ML20247M367
| ML20247M367 | |
| Person / Time | |
|---|---|
| Issue date: | 03/29/1989 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1724, NUDOCS 8904060040 | |
| Download: ML20247M367 (189) | |
Text
-
/?cO7= l729
/g..
O1G \\AL i
O UNITED STATES NUCLEAR REGULATORY COMMISSION
a ADVISORY COMMITTEE ON REACTOR SAFEGUARDS In the Matter of:
)
)
)
SUBCOMMITTEE ON MECHANICAL
)
COMPONENTS
)
O i
1 l
Pages:
1 through 157 Place:
Bethesda, Maryla h
()
Date:
March 29, 1989 f0i' tile _lieOi'ul0p0!ili i.e rit u
1 8904060040 890329 PDR ACRS T-1724 PDC O
HERITAGE REPORTING CORPORATION
/$
owaw-
\\
1220 L Street, N.W., Suke 600 0
Washinston, D.C. 20005 (202) 628-4888
1 1
UNITED STATES NUCLEAR REGULATORY COMMISSION 2
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 3
)
In the Matter of:
)
4
)
)
l 5
SUBCOMMITTEE ON MECHANICAL
)
COMPONENTS
)
6
)
7 Wednesday, March 29, 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 8:30 a.m.
12 BEFORE:
CARLYLE MICHELSON
(}
13 Retired Principal Nuclear Engineer Tennessee Valley Authority 14 Knoxville, Tennessee and Retired Director, Office for Analysis 15 and Evaluation of Operational Data U.S.
Nuclear Regulatory Commission 16 Washington, D.C.
17 ACRS MEMBERS PRESENT:
18 MR. CHARLEC WYLIE Retired Chief Engineer 19 Electrical Division Duke Power Company 20 Charlotte, North Carolina 21 ACRS COGNIZANT STAFF MEMBER:
22 E.
Igne 23 CONSULTANTS:
24 P.
Wohld 25 HERITAGE REPORTING CORPORATION -- (202)628-4888
2 1
NRC STAFF PRESENTERS:
2 B.
Baer O.
Rothberg
.3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 O
)
E'ERITAGE REPORTING CORPORATION -- (202)628-4888
3 1
-. _R _O C _E E D.I N G _S P
2 CHAIRMAN MICHELSON:
The meeting will now come to 3
order.
This is a meeting of the ACRS Mechanical Components 4
Subcommittee.
I am Carl Michelson, Chairman of the above ACRS 5
Subcommittee.
6 The other ACRS members in attendance are Jay Carroll 7
and Charlie Wylie, and ACRS consultant Peter Wohld.
The l
l 8
Subcommittee will continue its review of safety-related motor 9
operated valve testing and surveillance in order to assure 10 valve operability under design basis conditions for the life 11 of the plant.
12 Al Igne on my right is the cognizant ACRS staff
()
13 member for today's meeting.
The rules for participation in 14 today's meeting have been announced as a part of the notice of 15 this meeting that was published in the Federal Register on 16 March 15, 1989.
17 This meeting is b<ing conducted in accordance with 18 the provisions of the Federal Advisory Committee Act and the 19 government in the Sunshine Act.
20 We have received no written or oral statements from 21 members of the public.
It is requested that each speaker 22 first identify himself or herself and speak with sufficient 23 clarity and volume so that he or she can be readily heard.
24 Before we get started, I would like to just review 25 briefly for the Subcommittee where we are at.
We have met HERITAGE REPORTING CORPORATION - (202)628-4888
4 g g_
last I believe on February 28th to review this matter.
At 1
J 2
that time, it appeared that the staff was going to have some 3
further thoughts about the contents of the proposed generic 4
letter that covers this issue, and I believe the staff has, 5
well, the staff has sent to us recently a new version of the 6
generic letter, and I believe they have come here today 7
prepared to discuss it in detail.
8 The Subcommittee at the end of our meeting should 9
decide whether to bring this forward to the Full Committee for 10 their consideration, and if so, what the Full Committee 11 presentation should be like.
12 I would also like to get a reading from the
()
13 Subcommittee as to what the Subcommittee's recommendation 14 would be for the nature of a letter so at least I can prepare 15 a first draft of a letter for the Full Committee meeting which 16 is going to be next week at what time?
17 MR. IGNE:
April 7th, 2:15 to 4:15.
18 CHAIRMAN MICHELSON:
Okay.
Is that time agreeable 19 with the staff?
20 MR. BAER:
Yes.
21 CHAIRMAN MICHELSON:
At that time, we will make, 22 assuming the Subcommittee moves forward with this, we will 23 propose a full presentation to the Full Committee so that we 24 cari get a letter out at this, this meeting.
25 We have this morning some representatives of NUMARC HERITAGE REPORTING CORPORATION - (202)628-4888 i
5 1
which, who will have comments on this issue.
2 We also have in front of us a letter which was sent 3
to me by the Nuclear Utility Backfitting and Reform Group, the 4
letter of March 28th.
We should read that sometime between 5
now and the end of the NUMARC presentation so that we can ask 6
any kind of question.
Perhaps there are representatives of 7
this group here that we can ask questions of.
If not, then we 8
can have a Subcommittee discussion if one is needed.
9 Those are the only, that's the only outside input to 10' this meeting at this time as far as I know.
11 I believe then that takes care of my comments.
12 Charlie, do you have any comments at this time?
,( )
13 MR. WYLIE:
No.
14 CHAIRMAN MICHELSON:
Pete?
15 MR. WOHLD:
No.
16 CHAIRMAN MICHELSON:
Then I believe that we will 17 just proceed with the staff.
I believe Bob Baer is going to 18 kick it off and make the staff presentation.
19 I assume, Bob, that you will include CnGR in your 20 presentation?
Is that right?
You had a meeting I think.
21 MR. BAER:
We had the meeting before the February 22 28th.
23 CHAIRMAN MICHELSON:
You haven't had one since?
24 MR. BAER:
No.
25 CHAIRMAN MICHELSON:
I thought you were going back.
HERITAGE REPORTING CORPORATION - (202)628-4888
)
6
. cs 1
MR. BAER:
No.
We sent the letter back.
I will get 2
to that.
They have no problem with it.
3 CHAIRMAN MICHELSON:
It is going to be a little warm 4
in here today I suspect.
Certainly it doesn't seem like it is 1
5 cooling off any.
6 MR. BAER:
As Mr. Michelson indicated, the staff is 7
here this morning to discuss in situ testing of motor-operated 8
valves.
I have a slide that sort of outlines the presentation 9
we have prepared.
I am going to give an introduction and an 10 overall summary which I nill try and keep very brief.
11 (Slide) 12 MR. BAER:
And I will go through the eafety
()
13 significance from a risk perspective, and at that point, Owen 14 Rothberg, the task manager, will talk about the deterministic 15 bases for letter or the proposed action, some sensitivity 16 studies he did on cost / benefit analysis, address other 17 concerns brought up by the ACRS at a previous meeting, and 18 then go through the generic letter and in outline form answer 19 any questions you may have.
I 20 CHAIRMAN MICHELSON:
Okay.
Bob, do you want to hold 21 any detailed questions on the meaning of the generic letter 22 until we get to that last bullet?
23 MR. BAER:
Yes.
Prefer to.
24 CHAIRMAN MICHELSON:
Okay.
25 (Slide)
I HERITAGE REPORTING CORPORATION - (202)628-4888
7 1
MR. BAER:
As Mr. Michelson indicated, we met with 2
the ACRS approximately a month ago, February 28th, and they 3
expressed some concerns about the letter which we will attempt 4
to address today.
5 At that time, we had gone two CRGR and received 6
their comments which were in my opinion, well, we had already 7
redrafted the letter before going to CRGR to reflect the 8
results of the INEL test, and we got a bunch of detailed 9
comments from the CRGR that were largely format and tone 10 dealing with the tone of the letter.
11 We have made all those changes.
We have sent it 12 back to the CRGR.
We have been told that they are happy with
(}
13 the letter as written with a few very minor editorial changes, 14 and we are expecting a letter from CRGR confirming that in the 15 next few days.
16 From our point of view, the letter is ready to be 17 sent to NRR to be issued to licensees, and that's, of course, 18 subject to ACRS comments.
19 We do expect to or are hoping to go to the Full 20 Committee next week as indicated, and we will be requesting 21 ACRS comments and hopefully affirmative letter from ACRS as a 22 result of the Subcommittee meeting and that Full Committee 23 meeting.
24 Before I get into the details of the presentation, I 25 would like to try in just a couple of minutes to summarize the HERITAGE REPORTING CORPORATION - (202)628-4888 J
l l
l 8
g 1
staff views, and it is the staff view that there is a problem
-)
2 with MOV performance at operating reactors.
I will be 1
3 discussing that in a fair amount of depth today from both the l
4 risk perspective and a deterministic basis.
5 The letter we propose has carefully considered and 6
in most cases incorporated the comments we received from 7
industry, from other members of the staff, from ACRS, and from l
8 CRGR.
l 9
We believe that the action we are proposing will l
10 significantly improve plant safety, and although we also l
11 believe that one can justify the letter purely on 12 deterministic basis, we did perform a backfit analysis
()
13 previously in which we looked at, we look at cost and did a 14 sensitivity study which Owen will be discussing a little bit 15 later.
16 We have concluded that the proposed action fully 17 satisfies the intent and requirements of the backfit rule, 18 51--10 CFR 51.09, and that the actions will result in 19 substantial increase in the overall protection of the public, 20 and secondly, the costs are justified in view of the, of that 21 additional protection, and I guess we are going to try and 22 make that case today.
23 CHAIRMAN MICHELSON:
Bob, in our previous meeting, I 24 think way back in October, we had some problems at that time
}
25 you remember with the work that Brookhaven had done, the HERITAGE REPORTING CORPORATION - (202)628-4888 j
9 1
numbers they had,come up with versus the industry's claim for 2
the numbers.
3 Have you changed any of that since?
4 MR. BAER:
We have done some sensitivity studies on 5
the cost, and I am going to go talk about risk part as the 6
next part of the presentation.
7 CHAIRMAN MICHELSON:
Okay.
You will try to bring us 8
up to, you know, any modifications of that study you will tell 9
us about this morning?
10 (Slide) 11 MR. BAER:
Let me talk a little bit now about the 12 safety, what we see as the safety significance or the risk 13 perspective of the MOV problem, and what I am really going to
{}
14 do is try and summarize the work Brookhaven did.
15 They calculated a core melt frequency.
They use the 16 SARAH code and they based this on 1150 results for two plants, 17 Surry and Grand Gulf, and they varied the failure rate from 18 2.9 percent to 29 percent, and they used the best estimate 19 value of 8.7 percent, and that is, although they did not 20 obtain that from the results of the IE Bulletin 85-03, because 21 they were not very. not many of those results were available I
22 at that time, it turns out that their best estimate number is 23 very close to the fail.ure rate reported by the licensees.
24 Before I get into the results of the calculation, I j
25 guess I would like to try and address some points that have i
HERITAGE REPORTING CORPORATION - (202)628-4888
10
(
l 1
come up during past discussions, and the first thing T guess I 2
want to make a point is that I don't think anyone will look at l
l 3
risk studies or PRA studies as a go, no-go gauge.
l 4
I think, however, they do provide a great deal of i
5 insight regarding the potential risk significance of a l
l 6
particular issue or concern, and but like any other study, 7
there are certain limitation, certain aspects that are quote, 8
conservative, and certain aspects that are non-conservative, 9
and I want to point out some of these that do exist in the 10 Brookhaven study, and which I think was an excellent piece of 11 work, but they had to make certain simplifying assumptions and 12 some of these, and I think the majority of the assumptions 13 leads one to believe that the actual core melt frequency may
}
14 be higher than that predicted by Brookhaven rather than lower, 15 but the first assumption they made is that all the valves that 16 were covered by Bulletin 85-03 have been fixed and they have a 17 very low failure rate, roughly four tenths of 1 percent, which 18 is the value that was used in the 1150 studies.
19 Now there has been a fair amount of data recently 20 that indicates that this may not be the case, but that, that 21 was within, that assumption was made in the Brookhaven study, 22 and in fact the Brookhaven study said the greatest risk 23 reduction occurred from 85-03 by fixing those valves in the 24 aux feed system and high pressure injection in the PWR and the 25 RCIC and HPCI in the PWR.
1 HERITAGE REPORTING CORPORATION - (202)628-4888
.-___-____-_______a
1 11 1
The data probably will become available from the 2
INEL test which was not available at the time the Brookhaven 3
did this study and that was not factored in.
INEL has some 4
information that says that there may be, and I emphasize the 5
may, possible systematic problems in sizing actuators on 6
setting torque switches, that the basic equations used by the 7
valve manufacturers and by the operator manufacturer may not 8
be conservative, may be non-conservative in fact.
That was 9
not factored into the Brookhaven study.
10 The Brookhaven study also did not consider common 11 mode failures other than design failures, common mode of 12 either type; talked about the design failure, design type
()
13 common mode, but there is obviously other common mode failures 14 that result from improper maintenance, improper adjustments, 15 and erroneous or improper installation, and none of those were 16 factored into a PRA study.
17 It was just that the PRA study that varied, as I 18 indicated, the failure rates over a fairly wide range.
19 On the other hand, the Brookhaven study pointed out 20 to us, and they recognize this also, they possibly cou'Id 21 overpredict core melt frequencies because they did not 22 consider the possibility of recovering from an event by manual 23 operation of a valve, and that is obviously a possibility.
24 High stress conditions, and there is possibilities
}
25 of people making a further error, but certainly there is the HERITAGE REPORTING CORPORATION - (202)628-4888
11
,fm, 1
possibility of recovering from a valve failure by gcing down
()
2 and manually operating the system, and that was not factored 3
into this study, so those sort of things have to be I think 4
considered in making a judgment as to how to tr'*t or how to 5
use the results of the PRA study.
6 (Slide) 7 MR. BAER:
Let me talk a little bit about the 8
results of the BNL study.
First it would probably make a 9
little more sense to at least briefly talk about the failure 10 rates and how the range, and they did cover a wide range, was 11 derived.
12 NUREG 1150, as most PRAs, use a failure rate of
()
13 roughly four tenths of a percent per demand, and on a couple 14 of occasions, I and other people in my branch and in fact the 15 branch I headed up previously have tried to find out the 16 source of that data base, and we really have not been 17 successful.
It just seems to be a number that has come around 18 from a, been around for a long time, and i don't think it, 19 well, I am quite sure it doesn't have a firm basis.
20 NPRDS data which is done under non-flow, usually 21 done under non-flow results of Section 11 testing, done under 22 no flow, no delta P conditions, shows a failure rate of 23 roughly three times, two and a half to three times that under
]
24 the most favorable conditions, not under stringent event 25 conditi ons, but anyhow that's what NUREG 1150 used.
l HERITAGE REPORTING CORPORATION - (202)628-4888
13 1
Now go down to the best estimate--
7-(>
t 2
CHAIRMAN MICHELSON:
Excuse me.
NUREG 1150 doesn't 3
use the NPRDS number?
4 MR. BAER:
No.
It used the--
5 CHAIRMAN MICHELSON:
The traditional number?
6 MR. BAER:
Yes.
I am going to try and get some feel 7
for this.
We have looked at NPRDS data and made an estimate 8
of the number of valves and the number of times that they are 9
exercised, and we don't see'a number anywhere near as low as 10 four tenths of 1 percent.
11 CHAIRMAN MICHELSON:
One of our fellows, Mark 12 Stella, has been looking into this for you.
I don't know if
(}
13 he has talked to you much about it.
He is coming up with some 14 of the similar observation that indeed the origin of some of 15 these numbers is very unclear, but apparently most of them 16 just go back to numbers chosen during WASH 1400, so he is 17 trying to figure out how they chose them during that time 18 period, and they were not necessarily the strong technical 19 numbers then, either.
They kind of pulled them out of the air 20 apparently to some extent, but we are still awaiting his final l
21 report.
We will send you a copy when he finishes I guess for 22 your own edification.
23 MR. BAER:
I would be interested in that.
I know 24 O' fen Rothberg had had some conversations I think 25 with--sometime ago.
HERITAGE REPORTING CORPORATIOIT - (202)628-4688
14 1
CHAIRMAN MICHELSON:
I am curious now about similar 1
2 numbers for a number of other components in the plant where if 3
we are going to place great reliability upon the PRA and if, 4
you know, if the numbers in the data bases aren't too good, 5
then one has to wonder a little more about the believability 6
of the bottom line.PRA numbers where you are dealing with 7
sensitive components that are sensitive to the bottom line.
8 MR. BAER:
Right.
I certainly concur with that 9
statement.
10 Brookhaven for their analysis looked at the best 11 estimate value of 8.7 percent failure rate.
They looked at 12 the MOVATS data as, that largely was obtained during 85-03
(}
13 that MOVATS was willing to share with Brookhaven the data 14 without identifying the specific licensees, and depending on 15 the interpretation, their first interpretation was a higher 16 failure rate, more in the order of 30 percent, but it does 17 turn out that this is largely in terms of just torque switch 18 missettings, that you do have to have both misset, in order 19 for the valve not to either open or close, the torque switch 20 has to be underset and has to be not bypassed.
21 I always thought that maybe the original 22 interpretation was correct because I doing--like the fact that 23 if maintenance people made two errors, then the valve may 24 work, but anyhow, they concluded after further talk, l
25 discussion with MOVATS, that the data at the time seemed to HERITAGE REPORTING CORPORATION - (202)628-4888
15 1
support a failure rate on, again under the relatively high 2
delta P conditions of 8.7 percent.
3 I should say you determine high load because I want 4
to make the distinction that although 85-03 dealt with valves 5
that could be potentially subjected to high differential 6
pressures, that many of the valves in the rest of the plant 7
can have certainly moderately, at least moderate differential l
8 pressures and a lot of them are much bigger valves so the 9
loads involved can be quite large even though the--at moderate 10 delta P even though because of the large diameter, the force 11 goes up as the diameter squared and linearly with the pressure 12 differential.
()
13 As a high estimate, they used 29 percent, which I 14 think probably covers the ballpark pretty well, and they used 15 one 10th of that or 2.9 percent for a low estimate.
As I 16 indicated before, the 85-03 results that Dick Kiessel and NRR 17 has been compiling as they have been coming in, and they, they 18 stay pretty steady at 8 to 8 and a half percent creeping 19 upward as more and more plants report data, and so that very 20 much supports the 8.7 percent.
21 Some licensees have voluntarily expanded the 85-03 22 type requirements to other safety-related valves, and I get in 23 one case, one plant, maybe to all valves, and the data, the
("N 24 information we have, they have indicates they were observing u) 25 again about the same sorts of failure rate that one wouldn't HERITAGE REPORTING CORPORATION - (202)628-4888
l 16 1
expect baseo on what we are seeing, that the failure rate for f-N )S 2
these valves that are subjected, potentially subjected to only 3
moderate, what I will call moderate dealt Ps, will show 4
markedly different failure rate.
5 CHAIRMAN MICHELSON:
Bob, where do the numbers come i
6 from for NUREG 1150 since that is supposed to be state of the 7
art PRA, best data base, all the other good stuff?
8 MR. BAER:
I think, just as you said, I think these 9
numbers have been carried on since WASH 1400 days.
10 CHAIRMAN MICHELSON!
So NUREG 1150 is not 11 necessarily believable in this regard then, either?
12 MR. ROTHBERG:
Reference NUREG CR 1363, and I have
(~)'n 13 seen a number of around 3.78 or 4 times ten to the minus 3 14 MOV failure rate, that covers a lot more than just MOVs.
15 CHAIRMAN MICHELSON:
Sometimes these numbers are not 16 jiving very well with this experience.
17 MR. ROTHbERG:
No, sir.
18 MR. BAER:
Someone wanted to comment.
19 CHAIRMAN MICHELSON:
Is there a comment bacI there?
20 MR. CURRY:
Brian Curry from Philadelphia Electric 21 Company--I recently found a study IEEE 500-84, and it gives a 22 number of values for reliability of pumps and valves, and what 23 the basis of it is, but it recently came to my attention, so 24 you ought to take a look at that further.
That may provide 25 some guidance on where the basis is.
HERITAGE REPORTING CORPORATION - (202)628-4888
c f
17 1
CHAIRMAN MICHELSON:
Do you know if there is a, do O
2 you recollect any numbers in there to get a feel for--
3 MR. CURRY:
I can't give you a number.
I was, only 4
became aware of about it last week.
5 CHAIRMAN MICHELSON:
We will try to chase a copy 6
down then.
Thank you.
7 MR. WOHLD:
I would like to ask a question.
These 8
failure rates, it is my understanding these are primarily 9
related to switch settings?
10 MR. BAER:
I think that's, that was true BNL, the 1
11 data that BNL used, but as I said earlier, I think one maybe 12 ought to say that hey, we looked at a sensitivity study i
13 varying the failure rate from just under 3 percent to 30 l
14 percent to see what that meant in risk percent.
At least 15 that's how I would categorize the results, rather than to try 16 and say that any one of these numbers is exactly right, but I 17 think you are right, that they look at MOVATS data primarily 18 on switch, switch settings, because they add up--all the many i
19 valves had multiple failures, and as a result, if you just 20 added up failure rates, you started in some cases coming up 21 above a hundred percent, which is not obviously correct, just 22 individually added up failures due to switch settings and 23 several failures due to other causes.
24 CHAIRMAN MICHELSON:
I think I saw just a little 25 while ago that you have done some looking at data that i
HERITAGE REPORTING CORPORATION - (202)628-4888
18 1
included these other causes and you were still ending up about 2
8 and a half percent?
Is that correct?
3 MR. i:OTHBERG:
That's the slides.
4 CHAIRMAN MICHELSON:
Add in these other things other 5
than switch settings, it is not changing that 8.7 percent 6
number very much apparently?
7 MR. BAER:
The other thing that I mentioned was the 8
35, the results reported by the licensees for 85-03, when you 9
add up all the valves, all the situations that they call the 10 valve inoperable, and I guess in some cases you applied some 11 judgment.
If they showed a marked change in switch setting 12 and didn't call it inoperable, I guess you put it in the
()
13 inoperable column?
14 MR. KIESSEL:
It is only if they said it was 15 indeterminate that the valve was operable, then I went back 16 and looked at what kind of switch settings they had changed.
17 And if there was an order of magnitude change, kind of thing, 18 I assumed the valve would have been inoperable if they had 19 really tried it.
20 MR. BAER:
But most cases, am I enrrect in saying in 21 most cases in your calculation where you called it inoperable, 22 it was what the licensee had said it was inoperable?
23 MR. KIESSEL:
Correct.
24 MR. BAER:
When you add up the number of quote, 25 incperable valves, divide by the total number of valves, I l
I HERITAGE REPORTING CORPORATION - (202)628-4888
19 1
think the latest number is 8.4 percent or 8.44 percent.
~O 2
CHAIRMAN MICHELSON:
Then there is other factors 3
that aren't necessarily in there yet which make it even worse 4
such as degraded voltage conditions you might not see during 5
normal testing but might see during an event, and that sort of 6
thing is just making it even less reliable.
7 MR. BAER:
There have been plants that reported no 8
failures that have been shut down or have had significant 9
problems subsequently on some of those same valves, 10 Well, in any event, the NUREG 1150 numbers for the 11 PWR that was selected, Surry showed a failure, an overall core 12 malt. probability over here of 2.4 times ten to the minus 5, 13 and the comparable number for Grand Gulf was 2.9 times ten to
)
14 the minus 5, again based on that the 3.8 percent valve failure 15 rate.
16 CHAIRMAN MICHELSON:
Those numbers are also using 17 under the abnormal loads of events?
Isn't that the same 18 reliability number?
For instance, closing a HPCI steam line, 19 I suspect they are using that same No. 2.9 ten to the minus 5, 20 MR. BAER:
Or failure rate of four tenths of a 21 percent.
22 CHAIRMAN MICHELSON:
Yes, which--
23 MR. BAER:
Gave this, this is just purely the 1150 24 results, and as far as, at least my understanding is that 25 throughout--
HERITAGE REPORTING CORPORATION - (202)628-4888
l 20 l
1 CHAIRMAN MICHELSON:
Excuse me.
Yes; 3.,
- yes, U-l 2
the.38" percent is used, they were using.
1 3
MR. 23ER:
In the Brookhaven study, and it was not 4
a, no attempt to red,.
1150.
Everyone I think in the room 5
knows some of the problems there.
They merely varied valve l
l 6
failure rate over a range to see and look at the major 7
sequences and said what would be the result in terms of core 8
melt frequency?
And you can see the results tabulated here 9
for going through the PWR for a low estimate of 2.9 percent.
10 That was a change in core melt frequency up to 2.7 11 times ten to the minus 5 because the results of the change 12 would be the delta or.3 times ten to the minus 5 as the r
13 failure rate went up, again thinking this is a sensitivity 14 study, to their best estimate of 8.7 percent, you now see an 15 order of magnitude.
16 MR. MINNERS:
No--high.
17 MR. BAER:
I'm sorry.
You are right.
I jumped.
As 18 you go over here to 8.7 percent failure rate, you see a 19 further increase in core melt probability so that it is now a 20 delta of nearly five times ten to the minus 5, and if you use 21 the high valve failure rate, subject to the limitations I 22 mentioned before, you would now see an order of magnitude 23 increase in core melt frequency.
24 And li kewise, on the BWRs, the effect was somewhat 25 greater, that even the low estimate gave a roughly 3 times ten HERITAGE REPORTING CORPORATION - (202)628-4888
21 1
to the minus 5 delta and core melt best estimate was roughly
- 3 l
%.)
l 2
an order of magnitude, and the highest then would be even 3
another order of magnitude, and--for the Grand Gulf reactor, 4
and the message got, I get out of this is that this is an 5
important safety consideration, and as you said, Carl, maybe 6
there is a serious underestimating of core melt frequency in 7
PRA studies like 1150 if as we believe, the valve failure 8
rates that they were using are too optimistic.
9 That was about all I was going to say on the risk 10 perspective.
If there is no other questions, I will turn it 11 over to Owen for the reme'nder of the presentation.
12 CHAIRMAN MICHELSON:
Now let me ask, Bob, the
()
13 Brookhaven study included cost numbers and so forth.
Are you 14 going to discuss that later?
15 MR. BAER:
Owen is going to, although we did what I 16 would call another sensitivity study there, Owen did just 17 recently.
18 MR. ROTHBERG:
I am going to talk about the 1
19 deterministic basis for our generic letter first, and I would 20 like to add before I get started for the revurd thet this 21 deterministic basin is described in some detail in our package 22 that we sent to CRGR on August 23rd, 1988.
It was Section No.
/
23 8609150045, and this is why it discusses some of the same 24 things in this first bullet that I hrre about the 25 deterministic be. sis.
HERITAGE REPORTING CORPORAT7'" - (202)628-4888
22 1
(Slide)
O 2
MR. ROTIIBERG :
By regulations, and those GDC 3
criteria, essentially 21 is kind of pushed over to the side, 4
but that last number should be 21.
The GDC essentially gives 5
us guidance that safety-related equipment by virtue of the 6
fact that licensees make the commitment that these items vill 7
perterm, have to perform, and they have to be tested 8
meaningfully to show that they vill perform.
9 Specifically 10 CFR 50 Appendix B Criterion XI says 10 that, and it says that the test program must show that the 11 equipment will perform under design basis conditions.
12 Now the current testing that Section 11 performs, 13 the stroke timing test, doesn't give you assurance of future 14 operability because it does not tell you about degraded 15 conditions that occur within the valve or the motor operator.
16 What it tells you is that at this particular time, this 17 particular device moved within a specific time period.
l 18 Pete has gone through, as you know, exercises in 19 varying the allowable time that ys 1 might use and making the 20 test more meaningful, but as the test now stands, the stroke 21 timing test doesn't give you any assurance of future l
22 operability because you don't know when you have made that 23 test whether the valve will perform the next time.
24 There is the possibility, and that's been discussed 25 in several AEOD reports, of leaving a motor-operated valve i
HERITAGE REPORTING CORPORATION - (202)628-4888 i
23 1
inoperable after a stroke timing test just because of this 2
phenomenon.
The test itself isn't oriented toward any sorts 3
of diagnosis of internal problems, and also it doesn't predict l
4 degradation.
It doesn't tell you whether this particular 5
motor operator is degrading in place.
6 A good example is valve stem packing, have been 7
adjusting packing to prevent a mitigate a leak.
That puts 8
additional loads on the motor.
Those additional loads may be 9
either degrading the motor or if one does a really bad job on 10 the packing, can actually disable the motor-operated valve.
I 11 don't think anybody argues with that.
12 The stroke timing test, if the packing adtis ment is 13 done after the stroke timing test, again, you have no 14 assurance of operability.
15 CHAIRMAU MICHELSON:
Before you leave that slide, 16 let me.ask particularly on the first two bullets there, first 17 bullet establishes that, you know, these components have to 18 work.
They have to, need certain GDCs, has to be demonstrated 19 indeed they can function for those GDCs.
20 Then the second bullet talked about a particular j
21 test which I believe, correct me if I'm wrong, I believe that 22 test is accepted by, not by regulation, but certainly by--
23 MR. ROTHBERG:
It is not regulations--10 CFR 5055 24 AG.
25 CHAIRMAN MICHELSON:
Okay.
And now that test has j
HERITAGE REPORTING CORPORATION - (202)628-4888
24 1
proved to be inadequate.
The question that arises then is
)
%s!
2 well, clearly to meet GDC we have cet to do something better 3
than we said to do in the regulation, and that's something 4
better.
Is that a backfit then or what?
What takes 5
precedence in the regulation?
6 The fundamental requirements such as the GDCs or the 7
requirements that are prescribed such as Section 'l or how--
8 MR. ROTHBERG:
I'm not sure.
I guess a lawyer might 9
be able to tell you better than I.
10 CHAIRMAN MICHELSON:
Isn't that the crux of the 11 question about whether you have to consider this a backfit?
12 MR. ROTHBERG:
Yes, and as I pointed out, we did a
()
13 backfit analysis on this.
We said we didn't consider it to be 14 a backfit, but we did a backfit analysis nonetheless.
15 CHAIRMA" MICHELSON:
It is not clear to me that what 16 you do when you have written two, two parts in a regulation, 17 one a fundamental prescription and then another a particular 18 test to be done, and then you find out the test isn't any 19 good.
The fundamental prescription hasn't changed any.
The 20 requirement to meet GDC is the same.
It is just that you 21 picked a bad test to meet it.
22 MR. ROTHBERG:
That's true.
23 CHAIRMAN MICHELSON:
Now if you do that, is it a I
24 backfit to pick a better test?
25 MR. BAER:
Let me--I can't, I think Owen gave you HERITAGE REPORTING CORPORATION - (202)628-4888 L.
25 1
the right answer.
I think it is a legal rather than a 2
technical question, but there is a--the section that refers, 3
the part of the regulations that refer to ASME Section 11, 4
there is a precedent.
One came up on the BWR crack.
In fact, 5
the original reference to Section 11 dealt with in-service 6
inspection or concept was in-service inspection rather than 7
in-service testid?, and the same thing happened in terms of 8
the adequacy.
9 It was determined on the BWR reactor coolant system 10 the ASME Section 11 in-service inspection was not sufficient 11 to detect intergranular stress corrosion cracking and so 12 additional requirements were imposed.
13 CHAIRMAN MICHELSON:
Without a backfit?
[
14 MR. BAER:
Well, that was before the latest version 15 of 5109.
I am saying from a technical point of view, when you 16 find something that is wrong, I think there are mechanisms and 17 precedents for requiring additional action.
18 CHAIRMAN MICHELSON:
Has OGC gone, stirred through 19 this thing and reached a judgment on--
20 MR. ROTHBERG:
I believe one of the things--
21 MR. BAER:
They concur in the letter, and they 22 represented on CRGR.
23 CHAIRMAN MICHELSON:
We are being wishy-washy.
We 24 are saying we, well, don't think it is backfit, but we are 25 going to do a backfit analysis anyway.
That I way I don't HERITAGE REPORTING CORPORATION - (202)628-4888
26 t
,-)
know whether you think it is a backfit or not.
'If you really i
1 V
2 didn't think it was a backfit, why are you doing backfit?
3 MR. MINNERS:
What difference does it jaake, Carl?
4 CHAIRMAN MICHELSON:
Well, the backfit analysis, 5
that's getting us into a lot of arguments about cost and 6
benefits, so forth, which I think you made a very good case of 7
so far, but I think that on a deterministic basis, there is a 8
good case for doing this fix.
9 MR. MINNERS:
Let me tell you what we do.
The 10 instructions that we give to the staff is that basically you 11 are engineers, and you are supposed to be doing a technical 12 job, which means you look at the risks that can be incurred by
()
13 these things and the costs to fix them.
That's what an 14 engineer does, but you don't leave it at that because the good 15 engineer not only looks at his numbers that he is trying--he 16 looks at other qualitative factors, okay, and T think most of 17 the rules are that kind of a thing.
They are qualitative 18 factors.
You ought to have valves that work.
I mean that's a 19 pretty basic thing in your head, so I don't think that the 20 staff worries too much about the legal niceties of things.
We 21 don't check with the lawyers that we are not doing something 22 legal, but we are engineers, and we try to do it in a 23 technical engineering manner, and I think that's what we are 24 trying to present to you.
25 MR. ROTHBERG:
If I cauld get off my leash, I could HERITAGE REPORTING CORPORATION - (202)628-4888
27 1.
probably look up in 10 CFR 5055 and find a paragraph I think fs t )i 2
in there that says that the staff may impose additional tests l
3 where they find that that's necessary.
4 CHAIRMAN MICHELSON:
I don't doubt that.
5 MR. ROTHBERG:
And then in 51.09, what it does, it 6
says if it is or isn't a backfit, even if it is a backfit, a 7
backfit analysis doesn't have to be performed on the, under 8
certain conditions, and I hesitate to paraphrase the words, 9
but I believe it has to do with making a case for safety.
10 CHAIRMAN MICHELSON:
You never went through that 11 particular ritual or I don't believe.
I didn't see the 12 documents that indicated--
(}
13 MR. ROTHBERG:
What I did was I did the backfit 14 analysis, because I was so sure of the case for the 15 cost / benefit, and I wantod to make sure that all bases were 16 covered.
i 17 As it turns out, I have been criticized for doing 18 both, and I don't think I should be.
19 CHAIRMAN MICHELSON:
You realize, of course, it is 20 the backfit analysis and those costs that we are arguing about 21 and I think it doesn't really make a bit of difference.
I am 22 convinced of the technical merits of the case.
I am just 23 wondering why we even get into this question of whether it is q
24 a backfit or not and so forth and why you are doing the kind
'%./
25 of cost / benefit analysis like Brookhaven did.
HERITAGE REPORTING CORPORATION - (202)628-4888
28 a
1 MR. MINNERS:
Because I insist upon it.
If you do a g3U 2
proper engineering analysis, you have to look at risk-and you 3
have to look at the cost.
I don't care if somebody came up 4
and told me you have to do this for compliance with the rule, 5
I would still make them do a cost / benefit analysis, and if the 6
cost / benefit analysis came out unfavorable to me, I would say 7
hey, the problem is we ought to change the rule and not 8
enforce it.
So we are engineers and we are not lawyers and we 9
look at it from an engineering point of view, and if what we 10 want to do is at variance with the rules and regulations, we 11 will then go to the proper procedures to change that.
12 CHAIRMAN MICHELSON:
The technical fix may get bound
()
13 up in the legal questions.
14 MR. MINNERS:
There is no problem.
The lawyers will 15 tell you that if the rules are wrong, you can change them.
16 There are procedures for doing that.
17 CHAIRMAN MICHELSON:
Why don't they just leave it at 18 that?
19 MR. MINNERS:
At what?
20 CHAIRMAN MICHELSON:
The rule, if Section 11 doesn't 21 do the job, fix it.
22 MR. MINNERS:
Section 11 isn't wrong.
It is just 23 insufficient.
I don't think--nobody is suggesting not doing 24 the stroke testing.
Nobor.y is suggesting not doing the stroke
{
25 testing.
There is a suggestion you supplement that.
HERITAGE REPORTING CORPORATION - (202)628-48R8
29 1
CHAIRMAN MICHELSON:
It is the supplement that's--
2 MR. POINDEXTFR:
Excuse me.
My name is Tom 3
Poindexter from NUBARG, the backfitting reform group.
4 Our main concern is not whether it is or isn't a 5
We are concerned about just following the regulation 6
and coming up with a reliable value impact statement such that 7
the schedule is a valid schedule.
NUBARG doesn't dispute 8
whether this is a worthwhile thing to do on safety, but we are 9
concerned that the schedule may not be realistic because the 10 cost analysis is not that accurate.
That's our main concern.
11 CHAIRMAN MICHELSON:
The costs have nothing to do 12 necessarily with the schedule, does it?
13 MR. POINDEXTER:
The more hours you put in, the more
{}
14 you spend.
There is a relationship between those two.
15 CHAIRMAN MICHELSON:
I think what you have to do is 16 look at technical merits of the argument ar.d determine how 17 fast you have to tix it.
18 MR. POINDEXTER:
Yes, but if that--
19 CHAIRMAN MICHELSON:
Cost is the major 20 consideration.
There are--practicality is certainly a 21 consideration.
You have got to have time to do it.
22 MR. POINDEXTER:
That's what we are worried about.
23 CHAIRMAN MICHELSON:
I sense that in the letter, of 3
i 24 course, but I also sense that you thought they really hadn't O
25 done a proper economic evaluation, and I'm not sure that j
HERITAGE REPORTING CORPORATION - (202)628-4886
1 l
30 l
1 that's unnecessary if indeed there is a technical deficiency
-)
V 2
in the way the valves are working.
3 MR. WOHLD:
I guess in a way the Section 11 testing l
4 is necessary because I don't think it was ever identified in a i
5 regulation that Section 11 was sufficient.
It was one of the j
6 requirements.
7 CHAIRMAN MICHELSON:
Only one.
8 MR. ROTHBERG:
One of the things that is in the 9
change letter is a paragraph discussion of the limitation and 10 advantages of the Section 11 test, and that's I believe right 11 at the beginning of the discussion.
12 MR. POINDEXTER:
If it is determined it is truly a
()
13 backfit, then the safety analysis really should be prepared so 14 that industry has a chence to review that.
15 MR. ROTHBERG:
We have done that.
I gave you--
16 MR. BAER:
I would like to address that point 17 because I think one, we have done, Owen will present some 18 sensitivity studies where we have increased cost, and I still 19 think that we show a very favorable cost / benefit ratio, and 20 industry ends up with two different opportunities despite the 21 cost / benefit analysis.
22 This is done, Owen tried to do it on a generic 23 basis, which is difficult.
If any--the generic letter that we 24 are sending out is a request for information.
A licensee can 25 come back and say no, we don't think this applies to our plant HERITAGE REPORTING CORPORATION - (202)628-4888
31' 1
for the following reasons, and then the staff is obligated to j-'u) 2 do a plant-specific backfit for that plant, but I think we are 3
going to show that even with, or Owen will be showing that 4
even with large increases from the cost that Brookhaven used, 5
the cost / benefit ratio that we end up with is clearly.much 6
lower than a thousand dollars a man-rem, so I think the, I 7
really think we have met the test of 51.09 that there is a 8
substantial safety benefit and that there is a cost favorable, 9
favorable cost / benefit ratio, and I think we will try and 10 demonstrate that, and obviously people can dispute what we 11 said, but I think we are putting aside the philosophical 12 question as to whether it is legally necessary for us to do
()
13 it.
If we try to make adequate protection, like Warren says, 14 that's a messy place for engineers to try and tread.
I do j
1 15 think we do satisfy at least on a generic basis, the 5109 16 requirements.
17 MR. POINDEXTER:
Does the staff feel this is a 18 backfit, just so I am clear?
1 j
19 MR. BAER:
You will have to get a lawyer to answer I
20 that.
I think we have given our answer.
We think it is a 21 deterministic basis for requiring it, but we also think that 22 it passes the backfit type test.
23 MR. POINDEXTER:
Okay.
Thanks.
24 MR. ROTHBERG:
One thing I might point out is that 25 even though the Section 11 exists as a requirement as far as HERITAGE REPORTING CORPORATION - (202)628-4888
32 I
1 regional inspection goes, when there was something that went
(
)
2 beyond the reciuirements of Section 11 there was no hesitancy 3
to apply all of the Appendix B criteria ever since, you know, 4
Appendix B was created, and so applying the rules of Appendix 5
B is nothing new, so I personally don't think that this is a 6
7 I consider Section 11 necessary, but not sufficient, 8
and I think that's the way the regions look at it.
9 MR. MINNERS:
I think one of the purposes of 10 presenting this is it shows that we think we have the legal 11 authority to impose improvements in this area.
If there 12 are--with no rules at all which talked about testing, okay, I
/~';
13 think it would be improper t..
Jo the generic letter route and V
14 you would want to go maybe rule and regulation, but but we 15 think that we have rules which say hey, this is what you are 16 supposed to do, and all the generic letter is saying is they 17 are saying this is how you should do it.
18 CHAIRMAN MICHELSON:
Why don't you proceed?
19 MR. ROTHBERG:
Sure.
There have been, in addition 20 to the things we have discussed, there have been a number of 21 re,7 orts which point to the problems with motor-operated 22 va'.ves, and they are mentioned here, and they have been 23 occurring right up until the present day.
24 The AEOD reports are referenced.
We have these 73
(-)
25 recent test programs at Brunswick, Fermi and Surry.
The HERITAGE REPORTING CORPORATION - (202)628-4888
33 8
1 Bulletin 85-03 responses, as Bob pointed out, show that our 2
failure rates are compatible with what we originally showed.
3 CHAIRMAN MICHELSON:
.Give, in a nutshell tell us the 4
status now of the responses.
5 MR. ROTHBERG:
Excuse me?
6 CHAIRMAN MICHELSON:
Could you tell us in a nutshell 7
the status of the responses to 85-03?
8 MR. ROTHBERG:
Yes, sir.
I think they are in your 9
slide in the package that I gave you, and there is a summary 10 which I got from Dick Kiessel, three-page summary of final 11 report status of 85-03, and on the last page of that, there is 12 three paragraphs down at the bottom to caution, and then it 13 says for the 95 plants reporting to date, 7.96 is the average
{}
14 of the individual failure rates in percent; 15.35 is the 15 standard deviation, and 8.44 percent is the number of failure 16 rates in percent, that he took the number of inoperable valves 17 and divided it by the total number of valves.
l 18 CHAIRMAN MICHELSON:
Ninety-five plants are reported 19 to date.
Does that mean that 95 plants have finished their
{
20 work on 85-03?
)
21 MR. ROTHBERG:
I really don't know.
Dick might be 22 able to answer that question.
23 MR. KIESSEL:
They feel they have.
It does not mean 24 that the inspections have been completed or--
25 CHAIRMAN MICHELSON:
I just wondered because in our HERITAGE REPORTING CORPORATION - (202)628-4888
o 34 l.
I 1
new generic letter, we talked about not having to do any more g
l
-)
2 reporting.
I wondered how much has already been done, and 3
this seems tn infer that roughly 95 percent of it has been 4
done.
l 5
MR. KIESSEL:
No, sir.
6 CHAIRMAN MICHELSON:
No?
7 MR. KIESSEL:
This is only a bullet.
The generic 8
letter which does not--
9 CHAIRMAN MICHELSON:
I am talking about the Bulletin 10 85-03.
i l
11 MR. KIESSEL:
Okay.
The licensees feel that they 12 have done 95 percent.
()
13 CHAIRMAN MICHELSON:
Roughly it is nearly all done 14 now?
As far as the licensees are concerned, 85-03 is nearly 15 all done?
16 MR. KIESSEL:
Ninety-five out of a 127, 17 MR. ROTHBERG:
What I would like to talk about now, 18 and this is the question that you have asked t'ecently, is our 1
19 cost / benefit analysis, did we address some of the questions?
20 This slide refers to a handout which is the last few 21 pages of your handout entitled an evaluation of the potential 22 effects of an MOV prototype test program, et cetera.
If you
)
1 1
23 could turn to that, we could discuss it, and I will use this
]
l
/
24 slide to illustrate that.
I 25 (Slide)
HERITAGE REPORTING CORPORATION - (202)628-4888
35 i
1 MR. ROTHBERG:
We were asked what the effect of an O
2 industry-wide prototype test program that seemed to be 3
required by the Generic Issue 87 results might be, and so I 4
went and looked at that, and what I did was the first thing I 5
did was found out that it approximately costs $250,000 to test 6
the INEL valves each, and I used that as a starting point, and 7
then I decided well, the number of valves to be tested in the 8
program, what I did was I found out that we, in the EPRI 9
Marshall test, they tested seven valves.
We tested two more.
10 In these INEL tests it seemed reasonable to say that an order 11 of magnitude increase in the number of valves tested would be 12 about right, but I mean ten to a hundred, so I said a hundred.
[}
13 I wanted to be conservative, so I used $250,000 per MOV and a 14 hundred MOVs as a best estimate.
15 Now the numbers both high and low are used in the 16 same way that NUREG CR 5140 does.
The high and low are used 17 in the algebraic sense, so a low number is more negative, and 18 what it refers to is--and since negative numbers are costs, so l
l 19 you may have a, a numerically higher and an absolute value 20 higher under the low than high, and that's why they come out 21 that way.
That's why these headings are the way they are.
22 This is consistent again with NUREG 5140.
In any case--
23 MR. BAER:
Basically what BNL did was took the 24 lowest safety benefit and divided it by the highest cost, and 25 that's what--
(
HERITAGE REPORTING CORPORATION - (202)628-4888 I
36
)
1 MR. ROTHBERG:
This is essentially what we are g
2 doing.
It was.
$25 nillion was our best estimate for an 3
industry-wide test program in any case, and the range was from 4
ten to 47 million dollars.
That's how we arrived at that.
5 Another question that we were asked about was it 6
seemed like NRC was going to have to do somewhat more because 7
we would have to do these audits, and go around to the plants.
8 We would have to train some people, so this is where the 9
sensitivity study comes in, and on the next page of this 10 thing, which is entitled a sensitivity analysis, from NUREG CR 11 5140, what I did was if you will look in there, you will 12 see--I don't have a slide for this, but I just refer you to
(])
13 the, to the table.
I have the cost of the test program in, on 14 the same line with impacts, and then I have multiplied NRC 15 development, implementation, and operation costs by a factor 16 of ten.
17 Now this is not to say that we think that these 18 costs are going to go up by a factor of ten, but I wanted to 19 see what would happen if we would increase them by a factor of 20 ten, and I did that.
21 And when you factor in both of those, you get a 22 total, and the total impact--this is without the supplemental 23 attributes--this is without considering off-site property, or 24 on-site p1operty damage, or plant improvement.
That was never 25 included in this, any of this.
You see that the best estimate HERITAGE REPORTING CORPORATION - (202)628-4888
37 1
goes from $418 million saved to $384 million saved.
It's O
2 inconsequential in terms of a study like this, different 3
reference, and the range, instead of the high estimate, is 435 4
versus 453 million dollars, and the low is, we have previously 5
estimated $355 million, now we are estimating $297 million, 6
but I didn't stop there.
7 I took and did another exercise.
We had postulated 8
that the industry was going to save $409 million in this 9
study, and the next page whet I did was I wiped that out.
I 10 said suppose the industry doesn't gave a nickel on this, that 11 we are off by this half, S409 million figure, and I took that 12 out, and you see that on the last page that I have taken that
(}
13 out, and what that comes up with is instead of $197 million 14 savings, that would mean that the industry would call, the 15 industry's cost would be $112 million over whatever was saved.
16 All right.
17 Then I took that figure of $112 million and divided 18 it by the lowest estimate of man-rem saved--person-rim, excuse 19 me--saved, and that was 185,000 person-rim.
If you do that, 20 you come up with S106, $666, excuse me, per person-rim, and 21 that is well within guidelines that I have been told of a 22 thousand dollars a person-rim is cost effective.
23 Admittedly a thousand dollars is arbitrary, but that 24 seems to be the party line.
25 MR,. BAER:
We want to emphasize this is the minimum HERITAGE REPORTING CORPORATION - (202)628-4888 i
1
38 1
of what we felt was the low estimate of person-rim saved.
2 MR. ROTHBERG:
But if you were to use the other 3
numbers, you come up with higher numbers.
I think if you use 4
the first column, it comes up to less than a hundred dollars a 5
person-rim, so that's, that in our mind is the worst case 6
scenario from the point of view of proffering our case.
7 MR. WOHLD:
I guess by what you are presenting here, 8
Owen, I would have to admit I am having trouble following the 9
discussion here, but by what you are presenting, you are 10 adding the cost / benefit of a prototype testing program to the 11 generic letter?
12 MR. ROTHBERG:
I am adding the cost, not the
()
13 benefits.
Costs are negative.
Benefits are positive.
14 MR. BAER:
He is done basically two things.
He has 15 said as an additional cost, we will throw in the cost of the 16 prototype test program, and the high estimate of that was $47 17 million.
18 Then this final step was to say although we think 19 industry will save money wi th this program, because of 20 avoiding of maintenance problems, and the money already being 21 spent on valves, we will wipe that out and so we have only 22 costs, including this prototype program, no, quote, benefits 23 in terms of dollar benefits, and take the minimum person-rim
/~%
24 that we think would be saved and divide it by the total cost, O
25 and we come up with well less than a thousand dollars a HERITAGE REPORTING CORPORATION - (202)628-4888 u________________________._____._____________________
39 l
1 man-rem.
O 2
MR. ROTHBERG:
That's right.
Now I would like to--
3 CHAIRMAN MICHELSON:
Very conservative.
4 MR. ROTHBERG:
We think it is conservative.
I would l
5 like to say something else about the--I got a copy of that 6
NUBARG 1etter which refers to some cost estimates that came in 7
from NUMARC sometime ago.
8 I think that the NUMARC cost estimates probably have 9
to be revised because the generic letter, since NUMARC 10 reviewed it and came up with these cost estimates, it is 11 revised, maybe they might saying something to this in their 12 presentation.
"3 13 MR. WOHLD:
I think this addresses the question I (J
14 had at the last meeting, that this letter seemed to be 15 significantly expanded over the original intent, and what you 16 are doing here is adding the cost / benefit of the blowdown 17 testing.
18 MR. ROTHBERG:
Right.
We are adding the cost.
19 MR. BAER:
The costs, and factoring that inte the 20 cost / benefit.
21 MR. ROTHBERG:
That's right, and again, we think 22 that we have made our case.
23 Again, this is just a summary of this.
We addressed 24 in that last slide the--and in those four pages, the cost of 25 impact to train and audit on site.
As you can see, the NRC 1
HERITAGE REPORTING CORPORATION - (202)628-4888
40-1 costs even increased by an order of magnitude, are really not 2
going to be a factor in this in terms of overall industry-wide 3
cost.
It is not really a factor.
1 4
One of the questions that came up in the last ACRS-5 meeting was our definition of developed and decign basis.
If 6
you will note in the latest generic letter, the reference to 7
10 CFR 50.49 has been removed.
The definitions of design 8
basis and, design basis and--I'm sorry.
9 MR. BAER:
And safety related, right?
10 MR. ROTHBERG:
Excuse me.
Safety-related--I sorry.
11 Those definitions are consistent with what is in the 10 CFR, 12 and the standard review plan.
They are standard definitions.
()
13 Ue link them to 10 CFR 50.49 because they were concisely 14 stated there, and we don't have any problem with that, but 15 those definitions are what is in the standard review plan, if 16 there is a problem with those definitions.
17 MR. BAER:
They are also in Part 21 and 100 I think.
18 MR. ROTHBERG:
That's right.
19 CHAIRMAN MICHELSON:
Does the Part 21 and 100 have 20 those, the words external events in their definition of design 21 basis?
22 MR. ROTHBERG:
I don't--well, Part 100, of course, 23 talks about the, in terms of events, and I'm not sure whether
(-}
24 21 talks about external events.
\\/
25 CHAIRMAN MICHELSON:
Your definition in the proposed HERITAGE REPORTING CORPORATION - (202)628-4888
41 1
generic letter is, has not been necessarily a quote of the 2
parts of the regulations?
Plus external events is in 3
regulation?
4 MR. ROTHBERG:
We never had quotation marks around
)
5 it to begin with.
6 CHAIRMAN MICHELSON:
I thought you just said it was 7
right out of the regulation?
8 MR. ROTHBERG:
I took it verbatim out of the 50.49.
9 CHAIRMAN MICHEL3ON:
External events are already in 10 the regulation for design basis?
11 MR. ROTHBERG:
Yes, sir.
12 CHAIRMAN MICHELSON:
Okay.
Thank you.
That's quite
()
13 an interesting extrapolation of what the people that are going 14 to worry about valves are going to have to think about when 15 they decide on what requirements for a particular valve might 16 be.
It does include fire and flooding and so forth?
17 MR. ROTHBERG:
Well, I don't think--and we 18 specifically say that in our letter--that we intend to extend 19 the design basis.
20 MR. BAER:
We do not.
21 MR. ROTHBERG:
We do not.
22 CHAIRMAN MICHELSON:
What do you mean by the 23 definition?
That in other words, that's a theoretical 24 definition of design basis, but not one that you intend?
Is O
r 25 that what you are saying?
HERITAGE REPORTING CORPORATION - (202)628-4888
42 1
MR. BAER:
No, carl.
The problem--we have gone 2
through this on OEM code committees.
If you try, try and 3
relate it to ASME classifications, which is sometimes, is to 4
well--if you try and relate it to what an individual plant has 5
designated as safety-related, you get a wide spectrum because 6
of some of the older plants and you will find that many of the 7
older plants have very few valves or systems that have been 8
designated quote, safety-related, so if you don't then give a 9
definition that puts it in functional terms, like we have 10 tried to do in the generic letter, if you are strictly limited 11 to those valves that an individual plant has called 12 safety-related, you are going to find that in many of the
(}
13 older plants, hardly any valves are covered.
14 CHAIRMAN MICHELSON:
I was thinking of your fesign 15 basis definition, not your safety-related definition, Footnote 16 No.
2.
17 MR. BAER:
Oh, I'm sorry.
I misunderstood the 18 question.
19 CHAIRMAN MICHELSON:
That's where they talk about i
20 external events.
And then I thought you said well, we are not 21 going to change the design basis, so if external events were 22 not included in the design basis, then you would say that they 23 don't need to consider them in their valve operability 24 program.
You seem to say you do considor them in this here.
25 MR. ROTHBERG:
We added a sentence.
HERITAGE REPORTING CORPORATION - (202)628-4888
43 1
1 MR. BAER:
I thought the question dealt with 2
safety-related.
)
3 CHAIRMAN MICHELSON:
No, no.
4 MR. ROTHBERG:
We added a sentence which was--I'm 5
sorry.
We added a sentence which was in the text previously 6
having to do with the design basis for each plant or those 7
documented in pertinent licensee submittals such as the FSAR.
8 Now those words are in the text.
They were, we put them all 9
together, and that we feel clarifies it.
Those words are not 10 in 10 CFR 50.49.
11 CHAIRMAN MICHELSON:
So if external events were not 12 included in any licensing document that might have been 13 submitted, then you don't have to consider them in determining
}
14 valve operability?
15 MR. ROTHBERG:
That is correct.
16 MR. BAER:
Provided that they are needed to, to 17 perform the functions which is in Footnote 1.
18 CHAIRMAN MICHELSON:
Yes, and provided they are 19 needed for Footnote 1.
Okay.
I just want to make sure I 20 understood because it didn't come through clearly from just 21 reading the generic letter.
22 MR. WYLIE:
Let me ask a question to clarify 23 Footnote 2.
It basically is addressini design basis events, 24 but the last sentence, it seems to be a definition of design 25 basis for any type, and just help me a little bit.
A HFRITAGE REPORTING CORPORATION - (202)628-4888
44 1
I would understand then that where you use over here 2
in the text regarding design basis degraded voltage, this 3
definition is a, that it is the design basis for the plant 4
where it has been documented in pertinent licensee submittals 5
such as FSAR, and if you haven't done it, then you have got to 6
do it?
7 MR. ROTHBERG:
That's right.
8 CHAIRMAN MICHELSON:
I didn't get that.
If you 9
haven't done it, I don't think you have to.
10 MR. WYLIE:
If you don't have it, how do you do it?
11 CHAIRMAN MICHELSON:
I think what it is, if it is 12 not committed to now, this document will not change that.
()
13 MR. WYLIE:
That's not what I get out of it.
14 CHAIRMAN MICHELSON:
You read it--
15 MR. WYLIE:
You say my interpretation is correct?
16 MR. BAER:
No.
17 CHAIRMAN MICHELSON:
You can't be both, both can't 18 be right.
19 MR. BAER:
When you say if you haven't done it, you 20 have to do it, we say later in the letter that they don't have 21 to repeat 85-03 valves if upon review they believe that they 22 have covered their design basis, and we specifically spell out 23 the degraded voltage, but the intent is that degraded voltage 24 that is committed to in the license of that plant.
25 If a plant--I don't know that there is one--managed HERITAGE REPORTING CORPORATION - (202)628-4888
(
45 1
to go through the whole system and never committed to I
2 operation at a less than a hundred percent voltage, then all 3
of this'1etter will not require them to do anything more.
4 MR. WYLIE:
I'm sure they did, but I suspect that 5
the strict interpretation of the definition of degraded 6
voltage would lead you to something else.
7 MR. ROTHBERG:
Degraded voltage within the design 8
basis.
9 MR. WYLIE:
Well, that's my question.
This, as I 10 interpret what you are saying here, it isn't design basis 11 established for the plant even if FSAR submittals or other 12 submittals.
(}
13 MR. BAER:
Right.
Some projects, it is 80 percent.
14 MR. WYLIE:
If you haven't done--
15 MR. BAER:
They haven't committed to, they have i
16 never made a commitment.
17 MR. WYLIE:
Normal envelope of voltages is basically 18 of whatever you get under normal operation, with everything 19 working, or when you are, go into emergency power supplies and 20 you sequence loads on and so forth.
Now correct definition 21 would be that you analyze that and that's what you define.
I 22 suspect that hasn't been done in every case.
23 MR. ROTHBERG:
I think, well, I don't know.
In our i
24 Attachment A to the generic letter, there is an item 28 that 25 says degraded voltage will emphasize within the design basis.
l HERITAGE REPORTING CORPORATION - (202)628-4888 1
46 1
MR. BAER:
Maybe--I guess, Mr. Wylie, I guess maybe gsb 2
I have oversimplified and probably do have an oversimplified 3
understanding of what--most licensees I thought had a 4
commitment somewhere that their equipment would operate at 5
some fraction, usually 80 or 90 percent of nominal--
6 MR. WYLIE:
Some at 70.
7 MR. BAER:
Of nominal voltage, and that the 8
equipment was designed to operate at that voltage.
9 MR. WYLIE:
Some valves would be operating on 10 emergency power supplies.
When you are sequencing loads for 11 example, now is.that a design basis?
12 MR. MINNERS:
Certainly.
()
13 MR. WYLIE:
So that's what is intended.
That's what 14 you are analyzed to?
15 MR. ROTHBERG:
Whatever you committed to.
16 CHAIRMAN MICHELSON:
Let me make sure that I 17 understand your answer then.
38 MR. BAER:
I am not sure I understand the question.
19 MR. WYLIE:
All I said was that's the definition of l
20 design basis, and you say that further, that the valves have 21 to be analyzed against design basis degraded voltage and those I
22 recent terms you used.
l r
23 Then all I said was I'm interpreting that to mean 24 that it meets this last sentence on Footnote 2.
These are the 25 things that are documented in the FSAR or otherwise, and if l
HERITAGE REPORTING CORPORATION - (202)628-4888 L
47 1
you haven't done it, you have got to do it.
2 MR. ROTHBERG:
That's right.
3 MR. BAER:
If there was never a commitment, I guess 4
to use your example, if someone had made the commitment that 5
their design basis degraded voltage was 80 percent, and they 6
had never done a real detailed analysis of the voltage dip on 7
the sequenced equipment on to the diesels and it turns out 8
that when they put on certain equipment it goes down below,-
9 the voltage goes down below 08 percent, I don't think this 10 letter requires them to go reanalyze it.
If their commitment 11 was for 80 percent, somehow the staff in their review never 12 picked it up and there was a situation where it could go down
()
13 78 percent, I don't think this letter requires them to go back 14 and reanalyze, and that would be my interpretation.
I don't 15 know if there is a consensus on that, but I'm admittedly 16 looking at it as simple-minded.
There is a number somewhere 17 in the documentation.
18 MR. WYLIE:
You ere not going to re-review the 19 review of the design basis degraded voltage?
20 MR. ROTHBERG:
Absolutely not.
21 MR. BAER:
That's right.
22 CHAIRMAN MICHELSON:
The design basis includes such 23 things as particular tests that the vender said are needed, 24 the licensee said he would do?
Is that a part of the design 25 basis?
Or is that something else?
)
HERITAGE REPORTING CORPORATION - (202)628-4888
48 1
MR. MINNERS:
You mean pre-op tests?
2 CHAIRMAN MICHELSON:
Whatever kind of tests.
3 MR. MINNERS:
It doesn't compute for me, Carl.
I 4
mean I don't see how tests has anything to--deeign basis to me 5
is parameters.
6 CHAIRMAN MICHELSON:
Design basis is only the basic 7
parameters on a given component.
How many tests might be 8
performed or the maintenance that might be performed or-9 whatever?
Just the basic design?
10 MR. MINNERS:
If subjected to some strange test that 11 is outside the design basis, fails--
12 CHAIRMAN MICHELSON:
Some of these testa aren't very
(;
13 strange.
Some tests that I have in mind aren't strictly very 14 strange.
15 MR. MINNERS:
If they do some test which is outside 16 the design basis and it is failed, we don't intend to fix that 17 up.
ie don't really expect to-expand the design basis to 18 includ those test results.
19 CHAIRMAN MICHELSON:
Except in the case of Section 20 11 where you do intend to expand tests, is that right?
21 Section 11 is what utilities committed to basically on most of 22 these valves.
Now you are saying beyond--
23 MR. MINNERS:
All of the test parameters for Section 24 11 tests are well within the design basis as I understand it.
25 CHAIRMAN MICHELSON:
Oh, yes, certainly.
HERITAGE REPORTING CORPORATION - (202)628-4888
49 1
MR. MINNERS:
I am triking, I am trying to
{
l 2
understand your question, j
I 3
Are you talking about tests in which the test i
4 parameters are outside the design basis?
5 CHAIRMAN MICHELSON:
You said you would not expand 6
the design basis.
7 MR. MINNERS:
Right.
8 CHAIRMAN MICHELSON:
And I am just asking does that 9
include testing that was prescribed in a design basis?
I 10 think the answer is no.
Because if it were yes, you are 11 expanding, for instance, Section 11, testing of valve beyond 12 Section 11 and Section 11 is what is prescribed in the design
(~%
13 basis.
%)
14 MR. BAER:
I wouldn't--Section 11 design basis, 15 Section 11 is a series of tests that are required.
16 C" AIRMAN MICHELSON:
It was, the reason for the 17 question i what you included in the design basis, and the 18 answer is you do not include testing, for instance, as a 19 design basis.
20 MR. BAER:
Yes.
21 CHAIRMAN MICHELSON:
That takes care of it.
22 MR. WOHLD:
Owen, I just want to add something on 23 the design basis.
24 I have had a lot of opportunities to try and
,q
%)
25 interpret what design basis meant starting way back in the HERITAGE REPORTING CORPORATION - (202)628-4888
50 1
early '70s, and I used to write some of the FSAR material O-2 relating to commiting to design basis testing and so forth, 3
and quite often you use that when you don't know the details 4
of what you want to do.
It is words you can use and no ono 5
can argue with, but in interpretation, it is extremely 6
difficult.
7 When you look at the, for instance, your words here, 8
design basis documented in the FSAR, in'the implementation of 9
some very basic design bases, there are details in the design 10 of the plant that are not documented anywhere in the FSAR.
11 MR. BAER:
Such as the FSAR?
We are looking for I 12 think things that were committed to the NRC and not--I guess I
()
13 have a problem.
I think you can do testing to try and 14 demonstrate your method of design basis, but I guess I am, 15 like Warren, it doesn't compute for me to call a test a design 16 basis.
17 MR. MINNERS:
I think I understand what you are 18 saying, Pete.
I think I have run across that, too.
There are 19 some leads in the design which you can't find written down any 20 place in black and white.
I agree wi th that.
I don't know l
l 21 what I would do about that.
22 MR. WOHLD:
I think the problem that I could see 23 happening on this is that the generic letter goes to a l
l 24 licensing group within the utility to determine what tests it
{)
25 is necessary to take place, and that would give an entirely l
HERITAGE REPORTING CORPORATION - (202)628-4888
_--__-a
51 l
1 different slant than if an engineering group looked at the 2
testing requirements.
3 I guess my basic premise that I have been pushing l
4 several times is you have to take a common-sense approach to 5
the facts and figure out what the valve was installed for and 6
you can see if it meets that design requirement.
I think that 7
would be the first place to start.
8 I'm a little bit concerned when you said that, you 9
know, r".bmittal in the FSAR, for instance, if you say that you 10 are going to keep the fuel elements below 2300 degrees during 11 an accident, so forth, in the detailed implementing design 12 that may require a valve to work against a certain DP that it
{}
13 is not recorded any place, so it gets you down to having to do 14 a,
I guess a common-sense engineering evaluation.
15 MR. BAER:
I see that we had somewhat of the same 16 discussion I guess a month ago, and I see no problem with it.
17 In fact, as an engineer, that's exactly how I would approach 18 each of these valves.
I would look at things like pump-19 shutoff head and say yes, this is realistically the, what the 20 valve may have to open or close against.
21 As a maximum, there maybe other reasons why they 22 cannot even see that kind of delta P, but certainly it is hard 23 to envision on an opening stroke that it has to open against a 24 delta P greater than the pump shutoff head, or if it were an 25 isolation valve and a high pressure system, that it would have HERITAGE REPORTING CORPORATION - (202)628-4888
52 1
to close against any pressure higher than a maximum pressure
,,,3
\\-)
2 of that system, and that's, as an engineer, how my start, 3
would be my starting point.
l 4
MR. WOHLD:
Right, and I think we agree, but this, 5
this reference to the documented licensing requirements does 6
not start you in that direction.
It starts, starts the 7
utility with the FSAR review.
8 MR. BAER:
The words come largely I believe from 9
85-03 thoughts, and I do remember when Bulletin 85-03 went to-10 the CRGR, the concern was exactly the opposite of the one you 11 are expressing.
12 The concern by the CRGR was that we were going to be
()
13 creating new design basis events with this program, especially 14 with quote, mispositional valves or valves that were like 15 Davis-Besse.that got into the wrong position, and I think CRGR 16 was trying to make it clear that they merely had to, if a 17 valve got mispositioned, that the intent was that you had an 18 actuator and switch settings that would get it back into the 19 correct position, but you didn't have to start chasing down 20 completely new scenarios as a result of that.
l 21 Now I think my recollection of that's how those 22 words started, and I will turn to Dick, you were involved in 23 the bulletin.
24 HR. KIESSEL:
Yes.
That's essentially the same., but 25 it was intended that the licensee should go back to his basic i
1 HERITAGC REPORTING CORPORATION - (202)628-4888
53 1
design concept, and build forward from that.
('s V
2 They were to--let's take the case of a shutoff 3
valve, or an isolation valve, and let's put it in the steam 4
line to the HPCI turbine.
5 What is the maximum pressure that that valve will 6
have to close against?
We have to go back and look at change, 7
the various scenarios for the events where that might come 8
about, you know, and you can't just take normal reactor 9
operating conditions.
10 It also was intended to tie and cash any possibly 11 overlooked valves when a new design basis event had been added 12 or a new method of analyzing the event had suddenly said hey, 13 this valve has got to operate under these conditions.
14 Do you follow me?
Now one way of bounding is, that 15 can happen in many conditions, is to simply look at pump 16 characteristics.
You also have to look at system 17 characteristics on many of the valves.
18 CHAIRMAN MICHELSON:
I think what you have been 19 describing to us goes beyond what is in Footnote 2 because 20 there you are only talking about what has been submitted to 21 the NRC, and I think you said it is going, it has to go far J
22 beyond that.
23 MR. KIESSEL:
Okay.
24 CHAIRMAN MICHELSON:
You don't receive but a O.
25 fraction of all the design basis information for a given j
HERITAGE REPORTING CORPORATION - (202)628-4888
54 1
system, and that design basis information is where you find 2
out what the performance is expected to do, whether or not'it i
3 was ever submitted to the NRC.
4 MR. MINNERS:
We are not interpreting that to mean 5
that unless you have submitted every flow rate 6
pressure / temperature, et cetera.
That's not what is meant by 7
design basis.
8 CHAIRMAN MICHELSON:
I think that's right.
9 MR. MINNERS:
Design basis, somewhere in the 10 documentation it says this system has to work for this 11 accident.
12 CHAIRMAN MICHELSON:
If I want to read this legally,
()
13 the words say those documented as pertinent licensee 14 submittals such as, Right?
Only licensee submittals?
15 MR. MINNERS:
Agreed.
16 CHAIRMAN MICHELSON:
I am saying now it should have 17 been worded pertinent design basis documents and licensee 18 submittals because there is a lot of this that you are going 19 to, is never submitted to NRC that has established what the 20 requirements--
21 MR. MINNERS:
I don't disagree with that.
22 CHAIRMAN MICHELSON:
One at a time.
23 MR. KIESSEL:
Definitely you do not want that.
We 24 don't.
25 CHAIRMAN MICHELSON:
For what reason?
HERITAGE REPORTING CORPORATION - (202)628-4888
55 1
MR. KIESSEL:
One very good reason--when the 2
licensee went and sent his design spec to the valve 3
manufacturer, he probably ba11 parked it, and he ba11 parked it 4
very high, and we are not talking about using that kind of 5
condition to demonstrate valve operability.
We are talking 6
about using the actual conditions that the valve will see when 7
it is subjected to design basis events.
8 CHAIRMAN MICHELSON:
Now I agree that was more 9
reasonable, but it doesn't have to be submitted to NRC to be 10 the design basis.
In this, this footnote says only what is 11 submitted to NRC.
12 MR. MINNERS:
I don't understand.
Give me an
{J~)
13 example of something that is part of the design basis that is 14 not submitted to the NRC.
15 CHAIRMAN MICHELSON:
It is something that is not 16 submitted to the NRC?
17 MR. MINNERS:
Yes.
You are saying design basis, 18 which is part of design basis.
19 CHAIRMAN MICHELSON:
Such things as, for instance, 20 on a HPCI pump, when the turbine overspeeds and you have to 21 catch up with the overspeed electronically, and mechanically, 22 the governor, that raises the delta P on the valve 23 significantly.
That kind of sublety is probably never told to 24 the NRC.
You just say the thing will open against the, some 7g V
25 kind of other words.
HERITAGE REPORTING CORPORATION - (202)628-4888
{
56 1
MR. MINNERS:
Yes, and I would presume that those I
2 other words--
l 3
CHAIRMAN MICHELSON:
Design pressure expands, that's 4
allowed to go to I think 10 percent over design pressure 5
because it is a very one-shot proposition or very infrequent, I
6 so the code allows us to push the delta Ps up higher on the 7
components, but that kind of sublety simply isn't in what is 8
submitted to the NRC and shouldn't be, but it is what should 9
be considered when you do your design review and you determine 10 your design basis.
11 MR. MINNERS:
I would accept your words if you would 12 change it to say that kind of detail is not submitted to the
()
13 NRC, but I presume that when they submit it to NRC that they 14 have got a HPCI system that is going to work in an accident 15 situation, that that includes those subleties, that they don't 16 have to tell me every little sublety, but the general 17 statements mean that they have done the details to assure that 38 it will work.
19 CHAIRMAN MICHELSON:
Then I think you need to revise 20 a little bit your definition of design basis, and I think I j
21 would have no problem.
22 MR. MINNERS:
I think the CRGR would disagree.
I 23 think the CRGR would say putting in the extra words would 24 expand it.
25 CHAIRMAN MICHELSON:
You are only requesting to FERITAGE REPORTING CORPORATION - (202)628-4888 l
57 1
worry about what you go back to where the licensees, licensing f3 d
L 2
people will do this work, and not the engineers who did the 3
job, and you just stick strictly with whatever you happen to 4
have told NRC, whatever that might have been, and I think 5
that's wrong.
I think you have to go back and look at the 6
fundamental basis for your system, examine that, and determine 7
how to test the valves accordingly.
8 MR. MINNERS:
I think that's what we are asking and 9
I think that's what the words said.
10 CHAIRMAN MICHELSON:
Well, then we have reached an 11 impasse on this point.
12 MR.. ROTHBERG:
May I offer something in addition?
()
13 These words were taken out of Bulletin 85-03, review and 14 document the design basis for the operation of each valve.
15 This documentation should include the maximum differential 16 pressure expanded during the normal opening and closing of 17 valves, both normal and abnormal events to the extent that 18 these valve operation and events are included in the existing 19 approved design basis, i.e.,
the design basis documented in 20 pertinent licensee submittals such as the FSAR.
21 CHAIRMAN MICHELSON:
That is more encompassing 22 consideration than what you have in Footnote 2.
23 MR. ROTHBERG:
I think the words the design basis 24 documented in licensee submittals such as FSAR is correct.
25 CHAIRMAN MICHELSON:
Two parts--the real basis of HERITAGE REPORTING CORPORATION - (202)628-4888
58 1
which the utility, NRC, and both of them have to be included.
O 2
MR. ROTHBERG:
This doesn't say anything about 3
anything but submittals.
4 MR. BAER:
What are you reading from?
5 MR. ROTHBERG:
85-03, pt-a 4.
6 CHAIRMAN MICHELSON:
Read it once more.
7 MR. ROTHBERG:
I will start that these valves, 8
operations and events are included in the existing approved 9
design basis, i.e.,
the design basis documented in pertinent 10 licensee submittals such as FSAR analyses, and I will read the 11 rest of the parentheses--and fully approved operating and 12 emergency procedures, et cetera.
(}
13 Now that phrase, and fully approved emergency, 14 operating emergency procedures, et cetera, was taken out by i
15 CRGR at the last meeting when they went through the letter.
16 CHAIRMAN MICHELSON:
I know.
17 MR. ROTHBERG:
The phrase, that stands--and is CRGR 18 approved, is that the design basis documented in pertinent 19 licensing submitals such as FSAR analyses--it does not include 20 documentation.
l 21 CHAIRMAN MICHELSON:
Just for clarification, the 22 only thing you are expecting them to do is to abide by 23 whatever they committed to in their documented submittals to l
24 the NRC?
That's all you are really asking them to do for 1
l l
25 sure.
HERITAGE REPORTING CORPORATION - (202)628-4888
_-___________________-_____-_--_-_______-__________-__a
59 1
Now common sense says you do a lot more, but as far O
2 as what you are asking, it is only what was committed to to 3
the NRC.
4 MR. ROTHBERG:
This is a--
5 MR. BAER:
Rather than concentrate on the 6
footnote--that's a footnote that is something in background.
7 CHAIRMAN MICHELSON:
It is the definition.
That's 8
not just background.
That's the definition.
Two footnotes 9
that are key--what is search related and what is the design 10 basis?
That's not--it is a footnote, but it is the definition 11 of design basis, and then you talk thereafter about design 12 basis.
(,}
13 MR. BAER:
If you look at page 4, item A, maybe 14 that's a more explicit statement of what we are expecting 15 people to do on these valves.
16 CHAIRMAN MICHELSON:
Tae end of the sentence narrows 17 you back down to whatever was submitted to the NRC I 18 think--existing approved design basis.
Could very well be 19 interpreted to mean as submitted to the NRC.
20 MR. BAER:
Yes.
21 CHAIRMAN MICHELSON:
So it is the same thing as 22 Footnote 2.
I am just trying to make sure I understood it 23 because I think it is a serious narrowing of the situation in 24 some cases, particularly since they were not always done so 25 well in the past and a lot of these things were not even HERITAGE REPORTING CORPORATION - (202)628-4888
60 1
thought about.
Now we realire that, we are saying you don't f(
2 have to pick them up.
3 MR. MINNERS:
I understand what you are saying.
You 4
are saying serious, and if that's true, I guess maybe we do 5
something about it, but I haven't heard any examples of a f
6 serious difference.
7 I think you and I interpret what is submitted to the 8
NRC differently.
9 CHAIRMAN MICHELSON:
An example of a serious one is 10 the HPCI steam line.
I know for a fact that at least in one 11 project's case, those valves were not able to isolate breaks 12 down stream of the valves.
It was not in the design basis and
/'T 13 was never.
(/
I 14 MR. MINNERS:
Wait a minute.
How can you say that?
15 How can you say that?
This is in the design basis.
How can 16 you say it is not?
17 CHAIRMAN MICHELSON:
I said it is not--you go to 18 procurement specification, you will find no requirement
)
19 whatsoever for that on the operating--
20 MR. BAER:
That's what we want to avoid.
If someone 21 goes back--
I 22 MR. MINNERS:
That is not a design basis.
)
23 CHAIRMAN MICHELSON:
Let's back off for a moment.
24 MR. MINNERS:
Design basis is what was submitted to 25 the NRC which says that that valve will close during a double HERITAGE REPORTING CORPORATION - (202)628-4888
l 61 1
ended break after that line.
That's the design basis.
7~
(/
2 Now to say what they sent to the valve manufacturer, 3
design basis, no.
That's not submitted to the NRC.
What was 4
submitted to us was a statement which I am sure you can find 5
in the FSAR which is this valve will close during a double 6
ended break.
That's what--and that's the design basis.
l 7
I think that it is your use of words that bothers 8
me, Carl.
Design, you are using design basis with i
9 the--engineers use design base when--a valve manufacturer tells 10 him the design basis, we are not using that.
We are using it 11 in, pardon the expression, the legalistic, regulatory sense of 12 the design basis of the stuff that is documented in the FSAR
()
13 And other commitments.
14 CHAIRMAN MICHELSON:
Let's say for the moment that 15 the HPCI steam line had not been discussed in the FSAR, steam 16 line break.
17 MR. MINNERS:
Tnen we have got a problem.
18 CHAIRMAN MICHELSON:
Those cases, how are we going 19 to do that?
Some kind of ad hoc--
20 MR. MINNERS:
I guess we are not going to do that.
21 We are not going to do that.
5 22 CHAIRMAN MICHELSON:
This, that particular case it l
23 was never discussed in the FSAR, but you know, there is 24 thousands of pages that you have to think about 25 MR. MINNERS:
I would like to fix the whole world, l
HERITAGE REPORTING CORPORATION - (202)628-4888
62 1
Carl, but if the design basis of some plants is incorrect, 2
okay, this effort is not going to fix that.
But--
3 CHAIRMAN MICHELSON:
If it was found it hadn't been 4
discussed in the submittals, then how, what is the mechanism 5
for changing that?
6 MR. MINNERS:
If it is widespread, then we take it 7
on as a generic issue or something like that.
8 CHAIRMAN MICHELSON:
Cleanup line is the other one.
9 MR. MINNERS:
Or maybe on individual plants; there 10 was a case at--Shoreham was a case in which it was picked up 11 through the PAR or PRA, and that was taken care of.
12 CHAIRMAN MICHELSON:
It wasn't actually picked up
()
13 that way.
It was picked up by another mechanism.
14 MR. MINNERS:
But it was picked up.
Okay.
15 MR. BAER:
But that truly, fact of life, is that 16 truly is a backfit.
In some older plants you are talking 17 about valves, but there is plenty of other examples on older 18 plants that have design bases, especially in terms of external l
19 phenomena, that we wouldn't accept today, but the generic 20 letter was not intended to fix that.
21 MR. ROTHBERG:
Do they do the regulation?
22 MR. MINNERS:
Yes.
l 23 CHAIRMAN MICHELSON:
If the utility finds that, for 24 instance, he had not discussed the rupture of the HPCI steam
[}
25 line and therefore didn't have a particular requirement on HERITAGE REPORTING CORPORATION - (202)628-4888
63 1
that valve.in that regard, does he now record this finding
~
l 2
under the deficiency reporting process?
Is it considered a i
3 deficiency?
4 MR. ROTHBERG:
That's not--
5 MR. MINNERS:
I don't really know the answer, but I l
l 6
would doubt it, Carl.
7 CHAIRMAN MICHELSON:
You know, From a practical j
8 sense, isn't that a deficiency?
If you are unable to isolate 9
ruptures outside of containment with the so-called containment l
10 isolation valves?
Is that not a deficiency?
Report it as a 11 deficiency and then you pick it up on the plant-by-plant 12 fixes.
()
13 MR. MINNERS:
Are you referring to particular 14 regulation?
15 CHAIRMAN MICHELSON:
Yep.
16 MR. MINNERS:
I'm not well familiar with that to 17 respond to that.
18 CHAIRMAN MICHELSON:
Because it clearly, clearly 19 reactor water cleanup is the same thing.
If I hadn't, really 20 everybody thought about that, hadn't put it in as a 21 requirement, it never would have committed to do it.
Now I 22 realize that yes, those valves have to close and maybe in my 23 case these won't close, do I report that as a deficiency?
24 MR. MINNERS:
I thought that some licenses have done 25 that.
They have picked up things that--
HERITAGE REPORTING CORPORATION - (202)628-4888
64 1
CHAIRMAN MICHELSON:
I have seen a few LERs.
%)
2 MR. MINNERS:
Whether that is a requirement or not, 3
I would have to ask a lawyer.
I don't know.
4 CHAIRMAN MICHELSON:
Under this process now when 5
they find such things in their review find now I didn't really 6
commit to it, it is not under this generic letter, but clearly 7
it is a design deficiency--
8 MR. MINNERS:
If they do, it would be nice, but that 9
is not the intent of the letter for people to do that.
We 10 didn't think that would be an extra benefit we might get.
11 CHAIRMAN MICHELSON:
We won't get; what happens to 12 the residual?
13 MR. MINNERS:
That's a problem for another day.
We 14 fix them one at a time.
15 CHAIRMAN MICHELSON:
I think it is very likely that i
16 in the process of doing this, utilities are going to find 17 things that slipped in the cracks, and although not 18 specifically covered by the generic letter, clearly it is a 19 deficiency in these that needs to be fixed.
20 MR. MINNERS:
Most licensees, pre-dominance of 21 licensees are responsible I would think to fix those 22 situations.
If peop1.e want to cheat, it is very hard to 23 prevent that.
24 CHAIRMAN MICHELSON:
I am not thinking of cheating.
25 MR. MINNERS:
It it looks at an obvious deficiency HERITAGE REPORTING CORPORATION - (202)628-4888
65 1
and says to hell with it, I would call that cheating.
f_
2 CHAIRMAN MICHELSON:
I am looking at what are you 3
asking for?
4 MR. MINNERS:
We are asking them to follow their 5
commitments.
If they haven't committed to it, we are not 6
asking them to do anything in this particular generic letter.
7 Now if that's a pervasive problem within the 8
industry, well, fine.
Maybe we ought to look into it and 9
after it.
I don't think it is.
It hasn't been demonstrated i
10 that it is.
I don't know of any program now to do anything 11 about it.
12 I understand people's concern is that we are
()
13 supposed to be overlooking the total safety of the plants, but 14 as a practical matter, we can only take small bites and work 15 at it bit by bit, and that's what we are doing here, and if we 16 let things expand and fix all the problems--
17 CHAIRMAN MICHELSON:
It was your intention that such 18 deficiencies, if they should appear, are to be reported.
Then 19 this is a document where you tell them what mechanism to 20 use--if not the generic letter, then whatever the mechanism 21 is.
1 22 MR. MINNERS:
It was not our intent.
23 CHAIRMAN MICHELSON:
Not your intent they even tell 24 you about them apparently, at least you don't want them to 25 remain silent on what they do about it.
HERITAGE REPORTING CORPORATION - (202)628-4888
66 1
MR. ROTHBERG:
I think we are compatible with what
,g U
2 we did under Bulletin 85-03, that the design basis that they 3
did, reviews that were done from Bulletin 85-03 by extension 4
for all safety-related motor-operated valves would be the same 5
type of thing we are looking for.
l 6
MR. MINNERS:
I take that back, what I said before, j
7 There is a program to try to pick this up.
It is called IPE, I
8 when they are supposed to go out, look hard at their plants.
9 If they find safety significant deficiencies, vulnerabilities 10 or whatever the word, they are supposed to report those.
11 CHAIRMAN MICHELSON:
I thought they were supposed to 12 record them under any program for that matter, not just IPE.
()
13 Any time they find them, they will report them.
14 MR. MINNERS:
That's the regulation you are talking 15 about.
I am not familiar enough to that, though, but under 16 IPE, that certainly is the intent, and that's a requirement.
17 That's the 5054 letter, and they are required to report that 18 stuff.
19 CHAIRMAN MICHELSON:
I would have thought they 20 reported it under 5073 or--
21 MR. MINNERS:
You have to ask somebody.
22 CHAIRMAN MICHELSON:
Remind the licensees if they 23 find such--we are not trying to upgrade their design by this
(~g 24 bulletin.
They are obligated still to report such findings.
j
\\/
l 25 MR. MINNERS:
You have to ask somebody from AEOD HERITAGE REPORTING CORPORATION - (202)628-4888 L__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
l 67 1
about the reporting requirement.
7 2
MR. KIESSEL:
Carl, I think CRGR would say take that 3
statement out if we were to try and put it in.
4 The basis for my statement that they would tell us l
5 to take it out was that we did have similar statements on if 6
you found a valve with, that is inoperable, you will make the 7
appropriate reports in accordance with bla, bla, bla, bla, and 8
the CRGR conclusion was hey, they are supposed to do that 9
already.
We don't have to tell them what they are doing, or 10 what they are supposed to do, so therefore, take out that 11 statement.
12 CHAIRMAN MICHELSON:
That is one possible argument.
()
13 MR. BAER:
But it is clear to me that if they find 14 an inoperable valve, they have to report that under 5073.
It i
15 is not clear to me what Carl is aiming at because you said 16 hey, it is obvious the reactor water cleanup line should have 17 been part of the design basis, and I would not disagree with 18 that, but I'm not so sure that there are aren't a bunch of 19 other lines that go through the containment on systems that 20 normally aren't postulated to break that couldn't break, and I i
21 honestly wouldn't expect the licensee to go back through every 22 system that went through the containment and say here is a 23 bulletin--I'm sorry--a letter on motor-operated valves.
If I 24 have a line that goes through the containment, and has 25 motor-operated valve, I will look and see if I postulated HERITAGE REPORTING CORPORATION - (202)628-4888 J
1 68 I
l 1
breaks in those lines as opposed to just closing, isolate 2
under an isolation signal.
I honestly wouldn't expect the l
3 licensee to go through each and every system in that manner, l
4 but if you have a valve that didn't work, yes, I think that's 5
clearly reportable under 50--if it isn't, is not under--right; 6
if it is only a single valve, you are right.
1 7
CHAIRMAN MICHELSON:
One other clarification on the l
8 Footnote 2, when you say things, submittal as submitted, 9
licensee's submittal such as the FSAR, do you mean any 10 licensee submittal that was sent to the NRC, or just certain 11 kinds?
Any answer to your question, all that sort of thir.g is 12 covered?
{}
13 MR. BAER:
I think so.
14 CHAIRMAN MICHELSON:
Any document, in other words, 15 sent no the NRC was a part of licensee's submittal?
16 MR. MINNERS:
Yes.
17 MR. ROTHBERG:
Pertinent licensee submittals such as 18 the FSAR, I don't know that a letter, I guess a letter would 19 come under that.
I was thinking in terms of like the FSAR, 20 the tech specs, the environmental reports.
21 CHAIRMAN MICHELSON:
I am thinking the next lower 22 tier of documents, a very important set being the internal 23 flooding in which they went through and described the various 24 process, identified certain valves that would close and that 25 sort of thing.
Such valves now come under this program to HERITAGE REPORTING CORPORATION - (202)628-4888
69 l
1 whatever extent they said they would be operable under that 2
program?
3 MR. MINNERS:
Absolutely, yes.
Sent out other 4
generic letters and they have responded to those generic 5
letters.
They made--
6 CHAIRMAN MICHELSON:
Appendix R, there are certain 7
places where they talk about connecting this or that valve and I
8 doing this or that thing with it, and you know, that becomes a 9
part of the design basis because external events are included 10 by definition, and fires and external events.
11 MR. MINNERS:
I don't think fire protection 12 systems--I think we discussed this before; safety-related
()
13 valves.
14 CHAIRMAN MICHELSON:
External events, what you do 15 about it includes fire protection systems.
Again, you know, 16 this thing gets kind of makes you start talking about any 17 level of detail of the practical level where you have to start 18 working on it.
You don't start working with just an FSAR.
19 MR. ROTHBERG:
External events and natural phenomena 20 for which the plants shall be designed to ensure functions 1 21 through 3 above.
It is not just external events.
It is 22 external events that are needed to one maintain the integrity.
23 CHAIRMAN MICHELSON:
I understand that.
Of course 24 that's what Appendix R did demonstrate you can achieve at safe
)
25 shut down.
Any valves involved in that process that have to HERITAGE REPORTING CORPORATION - (202)628-4888 a
70 1
function a certain way to satisfy the Appendix R analysis now U
2 come under this valve operability program as I would read it.
3 MR. ROTHBERG:
These definitions again appear in one 4
form or another throughout 10 CFR.
I think that the licensees 5
have to understand what their list of safety-related 6
motor-operated valves is, or I don't know that licensecs have 7
had to take a particular problem identifying which valves are 3
or not safety-related in the past.
Certainly Davis-Besse has 9
had no such problem, have a complete list they say of 10 safety-related.
11 CHAIRMAN MICHELSON:
We haven't asked them to take 12 that complete list and do valve operability on it.
Now they
()
13 are going to go back and look carefully at what is meant.
14 MR. ROTHBERG:
Certainly these definitions I don't 15 think were at fault because there again 10 CFR--
16 CHAIRMAN MICHELSON:
I don't have any problem with 17 the definitions.
I just want to make sure that I understand 18 that how comprehensive this definition might be.
External 19 events are clearly in the definition that you have to 20 demonstrate safe shutdown.
That's items 1, 2 and 3 under 21 external event conditions.
Whatever component it takes to, 22 that's required to perform that function is now going to be 23 valve reliability and operability surveillance of some sort.
24 MR. ROTHBERG:
I guess what is being taken to task 25 is the way that 10 CFR talks about external events that could HERITAGE REPORTING CORPORATION - (202)628-4888
71 1
be.
fsO 2
CHAIRMAN MICHELSON:
Okay.
I think I understand.
I 3
think we are together.
I'm not sure.
4 MR. WOHLD:
What is the conclusion?
Fire system 5
valves are included?
6 MR. ROTHBERG:
I really don't know.
I can't say 7
that fire--
8 CHAIRMAN MICHELSON:
Fire, external events, whatever 9
you do to demonstrate safe shutdown for a given fire in given 10 location will have to be, will have to have some kind of 11 operability insurance on it.
12 MR. BAER:
Those aren't necessarily the valves in
()
13 the fire protection system.
14 CHAIRMAN MICHELSON:
I don't know which valves they 15 are.
You are not that explicit in your definition, You just 16 said external events and safe shutdown.
17 MR. BAER:
If I, I think in terms of a fire right 18 now, maybe I'm missing the point, but if the question with the 19 fire was it would disable certain safety-related equipment, 20 you still have to demonstrate you are shut down, and--
21 CHAIRMAN MICHELSON:
You also have to demonstrate 22 the fire doesn't spread.
You can put it out.
One hour in 23 some cases because you have only protective cable for one hour 24 in some cases, so if you don't put it out in one hour, the 25 cable that was protected for safe shutdown may no longer be HERITAGE REPORTING CORPORATION - (202)628-4888
72 1
protected for safe shutdown.
Even the fire out is part of the 2
importanct, is an important step.
3 ML. BAER:
Is.your concern then with the valve and 4
the fire protection system itself?
5 CHAIRMAN MICHELSON:
Right.
6 MR. BAER:
As opposed to things that I would call 7
safety-related valves anyhow in the service water system and 8
in the component cooling water system, to achieve in the RHR 9
system, you need a safe shutdown.
10 CHAIRMAN MICHELSON:
I am trying read your 11 definition, make sure I understand the scope.
I would say 12 yes, certain fire protection valves may well come under the
()
13 scope the way the definitions are written.
Now if that wasn't 14 your intention, then you need to revise the definition 15 somehow, say all external events are out of the definition.
16 Then I will argue as to why they are not in there, but they 17 are in there clearly.
18 MR. ROTHBERG:
Right.
We continue?
19 CHAIRMAN MICHELSON:
Yes.
20 MR. ROTHBERG:
There was a concern about the fact 21 that recovery actions weren't considered in, in our value 22 impact analysis.
Bob touched on it earlier.
23 We feel that the, there are offsetting--it is and it i
24 was not included indeed, but there are some offsetting 25 considerations.
i 1
HERITAGE REPORTING CORPORATION - (202)628-4888
73 1
First of all, the assumption made in the question I 2
think was that recovery actions would always be helpful and 3
useful to you, and therefore, that they would decrease your 4
core melt probability, but there are some considerations that 5
say that well, that might not be some.
For instance, the 6
operator action may not necessarily be the correct action, 7
I'm not saying that happens every time.
The guy is going to 8
do the wrong thing, but there is at least a percentage of the 9
time where somebody will do the wrong thing versus the right 10 thing, and I think we have had enough experience to know that 11 is the case.
12 The remote manual action that may be necessary may
()
13 not be possible because it may not be possible to get to a 14 valve to operate it either manually or from some non-control 15 room location, and as Bob touched on, the problem of common 16 mode failure was, was not considered in that you may have more 17 than one valve exhibiting this failure at the same time.
18 Because of that, we think that the value impact 19 analysis is not really impaired by the fact that we didn't 20 consider operator actions.
That's our argument for that.
21 MR. WOHLD:
I think one of the factors that would 22 support your argument, there is some, some utilities have not 23 considered--perhaps all utilities only considered the manual 24 actuator as a safety-related piece of equipment, and some 25 utilities will not maintain those and due to lack of training 1
HERITAGE REPORTING CORPORATION - (202)628-4888
74 1
on the part of the equipment operator, many of these things 7-2 you find are broken and unavailable, so you simply can't 3
handwheel some of these valves.
4 MR. ROTHBERG:
What I would like to do now is to 5
give you an outline of the generic letter as it stands today.
6 You were sent a copy of the memorandum from Beckjord 7
to Jordan, and the letter is essentially identical to that.
I 8
say essentially because there were some minor changes made.
I 9
classified them as mf by Jim Conran and Joe Zimm and Jack 10 Goldberg, the project managers, and Roy Woods asked me to make 11 a change in his discussion of the Generic Issue 87 concern.
I 12 can give you a copy of the Commission's later.
I have marked
()
13 them in red, but I don't think that they--for all intents and 14 purposes, the letter is as you see it, and that is the way it 15 is going to go back to, to NRR.
Again, I will be glad to give 16 you a copy of these markups, but what I want to say is that 17 the generic letter is as you have it in front of you 18 essentially.
19 The way the generic letter is now structured is 20 there is a background section, and in that background section 21 is the discussion of the bases for the letter, and I think we 22 have gone over the first page, and the first two footnotes, in 23 pretty much agonizing detail.
24 The second paragraph on the first page of the 25 generic letter refers again to some of the deterministic HERITAGE REPORTING CORPORATION - (202)628-4888
l l
75 p
1 arguments for the generic letter, and also the discussion at 2
the bottom of the page of the stroke timing test and its 3
advantages and limitation on that.
4 On page 2, we discussed the INEL blowdown tests, and l
l 5
the, again, the limitations of diagnostic testing to perhaps 6
serve to extrapolate up to design basis conditions.
7 We also point out to the fact that you may not be 8
able to accomplish some of the things that you want to l
9 accomplish with in situ testing.
Again, the prototype testing 10 program is alluded to at the bottom of that page.
1 11 Next in the discussion part we refer to the general 12 program, and we added words about how licensees may want to
(}
13 consider that these are long-term programs, that they are tied 14 into the maintenance and repair guidelines that are also tied 15 into the QA program that the licensee may have.
We tried to 16 do that, too.
17 In the recommended actions, scope paragraph, it 18 refers to all safety-related motor-operated valves and the 19 position changable valves.
Now the position changeable valves 20 are valves that might not be otherwise safety-related but are 21 not locked out.
A convenience valve that is for one reason or 22 another not locked out and is a motor-operated valve would be 23 in that category.
24 A valve that is operable from the control room which 25 is covered by a key lock I would not think would be considered HERITAGE REPORTING CORPORATION - (202)628-4888
76 1
to be-position change, position changable valve because I f
O 2
think that the key lock prevents the type of non-thinking 1
(
3 inadvertent operation that we are concerned about.
If there l
4 is the possibility, on the other hand, that that key lock 5
might not do the job, that the electric, through some 6
electrical fault the valve might become mispositioned, then 7
that should be looked at.
8 CHAIRMAN MICHELSON:
In your first paragraph under 9
recommended action, I think that's the first time you allude 10 to the possibility that balance of plant valves might also be 11 included, but apparently as near as I can read what you are s
12 saying, you ought to really think about it hard, but it isn't 13 really a requirement of the generic letter.
14 MR. ROTHBERG:
That's right.
One of the changes 15 that was made, and this was Jim Conran's request, is the way 16 that last sentence in the paragraph reads now, is although 17 this program should address safety-related MOVs and position 18 changable MOVs, as a minimum, NRC envisions that as part of a 19 good maintenance program, other MOVs in the balance of plant l
20 should be considered for inclusion in the program commensurate 21 with the licensee's assessment of their importance to safety.
l 22 I don't think that the thought has changed, but the l
l l
23 words are slightly changed.
It puts the burden on the i
1 24 licenses 25 CHAIRMAN MICHELSON:
I guess one says intuitively HERITAGE REPORTING CORPORATION - (202)628-4888 1
77 1
that there are some balance of plant valves that might be U
2 important to safetv, but those are not included under 3
definitions covered by Footnotes 1 and 2.
4 MR. ROTHBERG:
All we say is essentially that it is 5
up to the licensee.
6 CHAIRMAN MICI3LSON:
Under Footnote 2, what happens 7
in the case where one had included the operation of a certain 8
balance of plant valve to prevent a flood from getting out of 9
hand and blowing out walls and whatever?
Is that valve--now I
10 it is already in the FSAR submittal or in the licensing I
11 submittals, et cetera, that says that valve has to close 12 within, has to be closed within 15 minutes or we would indeed
( }
13 have a problem, but--well, close it within 15 minutes.
14 Is that valve in or out of the program since it is a 15 balance of plant valve?
16 MR. ROTHBERG:
I would say that, that it has to 17 close within 15 minutes, it seems that it has a search 18 function and therefore it is a safety-related versus--
19 CHAIRMAN MICHELSON:
Even though it is a balance of 20 plant, and my first statement isn't quite true then, and that 21 is that there are balance of plant valves wb'.ch will have to 22 be under this program, not just whether the licensee decides 23 or not.
24 That would be a valve that is to be, because in the O
)
25 flooding analysis that valve was given credit to prevent the HERITAGE REPORTING CORPORATION - (202)628-0888 l
78 1
flood from jeopardizing safety-related equipment ultimately,
/'"\\
U 2
but it is a BOP valve.
It is not a--
3 MR. ROTHBERG:
Sounds from your description like it 4
would be a safety-related valve.
5 CHAIRMAN MICHELSON:
You find a number of such 6
valves.
7 MR. ROTHBERG:
It sounds also like it would have 8
been somewhere in the licensee's commitments or his submittals 9
that that valve should have been safety-related all long.
The 10 letter, I guess--
11 CHAIRMAN MICHELSON:
It's a question of balance of 12 plant, s e e,,
We have got two definitions now,
(}
13 safety--three--safety-related valves, and design basis valves,.
14 and now balance of plant '71ves.
15 MR. ROTHBERG:
The nuance is that this letter is by, 16 in some fashion going to expand the number of ralves that are 17 safety-related.
We have no intention--
18 CHAIRMAN MICHELSON:
Want to make sure to pick up 19 the ones that are there that you have been told about, but 20 maybe didn't even get under programs in the past, is that 21 right?
Or valves out there which we were told about during 22 the flooding analysis which aren't getting in-service 23 inspection and so forth, and we are not going to go back and 24 make sure at least they are operable, Some of these are 25 manual valves.
HERITAGE REPORTING CORPORATION - (202)628-4888
79 1
MR. ROTHBERG:
They won't be in this.
2 CHAIRMAN MICHELSON:
That's a new issue.
Some of 3
these are manual.
If they are motor-operated, then those 4
would be under this program, even though they are balance of 5
plant valves.
6 MR. ROTHBERG:
Also does not cover hydraulic.
7 MR. BAER:
For the capability, shut down the reactor 8
and maintain the safe condition.
9 CHAIRMAN MICHELSON:
Maybe that sentence on balance 10 of plant valves ought to have been worded--you have got it in 11 there when you talk about important to safety, but you really 12 are saying that they don't have to be included.
They should
(}
13 be considered.
14 MR. ROTHBERG:
When the licensee assesses their 15 importance to safety, he is going to have to make that 16 determination.
I guess that, and I think this is sort of a 17 bowing to the maintenance rule.
18 CHAIRMAN MICHELSON:
We don't know what it will even 19 look like yet.
We have proposed maintenance rule, but--
20 MR. ROTHBERG:
Then in the recommended action, the 21 course of going now through the items, specific items, A,
B, 22 C,
first part of D, we tried to reproduce again from Bulletin j
23 85-03, review and document the design basis, establish the 24 switch settings, set the switches, and then revise the 25 procedures for setting the switches.
i HERITAGE REPORTING CORPORATION - (202)628-4888
80
)
1 CHAIRMAN MICHELSON:
Could I ask a question on that g-]g L
2 particular set?' The only problem I had was when you read item 3
C, it has a note down there that refers back to A and B, and 4
the question that came to my mind was A, A just says you 5
decide what your design basis is and then B, you establish the 6
correct settings.
It is in C that you take those settings now 7
and determine what your switch settings have to be.
You make 8
the switch setting changes I should say.
You change the 9
setting, but when I read this note, I'm not sure that for 10 instance, it says you don't worry about break outside of 11 containment right now for other reasons.
12 MR. ROTHBERG:
No.
Just worrying about testing for
()
13 that--in other words, it is still to be included.
14 CHAIRMAN MICHELSON:
Let me ask the question and 15 then you can tell me.
16 Clearly I am supposed to consider breaks when I 17 research the design basis.
I am supposed to consider breaks 18 when I decide what the correct settings are.
19 My question is do I also go and make the switch 20 changes for the postulated break?
21 MR, ROTHBERG:
Yes.
22 CHAIRMAN MICHELSON:
That part wasn't clear because 23 why didn't you refer to A, B and C?
24 MR. ROTHBERG:
This is correct.
}
25 CHAIRMAN MICHELSON:
I know it is, but it refers HERITAGE REPORTING CORPORATION - (202)628-4888 m_____________
._ a
81 1
back to A and B, and that didn't--
J 2
MR. ROTHBERG:
I see that.
You can do that.
3 CHAIRMAN MICHELSON:
That you are going to also 4
require that they change the switch settings to the values 5
they need for the, for these postulated breaks, is that right?
i 6
That was your intention?
I 7
MR. ROTHBERG:
Yes.
8 CHAIRMAN MICHELSON:
Okay.
That's I think the 9
correct intention and I just want to make sure that I i
10 understood that was included.
11 MR. ROTHBERG:
Yes, sir.
12 MR. BAER:
I'm sorry.
You are suggesting it says A,
()
13 B and C?
14 CHAIRMAN MICHELSON:
That would be nice except it is 15 already in C and maybe for editorial purposes, it needs to say 16 A,
B and this section or this item or something.
17 MR. ROTHBERG:
Sure.
18 CHAIRMAN MICHELSON:
Instead of referring to C.
19 MR. WOHLD:
A question on the position changeable 20 valves--what are you going to use for a design basis for, of 21 differential pressure on the position changable valves?
22 MR. ROTHBERG:
That will have to be determined by 23 the licensee.
They will have to determine what the pressures 24 and flows are that will be seen--if he choses to leave that 25 MOV as position changes, he has got an option.
He can lock I
HERITAGE REPORTING CORPORATION - (202)628-4888
82 1
out, but if he choses-not to, then he is going to have to 2
decide what pressures and flows that particular MOV is set to.
i 3
That's part of the design basis review.
4 MR. WOHLD:
That's outside of your Footnote 2, 5
though, as far as the documented design basis?
6 MR. MINNERS:
You people keep saying that, and I 7
think you misunderstand what design basis is.
You seem to be 8
implying that design basis are numbers--flow, temperature, 9
pressure, or something.
That is not the design basis that we 10 are talking about.
11 The design basis is the kind of statements that are 12 made in the FSAR, other documeucs, unich sa a this si& tem will j
()
13 operate under certain accident conditions without being 14 specific about the accident conditions.
It says it is going 15 to take a double-ended break, okay.
Usually the flows and 16 pressures are not specified and you don't need that, but the 17 design basis is the double-ended break so that is specified in 18 the document, so if you had a position changable valve, okay, 19 you would have to be able to be open under the double-ended 20 break conditions.
21 MR. ROTHBERG:
Safety-related syster.s, by the way.
22 MR. WOHLD:
I don't think--what was a mispositioned 23 valve and the need to recover from that, I don't think it is 24 part of the original design basis.
25 MR. ROTHBERG:
We are not talking about HERITAGE REPORTING CORPORATION - (202)628-4888
83 1
mispositioned valves.
They are other wise safety-related.
We 2
are talking about valves that are not locked out.
We didn't 3
use the word disposition.
We used the word position 4
changable.
The nicety is, that the nicsty is that a 5
mispositionable valve may be a safety-related valve that you 6
could get--valves in the Davis-Besse.
Those are 7
safety-related valves.
They have a design basis to open and 8
to close.
They have the function to open and close.
Those 9
wouldn't be included as position changable valves, but if you 10 have, you had a convenience valve upstream or downstream, and 11 that convenience valve was not locked out for one reason or 12 another because it was not convenient to do so, you would have
()
13 to make the assumption that somewhere along the line that 14 valve could be put in the wrong position and then would have 15 to recover from that.
The valve itself is not the otherwise 16 safety-related.
It just appears to be a safety-related 17 system.
18 What you want to do is make sure that when that 19 valve does get to the wrong position, that you can get it out 20 of the wrong position and get it into the right one.
- Again, 21 you have this option, so that option is to lock this thing 22 out, get it out of the way.
23 MR. WOHLD:
I understand what you are saying, but I 24 don't think that the original documented design basis 25 submitted to NRC will include the requirement to recover from HERITAGE REPORTING CORPORATION - (202)628-4888
84 1
a mispositioned changeable valve.
I think that as you, I (s/
2 recall assumptions in the past were the valvewas --
3 MR. ROTHBERG:
Talking about a component now.
4 Talking about'a system.
5 MR. MINNERS:
He is correct.
That's right.
You 6
could probably not find the FSAR where they said hey, if this 7
valve changes position, we can open it.
No.
That was not.
8 You are now telling them that's right, this is a new 9
interpretation of what you have to do.
That is correct.
10 MR. WOHLD:
It is an exception to your Footnote 2.
11 MR. MINNERS:
No, it is not an exception.
We are 12 saying that part of--
(}
13 MR. ROTHBERG:
I guess you could say it is an 14 addition.
15 MR. MINNERS:
It is your use of design basis that is 16 confusing it, okay.
The design basis is just the conditions 17 under which it operates.
It doesn't tell you what equipment 18 has to operate.
The design basis is the definition of what 19 parameters it has to operate under and not what equipment it 20 has to operate under.
21 MR. ROTHBERG:
Design basis intereste. are 22 conditions.
That's in the definition.
23 MR. MINNERS:
Design basis is the double-ended 24 break, and we are just saying now that proffering to us that 25 this system will perform its function under the design basis l
1 HERITAGE REPORTING CORPORATION - (202)628-4888 U
85 1
event conditions, and we you also have to consider position 2
changable valves.
Position changable' valves, yes, is.a new 3
twist which was not really looked at for--
4 MR. BAER:
Just to clarify or expand on what Warren 5
said, take the double-ended break as a design basis event.
If 6
you are, you now have an ECCS system supposed to mitigate that 7
event, and not that the valve its--ECCS has to open or close 8
against double-ended pressure, but if that valve, if there is 9
a valve in that ECCS system that can make it inoperable, but 10 by getting mispositioned you have to be able to have enough, 11 the actuator, the motor-operator has to be set up so that it 12 can get back to the correct position so the system could
(}
13 maintain, system can operate.
14 MR. ROTHBERG:
Again, we are being--
15 MR. WOHLD:
What I am saying is that I think that's 16 outside of the original design basis.
17 MR. BAER:
The design basis is the LOCA.
18 MR. MINNERS:
You have to be able to have a 19 functional ECCS system under a LOCA.
That's the design basis.
20 MR. BAER:
You have to mitigate a LOCA.
21 MR. KIESSEL:
It is not beyond the design basis for 22 the plant,.
It is beyond the definition of operable valve, and 23 that has been a long-recognized ratchet that the bulletin had, 04 and that the generic letter had.
25 MR. ROTHBERG:
He published the supplement to his l
HERITAGE REPORTING CORPORATION - (202)620-4888
86 1
bulletin just to clarify the point that we had always wanted
~3-]
2 these position changable valves in there, and by extension of 3
the bulletin, all safety-related motor-operated valves, we are 4
extending the consideration of position changable from the two 5
systems that he covered under his bulletin to all 6
service-related motor-operated valves, and therefore we look 7
at all the safety-related systems.
8 CHAIRMAN MICHELSON:
Could I give you an example?
9 You can tell me how that applies then to the case you just 10 exemplified.
11 The service water systems very often are, have 12 interties between non-essential loads and essential loads.
[')i 13 Non-essential loads are generally non-seismic piping and so 14 forth, but there is an isolation valve which you close il you 15 rupture the non-essential side, or if it gets excessive 16 drainage for whatever reason.
Now that isolation valve I 17 think can be usually shown to be safety-related.
18 Now normally it might be closed or maybe normally it 19 might be open.
I don't know.
Either situation could exist, 20 but in this position changeable situation, I guess that we 21 have to assume that if it were normally closed, and we had a 22 seismic event which it might disrupt some of that non-seismic 23 piping if we get a spurious signal for that valve to open, we 24 have to demonstrate that we can reclose it under whatever flow
-)
25 conditions are not existing out for the broken side of the HERITAGE REPORTING CORPORATION - (202)628-4888
87 1
non-seismic side of the system.
Is that the idea?
2 MR. ROTHBERG:
That is not a position changable 3
valve as we thought of it.
It is a safety-related valve.
4 CHAIRMAN MICHELSON:
Well, wait a minute.
Position 5
changeable can be safety-related I hope.
There are a lot of 6
those.
7 MR. ROTHBERG:
They can be.
The idea is that the 8
ability of that valve to close or to open is not related to 9
the fact that it is, it is just position changable.
It is a 10 safety-related valve and it has a function to close, and the, 11 it seems to me that the electrical system and the other things 12 that are necessary to make it change have to be designed I')
13 seismic event.
\\,_/
14 MR. MINNERS:
Can be either one I guess.
If it were 15 normally open, it would then be a safety-related valve.
16 CHAIRMAN MICHELSON:
Isolate it; normally closed.
17 MR. MINNERS:
Normally closed--
18 CHAIRMAN MICHELSON:
Then would be the position.
19 MR. MINNERS:
You have to say if it opens.
20 CHAIRMAN MICHELSON:
If you are keeping them 21 normally closed, you have to consider them under the position 22 changable situation, and whatever flow conditions exist, then 23 if it changes to its opposite position, you have to reclose 24 it.
I am not sure that's sublety that was ever necessarily 25 detailed to the NRC in the FSARs.
I'm not sure the utilities HERITAGE REPORTING CORPORATION - (202)E28-4888
88 1
necessarily ever detailed the flos, rates that have to close l
2 against them.
1 3
MR. MINNERS:
I'm sure not.
4 CHAIRMAN MICHELSON:
Under this program it still 5
won't be picked up.
Because it wasn't reported to you people, 6
then it doesn't have to be. picked up.
t 7
MR. MINNERS:
It was reported to us in a sense.
8 That service water system is reported to us as a required 9
system that must function during certain events.
The design 10 basis is that system has got to work during an earthquake.
11 CHAIRMAN MICHELSON:
So that's the design basis.
l 12 MR. MINNERS:
Design basis, got to work during
()
13 earthquake.
14 CHAIRtiAN MICHELSON:
That you might pick up some of 15 it that way.
16 MR. KIESSEL:
We did not change the definition of 17 design basis.
We changed the definition of operability, and 18 what is meant by an operable valve.
19 They may have never considered that valve to be 20 operable because its normal position is closed, and what we 21 are saying is hey, it doesn't make any difference what the
)
22 normal position of the valve is.
You have, must consider the i
23 fact that it can't be mispositioned, and that then catches it 24 under the position changable.
25 CHAIRMAN MICHELSON:
Okay.
Good.
Thank you.
There HERITAGE REPORTING CORPORATION - (202)628-4888
89 1
are a number of these interfacing valves.
I'm not sure they
\\
2 were even picked up under any ISI process.
3 MR. KIESSEL:
That's the problem.
They aren't j
4 covered under 5055AG.
5 MR. WOHLD:
The valve in your example, Carl, would 6
be considered passive, would not even be stroked.
7 CHAIRMAN MICHELSON:
That's right.
Motor-operated 8
header. motor-operators comes under this program.
If you have 9
blocked out the breaker, do it, then you could take it off the 10 list here.
11 MR. WOHLD:
What I was trying to say before, as far 12 as I remember, an inadvertently positioned valve was
)
considered a single failure in that you assume the other train 13 14 was available.
I don't know of any requirements in the FSAR 15 to re-position a mispositioned valve.
{
16 MR. MINNERS:
There is not.
That's why we put in 17 this generic letter we are telling you now we are 18 reinterpreting that and you must be able to re-position it.
19 That's a ratchet.
We agree.
That's a change, z
20 CHAIRMAN MICHELSON:
Got to define how many get 21 positioned.
{
22 MR. WOHLD:
Therefore, your Footnote 2 does not 23 apply.
24 MR. MINNERS:
It does apply.
]
25 MR. BAER:
The design basis event has changed.
It HERITAGE REPORTING CORPORATION - (202)628-4888
'90 1
is still a LOCA or a seismic event, or a main steam line break 2
or any of the extra 15 transients.
Those were not changed, 3
but we are saying if there is a valve in those systems that 4
could prevent one train from mitigating, the system from 5
working, yes, it is, you have to now consider it.
6 MR. KIESSEL:
Pete, we are not changing the 7
definition of safety-related, either..We are adding a whole 8
new category, position changable, and it is clearly spelled l
9 out in the first section under the recommended action.
10 CHAIRMAN MICHELSON:
Have to be a fairly large set.
11 MR. ROTHBERG:
It could be.
I don't know if it is.
12 I can't conceive that the licensee has many valves that he e's 13 hasn't locked up in the safety-related system.
g) 14 CHAIRMAN MICHELSON:
Air systems and water systems 15 and so forth where you are using them for dual purpose, is 16 non-safety.
17 MR. ROTHBERG:
If they are safety-related; otherwise 18 they would be in the program anyway.
19 CHAIRMAN MICHELSON:
Well, yes.
Not in terms of 20 testchility and so forth; they are classified safety-related 21 as--as Pete points out, not getting ISI.
22 MR. WOHLD:
I think there might be a sizeable number 23 of valves that aren't capable of operating against this
(
24 mispositioning.
25 CHAIRMAN MICHELSON:
Yes.
I think that could HERITAGE REPORTING CORPORATION - (202)628-4888
91 1
happen.
OV 2
MR. ROTHBERG:
Moving along, we came to paragraph E, 3
which along these same lines, we said that no change to the 4
existing plant design basis is intended, so that should be 5
inferred design basis review then should not be restricted to 6
determination of estimated maximum design basis pressure, but 7
rather they should look at the other parameters that are 8
necessary for the proper functioning of the MOV such as--and 9
then we ask, we gave an example.
The review should include 10 the effect on MOV performance of design basis degraded 11 voltage, and this again I believe was both an ACRS and a CRGR 12 comment.
r' 13 We are in paragraph F asking for documentation, and V) 14 we also put in a caveat about testing of the degraded voltage 15 conditions, and in G, we refer to our Attachment A, which has 16 grown to 34 items I think, or 33 items based on some 17 discussion I believe we had here the 28th of February, but the 18 list is now offered as an information item.
It is not meant 19 to be a checklist and/or a guidance to ar inspector.
20 We would like a program that addresses things like i
21 this, and I'm sure that the licensees are well aware of the 22 problems that they have had.
23 In H, we talk about trending, and in that, we j
24 referred people to NPRDS as an example of what is to be used.
O 25 On the top of page 6, Jim Conran called me yesterday HERITAGF REPORTING CORPORATION - (202)628-4888
92 1
, -)
and asked me to make another change, and he asked me to take
,U 2
out the phrase "a system such as."
He told me that that was 3
requested by Jordan.
The way the sentence reads now, it is 4
the NRC encourages the use of the industry-wide NPRDS for this 5
- purpose, j
I pointed out that that was sort of more 6
7 restrictive, but as Jim pointed out to me, the sentence says 8
the NRC encourages, and therefore the item shouldn't be 9
misconstrued as being, requiring these NPRDS.
In any case, we 10 did it.
We intended to do that, to take out that phrase such l
11 as.
12 CHAIRMAN MICHELSON:
There isn't any other game in
()
13 town at the moment, is there?
14 MR. ROTHBERG:
I don't know of one.
I'm not sure 15 that NRR is going to do the job.
16 CHAIRMAN MICHELSON:
I'm not, either, but that's the 17 only game, and it could do the job with proper modification 18 in it its present form.
19 MR. ROTHBERG:
At least it is industry-wide.
With 20 that in mind--
21 CHAIRMAN MICHELSON:
I would think, though, you 22 would have used a couple of adjectives such as the 23 appropriate, that system is going to have to be appropriately 24 adjusted to do this job.
You aren't going to accept it just
}
25 as is I wouldn't think.
The statement seems to say--
1 l
1 HERITAGE REPORTING CORPORATION - (202)628-4888
93 1
MR. ROTHBERG:
It is really not for us to accept.
2 It is the licensee's tool, and the carrot in that is in the 3
scheduling.
Down in paragraph J, we offer this carrot.
I 4
will get to that in a minute.
5 We have a schedule that the first round of tests I
j 6
bave to be done within five years or three refueling outages, 7
and within that time, what we would like licensees to do is to.
8 first send us a letter within six months commiting to the 9
program, and then within a year in that five-year interval, 10 get their schedule together.
11 CHAIRMAN MICHELSON:
Whatever they will do they will 12 have to demonstrate they can determine trends soon, and the
}
13 system doesn't do that.
14 MR. ROTHBERG:
Might be able to do it later.
15 MR. BAER:
Let me interject this paragraph is trying i
16 to reflect what we think is something that industry asked for.
17 The initial set of tests each plant has to do, but the 18 question of repetitive testing came up, and industry was 19 concerned about the burden in terms of manpower and cost and 20 scheduling, and said they very much like the concept of 21 determining the frequency of retesting based on actual 22 maintenance records, their own experience, and there was some 23 discussion of well, how do you get this on individual plant?
24 And it turned out we really like to share data, and what we 25 are trying to do here is say yes, we are encouraging you, HERITAGE REPORTING CORPORATION - (202)628-4888
94 1
industry, to go ahead and share data, and if in doing so, you g
.d 2
can justify less frequent testing, fine.
And as Owen said, 3
this is sort of a carrot, but we are not trying to mandate a i
4 system.
We are trying to say industry, we think you had a 5
good idea, and we are trying to give you the flexibility.
6 CHAIRMAN MICHELSON:
My only concern was if you 7
recognize that it is going to have to, cannot be used in its 8
present form, we should have so stated.
I would have added 9
the words the industry-wide NPRDS appropriately modified for 10 this purpose.
11 MR. MINNERS:
We went back to AEOD who are the 12 proper people to scan those words.
They gave us--and I guess
(])
13 they don't feel that way or--
14 CHAIRMAN MICHELSON:
You mean they think in its 15 present form by discontinuing to use it, it was okay?
16 MR. MINNERS:
I guess I don't know if that is what 17 they think.
18 CHAIRMAN MICHELSON:
If they think otherwise, this 19 should be slightly--
20 MR. BAER:
We will go back to AEOD.
21 CHAIRMAN MICHELSON:
Because I read this literally l
22 now to mean that use NPRDS, and that's right basically, but 23 reading it, some modification will be needed.
24 MR. ROTHBERG:
Encouraging--
25 MR. MINNERS:
We will talk to tnem, see if they l
HERITAGE REPORTING CORPORATION - (202)628-4888
95 1
would like to modify the words based on that comment.
O~
2 CHAIRMAN MICHELSON:
That's the concern.
3 MR. MINNERS:
We are not any experts on NPRDS.
4 CHAIRMAN MICHELSON:
The same people keep telling us 5
it isn't any good, and then you are endorsing it here.
6 MR. ROTHBERG:
As you say--in any case, I will refer 7
you to the last paragraph of J--the surveillance interval 8
should not exceed five years or three refueling outages, 9
whichever is longer, unless a longer interval can be 10 justified.
See item H for a particular MOV.
And that is 11 again carrot is that essentially trending and then maybe some 12 valves that after they do some work, we can allow them.
(}
13 CHAIRMAN MICHELSON:
Small-question on J--the first 14 sentence, the program for verification of the switch settings 15 on line item D, item D is really an item relating to 16 procedures, not the switch settings predominantly.
17 Wouldn't it be better to talk about verification of 18 the procedures outlined in item D?
Or was that--
19 MR. ROTHBERG:
Thats fine.
I don't have any 20 objection to that, but I think that these words--it's funny 21 how these things come about, but I think these words are i
22 pretty much out of 85-03, but procedures would be fine with 23 me.
24 CHAIRMAN MICHELSON:
85-03 ought not to be now the 25 sterling example of how to do it.
That was the first attempt HERITAGE REPORTING CORPORATION - (202)628-4888
96 1
to get something started, and I'm sure we have learned a whole 2
lot.
3 MR. ROTHBERG:
There was an attempt here to follow 4
as much of 85-03 as possible in order to allow licensees to 5
see where they could just extend their program.
That was the 6
idea.
The word procedures, I don't--
7 CHAIRMAN MICHELSON:
It is only because procedure is 8
more encompassing than sw.
"1 settings.
D really dealt with 9
procedures.
10 MR. WOHLD:
I just want to make a comment on 11 extending the bulletin.
12 I think there is a great danger at this point of
()
13 considering this as an extension of the bulletin.
A licensee 14 had an acceptable program under the bulletin, to simply extend 15 that program in light of the, of these Idaho tests and other 16 questions, I think there could be a great expense that that 17 would have to ne redone.
18 MR. ROTHBERG:
I think that is covered.
I think we 19 are covering where the Bulletin 85-03 fell short.
I think 20 that's covered in agonizing detail in the discussion.
21 MR. WOHLD:
Let me ask, in going through the 22 recommended actions, presuming it is possible to make the 23 switch settings independent of differential pressure, what you 24 see is a test requirements under that sort of condition?
25 In other words, if you bypass the open switch HERITAGE REPORTING CORPORATION - (202)628-4888
(
97 l
1 supplier which are the hard wire, is there any differential 2
pressure testing necessary?
3 MR. MINNERS:
How do you know the motor is going to 4
do it?
5 MR. WOHLD:
I guess what I am getting to is the I
6 original intent of the bulletin was the switch settings, and 7
the testing developed was around those switch settings.
I 8
don't know that the basic actuator size and capability was l
9 questioned to that great a degree.
10 MR. KIESSEL:
Pcce, we went through this at the last 11 ACRS meeting.
The original intent of the bulletin was not 12 just switch settings.
()
13 The original intent of the bulletin was to determine 14 the conditions under which the valve had to operate, determine 15 the switch settings that will allow the valve to operate under 16 those conditions, verify that the switch settings aren't made 17 properly, test the valve to demonstrate that it can be, that 18 it will operate properly, and finally, set up a program to 19 ensure that the valve will continue to operate.
20 MR. WOHLD:
The basis was to verify switch settings.
21 If you make the switch settings kind of independent of 22 differential pressure, what is the need in your having it full 23 voltage conditions?
You know, I don't--
24 MR. ROTHBERG:
Since you have locked out the 25 switches, what is the setting?
HERITAGE REPORTING CORPORATION - (202)628-4888
98 1
1 MR. WOHLD:
I am saying if you can develop a
(:)
i 2
technique to make the switch settings independent of 3
differential pressure, that reduces the need to do 4
differential pressure testing.
5 MR. BAER:
You've got to do some testing to show 6
operability.
7 MR. ROTHBERG:
Need to do--sometimes you won't be 8
able to do that in situ, will have to use some sort of 9
prototype test and extrapolation.
Hight be able to do it on 10 one ground in a plant and then extrapolate some of that data 11 to another valve even though it is not exactly the same, but 12 sooner ore later you are going to have to verify.
(
13 MR. MINNERS:
I think I have a problem when you say 14 make it independent of.
I think that's a problem.
We talked 15 to industry about that and said get rid of the damn switches, 16 and the answer was well, you might destroy the valve if you 17 don't have the switches.
18 MR. WOHLD:
If you get rid of the switches, that 19 gets rid of 90 percent of the concern under the bulletin.
20 MR. MINNERS:
You may destroy the valve.
You may 21 put out more torque.
22 MR. ROTHBERG:
You may not have a big enough valve.
23 Doesn't make any difference if you have switches or not.
24 MR. MINNERS:
Both ways switches can't necessarily 25 just crank it up and overpower it and get the other way.
You HERITAGE REPORTING CORPORATION - (202)628-4888
L 99 l
1 just can't presume that the actuater is big enough to operate l
2 the valve, l
3 MR. WOHLD:
I guess I don't know how much.
4 (A brief recess was taken.)
5 CHAIRMAN MICHELSON:
Go ahead, Owen.
6 MR. ROTHBERG:
The last portion of the letter deals 7
with reporting requirements, which it is mostly 8
administrative.
These were put in at the, essentially in 9
responding to NRR's requirements.
There were several changes 10 to item L, recommended by the lawyers, and I can read them to 11 you, and they are a little bit wordy, as things from lawyers 12 sometimes are, but what I would prefer to do is to just give
()
13 you a copy of it.
They really don't change any of the--
14 CHAIRMAN MICHELSON:
Can we get those right now?
15 MR. ROTHBERG:
Sure.
In item L, it reads now, for 16 any date that cannot be met, the licensee shall advise the NRC 17 of a revised schedule and provide a technical justification in 18 writing.
Then there is the phrase other realized or 14 alternative action along with the technical justifications 20 taken out, and a sentence was added instead, for any 21 recommendation that cannot be met, or that the licensee 22 proposes not to meet, the licensee shall inform NRC and 23 provide a technical justification, including any proposed 24 alternative action in writing.
25 Now that was again mandated by the lawyers, and they HERITAGE REPORTING CORPORATION - (202)628-4888
100 1
prefer that.
Then we made the next sentence, starting with the licensee, we took out the, and just said, licensees was 3
made a separate paragraph, and that reads almost identical to 4
the way it does now.
Except that the licensee should also 5
submit in writing was the paragraph, and at the end of that 6
paragraph, we took out for possible future inspection, again 7
included CRGR, they had said that was implied.
That was the 8
only changes that we made to that.
9 These are required actions in that NRC can mandate 10 that to 10 CFR, the commitment to the schedule, if the 11 licensee choses not to make the commitment, the NRC has to 12 decide what it wants to do from then on.
("N 13 Paragraph M instructs licensees to notify the NRC U
14 after that first round of tests has been completed, and then 15 we have a paragraph that instructs licensees, that supersedes 16 Bulletin 85-03.
17 We also have a paragraph that says that licensees 18 may use the results of tests that they used under Bulletin 19 85-03 hoping to avoid duplication, and then the rest is boiler 20 plate that under the OEM clearances, and things like that.
21 The 2,000 manhours per licensee is, was at one time 22 200 manhours.
The reason we did that is because now we are 23 allowing licensees to evaluate whether they want to have 24 alternative programs based on trending and that sort of thing, O
25 and it is going to take them some time to make those changes, HERITAGE REPORTING CORPORATION - (202)628-4883
101 1
so having done that, we have the reference, and the only other 2
thing is of course, Attachment A, which is nothing more than a 3
listing, and it is up to 33 items, this is nothing more than 1
4 a,
some guidance on what might be wrong inside MOVs.
5 That concludes what I had to say.
6 CHAIRMAN MICHELSON:
Now the information that is 7
developed under this generic letter is what is, the program is 8
defined to the NRC, I guess all the data is just retained on 9
site, is is that correct?
10 MR. ROTHBERG:
That is correct.
11 CHAIRMAN MICHELSON:
And that's the section now that 12 tests what the NRC will do with that on-site data.
[}
13 MR. ROTHBERG:
It doesn't really say what we are 14 going to do with it.
15 CH?,IRMAN MICHELSON:
Does it say we are going to do 16 anything, even look at it, may look at it, or can look at it?
17 MR. MINNERS:
Obviously we can look at it.
18 MR. ROTHBERG:
We can look at it.
19 CHAIRMAN MICHELSON:
Is that explicitly stated 20 anywhere here?
21 MR. KIESSEL:
Again, CRGR--
22 MR. BAER:
No.
23 MR. KIESSEL:
Asked to take out words to those 24 effect because one, it commits NRC to do something, and two, l
25 we c1 ready have that right, so therefore, it need not be l
HERITAGE REPORTING CORPORATION - (202)628-4888 L
102 1
reiterated.
O 2
CHAIRMAN MICHELSON:
What right do we have to ask 3
them to--let's ssy there are certain of this information that 4
we may wish to review in Washington.
5 Do you have the right to reproduce this at site or 6
something, and ship it off, or do you have the right to ask 7
for anything you wish, or how is that handled?
8 MR. KIESSEL:
I am not in the specter.
I can't 9
answer that.
10 CHAIRMAN MICHELSON:
It is kind of a question how 11 are you going to ultimately get the information?
You meaning 12 the NRC staff here in Washington.
{}
13 MR. MINNERS:
I don't think we are.
14 MR. KIESSEL:
There is no commitment for the NRC 15 staff.
16 CHAIRMAN MICHELSON:
You are never going to look at 17 any of this?
18 MR. MINNERS:
Never, never say never.
19 CHAIRMAN MICHELSON:
There is no intention to look 20 at any of this?
21 MR. MINNERS:
Correct.
22 CHAIRMAN MICHELSON:
Is that the idea?
23 MR. KIESSEL:
That is correct, by NRR staff.
24 MR. MINNERS:
By NRR.
25 CHAIRMAN MICHELSON:
Research, if they decide HERITAGE REPORTING CORPORATION - (202)628-4888
103 1
something looks interesting, or AFOD, how do they--
2 MR. BAER:
I think the concept is going to be there 3
is some inspection at the site.
That's how the NRR has 4
handled this on other issues is put together a team to go out j
5 and look at specific plants; maybe not every plant, but they 6
do have programs.
7 CHAIRMAN MICHELSON:
What I have in mind is that it 8
has been a sticky issue from the beginning almost as to the 9
accessibility of NPRDS data to the NRC staff, who can even 10 have a terminal like an acquisition system to data base all 11 the other questions about accessibility.
12 Then there is going to be a further question even
()
13 when you get accessibility, to what extent can you even 14 publish the results?
And I feel that that's kind of a 15 non-acceptable situation in a case like this because you are 16 going to have to have sufficient dissemination of the 17 information so that good decisions can be made in the future, 18 and I just wondered to what extent this information is going 19 to be deemed proprietary, available on site for the NRC to 20 look at, but not capable of being disseminated to our 21 contractors and so forth?
22 MR. MINNERS:
That doesn't concern me.
Proprietary 23 doesn't make any difference.
We can get proprietary 24 information and we can get proprietary information to our 25 contractors.
HERITAGE REPORTING CORPORATION - (202)628-4888
104 1
. CHAIRMAN MICHELSON:
Have a little trouble getting 2
it from NPRDS.
3 MR. KIESSEL:
No, sir.
4 CHAIRMAN MICHELSON:
Anybody in the Agency can get 5
it any time they wish?
6 MR. KIESSEL:
Yes.
7 CHAIRMAN MICHELSON:
You better inquire in more 8
detail on that.
9 MR. KIESSEL:
I just went through the training to 10 get it.
11 CHAIRMAN MICHELSON:
The staff people--
12 MR. KIESSEL:
Contacts were selected so that there
{}
13 were people in every division in NRR.
14 CHAIRMAN MICHELSON:
Go ask whoever you wish to ask 15 about the accessibility of NPRDS to our contractors.
One 16 contractor I know that can get it because there is a special 17 monitoring of the NPRDS and they are doing it.
18 MR. MINNERS:
We would have to give it to the 19 contractor.
You mean--as far as I know, the contractors are 20 not allowed direct access.
21 CHAIRMAN MICHELSON:
In order to give it, you have 22 to get it.
To get it, I guess you are going to have to go to 23 the utilities and reproduce it or something?
Is that the way 24 you think they will get it?
25 MR. MINNERS:
Yes.
HERITAGE REPORTING CORPORATION - (202)628-4888
l 105 1
MR. ROTHBERG:
Or on site.
2 CHAIRMAN MICHELSON:
Then you think that provision 3
is already implicit in our whole process?
l 4
MR. MINNERS:
Let me put it from our perspective or 5
from NRR's perspective.
NRR says they don't really have the 6
resources to review it, so why bother having people who send 7
it in?
8 CHAIRMAN MICHELSON:
I agree with that.
9 MR. MINNERS:
That's the basic premise that this is 10 set on.
11 CHAIRMAN MICHELSON:
If you wish to do some research 12 work with it, which I think you might wish to, should be
(}
13 pre-determined or pre-stated requirements that this become 14 available for such purposes without question.
15 MR. MINNERS:
I don't think, well, that would be 16 another change because that would be a burden on licensees if 17 we are going to ask them to provide information to us.
18 CHAIRMAN MICHELSON:
Ask them in the future.
19 MR. MINNERS:
We can always write them a generic 20 letter and say give it to us.
21 CHAIRMAN MICHELSON:
When you start trying to 22 collect this information, you may be surprised at its ready 23 accessibility.
24 MR. MINNERS:
We are already having that problem on 25 snubbers, okay.
If you ask for information, and people won't HERITAGE REPORTING CORPORATION - (202)628-4888
106 1
give it to you voluntarily.
2 CHAIRMAN MICHELSON:
This is a very important 3
program here, and the results are very important to make 4
future decisions, and I would think you would have some 5
reasonable provisions in here for proper acquisition of the.
6 information.
7 MR. MINNERS:
If we need it, we send them a generic 8
letter.
9 CHAIRMAN MICHELSON:
I guess that's what you have to 10 do.
You have to go through another generic letter to get it.
11 MR. ROTHBERG:
My personal experience with NPRDS is 12 that I have called up for pronounced and specific requests on
()
13 Limitorque operators and I have gotten it in very short order, 14 They sent me one stack almost a foot thick on Limitorque 15 things.
Limitorque, there is, was no trouble at all.
l 16 CHAIRMAN MICHELSON:
Did you do the searching 17 yourself?
l 18 MR. ROTHBERG:
No, as a matter of fact.
It was 1
19 easier than that.
20 CHAIRMAN MICHELSON:
I am not sure--that's the 21 easiest way, but not necessarily the more desirable way.
22 Researchers should be doing the research.
23 MR. ROTHBERG:
Again, there is a lot of--
(y-24 MR. KIESSEL:
There are 20 contacts through-'t the V
25 Agency.
HERITAGE REPORTING CORPORATION - (202)628-4888 L_ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
f 107 1
MR. ROTHBERG:
There are lots of ways to manipulate O
2 the data.
There is a handbook on how to do it.
3 CHAIRMAN MICHELSON:
I have the handbook on how to 4
do it.
I understand the system a long time ago, but the 5
concern has been it has not been readily accessible to our 6
contractors, for instance, and it is still not readily 7
accessible to our contractors, so the researchers are people 8
that you are hiring to do the research, cannot do the 9
searching.
You have to go through middlemen, which is a very
-10 poor way to operate.
11 MR. ROTHBERG:
You are right.
12 MR. MINNERS:
That may be a problem, but it is not
(}
13 one that this generic letter is designed--
14 CHAIRMAN MICHELSON:
I understand the generic letter 15 won't address that.
16 MR. ROTHBERG:
That's all I have.
17 MR. MINNERS:
And we ordinate it preferably with 18 AEOD, and that satisfies them, and they are presumably the 19 biggest user.
20 CHAIRMAN MICHELSON:
They have on-line people who 21 hire, goes out and hires people to do this work, is I think 22 where it is more difficult, but if they are happy with the 23 letter, that's fine.
24 Any questions?
Charlie, do you have any questions 25 on it?
Pete?
HERITAGE REPORTING CORPORATION - (202)628-4888
108 1
MR. WOHLD:
No.
Personally I think there should be bo 2
a great amount of involvement by the staff.
Maybe you don't 3
have the resources, but for instance, in the Bulletin 85-03, I 4
think there is a lot of valuable information that could have 5
been pulled together, and help direct this generic letter more 6
specifically, but that information is not available because it 7
was not asked for, but the way, with all the questions we have 8
on design basis, safety-related versus non-safety related and 9
so forth, unless there is some coordinated direction on this 10 thing, there is going to be a hundred widely differing 11 programs to meet this generic letter.
12 CHAIRMAN MICHELSON:
Who is this letter passed on to
()
13 now for the implementation stage?
14 MR. ROTHBERG:
The letter is ready to go back to NRR 15 under a cover of a memo from our office director, Mr.
16 Beckjord, to merely--and we have been holding it frankly 17_
awaiting to have this meeting.
18 CHAIRMAN MICHELSON:
But NRR will do the 19 implementing, whatever that is?
You can't speak for them as 20 to what they intend to do?
21 MR. BAER:
That is correct.
22 MR. ROTHBERG:
We have a fair idea.
There is a 23 generic communications list that lists Mechanical Engineering 24 Branch as having a temporary instruction to be prepared on 25 this.
HERITAGE REPORTING CORPORATION - (202)628-4888
109 1
CHAIRMAN MICHELSON:
They will pick up the ball and 2
run with it when it comes to interpreting and deciding what, 3
hcw they want certain things done, mechanical things?
4 MR. BAER:
That's the lead branch as we understand 5
it.
Of course, we will assist them if they want some help.
6 CHAIRMAN MICHELSON:
I see no other questions.
I 7
would like to thank the staff for giving us a fine 8
presentation.
I think that it is pretty clear as to where you 9
are headed, and I think the Subcommittee at least understands 10 the situation.
11 Our Full Committee meeting is scheduled for Friday, 12 April 7, 2:15 to 4:15, and during the Full Committee
{';
13 presentation, we would like to have the staff go through I 14 think much of the material you did this morning hopefully 15 without so many questions at least from us, but the questions 16 will be anticipated from other members, and I think that the 17 scope of what you covered this morning is about right.
18 Charlie, do you have any comments?
19 MR. WYLIE:
I think it's about right.
20 CHAIRMAN MICHELSON:
I'm sure that the Full 21 Committee is going to ask a lot more about cost / benefit and 22 that sort of thing maybe than we did, and you will have to be 23 prepared to field those kinds of questions, and beyond that, I 24 have no other words of wisdom as to what you anticipate.
1 g<w)
\\_
25 just don't know.
I have no good reading.
We have two hours.
HERITAGE REPORTING CORPORATION - (202)628-4888
b 110 1
MR. IGNE:
Two hours.
g-U 2
CHAIRMAN MICHELSON:
Two hours even; I would 3
anticipate that the staff will have availcble one hour of that 4
for, including presentation and questions, that we haven't-5 heard from NUMARC yet, but I assume that they may wish to talk 6
to the Full Committee, and basically that will give the other 7
hour, a fraction of that other hour.
If there is any time i
8 left, then we will have a Full Committee discussion before we 9
go.
10 Does that seem reasonable?
An hour is about all I 11 think you need.
Is that right?
12 MR. BAER:
Yes.
We will try and boil it down.
()
13 CHAIRMAN MICHELSON:
I think it will take a little 14 distilling down.
An hour will be okay.
15 Okay.
With that, let's go on to NUMARC's 16 discussion.
At the end of their discussion, we can find out 17 how much time they want for the Full Committee and then adjust 18 the schedule accordingly.
19 Who is going to speak for--I guess it is going to be l
20 NUMARC speaking?
21 MR. CALLAWAY:
With some industry rell,resentatives.
22 CHAIRMAN MICHELSON:
You can set up at the end of 23 the table here.
Why don't you sit right here, Pete, if you (v~}
24 want?
25 MR. CALLAWAY:
I am from NUMARC.
Mr. Brian Curry i
HERITAGE REPORTING CORPORATION - (202)628-4888
l l
111 1
from Philadelphia Electric, and this is Bob Elfstrom from the, l
2 from Toledo Edison.
3 First, I would like to express our appreciation to 4
the NRC staff and especially Mr. Rothberg for taking'our--
1 5
CHAIRMAN MICHELSON:
Pull that microphone up kind of 6
aiming towards your mouth.
It is fairly directional.
l 7
MR. CALLAWAY:
Also we would like to thank the ACRS 1
8 Subcommittee for giving us the opportunity to express our 9
viewn.
We believe that the flexibility that was-added to the 10 schedule and to the retest requirements should enable industry 11 to develop stronger programs and thus increase MOV 12 reliability.
(}
13 Our presentation here is aimed at trying to clarify 14 some points of the letter.
We at NUMARC are thinking about 15 having a workshop to aid in the industry's response to this 16 letter, so it is important for us to fully understand the 17 requirements of this letter.
18 My first comments are going to be concerning the 19 background section of the letter.
The subject of the generic 20 letter is safety-related motor-operated valves testing and 21 surveillance.
22 The question is since dampers are not valves, are 23 safety-related motor-operated dampers included in the letter's 24 scope?
25 MR. BAER:
Yes, by CRGR direction, and Mr. Sleezik, HERITAGE REPORTING CORPORATION - (202)628-4888
112 1
who is both that, was the office director, and member of CRGR, 7sd 2
pointed out that we have--well, I ought to go back.
We have 3
an exclusion, and CRGR asked us to, told us to take the 4
exclusion out, and Mr. Sleezik's comment was there has been 5
enough problems with control room habitability that he just 6
felt that I guess he said he is insecure with not having those 7
included.
I 8
CHAIRMAN MICHELSON:
Only if they are 9
motor-operated?
10 MR. BAER:
Only if motor-operated.
11 MR. CALLAWAY:
He found there have been many 12 problems with, with the dampers that were motor-operated
()
13 dampers?
14 MR. BAER:
I'm not that close.
I know there is 15 control room inspections that have been going on for the last 16 several years.
I'm not that familiar with the results.
I 17 guess he felt that way.
The problems were significant.
18 MR. ROTHBERG:
As far as that goes, his rationale 19 was not whether there were problems or not problems.
He felt 20 that, at least the words that I heard at the meeting were 21 these equipments are safety-related and therefore, they should 22 be looked at.
23 We have built in some flexibility in the retest 24 schedule so that they are not giving anybody any trouble.
It 25 may be that the--but as far as the design basis and looking at j
HERITAGE REPORTING CORPORATION - (202)628-4888
113 1
them initially, that was the meaning.
(J 2
MR. CURRY:
I have a question.
Is that limited to 3
dampers that are operated by Limitorque motor-operators, or 4
simply a damper that has an electric motor on it?
5 MR. ROTHBERG:
Electric-motor-operated they come 6
under the heading; are they considered to be valves, first of 7
all?
8 MR. CURRY:
I think you have already answered.the 9
affirmative, that you believe they are valves, that we have to 10 test them.
11 MR. ROTHBERG:
I'm sorry.
4 12 MR. CURRY:
My question is there are dampers our 13 there that have what are called electro-hydraulic actuators on
}
14 them that have electric motors, and as far as I know, there is 15 no diagnostic systems that have been developed for them at 16
- all, 17 MR. ROTHBERG:
Those electro-hydraulic actuators, 18 the best I understand them, what they do is they pump up the 19 motor there to pump a reservoir, and then the hydraulic 20 actuator is the actual mechanism for driving the, the valve.
21 I don't think those are envisioned to be, I didn't 22 envision them to be included.
When I spoke of an actuator, I 23 was thinking of the device that moves the disc.
24 MR. CURRY:
Okay.
So electro-hydraulics are not 25 included under your interpretation?
4 HERITAGE REPORTING CORPORATION - (202)628-4888
114 1
MR. ROTHBERG:
I wouldn't think so because we are, 2
we are not envisioning--what we were envisioning with this 3
generic letter was basically Limitorque and Rotorque 4
operators.
That covers what 98 percent of the industry.
We 5
had not considered some of these odd things.
We wanted to put 6
in a caveat to, to eliminate some of these things explicitly, I
7 but we can't do that.
8 CHAIRMAN MICHELSON:
That doesn't cover 98 percent 9
of the kinds of control devices you see in heating and 10 ventilating systems, though.
Very rarely are they Rotorques 11 or Limitorques, very rarely.
12 MR. BAER:
I think the answer is if it is a direct
()
13 motor driving the damper, I think it ought to probably be 14 included.
You know, licensees are always free to take an 15 exception also, and if it is a hydraulic device, 16 electro-hydraulic, that would tend to eliminate it and maybe 17 just state that in your response.
18 MR. ROTHBERG:
Another possibility is if it is that 19 much of a problem, I'm not sure that it would be considered to 20 be a valve.
I don't know.
21 CHAIRMAN MICHELSON:
The damper is not a valve 22 necessarily, although valves are used sometimes as dampers.
23 MR. MINNERS:
These are not dampers that we are 24 talking about.
()
25 CHAIRMAN MICHELSON:
What are they then?
HERITAGE REPORTING CORPORATION - (202)628-4888
115 1
MR. MINNERS:
They are valves.
They are used to O
2 close and isolate systems.
A damper is a piece of equipment 3
as I see it to modulate the flow.
4 CHAIRMAN MICHELSON:
It could be used for isolation, 5
too.
6 MR. ELFSTROM:
That's where our specific question 7
comes from.
This is Bob Elfstrom, Toledo Edison.
8 We have both kinds.
For example, we have 150 bound 9
glass fisher butterfly valves that are used as dampers for 10 positive isolation.
11 On the other hand, I probably have anywhere from 12 half a dozen to ten for lack of a better description dampers 13 in excess of venetian blind.
{}
14 MR. ROTHBERG:
Those fishers are not electro--
15 MR. ELFSTROM:
Limitorque operators; I have 16 Limitorque operated 8 inch butterfly valve used in the HVAC 17 system, has positive shutoff damper.
18 I also have 30 by 28 venetian blind type dampers 19 that are not positive shutoff that are operated by Limitorque 20 actuators.
21 The purpose of this question was to determine if 22 that was what was being covered.
When we had the explicit 23 exclusion, it was somewhat clearer, but now there is nothing 24 in here that defines it one way or the other, and while it 25 says valve, it is fairly obvious that it the fisher butterfly HERITAGE REPORTING CORPORATION - (202)628-4888
116 1
valve is a valve that can be used for air or used for water, 2
but it is not nearly so obvious whether the PAPCO venetian 3
blind type slat damper that is operated by Limitorque actuator 4
that does not have positive shutoff feature is to be l
5 considered as a valve.
l 6
MR. MINNERS:
What is a safety function of a damper 7
that doesn't have a positive shutoff?
8 MR. ELFSTROM:
They are essentially to close the, 9
divert flow from one section of ducting to another.
It does 10 not have to be a positive seal.
The positive seal is for 11 sealing off, controlling ventilation.
12 MR. MINNERS:
It doesn't really have a safety
()
13 function as I view it.
I can't envision--
14 MR. ELFSTROM:
Yes, sir, it does.
They are 15 definitely safety-related.
16 MR. MINNERS:
What is the safety function?
17 MR. ELFSTROM:
You are basing, you are basing your 1
18 ventilation on flow rates, and at that particular point in 19 time, if you want to redistribute your flow rates due to an 20 accident condition, okay, you don't have to have a positive l
21 seal.
You just have to maintain a specific flow rate going 22 from one particular section to another.
These--
23 MR. BAER:
Is it really that hard to test them?
I
)
l 24 have a two-zone heating system in my house and the venetian
}
25 blind dampers work, operate all the time until the relays get i
HERITAGE REPORTING CORPORATION - (202)628-4888
117 l
1 screwed up, l')
V 2
MR. ELFSTROM:
Exactly.
The problem we deal with is 3
the base line testing that you get into, for example, I have I
4 got a 5 inch column of water pressure drop in my design delta 5
P for this these dampers, and it is almost impossible to test l
6 now the, because of the arrangement of the fans and what you i
7 can do to harm them.
8 Now on the other hand, the OEM supplier has done 9
significant amount of work.
They probably have some of the 10 best sets of equations for the operation of these things that 11 I have seen as compared to valves.
They are very detailed and L
12 they have done a lot of testing, and we haven't had any
{}
13 difficulty with them as far as operational.
14 The difficulty you derive is standard problems you 15 would derive from any Limitorque operator that isn't 16 maintained properly or it has Limitorque associated problems, q
17 damper associated problems.
Functionality of the damper end 18 of the device has not been a problem, so you could use a 19 diagnostic system, for example, to monitor for degradations, 20 but you can't quantify the numbers out of it as to what kind 21 of force you are applying on this damper.
22 MR. ROTHBERG:
It is too small?
23 MR. ELFSTROM:
It is a quarter turn operation.
Just 24 like a butterfly valve, we don't have any ways of really 25 measuring direct torque that is being put on the stem.
HERITAGE REPORTING CORPORATION - (202)628-4888
118 1
MR. ROTHBERG:
Could you envelope it?
Could you say 2
that because it develops this much--
3 MR. ELFSTROM:
That's the way--
4 MR. ROTHBERG:
Essentially--
5 MR. ELFSTROM:
That's essentially the way we have 6
done, we have addressed these in 85-03 response, and we had 7
some difficulties with them figuring how you actually do 8
address them, but you essentially bound it by saying yes, I am 9
putting out this much by calculation through the gear train to 10 this, the stem of this, you know, device, and by the 11 calculations that have been done, supplied to me by the OEM 12 vender, this is adequate.
f'T 13 CHAIRMAN MICHELSON:
What is the usual rating for,
\\_/
14 you know, you have certain butterfly valves and whatever, used I
I 15 in ventilation systems that are designed to standard pressure 16 boundary codes as opposed to the other kinds of devices, the 17 butterfly--not the butterly but the venetian blind type which 18 is not designed for normal ASME code, and I think it is only 19 the devices that are code designed for, for the pressure
)
20 boundary that we are talking about here.
That's what I 21 thought.
I never considered that you are including all the l
22 other diverter type of designs.
23 MR. ELFSTROM:
That's the problem we are trying to 24 clear up.
It was clear when we had the disclaimer, but now 25 that it is not there anymore, the title specifically says HERITAGE REPORTING CORPORATION - (202)628-4888
119 1
valves, and now you have to make a determination what your
-l 2
definition of a valve is.
Normally a flow diverter type
{
3 venetian blind type damper is not considered as a valve.
1 4
MR. ROTHBERG:
You have to decide if it is a valve.
l 1
5 Then you have to decide if it is a motor-operated valve.
Then
]
6 you have to decide--
7 MR. ELFSTROM:
We know the fact the Limitorque--the 8.
generic letter because it is a Limitorque operated and
)
9 safety-related, but now that, the question is is this your 10 definition of a valve?
11 MR. ROTHBERG:
That's a good question.
12 MR. ELFSTROM:
We are not, all we are trying to do--
{}
13 MR. ROTHBERG:
I thought I knew what a valve was.
14 CHAIRMAN MICHELSON:
Not designed to ASME code, so 15 they ASA I guess, but they are solid valve body.
You are not i
l 16 talking about--
17 MR. ELFSTROM:
The ones I am trying to get 18 clarification on are the kinds that are not valves by design.
19 They are essentially, like I said, venetian blind inside of a 20 frame run by gear trains and they are not positive shutoff 21 device.
22 CHAIRMAN MICHELSON:
I have not considered those to 23 he included.
24 MR. ROTHBERG:
You had a discussion where you talk 25 about weir gates, I guess it Brian, weir gates, and my thought HERITAGE REPORTING CORPORATION - (202)628-4888
120 1
1 after thinking a little bit was that maybe they are not 7gV-2 valves, or a weir gate in my mind is not a valve, and that 3'
~ might be what you would do.
4 MR. ELFSTROM:
In our--this idea of evaluation, of a 5
workshop, as you have seen through the discuss.ons that have 6
been had, there is going to be difficulties with 7
interpretation and some confusion especially for utilities who 8
have not had a lot of experience in this area, and what we are 9
trying to do is set up a little bit ahead of time so that some 10 of these things are clear in our minds, because I agree that I 11 feel comfortable in coming bank and saying I am going to do 12 these valves and look, by the way, I have--these things are
()
13 Limitorques, but they are not valves so I am not going to do 14 this.
These are not part of the scope of this, and I just 15 want to find out what the feeling was.
16 CHAIRMAN MICHELSON:
Other than butterfly valves?
i i
17 MR. ELFSTROM:
Yes, sir.
l 18 CHAIRMAN MICHELSON:
What kind of other valves are 19 involved?
20 MR. ELFSTROM:
These are the PAPCO dampers, the 21 venetian blind slats.
22 CHAIRMAN MICHELSON:
That's the kind you would like l
23 to exclude, but the, those that you think ought to be 24 included, they are all butterflys?
)
25 MR. ELFSTROM:
Yes, that are used as dampers, but HERITAGE REPORTING CORPORATION - (202)628-4888
_ - __ _ _ ___-____ _________- _______ ______ _____ = ______________ - _ __- _ _ _ _ _ _
121 1
they are positive shutoff device and they are code designed
(~h 2
valves.
3 CHAIRMAN MICHELSON:
I think that's another key 4
word--positive shutoff device.
Dampers are never considered.
5 MR. ELFSTROM:
Again, it's kind of what you 6
determine your definition of valve as.
7 CHAIRMAN MICHELSON:
I thought all we included were 8
the positive shutoff devices operated by Limitorque operators 9
or--
10 MR. ELFSTROM:
Rotorques.
11 CHAIRMAN MICHELSON:
I didn't think we ever intended 12 to include all the ventilation.
13 MR. ROTHBERG:
The whole idea of the dampers, and 14 frankly the whole idea of dampers and weir gates and things, 15 that is something we have, did not envision considering.
16 Motor-operated valves, piping system, we are thinking of 17 things like water.
We were not thinking in terms--or steam.
18 We were not thinking in terms of pressure and air and that 19 sort of thing.
If it is, if it is service-related, and I 20 guess that's a butterfly valve, and seems to me that we 21 wouldn't be I guess having that as part of the definition.
22 On the other hand, these things like venetian blind 23 slats and things like that don't sound like valves to me.
24 Weir gates don't sound like valves to me, but I guess there O
1 25 could be other interpretation, but I think that what we have i
HERITAGE REPORTING CORPORATION - (202)628-4888
{
l
122 1
done in the letter is allow sufficient leeway for a licensee 2
to make his own interpretation for, within reason.
I don't 3
think that really affects the program.
I think 99 percent of 4
the program is all these conventional things.
5 MR. ELFSTROM:
We agree.
6 CHAIRMAN MICHELSON:
Isolation valves on the air 7
side, on the environment side, are the type he is referring 8
to.
They are all butterfly.
9 MR. ELFSTROM:
Butterfly valves.
10 MR. CURRY:
One thing about that definition, the 11 device or a valve that may be existing might be a big 12 assistance because in the case of containment atmospheric
(}
13 control systems, you have a pipe coming out of containment to 14 the valves and after that transitions to the ductwork, and 15 that would, you know, give you a limited scope and it is 16 reasonably easy to define what it is.
17 CHAIRMAN MICHELSON:
In some cases, people run the 18 pipe through the plants, spiral long pipe, a hundred feet.
19 MR. CURRY:
It is considered a valve until it gets 20 out of the piping system, and what I am thinking is I don't 21 think that pipe out to that valve from the containment--
22 MR. CALLAWAY:
We just want to have that clear.
23 MR. ELFSTROM:
The logic here is that you are 24 thinking along the same lines that we were, and the licensee 25 should take action to, if they have these types of devices, to E
"" "^ * " - (
- -4***
123 1
provide that in their response.
2 CHAIRMAN MICHELSON:
Heating and ventilating system 3
is a real interesting one in which to look at all various kind 4
of valves.
You have chiller systems and so forth and you 5
start wondering about the valves on the chilled water, on the 6
chilled water side and chiller package itself.
A lot of those 7
are motor-operated.
They are a fair number.
There is a, 8
quite an extensive utilization in the smaller stuff.
It is 9
safety-related, and I assume it is addressed under this 10 program, as long as it is motor-operated, and then the 11 ventilation system is the same except they want to know where 12 do you cut it off?
And I think that it is clear I would cut
()
13 it off at the, when you go to a blind type damper, venetian 14 blind, if you use a motor-operated butterfly valve as a 15 damper, it is under the program.
If it is safety-related, 16 yes, but those are details which I assume we can sharpen up.
17 I think a list of questions of the staff some day would be 18 appropriate in which to work out the fine structure.
19 MR. CALLAWAY:
Okay.
The next question we had a 20 concern that--
21 CHAIRMAN MICHELSON:
Could you speak up just a 22 little?
23 MR. CALLAWAY:
The next question we had pertained to 24 the industry-sponsored test program which we believe will 25 increase the cost substantially, but the question is why are HERITAGE REPORTING CORPORATION - (202)628-4888
124-1 you addressing an ladustry-wide issue in a generic letter 2
which is directed specifically at individual licensees?
3 MR. MINNERS:
Why not?
Let me turn the question to 4
you.
Why shouldn't we do it this way?
Is there something 5
that says we shouldn't do it that way?
6 MR. CALLANAY:
I think if we are going to have an 7
industry-wide initiative, it should include people like EPRI.
8 For instance, in the one-day workshop that you had recently 9
discussing the test results of the blowdown test result, I 10 think is a better forum for discussion of the industry about 11 the initiatives.
I'm not sure what the individual licensees 12 are going to do.
()
13 MR. MINNERS:
I would invite NUMARC to take the 14 initiative.
If you want to arrange an industry-wide whatever, i
15 I think the staff, you could invite staff and see if we want 16 to participate.
I imagine that we would, 17 MR. CALLAWAY:
All right.
18 MR. MINNERS:
So we are, we are not--I think it is a 19 matter of mechanisms.
I think a generic letter is a legal 20 form that needs to be decided upon.
I don't think that we 21 want to go to a rulemaking action wh :h s really the ultimate 22 in generic form, so but I think we can, you know, accommodate 23 ourselves to the generic format and style, deal with the
(
24 industry.
25 MR. ROTHBERG:
I assume you are talking about the HERITAGE RSPORTING CORPORATION - (202)628-4888
____---_-_-_________________]
125 l
1
. paragraph on the top of page 3 which is in the, specifically O
2 we put that in the background section.
It says therefore, 3
licensees "are also encouraged to consider the need for 4
industry-sponsored MOV test programs to ensure that. diagnostic 5
techniques can be used to determine switch settings to ensure 6
operability of those safety-related and MOVs for which it is 7
not practical to test at design basis conditions.
8 MR. BAER:
Is that the paragraph you referring to?
9 MR. ROTHBERG:
Top of page 3.
10 MR. MINNERS:
We have worked on other generic issues 11 vith NUMARC in which they have produced generic documents for 12 the use by the individual licensees, and then the licensee
()
13 just comes in and he references that and uses that forum.
We 14 have accepted it, so we have had these kind of mechanisms 15 worked out and I think that's all that we are saying.
You 16 guys'want to try doing it, that's acceptable and probably 17 preferable to us rather than dealing with each licensee, 18 case-by-case.
19 CHAIRMAN MICHELSON:
Ultimately you have to deal 20 only with the licensee.
21 MR. MINNERS:
Only licensed licensees.
22 CHAIRMAN MICHELSON:
That's the only legal 23 arrangement there is.
They can all speak with one mind if j
24 they wish.
25 MR. BAER:
And certainly, you know, owners; we have HERITAGE REPORTING CORPORATION - (202)628-r388
126 1
done the same with owners groups, too.
2 MR. ROTEBERG:
I also want to emphasize that we put 3
this in the background section and we offered it as a 4
suggestion only.
We didn't put it in the recommended action.
5 MR. WOHLD:
This paragraph reminds me the switch 6
settings--in this paragraph here you are referring to th9 7
correctness of switch settings, but in the testing.that you 8
are looking for, you really are not looking for svitch 9
settings.
You are looking for the capability of the unit, 10 whether it is switch setting limited, or whether it is 11 operator limited or limited by the valve characteristics.
12 MR. ROTHBERG:
That's
- 4. good point.
Perhaps it is a
()
13 little too restrictive to say switch setting.
It may be that 14 correct adjustments might be a better word perhaps.
15 MR. WOHLD:
You are really interested in valve 16 capability, actuator size or the limit switch.
17 MR. ROTHBERG:
I have no problem with changing it.
18 I have to check with the people who do that.
It may not get 19 done because of the way this thing is--you have to get 20 consensus of a lot of people.
I will take a waack at it.
21 MR. WOHLD:
My concern on this is that the plant 22 manager that lives out there in Toledo Junction reading this 23 is going to get a new graduate from school and have him go out 24 and adjust a bunch of switch settings, do some DP testing, and 25 feel like he is home free, and I think it is unbalanced toward HERITAGE REPORTING CORPORATION - (202)628-4888
'137 1
switch settings because of the history of the way that we got V
2 where we are, t
3 MR. ROTHBERG:
Adjustments for corrective action--I 4
have no problem with that.
5 MR. WOHLD:
You are really not extending the 6
bulletin.
You are coming up with a lot of requirements--
7 CHAIRMAN MICHELSON:
Proceed, keeping in mind that 8
there is another meeting that starts at one o' clock, and the 9
rest of the time will encroach upon somebody's lunch hour.
We 10 can go until one o' clock if necessary.
11 MR. CALLAWAY:
The next thing we have to talk about 12 is cost.
Owen mentioned that perhaps we would decrease the
()
13 costs from our initial figures, and in some respects we may 14 because we have more time now if you do the testing, and some 15 of the retesting since it is based on trending, that we may 16 end up doing fewer tests, but since our costs are based on the 17 number of hours that are being tested, it looks to me like now 18 that we have to include position changeable non-safety valves 19 and perhaps valves that influence external events, it seems to 20 me that the cost may increase more than decrease, and Brian 21 has done some calculations here, and we would like to present 22 those.
23 MR. CURRY:
What we did is we looked at the 24 sensitivity analysis that the company did and we didn't, you 25 know, it was done as we are sitting here so it is not quite as HERITAGE REPORTING CORPORATION - (202)628-4888
128 1
thorough as we wanted it done, but using the figures that show g-
'V 2
up here, that the staff assumes that the worst or the highest 3
cost to the industry to implement this thing as being $52.6 4
million dollars on that net. permanent value basis.
5 Now what we do, the worst says average case--this is 6
what we, from what we have seen in our costs, 85-03, that the 7
net present value to test the valves does not fit, not even 8
close to S 52 million.
It's $1,868,382,000 and some change.
9 MR. MINNERS:
How do you get that number?
10 MR. CURRY:
We took the number as our estimate.
It 11 costs around 500 to 70 million dollars to test all the valves 12 the first time around.
()
13 MR. MINNERS:
How do you get that number?
14 MR. CURRY:
We came up with a number that, from the 15 engineering basis, it costs around $5,000 to test the valve or 16 excuse me, $5,000 to do the engineering evaluation, and then 17 valve test costs run around $15,000.
One of the--
18 MR. MINNERS:
How many valves are you testing?
19 MR. CURRY:
A hundred and 50 valves per plant; now 20 one of the things we saw in the cost / benefit study initially 21 is that the study assumed that if I had 150 valves in a plant, 22 I started with valve No.
1, and as soon as I would start with 23 that one, I moved on to 2 and on to 3 then on to 4, if there
/'s 24 were no breaks there at all, these valves that we are testing U
25 are used for maintenance, used for locking purposes, they have HERITAGE REPORTING CORPORATION - (202)628-4888
129 I
1 got to remain operational, even when the plant is shut down, 2
and thus we have a lot of restraints on when a valve can be 3
tested.
4 It cannot be a case where I roll my testers in and 5
half a week later they are done and out of there.
There l
6 appears to be a good chance that refueling outages would end 7
up being limited by valve testing.
I have got to teet these 8
valves before I can come back again.
That's one of the 9
biggest reasons that our costs to test a valve is a lot 10 higher, because we are aware of the--it taker time.
We can't 11 take people to test valves and do other things with them.
It 12 takes people's time.
It is real costs we have got to spend,
()
13 and that's how we came up with this.
14 MR. ROTHBERG:
Rather than bickering aboit whether 15 your costs are higher than our costs and driving tha*., NUREG 16 CR 5140 documented our costs, where we got our information 17 from and why we made the choices that we made, and I package 18 the sensitivity analysis that they just did with that.
19 If you want to have NRC consider the alternative, I 20 would suggest that you document it in the same way where you 21 get a set of costs and send it to NRC would be my thought on 22 the subject.
l 23 I looked at your letter originally.
Well, it had, 24 it had limit lines, but it didn't really tell us where the 25 information came from, where the cost data came from.
HERITAGE REPORTING CORPORATION - (202)628-4888
1 130 1
MR. CALLAWAY:
What we did, we looked at 24
..O 2
different power plants and we asked them what it cost them to 3
do the bulletin.
4 MR. ROTHBERG:
And they gave you a number?
5 MR. CALLAWAY:
They gave us numbers, and we arranged 6
those numbers and we came up with a figure that I thought was 7
high, but it turned out to cost about $20,000 to test a valve.
8 MR. BAER:
My understanding was just what Brian 9
said, that a lot of that was because you couldn't test very 10 efficiently.
You had to keep certain systems running, and I 11 guess I thought that when we went to three refueling outages 12 to accomplish the one-time through program, that that would
()
13 help considerably in that regard.
14 MR. CALLAWAY:
I think it will help.
15 MR. BAER:
The cost didn't go down.
16 MR. ELFSTROM:
Won't decrease very much because we, 17 those are intangibles to a large degree.
We know they cost us 18 more.
It is just like we didn't include the cost of 19 scaffolding records because we didn't keep records that are 20 that accurate, but the records that we did use in coming up 21 with these cost figures are fairly accurate, and they are 22 really based on the average amount of time that it takes to 23 test the valves with the amount of people involved in the 1
24 field costs are biggest, are the biggest contributors, and
]
25 again, you are not even considering their consideration of the l
4 HERITAGE REPORTING CORPORATION - (2021628-4888 I
l l
131 1
1 1
time'of the operation individual who has to do the tag-out,
. f,b l
2 for example, so we feel that based over the 24 plants we had, 1
i 3
we had numbers that were as high as $45,000 per valve for some 4
plants, it's, it's a fairly conservative estimate, and the 5
problem that, the derived problem here is the bottom line l
6 that we end up with.
It is magnitude.
It just is monstrous.
7 MR. BAER:
Let me just, let--bear with me for a 8
second, because I looked at the 570 billion, million, half a 9
billion, and even said that's a one-time cost.
We use 10 everything Owen did this morning, intentionally used our 11 minimum safety benefits.
If I used our best estimate safety
.2 benefits, it can accumulate to that, but I wanted to explcre
()
13 your original total of 1.8 billion which apparently hasn't 14 changed, did not have any present worth factoring into it at 15 all, and we have given you now a more relaxed schedule.
The 16 number hasn't changed, even though now you say it does include 17 present worth.
I ar, confused on that.
18 MR. CURRY:
The original figure that we showed on 19 our our slide was 3.3 million.
That was not a net present 20 worth number.
The 1.8 billion is a net present worth number, 21 and while we spread the time out, we still have to spend the 22 money to test the valve, and whether you do more or less 23 valves per hour, the cost per valve is still there.
24 MR. BAER:
I am talking about the future, the 25 subsequent, the repeated, which was not talked about really, HERITAGE REPORTING CORPORATION - (202)628-4888
132 requiring testing.
We just make sure that you look at them L
2 periodically.
3 MR. CURRY:
I'm--let me go and speed up and see if 4
we'can get there.
If we said that position changable valves 5
would increase our numbers by 10 percent, it comes out to 6
roughly $2 billion.
It is something like 39 times higher than 7
the 52 million number.
8 Now using the beneficial and most beneficial 9
estimates, the biggest savings to industry, and assuming that 10 we can reduce our testi.; cost by 50 percent, now we are just 11 going to chop it by 50 percent, we are down to about $1.5 12 billion to do this over the entire timeframe.
Using, removing
()
13 our estimated or using our estimated cost and removing your 14 cost to do the thing, which was low, $24 million--
15 MR. CALLAWAY:
The benefits.
16 MR. CURRY:
We came out that the benefits, which is 17 shown here as $435 million, changes to a cost to industry of 18
$1 billion, and using the same number of how many dollars per 19 person-rim, it now comes out to $5,897 per person-rim.
20 MR. BAER:
I don't know if I follow all your 21 numbers, but I can do the last number.
If you have a billion 22 dollar total cost, and if you's a best estimate of a million 23 man-rem, it comes out exactly a thousand dollais per man-rem.
(
24 That's as good at arithmetic as I still am.
25 MR. CURRY:
I am looking at your number that, that HERITAGE REPORTING CORPORATION - (202)628-4888
1 133 1
we are going to have 185,000 person-rim.
fs
(_J 2
MR. BAER:
That was our low estimate.
3 MR. ROTHBERG:
Use best estimate.
4 MR. BAER:
If you look at the best estimate--
5 MR. CURRY:
There's a mistake.
Where is it?
6 MR. ROTHBERG:
Right there on the best.
7 (There was a brief pause in the proceedings.)
8 MR. CURRY:
Okay.
Calculator?
9 MR. MINNERS:
Mr. Chairman--
10 MR. BAER:
Take my word for it.
It is a thousand.
11 Mk. MINNERS:
I think this is not a--
12 CHAIRMAN MICHELSON:
If you have a case to make,
()
.s. ' ant.
Let the staff 13 document your case, send us the 14 review the-document along with ourselves, and we can ask 15 questions on it, but just to sit, go back and forth on the 16 details of it, it is not, not fruitful for this Committee, so 17 if you do have a concern or want to make a case, you are 18 certainly invited to come in and make a spiel to the Full 19 Committee when we meet.
20 If it is important enough, we can have another 21 Subcommittee meeting, but I would rather not since we have to 22 move on this quickly, but you do have to keep in mind that we 23 have to see things kind of packaged up concisely.
We don't 24 have the time.
We have another meeting at one o' clock, and so 25 we don't really have the time to debate the points.
HERITAGE REPORTING CORPORATION - (202)628-4888
134 1
I think we have heard some of your concerns.
If you O
2 have others that you might be able to sum up fairly concisely, 3
we would like to hear those, but if you have a real concern 4
with the costibenefit analysis, for instance, then do your own 5
analysis with appropriate substantiation, and send it to the 6
staff and us for consideration.
We will certainly be willing 7
to entertain it, but I don't think we can kick it around the 8
table.
9 MR. BAER:
If you could send something in, let us 10 look at it, maybe we could sit down sometime before the Full 11 Committee meeting, try and arrange for something next week.
12 CHAIRMAN MICHELSON:
The Full Committee will meet on
()
13 this next Friday, a week from Friday, at 2:15 in the 14 afternoon.
15 MR. BAER:
I have got two other Full Committee 16 meetings next week which I think are currently scheduled for 17 Thursday, so I might ask for something like Wednesday or 18 something.
19 Why don't you see if you can send us in something 20 fairly quickly and at least let us look it over?
And if 21 nothing else, we can get back to you on the phone certainly 22 and maybe sit across from the table.
23 I agree with Warren.
It is pretty hard to do.
As 24 you had difficulty following Owen's numbers, we are having 25 difficulty following yours.
HERITAGE REPORTING CORPORATION - (202)628-4888 i
135 1
1 MR. CALLAWAY:
Right.
Okay.
The next point that we
,_c 2
want to bring up was with the position changable valves.
3 Our question was if you actually meant to include 4
non-safety valves that were in safety systems?
That question 5
has been answered.
Frankly we couldn't believe that you would 6
want to include valves that may or may not be designed to 7
operate under accident conditions.
8 MR. KIESSEL:
If they are not designed to operate 9
under accident conditions, they better be locked out.
10 MR. CALLAWAY:
You know, there are many aspects of 11 safety in a power plant that depend upon the procedural 12 methods, and most of those valves in all, all the valves in
(}
13 the power plant are included and very extensive valve line-ups 14 have to be signed off and verified and so--
15 MR. MINNERS:
People still take off with their flaps 16 up.
17 MR. KIESSEL:
That does not, that does not preclude 18 the operator, the plant operator--
19 MR. CALLAWAY:
Of course it doesn't.
20 MR. KIESSEL:
In crisis condition pushing the wrong 21 buttons, getting his system misaligned.
22 MR. CALLAWAY:
Anyone can do that, but what--that 23 does have an influence, and it has an influence on the design 24 of the power plant, and they may not have taken them to 25 account in mispositioning of these position changable valves HERITAGE REPORTING CORPORATION - (202)628-4888
136 1
in the design of the power plant.
It may be outside of the O
2 design basis documentation because they assumed that certain 3
valves are going to be positioned manually and left that way 4
according to administrative procedures, and so we are dealing S
here with an issue that is outside of the, outside of the 6
design basis of the power plant.
I am not saying it is.
7 MR. KIESSEL:
It is not outside the design basis of 8
power plant.
The system must function for certain design 9
basis events.
I 10 MR. CALLAWAY:
The system must function per a 11 design, and if you look at the design of a lot of systems, it 12 has certain valves are normally open, and certain valves are
{}
13 normally closed.
14 MR. KIESSEL:
And we all know they never get 15 mispositioned!
16 MR. CALLAWAY:
I am not saying they don't get 17 mispositioned, but I'm sayir.; because the system is designed 18 with certain valves being open or closed, it may or may not 19 fall into the single failure criteria that is in the design 20 basis review.
I am saying there may be, there may well be a 21 spot there, and--
22 MR. BAER:
I guess if it is operationally possible 23 to put those in position and remove power, which I can 24 remember doing ECCS reviews where certain mini-flow valves on 25 Westinghouse plants, this was required to meet single failure, HERITAGE REPORTING CORPORATION - (202)628-4888
137
_1 I think that's certainly an acceptable approach.
7s V
2 MR. CALLAWAY:
The problem with that is now you have 3
the lights off on the valve and you can't tell if the valve 4
has been mispositioned or not.
If you take the power off the 5
valve--
6 MR. BAER:
There is plenty of valves where that has 7
been done.
8 MR. CALLAWAY:
I realize that is a common practice, 9
but for a lot of utilities it may not be favorable.
10 MR. BAER:
Every time this issue comes up, I guess 11 all the LERs--and I start sending Owen LERs where things got 12 mispositioned, not the operator action, but there is two I
(}
13 sent you yesterday which you may not have received, inverter 14 failure in one case, and I forget what the other one was, but 15 almost ever stack of LERs I get there are valves that have 16 moved.
Inverter failures, jumpers, one might have been 17 degraded voltage, momentary different voltage, and--
18 CHAIRMAN MICHELSON:
Control system related.
19 MR. BAER:
Yes, and so it really isn't just 20 administrative procedures and operator.
I mean administrative 21 procedures and line-ups, that can get them mispositioned.
22 MR. CALLAWAY:
I appreciate that.
I'm just saying 23 that if there is systems initially designed assuming certain 24 valves would be left in certain positions, it may well be p(/
25 difficult to find the design basis.
HERITAGE REPORTING CORPORATION - (202)628-4888 L
138 1
MR. KIESSEL:
What ve are saying is'we have enough.
(
2 evidence to indicate that that assumption is not valid.
3 MR. CALLAWAY:
Regardless of validity of that, you 4
know what--the designer problem, the design of the system, 5
that is how it was designed.
6 MR. MINNERS:
We agree with that, but the system j
7 is--so what we now found out later, afterward we made a 8
mistake and decide the system is wrong.
Didn't consider 9
everything we should have considered.
Now we are trying to 10 play catch-up and fix it.
That's what we are trying to do, 11 and we admit that.
It is a backfit in that sense.
12 MR. ROTHBERG:
The design basis remains the same.
13 MR. CALLAWAY:
Right.
It hasn't changed.
14 MR. JOHLD:
The original design basis of the plant 15 did not include present positioning these valves so that they 16 may not be qualified.
It is only by chance are they going to 17 be qualified.
For instance, look at the safety injection 18 valves to the core.
19 MR. ROTHBERG:
That is a big safety problem.
It is 20 a system safety.
21 MR. WOHLD:
The valves are positioned, assured for 22 line-up by systematic checks, but I think it is a matter of 23 chance whether those are qualified to be repositioned during 24 the--so this may be a much greater problem.
This isn't an oh, 25 by the way.
This could be a significant issue.
HERITAGE REPORTING CORPORATION - (202)628-4888
139 1
CHAIRMAN MICHELSON:
These are details which we just
.f-(/
l
-2 simply don't have the time to get into today, nor is it 1
3 appropriate for us to get into this kind of detail.
I think 4
they would like to hear from NUMARC as to what their general 5
problems are, with soine specific examples, but let's not get 6
into debating the details because that's going to have to be 7
worked out somewhere else.
We just don't have the time today, 8
keeping in mind that we have another meeting at one o' clock.
9 MR. CALLAWAY:
Okay.
Thank you.
Bob is going to 10 talk about the design basis review.
I'm sorry.
Brian is 11 going to talk about design basis review, 12 MR. CURRY:
From our conversations today, I think
()
13 this should be--we do not have a problem here, but what you 14 are really asking for is a system-based review, and that you 15 don't want us to simply do a docur.ent with you of the original 16 purchase specifications that were supposed to take into 17 account credible scenarios--valve, licensing commitments, and 18 at the same time, abnormal conditions or operation would be 19 used to bound normal operation where it is possible, and that 20 supply to the MCC with elevated temperature and degraded 21 voltage should be there.
22 Is that a correct, you know, interpretation?
23 MR. BAER:
I think basically so, although that 24 list--
25 MR. MINNERS:
Take it a little slower.
HERITAGE REPORTING CORPORATION - (202)628-4888
140 1
MR. ROTHBERG:
I regard that specifically, no 2
change.
Let's see the design basis review--I am rear.ing from 3
item E, paragraph E, generic letter, on page 5, the design 4
basis review should not be restricted to a determination of 5
estimated maximum design basis differential pressure.
The 6
design basis review should include an examination of the 7
pertinent design and installation criteria that we used in 8
choosing the particular MOV.
For example, review should 9
include the effect of degraded voltage, on like that.
10 That's what I had in mind.
If flow is more 11 important than pressure, then flow is the criterion.
If the 12 design basis for the plant was 80 percent degraded voltage,
()
13 then that sF7uld have been included, and I would have thought 14 that that rould have been included from the beginning.
15 We are not asking particularly for a witch hunt, but 16 on the other hand, what we are asking is that licensees be 17 aware of the pertinent design criteria from which they set out 18 their motor-operated valves and choose the size of those motor 19 operators because what you have to do is compare the very 20 minimum, the size of the motor operator that you have, with 21 the requirement to open and close against these conditions.
22 That's what we need, and I don't believe that anybody can
)
23 justify saying well, we don't know this or we can't find this, 24 because to me, that's unacceptable that a licensee should not 25 know what the design basis is for the operation of these I
i l
HERITAGE REPORTING CORPORATION - (202)628-4888 j
i:
141 1
motor-operated valves.
l 2
MR. ELFSTROM:
We are not saying that.
We are just l
3 trying to, to give you our insight on it.
4 CHAIRMAN MICHELSON:
Let me go back and remind you 5
this is an ACRS meeting, not a meeting between the licensee, 6
representative of the licensees and the staff.
We expected 7
the staff to tell us what problems they had, what position 8
they had on this final document.
We expected you to tell us, 9
not debate back and forth with the staff as to what they meant 10 or what they are doing and whatever, so the presentation is to 11 the ACRS, not to, not a question and answer period with the 12 staff.
()
13 You can have that all you want at another time and 14 place, but right now we would like to know what industry 15 complaints are, what their problems might be and whatever with 16 this particular document and as it is prepared now, can we 17 move forward?
And if you have any pitch you want to make to 18 us, please do so, and if you have got problems with staff and 19 their interpretation, deal with the staff on that, so I would 20 expect that we would not discuss this further with the staff 21 but discuss it with the ACRS, so your presentation is to us 22 and not to the staff.
23 MR. CURRY:
We do have one problem area in that we 24 have a hard time understanding how installation criteria 25 affect the design basis.
HERITAGE REPORTING CORPORATION - (202)628-4888
142 1
CHAIRMAN MICHELSON:
Could you give us a little more 2
of an example?
3 MR. CURRY:
An example would be how I weld a valve i
4 into a pipe doesn't seem to affect the design basis and it l
5 doesn't seem to affect switch set settings.
6 MR. MINNERS:
Is that a question or a statement?
7 MR. CURRY:
It is a statement.
He asked where we 8
have a problem.
We have got a problem there.
I don't now how 9
to interpret it.
10 MR. BAER:
I don't understand the problem.
11 CHAIRMAN MICHELSON:
Would you explain what your 12 problem is that prompted this particular comment?
()
13 MR. CURRY:
We don't know how to address the 14 bulletin with respect to installation criteria affecting 15 design basis.
We don't see that it--
16 MR. MINNERS:
You left out a word.
We have all of 17 these bureaucratic words we use.
If you leave them out, we 18 get into trouble.
The pertinent word is pertinent.
Paragraph 19 E halfway down, design basis review should include an 20 examination of the pertinent design installation criteria.
21 Now if you have got some welding criteria and that's not 22 pertinent to switch settings, hey, okay.
23 MR. CURRY:
We have a hard time understanding 24 what--anything that is in installation of criteria is 25 pertinent.
HERITAGE REPORTING CORPORATION - (202)628-4888
143 1-M R., MINNERS:
We do, too, but you have-got to O
2 realize we are bureaucrats.
We don't know the details.
We 3
have got this all-encompassing statement to say hey, it is up 4
to you guys-to figure out what is pertinent.
We don't know.
5 I don't know.
6 MR. CURRY:
It's a God, country and motherhood 7
statement?
8 MR ~. MINNERS:
Right.
9 CHAIRMAN MICHELSON:
I think I, much of what we deal 10 with is a little in that degree.
It has to be because we are 11 doing a very high level.
12 MR. MINNERS:
That's right.
I didn't write the
(}
13 words.
You didn't have any specific installation criteria 14 that you thought might affect it?
15 MR. ROTHBERG:
Well, at the point I wrote the word 16 installation criteria, I was thinking of wiring at the time.
17 The words got changed around and some of the wiring business 18 got put elsewhere, and eliminated, and what this is what, what 19 resulted, and I think you are characterizing it as a general 20 statement of, of what might be pertinent to affect the 21 operation of the motor-operated valve.
22 MR. MINNERS:
The specific example was the wires 23 weren't big enough.
24 MR. ROTHBERG:
Precisely.
That was the thing I had 25 worked on at that time.
HERITAGE REPORTING CORPORATION - (202)628-4888
144 I
1 CHAIRMAN MICHELSON:
Very limited; could we proceed gV i
2 on with some of your other points?
I think this one has been j
3 understood.
q 4
MR. ELFSTROM:
This is addressing paragraph B which 5
is discussing switch setting methodology.
We have essentially 6
taken the information you derived from paragraph A, which is 7
you do your design basis review, you end up with differential 8
pressures, any other pertinent requirements, and then you 9
establish correct switch settings.
10 At this particular point is the point where the 11 licensee would take the information derived from this design 12 basis review, and plug it into an equation to come up with the 13 appropriate switch setting for the valve.
14 Questions concerning the equation for gate valves I 15 think has been well documented at this point in time, and we 16 feel that the licensees are going to have to do some 17 additional analysis maybe to accommodate these questions that 18 have been raised.
19 The difficulty we have at this point is probably 20 anywhere from 40 to 50 percent of the actuators and valves 21 covered under this generic issue are quarter turn valves, L
l 22 butterfly valves, plug valves.
23 We have two distinct difficulties there.
We want f
24 the lack of diagnostics to directly measure torquing input to 25 the valve stem which is what you are provided with to l
HERITAGE REPORTING CORPORATION - (202)628-4888
145 1
determine operability, and the second is that at this point in
( )
2 time, all we really have other than ensuring the proper 3
operation diagnostics can still do that as far as the trending 4
and showing glitches in the physical operation of the assembly 5
itself, but at this point in time, all the industry still 6
really has is the equations to go in and set these OEM valves 7
up to this type of valve, and while delta P testing which gets 8
into paragraph C, will be done as much as possible, we now are 9
getting into the range of valves where you are talking 24 inch 10 butterfly valves, for example, something which now gets 11 service water system, gets very difficult to establish 12 conditions to do this adequately with, and our concern is that
(~}
13 this whole generic letter is really pushed toward rising stem LJ 14 valves, and it doesn't really address any specifics concerning 15 quarter turns and the, and the inability of the industry to do 16 adequate diagnostic testing to this extremely, validate these 17 settings versus OEM equations.
18 We don't feel at this point that the OEM 19 equations--we do { eel they are conservative in all the i
20 applications that we have run across at this particular 21 juncture, so we feel I guess at this point that the OEM 22 equations for non-rising stem gate valves seem to be adequate 23 when everything in the weather is pointed in the opposite fs 24 direction, that OEM equations are inadequate.
\\s) i 25 CHAIRMAN MICHELSON:
Have you looked the at the l
HERITAGE REPORTING CORPORATION - (202)628-4888
146-1 results of the testing that was done at--where was that--Idaho
.LJ 2
or INEL, which clearly indicated that there was a Int that 3
wasn't well understood?
How we handle that problem I'm not 4
sure, but you can't use the kinds of approaches you have been 5
using on gate valves or they won't work.
6 MR. ROTHBERG:
Mr. Chairman, the NRR approach as 7
stated by Ernie Rossi was that it is really not our problem in 8
NRC to solve the various difficulties that industry may have 9
in determining what their design basis forces and thrusts are.
10 They have equipment in the plant that they have made 11 a commitment to have operationally ready from the design basis 12 conditions, and I think what I am hearing here is that they
()
13 don't know what those conditions might be.
14 MR. ELFSTROM:
No, sir, that is not what I am 15 saying.
Item A, you do your design basis review, you come out 16 with those conditions.
The next thing you have to do with 17 them is apply those conditions to essentially a switch 18 setting.
l 19 MR. ROTHBERG:
Didn't you do that to begin with?
20 How did you do that when you set up the valve?
When you chose 21 and purchased and installed the valve?
22 MR. MINNERS:
He used design equations.
23 MR. ROTHBERG:
You are saying now that you may not 24 have a, you may not rely on those design equations?
They may 25 not be right?
i HERITAGE REPORTING CORPORATION - (202)628-4888
147 1
MR. ELFSTROM:
That's not what I am saying.
What'I g-
%s 2
am saying is for certain types of valves, specifically riser 3
stem gate valve, there has been testing that has shown there 4
may be a lot of conservatism in the equation, 5
MR. BAER:
I want to go back.
The bulletin 6
originally was carefully written.
You won't find the word 7
diagnostic system or equipment in there anywhere I don't 8
believe.
9 MR. KIESSEL:
In the requirements.
10 MR. ELFSTROM:
We will find it --
11 MR. BAER:
In the discussion perhaps, okay.
I stand 12 corrected.
But there wasn't a lot of diagnostic systems
()
13 around when the bulletin was written.
Industry has learned a 14 lot about rising stem gate valves as a result.
15 Now I admit there is not yet a comparable set of 16 knowledge about the other types of valves, but I am sort of 17 like with--when somebody has selected these valves, and 18 hopefully the valves are operable if called upon to mitigate i
1 19 an accident, and I guess there is a concern that they won't i
20 operate if an accident or an event occurs, and saying hey, we 21 don't know how we are going to go about selecting the switch 22 settings and really kind of says we don't know how we did it 23 originally, we are not sure that the equipment, the original l
24 equipment manufacturer's equations were right isn't very
[}
25 reassuring.
l 1
l l
l HERITAGE REPORTING CORPORATION - (202)628-4888
)
148 i
1 MR. ELFSTROM:
That's not what we are saying.
,s i
'V 2
CHAIRMAN MICHELSON:
What are you saying then?
3 MR. ELFSTROM:
For example--
4 CHAIRMAN MICHELSON:
Say it again.
5 MR. ELFSTROM:
We as an industry know that there 6
have been a rising stem gate valves tests that have been done 7
that show that the equation for whatever reason may not be 1
8 conservative in all cases.
That is fairly common knowledge at i
9 this point.
The letter, again the vast majority of it, it 10 really leans toward rising stem valves.
11 For example, in item C, this letter is not intended i
12 to establish recommendation for valve testing for the
()
13 condition of simulated line break--however, the break in the 14 line should be considered in the analysis described in item A 15 and B above.
The MOV operations rely on the design basis.
By 16 the item 24 inch butterfly valve which is Mr. Michelson's 17 example, that isolates my seismic qualified, the service water 18 system from the turbine system at this particular point in 19 time, okay.
It is put there to mitigate a line break.
The 20 only thing I have available to me at this point in time is the 21 OEM equation, for that type of valve.
I cannot measure 22 quantitatively the torque that I am putting out on to the 23 valve stem.
I can do it by calculation.
That's the way it 24 was originally sized in the first place.
25 CHAIRMAN MICHELSON:
The butterfly you are saying?
HERITAGE REPORTING CORPORATION - (202)628-4888
149 1
MR. ELFSTROM:
Yes, sir.
All we are saying, at this b,~s 2
particular point in time, there are not any means other than 3
the means that we have available to us.
We can refine them by 4
making sure our design basis conditions are correct, but other 5
than that, that applying that to the switch settings on these 6
valves, I don't know what else the industry can do at this 7
point.
8 CHAIRMAN MICHELSON:
We can develop sufficient test 9
information to validate what you have been assuming in the 10 design basis thus far.
I think you're acquainted with the 11 tests that were done at INEL and the butterfly valve.
12 MR. ELFSTROM:
Specific type.
f~T 13 CHAIRMAN MICHELSON:
Which brought some particularly V
14 interesting problems to light; now the NRC does testing for 15 the purposes of deciding whether or not a problem may exist, 16 not for the purposes of solving the industry's problems for 17 them.
The industry has to on out and solve the problem.
I 18 don't think the NRC is going to go out and solve it.
1 19 MR. MINNERS:
What I hear you saying is you have a j
20 problem, and I agree you have a problem.
21 CHAIRMAN MICHELSON:
That's what I hear, too.
You 1
22 can't look to the NRC to solve your problem.
23 MR. ELFSTROM:
What I am saying is we don't feel we 24 have a problem for the quarter turn valves.
I 25 MR. MINNERS:
You think the equations are i
HERITAGE REPORTING CORPORATION - (202)628-4888
1 150 I
1 satisfactory.
l
.]
2 CHAIRMAN MICHELSON:
If you can present test data or j
3 other substantiation to say there isn't a problem, that's
)
4 great.
5 MR. ELFSTROM:
Then this generic letter is telling 6
us we have to do a test program to validate?
L 7
CHAIRMAN MICHELSON:
If you haven't done one, I 8
think it is suggests you better do one.
9 MR. CALLAWAY:
What we are saying, the letter
-10 implies the OEM equation is not conservative, and we are 11 saying that we feel it is.
12 CHAIRMAN MICHELSON:
You are talking butterfly
(}
13 valves?
14 MR. CALLAWAY:
Yes.
15 MR. MINNERS:
What is your basis for that feeling?
16 MR. CALLAWAY:
Well, the basis for that is all the 17 testing that has been done at nuclear power plants on quarter 18 turn valves, and the fact that we have--
19 CHAIRMAN MICHELSON:
Where is this documented?
20 MR. CALLAWAY:
I am not talking about prototype 21 testing.
I am just talking about a surveillance testing.
22 CHAIRMAN MICHELSON:
They are not done under the 23 kind of full conditions.
You really have to tell us where the 24 documentation is that they can be reviewed independently to 25 assure that, the Agency that you have done it and done it l
HERITAGE REPORTING CORPORATION - (202)628-4888
151 1
correctly and that it is--substantiate your case when you 2
submit that documentation you don't think there is a problem.
3 I haven't seen that documentation.
I don't see it much 4
anyway, but I haven't heard of its existence in terms of 5
within the Agency.
6 You have to substantiate these things to the Agency, 7
not just people that have done tests and say it is okay..
8 That's not peer review at all.
9 MR. CALLAWAY:
Trying to prove a negative isn't 10 easy.
If you go out there and do tests, surveillance, valve 11 surveillance tests, if you find that you have very few 12 problems with butterfly valves which have been set up through
()
13 an OEM equation, you can conclude through that that you have 14 conservatively set--
15 CHAIRMAN MICHELSON:
Not at all; not at all.
16 MR. CALLAWAY:
You don't have--
17 CHAIRMAN MICHELSON:
I don't think this is the 18 place.
If you show that the tests were under simulated 19 accident situations for which these valves are now expected to 20 to perform, then you could--I am not aware of such a test.
21 And maybe they have been done, other words, isolating a 22 break and so forth that these va3ves, if that has been done, 23 that's great.
I don't, I haven't seen that.
Maybe you could 24 send us a--the industry only has to document a defendable 25 case.
I believe that's all the staff is asking for.
HERITAGE REPORTING CORPORATION - (202)628-4888
l 152 1
MR. ELFSTROM:
At the same time, sir, you are O
2 specifically saying something that the letter specifically 3
excludes, and that's line break.
4 CHAIRMAN MICHELSON:
That's something coming up 5
later.
That's not here right at present.
6 MR. ELFSTROM:
Under those conditions; now we have 7
to rely on analysis.
8 CHAIRMAN MICHELSON:
Or test.
l 9
MR. MINNERS:
Okay, for the moment.
10 MR. ROTHBERG:
You don't have to.
You could run the 11 test if you want to.
12 MR. ELFSTROM:
Yes.
I suppose we can.
Some of
(}
13 these are going to take a while to get to a facility that will 14 handle it.
15 MR. ROTHBERG:
I understand that, but the point was 16 that this note was put in there as a relaxation of our 17 requirement that said we recognize that you can't use your 18 facilities and test stand, and so we are going to use, you can 19 rely on, on analysis, but you can't have it both ways.
If the 20 analysis is flawed, you have reason to believe that the 21 analysis is flawed, then you have to do something else it 22 seems to me.
23 MR. ELFSTROM:
Oh, I don't disagree.
I guess it 24 just depends on--
25 CHAIRHAN MICHELSON:
This does not exclude these HERITAGE REPORTING CORPORATION - (202)628-4888
153 l
t 1
items from consideration.
It only is the question of what l
2 kind of argaments that you might be able to make presently in 3
them.
You don't have test data in some of these areas.
4 MR. ROTHBERG:
That's right.
5 MR. BAER:
On the particular valve that you are 6
mentioning, if you don't have line break, you are not going to 7
get phased flow.
This.is subcooled surface water, and isn't, 8
if there is a test that could be run with you know, the pump 9
at that, service water pumps running, and see if--again on the 10 non-safety side of it--
11 MR. ELFSTROM:
The particular valve I have tested 12 under delta P conditions, and with full flow that I can get
()
13 from my service water, the system.
14 MR. BAER:
Doesn't that give you much better 15 assurance?
16 MR. ELFSTROM:
Gives me very reasonable assurance 17 that it works.
My difficulty with it is I still don't have 18 any--we are talking about validating the equation.
To be able 19 to validate the equation, I have to have quantifiable data, 20 that I can't say this is a go, no-go test.
I can't measure 21 the amount of torque it took me to--
22 MR. BAER:
As an engineer, I tend to be analytical.
23 I would rather have the valve tested than the equation l
24 validated.
You know, someone could go out and somehow l
25 magically test every valve in the power plant and tell the HERITAGE REPORTING CORPORATION - (202)628-4888
154 1
peak conditions, that to me is better proof than--
2 MR. ELFSTROM:
Go, no go testing I am real happy 3
with.
I won't have a problem.
Unfortunately, I get can't get 4
there from here.
5
\\
CHAIRMAN MICHELSON:
Whatever case, it is necessary 6
to present a quantifiable case that you believe that the valve 7
will operate.
Some kind of analysis has to be done, and it 8
has to be done on the basis of some reasonable number o 9
and so forth, not just speculations.
10 MR. ELFSTROM:
Yes, sir.
11 CHAIRMAN MICHELSON:
And I think that that exercise 12 has to be performed and it has to be shown how close to that lll 13 situation you can test it.
In some licensees, in 14 situ--clearly you can't break a pipe in order to test it and 15 so you do some kind of in situ testing and so that is 16 extrapolated to the design basis conditions to some 17 comparability.
Those exercises have to be carried out as I 18 read the the letter.
19 The Full Committee, which will be at 2:15 on Friday, 20 presently, is that correct?
21 MR. IGNE:
Two fifteen.
22 CHAIRMAN MICHELSON:
Does the industry wish to make 23 any presentation to the Full Committee?
24 MR. CALLAWAY:
Let me get back to you on that.
25 CHAIRMAN MICHELSON:
We will need to know fairly
- _ - - - - - - - -HERITAGE REPORTING _CORPORATTI%f---1^^^"^
^^'~
l 154 f
I peak conditions, that to me is better proof than--
' 0 2
MR. ELFSTROM:
Go, no-go testing I am real happy 3
with.
I won't have a problem.
Unfortunately, I get can't get i
1 4
there from here.
5 CHAIRMAN MICHELSON:
Whatever case, it is necessary 6
to-present a quantifiable case that you believe that the valve 7
will operate.
Some kind of analysis has to be done, and it 8
has to be done on the basis of some reasonable number of facts 9
and so forth, not just speculations.
10 MR. ELFSTROM:
Yes, sir.
11 CHAIRMAN MICHELSON:
And I think that that exercise 12 has to be performed and it has to be shown how c3ose to that
()
13 situation you can test it.
In some licensees, in 14 situ--clearly you can't break a pipe in order to test it and 15 so you do some kind of in situ testing and so that is 16 extrapolated to the design basis conditions to some 17 comparability.
Those exercises have to be carried out as I 18 read the the letter.
19 The Full Committee, which will be at 2:15 on Friday, 20 presently, is that correct?
21 MR. IGNE:
Two fifteen.
22 CHAIRMAN MICHELSON:
Does the industry wish to make 23 any presentation to the Full Committee?
24 MR. CALLAWAY:
Let me get back to you on that.
25 CHAIRMAN MICHELSON:
We will need to know fairly HERITAGE REPORTING CORPORATION - (202)628-4888
155 1
soon, next'two days, and certainly no later than Monday next
(/
2 week.
No.later than Monday of next week we need to know if 3
you are on the agenda or not.
4 MR. CALLAWAY:
Yes, sir.
5 CHAIRMAN MICHELSON:
How much time--it will not be 6
more than an hour.
The entire presentation is some two hours.
7 MR. CALLAWAY:
Okay.
Thank you.
8 CHAIRMAN MICHELSON:
So it will be at most half the 9
time.
The staff is going to have a presentation for one hour, s
10 and if there is any time left after the, whatever 11 presentations are made, then we will have the Subcommittee 12 discussion.
()
13 Are there any, any questions, Charlie, or input?
14 Now what the Subcommittee needs to consider here is the, 15 whether to bring this forward to the Full Committee at this 16 time or not.
17 Do you have any problem with that?
18 MR. WYLIE:
No.
I think we should proceed with it.
19 CHAIRMAN MICHELSON:
I agree also.
You should bring 20 this forward.
There is a number of small questions about 21 information that maybe is still lacking, but I believe that we 22 have sufficient information with which to review this subject 1
23 at Full Committee.
24 I think that--I don't know what the advice of the 25 staff or the, or NUMARC as to what might come up, but f
HERITAGE REPORTING CORPORATION - (202)628-4888 L____-_____.
156 l'
basically the staff is going to make a presentation in their 2
handout in a condensed form, so in one hour, whatever 3
questions and discussion there is.
1 4
I would strongly recommend that NUMARC wishes to 5
make a presentation, reasonably formal, to'the point what is 6
your problem, what are your arguments, what are your. concerns, 7
what would you like the ACRS to do and not to do, or the NRC, 8
but it will be pretty much a straightforward presentreion with 9
questions.
It will not be any discussion between you and the 10 staff per se or the staff and you.
11 Normally in our Full Committee meetings we don't 12 allow the staff and participants to start doing any detailed
()
13 amount of chit-chat.
Certainly you can ask, if there is a 14 real question you can ask it and the staff will be willing to 15 answer and vice-versa.
Basically you are speaking for the 16 Committee and not to the staff, so keep that in mind.
17 If there is time remaining after the, after the two 18 presentations, then we will have a Full Committee discussion l
19 as to the nature of the letter and so forth, that I would then 20 have to draft for Full Committee consideration on Saturday.
I l
21 guess I haven't seen the schedule yet.
I think there is f
(
22 something still scheduled for Saturday.
23 MR. IGNE:
Yes.
24 CHAIRMAN MICHELSON:
Letter writing is still
)
25 scheduled on Saturday, so the letter writing would be--at the 1
l HERITAGE REPORTING CORPORATION - (202)628-4888
i 157 1
time we do schedule to write a letter, it is nice if the staff ss 2
wishes to come down and answer any questions.
If the 3
questions arise, it is nice if a NUMARC representative wants 4
to be available to answer any question if one should arise, 5
but it will be pretty much straightforward presentation.
6 I would recommend that you present to the Committee 7
what your arguments are, what your positions are, what your 8
recommendations are in a rather well-packaged, concise manner 9
so that they can understand where you are coming from and what i
10 your concerns might be.
1?
At this time, I think unless there are other 12 questions on the part of the Committee, I believe we are f}
13 finished with, with the, with the staff and with the NUMARC 14 people, and so with that, I will declare that this meeting is 15 adjourned.
16 (Whereupon, at 12:45 p.m.,
the meeting was 17 adjourned.)
18 19 20 21 22 23
(:)
25 HERITAGE REPORTING CORPORATION - (202)628-4888
1 CERTIFICATE 2
3 This is to certify that the attached proceedings before the 4
United States Nuclear Regulatory Commission in the matter of:
I 5
Name:
Advisory Committee on Reactor Safeguards--Subcommittee on Mechanical Components i
7 Docket Number:
8 Place:
Bethesda, Maryland 9
Date:
March 29, 1989 10 were held as herein appears, and that this is the original 11 transcript thereof for the file of the United States Nuclear i
12 Regulatory Commission taken stenographically by me and, 13 thereafter reduced to typewriting by me or under the direction 14 of the court reporting company, and that the transcript is a 15 true and accurate record of the foregoing proces ings.
16
/S/
w' 17 (Signature typed):
Catherine S. Boyd 18 Official Reporter 19 Heritage Reporting Corporation 20 21 22 23 24 1
25 i
O Heritage Reporting Corporation f
(202) 628-4888
PROPOSED GENERIC LETTER IN SITU TESTING OF MOTOR-0PERATED VALVES PRESENTATION TO THE MECHANICAL O
COMPONENTS SUBCOMMITTEE OF THE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS MARCH 29, 1989 ENGINEERING ISSUES BRANCH 0FFICE OF NUCLEAR REGULATORY RESEARCH ROBERT L. BAER, CHIEF OWEN ROTHBERG, SENIOR TASK MANAGER 1
._ J
O
SUMMARY
OF PRESENTATION O
INTRODUCTION AND OVERALL
SUMMARY
R. BAER 0
SAFETYSIGNIFICANCE(RISKPERSPECTIVE).
R. BAER DETERMINISTIC BASIS FOR LETTER. 0.ROTHBERG O
O COST BENEFIT ANALYSIS.
O. ROTHBERG OTHER ACRS CONCERNS.
O. ROTHBERG U
OUTLINE OF GENERIC LETTER. 0.ROTHBERG i
O 2
I O
SUMMARY
OF RECENT ACTIONS CONCERNING GENERIC LETTER 0
LAST ACRS MEETING ON 2/28/89.
EXPRESSED CONCERNS 0
Q CRGR APPROVED LETTER, ESSENTIALLY AS WRITTEN IN 3/9/89 ENCLOSURE OF MEMO FROM BECKJORD TO J0RDAN l
U LETTER READY TO BE SENT TO NRR ACRS COMMENTS AND APPROVAL REQUESTED 1
O 3
l 1
I o()
I MOV SAFETY SIGNIFICANCE CALCULATED CORE MELT FREQUENCY (CMF)
USED SARAH CODE i
VARIED MOV FAILURE RATE FROM.029 TO.29 0
BEST ESTIMATE VALUE OF.087 IS VERY CLOSE TO LICENSEE REPORTED IEB 85-03 RESULTS CMF MAY BE UNDERPREDICTED
]
ALL MOVs COVERED BY IEB 85-03 WERE ASSUMED TO HAVE FAILURE RATE OF.0038 0
POSSIBLE SYSTEMATIC DESIGN PROBLEMS IDENTIFIED BY INEL TESTS NOT CONSIDERED COMMON MODE FAILURES NOT CONSIDERED CMF MAY BE OVER PREDICTED REC 0VERY FROM EVENT BY MANUAL OPERATION OF VALVE NOT CONSIDERED 4
M0V SAFETY SIGNIFICANCE COREMELT PROBABILITY PER REACTOR YEAR PLANT TYPE NUREG-11SO NUREG/CR-5140 M0V FAILURE RATES LOW HIGH BEST PWR 2.4E-5 2.7E-5 2.8E-4 7E-5 BWR 2.9E-5 6.4E-5 1.4E-3 2E-4 M0V FAILURE RATES NUREG-1150
.0038 NUREG/CR-5140 LOW EST:
.029 l
NUREG/CR-5140 BEST EST:
.087 NUREG/CR-5140 HIGH EST:
.29 BASIS FOR FAILURE RATES M0 VATS DATA IEB85-03 RESULTS ADDITIONAL MOV TESTS BY SAME LICENSEES l
O 5
O DETERMINISTIC BASIS FOR ACTION BY REGULATION, SAFETY-RELATED COMPONENTS MUST BE OPERATIONAL AND IN SITU TESTING MUST VERIFY OPERATIONAL READINESS 10CFR50, APPENDIX A, GDC CRITERION 1, 4, 18, 21 10CFR50, APPENDIX B, CRITERION XI CURRENT IN SITU TESTING, AS OUTLINED IN ASME CODE SECTION XI, DOES NOT PROVIDE SUFFICIENT ASSURANCE OF DESIGN BASIS OPERABILITY.
STR0KE-TIMING TEST IS NOT ORIENTED TOWARD DIAGNOSIS OF SPECIFIC DEFICIENT CONDITIONS.
m MOV DEFICIENCY DOCUMENTATION AEOD REPORTS C603, S503, E501, E305, C203 RECENT TEST PROGRAM PROBLEMS AT BRUNSWICK, FERMI, SURRY BULLETIN 85-03 RESPONSES ADDITIONAL MOV DIAGNOSTIC TESTING BY SOME LICENSEES INDICATES SIMILAR FAILURE RATES FOR ALL MOVs Gk.)
6
O l
I
)
I l
l I
1
)
COST-BENEFIT ANALYSIS 0
ADDED COSTS OF AN INDUSTRY-WIDE PROTOTYPE TESTING PROGRAM FOR MOVs AT DESIGN-BASIS CONDITIONS BEST HIGH LOW
-$25,000,000
-$10,000,000
-$47,000,000 O
(]
EFFECT OF ADDED NRC COSTS IS "SMALL" O
EVEN IF NRC POSTULATED INDUSTRY SAVINGS IS ELIMINATED, WORST-CASE ESTIMATE IS THAT PROGRAM WILL COST ABOUT $600 PER PERSON-REM SAVED. THIS IS CONSIDERED COST EFFECTIVE.
I O
7
"O 1
1 l
l r
ACRS CONCERNS 0
NRC COST IMPACT TO TRAIN PERSONNEL AND AUDIT ON-SITE SENSITIVITY ANALYSIS INDICATES THAT Q
COST IS SMALL AND !'AS NO IMPACT ON OVERALL CONCLUSION DEFINITION OF " SAFETY-RELATED" AND " DESIGN-BASIS" IS NOT DEPENDENT ON 10CFR 50.49. THESE DEFINITIONS ARE CONSISTENT WITH WORDING IN PARTS 21 AND 100 AS WELL AS THE SRP.
1 o
8
~
l
, (~)
\\_/
PREVIOUS ACRS CONCERNS (CONTINUED)
U THE STAFF WAS ASKED TO CONSIDER THE EFFECT OF REC 0VERY ACTIONS ON THE CORE MELT FREQUENCIES POSTULATED IN THE VALUE-IMPACT ANALYSIS RECOVERY IS CONSIDERED TO BE EQUALLY OFFSET BY THE FOLLOWING CONSIDERATIONS:
- OPERATOR ACTIONS MAY NOT NECESSARILY BE THE CORRECT ACTIONS ESPECIALLY IN
]
" STRESS" SITUATIONS EQUIPMENT MAY NOT BE ACCESSIBLE TO ACCOMPLISH PANUAL, REMOTE REC 0VERY ACTIONS
- CONM0N-MODE FAILURE AS DISCUSSED EARLIER O
9
()
GENEnIC LETTER OUTLINE l
ESSENTIALLY AS IN 3/9/89 MEM0 BECKJORD TO JORDAN 0
BACKGROUND DISCUSSION OF BASES FOR LETTER DISCUSSION OF INEL BLOWDOWN TESTS DISCUSSION OF GENERAL PROGRAM RECOMMENDED ACTIONS SCOPE - SAFETY-RELATED MOVs, POSITION CHANGEABLE MOVs OTHER MOVs AT LICENSEE'S OPTION REVIEW & DOCUMENT DESIGN BASIS ESTABLISH SWITCH SETTINGS
()
SET SWITCHES REVISE PROCEDURES FOR SETTING SWITCHES NO CHANGE TO EXISTING DESIGN BASIS DOCUMENT TESTS & REVIEW 85-03 MOVs FOR DEGRADED VOLTAGE LIST OF DEFICIENT / DEGRADED CONDITIONS TRENDING AND ANALYSIS O
v 10
(
GENERICLETTEROUTLINE(CONTINUED) i 0
SCHEDULE ACCOMPLISH FIRST TEST OF ALL MOVs WITHIN l
l 5 YEARS OR 3 REFUELING OUTAGES. SETUP l
SCHEDULE IN FIRST YEAR.
SUBSEQUENT TESTING EVERY S YEARS OR 3 l
REFUELING OUTAGES, OR BASED ON TRENDING, If JUSTIFIED.
REFUELING OUTAGES THAT START WITHIN 6 MONTHS OF LETTER DO NOT COUNT.
O REPORTING REQUIREMENTS REPLY WITHIN 6 MONTHS
,]
SEND LETTER IF SCHEDULE CHANGES BASED ON TRENDING KEEP DOCUMENTATION ON SITE SEND LETTER AFTER ALL MOVs HAVE BEEN TESTED ONCE O
il
8aDe No."
1
.03/23/89 FINAL REPORT STATUS OF IEB 85-03 i
SUMMARY
BY FAILURE HATE f
AS OF MARCH 23, 1989 NUMBER-NUMBER NUMBER TOTAL BYPASS TORQUE LIMIT NUMBER HUMBER PER CENT UTILITY NUMBER SVITCH SWITCH (SWITCH OTHER INOP INOP PLANT NAME REPLY VALVES CHANCED CHANCED. CHANGED DEFECTS VALVES VALVES CINNA 07/08/88 30 30 29 2
30 25 83.33 ROBINSON 2.
01/22/88 11 0
6 0
0 9
81.82 ARKANSAS 1 02/29/88 34 0
34 0
1 19 55.88 FARLEY 2 08/15/88 37 18 35 25 4
15 40.54 SEQUOYAH 2 08/05/88 21 0
18 0
0 8
38.10 MILLSTONE 3 06/27/88 33 0
32 0
0 12 36.36 SUSOUEHANNA 1 01/12/88 22 11 11 11 10 6
27.27 FARLEY 1 08/15/88 37 15 36 15 3
10 27.03 ARKANSAS 2 02/29/88 41 0
39 0
0 11 26.83 MILLSTONE 2 06/17/88 23 0
7 0
0 6
26.09 l
OCONEE 1 01/14/88 14 14 12 14 0
3 21.43 COOPER 09/30/88 19 0
8 0
0 4
21.05 VNP 2 12/18/87 21 0
20 0
0 4
19.05
'SUSOUEHANNA 2 01/12/88 22 1
12 0
5 4
18.18 KEVAUNEE 01/04/88 18 0
18 0
18 3
16.67 QUAD CITIES 1.
01/15/8B 12 0
11 0
0 2
16.67 l
OU81 TIES 2 01/15/88 12 4
9 0
0 2
16.67
]
HAD M NECK 06/27/88 14 0
10 0
0 2
14.29 OCONEE 2 01/14/88 14 14 11 14 0
2 14.29 OCONEE 3 01/14/88 14 14 12 14 0
2 14.29 DRESDEN 2 01/15/88 8
0 6
0 0
1 12.50 l
PERRY 1 03/31/88 24 11 19 0
21 3
12.50 DAVIS-BESSE 1 04/22/87 165 26 0
19 164 17 10.30 SEQUOYAH 1 08/05/88 21 0
8 0
0 2
9.52 FITZPATRICK 01/15/88 24 0
14 0
3 2
8.33 PALISADES 01/15/88 24 0
23 0
0 2
8.33 BAN ONOFRE 1 09/01/87 12 1
3 0
0 1
8.33 NORTH ANNA 2 01/04/88 25 25 24 0
0 2
8.00 INDIAN POINT 2 02/26/88 13 0
11 0
0 1
7.69 VOLF CREEK 01/14/88 34 34 27 34 0
2 5.88 TROJAN 12/15/87 35 1
26 3
35 2
5.71 210N 1 01/15/88 35 0
5 0
0 2
5.71 PRAIRIE ISLAND 2 04/08/88 21 0
19 0
0 1
4.76 RIVER BEND 1 02/17/88 21 1
7 3
4 1
4.76 SAN ONOFRE 2 09/01/87 23 22 11 0
0 1
4.35 BRUNSVICK 1 05/25/88 25 0
0 0
0 1
4.00 NORTH ANNA 1 01/04188 25 25 24 0
0 1
4.00 BRUNSVICK 2 05/25/88 26 0
16 0
0 1
3.85 RANCHO SECO 01/19/88 32 18 14 0
23 1
3.12 FERMI 2 07/25/88 33 0
21 0
22 1
3.03 C A L[^}A Y 1
04/12/88 34 24 13 1
4 1
2.94 210h-/
01/15/88 35 0
0 0
0 1
2.86 BEAVER VALLEY 1 04/18/88 24 0
0 0
0 0
0.00 BEAVER VALLEY 2 07/31/87 16 7
10 0
11 0
0.00 BIC ROCK POINT 01/20/89 1
0 0
0 0
0 0.00
,Page No.
2 03/23/89 FINAL REPORT STATUS OF IEE ES-L:
SUMMARY
BY FAILURE RATE
)
7~3 i,)
AE CF MARCH 23, 1989 NUMEER NUMBER NUMBER TOTAL BYPASS TOROUE LIMIT NUMBER NUMBER PER CENT UTILITY NUMBER SVITCH SVITCH rSVITCH OTHER INOP INOP PLANT NAME REFLY VALVES CHANCED CHANCED CHANCED DEFECTS VALVES VALVES BRAIDV00D 1 01/15/88 16 0
0 0
0 0
0.00 BRAIDV00D 2 01/15/88 16 0
0 0
0 0
0.00 BYRON 1 01/15/88 16 2
0 0
0 0
0.00 BYRON 2 01/IL/88 16 5
0 0
0 0
0.00 CALVERT CLIFFS 1 01/28/88 12 0
10 0
0 0
0.00 CALVERT CLIFFS 2 01/28/88 12 0
3 0
0 0
0.00 CLINTON 1 03/02/87 23 14 12 6
20 0
0.00 CRYSTAL RIVER 3 02/15/88 17 1
2 2
17 0
0.00 DIABLO CANYON 1 09/14/87 27 0
19 0
0 0
0.00 DIABLO CANYON 2 09/14/87 27 0
19 0
0 0
0.00 DRESDEN 3 01/15/88 8
0 7
0 0
0 0.00 DUANE ARNOLD 01/15/88 24 0
22 13 17 0
0.00 FORT CALHOUN 1 08/07/87 8
6 8
2 0
0 0.00 FORT ST. VRAIN 01/14/88 13 13 0
0 13 0
0.00 CRAND CULF 1 02/27/88 23 0
15 0
0 0
0.00 HAR 1
03/04/87 28 0
0 0
0 0
0.00 IN POINT 3 01/15/88 10 10 9
4 3
0 0.00 LAS LE 1 01/15/88 18 0
1 0
0 0
0.00 LASALLE 2 01/15/68 19 0
7 0
0 0
0.00 LIMERICK 1 11/17/87 34 0
0 0
0 0
0.00 MAINE YANKEE 12/15/87 20 0
19 0
20 0
0.00 MCCUIRE 1 03/01/89 43 0
31 0
0 0
0.00 MCCUIRE 2 03/01/89 43 0
34 0
0 0
0.00 MILLSTONE 1 06/27/88 9
0 8
0 0
0 0.00 MONTICELLO 03/15/88 19 10 11 10 0
0 0.00 NINE MILE POINT 1 09/18186 32 26 5
1 9
0 0.00 NINE MILE POINT 2 11/13/87 21 0
13 0
0 0
0.00 OYSTER CREEK 1 01/15/88 20 19 17 0
0 0
0.00 PALO VERDE 1 01/15/88 33 0
26 0
0 0
0.00 PALO VERDE 2 01/15/88 33 0
26 0
0 0
0.00 PALO VERDE 3 01/15/88 33 0
31 0
0 0
0.00 POINT BEACH 1 06/30/88 24 0
6 0
0 0
0.00 POINT BEACH 2 06/30/88 18 0
4 0
0 0
0.00 PRAIRIE ISLAND 1 04/08/88 21 0
19 0
0 0
0.00 UALEM i 01/05/88 17 0
9 0
0 0
- 0. 00 SALEM 2 01/05/88 17 0
12 0
0 0
0.00 SAN ONOFRE 3 09/01/87 23 20 20 0
0 0
0.00 SEABROOK 1 11/3C/87 30 7
5 6
22 0
0.00 SOUTH TEXAS 1 06/10/08 19 0
0 0
0 0
0.00 SOUTH TEXAS 2 06/10/88 19 0
0 0
0 0
0.00 ST / *iCI E 1 01/14/88 29 0
23 0
0 0
0.00 ST.\\ 5CIE 2 01/14/88 29 0
26 0
0 0
0.00 SUKMER 1 10/06/87 14 0
2 0
0 0
0.00 SURRY 1 09/30/88 29 0
0 0
0 0
0.00 SURRY 2 09/30/88 29 0
0 0
0 0
0.00
Page:No.'
3 L D3/23/89 FINAL REPORT STATUS OF IEE 85-03 7
SUMMARY
EY FAILURE RATE AS OF g
MARCH 23, 1989 NUMBER HUMBER NUMBER TOTAL BYPASS TORQUE LIMIT NUMBER NUMBER PER CENT UTILITY NUMBER SVITCli SVITCH (SVITCH OTHER INOP INOP
. PLANT NAME REPLY VALVES CHANGED CHANCED CHANGED DEFECTS VALVES VALVES TM1 1 06/25/87 12 10 12 1
1 0
0.00 TURKEY POINT 3 01/14188 14 0
14 0
0 0
0.00 VOCTLE 1 09/04/87 50 0
0 0
32 0
0.00 V0GTLE 2 02/03/89 51 0
0 0
25 0
0.00 VATERFORD 3 12/21/87 20 4
19 0
16 0
0.00
- Total ***
2323 463 1197 200 553 196 Caution - A sero entry for the number of switch changes or for the number of other defects does not necessarily mean that none occurred.
It l
probably means that the licensee simply did not report any data on the topic.
To be sure, see the Comments Report for the facility.
For the 95 plants reporting to date:
O 7.96 is the average of the individual failure rates in percent, 15.3543 is the standard deviation of the individual failure rates'in percent.
8.44 is the average failure rate in per cent.
For the 42 plants reporting inoperable valves to date:
18.01 is the average of the individual failure rates in percent.
21.1899 is the standard deviation of the individual failure rates in percent.
17.13 is the average failure rate.
O(_/
0 EVALUATION OF THE P0TENTIAL EFFECTS OF A M0V PROTOTYPE TEST PROGRAM AS WELL AS INCREASED NRC EXPENDITURES ON THE VALUE-IMPACT ANALYSIS DESCRIBED IN NUREG/CR-5140 ACRS has asked RES/EIB to assess the potential impact of an industry-wide MOV 1
prototype testing program on our value-impact analysis that was presented in i
The value-impact analysis as described in NUREG/CR-5140, does not include consideration of such a test program.
Further, ACRS expressed a concern that NRC expenditures might be more than previot. sly anticipated because of the need for training and increased on-site surveillance.
The following analyses were performed in response to the above mentioned concerns. The sign conventions and definitions of "High," "Best," and " Low"
)
categories were used as in the NUREG report.
1.
Industry prototype testing program:
NRC/INEL cost to perform blowdown tests on two MOVs was approximately
$500,000 or $250,000 per M0V. This is considered to be a "high-end" figure because 'of its research orientation and the use of expert personnel. However, in the interest of estimating bounding impacts, a test cost of $250,000 per MOV was assumed as a "Best" estimate. The cost O
was assumed to vary in the range of $125,000 per M0V to as much as
$375,000 per M0V.
The number of MOVs tested under full scale blowdown or other design basis conditions was assumed to be approximately 100.
So far, INEL has tested two MOVs and the EPRI-Marshall test series involved blowdown tests on 7 MOVs. An order of magnitude increase in MOVs tested seemed reasonable.
l The number of M0Vs was then assumed to vary between 75 and.125 with a "Best" estimate of 100. Therefore the range of impact is:
Best High Low
-$250,000/MOV
-$125,000/MOV
-$375,000/MOV x 100 MOVs x 75 MOVs x 125 MOVs
-$ 25,000,000
-$ 9,375,000
-$ 46,875,000 say say
-$ 10,000,000
-$ 47,000,000 2.
All of NRC's costs were arbitrarily multiplied by a factor of 10 in order to estimate their potential impact on the overall analysis.
It is not suggested that the NRC's costs would be that large. Rather, an attempt was made to bound the NRC's costs and then assess the effect on the overall conclusion.
Q As is seen from the summary on sheet 2, the effect of the test program and increased NRC costs is small in comparison to the savings.
d5 g
g 9
7+ g /
(
E
[
4 T
6 4
3 6
6 7
8 9
8 8
9 A
h E
E E
E E
g E
E E
F F
E E
E E
M g
5 7
0 2
5 8
9 7
6 9
8 8
8 6
7 5
4 8
7 o,
4 7
2 9
0 5
2 I
i 1
1 1
1 T
H
(
S 7
6 6
7 7
2 4
1 8
5 1
5 4
2 E
u q7 rf r,
6 /s h
6 3
3 6
6 8
8 7
8 8
j t
E E
E E
E E
E E
E E
E E
E E
9, s
4 3
2 5
4 9
5 5
7 5
0 7
1 7
a 0
8 9
0 0
5 5
3 0
6 e
q 4
8 4
7 B
e 1
5 1
1 1
3 4
6 9
6 1
7 1
8 h
)
)
y d
s T
r e
e C
a d
t m
i u
o o
to AP m
o b
M g
l u
v i
g g
I S
a r
t D
t e
t E
c s
a I
a o
F p
d l
I m
(
a T
I t
N E
n A
e U
e U
L m
Q A
e
)
V l
L V
m p
A e
L p
T r
A u
D
)
T s
1 0
n
)
l O
)
1 o
l a
T t
s s
a t
u r
5 r
t n
o a
e n
e h
l
)
e p
e m
t l
l l
m e
i o
a b
n e
l w
D t
a i
l p
(
n T
E p
p n
e T
d p
u E
i
- n U
e u
S U
e B
d S
(
L g
l I
i
(
A n
p R
o
)
V i
p T
v
)
l v
u g
T a
l a
L a
S A
a n
A S
(
l e
t o
T
(
)
o s
n i
O l
)
t T.
o e
t T
S a
1 n
t.
d d
a T
t s
e
(
i r
C n
t m
b c
e A
e n
e S
c p
P m
e v
E A
O M
e m
o U
(
(
I n
l e
r L
o p
l p
A e
e i
p p
m V
r r
t u
p I
T u
u a
S u
s s
t n
n
(
S y
o o
n o
o
(
t T
p p
e i
i y
i "t
x x
m t
t t
t y
l E
E e
a n
a r
t i
V h
l r
e t
n e
r b
t t
l l
p e
m n
o p
e a
T l
a a
m p
p e
i o
p l
a n
n I
o o
m t
r o
i e
o o
l e
a P
r a
I i
i y
y e
l r
P v
I t
t r
r v
p e
e A
c a
a t
t e
m p
t e
E i
p p
s s
D I
O i
t t.
l u
u u
u s
i n
b c
c d
d C
C C
f s
a u
c c
n n
R R
R f
n l
P O
O i
I N
N N
O O
P
3
( }
As a further exercise, all of the potential savings due to improved industry operations were assumed to be eliminated. Taking the low estimate (worst case),
the $409,000,000 savings was assumed to be eliminated. This results in an overall industry cost of $112,000,000 rather than a savings of $297,000,000.
If this $112,000,000 cost is divided by the lowest estimate of person-rem saved, the result is a cost of $606 per person-rem. This calculation is shown on the summary on sheet 4.
Thus, it was concluded that the program will be cost ffective.
l
(
ff 57.
Owen Rothberg, Task Manager Engineering Issues Branch Division of Safety Issue Resolution Office of Nuclear Regulatory Research O
O
o.g 4
9 7
f(, /
E 3
T 6
4 3
6 6
7 8
9 8
8 9
A h
E E
E E
E E
E E
F F
E E
E E
M g
5 7
0 2
5 8
9 7
6 9
8 8
8 6
7 5
4 8
7 o,
4 7
2 9
0 5
r I
i 1
1 1
y T
i c
S 7
6 6
7 7
2 4
1 8
5 1
5 4
2 l
(
E u
/g 4
6 7/
h 6
3 3
6 6
7 8
/
8 7
8 8
t E
E E
E E
E E
E E
E E
E E
E 9,
s 4
3 2
5 4
9 5
5 7
5 0
7 1
7 g
0 8
9 0
0 e
q 5
5 3
0 6
7 4
8 4
B 1
5 1
1 1
3 4
6 9
6 1
7 1
8
{7,
)
)
y d
s T
r e
e CA a
d t
ggg m
i u
PM m
o b
u v
i g
g I
S a
r t
D t
e t
E c
s a
I a
o F
p d
l I
m
(
a T
I t
N E
n A
U e
U qf L
m Q
il A
e a
)
V l
L V
m p
A e
L p
T r
A u
O
)
T s
T 0
n
)
l O
)
1 o
l a
T t
s s
a t
u r
5 r
t n
o a
e n
e h
l
)
e p
e m
t l
l l
m e
i o
a b
n e
l w
D t
a i
l p
(
n T
E p
p n
e T
d p
u E
i m
U e
u S
U e
B d
S
(
L g
l I
i
(
A n
p R
o
)
V i
p T
v
)
l v
u g
T a
l a
L a
S A
a n
A S
(
}
e t
o T
(
)
e s
n i
O l
)
t T.
o e
t T
S a
l n
d d
a T
t a
e M
(
i r
C n
t m
c e
A e
n e
S c
p P
m e
v E
A O
M e
m o
U
(
(
I n
l e
r L
o p
l p
A e
e i
p p
m V
r r
t u
p I
u u
a S
u f
s s
t n
n
(
S y
T o
o n
o o
(
t p
p e
i i
y i
't x
x m
t t
t t
y l
E E
e a
n a
r t
i M
h l
r e
t n
e r
b t
l l
p e
m n
o p
e a
T l
a a
m p
p e
i o
p l
a n
n I
O o
m t
r o
i
\\
e o
o l
e a
P r
a l
i i
y y
e l
r P
v I
)
t t
r r
v p
e e
A L
c a
a t
t e
m p
t e
B i
p p
s s
D I
O i
t t
l u
u u
u s
i n
b c
c d
d C
C C
f s
a u
c c
n n
R R
R f
n l
P O
O I
I N
N N
o O
P
ew SNUS 3c A
CORPORATION Ify[a"L,ip=fTs""""
SDo 89j,30 e, p 5 1
e ua 7, u. %\\\\ a Mr. E. Igne ACRS 7920 Norfolk Avenue Bethesda, Maryland 20814
Dear Mr. Igne:
Per our phone conversation, I have attached my comments on the draft copy of NUREG/CR-5140. I hope that these comments would be of use to you in your work on the MOV generic letter issue. Please contact me if you have any questions on this matter.
Sincerely, b
Dennis W. Henneke Consulting Engineer DWH/jd i
Attachment O
/
e v
4 ATTACHMENT A review of NUREG/CR-5140 was performed by the ANPP Reliability and Risk Analysis Group to determine the adequacy its results.
Since a limited time was available, the review was not comprehensive. The review indicates that both the methodology and data used as the basis for this report are improper and the results are invalid.
The following is the justification for this conclusion.
Summarv 1)
BNL us'ed Draft NUREG-1150 results that have since been modified.
2)
BNL analysis was not performed to the same level of detail as the original analysis, especially regarding recovery analysis.
3)
The data used for MOV fails to open/close with design DP is probably an order of magnitude too high. This data is based upon unreviewed data that appears to have been interpreted very conservatively.
4)
Applying the derived failure rate to all MOVs in all sequences is inconsistent and inaccurate.
5)
The analysis performed for upper and lower bound core damage and person-rem estimates is inconsistent.
Method 1)
BNL used Draft results for NUREG-1150 that have since been revised and significantly modified. Although this does not fully invalidate the results used by BNL, the BNL results and dominant cutsets should be
. O reviewed, in detail, by the present NUREG-1150 analysis for accuracy and consistency.
Discussions with analysts for the Surry and Grand Gulf PRAs indicate that they were not contacted by BNL for input or review.
Per phone conversations, they agree that invalid methodology and data was used by BNL, as indicated in the comments below.
2)
By modifying the base failure rate for MOV fails to open/close, BNL has brought to the top, cutsets that were previously not important.
If, in the original analysis, these cutsets were in the dominant cutset list, a
recovery analysis would have been performed on them. For example, credit on dominant cutsets.
for manual operation of MOVs would have considered NUREG-1150 ' typically credits a 0.3 human error probability for performing this action if sufficient time and accessibility were available.
BNL does not give 'any credit for this recovery or many other possible recoveries.
the original 3)
Additionally, conservative assumptions were taken throughout NUREG-1150 analysis, and many of those assumptions were left in and not recovered since the affected cutsets did not show up in the top cutsets.
If sequences were found to be dominant, more effort on recovery would have been considered. One example was the requirement for high pressure recirculation during a small LOCA.
If this sequence was shown to be dominant, recovery by depressuritation and use of LPSI recirculation would have been considered.
When BNL modified the failure races, this sequence becamc a major contributor.
If BNL had completed the analysis
(_,
correctly, they would have looked at recovery and the overall effect due to this sequence would have been significantly reduced.
1
~,,
D ATTACRMENT g
(Continued)
V 4)
BNL applied the MOV failure to open/close under high DP to all MOVs in l
the PRA results used.
Of the major three valves that impact the PWR results (as listed in 5.1.4.1.3 of the report), one valve (RUST suction valve) does not see any significant pressure during stroking, and tha other two (LPSI to HPSI cross tie and sump suction valves) stroke with a low DP present.
This is contrary to the intentions of the authors who indicate on page 5-9 of the report, that this failure rate is only valid when an abnormally high DP is present.
It is incorrect to use the derived MOV trilure rate for all MOVs in the PRA. Prope'r analysis would look at the conditions affecting each MOV in the analysis and select only those MOVs that actually have an abnormally high or design DP present.
Valves with a lower DP present would probably have a significantly lower failure rate than those with a high DP present. Valves with very little DP present should not have their rates modified from the original NUREG-il50 values.
Data 1)
There is insufficient justification for the derived rate for MOV failure to open/ close on demand with high DP present. There have been numerous studies on MOV failures and none of these studies indicates any failure rate of the value presented. These studies included MOV demands where design DP was present during stroking (see 2 below).
" Data Summaries of LERs of Valves at U.S.
Commercial Nuclear Power
/
Plants" was a major study that looked at MOV failures.
This study
\\. ')
yielded no generic failures even near the magnitude that is implied by BNL.
2)
The base failure rate as used in NUREG-1150 had an unoer bound of 1E 02.
The failure rate presented by BNL is 8.7E-02.
If this failure rate were true, approximately one in ten valves opened with design DP across it would fail to open.
Many of the surveillance tests and startup tests performed at PVNGS and at other plants require MOVs to function with design DP across them (Ref. Reg. Guide 1.68, Section C3). No data has surfaced to indicate failures even in the same order of magnitude.
Any failure rate used in this study should be based on actual plant data / testing.
3)
The basis for the MOV failure rate is given on page 5-8 of the NUREG.
The "MOV fails to open" presented is based on MOVATS test data as shown in Appendix B.
This MOVATS data, as stated on page 5-7 of the NUREG, is unreviewed data.
The demand failure calculated is the product of torque switch bypass incorrectly set and torque switch set too low.
The interpretation of the dcta by BNL appears to be that all of these failures, as reported by MOVATS, are fatal failures.
In fact, the torque switch and the torque switch bypass both are set with conservative design margin.
A significant proportion of each of these failures would l
probably only result in a decrease in the design margin and not failure of the valves to stroke.
/N I
sl l
1 1 l
-~
L.., 4
?
s k
ATTACHMENT (Continued)
'4 )
Additionally, BNL interpreted the MOVATs data, reported as torque switch (TS) bypass incorrectly
- set, as TS bypass set too low. No portion was interpreted as TS set too high. This yields significantly conservative results for MOV rates used.
5)
The Bants for high and low estimates for the MOV failure rate used are insufficient. The low estimate failure rate used by. BNL is< still above the high (95%) estimate presented in the NUREG-11,50 analysis. A more accurate low estimate of dose savings would use the original NUREC-1150 failure rate and would yield zero person-rem saved. The inaccuracy yields incorrect (high) core damage estimates.
Additional Comments 1)
BNL recommends all passive valves be tested along with the active valves, due to the referenced Davis-Besse failure being on a supposed passive valve. Actual savings for testing passive valves is minimal.. Testing of passive valves, fr. most instances, is unwarranted.
2)
The Dose Analysis used by BNL was not completely reviewed by ANPP, but some inconsistencies seem evident, especially in regards to upper and lower bound calculations.
The error bounds presented in many of the published PRAs appear to be much greater than BNL reports.
This area should be reviewed further, possibly by the NUREG-1150 authors.
l O
\\
3 __
-m
,..{,. ; -
4 l
i e
f y
Arizona Nuclear Power Project P O Box 52034 e PHOENIX. ARIZONA 85072 2034 j
1 1
1 l
167-02957-CRS/DWH
]
October 26, 1988 f
l l
1 Mr. Tom Tipton
)
NUMARC 1776 Eye Street, N.W.
Suite-300 Washington, D.C.
20006-2496 Dpar. Mr. Tipton:
Subject:
NUREG/CR-5140 Advanced Copy Review File:
88-163-419 Per your request, the ANPP Reliability and Risk Analysis Group has performed a review of the advanced copy of NUREC/CR-5140. This NUREG, as you know, is being used, in part, as the basis for expanding the requirements of IEB 85-03 to all safety-related MOVs.
Although the. review was not a comprehensive effort, the review comments (attached) bring out many questions re5arding the validity of the results of this NUREG.
If you require any further assistance in this area, please contact Dennis W. Henneke or C. R. Stevens at (602) 371-4608.
Very truly yours, l
E. C. Sterling l
Manager, Nuclear Engineering ECS/DWH/dlm Attachments cc:
C. R. Stevens D. W. Henneke R. G. Kershaw J. E. Allen j
D. B. Karner l
W. F. Quinn l
A. C. Rogers l
O)
I v
i I
I J
~
NUCLEAR UTILITY BACKFITTING AND REFORM GROUP i
SUITE 800 i
1400 L STREET. N. W.
WASHINGTON, D. C. 20005 3502 TELEPHONE (202) 374 5700 March 28, 1989 Mr. Carlyle Michelson U.S.
Nuclear Regulatory Commission Advisory Com:nittee On Reactor Safeguards, Mechanical' Components Subcommittee Chairman Washington, D.C.
20555 subj Nuclear Utility Backfitting and Reform Group (NCBARG) Comments On NRC Draft Generic Letter, " safety Related Motor Operated Valve Testina and surveillanceu
Dear Mr. Michelson:
As you may be aware, the Nuclear Utility Backfitting and Reform Group (NUBARG) is an active industry organization which monitors the NRC's implementation of the backfit rule (10 C.F.R. 5 50.109).1/
We do not consider NUBARG an adversary to the Staff in that we have supported the backfit rule as evidenced by our filing of court briefs supporting Staff positions regarding ongoing proceedings challenging the rule.
However, we have also questioned Staff actions whenever we believed that the'backfit rule was not being correctly applied.
Consistent with the above, NUDARG has been monitoring the progression of the NRC Gtaff's draft generic letter on motor-operated valve (MOV) surveillance and testing.
We have attended all ACRS full and subcommittee meetings that have addressed this issue.
Most recently, we attended the February 28, 1989 ACRS Mechanical Components subcommittee briefing and offer several thoughts regarding your subcommittee discussions.
In brief, we agree with your concerns and applaud your willingness to challenge the accuracy of the Staff's value-impact analysis.
NUBARG has also been concerned with the analysis and would concur with the rejection of the generic letter pending the Staff's 0
2/
uusino co 9riee 27 nac 11ceo ee -
Mr. Carlyle Michalcon March 28, 1989 Page 2 preparation of a more accurate and realistic value-impact estimate.
I.
The Draft Generic Letter May Involve a Backfit After review of the draft generic letter, NUBARG concludes that the Staff appears to be requiring testing beyond previously accepted testing criteria (ASME Section XI).
Therefore, its actions may be considered a backfit.
The backfit rule states, "Backfitting is defined as a modification of or addition to.
the procedures or organization required to design, construct, or operate a facility which may result from a new or amended provision in the Commission rules or the imposition of a r2gulatory Staff position interpreting the Commission rules that is either new or different from a previously acolicable Staff oosition."
(emphasis added)
We believe that the Staff has modified its position concerning actions necessary to satisfy regulatory requirements regarding the operability of motor operated valves.
The Staff
- states,
...it is now recoanized that the Sec_ tion XI testing m
alone is not sufficient to provide assurance of MOV operability."
i (emphasis added).2/
It is our understanding that ASME Section XI l
had previously been considered by the Staff to be an acceptable method of ensuring compliance with 10 C.F.R. Part 50, Appendix A, General Design Criteria 1, 4,
18, 21 and 10 C.F.R. Part 50, Appendix B, Criteria III, XI, and XVI.
However, the draft i
generic letter appears to alter that acceptance.
l The draft generic letter contains several additional discussions that have the potential for being a backfit issue.
First, we have noted that the generic letter requests information on misadjustments.
This item could become a backfit issue in that such an analysis is not quantifiable and may result in design basis assumptions being modified.
Second, we note that the draft generic letter states that "...some MOVs may be subjected to mechanisms and loads that were not previously accounted for.
INEL's preliminary conclusions indicate that industry sizing equations [likely approved by the NRC] may not be conservative for all design basis conditions."
NUBARG believes that if the NRC withdraws its approval of MOV load-related equations that were either explicitly or implicitly approved by the NRC, it could be considered a backfit.
f3 2/
NRC Memorandum to Edward L. Jordan, CRGR Chairman from R.
(,)
Wayne Houston, Director, Division of Safety Issue Resolution,
" Final Draft Generic Letter - Safety Related Motor Operated Valve Testing and Surveillance", March 9, 1989, at p.
2.
i Mr. Carlylo Micholson March 28, 1989 Page 3 A backfit analysis is not required when the Staff finds and declares (with a documented evaluation supporting its finding) either that a modification is necessary to bring the facility into compliance with a license, commission rules, or orders, or that an immediately effective regulatory action is necessary to ensure that the facility continues to provide adequate protection of the public health and safety.
10 C.F.R. 5 50.109 (a) (4).
However, regarding the subject draft generic letter, we have not found a documented Staff evaluation satisfying this requirement and exempting the Staff from performing a full backfit analysis.
Even though the Staff has stated that it does not believe this action constitutes a backfit, it has performed a detailed value-impact analysis (consistent with the requirements of 10 C.F.R.
f 50.109(3) discussing the estimated monetary gain / loss resulting from the generic letter.2/
Notwithstanding our backfit concerns, NUBARG generally agrees with the Staff regarding the need for licensees to ensure that safety-related MOVs are capable of performing intended design functions.
Therefore, we have not challenged the draft generic letter as a backfit.
However, our agreement with the generic letter is premised on an accurate Staff assessment of the y'
time and cost necessary to implement its programs and on the
\\
development of a realistic schedule for complying with the generic letter.
We further discuss below, our concerns regarding this issue.
II.
The Staff Has The Responsibility To Ensure that Its value-Impact Analysis Accurately Reflects Monetary and Manpower Expenditures I
The Staff (through Brookhaven National Laboratory in NUREG/CR 5410) has estimated that implementation of the generic letter (not including subsequent MOV testing) would initially cost between 195 and 414 thousand dollars per reactor (24.7 to 52.5 million dollars total for the industry).
Moreover, the Staff estimates that the remaining plant-lifetime savinas per reactor will be 3.28 to 6.67 million dollars per plant (417 million to 847 million dollars total for industry).4/
During the February 28, 1989 Subcommittee meating, the Staff acknowledged that its required man-hour estimate was erroneous by a factor of i
10.
Accordingly, the Staff has increased the estimate from 200 to 2,000 man-hours per licensee response.
The additional hours 2/
NUREG/CR-5140, "Value-impact Analysis For Extension of NRC Bulletin 85-03 to Cover All Safety-Related MOVs," July 1988.
gsO 4/
Analyses appaar to have generally used 127 reactors as a baseline number for calculating industry effects.
Mr. Carlyle Micholcon l
March 28, 1989 Page 4 include assessment of new recommendations, searching data.
sources, gathering and analyzing data and preparing required letters.
However, the increased man-hours do not include additional time for implementation of the requested actions.
Assessment of the costs involved in implementing requested
)
actions is a major area where we believe the Staff has erred in its value-impact estimate.
Moreover, the 1800 additional man-hours for analysis and document preparation will surely result in
)
additional testing and perhaps installation modifications.
All i
of the above factors should be reconsidered.
On November 18, 1988, the full ACRS discussed the industry cost issue.
At that meeting, NUMARC representatives (speaking on behalf of industry) estimated that the initial cost (to industry) for implementing the generic letter would be 570 million dollars (4.49 million dollars per reactor).
NUMARC also estimated that the total remaining-life cost to industry would be 3.3 billion dollars (25.99 million dollars per reactor).5/
These significant Staff discrepancies and errors lead us to believe that the Staff has not conducted an adequate value impact analysis.
Moreover, the Staff's consideration in this case of
" speculative cost savings" could set a precedent that effectively
/'a reads the cost-justification standard out of 10 C.F.R. 5
(-)
50.109 ( a) ( 3 ) of the backfit rule.
Since you have expressed concern regarding the disparity between industry and NRC estimates (the difference between the two exceeding 2.5 billion dollars) we are confident that you understand NUBARG's position.
We urge the ACRS to continue to request the NRC to justify its value-impact assumptions and conclusions regarding this matter.
III. Conclusion It appears that the Staff's value-impact estimates may be substantially in error.
The backfit rule will be ineffective if the Staff is not required to accurately characterize issues as 5/
Inside N.R.C.
quoted a Davis Besse Senior Nuclear Specialist who stated that previous testing of 165 MOVs at Davis Besse cost at least 2.5 million dollars -- inferring that a full program would cost significantly more than the Staff's estimate.
This actual example is more consistent with NUMARC m) estimates and far exceeds NRC value-impact conclusions.
Inside N.R.C.
Vol. 10, No. 24, November 21, 1988 at 2.
See also Attachment 1 to this memorandum.
l
Mr. Ccrlylo Michalcon March 28, 1989 Page 5 backfits when appropriate and prepare realistic and accurate value-impact analyses (pursuant to the backfit rule) to justify its actions.
We would therefore concur with rejection of the draft MOV generic letter by the ACRS pending a more realistic estimate of required licensee resources and subsequent revision of the Staff's proposed implementation schedule.
Sinc 1
1 1
Nichol S
Reynolds Daniel tenger Counsel tc he Nuclear Utility Ba fitting and Reform Group
(
Attachment cc:
Victor Stello, Jr. (NRC)
Dr. Thomas E. Murley (NRC)
Edward L. Jordan (NRC) l O
G f
Mr. Carlylo Michsleon M2rch 28, 1989 Page 6 Comparison of NRC v. EPRI value-impact Estimates ($)
INITIAL (COST)/ SAVINGS TOTAL (COST)/ SAVINGS Per Industry Per Industry Reactor Reactor NRC
($195-414
($24.8-52.6
$3.3-6.7
$417-847 thousand) million) million million EPRI
($4.5
($570
($26
($3.3 million) million) million) billion)
Diff.
$4.3-4.08
$545-517
$22.7-19.3
$2.8-2.45 NRC/EPRI million million million billion O
I
.