees in our public workshops on 89-10. 10 If you don't have any further questions, I will l 11 centinue on with the review of the motor-operated valve action i 12 plan. (') 13 CHAIRMAN MICHELSON: Proceed. i 14 MR. SCARBROUGH: Thank you. 15 MR. WOHLD: Can I ask a follow-up question on i 16 applying the prototype data to in situ valves? 17 I--the gist of the workshop seemed to indicate that 18 if you had enough margin, that you could, at some point you 19 could say that the valve was good enough you didn't have to do j 20 a DP test. Would that be true? 21 MR. SCARBROUGH: Under a case-by-case situation, if 22 they were able to show that they had a grossly oversized 23 actuator, and they could justify it without performing the [} 24 high DP test, they would still need to do the low DP test to 20 make sure that the valve was set up properly and plus they
t 57 s.t ( ) 1 could do their continuing surveillance they need to do every 2 five years. They still needed to do that, but if they could 3. justify it analytically that.they have a grossly oversized 4 valve, you know, we would be receptive to that presentation, 5 but it would be on a case-by-case basis. 6 MR. WOHLD: Grossly oversize, you would say we could 7 handle friction factor of one and would back off? 8 MR. SCARBROUGH: I think that's what we are looking 9 at. The friction factors range from.38 to.74, and we have, 10 I have seen one question that came in on Generic 89-10 which e 11 said what if we assume a conservative friction factor of.6? 12 Is that okay? Can we now not do any testing. We have seen () 13 that.6 is not conservative anymore. You are going to have to 14 go up toward one. 15 MR. WOHLD: Thank you. l 16 MR. SCARBROUGH: Just to give you a little bit of 17 background on the action plans, I was assigned to review the i 18 past and ( trrent MOV problems, the regulatory requirements in 19 this area, the NRC and industry activities that address the ] 20 motor-operated valve problems, and determine where additional 21 effort was needed, and then I was asked to develop plans that 22 coordinate the NRC activities that are, will ensure valve 23 problems are addressed to provide for the assistance to the () 24 industry in its efforts to resolve valve issues. 25 In performing this task, I discussed NRC activities
58 (]{ l' with= personnel from AEOD, Research, and other branches within 2 NRR, provided drafts of the plans to those individuals, and 3 incorporated many of their comments. The plans, in general, 4 there is one from motor-operated valves and one for check 5 valves, provide for an organized approach to resolve concerns-6 for valve performance. 7 The Mechanical Engineering Branch will coordinate 8 NRC activities but will not duplicate the work of other 9 groups, so for example, the intersystem LOCA problem will l 10 continue to be led by the Risk Applications Branch and 11 Research will continue to prepare regulations and regulatory i 12 guidance. We will work for them on the preparation of those j i () 13 documents, and AEOD will continue to review operating 14 experience. l l 1 15 Why don't I go ahead and give you some background on l l l 16 what this plan provides? 17 (Slide) i l 18 MR. SCARBROUGH: As background, the plans talk about 19 the past problems with motor-operated valves and they include l 20 switch settings incorrectly set as you are fully aware of, 21 excessive wear, and also thermal overload, which it seems to 22 be a nagging problem that won't go away. The causes are I 23 addressed. There are many--33ck of coordination between-- 24 DR. CATTON: What is thermal overload? 25 MR. SCARBROUGH: When the motor itself starts to
59 /~N 1 overheat; the lack of coordination between plant organizations N-]. 2 has led to improper sizing in some instances. 3 Inadequate engineering analysis, we have heard about 4. that from KWU people where they talk about the valve thrust 5 equation, and possibly our setup of that equation isn't 6 exactly proper. 7 Inadequate training and procedures, we have seen 8 that cause problems with motor-operated valve performance, and 9 in-service test criteria, that is in Section 11. The stroke 10 time and seat leakage alone do not tell us a whole lot about 11 the condition of the MOV. 12 There has been several events which has-- 13 MR. CARROLL: You left off one that I wish you guys j (} 14 would stop ignoring, and that is thermal binding of J 15 motor-operated valves. That's a real problem. j i 16 MR. SCARBROUGH: In terms of the motor itself? l 17 Thermal binding where? 18 MR. CARROLL: No. When you, when you close the l ) 19 valve cold and then heat it up and it gets differential l 20 expansion, you may not be able to pull it out of its seat. l 21 MR. SCARBROUGH: Right. That's one of the things 22 that Generic Letter 89-10 will address because it will have to 23 perform to that testing at design basis conditions, 24 temperatures, flows. 25 MR. CARROLL: That won't do it.
60 1 I' 1 MR, SCARBROUGH: I will hear more about that. I \\df 2 thought we would handle it. 3 MR. CARROLL: It is listed in the table. I guess we 4 asked you guys to put it in the table, i t 5 CHAIRMAN MICHELSON: You have to understand it. 6 MR. CARROLL: I have to understand what the words 7 mean I guess. 8 MR. SCARBROUGH: Okay. The Davis-Bessi event 9 occarred in June of '85, which was a, started off as a loss of 10 main feedwater, and by inadvertent operation, the auxiliary 11 feedwater discharge valves were shut and they weren't able to 12 reopen them. The bypass switch wasn't set for a sufficiently () 13 long period. 14 The Catawba event, we just talked about 15 Millstone--all events that have raised our awareness of 16 motor-operated valve problems. 17 (Slide) 18 MR. SCARBROUGH: In terms of regulatory 19 requirements, there is quite a few. The Appendix A and 20 Appendix B criteria, Part 50, cover motor-operated valve 21 performance in a number of areas, but in very broad terms 22 which need to be clarified. 23 Section 50.55A applies the criteria of Section 11 to 24 code class valves, but it does not cover all of the aspects of } 25 Appendix A or and B, Appendix J addresses containment
61 y (3 1 isolation testing, and Section 50.36 is the general section'on i .g 2-preparation of standard or technology expansion of technical 3 specifications. 4 (Slide) 1 l 5 MR. SCARBROUGH: The NRC activities are, there is 6 many activities going on with NRC on various aspects of MOV 7 performance--the Mechanical Engineering Branch performance 8 evaluation, special inspection branches, risk application i 9 branches, these have 7 04 to a number of documents 10 issued--Bulletin 85-U1, Generic Letter 89-10, Generic Letter i 11 89-04, and the pressure isolation valve program, IS LOCA, the 12 Office of Research. (} 13 MR. CARROLL: Tell me about that again. 14 MR. SCARBROUGH: IS LOCA program, intersystem LOCA, 15 MR. CARROLL: Okay. 16 MR. SCARBROUGH: That's Rich Barrett. He has a 17 program that they are developing. The Office of Nuclear 18 Regulatory Research has several programs in terms of MOV 19 perform. Generic Issue 87-10 and 04 which had a large part in 20 the development of 85-03 and 89-10; a performance program 21 which has been conducted by E Tech, and evaluates various 22 aspects of MOV performance. One is the use of motor signature 23 monitoring to measure MOV operability, and that program is C 24 nearing completion from my last discussion with the contact in 25 Research on that program.
o 62 l ()~ 1 MR. WOHLD: 'Who is that contact? 2 MR. SCARBROUGH: Jerry Weidenheimer; the nuclear 3 plant age research program being conducted by Oak Ridge, Phase 4 1 addressed failure modes and parameters, and Phase 2 looked I 5 at effective methods for monitoring MOV conditions. 6 NUREG/CR 4234, Volumes 1 and 2,. Volume 2 was just 7 issued very recently and looks like it has some good l 8 information in.it. I haven't had a chance to read it in 9 detail yet. 10 The Office for Analysis and Evaluation of 11 Operational Data has had several reports related to MOV 12 performance. The major report was a case study report, C603, () 13 issued in December 1986. It reviewed MOV operating 14 experience, applied recommendation for resolution of concern 15 for MOV performance. 16 SECY paper 89-165 was issued in May of this year, 17 and the EDO indicated the responsibility for the overall 18 coordination of NRC activities and the oversight of industry 19 has now been transferred from AEOD to NRR. 20 (Slide) 21 MR. SCARBROUGH: Other activities that are ongoing I 22 includes the EPRI work, preparing technical repair guidelines 23 for Limitorque SMB-000 that was issued in January of this 24 year. They were also working on some application guidelines {} 25 for MOVs which hopefully will be issued later this year, j
') 63 '\\ 1 INPO has stepped up its plant visits in terms nf [G 2 motor-operated valve operability, and prepared a summary of 3 key elements for comprehensive MOV program in October.of last 4 year. 5 NUMARC has also been involved in MOV performance. 6 They are preparing guidelines and implementation of Generic 7 Letter 89-10 looking at signature tracing, technique 8 improvements, MOV design improvements, and IEEE maintenance 9 practices has also been issued this year. 10 MR. WOHLD: You say NUMARC has something out on 11 89-10 already? 12 MR. SCARBROUGH: Not yet. It is still in the works () 13 I believe. 14 MR. WOHLD: Okay. 15 MR. SCARBROUGH: The signature tracing techniques 16 includes MOVATS and VOTES and others. And the IEEE document 17 was issued earlier this year and looks at MOV experience, i 18 installation and maintenance guidance and various signature 19 tracing techniques. 20 ASME has several documents that it has in 21 preparation. The first is OM-08 on performance testing for 22 motor-operated valves, OM-10, which was recently endorsed by 1 23 addenda to the code, and then provides in-service testing 24 criteria. QME-1 looks at the qualification of motor-operated 25 valve assemblies, and that is currently in preparation. You
i 64 1 can see there are several activities ongoing within industry {} 2 as well, but they aren't coordinated in their efforts, either. 3 MR. CARROLL: Does NUMARC agree with you? 4 MR. SCARBROUGH: Clyde is here. We can ask him. 5 MR. WOHLD: You've got to give Clyde some time. 6 MR. CARROLL: That was a rhetorical question. 7 MR. SCARBROUGH: Hearing no objection, I will just 8 continue. 9 The proposed action plan itself has several parts to 10 it. Let me show you. This is kind of a very quick summary of 11 the various efforts that will be performed within the plan. 12 Mechanical Engineering Branch is assigned the lead for the () 13 effort. MEB will lead or direct MOV audits and inspections as 14 its expertise and resources allow. It also will be involved 15 in coordination of MOV assignments and inspections by other 16 headquarters NRR units, and will also follow, review the 17 results of MOV audits and inspections by region personnel. 18 AEOD will continue to record on MOV experience, and in this 19 case, MEB will interact with industry. An example of this was 20 a recent letter to NUMARC that MEB prepared. 21 And Research will continue to prepare rules and 22 regulatory guidance with standards from HEB as needed. 23 (Slide)
- q 24 Mn. SCARBROUGH
The next couple of slides gives you (/ 25 a detailed rundown of the various activities that need to be
65 () 1-coordinated or involved with by the Mechanical-Engineering 2' Branch, starting off with what I consider the cornerstone of 3 getting our arms around the MOV issue which is to prepare 4 revision of the in-service testing regulation, regulatory 5 guidance that will go along with it, and this is an example 6 list of various areas that need to be addressed by the rule ] 7 change or the regulatory guidance, importantly the scope we 8 need to make sure that the industry understands the scope is 9 noc limited to code equipment for in-service testing. 10 CHAIRMAN MICHELSON: What does that mean? 11 MR. SCARBROUGH: Currently the Section 50.55A only 12 applies Section 11 testing to code equipment, code Class 1, 2 () 13 or 3 code pumps and valves. That is not nearly the 14 safety-related class of equipment, and it needs to be clear 15 that that equipment needs to be included and also the 16 equipment that is in safety-related systems, that could 17 function in a way that would interfere with the ability of the 18 safety-related system to perform its function also has to be 19 in that program, include testing, so what we are doing is-- 20 CHAIRMAN MICHELSON: Are you talking beyond pumps 21 and valves then? 22 MR. SCARBROUGH: No. We are only going to-- 23 CHAIRMAN MICHELSON: How would you have a, what (} 24 would be an example of a Class 3 system that did not have ASME l 25 valves in it? i _____m ._m -mm a mna n a A .A aaa m
66 i (~} 1 MR. WOHLD: . Talking about primarily non-code plants? s-i 2 MR SULLIVAN: Not necessarily; for example, in many 3 of the diesel generator auxiliary systems, they are not code 4 class systems but safety-related pumps and valves. 5 CHAIRMAN MICHELSON: That would be an example all 6 right because much of that stuff is not covered with the code 7 at all. It is safety-related all right--diesels, freon 8 compressors, thing likes like that. Packages all have a 9 number of valves on them that are not in the, normally not 10 covered by the code. 11 MR. WOHLD: Older plants, I don't know, maybe newer 12 plants, too, but I know in some cases they provide a diesel '() 13 cooling. The pumps and valves in the intake structure leading 14 up to the skids were not code because the philosophy was if 15 they are going to tie into a non-code pipe, why make the rest 16 of it code? So they made it non-code all the way. 17 MR. SULLIVAN: You are right. It is not just skid, 18 but-- 19 MR. BAER: I might comment on the O&M codes. We 20 have been having a concerted effort to have the schemes 21 written in terms of the functional requirements rather than 22 the ASME classification for just those sort of reasons. 23 MR. SCARBROUGH: Part of the rulemaking effort will 24 also involve OM-10, and I believe that's in process right now, ) 25 and also clarify that testing is intended to demonstrate \\ .-,,,nn -na
67 j {}' 1 operability and not just purely look at whether or not it 2 meets a stroke time criteria. 1 3 Other areas that may be incorporated into regulatory j I 4 guidance later is Generic Letter 89-10 and 89-04 provisions, ) 5 in-service testing program descriptions, the training and 6 other in-service testing guidelines. 1 i 7 MR. WOHLD: My question was that what you are saying 8 here is that it looks to me like you are departing from the l i 9 past practice of letting the code be responsible for defining i 10 this testing here and take more responsibility within NRC and l I 11 saying you have, okay, you have got the code, here are these 1 12 other requirements. O) 13 ,MR. SCARBROUGH: That is correct. 14 MR. WOHLD: Thank you. I 15 MR. BAER: We are also trying to push the code 16 committees a little harder, too. Trying to do both. 17 MR. SCARBROUGH: So these other areas that beyond 18 the, just a rule change itself, we are looking at other areas J 19 of MOV performance, maintenance, that could use some improved 20 regulatory guidance to thermal overload issue, try to resolve 21 that one way or another. 22 Guidance for NRC personnel, inspection manual 23 chapters where we are weak in that area. That was one comment 24 I received from people who were out there doing inspections. 25 There is now a good resource of guidance in that area.
( qi; 1 .;,~ t 68 i, ; j l' Prepare technical guidance-for Generic LetterL89-10, L 2" we.are in the process of now preparing the TI for V, t 3-implementation of' Generic Letter 89-10. The workshops we != L 4 have, there should be workshops that are complete.now. We are L l 5 - in,the process of preparing meeting minutes. We will go j ) 6 through the vast majority of questions and respond to'them-in 1 1 7 - detail, and indicate our current positions on all of the l i 8 , questions that came in, j ~! 9 Next year.we hope to' perform a review of someLof the 10 program descriptions and will-be at the plant sites to ensure l 11-that the licensees are implementing the generic letter as we-12 had intended. () 13. And then also as part of the 89-10,ef' fort,.whether 14 it is industry testing of'a generic nature, we will be '15 available to look over that testing to ensure that it is 16 applicable to different types of MOVs, and-conducted properly. 17. As I mentioned, the MEB will be involved in MOV '18 audits and inspections of a general nature beyond 89-10 and 19 help coordinate other MOV inspections by NRC personnel. 20 We hope to hold coordination meetings on MOVs 21 quarterly, and to ensure efficient, effective regulatory 22 approach in terms of MOVs throughout the Agency. 23 We are also reviewing research program results, and 24 look at new program proposals, and we also assist industry in 25 its efforts for MOV performance by various meetings and l ________m m_m A hn A A A A A ^^n M
69 a () 1 reviewing guidelines in_ areas in that nature where we can be 2 of assistance. 3 We also hope to review some of the INPO plant visits 4 and find out where their strengths are, and where we may be 5 able to adjust our program. l 6 MR. WOHLD: Tom, can I slow you down for a second? 7 -MR. SCARBROUGH: Sure. 8 MR. WOHLD: What, are there any efforts or do you 9 expect any efforts of people to come in and present programs 10 to you to get some sort of approval as opposed to waiting for 11 an on-site' audit, or are there any venders coming in? I 12 forget'the process. In the licensing process, don't you have () 13 topical report? 14 MR. SCARBROUGH: No. We intend that we will review 15 the programs on the plant sites at the plant site in, possibly 16 through a generic industry effort where they are developing 17 some sort of guidance where we could be of assistance in that 18 area to them. 19 I would be receptive to that, but we are not looking 20 at doing pre-programmed reviews of MOV programs. We intend to 21 do more inspections at the plant sites rather than 22 pre-approval. 23 MR. WOHLD: Would you consider approval up front if l f} 24 someone presented that to you, an Owners Group or a NUMARC? 25 MR. SCARBROUGH: If it is an industry effort that's
70 ~ 1 a generic effort, certainly. We are coing to try to work with {} L L 2 industry and let them know where we have problems with what l' 3 they are doing or we, we have approval for what they are 4 doing, so we will definitely be receptive to that. 5 MR. WOHLD: Thank you. 6 MR. SCARBROUGH: This is, the next item is 7 evaluation of industry efforts so we can cooperate very much. 8 We still have to evaluate whether or not they are actively 9 pursuing problems that come up and whether or not we need to 10 adjust our program in response to some problem that isn't 11 being resolved as quickly as necessary. As new design and 12 testing improvements arise, and MOV experience occurs, we may () 13 have to modify our program to reflect that new experience and 14 improvements. 15 Where technical specifications will be revised, MEB 16 will be available for assistance as needed. 17 Also in terms of plants technical specifications 18 where a licensee proposes to modify its plant technical 19 specifications and the project manager needs assistance in 20 that area, we will be available to assist in the review. 21 The valve review group, which is currently headed by 22 Jerry Weidenheimer, we intend to be fully cognizant and 23 participate in that group to a large extent, and we consider 24 that an important part of this whole program. 25 We also intend to continue to participate in the
i . 71 ' ~ j! 1 co' des and standards development. We currently are and we.will- '2-continue ihat working.with Research and other groups within. 3 NRC to-help prepare those codes'and standards in a manner that- .4 is. acceptable to NRC. 5-Bechtel/KWU Alliance, they have quite a bit of good 6 information that we can use. We have already seen someLof 7 their presentations, and we continue to, will continue.to 8 . follow their efforts. 9 And finally, a Commission paper on MOV status, and 10 what our activities are, we will prepare those on a periodic 11 basis. We have one:in~ process right now which we hope to send 12 down to the Commission in October, this month. 13-MR. WYLIE: Let me back up to your item C on thermal. f -( j; 14 overload. 15 What is the issue there? 16 MR. SCARBROUGH: I'm sorry? '17 MR. WYLIE: Thermal overload protection, it is 18 basically the overload protector in the starters I assume. 19 There is some problem. What is the problem? 20 MR. SCARBROUGH: The problem there is they either 21 trip too early or they don't trip soon enough. You either 22 trip them before the valve can do its job safely without 23 damaging the equipment, or too late where it has already 24 ' burned itself out. 25 MR. WYLIE: That's the question of valve not being ________m
72 l ~' (G 1 adequately sized to begin with, not the overload. 2 MR. SCARBROUGH: Whether or not the motor can handle 3 the torque that is required. 4 MR. WYLIE: I understand, but the overload protector 5 is there to protect the motor from burning up, and if it is 6 undersized, it is going to burn up. I just fear-- 7 MR. BAER: Let me try, or Owen--Owen wrote a topical 8 on this. Maybe I will introduce it and let him talk about the i l 9 detail because there has been a reg guide out for a number of 1 10 years which the staff, which is written in a manner where the 11 licensee can interpret it in a way, and the staff has in the I 12 past encouraged this interpretation, where you size the j ) rs 1 (,) 13 thermal overload set points at such a point that they will j 14 never get activated, either that or not use them, so that 15 there's an old philosophy the valve ought to working without ) l i 16 realizing that this was slowly degrading the valve motor, the i 17 actuator motor in many cases, and Owen had done a little study l i 18 of this. 19 MR. ROTHBERG: We put out you NUREG, 1296, and 20 essentially what we said was it looks like the regulatory 21 guide provides guidance. There are three alternatives. The i 22 first is to remove thermal overload protection except for 23 testing. 24 The second is to allow thermal overload protection {) 25 to be in the system, but to take it out for accident signal,
73 ..()' 1 and the third alternative was to set all thermal overload, 2 quote, with all uncertainties resolved in favor of completing 3 the safety-related action. 4 What people had done was interpreted that third 5 criteria to say oh, you mean that we will set these thermal 6 overloads so high that they will never work and by extension 7 of that, we have only provided, for instance, Davis-Bessi 8 doesn't have overload on motor-operated valves. 9 What that means, of course, is that the motor itself I 10 then becomes a thermal overload because that's the, that's the 11 way it works out, but we recommended in the NUREG, people 11 2 realistically size thermal overload protection and IEEE 741 () 13 has an appendix being added to it to provide some guidance i 14 finally for setting thermal overload protection on l 15 intermittent duty operated valves. 16 What I guess Tom is alluding to in this is not to 17 changing the regulatory guides, but to overseeing people using 18 the more realistic criteria for thermal overload protection in 19 their plants rather than just rerouting the systems. It is 20 going to be kind of hard to do because the tech specs on a 21 number of these plants, especially the newer ones, call for 22 taking the thermal overload out of the circuit entirely, and 23 there were a number of--in plants, new plants, they put the 1 24 thermal overloads in, and they just are, the way they are set {} 25 now, they are useless. 1 l --n nn,e
1. 74-(). 1 CHAIRMAN MICHELSON: I thought they were going to l 2 bypass.them only for the ECCS, only when the ECCS signal. I i 3 MR. ROTHBERG: That's the second option. ) 1 4 CHAIRMAN MICHELSON: Nobody implemented--did many 5 people implement that option? l 6 MR. ROTHBERG: There are several people that do '7 that. However, the problem with those, with the bypassing 8 except for the ECCS is that it is, the best I know, only done 9 in BWRs because it is very difficult to come up with an 10 accident scenario, control--it is a pretty restricted thing. 11 GE for many years advocated taking thermal overloads 12 out of safety-related motor-operated valves, and so there is a () 13 -bit of confusion about what is to be done. 14 What we tried to do in NUREG 1296 was tell everybody 15 about exactly what was going on, give a summary, and say look, 16 you have got this opportunity to provide thermal overload 17 protection that will protect the motor operator without 18 causing spurious trips, and given guidance on how to set up 19 your motor-operated valves, thermal overload protection, it 20 appears to be doable. That's the type of thing that we would 21 like to see people do. 22 CHAIRMAN MICHELSON: Two thermal overload protection 23 methods--one was a simple thermal overload device which was a 24 separate device. {} 25 The other was buying breaker for thermal overload
t L 75 ( (' } - 1 provisions, which were much more precise. 2 MR. ROTHBERG: By motor case breaker, you are l 3 speaking of thermal overloads inside the motor itself? 4 CHAIRMAN MICHELSON: In the motor case breaker, I L 5 thought those came with thermal overload-- 6 MR. WYLIE: They are the breakers with thermal 7 overload protection and magnetic protection, but the more 8 conventional way I.have always done it is to use thermal 9 overload elements which have heaters in the starters. They 10 are pretty good. They are better really than the breakers,- 11 but the problem is obviously if the overload protector is 12 properly sized, and it actuates, it is because the valves got () 13 something wrong with them. It is binding or something. There 14 is the question of whether you want to sacrifice the motor for 15 operability of the valve. 16 MR. ROTHBERG: The problem to that, sacrificing the 17 motor, is that you are making an implicit assumption, and that 18 is that the valve will go to its proper position and then the 19 motor will die. 20 MR. WYLIE: Not necessarily. You have, you can set 21 the overload protection the other condition. l 1 22 MR. ROTHBERG: That is correct, that the problem in l 23 the past has been that most of the guidance for setting 24 thermal overload has been for continuous duty motors. Until 25 this appendix to IEEE 741 came along, there were just a couple
!~ '76
- (')t 1
of papers, one by Formica, and maybe another paper by,a fellow 2 named Baxter. Those were the only two guidance papers that 3 gave any guidance about how to set the thermal overload for 4 duty motors, and they weren't--IEEE 721 seems to go a much 5. longer way to giving people a, a thermal overload protection 6 that, that is realistic without causing spurious trips. 7 MR. WOHLD: Just a couple comments.on thermal 8 overload--the problem I see is that the thermal overload is a 1 '9 hundred yards away from the winding, and unless the thermal 10 characteristics and heat-up and decay characteristics and time 11 constants are the same for the thermal overload as they are 12 for the motor rotor, I don't see that they have a valid () 13 thermal protection. 14 MR. ROTHBERG: I agree. 15 MR. WYLIE: It is a limit. It limits the motor. 16 But--the heater is another one, heats up and limits the motor. 17 MR. WOHLD: But it, I don't, I don't expect it to 18 mimic the thermal decay contact. What I see is-- 19 MR. WYLIE: Not precisely, but you are talking about 20 for, if you start it frequently, it retains the heat and keeps 21 building up. 22 MR. WOHLD: For instance, if you are doing MOVATS 23 testing, you exercise the valve and hook up some different 24 connections, exercise it again, I see the motor heating up, {} 25 heating up, heating up, and the thermal overload heating up l
77 1 and cooling.down. 2' HR. CARROLL:- That's'right. So what? L 3' MR. WOHLD: 'Well,.the windings can overheat and the I; l 4. thermal: overload will never protect it,'so you can, even if 5 you have a properly sized thermal overload, I think it'gives a ~ 6 false sense of confidence. You have to observe the duty cycle. l 7 of the motor. 8-MR. CARROLL: Sure, it'is true a lot of, most-ECCS 9 in power plants', you can't just keep. starting them..All the 10 big motors have limitation on how many starts per hour. 11 MR. WOHLD: You' don't have thermal overload 12 -protection on large motors. () 13 MR. WYLIE: I mean--I will tell you what. The 14 former utility I was with provided lock rotor protection. In 15 large motors, pumps and what have you, you actually put 16 thermal couplings in the windings, but the--and.you alarm 17 those, the operator. You didn't trip on them. You tripped on 18 short circuit and locked the rotor. 19 Now the smaller ones like the valves, that's a 20 problem because you can't, normally you don't put 21 thermocouple in the winding of the valves. 22 CHAIRMAN MICHELSON: Gentlemen, we have got a i 23 concerted time. We can either talk about thermal overload or d } the KWU tests that are being performed in Germany. We have 24 25 some preliminary results. It is not on the agenda, so we have ) .---~-.. ----
_____ _ m 78 (' 1 got the choice We don't have time for both, w] 2 MR. WYLIE: I'm just curious what the problem was. 3 CHAIRMAN MICHELSON: Roy Woods I understand is here 4 and is going to tell us a few minutes about what they are 5 finding so far in the KWU test, which would probably be quite 6 pertinent and you wanted to fit it'in at this time as I 7 understand it. 8 MR. SCARBROUGH: That illustrates why we flagged 9 thermal overload, because it is hanging out there and needs to 10 be resolved in some fashion, and that's why it is called out 11 in the action plan. 12 CHAIRMAN MICHELSON: I assumed you had finished. () 13 I'm sorry. 14 MR. SCARBROUGH: I had. Thank you. 15 CHAIRMAN MICHELSON: Okay. 16 HR. CARROLL: I had one comment I wanted to make. 17 It seems to me something missing from your program is that I 18 think all of the recent activity has sent up a false message 19 to the, to the PRA community. I think they are very confused 20 about how reliable or unreliable these valves are. I think 21 failure has been ill-defined. I think there is a lot of 22 problems. It seems to me that somebody ought to sit down and 23 try to, try to put something together as to the current 24 situation with respect to MOV reliability. 25 CHAIRMAN MICHELSON: For PRA purposes.
L e 79 >~3 1 MR. CARROLL: For PRA purposes. \\.) L 2 MR. SCARBROUGH: Okay. Thank you. We will pass 3 that along to the PRA people, 4 .MR. CARROLL: That's not-- 5 CHAIRMAN MICHELSON: That's not going to do any 6 good. They won't do anything with it. 7 MR. CARROLL: I think the staff should have by now a 8 pretty good feel for how reliable these stupid valves are or 9 aren't. 10 MR. SCARBROUGH: I don't have a number, though. I 11 don't have a, you know, number that they can plug in. 12 MR. CARROLL: I think you could, for example, you 13 could certainly say the Brookhaven number of 29 percent 14 failure, right, is totally out to lunch. 15 MR. BAER: That was an upper bound. Here we are 16 using 8 percent. 17 MR. CARROLL: I know, Bob, but I hear the number 18 being quoted. I think the other thing you have got to do is ) 19 look at what failure really means. 20 If a valve goes 99 percent closed, it still has a 21 little leakage through it, it sure as he]l has quote, failed, 22 but in terms of safety significance, it may be trivial, so I 23 kind of think the people that have been involved in this issue 24 on the staff need a paper of some sort to provide some 25 insights to PRA people who, you know, many of whom are number
a i i 80 l 1 /^ 1 crunchers who wouldn't know a valve if they stumbled over one, \\~)\\ 2 but they put numbers into them in the PRAs. 3 MR. SCARBROUGH: That's a good point. If they are i 'i 4 all using numbers which are totally unsupported by data, we 5 know then they need to change their number. 6 MR. CARROLL: That's right.
- /
CHAIRMAN MICHELSON: Go ahead. Would you give us 8 what you know about the dates of KWU teste? 9 MR. WOODS: I'm Roy Woods with the staff. As you 10 probably know, we are right in the middle of the testing 11 program now. 12 We are testing three valves typically of the reactor (} 13 water cleanup system, so those are tests in the water. They 14 are 6 inch flexible wedge gate valves, 900 pound and 600 pound 15 class. Both are represented. 16 Then we are going to test three valves of the HPCI 17 steam line. Those will be 10 inch flexible wedge gate valves, 18 So far, we are most of the way through the water tests. I 19 understand Dr. Catton is doing one of the steam tests. I am 20 going up for all the steam tests. Okay. 21 We have tested, let me--I am not going to identify 22 the valves on the record like this, but you can recognize 23 them. We have tested what we will call valve A. It is the 24 same as the previous valve A--900 pound class valve. We 1O 25 tested it at normal operating pressure and normal operating
? N j p;
- is li 4-81 L
)[ 1 temperature, which means,a thousand psi gauge in like 10 2 'subcooled. I think that's 30 degrees Fahrenheit. We set'the-13 torque switch to provide a stem closure thrust that we believe 4-would cause the valve'to close and seat to isolate the flow L 5 and-seat. 6 Based on all.of our experience from the previous 7 tests, and we added a thousand. pounds stem closure force on. 1., 8 . top of what we thought it would take based on that previous 9 experience. 10 The valve'didn't seat. The valve theoretically-11 isolated the-flow, meaning that the lower surface of the seat 12 on the gate came down to the upper part of the seat on the q () 13 bottom of the valve body, and this just touched 11t,'and you 14 could say theoretically that should have isolated the flow, 15 but there'was considerable acoustic emission, in other words 16 noise, sand it was able to measure the flow.down that range, l l jl 17 but it didn't isolate. It is still exposing equipment, so 18 your question, is that a failure or not, well, that's up.to i 19 the licensee to argue, but it certainly wasn't a success in j ') 20 the sense that it met, it did what it was designed to do. l 21 CHAIRMAN MICHELSON: This was under pipe break 22 conditions? 23 MR. WOODS: Full-sized breaks simulated downstream. (} 24 MR. CARROLL: Was the valve at that point 25 disassembled to see if there was a piece of weld rod or i I l
82-1 [ '1 .'something under the-- 2L 'MR. WOODS: I. hope the contract they 'are--we are i q -3 paying large bucks so I hope it does a better job of setting , 4' it up'than to set it up with a piece of weld rod, but i l 5L .certainly it was looked at. s
- 6 MR. CARROLL:
It was disassembled? 7 MR. WOODS:.Yes. I 8 MR. CARROLL: What did they find? 9 MR. WOODS: The valve is basically destroying ~ 10 itself. You1can tell from the' data that it wasn't smooth' 11 friction where-it is sliding smoothly closed. 'It was--and the 12 compression in the stem, you can see large noise'where the. '( J- _ 13 valve is hanging and sticking. When'you pull it harder, it 14 jumps and it. jumps again and so you are actually--it is, if 15 you want to characterize it as a friction, it would be like a 16 .8 coefficient, but it is not friction. It's more like 17 gulling or extrusion or whatever. L ' 18 MR. CARROLL: There was evidence the gulling on the 1 19 guide is-- 20 MR. WOODS: We haven't disassembled the last time O l 21 talked to them enough to look at guides because it is no 22 longer critical path. We were criticized roundly on the first 23 round for testing damaged valves. This is a damaged valve. 24 It is not being further tested. It is on the scrap heap to be ) 25 looked at later, so there was damage on the surface of the
83 i <~ 1 disc, which we can see by looking in, but as far as totally k. 2 disassembling it and measuring the guides and whatever, to my 3 knowledge, they haven't done that yet. 4 MR. CARROLL: Okay. You have said that the leakage-5 was detected by acoustic emission. 6 MR. WOODS: I could hear it roaring. There are 7 acoustic monitors on it. 8 MR. CARROLL: Did people actually measure the 9 percent flow? 10 MR. WOODS: You can't measure it down in that range. 11 They are not set up to measure that type of flow. 12 DR. SIESS: Steam? ~ 13 MR. WOOD: It is is a' water test. 14 DR. SIESS: You can't measure water to put a cup 15 down there? 16 MR. WOODS: It's 530 degree water coming out. If 17 you want to stand there, it's all right with me! 18 DR. SIESS: That tells me something. There are more 19 water than I care to stand by. 20 MR. WOODS: Yes, sir, indeed. Now this particular 21 test is not like the ones at Wyle where you can see a video 22 camera coming out in the trees because it is a different 23 facility and they have a quencher on it, so it is going nicely 24 into a facility designed to collect it. 25 MR. CARROLL: But couldn't somebody do a heat l
84 1 balance or something? l (~)N \\- i l 2 MR. WOODS: We have all kinds of data. We may be 3 able to get some idea what the flow was. I can't tell you as 4 I sit here exactly and get-- 5 CHAIRMAN MICHELSON: The downstream conditions for 6 this test were somewhat different than the Wyle test. The 7 Wyle was a guillotine type break whereas this is really just a 8 dumping into a large volume system, but still a fair back 9 pressure, is that correct? 10 MR. WOODS: I think the back pressure is negligible. 11 CHAIRMAN MICHELSON: How much was it? Was it big 12 enough volume to give very low back pressure? () 13 MR. WOODS: All you have to do is raise the bottom 14 pressure two or three pounds to compensate for that and you 15 have the same test. It is just a--I guess noise-wise, they 16 are not allowed to, in that location, to make that kind of 17 record.of the countryside. l l 18 I have two or three other tests to tell you about. l l 19 CHAIRMAN MICHELSON: Go ahead. 20 MR. WOODS: This was the first one we did. We are 21 not testing that valve any further. Incidentally, that was 22 the same as the previous valve, indeed the identical same 23 valve, but it has been refurbished. 24 MR WOHLL: Was the signature pretty much the same? 25 MR. WOODS: I believe it was. That the difference I (
1. 85 ]) 1 is in this case, you have a little bit greater load on the 2 valve probably because they were able to maintain the pressure 3 more consistently during the blowdown. It is much larger. It 4 is capable of doing the test you really wanted to do. He got 5 some pressure losses in the system previously. 6 MR. CARROLL: This was the guy as I remember with 7 the very wide gauge or wide-- 8 MR. WOODS: That is correct. 9 MR. CARROLL: Okay. 10 MR. WOODS: This still seems to be related to the 11 fler:.ibility of the valve, somewhat partly at least due to the 12 gap that you referred to. () 13 The second valve we tested, let's call it valve W, 14 we had not tested it before. It was a 600 pound type valve, 15 which means it is a wider weight valve, less robust, that it 16 has a smaller stem, smaller diameter, and the limit on how 17 hard you can try to close this valve is the column influx in 18 the stem, and so we were told by the manufacturer or his 19 representative--I understand they no longer make valves for 20 this purpose, but we were told what maximum thrust we could 21 set the torque switch for, conversion, and it was 22 considerably, vastly lower than the amount we were able to put 23 on this valve A. I think like 10,000 pounds less, 50 percent, () 24 plus the fact that it is a more flexible valve, so we believe 25 that the probably is it might tend to take more thrust. We ________-_mm .--m m a o m am m
n a 86 13 t 1 .couldn't even get anybody to take the bet for three beers or 2 -any amount of moneyLthat this valve would close. Guess what?. '3 It closed three times just fine. 4 MR.' CARROLL: Did it have much less gap? ,5 MR. WOODS: I believe it is a lower gap. 6-MR. CARROLL: I would have taken your bet. 7 MR. WOODS: 'You weren't there. You would have-
- l
-8 gotten three beers! 9 MR. CARROLL: It closed three times? 11 0 MR. WOODS: It closed a thousand pounds, 10 degrees 11 subcolled, 100 pounds, 10 degrees, and then even to though it-12 is'beyond the' design that this would have to see in the i
- (])-
13 background, at least we ran it at 400 pounds, and it closed. l 14 That last time we lost some of the data because they, they, 15 put the vibration monitors on and.at 20Gs it quit, and also 16-the stem compression gauge gave out, but it worked. 17 MR. CARROLL: The signatures looked pretty much the i i 18 same on that? i' l 19 MR. WOODS: No. They looked quite different. It's 1 20 clearly a different valve. It is clearly-- 21 MR. CARROLL: For this valve, the three tests on 22 this valve? 23 MR. WOODS: They looked similar to themselves, but j 24 different than valve A. 1 25 MR. CARROLL: But the difference being it was a l i
87. L () 1 smooth signature and valve A was lurching. 2 MR. WOODS: Concave in a different direction; it 3 doesn't have the same shape at all, which leads you to believe 4 what you are looking at is different surfaces riding on things 5 that are inherent to that design. It is not coming down on 6 the guides and then transferring in the same way that it does 7 on the other valve because it is not the same internal there. 8 CHAIRMAN MICHELSON: That this was a different 9 vender. 10 MR. WOODS: Different vender, totally different 11 valve. Still a flexible wedge gate valve. Otherwise it is 12 different. Both of these valves had the normal size operator () 13 -on for that application and that surface whereas the previous 14 type valve A ran a much larger, and I guess I should say that 15 if we had set valve A at the maximum torque switch that would 16 have been allowable, there would have been plenty of thrust to 17 close the valve. It destroys itself in the process, but it 18 would have closed. 19 Now we are currently running valve V, which is the 20 same as valve B we ran before. The first test has been 21 examined and that was a thousand pounds, 10 degrees subcooled. 22 They worked fine, just about like it did before, like we 23 expected, and we are now running a series of tests at a 1 (]} 24 hundred degrees subcooled at various pressures. In fact, I 25 expect a call within the next two or three hours to update me j l
b j ?- L 88 L )( -1. J onithat.
- 2' That's'all I have.-
3L 'MR. CARROLL: What was the implied friction factor, 4 Roy,-on-valve W? l Is MR.-WOODS: It looked like it was around 3, but at~ ( 6 the time there was'an error. We had put in-the actual stem-7
- packing--I'm sorry.- We put in the maximum stem packing and 8-the actual stem backing load is less, so'that would tend to 9
affect what the code'would back out at friction factor. . 10 -Honestly I don't'know, but it wasn't usual. 11 MR. WYLIE: Are the results going to be published in 12 a NUREG or something? ([ 13 MR. WOODS: Certainly. The results of the. previous 14' Ltest, by the way,-are available now in the Public Documents 15-Room as an informal contract report. 16 MR. CARROLL: We have got a copy? 17 MR. WOODS: The people that were at the February 1st 18 meeting got copies. It is now available to anyone who wanta 19 it, and the electronics data is available just because of the 20 copying. 21 MR. WYLIE: NUREG, these are NUREGs? 22 MR. WOODS: It's not a NUREG yet. We are currently 1 23 in the process of publishing it as a NUREG. At the moment it L( } 24 is available in a less formal case. It is the same data. We 25 haven't made a NUREG out of it yet. I 1 .mm.m .m m mmmm
89 ( ) 1 CHAIRMAN MICHELSON: Any other questions? 2 MR. WOODS: Dr. Catton, did you get.the information i 3 you needed? 4 DR. CATTON: Yes, which was your phone number. 5 MR. WOODS: Okay. You probably know it and I don't. 6 That's fine. I don't want itt 7 MR. CARROLL: Do you think your new philosophy, Roy, 8 of not testing damaged valves is appropriate? 9 MR. WOODS: Well, certainly you don't know what you 10 are getting if you test damaged valves. You don't know what 11 you are testing. WE can't be criticized this time for testing 12 a damaged valve. We put it on the scrap heap. () 13 MR. CARROLL: Yep. 14 CHAIRMAN MICHELSON: Thank you, Roy. We will 15 adjourn until ten to. 16 (A brief recess was taken.) 17 CHAIRMAN MICHELSON: We are ready now to start with 18 the check valve discussion, and I believe Mr. Scarbrough is on 19 again? 20 MR. SCARBROUGH: Yes. 21 CHAIRMAN MICHELSON: Good. Will you tell me, just 22 for my edification, now what part of NRR has got the 23 responsibility for the valve? That's Tad Marsh, Mechanical 24 Engineering Branch? {} 25 MR. SCARBROUGH: Yes, sir.
'90 p l(f
- 1 CHAIRMAN MICHELSON:
Are you a sectionLin that ~ 2 branch? 3-MR..SCARBROUGH: I am in the'in-service testing 4 section. 5 CHAIRMAN MICHELSON: It is not in one section? .It. l I 6-is all over'the branch? i 7 MR.'SULLIVAN: It is one section. 8 -CHAIRMAN MICHELSON: What.section is that? 9 MR. SULLIVAN: Quality in-service testing section. ) 10 LCHAIRMAN MICHELSON: -That's where all the work.will 11 be for NRR? ) 12 MR. SULLLIVAN: That's where the MOV work will be. I (f. 13 The check valve work, it is not as clear, so-- 14-CHAIRMAN MICHELSON: It will be within the -15 Mechanical Branch? 16 MR. SULLIVAN: Yes. That's true. 1 17 CHAIRMAN MICHELSON: Just want to get the suppliers l J l 18 straight. 19 MR. SULLIVAN: That's as straight as we can make it j i .1 20 right now. l 21 CHAIRMAN MICHELSON: Good. Thank you. l 22 MR. SCARBROUGH: The check valve plan was developed 23 at the same time as the motor-operated valve plan, and so it 24 has some very similar aspects to it, as you will see. 25 (Slide)
- l
-l 91~ l 1 MR. SCARBROUGH: And the. reasons for developing the 7-(_/ 2 plan were the same as for the MOV plan. The same process was i 1 3 performed for doing the documents, discussing problems with i 4 NRC personnel, and the plants themselves start off. describing-5 past problems, failures of locking devices such as the nuts l 1 6 and the pinch pins; the wear issue, excessive wear, caused by { l 7 various factors in the installation; sticking of the valves 8 and leaking of the valves. ) l 9 The causes as in, within MOVs were 10 varied--inadequate design, poor applications in sizing and 11 location, and categorization. Some valves that should have 12 been categorized as leakage testing type of valve required to (3 13 prevent leakage weren't categorized as such and did not (/ 14 undergo leak testing. l 15 The in-service testing criteria in Section 11 is not 16 very clear, and also is one of the causes of the problems that 17 we have had. There has been many events relating to check 18 valve failures. The most significant could be called the San 19 Onofre event in November of '85 where failures of check valves 20 in the main feedwater line resulted in some significant damage 21 to the secondary plant system, and it was caused by inadequate 22 sizing of the check valves, and the poor installation location 23 led to excessive wear. 24 Some of their analyses found that some of the discs 25 had been completely separated from the, from the hinge arm
92 ) ,em ; 1 itself,-but since they weren't used in back flow distribution, l 2 they never were discovered. 3 Clavert Cliffs had two events which were very 4 similar--failure of check valves in the steam line to turbine 5 auxiliary feedwater pumps. It could have led to simultaneous i 6 blowdown of both steam generators into the containment ) 7 building, and part of the problem was that they weren't tested 8 for reverse flow. i 9 Crystal River, in June of last year, we had a 10 failure of check valves once again in emergency steam / water 11 lines. There is back leakage of check valves, and this was a 12 continuing problem that they just couldn't get their hands on. } The Arkansas Nuclear 1 event in January of this 13 14 year, failures of a check valve in a high pressure injection 15 line, and resulted in radioactive water being circulated 16 outside the containment and back into another direct cooling 17 system and it was caused by excessive wear. It resulted in 18 the seat not becoming fully closed properly. 19 Commanche Peak was a different type of problem. 20 There they had improper assembly of the check valves. 21 They--the way the manufacturer had described how to assemble 22 the valves, they had a unique type of assembly arrangement, 23 and they hung up such that they wouldn't close, so that was a 24 very significant common mode failure for these types of check q %) 25 valves.
93 1 CHAIRMAN MICHELSON: That particular failure would f k 2 not have been detected by disassembly to see how things were L 3 working as I understand it? l 4 MR SCARBROUGH: Unless they disassembled it'and put l 5 it back the exact same way. l 6 CHAIRMAN MICHELSON: Worked until they reassembled 7 it, then it would cease to work again. 8 MR. SCARBROUGH: That is correct. 9 MR, SULLIVAN: They were assembled right from the 10 very start, so because the procedure that the licensee had was 11 inadequate. 12 CHAIRMAN MICHELSON: I thought some were assembled 13 right and some weren't. gs 14 MR. SULLIVAN: You may be right. I don't know just 15 how, why they-- 16 MR. CARROLL: I think they went through some kind of 17 rehabilitation program after the valves were installed and 18 they reassembled them, using the screwed-up procedure. 19 MR. WOHLD: Tom, question--I would like to Paow if 20 you can characterize what the open. if there is, how much 21 opening concern there is for check valves versus closing 22 concerns. I 23 In the ASME code, I guess 90 percent of the 24 philosophy is directed toward opening, testing for proper 25 opening, and about 10 percent for closing, but I think the l l I
.94 f) 1 ' opposite is true of the nature of the problem. Is opening a 2 problem? The only, for instance, the only opening I know of 3 is_where the disc totally falls loose and blocks the 4 discharge. Other than that-- 5 MR. SCARBROUGH: I haven't seen events. I have 6 reviewed--there is not much concern for opening problems. 7 These problems mostly relate to closing problems where we had 8 back leakage, back through the systems that weren't designed 9 for such. 10 MR. WOHLD: Thank you. -11 CHAIRMAN MICHELSON: There have been scenarios where 12 they failed to open more than one. I have read several LERs. () 13 In one case a bunch of clam shells got in the back. Removed 14 them. 15 DR. CATTON: There was another one. It; CHAIRMAN MICHELSON: Corroded up badly, wouldn't 17 open. 18 DR. CATTON: It rusted open 19 CHAIRMAN MICHELSON: He is calking about failure to 20 open. ] 21 DR. CATTON: These stayed open. 22 CHAIRMAN MICHELSON: There are lots of failures by 23 staying open. () 24 HR. WOHLD: Closed I think is more prevalent. 25 CHAIRMAN MICHELSON: Yes, failure to close.
95 1 DR. CATTON: Somehow having a check valve in your -I' 2 system that is rusted open just boggles the mind. 3 MR. WOHLD: There is not much requirement in the 4 ASME code to check for closure. I have got an example I can 5 bring up. 6 CHAIRMAN MICHELSON: One of the things that bothers 7 me from our discussion this morning, I'm not sure whether 8 failing to open is a design basis scenario. It may be it is 9 not a design basis scenario. In the past, rarely have you 10 ever done an analysis in which you postulate failure of a 11 check valve to open, so therefore, if it starts becoming a 12 problem, you are putting it in--I think you have got to do a g-13 backfit analysis. %)g 14 MR. BAER: I think Pete is right. I think many 15 licensees characterize it as a loophole in the code, but the 16 code says define the safety function, and if someone defines 17 the safety function as the pass flow, then the only test they 18 do is to make sure they can pass flow, so if it is stuck open, 1 19 it passes every time. l ) 20 CHAIRMAN MICHELSON: It is not a design basis l' l 21 scenario in that case perhaps to close. l 22 MR. BAER: That is correct. 23 CHAIRMAN MICHELSON: Therefore, if you want to 24 implement that as a scenario and put some requirements on 25 closure, you have got a backfit. l l l
96 [) 1 MR. WOHLD: The loophole in the code is I think more 2 ' specific. There is two tests you do. One is for a' stroke 3 test, and the other is a leak test, and characteristically, 4 the only check valves specified as Category A, which 5 identifies leak test requirement, are containment isolation 6 valves, so if the valve is not a containment isolation valve 7 and it's, it doesn't get any leak test, and then the normally 8 closed valves, the quote normally closed valves do not have a 9 stroke requirement because the code assumes well, they don't 10 need the stroke. 11 Unfortunately, that's not true in most cases, but if 12 the valve is in a stand-by situation normally closed and it is () 13 not a containment isolation valve, it gets no test at all. 14 MR. SCARBROUGH: That may change, though, as we will 15 see in my--the regulatory requirements are included. 16 MR. BAER: Tom, Frank wanted to-- 17 MR. CHENY: Let me make one more comment on that. 18 The OM-10 in-service testing document which has just been 19 referenced in Section 11 makes it very clear that if the check 20 valve has a safety function to close, that it must do a 21 closure verification test, and it is not, not a leak test like 22 we are talking about as Section 11 used to be, which was 23 somewhat ambiguous. OM-10 is a great improvement on that and (} 24 is very clear. 25 CHAIRMAN MICHELSON: Where it stated in your
97 1 licensing, approved licensing documents that as a requirement 2 for open or closed as the case may be. 3 MR. SULLIVAN: One way to handle this issue is.to 4 wait until they get into Generic Letter 89-04 because we did 5 try to specifically clear up the confusion that's in IWB on 6 this issue. All I can tell you now-- 7 CHAIRMAN MICHELSON: No, no. Let's just wait. 8 MR. SCARBROUGH: The regulatory requirements are 9 very similar to MOV requirements such as general requirements 10 for testing and proper design, and Appendix A and B, and then 11 section 505A, and the Appendix J, on leak testing steam 12 isolation valves. (} 13 (Slide) 14 MR. SCARBROUGH: The NRC activities on check valves 15 has had a long history. There is several branches that have 16 been involved in various check valves activities. The main 17 event started following San Onofre where the Commission issued 18 a staff requirements memorandum in February of '86 requiring l l 19 that performance of a generic analysis of check valve 20 failures. i 21 In April of that year, there was a meeting between l l 22 NRC and industry where this problem was discussed, and it was 23 indicated that the industry should work more diligently in l 24 resolving this problem. (%s) 25 In December of that year the Surry Unit 2 feedwater
i l 98 l 1 line rupture occurred. It didn't include, wasn't caused by .f, . (v)' 2 check valve failure, but the analyses found check valve 3 problems and it re-initiated the concern for check valve i 4 performance. 5 CHAIRMAN MICHELSON: I thought there were some check 6 valve failures, a check valve failure that line-- 7 MR. SCARBROUGH: There was a failure, but it wasn't 8 determined to be a significant contributor to the event. 9 CHAIRMAN MICHELSON: No. It didn't cause the 10 erosion or anything, or the break, but there was a failure of 11 a check valve, wasn't there? 12 MR. SCARBROUGH: Yes, sir, there was a check valve .,fm 13 failure. 14 As a result of that, the staff began to develop a 15 plan for inspections of plants with check valves. That was in 16 March of '87. Several SECY papers were written. They are 17 listed there, which inform the Commission of the results of i 18 some of those check valve inspections. 19 Some of the problems found were insufficient 20 preventive maintenance, and not all, all check valves were 21 tested, and all the safety functions were being tested in t 22 check valves, and an information notice was issued in August 23 of that year which described some of these results, indicated 24 that all safety-related check valves were weren't in this 25 service testing program. We weren't testing their ability to
99 [ l' p- . perform all their safety functions. 2 'Along came Generic Letter 89-04 in April of this 3 year which is addressed check valve back flow and forward' flow 4 disassembly issues, and Ted Sullivan will talk about that in a 5 few minutes. 6 The pressure isolation valve program also addressed 7 check valves in the sense of where they are a barrier between. 8 high pressure-low pressure systems. 9 Another problem that has come up concerning the 10 Anchor Darling valves with type 410 stainless steel bolts, a 11 Bulletin 89-0.2 was written to request that licensees inspect 12 this specific model Anchor Darling check valves and other 13 check valves which may contain this type of bolting material. 14 (Slide) 15 MR. SCARBROUGH: Research also has several 16 activities ongoing related to check valve performance. 17 Generic Issue 105 on interfacing systems LOCA, there has been 18 two NUREGs issued in this area that provide the results of i i 19 probabalistic risk assessments or core melt risk--NUREG/CR 20 5102 and 5124, once again, the valve performance program, and i l 21 Research also had some check valve activities related to it, j l 22 for example, evaluation of acoustic emission, to quantify 23 valve leakage, and also visual in-service inspection, the I 4 24 boroscope type of analysis. 25 The nuclear plant age research program also has a j l l l
100 f~) - 1 check valve aspect to it. Phase 1 that was completed (> '2 concerned failure modes and parameters. There was a NUREG/CR 3 4320, Volume 1, issued in December of 1985, which addressed 4 Phase 1, and Phase 2 is currently ongoing. NRR is evaluating 5 different monitoring techniques--acoustic, ultrasonic, 6 radiographic, and some others, and its report is due out early 7 next year. 8 AEOD has been involved with check valve activities 3 evaluating operating experience, and has participated in some 10 inspections of check valve activities at plants. 11 (Slide) 12 MR. SCARBROUGH: Other activities outside the NRC () 13 include the INPO efforts, SOER 86-3, issued in October of 14 1986, which addressed the San Onofre problems and other check 15 valves problems at various other plants, the need for training 16 relating to check valves, and preventive maintenance of check 17 valves. 18 They held a workshop in 1986 to discuss the current 19 problems of the San Onofre event and future plans for 20 improving the performance of check valves. 21 INPO has also been active in plant visits concerning l l 22 check valves to try to improve their performance. 23 EPRI also has activities ongoing. They prepared an 24 application guideline for check valves that was issued in l l 25 January of 1988, EPRI No. NP 5479, and it discusss various
101 [~h 1 c.spects of check valves--sizing, type, orientation, u) 2 performance: characteristics, check valve degradation, and 3 other problem areas. 1 i 4 The Nuclear Industry Check Valve group, NIC, was 5 formed in April of this year for an exchange of technical 6 information relating to check valves, and also it is our 7 understanding they are doing some analyses of various 8 non-intrusive techniques for evaluating the condition of check 9 valves. 10 DR. CATTON: They also plan to have an experimental 11 program to pursue. I'm sure they wouldn't appreciate you 12' having them under EPRI up there. l( ) 13 MR. SCARBROUGH: That is a--this slides was retyped 14 yesterday late afternoon and I did not get a chance to 15 proofread it. 16 DR. CATTON: They were very clear on the fact that 17 they were not part of EPRI. 18 MR. SCARBROUGH: Right. I apologize to them. My 19 master does not have that, has all of these over here. 20 MR. SULLIVAN: That wasn't our intent. You can say 21 the same thing about the ASME. I 22 DR. CATTON: I understand. l 23 MR. SCARBROUGH: I have to proofread my stuff. 24 DR. CATTO: At their meeting where they were {} 25 describing the testing program they are going to have, and it _ _ _ _ - _ - - - _ _ _ - _ - - _ - _ _ _ _ - _ _ _ ~ _ - - _ _ _
102 r" 1 became very clear that they wanted everybody to know that EPRI (.)) 1 2 involvement was only one of administrative and handling their 3 contract for them. They were separate and independent, i 4 MR. SCARBROUGH: You are perfectly correct. I will 5 get Al Igne a corrected copy which is in tune with the one I'm 6 reading from. 7 The other diagnostic technique improvements, that 8 they are involved with the Checkmate program, by MOVATS and 9 radiography technique developed by Schomberg. 10 Design improvements, we are looking at integral arm 11 and disc to try to eliminate some of the locking devices, and 12 the ASME OH-10 document has in-service testing criteria for () 13 check valves in it. 14 (Slide) 15 MR. SCARBROUGH: The proposed action plan for check 16 valves is listed in the same manner as the MOE plan. Some of a 17 the technologies are the same and some are quite different. 1 [ 18 Mechanical Engineering Branch has the coordination l 19 and responsibility for the activity, starting off, check valve l 20 inspections, performed as resources allow, and coordinate 1 21 inspections by other NRC units, review research program 22 results, and AEOD records and interacts with industry on check 23 valve issues. 24 We also hope to have coordination meetings on a 25 quarterly basis. We may try to combine these with MOV l
103 - ('} 1 coordination meetings to be more efficient in our resources. \\s 2 The revision of the IC regulation ongoing for the 3 MOVs will encompass the same type of issues as the check '4 valves. Revise it, regulations clarify the scope, to the 5 ' endorsement of OM-10 with any necessary exceptions, and to 6 make clear that it is, in-service testing is intended to 7 demonstrate the performance of all safety functions. 8 Also we are going to-- 9 CHAIRMAN MICHELSON: This question of safety 10 functions, is that going to be thought about somewhere else in 11 view of our discussion this morning? What is the safety 12 function of a check valve? () 13 Normally I think only in the closing directions is 14 it considered to perform a safety function,.but clearly it 15 also implicitly must open and overflow, but it is there in 16 case something goes wrong and it has to check against the 17 reverse flow, so what is the safety function? 18 MR. SCARBROUGH: If the check valve has to open to 19 allow flow of emergency core cooling water or some water to 20 the steam generators, then it has a safety function to make I 21 sure it is open and they will have to make sure that they can 22 perform that function. 23 CHAIRMAN MICHELSON: That is in the original design 24 basis. Do you think the check valves have to open as well as 25 close?
1 L. '.I i b 104' 4 1 MR. SULLIVAN: Yes. -2~ MR..SCARBROUGH: If'it is'in:the line and they ^ 3 assume that it' opens. 4 CHAIRMAN MICHELSON: ' Implicitly. clearly it has to i 15 open. I don't argue that. 6-MR' BAER: 'Again, I think'if you think of' design ' j 1 7 basis in terms of'the basic scenario that the system is trying. ) i 8 to mitigate,Rit is in the design basis. 9 CHAIRMAN MICHELSON: It is probably there to check 10 closed against some reverse- - 11-MR. BAER: I mean if it is in the' aux feed system 12 and the function is to supply water to steam generator, that (])- 13 valve has.to open.
- ~
14 CHAIRMAN MICHELSON: That is true of all valves. 15 Isolation valves have to open, too, as well as close nicely 16 I'm sure. 17 MR. BAER: I don't think there.has ever been a 18 question that when they test check valves, they test to make 19 sure they. pass flow. 20 CHAIRMAN MICHELSON: No. I think that was the 21 question raised just a little while ago, and you don't test to 22 see if they open. You test to see if they close. 23 MR. SULLIVAN: I think what Pete was saying was 24 there tends to be more emphasis on the open, testing for 25 opening than testing for closing-.
i I 105 i 1 i rx ( J' 1 MR. BAER: That's my impression also. 1 I 2. CHAIRMAN MICHELSON: The point is if the disc is 3 missing, it tests open fine. 4 MR. SULLIVAN: Right. 5 CHAIRMAN MICHELSON: If it wasn't a safety function, 6 they have to be leak tight, then they never test it to see if 7 it closed. 8 MR. SULLIVAN: The reason the confusion has come up 9 is because IWP refers to valves that are normally open and 10-valves that are normally closed, and when it talks about 11 valves that are normally closed, it says the test you should 12 do is test it to make sure it will open, and vice-versa. () 13 What the code implies is that about a dozen have 14 safety function in both directions. It implies that the valve 15 is not bimodal and so people have tended to put more emphasis 16 on reading that code literally. If the valve was normally 17 closed, they would just test it to make sure it opened. 18 What we tried to do in the generic letter was step l 19 back from that Generic Letter 89-04 that is, and say that the l l 20 code requires, at least our interpretation is that the code 21 requires that the check valve be exercised to the positions in 22 which they perform safety functions. We were saying that the i l l 23 valve is probably bimodal, do both tests, and we have also () 24 assumed in our implementation of IST that that is part of the 25 design basis. We didn't question that. I
1
- 1 106 1~
' CHAIRMAN MICHELSON: You don't have anywhere where ); 2 it is explicitly stated?L '3 MR, SULLIVAN: No. I think where it comes from is ,4 the way the system is used. 5 . CHAIRMAN MICHELSON: Is that considered a backfit?. i 6 Go back and require that test?. 7 MR, SULLIVAN: We--this 89-04 is a backfit, but 8' probably from a different point of' view than you~mean. 9 ' CHAIRMAN MICHELSON: It is considered a'backfit. 10 MR. SULLIVAN: It is a backfit, but from the point 11 of view that we had never issued generic regulatory guidance 12 on this subject,-is what we also called a compliance issue, so (]) 13 it is a hybrid type of backfit issue. 14 CHAIRMAN MICHELSON: Why isn't 89-10 the same 15 ' category? 16 MR, SULLIVAN: I can't answer that. 17 CHAIRMAN MICHELSON: It was considered a backfit. 18 MR. KIESSEL: Woe. Woe--89-10 was considered a 19 backfit, and one of the driving elements was the inclusion of 20 position changeable valves, but Owen will point out that we-- 21 CHAIRMAN MICHELSON: That is part of the argument on 22 why you didn't want to, you know, go back and pick up these H23 other scenarios was that you didn't want to consider that a 24 backfit. 25 MR. KIESSEL: No, sir, because this bulletin or this
107 .gf( j 1 generic. letter was limited to the valves. It was not intended 2 to go beyond the valve act itself, and the backfit then would 3 have to be applied to the scenario which was not addressed in 4 any way in the approach to the generic letter. 5 CHAIRMAN MICHELSON: The valve that is not on the 6 . scenario, that valve has to be? 7 MR. KIESSEL: Correct. 8 CHAIRMAN MICHELSON: I would argue that closing or 9 opening of a check valve might be a scenario that was never 10 included originally and now we are saying it has to be met. 11 MR. SULLIVAN: You could probably get the same 12 issue. /'_T () 13 CHAIRMAN MICHELSON: Sure. 14 MR. SCARBROUGH: Okay The next item there is you 15 also look at regulatory guidance in other areas--available 16 diagnostics, IST program, training, various categorization 17 areas. 18 We also evaluate the need for improvements to 19 guidance for NRC personnel. 20 Another weakness was inspection manual chapters. 21 There is not a good manual for check valve inspection. l 22 We also assist the industry in resolving check valve 23 concern. We will have meetings with industry, and review (} 24 guidelines that they propose. 25 We also hope to have sample reviews of INPO plant
108' .1 ' visits to determine where their strengths are. u 2 We alsofevaluate industry efforts as part of our p -3 ' regulatory mission to determine where they are performing 4. resolutions of check valve problems and where1that resolution- ~ L is not-proceeding as quickly'as necessary.. 5 6 We-will also assist in revision'of tech specs-7 standard or plant tech specs as necessary. Also the NRC Valve 8. Review Group will also take up the issues of check valves in 9 .that group, and we plan to have that to be a very active group: 10 within the-Agency. 11 We will also participate in the need, participate in 12 the development of codes and standards and emphasize the need i.. 13 for in-service testing standard for check valves, and our 14- . understanding is that need is.now recognized and there.is a-15
- move to proceed in that direction.
16-We will.also follow the improvements in design and 17 testing check valves, and valve' experience, determine the need 18 for revisions to the NRC efforts; and finally, to prepare a 19 Commission paper on a periodic basis to provide an update of i 20 the status of our activities to the Commission, and as I said, 21 for MOVs our first Commission paper is due this month and we 22 are in the process of developing that at this time. 23 And that's basically the two programs. I will be 24 glad to answer any questions about them if I can. 25 MR. WOHLD: A quick quest' ion related to the changes --_m
109 q i i l 1~T 1 that.are~in plant such as in OM-10. Once OM-10 goes in, that G change will be related I presume in new IST programs. For 2 l; 3 some people that had just revised.their program at the time l: 4 IST comes out, OM-10, it is going to be ten years before they 5 update their programs to handle these other check valves? 6 MR. SULLIVAN: Pete, I think that depends on how 7 this rulemaking that Tom talked about earlier is framed, and I 8 don't think we are in a position to answer the question one 9 way or the other. 10 If it continued, if the new rule carried forward the 11 current approach, then it would be some time before it got 12 picked up. If the rule is more aggresive than that, then it (') 13 might be sooner. 14 MR. SCARBROUGH: Thank you. 15 CHAIRMAN MICHELSON: I believe we are now ready for 16 the final speaker on in-service assessment, Mr. Sullivan. 17 MR. SULLIVAN: A few minutes ago I said that I would 18 talk some more about check valves, and the relationship to the 19 code or codes when I made my pitch on 89-04, but I think I 20 just did try to address that when it came up again, so my 21 intent is not to go back into that. 22 As I understand the reason now we were asked to come 23 down to talk about 89-04 is basically because we hadn't really /~% 24 done that before issuing the letter, and the treatment or V 25 coverage of it in June was a little too brief, so my intent is
l i 110 . [~} 1 to quickly go through the generic letter, at least in outline v 2 ' form. I'm not going to assume too much, but I am going to try 3-and go through it in a kind of a broad fashion, i 4 (Slide) ) 5 CHAIRMAN MICHELSON: I think I would like to give 6 some guidance to the Subcommittee only. We did not write a 1 7 letter on 89-04. I think one of our determinations today has 8 to be whether or not we should even write a letter as to the 9 content. 10 MR. CARROLL: Was that, the pending issuance of a 11 letter, brought to our attention? We said no, we don't want 12 it? i () 13 CHAIRMAN MICHELSON: No, It just came out without 14 our even knowing it happened. It was never brought to our 15 attention. Copies were not even sent to us as I understand 16 it. It just suddenly appeared. In fact, I think we heard 17 about it at a Subcommittee meeting and asked about it and 18 that's how we found out that this happened. 19 MR. SULLIVAN: I am a little bit surprised about one 20 statement you made. 21 Aren't you on the distribution for all reg guides? 22 CHAIRMAN MICHELSON: We are now. We weren't then. 23 This one came along, a cou;,le others came out at the same time 24 that we didn't get, either. We suddenly realized we weren't \\ 25 getting them, and so that's, it is straightened out now. I
.m '111 1 think it will'not happen.again. s}p j 2. MR. SULLIVAN: When we made the distribution'for 3 that', we'were told there were 1500 people in' distribution for ~ -4 .all generic' letters and'information notices and so forth, so 15-we assumed that it covered the world. 6 CHAIRMAN MICHELSON: Not quite. 1' 7 MR. CARROLL: There are more people in the world i 8 than 1500:by last count! 9 MR. SULLIVAN: Briefly, sort of by way of background H 10 on the generic. letter, prior to the issuance of the generic \\. 11 letter, the NRC had never sent out any written generic l 12 technical guidance on IST, and the only technical guidance if l }}l 13 .you will'that was issued by NRC was technical guidance by way i 14 of holding meetings with licensees or by way of writing up 15 issues in SER, but that's plant-specific, not generic, so we 16 felt a need to do something about that. That's a part of the 1 17 reason we sent the letter out. l 18 We also felt a need to try and do something to l 19 straighten out the huge backlog of program, IST program 20 revisions, and the relief requests. At the time the generic 21 letter was issued, there were only 14 plants covering 23 22 reactors that had up-to-date SERs. l i 23 Of course, we had sent out more SERs than that, but 24 they waren't up to date because IST program revisions had come ) l l 25 in, staff hadn't had a chance to look at them in detail, and
l 112 i and so the SERs were gradually becoming more r 1 < evaluate them, 2 and more out of date. 3 I think what I just went back to--I'm not sure how 4 well it is understood--is that in addition to the ten-year 5 updates that the regulations require, the utilities frequently 1 6 send in--not all of them, but a couple will send in revisions 7 every couple of years, or even more frequently than that, and 8 so there is a huge deluge of paper that comes in under IST, 9 and the staff really wasn't able to keep up with it. 1 10 There is also sort of a conflict that is generated l 11 by that, and that's because of the tech spec that exists.in 12 about 85 percent of the plants. In the standard review plan, l () 13 it is tech spec 405, and that tech spec says follow the ASME 14 code except where written relief is granted, and what that 15 means is that the relief request really shouldn't be 16 implemented unless the staff has evaluated it, and since we 17 weren't able to keep up with that workload, there was a 18 conflict. Utilities basically were having to proceed with 19 the, along the principle that since we hadn't responded, then 20 NRC must not have a big problem with what was being sent in. 21 And also akin to that is the fact that since we 22 weren't able to keep up with all of this work, when the 23 inspectors went out to inspect against a program, they didn't 24 have the benefit of what NRR thought of various relief 25 requests.
p V 113 ) -1 CHAIRMAN MICHELSON: Could you give us an. idea of 2' what the-time delay we are talking about-in failing to respond to the licensee? 4 MR. SULLIVAN: I'm not sure what you mean by time 51 ' delay. .6-CHAIRMAN MICHELSON: How long? They send in-and you 7' Lwere supposed to respond. How many days or weeks and months 8' went by and you didn't respond, or maybe years? 9 MR. SULLIVAN: Your question I think implies that 10-there was some time that we were always behind, but I think-- 11 CHAIRMAN MICHELSON: No, no. I am trying to find '12 out, you'said that the licensee kind of worked on the ' ( [- 13 assumption that since you didn't say anything, it must be all 14 right, and how long did he wait before he decided it was all 15 right? 16 MR. SULLIVAN: Well, I think that licensees never 17 waited. They basically wrote programs, sent them into NRC, 18 wrote procedures against those programs a,nd put them right-- 19 CHAIRMAN MICHELSON: You weren't holding them up at 20 all? l 21 MR. SULLIVAN: No. 22 CHAIRMAN MICHELSON: They just were uncertain as to 23 whether you were happy with them? (} 24 MR. SULLIVAN: That is correct. 25 MR. WOHLD: They were basically in violation of
114 r 1 NRC's requirement, but not because of their own fault. It was tx 2 a difficult inspection problem. 3 CHAIRMAN MICHELSON: I gather this went on for 4 perhaps years, not weeks or months? 5 MR. WOHLD: Some plants submitted plan year programs 6 that have come tc pass and still haven't had anything. 7 CHAIRMAN MICHELSON: Ten years went by? 8 MR. SULLIVAN: Sure. There are some plants that 9 have never received an SER. Most plants have at least had, I 10 would say 95 percent of the plants have at least had the 11 benefit of NRC holding a fairly lengthy, probably in the order 12 of two-day meeting, usually at the site, although not always, (} 13 and during these meetings, they would get, at least verbally 14 get the benefits of our reaction to their programs. 15 Sometimes that didn't really do much good because if l 16 IST is relegated to a newly hired engineer, and there weren't 17 any major problems, he would maybe get promoted to something l 18 different, and he would have new people coming in, so whatever l l 19 we pass on to the utility was frequently lost by change of 1 20 personnel. 21 CHAIRMAN MICHELSON: Is it safe to assume that the 22 NRC did review these early on enough to determine they didn't 1 23 have any real problem with it, or is it, you know, you don't 24 want to let something like that go on for ten years without 25 even looking at it to see if you have got a problem with it or .-~~.-_- - _ -
VQ ~ ~ ~ j 'l l+ G= 2 n ~ 'I v 115 -l i 1 q[, c 1 .you clearly!didn't have'enough time to respond, at least I: 2 assume you had enough time to-check it,Jmake sure you had no ~ -3 . problem? .) -] 1 4 ~MR. SULLIVAN: Sometimes we didn't have enough time 5 toldor that, either. 6 CHAIRMAN MICHELSON: That's kind of a dangerous mode c; 7 of operation,~isn't it? .8 MR. SULLIVAN: Well, that's part of why we sent out j ) l f 9 the generic letter, to try to solve some of that problem. l j-j .s 10 -MR.-WOHLD: I guess one o'f the feelfngs I had : on the. .l 11-program, the' law I don't think requires that NRC review these 12
- entire programs.
It requires the review of the relief I I 1 ) 13 requests, but NRC took it upon itself to do the entire 14 program, so I think that held down the program with' time- .15 limits. ] i 16 MR. SULLIVAN: I think that's true, but we felt that i l 17 further review of the. programs, we were finding enough ] i 18 problems that we thought it was good to do the whole job i i 19 instead of a partial job. 20 In hindsite, we may have been oble to keep up with 21 relief requests if we had just reviewed tnem. I'm not sure if 22-the industry would be any better off or worse off. I don't i 23 know. .(} 24 MR. WOHLD: I agree. 25 MR. SULLIVAN: I have listed a couple et purposes
ll 4 116' i 1 f 1 there for the generic letter, but they really reflect what I' I 2 was showing before in the background. 3 We were' attempting through the generic letter to L 4' provide guidance on some areas that we felt were consistent 1 J 1 5 problem areas in programs. l 6 We were providing a mechanism or are providing a l 7 mechanism for licensees to get their IST programs or their-8 relief requests. approved without, requiring a detailed review 9 on the part of staff, recognizing that we weren't able to keep i i '10 up with them, and.we.were trying to clarify the approval 11 status for IST programs. By that what I mean is we were 12 trying to convey the message that tech spec 405 is something' l L() 13 that should be met, and we want to.get utilities in the i 14 framework that this isn't going to go on being informally-15 waived indefinitely. 16 The eleven areas that I referred to in the first 17 bullet basically fall into three different types of areas. 18 They fall into interpretations of the code, and an example of 19 that is the one we' talked about previously where we clarified 20 that the code doesn't mean that just check one function of the 21 valve if the valve has two safety functions. In other words, 22 we clarify that a valve isn't just to be stroked open if it 23 has a safety function for closure, so that would be an example 24 of a code interpretation. q%J' 25 We have two positions on tech spec interpretations.
1 117 (} 'l An example of that is we wanted to clarify when the-actions 2 statement should start for a pump or valve that is declared 3 inoperable. 4 There is a little bit of confusion that 5 arises--maybe not a little bit--a lot of confusion that arises 6 from code language that seems to, to give a licensee several 7 days after a, a component fails an IST test before they 8 declare the component inoperative, and our position in the 9 generic letter was to say if the component is a tech spec 10 component and'it is inoperable, you declare it inoperable when 11 you know it is inoperable, not several days later. 12 And then we have six different positions. Some of () 13 these overlap. They do more than one thing. We have six 14 positions where we have given alternatives to code 15 requirements, and our approach there is to say if a utility 16 follows these alternatives, your reliefs are automatically 17 approved. You don't need NRC TO review anymore, so that was 18 put in there to cut down on some of our effort, to remove some 19 uncertainty on what NRC views were in these technical areas. 20 The generic letter gets a little bit complicated because all 21 plants aren't in exactly the same situation relative to 22 review. 23 (Slide) (} 24 MR. SULLIVAN: So what I've shown there are in the l 25 generic letter we grouped plants into three different types of i
li. [ 118 l { 1 ~ plants.- We called them Table 1 plants, Table 2 plants, and 2 plants not on'either table. This~ slide covers the first two 3 groups. 4 -Table 1 plants are plants where NRC had~ completed a' 5 significant amount of_the. review, the SER:was in writing, and i 6 we felt we didn't want to throw that effort away, so we list'ed 7 those plants so they would know who they were, and we told-8 them that we would be getting.back to them in an SER'as to-9 exactly what they should be doing, but'we also told them that 10 all the positions in the generic letter don't strictly deal 11 with program areas. Some of them dealt with procedure areas', 12 and that they ought to look at the letter from that point of '(]J 13 view. 14 Table 2 plants were, I think there is 14 of them 15 where we said that the, the SER was up to date. We had done 16 an SER against the program that was currently submitted, and 17 that constituted the approval of their relief requests, but 18 again, there were procedural areas that they should make sure 19 that the procedural areas in the generic letter were reviewed 20 and corrected if necessary. 21 (Slide) 22 MR. SULLIVAN: Then we have probably the bulk of the 23-plants, about half the plants, that aren't on either table, 24 and what we said in the generic letter was that the generic 25 letter itself would constitute approval of the relief
i 119 (( ) 1 requests, and essentially the program if they would take the 2 attached positions and make sure they met them. 3 And in doing that, we were saying that all those 4 relief requests that didn't fall within the attached positions 5 were essentially grandfathered. 6 Now there is a couple of rationales behind why we 7 were doing that. One of them is, as I was snying previously, 8 that silence implies approval, and we had been silent on those 9 relief requests. There was some question about whether we 10 would ever have gotten around to doing them all, to change 11 that quote, silence, and the third rationale was that we felt 12 that by being specific on those eleven areas that were () 13 attached to the letter, we were going after the most 14 significant issues, so that's basically the rationale for 15 saying that the relief requests that weren't in the generic 16 letter, that weren't covered by the positions in the generic 17 letter, were essentially grandfathered or approved anyway. 18 What we also asked licensees to do was write in and 19 tell us how they were reacting to the generic letter. In 20 other words, let us know that they were going to conform to 21 the attached positions. Let us know if they had any new 22 relief requests that weren't previously docketed that came 23 about because they started looking at the program more {} 24 closely. 25 We said if they did, we would look at them, and i
e E 120 ' () 1 that's sort of an area where we can't make a generic letter 2 supersede regulations. The regulations are there. They say i { 3 that we are supposed to review relief requests,-that licensees 4 have an obligation to send them in, and so if new relief 5 requests came about as a part this whole process, we would 6 review them. 7 And then we essentially have the same approach down 8 the road. We tried to write the generic letter so that it 9 wasn't just something that would fade away. We said that for 10 the future, if revisions come in or updates come in, and they 11 have relief requests that are covered by the Attachment 1 12 areas, that is the generic letter positions themselves, then () 13 those positions continue to apply. The alternatives that are 14 approved in those positions would continue to be viable, and 15 relief requests that were new that fall outside the generic 16 letter we would review, so that's what we are saying up there 17 on the top. 18 We also advised people in the generic letter that we 19 would be conducting inspections, and that we would, we would 20 mainly focus on the generic letter, but other areas that 21 didn't look right, we would also look into them as well. 22 (Slide) 23 MR. SULLIVAN: And then I have a little bit of {} 24 information that you might call status on the generic letter 25 since we have sent it out.
l t ; p 121 ( 11 We sent this out in April 3rdt on April 3rd of;this ~ fyear. In June, we held workshops or road shows similar to the 2 3 ones-that we referred to earlier on Generic Letter 89-10. We i. '4 went to:four-of five regions,: held-meetings at-airport motels 5 and hotels. The format:for the' meetings were basically the .6 -same as what we did for 89-10. We basically gave some short 7. presentations on the generic letter, and then opened the 1 8-meeting'up'for questions from attendees. '9 We gave responses. We have since put together a 10 meeting summary in the form, in a QA format. Unfortunately, 11 we haven't been able to' send out the summary yet because it 12 essentially constitutes generic guidance. It was not possible ( 13 to limit the meeting summary to simply areas very explicitly 14 covered by the generic letter. We got into all sorts of 15 questions on IST. By writing-answers, we are giving generic 16 guidance, and our management decided that it was important for 17 this document to be sent to CRGR, so right now we are 18 scheduled to go to CRGR to brief them on the meeting summary 19 next week on October lith, and we anticipate being able to 20 send it out shortly thereafter. { 21 There have been eight SERs issued since April. 22 These are just SERs that we have issued on Table 1 plants, and 23 we anticipate or our current schedule is that we will issue (} 24 the remainder of the SERs by the end of this calendar year. 25 That's the end of my presentation. I you have any
122 () 1 questions, I'll be glad to respond. 2 CHAIRMAN MICHELSON: Any questons? 3 MR. WOHLD: Yes. I've a few questions. If a 4 licensee uses this generic letter to get an approved program, j l 5 then all of these provisions in the attachment apply a hundred 6 percent? Is what you are saying? 7 MR. SULLIVAN: Well, if--suppose a licensee writes 1 8 in and he says I had a program in here before April 3rd, and 9 it had 50 relief requests. Okay. He may say that 20 of them 10 are not covered by the Attachment 1 positions, and I'm not 11 . changing them. I like them, and I'm not going to change them, 12 so those 20 are basically grandfathered. () 13 The other 30 he may say they are in areas covered by 14. He may say that 28 of them go along with your 15 attachment, but two of them I just can't follow your 16 attachment. I can't, you know, it doesn't really fit my plant 17 well, so he sends in two more relief requests. Then we will 18 have to review them. 19 Basically he will be in a position where 48 of the 20 50 would be approved, and then the whole 50 of them would be 21 approved pending our disposition of those two relief requests 22 in my example. 23 Does that answer your question, Peter? You have (} 24 something else in mind? 25 MR. WOHLD: I'm still confused as I work through
123 () l' this thing. 2 If I read your attachment one here, basically it 3 covers the back flow testing of check valves, that the code 4 now has a loophole in it? 5 MR. SULLIVAN: Right. We really tried to clarify 6 what we, how we interpret the code. 7 MR. WOHLD: I'm just wondering what the regulatory 8 things are therein. Is it a nice thing that they ought to do? 9 What if an inspector finds out they are not testing some 10 valves? Are these not considered inoperable under tech specs? 11 You know-- 12 MR. SULLIVAN: I think even before the generic (') 13 letter was sent out, I note from one of my experiences going i 14 to a plant, we found check valves in steam lines to the 15 auxiliary feedwater turbines were valved up. It was an old 16 plant that had an external operator on it, and it was, it was 17 very stiff, and we challenged the operability of that valve. 18 I don't think that the generic letter needed to be 19 issued for the inspector to challenge that particular problem. 20 The utility recognized that that valve needed to be operable 21 in order to, you know, avoid blowing down a couple of steam 22 generators in one possible scenario. I 23 Usually our experience has been that the code is 24 misleading, but when you talk to a utility about the fact that {} 25 a check valve has a safety-related function in both {
L ' L 124 ..:(} 1 directions, my experience has been that they have been 1: 2 cooperative in recognizing that and revising their program to ) 3 account for it. I don't know if your experience was 1 4 different. 5 MR. WOHLD: My experience was the opposite, you 1 6 know, small number of experiences. I had a document, an 7 inspection report sometime ago, and I don't want to mention 8 the plant name particularly, but they took a position, and 9 this was in a pre-op test phase, they took a position they 10 were not going to pre-op test these check valves for closure 11' because the code didn't require any closure testing on down to 12 the pipe, so why start out with testing when you don't have to () 13 do it later on? And they were pretty adamant about not 14 testing these valves, and it took about a year and a hall and 15 they finally still adamantly would not put these in a code 16 program, but they documentede in their procedures they said 17 this is the quote, Pete Wohld's list of check valves to test. 18 I've got a list of the valves that I went through 19 when I was at the site at that time, and we find there was 225 20 safety-related check valves, and very interestingly, the code 21 required closure testing of only 10 out of 225 check valves. 22 The way I analyzed this, there were 31 that were tested 23 because of the high-low pressure interface, but the code 24 doesn't require testing of valves that prevent an accident, ) 25 quite strange strangely. The accident prevention is not part -__--___- _-_-__--- _ _ - ~ _ _ - - - - -
125 l ) 1 of the scope in the valve program, so the pressure isolation 2" valves, 31 of those were'added by NRC's WASH 1400 study and so 3 forth, but less than, what is it, one out of 14 were checked j 4 according to the code, and that was it, and they refused. l 5 They understood the technical aspect of it, but they were 6 saying hey, Mr. Inspector, you make your requirements right 7 and we will do the testing according to your requirements, but 8 they flat refused to put those in the code program. 9 DR. SIESS: They recognized the importance to safety. 10 of this? But they refused to do it until they were ordered by-11 the NRC? 12 MR. WOHLD: They refused to put it in their quote, () 13 official code program. 14 DR. SIESS: Is that different from not doing it? 15 MR. CARROLL: Probably. 16 MR. WYLIE: Probably. 17 MR. WOHLD: They, with a lot of jawboning on my part 18 perhaps, they agreed to put it in their program, but they 19 refused to put it in their ASME code program. 20 DR. SIESS: That's an interesting point. I am just 21 wondering now nhy the industry that we repeatedly tell are 22 responsible for safety, the Commission says this at least once 23 a month, and every once in a while ACRS says it, are 24 they--that they have responsibility--that they would refuse to {} 25 do something that apparently is important to safety, or is it
,i u 126 ( .~ 1 simply that they.are willing to do'-it but they don't want to 2 do it'according'to some-rules that will get'them in trouble 3-every time they put a comment around the place? What is the p 4 . reason for this? 5 MR. SULLIVAN: Sometimes-- 6' DR. SIESS: Not doing it or-- 71 MR. SULLIVAN: Sometimes what they are trying to.do,.
- 8' at l' east 5in Pete's example, is they are trying to put it in a-9 different program that is notecovered by the regs.
It is a 10 program that they administer, and that way they feel that they 11" are going to be able to avoid enforcement problems. 12 DR. SIESS: That's good for them, but it is bad for. O 13 'tae Due11c7 14 MR. SULLIVAN: Just good for them. I don't know 15 whether it is bad for the public or not. 16 DR. SIESS: Well,'I-mean what are we interested in? 17 What is good for the utility, what is good for NRC, or what is 18 good for the public? 19 CHAIRMAN MICHELSON: You are interested in making 20 sure that it is done if it needs to be done, and if it is not 21 in their program, it is not enforceable. It may or may not be 22 done. 23 DR. SIESS: Is that saying the plants aren't safe 24 unless the NRC is over standing there watching every one of [ 25 them?
127 " /^)\\ - 1 CHAIRMAN MICHELSON: 'The industry is saying that'it-A~ 2 isn't in the, if you think it is necessary, put:itLin the 3 regulations. If it isn't in the regulations-- '4' DR. SIESS: That isn't what I heard. I heard.that. ) 5 the industry says that if~you'think it is.necessary for you to 'l 6 inspect it, then you put it in the regulations, not that if-i 7 you.think it is necessary that it be done, because I heard J 8 they were doing it'. They;just weren't.doing it under an NRC i 9 enforceable program, j q 1 10 CHAIRMAN MICHELSON: Nor are we assured that they 11 will continue to do'it. 12 DR. SIESS: Then this is only one of probably 10,000 13 things at a nuclear power plant that NRC cannot be assured is l l() 14 being done. 15 CHAIRMAN MICHELSON: But they aren't surely 16 important for us to have regulations governing it. 17 DR. SIESS: The NRC keeps saying that the utility is 18 supposed to be responsible for safety. The NRC really doesn't 19 believe that? 1 20 MR. BAER: Well, Chet, you seem to be looking at us. 21 There is an industry representative over there from NUMARC. 22 You might look at him as you talk. 23 MR. CHENY: Can I try that on first? L 24 DR. SIESS: I got an answer from you that they are 25 willing to do it. They just didn't want it to be a burden. l
J 128 [G~l 1 MR. BAER: Not from me--maybe from Pete. 2 DR. SIESS: I got it from somebody. 3 MR. CHENY: I think by not putting it in their ASME 4 code, they avoid the language that Ted was talking about 5. earlier, that is, 85 percent of the tech specs they test the 6 valve under non-go program and it didn't work right. I don't l 7 think that they are required by anything that is legally 8 enforceable to declare inoperable per tech specs. Okay. And 9 shut the plant down and declare the system inoperable or 10 whatever is related, to take action'on it. They can take it 11 in their own good scheduled time I think is the way it is 12 handled. () 13 DR. SIESS: I thought he made the point under ASME 14 codes they were allowed more time before they declared them 15 inoperable? 16 MR. CHENY: That's the difference. There is a delta 17 difference of opinion in time. What I am talking about is 18 they set their own schedule, and it is not enforceable by us 19 to take any action on it if it is not in the ASME code 20 program. 21 MR. BAER: Take Pete's example of a valve that 22 separates a high pressure and low pressure interface, and one 23 of those two valves is found by the licensee to be inoperable. ) Under their own program, they could go for months without 24 1 25 finding it.
5' l 4 l 1129 I ({ 1 The thing:that Ted was talking'about I think'is the k -2 difference between doing the analysis-in'a couple of days,.and l '3 then declaring it inoperable, or declaring it inoperable 4 immediately, but I think it is a couple of days delta, right.- i 5 MR. SULLIVAN: Right. .j 15 DR. SIESS: That does make'a difference in risk. L-7 MR. BAER: The couple of days, I personally wouldn't
- 1..
l 8 worry about. -The months I would. 9 MR. CARROLL: Even though there is one valve that is a 10 closed,-and leak. tight, and you are not doing anything'to 'I 11 disturb it? q 12 MR. BAER: Hopefully you are not doing:anything to () 13 disturb it, t 14. DR. SIESS: As I recall the-- j 1 15 MR. BAER: Not directly. l 16 DR. SIESS: As I recall from the SEP program, there j 1 17 were a number of plants that did not have two isolation valves j 18 on some pipes. 19 MR. ROTHBERG: You are talking about the reactor I 20 cooling pressure boundary I believe. 21 DR. SIESS: I'm talking about containment isolation 1 22 valves in this case, which everything is a real big issue with 23 us because if containment isolation valves don't work, you get { l 24. off-site dose that exceeds Part 100, and that's, of course, I (} 25 bad.
130 V(D-1 It turned out when you-looked at it from a risk 2 point of view, not from a design basis accident point of view, 3 it didn't make any difference whether it was two isolation 4 valves. The probability the pipe would fail on the wrong side 5 of either one of them, this is outside containmen*. is so much 6 greater than the probability of the valves wouldn t work, so 7 are we talking about risk, or now are we talking about rules? 8 MR. BAER: The example I was citing before was what 9 I was, I thought Pete was talking about was not 10 necessarily--it may be a containment isolation valve, but if 11 the valves fail, both of them failed, you would now 12 potentially have a LOCA that bypassed containment. That's the () 13 sort of valve that I would worry about, and I would worry 14 about if they only had one for an extended period of time, 15 because they could be other. 16 DR. SIESS: Event V valves according to the 89-04, 17 it is a category listed in 89-04. 18 MR. BAER: I'm always troubled by that nomenclature. 19 To some extent, any high pressure-low pressure valve is an 20 event V valve. To other people, the event V valves are those 21 that were specifically ordered by the NRC. For licensees to 22 take actions, I'm not sure which definition they use. 23 DR. SIESF: That's the definition I think was used, 24 somebody used in 89-04. {} 25 MR. BAER: The second one.
V' i 131' g Fl' .(); 1 DR. SIESS: I.think. 2 MR. BAER:. Only ones theyLwere ordered to treat. { '3 DR. SIESS: Whatever.NRC called them is1what they 4 are.' That's the easiestLway.to'take_ care of it. 5: CHAIRMAN MICHELSON: 'Does that take care ~of'any of 6' the other-questions?' 7 MR. CARROLL: I' guess.I would'like to see the' ~ 8 meeting minutes and the' interpretations'after CRGR'gets done R 9 if you can arrange that, A1. I don't know if you can. 10 MR. SULLIVAN: We will.make sure.you are on 11' distribution. 1 L 12 MR. CARROLL: Well, the minutes after they get l-([ 13 finalized. 14 MR. SULLIVAN: We will make sure you are on' 15 distribution, or A1. l l 16-MR. CARROLL: These are minutes of the workshops 17 that were held that have to go through CRGR before they are l 18 released. 19 MR. SULLIVAN: And when we do the minutes for 89-10, it 20-we will likewise send them to you. 21 CHAIRMAN MICHELSON: Any other questions? What I 22 would like to do is close, to terminate the meeting now, but 1 L 23 have a Subcommittee discussion which need not be on the ) ~ 24 record, so the record is closed. 25 (Whereupon, at 12:05 p.m., the recorded portion of
_7 j'.- >> i:
- / -
- 7. (
1. the meeting'was. adjourned.) '2- , c . 4-5- 6 7 s. 8-9 10 t. 11 12 0 13 14-15 16 17 18 19 20 21 22 23 O' 25
t CERTIFICATE l' 2 3 This is to certify that the attached proceedings before the 4 United States Nuclear Regulatory Commission in the matter of: 5 .Name: Advisory Committee on Reactor. Safeguards, Subcommittee n Me hanical Components 6 7 Docket Number: 8 Place: Bethesda, liaryland 9 Date: October 3, 1989 10 were held.as herein appears, and that this is the original .11 transcript thereof for the file of the United States Nuclear 12 Regulatory Commission taken stenographically by me'and,. '~ thereafter reduced to typewriting by me or under the direction 13 14 of the court reporting company, and that the transcript is a 0 15 tree and acc rate record of the fere,eino preceedines. 16 /S/ (#Wrw-.a n 7-17 (Signature typed): Catherine S. Boyd 18 Official Reporter 19 Heritage Reporting Corporation 20 21 22 23 24 i 25 -l O Heritage Reporting Corporation (202) 628-4888 l
L,4 c J [ y 1 e s. -. 0 BACKGROUND ON GENERIC LETTER-(GO NO WRITTEN TECHNICAL. GUIDANCE 0N IST HUGE VOLUME OF PROGRAMS / REVISIONS /R~ LIEF REQUESTS RELIEF REQUESTS UNREVIEWED AND IMPLEMENTED WITHOUT PRIOR NRC APPROVAL CONTRARY TO TS 4.0.5 INSPECTION EFFECTIVENESS HAMPERED .O-l LOL
i!-, "< - l :,- ..p PURPOSE OF GENERIC LETTER (GL) w PROVIDES GENERIC' GUIDANCE ON ELEVEN SIGNIFICANT: IST PROBLEM AREAS y .PROVIDES GUIDANCE ON DEVELOPING' ACCEPTABLE IST PROGRAMS CLARIFIES APPROVAL STATUS OF.IST PROGRAMS (1.E., RESOLVES TS 4.0.5 ISSUE). IST PROBLEM AREAS: CODE INTERPRETATIONS TECH SPEC INTERPRETATIONS APPROVED ALTERNATIVES TO CODE i O-
en q[ Q' ' APPROACH USED IN GENERIC LETTER'(GL)'89-04 L. 1
- THREE GROUPINGS OF. PLANTS TABLE 1 PLANTS
~ SER NEARING'. COMPLETION SER CONSTITUTES APPROVAL TABLE 2 PLANTS-SER ISSUED ON CURRENTLY. SUBMITTED. PROGRAM SER CONSTITUTES APPROVAL. ' TABLE 1 AND 2 PLANTS DO NOT NEED TO RESPOND TO GL ' ' NEED TO ENSURE PROCEDURES CONSISTENT WITH GL j O ---_.__-__L_______._.,_______.___._._.___
f:y .f x( y [ , PLANTS;NOT ON EITHER TABLE. '* 'GL CONSTITUTES APPROVAL PROVIDED LICENSEES: .[. -REVIEW PROGRAMS AGAINST ATTACHED POSITIONS, AND -CONFIRM.'CONFORMANCE WITH ATTACHED POSITIONS IN SIX MONTHS,-AND -MAKE ANY MODIFICATIONS WITHIN SPECIFIED TIME' NRC WILL NOT ISSUE SERS ON -CONFORMANCE WITH ATTACHED POSITIONS O- -REtiEr RE0ueSTS DOCxETED 8Ee0RE AeRIt 3, 1989 NRC WILL ISSUE SERS ON -NEW RELIEF REQUESTS O
/-6 PROGRAM UPDATES / REVISIONS
- 'FOR PROGRAM CHANGES COVERED BY ATTACHED POSITIONS
-LICENSEE MAY IMPLEMENT BEFORE NRC REVIEW -SER WILL'NOT'BE ISSUED
- -STAFF WILL EVALUATE DURING INSPECTIONS FOR NEW RELIEF. REQUESTS
-STAFF WILL EVALUATE PER 10 CFR 50.55A (G) INSPECTION AND ENFORCEMENT INSPECTIONS'TO BE CONDUCTED FOR CONFORMANCEWITH 10 10 CFR 50.55A. AS EXPLAINED'IN GL -FOCUS ON ATTACHED POSITIONS -0THER AREAS MAY BE INSPECTED 10-
7 O STATUS ON GL WORKSHOPS ON GL 89-04 HELD IN REGIONS IN JUNE 1989 MINUTES OF GL WORKSHOPS SENT FOR CRGR CONSIDERATION ON SEPTE14BER 6, 1989 CRGR BRIEFING ON OCTOBER 11, 1989 EIGHT TABLE 1 SERS ISSUED SINCE APRIL REMAINDER OF TABLE I SERs TO BE ISSUED BY JANUARY 1990 O O
g-- t l l-PROPOSED ACTION PLAN FOR MOTOR-OPERATED VALVES (MOVs) O l l O l l l
l l 1 l BACKGROUND l-
- PAST PROBLEMS
- CAUSES
' "^VIS-ESSE 6/9/85 O
- CATAWBA 3/14/88
- MILLSTONE 3 2/17/89 l
1 i ) O I
'gw-REGULATORY REQUIREMENTS
- APPENDIX A TO 10 CFR 50 APPENDIX B TO 10 CFR 50 l
SECTION 50.55A i l l Q APPENDIX J TO 10 CFR 50 SECTION 50.36 l l l O
fyyW ~- ~' , ~ ^ ~~ 1 L ] NRC ACTIVITIES- ~l 0FFICE OF NUCLEAR REACTOR REGULATION (NRR) { l t. (E.G., MECHANICAL ENGINEERING BRANCH, PERFORMANCE EVALUATION-BRANCH, SPECIAL INSPECTION BRANCH AND RISK. APPLICATIONS BRANCH l-
- BULLETIN 85-03 L
- GENERIC LETTER 89-10
- GENERIC' LETTER 89-014
- PRESSURE ISOLATION VALVE PROGRAM l
OFFICE OF NUCLEAR REGULATORY RESEARCH-(RES) O
- GENERIC ISSUE 87, FAILURE OF HPCI STEAM LINE WITHOUT ISOLATION
- VALVE' PERFORMANCE PROGRAM
-t
- NUCLEAR PLANT AGING RESEARCH PROGRAM I
i 0FFICE FOR ANALYSIS 8 EVALUATION OF OPERATIONAL DATA (AE0D)
- CASE STUDY REPORT C603 (12/86) i
- SECY 89-165 (5/31/89)
] I ) l O'
. Ji ' ;. L <3 f 1 1: -,oj + INDUSTRY ACTIVITIES m.. v. E PT R I.
- TECHNICAL REPAIR' GUIDELINES'FOR LIMITORQUE SMB-000
- -APPLICATION GUIDELINES-I N'P 0-
- PLANT VISITS
- 10/4/88 KEY ELEMENTS OF COMPREHENSIVE MOV PROGRAM O
N U M A R.C
- GUIDELINES FOR IMPLEMENTATION OF GL 89-10 SIGNATURE TRACING TECHNIQUE IMPROVEMENTS MOV DESIGN IMPROVEMENTS IEEE MAINTENANCE PRACTICES ASME
- OM-08 L-
- OM-10 O-
- QME-1 L- =_--
m, y, M: m:!u !*l], -PROPOSED l ACTION) t: 4
- REGULATORY. EFFORT LED.BY MECHANICAL ENGINEERING BRANCH OF NRR LEAD / DIRECT MOV AUDITS AND INSPECTIONS 1
COORDINATE:MOV AUDITS AND INSPECTIONS BY OTHER NRR UNITS 'FDLLOW AND REVIEW RESULTS OF MOV AUDITS AND INSPECTIONS BY REGION PERSONNEL (]) AEOD. REPORTS ON MOV EXPERIENCE (MEB INTERACT WITH-1NDUSTRY i... RES PREPARE' RULES.AND REGULATORY GUIDANCE .( O
i J -SPECIF'IC' ACTIONS,.0F LEAD NRR UNIT A.- PREPARE-IST REGULATION REVISION AND' REGULATORY GUIDANCE
- SCOPE.0F IST
- 'OM-10 ENDORSEMENT
' TESTING TO DEMONSTRATE OPERABILITY
- GENERIC LETTER 89-10' PROVISIONS GENERIC LETTER 89-04 PROVISIONS
- IST PROGRAM DESCRIPTION
- IST PROCEDURE PREPARATION
- TRAINING
-j
- OTHER IST AREAS, AS NEEDED B.
EVALUATE.NEED FOR REGULATORY GUIDANCE IN OTHER MOV () AREAS C. RESOLVE THERMAL OVERLOAD PROTECTION ISSUE D. PREPARE GUIDANCE FOR NRC PERSONNEL
- INSPECTION MANUAL
- STANDARD REVIEW PLAN E
PREPARE TECHNICAL GUIDANCE FOR GENERIC LETTER 89-10 F. HOLD WORKSHOPS ON GENERIC LETTER 89-10 G. REVIEW GENERIC LETTER 89-10 PROGRAM DESCRIPTIONS l O
I w.- L: ) H. REVIEW INDUSTRY TESTING FOR GENERIC LETTER 89-10 1. LEAD / DIRECT MOV AUDITS AND' INSPECTIONS 1 J. LC00RDINATE MOV INSPECTIONS'BY OTHER NRC PERSONNEL .K. HOLD.NRC COORDINATION MEETINGS (QUARTERLY) L. REVIEW RES PROGRAM RESULTS AND'NEW RESEARCH PROPOSALE 1-N.- . ASSIST INDUSTRY INCLUDING MEETINGS AND REVIEW OF GUIDELINES N. REVIEW SAMPLE 0F INPO PLANT VISITS 0. EVALUATE INDUSTRY EFFORTS P.- EVALUATE NRC PROGRAM FROM DESIGN AND TESTING IMPROVE-MENT AND MOV EXPERIENCE 10 Q. ASSIST IN REVISION OF STANDARD TECHNICAL SPECIFICATIONS, AS NECESSARY R. ASSIST IN REVIEW OF PLANT TECHf41 CAL SPECIFICATIONS i S. CONTINUE NRC VALVE REVIEW GROUP T. PARTICIPATE IN CODES AND STANDARDS DEVELOPMENT U. FOLLOW BECHTEL/KWU ALLIANCE V. PREPARE COMMISSION PAPER ON PERIODIC BASIS O
-f O 1: . J PROPOSED ACTION PLAN FOR CilECK VALVES O l l O
kt [- 1 . Q,/ BACKGROUND PAST PROBLEMS CAUSES SAN ON0FRE 1 11/21/85 CALVERT CLIFFS 3/17/88 AND 10/29/88 O CRYSTAL RIVER 3 6/21/88 ARKANSAS NUCLEAR ONE 1 1/20/89 COMANCHE PEAK 4/23/89 AND 5/5/89 O
'O REGULATORY REQUIREMENTS '[.
- APPENDIX A TO 10 CFR 50 I
- APPENDIX B TO 10 CFR 50
- SECTION 50.55A O
APPENoix a 10 10 CeR 50 l
- SECTION 50.36 l
l I 1 \\ ~ i I e 1 O i l
f 7 r N' R C ACTIVITIES 'NR'R' (E. G., MECHANICAL ENGINEERING BRANCH, MATERIALS ENGINEERING BRANCH, PERFORMANCE EVALUATION. BRANCH,'AND RISK APPLICATIONS BRANCH-
- AFTER SAN ONOFRE, STAFF REQUIREMENTS MEMORANDUM ON L
-2/7/86 REQUIRED PERFORMANCE OF GENERIC ANALYSIS i: OF CHECK VALVE FAILURES e
- 4/7/86' MEETING BETWEEN NRC AND INDUSTRY
- 12/9/86 SURRY UNIT 2 FEEDWATER LINE RUPTURE
- 3/30/87 STAFF INFORMS COMMISSION OF PLANS FOR INSPECTIONS
- SECY 87-168-(7/9/87), SECY 88-74 (3/9/88),
SECY 88-297 (10/18/88) REPORTS ON CHECK VALVE INSPECTIONS
- INFORMATION NOTICE 88-70 (8/29/88)
- GENERIC LETTER 89-04 (4/3/89)
- PRESSURE ISOLATION VALVE PROGRAM
- ANCHOR DARLING CHECK VALVES WITH TYPE 410 STAINLESS STEEL BOLTS (BULLETIN 89-02)
O
9 { o ' (m. ' NRC ACTIVITIES (CONTINUED) RES GENERIC ISSUE 105, INTERFACING SYSTEMS LOCA AT LWRs VALVE PERFORMANCE PROGRAM NUCLEAR PLANT AGING RESEARCH PROGRAM AE0D EVALUATION OF OPERATING EXPERIENCE INSPECTION PARTICIPATION O i \\
7
- q L
3 -F lNDUSTRY ACTIVITIES I N)P.0 o SOER 86-3 (10/15/86)
- OCTOBER'1986 WORKSHOP' PLANT VISITS p:
lEP-RI APPLICATION GUIDELINES FOR CHECK VALVES (1/88) L NUCLEAR INDUSTRY CHECK VALVE GROUP (NIC) FORMED 4/89' DIAGNOSTIC TECHNIQUE IMPROVEMENTS DESIGN IMPROVEMENTS ASME. OM-10 i O
(}) PROPOSED ACTION REGULATORY ACTIVITIES LED BY MECHANICAL ENGINEERING BRANCH OF NRR SPECIFIC ACTIONS A. DIRECT / LEAD CHECK VALVE INSPECTIONS B'. COORDINATE CHECK VALVE INSPECTIONS BY OTHER NRC UNITS C. REVIEW RES PROGRAM RESULTS AND AEOD REPORTS (INTERACT WITH INDUSTRY) D. HOLD NRC COORDINATION MEETINGS (QUARTERLY) E. REVISE IST REGULATIONS TO CLARIFY SCOPE, TO ENDORSE OM-10, AND TO MAKE CLEAR IST INTENDED TO DEMONSTRATE PERFORMANCE OF SAFETY FUNCTIONS F. EVALUATE IST REGULATORY CRITERIA AND AVAILABLE DIAGNOSTIC TECHNIQUES G. PREPARE REGULATORY GUIDANCE FOR IST PROGRAMS, TRAINING, CATEGORIZATION AND OTHER AREAS H. EVALUATE NEED FOR IMPROVEMENTS TO GUIDANCE FOR NRC PERSONNEL:
- INSPECTION MANUAL
- STANDARD REVIEW PLAN 1.
ASSIST INDUSTRY IN RESOLVING CHECK VALVE CONCERNS, INCLUDING MEETINGS AND REVIEW OF GUIDELINES J. SAMPLE REVIEW OF INPO PLANT VISITS K. EVALUATE INDUSTRY EFFORTS TO RESOLVE CHECK VALVE PROBLEMS L. ASSIST IN REVISION OF STANDARD TECHNICAL SPECIFICATIONS AND REVIEW OF PLANT TECH SPEC CHANGES, AS NECESSARY ) i M. CONTINUE NRC VALVE REVIEW GROUP { N. PARTICIPATE IN DEVELOPMENT OF CODES AND STANDARDS J (EMPHASIZE NEED FOR IST STANDARD FOR CHECK VALVES) O. FOLLOW IMPROVEMENTS IN DESIGN AND TESTING, AND CHECK VALVE EXPERIENCE TO DETERMINE NEED FOR CHANGE IN NRC EFFORTS P. PREPARE COMMISSION PAPER ON PERIODIC BASIS
o ff ~'
- 1..i U[
4 '5 Qi ,.h jO .MOV, FAILURE T0'CLOSE AGAINST DIFFERENTIAL PRESSURE .-u- ~!i D } ' EVENTS: i CATAWBA-3/14/88-t MILLSTONE 3 2/17/89 f-L ACTIONS 8/14/89 LETTER TO NUMARC FROM L. B.-MARSH REQUESTING-THAT EPRI PERFORM RESEARCH ADDRESSING-DESIGN AND QUALIFICATION OF MOVS 1 'TAFF INFORMED' LICENSEES AT WORKSHOPS ON GL 89-10' S THAT APPLICATION OF TEST DATA FROM PROTOTYPE MUST LBE' JUSTIFIED EVEN IF MOVS ASSUMED-TO BE. IDENTICAL 1
- .0
'{ ,,s. .,(. lm. PRESENTATION TO THE MECHANICAL COMPONENTS SUBCOMMITTEE OF THE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS CONCERNING GENERIC LETTER 89-10, " SAFETY-RELATED MOTOR-0PERATED VALVE TESTING AND SURVEILLANCE" Robert L. Baer, Frank Cherny, Owen Rothberg O Office of Nuclear Regulatory Research Division of Safety issue Resolution Engineering issues Branch October 3,1989 O {t.CRSH2721)
$ii 1 ,, ? :, i u + 1 ACRS CONCERNSLAND STAFF RESPONSE 1 1 ACRS LETTER T0 CHAIRMAN ZECH DATED MAY 9,1989 1
- FOUR CONCERNS USTED
.r STAFF MEMORANDUM 1TO ACRS DATED JULY 17, 1989
- STAFF COMPUED-WITH CONCERNS 1,4, AND PART OF 3
- CONCERN 2 WAS CONSIDERED TO IMPLY A BACKFIT, IF INCLVDED IN GENERIC LETTER 89-10 C. MICHELSON MEMORANDUM TO ACRS MEMBERS DATED SEPTEMBER 4,1989
' DISCUSSED CONCERN 2 0F MAY 9,1989 ACRS LETTER
- ASKED STAFF TO MEETING TO DISCUS $ 5 QUESTl0115 (CRSE92722)
E L B A w T C .E O S LS L I N E ANP I D F S S CCOP N E O N ENIS P A I S IL O O E Y S 9 SP E E R 8 HEP C 0 NIT I DCH E I I T B R N C 9 E S iC AN OD D E F l I 1 I P l T A1 MSR N A - T U N R T 3 E I I T H A A OE D N 7 G S N 9 I I S AS E WE A E 8 T Cl A0 I I S A D - G L 5 M C T N E T I NO R T I B F S HR E L D N P ' OT 4 E OG O E l E B I H N N OAT MI N D R 0 4 R T G N i D I MT E S A0 R O 5 t I S R A O R E L S T MR O E A 1 I F E SI N R R N R L PO S - G 2 S E T E S E T C T N 7 O U AE E O MP 9 E 4 E B ODS I S E I C CI V C R E 8 R t E O D N ML R I 1 I R E VS I L M E E SE OW E O V I N D H M E RO T I C C E F E C Q E T O S R OS4 N U F T A T N YC E I t S) S T O C O AH SA G S I T A N0B R P T P 4 A L SE D A1 O t L F N AL R T OE SR E 0B I OO D U I TN NA S C E T AO 1 MS OS - NV R I C G T S T N AI F N GR E N I C NR E E A S E E T E I 8S U T I E GE S 9 T O Y I R S T I M D M S E S E T R R MD O B AX I S S N N T AR D ,L AR E E AD D I E E E N T I WN T R O M BE T O D G A N O l N E N C L F E A A C I M TE O l S O A T NT HI HE E F ULM P D D D A O G E O T T T MS 4 GE P 4 L C I L i A L C A U S UY UA T ESD A N S E t F C E R TL G E E C OL D L N P I U N S E E DI A O L I O R S OI 4 E S t R T D D E T R S EN E C R R N E SE W S AI WE F H N SE S E E E E H NO4 S R E L A C F C F E E I t C T T iE D I H l S CE CB N AI B E Vi I G A VN T MM I F NO E (A NH C A R E l L N T I ETE N H D N G I S E S S N D O E I C M I DG WU E S N S MS I N O CD I I I M S E S G E OI F E E E C M_ S A D A E N AB E C P E R CC E OT I L D D OS R N E N O A E N D Y P P I E R E E R T E C I HN F S I S iS C N E HD H M1 I ,B A I A E P GE C T V' L r AE P 4 I Y I t T NR l I E S D E EE I FE O O4 E G R P OE E C l l t N H S i i D I L D T T T MOD A AAI CP I T 2 3 1 l
j1 1 1 S E S D G 'V E R N M E AO U T HM S S T C M N E 'R S S L E I F O O A SYOR T I C T E AN F T S I E H BAT D I ANER L T ND N WI RE NDE E S AN CE GNMI O FC TE E I T E T I Y G R I A S I I MST D .CE E E G N D E RNHE NI DMU EN J E E T H E T O DDA T G SE C U V CS E L E S I I A N EI A H E VB ,S WF R T 4 E T Ht A E R N I T O D SI T l CTR E E H W I I S SL OO T N V E F A EPNF E O Y.O B Y E P E M G L N L SASR AS C PS NY P N E E RI I PN GL P E YOH A S )4 S AOf P ACL DT P A 2 I T B 7 I S 2 M L N 0 E s T E 3 R 3 C D O .l N e E E E C T 1 S G s i R T e D D A - N I c R G N H 9 OH W R S A N R N F A OI O AT U ( E 8I T D S T C T 2 C I I SS 2D RD W N E A S EX S N EN G I T E E S E HT O EN NI C N M T T O E C CIT O D E O I MOL NS F I E T P HR T OT DAI X F S X T P S C CN D E E E E SI AR E A R U FE U R 'V E OE T QD OH Q ON SCH S L T O ME N A TCT OU N OT TOION T G E F F I N VHO RWT O .E l R E AD E T S E E T E Y S E WE NR WT O E D D I SI S U S E E E S R NR O NSMV NWOA D T A H L S A NEO PH T U E E L C S C PS E SE E HI MI H ECO O H AWRST WT T L D T I 4 5 g lllll lil
N-
- u pu p
- o UNITED STATES F,'n NUCLEAR REGULAYORY COMMISSION
.f Mg ADVISORY COMMITTEE ON' REACTOR SAFEGUARDS %..V [o' WASHINGTON, o, c. 20555 September 4,1989 MEMORANDUM FOR: A,CR Members FROM: r.1 rehelsen, Member
SUBJECT:
SELECTING MOV SWITCH SETTINGS AND DEMONSTRATING f VALVE OPERABILITY UNDER GENERIC LETTER NO. 89-10 The recornmended actions portion of Generic Letter No. 89-10 on, " Safety-Related Motor-Operated Valve Testing and Surveillance," requires a program which includes the following provisions: a. Review and document the design basis for the operation of each MOV. This documentation should include the maximum differential pressure expected during both the opening and closing of the MOV for both normal operations and abnormal events, to the extent that these MOV operations and events are included in the existing approved design. basis. Regarding item a., no change to the existing plant e. design basis is intended and none should be inferred. 3 The design-basis review'should not be restricted to a determination of estimated maximum design-basis () differential pressure, but should include an examination of the pertinent design and installation criteria that were used in choosing the particular MOV. For example, the review should include the effects on MOV performance of design-basis degraded voltage, including the capability of the MOV's power supply and cables to' provide the high initial current needed for the operation of the MOV. The ACRS letter of May 9, 1989 (" Generic Letter on Safety-Related Motor-Operated Valve Testing and Surveillance") included the following recommendation: 2. Although no change in the existing plant design basis is intended by the generic letter, we recommend that each licensee be reminded to review the design basis governing the selection of each MOV from the viewpoint of completeness and adequacy in light of current regulatory requirements. In the meantime, and to the extent possible, current requirements should be reflected in selecting MOV switch settings and demonstrating operability. The staff response to this recommendation was contained in a letter from V. S t e l'l o, Jr. to R. F. Fraley (" Generic Letter () on Safety-Related Motor-Operated Valve Testing and Surveillance"). The staff reply to recommendation 2 in the ACRS letter was as follows:
9/4/89 ?(. ACRS recommendation 2. regarding design basis review is considered to be partially incorporated in the generic letter (action items a. and e.) as now written. The wording was chosen to avoid any ambiguit3 about the NRC l position that no changes to the existing design basis i are intended. Within that context, it was recommended that licensees review their motor operated valve design parameters to be sure that the equipment fully satisfies the design be, sis. The ACRS recommended that ",..each licensee be reminded to review the design basis governing the selection of each HOV from the viewpoint of completeness and adequacy in light of current regulatory requirements" (emphasis added). This implies that older plants are expected to revise their design basis, which would considerably broaden the scope of the effort for many plants. This was not the intent of the generic letter. Accordingly, the wording of recommendation 2 was not incorporated into the generic letter. When discussing recommendation 2, the Committee chose to delete a second sentence which stated, "If reportable deficiencies in the design basis are encountered, they should be handled using existing regulatory procedures." The purpose of the deleted sentence was to emphasize that f~T although only the original design basis requirements apply, (_/ if reportable deficiencies are found during the required design review process, they should be documented in accordance with existing requirements for reporting and resolving such deficiencies. Thus, the question of backfit l under Generic Letter No. 89-10 would not be an issue. I believe the Committee felt that the existing requirements for reporting were self evident and need not be emphasized; therefore, the sentence was deleted. The third sentence of recommendation 2 is really the point of the recommendation. I believe this sentence is saying that HOV switch settings and operability testing should, to the extent possible, be based on a current understanding of the valve's functional requirements. This current understanding would correspond to the original design basis requirements if no reportable design deficiencies are f our.d during the design review. If reportable deficiencies exist which may be corrected (at least in part) by changes in MOV switch settings and in situ testing, then I believe the sentence is recommending that such corrective actions should be taken pending final resolution of the deficiency. It is t my view that this would be the prudent thing to do. l The staff rejected recommendation 2 on the basis that it l constitutes a backfit. It is my view that the staff f (' ) _ misunderstood the intent of the recommendation. The Mechanical Components subcommittee has scheduled a subcommittee meeting for the morning of October 3, 1989, to consider Generic Letter No. 89-04, " Guidance on Developing 2 i
I 9/4/89 L,] Acceptable Inservice Testing Programs." A discussion of Recommendation 2 will be added to the agenda and the l subcommittee will report its findings to the Committee i during the October meeting. i By copy of this memorandum, the staff is being asked to come prepared to discuss the foll5 wing questions concerning Generic Letter No. 89-10: 1. What is meant by the terms, " existing approved design basis" as used in requirement "a", and " existing plant design basis" as used in requirement "e"? What 153C$$ documents might define the design basis? 2. If deficiencies in the design basis are found during the review under requirement "a", what should be the J) Dei process for documenting and reporting? O 3. How will the regulatory resolution of such design deficiencies be handled? 4. Can interim corrections of design basis deficiencies be implemented before final resolution? 5. How does the staff view the question of adjusting MOV /~ switches and in situ testing procedures to reflect the (-}/ current understanding of valve functional requirements \\ instead of design basis requirements when the design basis is found to be deficient? For example, assume that during the design basis review of the Reactor Water Cleanup System for a BWR it is found that the containment isolation valves were not specified to isolate a downstream pipe break even though the flooding analysis for the plant as.cumed such a break. How would the regulatory process handle this deficiency under the Generic Letter? If the motor operator on the valve is large enough, The problem might be corrected by adjusting the torque switch to a new calculated value. Could this change be implemented before final resolution of the deficiency? Is such a change considered a backfit? Should such a change be a requirement in the Generic Letter? Is the staff aware of any cases wherein a licensee has not specified a downstream pipe break as the design basis for isolation valves on systems such as HpCI, RCIC, or RWCU? 3 \\
L' s cc: Tad Marsh, NRR Owen Rothberg, RES Gerald Weidenhammer, RES Robert Bosnak, RES Robert Baer, RES I Earl Brown, AEOD TEd Sullivan, NRR Richard Kissel, NRR Peter Wohld, ACRS Consultant C. P. Siess J. C. Carroll C. Wyl ie l E. Igne R. F. Fral ey R. Savio R. Major G. Quittschreiber 1
'~ [<%pcn%q 4 UNITED STATES k NUCLEAR REGULATORY COMMISSION WASHINGTON D.C.20555 7,; y JUL 1719B9 MEMORANDUM FOR: Raymond F. Fraley, Executive Director Advisory Committee on Reactor Safeguards FROM: Victor Stello, Jr. Executive Director for Operations
SUBJECT:
GENERIC LETTER ON SAFETY-RELATED MOTOR OPERATED VALVE TESTING AND SURVEILLANCE By a letter dated May 9, 1989 from F. Remick, Chairman ACRS to L. Zech, Jr., Chairman, NRC, the Advisory Committee on Reactor Safeguards concurred with the need for and scope of the subject generic letter. The ACRS also offered four recommendations for incorporation in the generic letter. The purpose of this memorandum is to inform the ACRS as to the staff's response to these recommendations. The NRC staff has added two paragraphs to action item f, of the generic letter in response to the ACRS recommendations, as follows: " Alternatives to testing a particular MOV in situ at design basis pressure or flow, where such testing cannot practicably be performed, could include comparison with appropriate design basis test results from other MOVs, either in situ or prototype. If such test information is not currently available for the particular M0V, analytical methods and extrapolations to design basis conditions, based on the best data available, may be used until test data at design basis conditions becomes available to verify the operability for the MOV. If this two-stage approach is followed, it should be accomplished within the schedule outlined in item 1. and would allow for MOV testing and surveillance to proceed without excessive delay. Testing of MOVs at design basis conditions need not be repeated unless the MOV is replaced, modified, or overhauled to the extent that the licensee considers the existing test results to be i unrepresentative of the MOV in its modified configuration." The additional text is intended to incorporate recommendations 1. and 4. and part of recommendation 3. of the May 9,1989 ACRS letter. ACRS recommendation
- 3. was partially incorporated in that alternatives to in situ design basis testing would be permitted based on the availability of reliable prototype test data if in situ design basis testing was impractical.
In situ design basis testing is still recommended if practical. O
b U ACRS recommendation 2. regarding design basis review is considered to be partially incorporated in the generic letter (action items a. and e.) as now ( written. The wording was chosen to avoid any ambiguity about the NRC position 1 that no changes to the existing design basis are intended. Within that- ] context, it was recommended that licensees review their motor operated valve ] design parameters to be sure that the equipment fully satisfies the design basis. The ACRS recommended that "...each licensee be reminded to review the design basis governing the selection of each M0V from the viewpoint of com-pleteness and adequacy in light of current regulatory requirements" (emphasis added). Tnis implies that older plants are expected to revise their design basis, which would considerably broaden the scope of the effort for many plants. This was not the intent of the generic letter. Accordingly, the wording of recommendation 2 was not incorporated into the generic letter. ngadO M M py Stdh ~2/. Victor Stello, Jr. Executive Director for Operations cc: Chairman Carr Commissioner Roberts Commissioner Rogers Commissioner Curtiss SECY p( [MEMD STELLO TO FRALEY/GLMOVS] EIB:DS$'d,EIB hn{ers usto tsp s DD-R DD.. S EIB:DSIR ORothberg/mb FChernr RBaer 06/ /89 0 7 0/89 Of/4/89 07/}/89 Op//o /89 j /89 D:R N ord V te lo / /89 fj89 0FFICIAL RECORD COPY f?n,w)b dc 600 Ccc4C C. mV ~
I ,g. a \\e I .t'1 7_ ,.,3..[ [ g); Distribution-RE5 Chron 1RES Circ M .EIB r/f U PDR-1 - G. - Arlotto R.' Baer. C.. Bartlett-- ' C. Berlinger E.LBrown-R.'Bosnak ' ~ ~ - F. Cherny F. Gillespie. .W. Hodges R. Houston A.=Igne R. Kiessel -R. Kirkwood R. Major. L :. L. Marsh W. Minners .T. Murley T. Novak W. Parler T. Ross O cRothberg J. Conran 'T.-Speis E. Beckjord EDO Rdg-MBridgers(ED0-4526) PDorm .O
x ll,, t- [ '*,o, . UNITED STATES fw ? f*Y D J NUCLEAR REGULATORY COMMISSION - '{ [$ ' ADVISORY COMMITTEE ON REACTOR sAFEGUAhDs - W ASHINGT ON, D. C. 20555 O g i.....f y May 9, 1989 .The Honorable Lando W. Zech, Jr. ' Chairman U.S.~ Nuclear Regulatory Commission Washington, D.C. 20555.-
Dear Chairman Zech:
GENERIC' LETTER ON SAFETY-RELATED MOTOR-OPERATED VALVE TEST
SUBJECT:
l SURVEILLANCE 'During the 349th meeting of the Advisory Committee on Reactor Safeguards, May we discussed : the. subject generic-letter. This matter was also-3-6, 1989, October 6-8, 1988, considered during' our meetings on September 8-10,1988, and April 6-8, 1989. Ourf Subcommittee on Mechanical Components discussed this issue during several recent' meetings, including one on-May 3,1989. During these meetings, we had the benefit of discussions with representatives of the NRC staff'and its consultants and the Nuclear Management and Resources Council. We also had the benefit of the document referenced. We have been following the NRC staff's activities concerning the industrywide roblem of deficient performance of safety-related motor-operated valves p(MOVs) for several years _and have held numerous-meetings to discuss this issue. : We consider the apparent unreliability of such valves and the poten- .tial inability of some' valves to function under design-basis conditions to be-a s,ignificant safety issue of high priority. Further, we are concerned that the stroke-timing. test prescribed by the regulations is not a valid test of This' is a test that con-valve operability under design-basis conditions. sists of stroking the valve open and closed, usually without flow or elevated pressure, and measuring the stroke time. Because of similar concerns, the staff issued Bulletin 85-03, which required a special operability assurance program for certain MOVs in two high-pressure safety systems (i.e., high-pressure injection and auxiliary feedwater sys-tems). This program was to ensure that the switch settings for the motor operators on these valves would be selected, set, and maintained so that the valves will be capable of performing their intended design-basis functions for the life of the plant. ~ The staff is now preparing to issue a generic 1etter to extend the scope of It also Bulletin 85-03 to all safety-related and position-changeable HOVs. suggests that other MOVs in the balance of plant be considered for inclusion, commensurate with the licensee's assessment of their importance to safety. O g
c. 4 7 O '"e "e e< e ' de " zech ar. sex 9. 1989 W concur with the need for and scope of the proposed generic letter in order to formalize a program to deal with this problem. However, we.believe that it should be-revised to incorporate the following recommendations: The matter of MOV testing and surveillance should be approached in two L 1. stages. For the first stage, each licensee should perform a review and develop documentation of the existing approved design basis governing the selection of each MOV and establish the " correct" MOV switch set-tings using the best. available data and calculational methods. The y' in-plant MOV settings should be changed to the selected values and the '
- /
design-basis operability demonstrated, to the extent practical, using in situ tests and state-of-the-art testing procedures, extrapolation l. l techniques, and diagnostic equipment. For the second stage, each licensee should = complete its program for ~ demonstrating the operability of each MGV by testing under design-basis conditions (preferably reflecting current regulatory requirements as noted in item 2 below) or by using an acceptable alternative. This may require extensive out-of-plant prototype testing and analysis. 'Although no change in the existing plant design basis is intended by the 2. generic letter, we recommend that each licensee be reminded to review the design basis governing the selection of each H0V from the viewpoint of completeness and adequacy in light of current regulatory require-O' In the meantime, and to the extent possible, current require-ments. ments should be reflected in selecting MOV switch settings and demon-s strating op'erability. g The present draft of the generic letter appears to permit alternatives 3. to in situ design-basis testing only if it is precluded by the existing We consider this requirement to be too restrictive plant configuration. / and recommend that reasonable alternatives be permitted at the option of V the licensee even if in situ testing is possible. The generic letter does not clearly state the circumstances under which 4 a demonstration of operability (e.nder design-basis conditions might ne u to be repeated in the future g., af ter a major maintenance or modi- / y fication is performed). We believe that this needs to be clarified. i Our intention in recommending a two-stage approach is to encourage an early implementation of the immediately achievable portions of the generic letter while work proceeds on a reasonable schedule to develop the required calcu-lational and testing capabilities and to complete the tests required to full implementation. We believe that; a two-stage approach will achieve ensure a more orderly achievement of the objectives of the generic letter. O ./
(' ) The Honorable Lando W. Zech, Jr. - Priority consideration should be given in both stages to those MOVs that the licensee considers to have the greatest impact on plant safety. Sincerei d h orrest J. Remick Chairman
Reference:
Letter dated April 26, 1989 from E. S. Beckjord, Office of Nuclear Regulatory Research, to T. E. Murley, Office of Nuclear Reactor Regulation,
Subject:
Transmittal of Generic Letter on Motor-0perated Valve Testing and Sur-veillance 'b) m O I O
Y ~ ,p#* UNITED STATES c' NUCLEAR REGULATORY COMMISSION
- 4. ' h M A&HINGTON, D. C,70555
.M(( L WG 141999 Mr. Thomas E. Tipton, Director Operations, Management and Support Services Division Nuclear Management and Resources Council (NUPARC) Suite 300 1776 Eye Street, N.W. Washington, D.C. 20006
Dear Mr. Tipton:
SUBJECT:
FERFORPAHCE OF MOTOR-OPERATED YALVES As you well know, the proper performance of motor-operated valves (MOVs) is essential to the safe operation of nuclear power plants. Over the last few years, the need to improve the performance of these valves has become apparent. In this regard, the hRC staff and nuclear industry have efforts unoerway to provide increased confidence that HOVs will perform their safety functions in a reliable manner. The NRC stati activities aimed at enhancing MOV performance include prepration of NOV regulatory guidance, increased emphasis on MOVs during inspections, and direction of MOV research programs. For the industry's part, the Institute of Nuclear power Operations is performing more detailed reviews ,k]ofMOVactivitiesduringplantvisits. Further, the Electric Power Research Institute (EPRI) has issued technical repair guidelines for a Limitorque valve actuator and 1$ preparing guidelines for MOV applications. Although these and othcr activities by the staff and industry have been beneficial, additional effort is needed to. resolve remaining MOV concerns. One aspect of MOV performance requiring attention by the industry involves the initial design and qualification of these valves. Concerns in this area have been raised by on-going research and plant operating experience. For example, preliminary results of NRC research reveal that the thrust required to close a valve under full flow conditions may be much larger than anticipated. As a consequence, there is a potential for MOVs being unable to close when necessary. These research results were discussed at a meeting of valve experts on February I of this year. A summary of that meeting is attached for your information. The v611dity of the research results is reinforced by the failures of HOVs actuators to close the valve that have occurred at operating nuclear power plants. Such events at Millstone Unit 3 on February 17th of this year and at Catawba Unit 2 more recently on March 14th are described in Licensee Event Reports 423/89-005 and 414/89-010, respectively. Among the possible contributors, individually or in combination, to the MOV failures to close are valve design, stem thrust analysis, and switch settings. One particular area of concern is that valve friction factors supplied by the manufacturer have been found to differ significantly fron licensee-determined values.
h Mr. Thomas E.. Tipton On June 28, 1989, the NRC staff issued Generic Letter 89-10, " Safety-Related Motor-Operated Valve Testing and Surveillance," that will have holders of operating licenses and construction permits establish a program to ensure that a switch settings for safety-related MOVs and certain other MOVs important to plant safety are selected, set, and maintained properly. Generic Letter 89-10 intends that in situ testing be performed under design. basis conditions for MOVs within its scope. Where such in situ testing is not practicable, the capability of the MOV to perform its function may be determined by analysis of in situ testing of other similar valves or a ptototype. The Generic Letter also provides for the establishment of a program to maintain ttie proper MOV switch settings and to help ensure continued MOV operability under design basis conditions. The performance of industry research with respect to MOVs could provide information helpful to the individual licensees in the imple-mentation of the Generic Letter. To improve our knowledge of MOV performance under various flow conditions, the NRC staff is conducting additional limited research relative to the thrust requirements for MOV operation. It is our understanding that EPRI is in the process tc identifying aspects of MOV performance that should be investigated by the industry. In light of c'ur research results and plant operating experience, it is recommended that the industry establish a research program to evaluate the adequacy of MOV design and qualification. 3f you have any questions on the need for this effort by the industry, please contact Tom Scarbrough of my staff at (301) 492-0916. Sincerely, t [., % T4 h Ledyard B. Marsh, Chief h Mechanical Engineering Branch Division of Engineering & Systems Technology Office of Nuclear Reactor Regulation
Enclosure:
As stated cc: Clive Callaway O __}}