ML20138B118
| ML20138B118 | |
| Person / Time | |
|---|---|
| Issue date: | 10/08/1985 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1453, NUDOCS 8510150092 | |
| Download: ML20138B118 (30) | |
Text
,
OR GWAL - MN/V3 m
O UNITED STATES NUCLEAR REGULATORY COMMISSION IN THE MATTER OF: DOCKET NO:
ADVISORY COSD1ITTEE ON REACTOR SAFEGUARDS SUBCOFDiITTEE ON RELIABILITY ASSURANCE (VALVES)
O LOCATION: WASHINGTON, D. C. .
PAGES:1 - 300 DATE: TUESDAY, OCTOBER 8, 1985
,,r 4 Q @ ilk [ ,I ,," .
{
sad -
r 9
)() Not 6310% "'"
g' N s AG-FEDERAL REPORTERS, INC.
b\ Official Reporters 444 North Capitol Street Washington, D.C. '.0001 gO1 2 851008 (202) 347-3700 T-1453 PDR NATIONWTDE COVERAGE
3 CR2'4 59'4 .0 1 l CRT/dnw.
1 UNITED STATES OF AMERICA
~2 NUCLEAR REGULATORY COMMISSION 3 ADVISORY COMMITTEE ON REACTOR SAFEGURADS 4 SUBCOMMITTEE ON RELIABILITY ASSURANCE (VALVES) 5 Nuclear Regulatory Commission 6 Room 1167 1717 H Street, N.W.
7 Washington, D. C.
8 Tuesday, October 8, .19 85 9 The subcommittee met at 8:30 a.m. , Mr. Carlyle Michelson 10 presiding.
11 i PRESENT:
12 MR. JESSE C. EBERSOLE
() 13 MR..GLENN A. REED MR. DAVID A. WARD 14 MR. CARLYLE MIC$ELSON DR. WILLIAM KERR-15 ,
16 MR. HENRY JONES, ACRS Consultant l 17 ,
i 18
.19 20 21 22 23 noens,e, i .
25
v.
t
() PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSIONERS' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS TUESDAY, OCTOBER 8, 1985 The contents of this stenographic transcript of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards (ACRS), as reported herein, is an uncorrected record of the discussions recorded at the meeting held on the above date.
No member of the ACRS Staff and no participant at.
} this meeting accepts any responsibility for errors or
~
inaccuracies of statement or data contained in this transcript.
4 i
t J
r- >
24594.0 2
.BRT r~ l (j 1 PROCEEDINGS 2 MR. MICHELSON: The meeting will come to order.
3 This i:s a meeting of the ACRS Subcommittee on Reliability 4 Assurance. I'm Carl Michelson, Subcommittee chairman of 5 this meeting. The other members in attendance are Jesse 6 Ebersole, Glen Reed, and David Ward. We have an ACRS 7 consultant, Henry Jones, with us also. The Subcommittee 8 will continue discussions of valve reliability. A risk 9 perspective on valve performance will be sought. Also to 10 be studied is the importance of valves from a safety 11 standpoint.
12 Richard Major is the ACRS staff member for this 13 meeting.
O 14 The rules for participation in today's meeting 15 have been announced as part of the notice of this meeting 16- that was published in the Federal Register on September 24, 17 1985.
18 It is requested that each speaker first identify 19- himself or herself and speak with sufficient clarity and 20 volume so that he or she can be readily heard.
21 We have received no written comments or requests 22 for time to make oral statements from members of the public.
23 We will now proceed with the meeting.
24 I wanted to mention now that we have had several 25 of these meetings, I think after this meeting it would be t-ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee 800 336 4646
-24594.0 3 BRT.
- g,p) . 1 appropriate to start to put together some thoughts about 2 'what'we.have learned thus far,-:and where we should perhaps
- 3 'goffrom"here in terms of recommendations for further 4~ meetings,.further action or whatever.
5 We really have focused, now, only or at least 6- primarily on motor-operated valves. I would still like to
.7 retain that focus, simply to narrow the subject of 8 discussion. There are many, many interesting problems 9 associated with air-operated 'ralves, hydraulic-operated 10 valves, pressure-operated valves, et cetera, which we may 11 or may not want to get into later, but for the moment we 12 will focus on the motor-operated valves.
13 At the end of the day we should kick around what,
-O 14 if anything more, we would like to hear and then a best 15 . course of action.
16 I was -- I will'suggest that I tried to put -
17 together some type of a document describing very briefly 18 what we have learned to date, or what we have done and what 19- we have learned to date, and'I propose some - propose some 20 recommendations for full committee ' recommendation.
21 What. I'll try to do is put together this 22- document'and circulate it to the Subcommittee'and get all 23 .your thoughts together so we have a subcommittee consensus 24 'to send forward to the full committee on what we think 25 should be done.
i O
l .
I
- ACE. FEDERAL REPORTERS, INC.
i : 202 347 3700 Nationwide Coverage 800 33Ml646
(24594.0 4 BRT
() 1 With that in mind, are there any discussion or 2 other comments?
3 MR. EBERSOLE: Carl, do you intend to go into 4 the other questions of valves in a more or less independent 5 fashion? There are a number of critical valves which are 6 'not motor-operated.
7 MR. MICHELSON: I think motor-operated valves 8 will giveaus a flavor for the kind of difficulties that one 9 might have, so it's a question of does the Subcommittee 10 really want to pursue it further for other types of valves --
11 that's part of the Subcommittee decision and recommendation 12 to the full committee.
13 MR. EBERSOLE: For instance, the PORV are oddball O 14 valves, not in the red class, and the hot solenoid valves, 15 the boilers -- have no clear mode of safety. They can get 16 you in trouble open or shut. So some sort of motion of the 17 reliability of these should be associated with how you work 18 them into full valve grids or whatever, and some sort of 19 standard practice.
20 MR. MICHELSON: Yes. If we want to look into 21 PORVs from the mechanical viewpoint, we can.
22 MR. REED: It seems to me there may be two 23 plateaus in this whole valve consideration. Since we have 24 been using the word'since the last meeting, " cosmic," it 25 seems to me there's a cosmic plateau with respect to valves n
U ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 80(k336 6M6
r 24594.0 5 BRT g
t_/ 1 and then there's a mechanical detail type.
2 I have to believe if we stay oriented on 3 motor-operated valves, we are talking mechanical detail 4 plateau. If we want to get cosmic, which I think is going 5 to be the longer term and the more subtle issue, I think we 6 have to talk the utilization or application of certain 7 valves for certain duties and certain environments, like 8 the PORVs and the borated, hydrogenated environment.
9 It seems to me that the ACRS, in their judgments, 10 ought to focus on the cosmic plateau. I think I go along 11 with what Jesse is hinting at here. If I was looking for 12 cosmic plateau I would be looking for PORVs because they
,- s 13 are important to the depressurization process.
! )
~~
14 MR. MICHELSON: Think what one has to decide, of 15 course, how to approach the problem. The motor-operated 16 valve is also very important to K heat removal, as evidence 17 that Davis-Besse, for instance, had both motor-operated 18 valve problems and PORV problems.
19 MR. REED: We know motor-operated valves are 20 going to be around. It's the mechanical performance, the 21 mechanical plateau that's the problem there. I'm wondering 22 if that's ACRS' duty. Maybe that's the duty of somebody 23 that's getting it in the butt with respect to either the 24 maintenance or the manufacturing or those kind of 25 mechanical little gear and pin-type things.
U,a ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverase 800 336-6646
7 24594.0. 6 BRT
,~
j 1 MR. EBERSOLE: I think I could even extend it to 2 the main steam stock valve on turbines. There's two modes 3 of those. One performs an automatic sectionalization 4 process, ones that are not commonly manifolded, and that 5 gi"es you a valuable asset to stop steam flow out of both 6 steam generators over and above those you get with the 7 isolation valve. So that's a standardization involved, 8 should we have main steam stop switch standardized in the 9 system.
10 MR. MICHELSON: We'll think about it through the 11 day because by the end of the day I think we should decide 12 what we want to delve into in the future and at what level 7- 13 do we want to delve into it, you know. We have looked at
('~' /
14 these valves, here, rather in some depth, simply to try to 15 understand what-all is involved in the problem. But I 16 don't think we should continue to look at that kind of depth, 17 or we don't have the time, you know, on many more items.
18 With that in mind, let's proceed .iith our agenda.
19 The first speaker is going to give us some 20 perspective on the risk contribution. This is a request of 21 the Subcommittee from our last meeting. We would like to 22 hear a little more about what is the level of risk involved --
23 to what extent do valves contribute to risk.
24 We have with us this morning Dale Rasmuson, of 25 Research, I believe it is. He's going to -- he's taking
,/
t' ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
24594.0- ,
7 ERT-
[t 1 Mr. Burdick's place, who was not able to be here this 2 morning. So, Dale, could you give us a few minutes and 3 introduce your speaker?
4 MR. RASMUSON: Thank you. Last night my 5 management came to me and cornered me and said could you 6 come down, Dr. Burdick is ill. So I know he was preparing 7 some remarks. I know he would like to address, share those 8 with you sometime.
9 I think it is important when we talk -- start 10 thinking about PRAs, that we don't necessarily think all 11 the time of bottom line numbers.
12 I think when I talk about PRAs, I think of a 13 whole bag of techniques that I have available, and we can O 14 learn a great dea) of information from qualitative analyses 15 as well as the risk -- the quantitative analyses, where we I 16 start putting probabilities with these things.
17 Let me just share a brief analysis that we had, 18 when-I was at the Idaho National Engineering, that we did 19 in conjunction with Duke Power Company. We were developing 20 methods for DOE, at this time, dependent failure analysis 21 methods, common cause failures, if you will, and we took 22 the auxiliary feed water system of Maguire and did a 23 qualitative common cause failure analysis. We did not 24 assign any probabilities to this. But what we were looking 25 at, we developed a bulk tree down to the component level O
ACE-FEDERAL REPORTERS, IbC.
102 347 3700 Nationwide Coversee 80)336 4646
a
~ 24594.0 8 BRT s
l 1 and then we assigned susceptibilities to the various 2 components, such as: are they susceptible to moisture or 3 to grit or to other types of environments, or even common 4 links, things that would link them such as maintenance 5 practices, maintenance procedures.
6 And usi.ng the techniques we developed, we were 7 able to identify critical areas, and critical sets of 8 components.
9 Since we were talking about valves, we were 10 looking at -- there was a group of valves we were looking 11 at, there was like about 50 valves that were manufactured 12 by the same manufacturer at this time and we started
,y 13 looking at them.
14 There were all the types, and we were 15 experimenting with them. But we found that out of 2000 cut 16 sets, we could concentrate on four valves that, if we 17 concentrated on these four valves, we could sort of focus 18 in on the problem.
19 Well, as you started looking at it, two of the 20 valves were locked open, so you could say, well, we want to 21 make sure that they are in a locked open position; one 22 valve was a check valve and the other valve was a 23 motor-operated valve. And, so, out of 50 valves we had 24 come down to one crucial valve and this was without any 25 probabilities.
s-ACE. FEDERAL REPORTERS, INC.
202 147-3700 Nationwide Coversee 800 336 6646
~
24594.0 9 CRT
()) 1 Probabilities can help us also to prioritize our 2 activities. Currently, in Research, I am the program 3 manager for some methods that -- for dependent failure 4 analysis where we are actually going to be using this on i 5 LaSalle, we-have what we call the risk methods integration 6 and evaluation program, and they, these methods, will be 7 utilized there.
8 And again, I think it's the qualitative insights -
9 that we can gain from these things, not necessarily the 10 bottom line numbers, that are the most important things 11 that we can have.
12 Along with these programs we have a root cause 13 -program that we are concentrating on. We are cooperating O 14 with Chip Vora, there. As we start looking at root causes, 15 in the aging program, we find that they are converging 16 together.
17 We have a reliability program, reliability 4
l 18 assurance program, which Dr. Burdick was going to discuss, 19 some of the details of that, with the Subcommittee, and I 20 know he wants to come back and talk about that.
21 Closely related to that is the tech spec program.
22 But one of the things that we find is very useful is 23 calculating component importance, in PRAs.
24 There are many different importance measures, i
25 and we have asked David Campbell, from JBF Associates, in O
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide coversee 800 3 4 4646
24594.0 10 BRT I
s i s ) 1 Knoxville, to share with you some insights that they have l 2 gained from the ANO-1 IREP study.
3 In this particular study, common cause failure 4 analysis was not addressed in very much detail, only for 5 very selected components such as common mode failures of 6 the batteries. One other activity that I am very 7 interested in right now, very excited about, is a joint 8 effort we have going on with EPRI with actually the 9 quantification of dependent failures. Carl Fleming of PL&G 10 is the principal investigator for this particular EPRI 11 contract. We are working closely with them to develop a 12 guide that reflects both EPRI's and NRC's methods. So I 13 think we will have some very interesting things coming out 7~
\'~' l -
14 in the near future on some of these things.
15 I'd like to now introduce David Campbell, and 16 his topic is valve. Importances from the ANO-1 IREP.
17 MR. CAMPBELL: Thank you. I'm David Campbell 18 with JBF Associates. I have some slides here. I guess I 19 can work just from the handouts.
20 The topic of my presentation is valve 21 importances from the ANO-1 IREP.
22 As part of the work we are doing under one of 23 the dependent failure analysis records programs that Dale 24 mentioned and also under a program to take the results of 25 PRAs and translate them into a format that's useful for INE ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 80433 H 646
L24594.0 11 BRT n
(_)' 1 decisionmaking, we have developed some software for 2 calculating the importance of various component types 3 within nuclear power plant PRAs, and also for -- we have --
4 the same' methods can be use'd to calculate the importance of 5 various root causes of failures in PRAs.
6 (Slide.)
7 I'll be presenting some results, from using the 8 software on the Arkansas nuclear 1 IREP.
9 (Slide.)
10 A little bit of background first. We used a ,
7 11~ reduced core melt frequency model from the IRPE study. We 12 took the dominant accident sequences that represent about 13 80 percent of the total core melt frequency and the results
,f-)s (m
14 you are going to see are based on this reduced model. It's j'
15 not the complete model of the plant.
I 16 Again, this is based on the IREP study of ANO.
17 We have broad definitions of valve types. We have broken 18 it down into five different types of valves motor-operated 19 valves, air-operated valves, relief valves, check valves, 20 and manual valves.
21 You can certainly make a finer distinction or 22 finer resolution in terms of valve types, and you might get 23 some different insights.
24 Also, I think it's important to keep in mind 25 that we have not considered root causes of failure in this.
O ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Cowrese 800 336 6646
24594.0 12 BRT L
(~'\ 1 I'think it's probably -- it's important to consider root
( ,/
2 causes of failure just in defining your different component 3 groups. It would make sense to me to define a group of 4 components such as they do share the same root causes of 5 failure in about the-same fractions. That would imply to 6 me that they really are legitimately a group of components, 7 and reliability assurance efforts that-apply to one would 8 apply to the group as a whole.
9 (Slide.)
10 MR. EBERSOLE: Is your study based on single 11 failures?
12 MR. CAMPBELL: We took the results from the ANO 13 PRA, just directly. They did not do any kind of dependent 14 failure analysis, for the most part. It tends to be an 15 analysis of individual random failures, for the most part.
16 (Slide.)
17 I have presented these results in the slides for 18 two different importance measures. I'm going to define 19 both. The first is Fussell-Vesely importance, the fraction 20 of the core melt frequency that involves failure of any 21 component in the specified group.
22 So, what we are really doing, we are looking at 23 cut sets in these dominant.c.ccident sequences and if, for 24 example, we are looking at motor-operated valves in the 25 high pressure injection system, we'll take any cut set that O
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Covereer 800 336 4 646
7 24594.0 13 BRT
_j 1 contains one of those valves and we'll count that into the 2 frequency that we are using to calculate this importance.
3 This portion is a fraction of the total core melt frequency 4 that involves those failures.
5 The other measure of importance is Birnbaum, 6 it's the sensitivity of the core melt frequency to changes 7 in the failure probability of components in the specified 8 group. This is more like a partial derivative of the core 9 melt frequency with respect to the failure probability of 10 the valves in a group.
11 One conclusion I have been able to reach after 12 taking a look at these results a little more closely is
,s 13 that both of these importance measures rank the different t )
'~'
14 valve types, and valve types within the different systems, 15 about the same in almost every case. So if you are 16 confused by either one of these definitions of importance, 17 but you have a good feel for the other one, you can just 18 look at that and ignore the other one and they are going to 19 rank things the same.
20 MR. EBERSOLE: Could you make the same statement 21 about valves versus pumps?
22 MR. CAMPBELL: I really don't know about that.
23 I haven't looked at pumps.
24 (Slide.)
25 Taking a look first at motor-operated valves, in p
\m ACE. FEDERAL REPORTERS, INC, 202 347 3700 Nationwide Coversee m336-6M6
F 24594.0 14 ERT 1 looking across the different valve types it is evident that 2 motor-operated valves are the most important. That comes 3 up in every system except for the reactor coolant system.
4 The only reactant coolant system valves that 5 were modeled in this IREP were the pressurizer relief 6 valves. It turns out we'll see in later slides the PRVs 7 are the cost important valves in the whole plant. But for 8 the rest of the systems, motor-operated valves tend to be 9 most important. So I think these results back up what was 10 said earlier about the importance of the various valve 11 types. In fact, PORVs and MOVs are the most important 12 valve types in the plant.
,- s, 13 In the dif ferent systems, well, again, the --
'~
14 it's hard to put a real strict interpretation on the 15 absolute value of the Birnbaum importance. They are used 16 for ranking things relative to each other. Fussell-Vesely 17 importance can contribute to a fractional addition to core 18 melt frequency.
19 For example, this 5 E-2, what that means is 5 20 percent of the cut sets or 5 percent of the contribution to 21 core melt frequency is cut sets that include failures of 22 motor-operated valves in the emergency feedwater system.
23 MR. WARD: That's all motor-operated valves in 24 the emergency feedwater?
25 MR. CAMPBELL: That's right.
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverese 800 336 6646
I
,24594.0 15 BRT
(~N T ,!. 1. MR.. ITARD: Not each individual?
2 MR. CAMPBELL: That's right. They are all 3 together here.
4 I suspect, I'd have to do a little more checking 5- to verify this, but it will tend to be one or two critical 6 valves in each system that are predominating this. The 7 high pressure injection, the low pressure injection, these 8 two numbers are the same and it's really one valve, it's 9 the borate water storage tank outlet valve that's dominant ;
10 in this contribution there.
'll MR. EBERSOLE: So it's 5 percent?
12 -MR. CAMPBELL: Yes.
13 One thing to be careful of, it's not appropriate O. 14 to add these numbers and call that a total contribution for 15 motor-operated valves. You'd be doing some double counting 16 there. That would be an upper bound. And probably a very 17 conservative upper bound on the total contribution of 18 motor-operated valves.
19 MR. EBERSOLE: That's the valves in the 20 -emergency system, like the ECCS systems right?
21 MR. CAMPBELL: That's right.
22 MR. EBERSOLE: Those systems are not on duty 23 except, hopefully, in very rare instances and this 24 sensitizes the valves significantly. I look down at 25 service water where valves are working all the time and I O
ACE FEDERAL REPORTERS, INC.
202 147 3700 Nation.We Covweee 8001364M6
24594.0 16 ERT 1 see 8 percent. Does that reflect the fact that I have an 2 active system running all the time? Do you know what I'm 3 trying to do, I'm trying to cut apart the significance of 4 the system, because it reflects the significance,of the 5
valves. That gets me into trouble because that doesn't 6 give me --
7 MR. CAMPBELL: You are absolutely right. The 8 significance of the system is an important consideration 9 here.
10 MR. EBERSOLE: But the valve, you might have a 11 good valve in a bad system or vice versa.
12 MR. CAMPBELL: That's right. What you said
,-s 13 about service water is correct, but I wouldn't claim that 14 the IREP models things in enough detail to be able to spot 15 the difference between an operating system and a standby 16 system. .
17 MR. EBERSOLE: Then it's not worth very much.
18 MR. CAMPBELL: You have to take a look at the 19 boundary conditions put on a PRA in the beginning of an 20 accident. The system is in some configuration, and it 21 needs to either stay or change to a configuration 22 appropriate to mitigate the accident. From that point of 23 view the service water system doesn't look too much 24 different from others.
25 Certainly the components may tend to wear out n
U ACE.FEDERAI, REPORTERS, INC.
202 347 1700 NationwWe Coversee 80rk)) m 44
24594.0 17 BRT 1 since they are being used --
2 MR. EBERSOLE: Isn't an accident loss of the 3 service water system --
4 MR. CAMPBELL: That is one potential initiating 5 event but it is fairly infrequent and I'm not sure it shows 6 up in the dominant sequences that we used in these 7 calculations.
8 MR. WARD: Are these valve failures treated as 9 initiating? I mean, are some of the initiating events the 10 valve failures?
11 MR. CAMPBELL: Only -- well, in this caso it 12 could be valve failures in the service water system is an 13 initiating event. The only other initiating event is, I (s) 14 guess would be pressurizer relief valve opening. That's 15 not on this slide. That's on the slide for relief valves 16 but I think those would be the only accident initiating 17 events.
18 MR. MICHELSON: I guess it's correct to assume 19 from looking at this chart that as far as motor-operated 20 valves are concerned, they are not a very high contributor 21 to risk?
22 MR. CAMPBELL: Well, again, wo have broken this 23 down into systems. Liko I mentioned, if you added those 24 contributions here you could get somowhere on the order of 25 30, 35 percent. That is cartainly an upper bound.
b ACE. FEDERAL REPORTERS. INC.
202 347 1700 Nation *We Coversee 800116-M44
24594.0 18 BRT
( )\ 1 MR. MICHELSON: As you point out, though, you 2 really can't add these.
3 MR. CAMPBELL: That's right. That would be a
,4 conservative upper bound.
5 MR. WARD: It sounds like it's not that far off, 6 though.
7 MR. CAMPBELL: Factor 1-1/2 to 2 above what the 8 actual contribution would be.
9 MR. WARD: Okay.
10 MR. EBERSOLE: Where you say " emergency 11 feedwater," usually, you mean auxiliary feedwater, right?
12 MR. CAMPBELL: Yes. B&W plants call it --
13 MR. EBERSOLE: You don't look at main feedwater O 14 valves? ,
15 MR. CAMPBELL: Only as potential accident 16 initiators. In fact in this IREP study, loss of main 17 feedwater was just treated as a single ovent and not broken 18 down to its causes so we couldn't get out the contribution 19 for valves versus other causes.
20 MR. EBERSOLE: Have you rubbed out the operator 21 error contribution or is it still in here?
22 MR. CAMPBELL If there's an operator error that 23 works directly on an individual valve that was modeled, it 24 would be in here.
25 There are some operator errors that tend to be O
ACE. FEDERAL REPORTERS, INC.
M)47 3700 Nellonw6de Coverage M)M
24594.0 19 BRT p
(_) 1 very important.in the IREP sequences, and sometimes they 2 involve failing valves as well as other things, and 3 involves a large number of other different components. We 4 didn't include them. -
5 MR. MICHELSON: I'm a little confused by the 6 fact that you seem to think that you can add these up and 7 divide tur a factor of 2 or 3 or something.
8 There's no way can you add these up. Each of 9 them is 100 percent. Each is, like the first one, 10 emergency feedwater, that's 5 percent of those 11 contributions from emergency feedwater, all of which add up 12 to some number, you know, like 100 percent.
13 MR. CAMPBELL: No, no, that's 5 percent of the (a~) 14 total core melt f requency and it includes f ailures of 15 valves in the emergency feedwater system. That's how it is 16 defined.
17 So it's not keyed, necessarily, to that system 18 failure versus all others.
19 MR. MICHELSON: No. But it's not keyed to the 20 fact that you have to add -- you can add all these others 21 up and come up with --
22 MR. CAMPBELL: They do represent fractional 23 contributions to core melt frequency. They all represent a 24 sum of frequencies of cut sets divided by the total 25 frequency.
()
- ACE FEDERAL REPORTERS, INC.
202 347 3700 Nesionwide Coverage 80S 3 4 4444
24594.0 20 BRT
) 1 If you added all these numbers, what you would 2 effectively be doing is double counting and sometimes 3 triple counting some of these cut sets and that's why you 4 would overestimate.
5 MR. MICHELSON: By a long, by a considerable 6 amount of thinking, don't you?
7 MR. WARD: He said it's just by a factor of 8 1-1/2 or something.
9 MR. MICHELSON: I thought he said 2 to 3.
10 MR. CAMPBELL: Probably 1-1/2 to 2. I have 11 enough information in my briefcase to be able to answer 12 that.
~x 13 MR. WARD: Sort of a bottom line -- I shouldn't
( i
, 14 use the term " bottom lino" talking about PRA, but you are 15 saying from this you'd say that something like 15 porcent 16 of the core melt frequency contribution is from Movs, all 17 the MOVs in this plant?
18 MR. CAMPBELL: That's right.
19 MR. MICHELSON: Okay. That makes senso.
20 MR. CAMPBELL: Again, that's not just MOV 21 failures, though. That's cut sots containing MOVs. It may 22 contain some pump failuros, too. You have to be very 23 careful about the way you interpret it.
24 HR. EDERSOLE: It may contain some pump failuros.
25 We are dealing with fluid flow and valven either permit it 79
\j ACE. FEDERAL REPORTERS. INC.
202147 370t) Nationwlde Coversee 80013HM6
24594.0 21 CRT
() 1 or does not. And it's pumps or no pumps. That's about the s
2 whole show. .I gather pumps are more important?
3 MR. CAMPBELL: We didn't do an analysis of pumps.
! 4 MR. WARD: I guess I didn't understand. The cut t
5 set contains the MOV, but a MOV failure is not necessarily 6 involved; is that what you are saying?
7 MR. CAMPBELL: It's not necessarily the only 8 thing involved.
9 MR. WARD: Oh.
10 MR. CAMPBELL: There would be other failures in 11 the cut set.
12 See, all we have to do to include a cut set is 13 to have a MOV in the emergency feedwater.
O 14 You can take a look at the Birnbaum importance, l
15 that gives you a little more of a feel in terms of how bad 16 things could get, if you will, in terms of if all the 1
17 valves, the MOVs in the emergency feedwater system fail.
18 What this number is actually saying there is the core melt 19 frequency would go up a couple of orders of magnitude.
20 That's if you went in and said the failure probability of 21 all these MOVs in the emergency feedwater system is 1, 22 that's how bad things could get. So, the two different 23 importance measures, and they do have different 24 interpretations, and probably in trying to decide how
- 25 important, really, valves are, it's useful to take a look O
ACE FEDERAL REPORTERS, INC.
302 347 3700 Naanonwide Coveress 80s.3 4 4646
'24594.0 22 BRT
.n
.I
) 1 at both of them.
2 MR. EBERSOLE: Is a failure viewed as a failure 3 to open or a failure to shut or both?
4 MR. CAMPBELL: It depends on whether it was 5 supposed to open or supposed to shut.
~
6 MR. EBERSOLE: So it's a particular mode.
7 MR. CAMPBELL: We have not discriminated against 8 any failure modes.
9 MR. EBERSOLE: Some are rigged to open, if they 10 can.
11 MR. CAMPBELL: That's right.
12 MR. WARD: Let's see, in the IREP analysis, is 13 there any common cause approximating or --
O 14 MR. CAMPBELL: Very little. No beta factor 15 analysis or anything like that where it was evident that 16 there was potential for a common causing event, they went 17 ahead and modeled that individual event. They did not do a 18 systematic common cause failure analysis of any kind.
19 MR. WARD: To me it's pretty important to try to 20 understand this because if PRA has something to tell us 21 about how important the problem of, you know, valves --
22 particularly motor-operated valves is, this is it. This is 23 what it's telling us.
24 David, this is one plant and it's just the IREP.
25 Have you got any comment on about how universally this sort O
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee 800 336 6646
_ - _ - - _ - _ - _ _ ~
y-24594.0 23 BRT-rx
(_) 1 of conclusion that might be drawn'from this could be i 2 extended?
3 MR. CAMPBELL: We certainly haven't done any of 4 this type of analysis for any other plants. Based on the 5 results of PRAs I have seen, I would certainly assume that 6 this is not atypical certainly for a PWR system. They, 7 from one plant to another, certainly tend to be very 8 similar in design. And PRAs tend to, I guess, take -- they 9 do a lot of averaging and take a very broad overview type 10 of look at things like valve failures, not getting down to 11 details of causes, and so forth.
12 So, I suspect that the results would not be 13 radically different from any other PRA.
O 14 MR. MICHELSON: What.do you use as a source for 15 your reliability information on valves?
16 MR. CAMPBELL: In this study they used the WASH 17 1400 data base as a starting point. Then they compared the 18 probability from the WASH 1400 data base to that which they 19 would expect in an experience in Arkansas nuclear 1, and 20 where it was significantly different they went ahead and 21 used the Arkansas data to update the WASH 1400 data.
22 What that means, for every valve type here, it's 23 WASH 1400 data except for air-operated valves. That's the 24- one valve type where Arkansas had experienced enough 25 failures to warrant that --
O ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide coversee 800 336 4646
24594.0 24
- BRT
() 1 MR. EBERSOLE: Let me ask you, the statistics on 2 reliability are obtained simply by floating the valve from 3 open to shut? How much did you depress that reliability 4 data 'towards a .more realistic -- when it's really doing its 5 thing?
6 MR. CAMPBELL: I guess it's the -- I'm certainly 7 not the best person in this room to answer that question.
8 In terms of statistics, the only way I know to check that 9 number.is to take a look at actual demands on systems and 10 try to get an estimate of valve performance in real demand 11 situations.
12 MR. EBERSOLE: In duress.
13 MR. CAMPBELL: Yes.
O 14 MR. MICHELSON: But we don't have many duress 15 situations to accumulate data.
16 MR. CAMPBELL: You don't have many but if you
-17 take them all you can get a pretty good idea of whether 18 your number is in the right ball park as least.
19 I know we did the sr.me thing on a diesel 20 generator study a few years ago. We took all the diesel 21 generator reliability data to estimate failure 22 probabilities for starting and running for diesels and 23 various other common cause failure parameters, and then we 24 took a look at actual experience during losses of off-site 25 power, any type. An emergency bus is de-energized and the O
ACE FEDERAL REPORTERS, INC.
202 347 3700 Naionwide covwnee 800 336 6646
124594.0 25 BRT' (3- We were able to get a
(_) 1 diesel is called on to start a load.
2 pretty good check for diesels, if we took all the 3 experience where they are actually demand. You really do 4 have'enough to get a ball park estimate.
5 MR. MICHELSON: It should be recognized, of 6 course, that valve technical specification surveillances 7 are done under nominal load conditions, not even water 8 flowing or anything, just the simplest possible, quietest 9 posrible test. From that you can't derive good data as to 10 how the valve performance, if it steadily has to open 11 against full differential or whatever.
12 MR. CAMPBELL: That's right.
l 13 MR. MICHELSON: How do you, in your mind, O' 14 compensate for that factor since when these systems are 15 really needed, there are high pressures across the disks or 16- they have to isolate breaks or whatever the case may be.
17 How do you rationalize that all this nominal load data is 18 f giving you a good, warm feeling about ability to close 19 under accident conditions?
20 MR. CAMPBELL: Again, I wouldn't necessarily 21 make that rationalization. I did not perform the IREP 22 study.- If I were to really try and answer that question, 23 and I think it is a valid concern, I would go back and try 24 to check how these valves really performed check their 25 reliability-performance under actual operating conditions.
n v
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 66d6
24594.0 26 EERT 1 MR. MICHELSON: We get one once in a while in 2 the form of an unusual event and then we find out they 3 don't work. And then we are surprised because all this --
4 I mean, you know, if you took the Davis-Besse event and 5 tried to calculate probability, which maybe you have done, 6 I'm not sure where you'd come up with that one based on 7 UASH 1400 data.
, 8 Somehow, it always leaves me cold to look at all 9 these good numbers from studies based on valve performances 10 under nominal load conditions. Somehow I ask: Well, what 11 happens if we had a real event? How would these systems
- 12 all work?
3 13 MR. CAMPBELL: Well, again I think you need to
~) 14 take a look at real events. Certainly real events where 15 things fail are brought to everybody's attention. Real 16 events where things work tend to go unnoticed.
17 I think, if you can go back to the data, and it 18 is a large job to go back and try and evaluate the number 19 of demands on various types of safety systems, if you do go 20 back and take a look at the number of demands and the 21 number of different valves that are challenged in each 22 demand, and then come back and look at the times they 23 failed during actual demands, you don't have necessarily !
24 enough statistics to have a high confidence in any 25 reliability parameter you estimate that way, but you can p]
\-
i ACE. FEDERAL REPORTERS, INC. '
202 347 3700 Nationwide Coverses 800 330u46
-24594.0 27 BRT.
n
( ,) 1 get a good feeling whether you are in the ball park or not 2 with the figures, say, you get out of a WASH 1400 data base.
3 I agree about your comment with what we are 4 doing with some of these bases, and even more important i
5_ what we have tended to do in the past with PRAs, we acted
'6 like everything fails independently. Every time we have a 7 serious situation like Three Mile Island or Davis-Besse or 8 any other you can name, what we've really got is one or 9 more dependent failure mechanisms; the common cause of 10 multiple component failures is active.
11 Those types of things for the most part were not 12 even modeled in this IREP study. Until you've got a PRA 13 that has that in there, you can't get out the right k
14 conclusions that will really show you what's going on here.
15 MR. WARD: Let me ask you another question, 16 David. If you went through the -- the IREP, the .
17 calculations again and plugged in numbers for valve 18 failures, a factor of 10 greater than you used, how much 19 would that change the Fussell-Vesely importance. I guess 20 that's what the Birnbaum importance number is supposed to t
21 be telling you, but I can't figure it out.
22 MR. CAMPBELL: The Birnbaum can show you what 23 kind ~of an increase you would get, if you took the' failure 24 _ probabilities to 1.
25- On the emergency feedwater system, at 5 times 10 l
l r
ACE-FEDERAL REPORTERS, INC.
t 202-347-3700 Nationwide Coverage 800-336-6646
24594.0 28 BRT p
(_) 1 to the minus 2 there, what that means is -- what that 2 number is, is what the core melt frequency would be if 3 .every valve in the emergency feedwater system were 4 guaranteed to be failing. That's as compared to 3 times 10 5 to the minus 4 under nominal conditions.
6 MR. EBERSOLE: That's the reliability -- either 7 it works or'it doesn't work.
8 MR. CAMPBELL: Yes. In this case if all the 9 motor-operated valves fail, the system would not work, so 10 that would show the effect of that system failure on core 11 cielt frequency. It would guarantee the failure.
12 MR. WARD: Okay. So that's, if you assumed all 13 the valves failed in that system, that's the -- that's just O . 14 sort of the demand -- well --
15 MR. CAMPBELL: It's almost, in this case since 16 the valves would fail the system, it's like saying if I 17 just didn't have an emergency feedwater system, how.high 18 would, core melt frequency for this plant be? It gets right 19 on up there.
20 That doesn't really help much with my ftR. WARD:
21 question. j
=
\
22 MR. EBERSOLE: You did this on ANO-l?
23 MR. CAMPBELL: Ye's'.
1 24 MR. EBERSOLE: If you-spent to ANO-2 you'd have 25 trouble -- oh, no, Arkansas put a-palve in, didn't they?
( n/
\_
l l
' ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 2 336 4 646
24594.0 29 BRT:
_1 MR. WARD: Might still have trouble, but they 2 got it. Can you help me any with some sort of a crude 3 estimate of what a factor of 10, in failure probability 4 would be? ,
5- I don't know what the numbers, but instead of 6 failing 1 percent of the time, what if I assumed they 7 failed 10 percent of the time and then make the 8 calculations? How does that change the core melt 9 contributions?
10 MR. CAMPBELL:- Off the top of my head I'd say it 11 would probably about double it.
12 MR. EBERSOLE: That number, S E-2, looks 13 suspiciously low to me. If I abandoned low pressure O 14 depressurization of the boilers, the only way I can get 15 water into the secondary main and emergency feedwater; 16 right? That's the only way I get the heat out. That's the 17 only way I ever get the heat out.
18_ That makes that number look suspiciously low.
19 MR. WARD: That's a pretty-high. number. That's 20 once in 20 years it's going to happen.
21 MR. EBERSOLE: I know that, but it still looks 22 low.
23 MR. MICHELSON: I'm surprised that the service 24 water being about the same order of magnitude, if I 25 understand the relationship here properly. It just says if D)
%s ACE-FEDERAL- REPORTERS, INC.
202-347-3700 Nationwide Coverage 8043364646
I 24594.0 30 BRT
/m,
' ts_) 1 I' lost all service water, that's really -- the importance 2 of service water is only a 5 percent contribution to core 3 melt. Is that right?
4 MR. EBERSOLE: You are saying it's a reliability 5 system, though.
6 MR. MICHELSON: No, no, not under the Birnbaum.
7 MR. CAMPBELL: This is an actual measure of core 8 melt frequency if all these valves in the service water ,
9 system fail. Core melt frequency would go up that high.
10 MR. MICHELSON: That's essentially saying -- I 11 think if you fail them all, you'll probably have a loss of 12 cooling water. If you' fail them all closed.
7-
. 13 Is this saying that's only a 5 percent
(/ 14 contribution to core melt?
15 MR. CAMPBELL: It's hard to say exactly what a 16 failure of all these service valves'would do during normal 17 operation.
18 MR. MICHELSON: You'd be in deep trouble 19 eventually because you'll lose all your systems because 20 they all depend on cooling water from'the service water.
21 MR. CAMPBELL: You are assuming they are going 22 to transfer closed, I guess?
23 MR. MICHELSON: Yes. Yes. .
24 MR. CAMPBELL: That's not necessarily what is 25 going on here.
/'N V
l
! ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 900 336 4 646
24594.0 31 BRT
() 1 MR. MICHELSON: I thought that is your 2 postulation, worst direction failure.
3 MR. CAMPBELL: The IREP did not look at the 4 service water system during normal operation. It's only 5- under accident conditions. There's a transfer to the 6 correct configuration.
7 So not even all valves in the service water 8 system are considered in this number.
. 9 MR. MICHELSON: Do you have a comment?
10 MR. RASMUSON: Yes, let me just give a comment 11 here. When we are looking at cut sets for an accident 12 sequence, we've got maybe one component from the, you know, 13 from the emergency core cooling system and maybe we've got
(_s t.
'~
14 something c,oming up from the emergency -- auxiliary 15 feedwater system and then maybe another component from the 16 service water system. So what we are doing, one of the 17 reasons that these things are all looking close to the same 18 is I've got a valve maybe from -- got a cut set that maybe 19 is made up of three valves, and so one from each of these 20 systems and so when I'm calculating the contribution for 21 that cut set, you know, that particular cut set would be 22 showing up here in the emergency system, and also down here 23 in the service water system, because the valve is -- which 24 system he's looking at, you know, that could impair in both 25 of those things.
l ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6646
24594.0 32 E RT.
(_) 1 So these numbers , it's not surprising to me that 2 they tend to be --
3 MR. MICHELSON: On my part I think it's just 4- ignorance of the details of what you've done here, trying 5 to look at the answer and not really know how you got there.
6 I think we'll have to move on, although I agree 7 with David's comment earlier, this.is very important to 8 understand but I'm not sure we are going to ever understand 9 it without going back and reading and looking at how you 10 got to this point because it's -- it is a little bit 11- confusing to look at just these answers.
' l:2 Right now I don't believe I can draw any
- 13 conclusions from these answers.
%).
14 MR. CAMPBELL: It's difficult. You've asked i
15 some questions that.could be answered. For example, if you 16 increase the failure probability of these by a factor of 10, 17 what would be the effect? Those kinds of questions can be 18 answered fairly easily.
! 19 If there are specific questions you would like 20 to have answered that you believe could be answered by PRA, 21 possibly if -- if we do come back or Gary Burdick does come i 22 back at a later date, we might be able to answer some of 23 those questions.
2:4 MR. MICHELSON: I think we'll have to move on, 25 though.
O ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646
24594.0' '
33 BRT
.() 1 MR. WARD: I guess I would sort of like to try 2 to understand it better, perhaps the answer to that
.3' specific question would help. Because, you know, I still 4 don't know how concerned to get about the problem of valves, ,
-1
-5 unless --
6 MR. MICHELSON: I don't either and I'm having
~
7 difficulty concluding from this chart whether I should be 8 concerned.
9 MR. WARD: It's telling us something, I think.
10 I don't know how universally applicable the conclusion is t_
11 or how much uncertainty there is attached to it.
,' 12 MR. EBERSOLE: There's so much of the color of
~
13 the individual system. design in here, I can't make much out
( 14 of it. ,
15 MR. MICHELSON: Can you send us, do you have the 16 report or basis for the calculation?
17 MR. WARD: Could I respond to Jess a minute?
, 18 Although, Jesse, he gave-his opinion that he thinks most 19 PWRs would look something like this. So that says it's not, 20 in his opinion, it's not all that dependent on the details.
21 MR. EBERSOLE: Does this reflect the fact if I 22 lose service water, as a case in point, I'm going to have a 23 core melt for sure?
'- 24 MR. . CAMPBELL: No, it doesn't.
25 MR. EBERSOLE: Tell me how I'm not.
i
()
l ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336 4646
24594.0 34 bRT
() l- MR. CAMPBELL: You are not going to have a core 2 melt if you lose service water?
3 MR. EBERSOLE: I would like to know.
4 MR. CAMPBELL: I would have to take a look at 5 that. There is a separate event tree for loss of service 6 water initiating the event. I'm not exactly sure what type 7 of recovery can be --
8 MR. EBERSOLE: As far as I know, it's the next 9 best thing to DC power f ailure, which is the worst. I 10 suppose the service water comes next.
11 MR. CAMPBELL: There's a lot of other things 12 that need to be kept in mind here. We have looked at 13 dominant sequences, not at the whele contribution to core e
14 melt frequency. We have not looked at any possible 15 recovery actions. A lot of these things that could fail, 16 possibly you could recover from.
17 MR. WARD: You haven't credited r.ny recovery?
18 This is just:the old IREP analysis?
19 MR. CAMPBELL: That's right.
20 MR. EBERSOLE: Maybe that's not a very good base.
21 MR. REED: Just to try to get what Jesse is 22 saying for a minute, it's my recollection that loss of 23 service water is not compounded with primary -- in a PWR, 24 with primary loop failure. Therefore, you would go on 15 auxiliary for the feed. And steam blow-off.
(
(
f ACE-FEDERAL REPORTERS, INC.
( 202-347-3700 Nationwide Coverage 800 33H646
'24594.0- 35
- ;BRT l-h -11 MR..EBERSOLE:
But the auxiliary services are 2, provided by service. water such as environmental cooling,
['
3 seal' cooling -- i r
4 MR. REED: I know you are saying that, but 5 there's:been a big push on in the last five years or more !
4' 6' to have~the seals on the' pumps cooled by fin coolers and i i ..
7 this sort-of thing, so I think that's supposed to be h 8- covered.
!' . i i ~. 9 MR. MICHELSON:
I kind of agreed with you, I 1
10 thought,- until I started inquiring around and I find the I -11 utilities really still have to have service water for ;
- j. 12 auxiliary feedwater. Some of them would get by two hours i ;
13 and so forth but they've:got to get'a recovery mode going. i
.- 14 It's not indefinite. But I think that's -- you can't !
15 conclude anything from this brief information'.
l_ 16 -
'MR. CAMPBELL The conclusions you are trying to i . .
17 make I don't'think can be made from these. slides. There i
- 18 are some conclusions that.can be made. 7 i
I 1-
- 19. The first, the relative-rankings, here, using ,
- 20. .either one of the importance measures, are fairly valid. _
. i 21 For. example, the MOVs in the individual water system are lc 22 more important; than . in the reactorL building spray system. f f
?
23 'There's-no question about that. If you are going to worry !
I 24 about MOVs, don't go look at the reactor building spray l 25 system,-look at the service water system.
~
O i-l' !
t
[
ACE-FEDERAL REPORTERS, INC. [
an.w.mo - N.u.-ie.co m6mm x__-_=._- -:_-____=___-_:___---__-- _ - _ - - _--__--____--.____:
24594i0 36 BRT.
i f')'
s.
1 The other conclusions that you can make from the .
2 relative ranking are that motor-operated valves are the 3 mest important. I think check valves and manual valves 4 tend to be next.
5 The PORVs come up important. It is one valve,
- 6 though, pressurized relief valves, that's what is driving 7 that.
8 MR. EBERSOLE: But look at pressure'-- it's high 9 but yet why? You have to of some sort of depressurization 10 failure to lose inventory to need those things, yet it 11 shows as a fairly high contributor.
12' MR. CAMPBELL: Our -- failure of high pressure 13 injection was involved in almost all the dominant sequences f'} 14 at Arkansas. That doesn't --
i 15 MR. EBERSOLE: What was the antecedent accident?
I 16 MR. CAMPBELL: Okay, it was -- tended to be i
17 small, various transients and very small local --
4 18- MR. EBERSOLE: Transients? What do you need
]
19 high. pressure for?
20 MR. CAMPBELL: I think it's probably looking at 21 -the relief valves lifting and high preasure injection may -- l 22 you may not necessarily even.need it but it's certainly a 23 backup if emergency feedwater does fail. That's being ,
24 factored into the sequences, that it is there as a backup. ,
25 MR. EBERSOLE: It's probably driven in by PRV i
- ^\ i U .
1 i
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide coversee 800 336 6646 .
24594.0. 37 BRT
) 1 failure.
2 MR. CAMPBELL: I would suspect that's a major 3 part of it, yes.
4 MR. EBERSOLE: Instead of pipe failure.
5 MR. CAMPBELL: Loss of seal cooling is loss of 6 high pressure injection, I guess. Not necessarily.
7 MR. EBERSOLE: And emergency room cooling, 8 that's driven by loss of normal emergency systems,.isn't it?
9 MR. CAMPBELL: That and loss of off-site power 10 which shuts down the normal HVAC. Yes.
11 Again, I think the conclusions-that can be drawn 12 from all the slides, and I guess I won't show them because
, 13 you have them in your handout package, the conclusion that 14 can be drawn, or the conclusion that can be drawn, revolves 15 around anything you can conclude based on relative rankings 16 of different groups of components that are shown here.
17 Again, we can consolidate these groups. If we 18 had more data we could probably refine the groups which 19 might, I think, be more appropriate. I believe in some 20 cases we have got things too consolidated already.
21 For example, if you take the high pressure 22 injection system, take a valve on the section side of the 23 pump and valve on the discharge side of the pump and lump 24 them into the same thing, call them the same valve type, I 25 think you are making a mistake.
r ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 4 646 h_
24594.0 38 BRT
( ,) l- So, probably, more attention would need to be 2 paid to the refining of these groups, as much as anything 3 else. Then the picture tends to get harder to interpret 4 when you do that.
5 Instead of five groups of valves, now you've got 6 10 or 20, maybe so the conclusions get more difficult.
7 Again, anything you can conclude based on 8 relative standings, here, I think is appropriate.
9 Are there any other questions?
10 MR. MICHELSON: Any more questions? I don't 2
11 have any more questions but I~must honestly admit I don't 12 really understand it either, because I think we've gotten --
13 we just don't have quite enough in front of us.
14 - I.s it possible to get a little more written 15 material or something?
16 MR. CAMPBELL: If you have specific questions 17 that you think could help you clear up'the way you want to
'18 interpret this, I think we could probably go back and do l 19 some calculations to answer those questions.
20 For example --
21 MR. WARD: Got a couple. Excuse me, go ahead.
22 MR. REED: On your last slide, reactor coolant, 23 is that the PORV? Is that what that's describing?
24 MR. CAMPBELL: Yes.
25 MR. REED: PORVs could be at 10 percent, a 10 l CE) i
[
l ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 3%4M6
n j
24594.0 39 ERT
() l percent contribution to core melt risk; is that what you 2 are saying?
3 (Slide.)
4 MR. CAMPBELL: Yes. That's exactly what that is.
5 They are involved in 10 percent of the dominant accident 6 sequence occurrences.
7 MR. WARD: I guess some specific questions, you 8 know, that would help me would be, you know, if there could 9 be some sort of -- just using this, the analysis that has 10 been done, some sort of sensitivity analysis to -- you know, 11 failure probability, the factor of 10 thing, for example.
12 And then a couple of the other, is there 7-13 somewhat -- limitations to the IREP in that common cause
( )
\'# wasn't handled very well. There are more -- I guess some 14 15 better ways to estimate common cause effects now. If we 16 could get some handle on that. And, also, it seems to me 17 for valve failures, crediting recovery is particularly 18 important -- a handle on that.
19 Then some sort of a further opinion or estimate 20 or something, on the generic applicability of these results.
21 MR. CAMPBELL: Some of the questions you just 22 asked could be answered in a very straightforward manner, 23 doing some more calculation, computer programs. Some can't 24 really be answered other than almost an expert opinion type 25 thing; for example the common cause effects, to say what 7,
x_
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 3364M6
'24594.0 40 BRT
([) I that really is. That hasn't been done in a PRA in such a 2- manner that you can pull that out.
3 On a system level analysis, that type of thing 4 has been done. ,
5 MR. WARD: We have to form an opinion on that 6- sort of thing so it may as well be from experts.
7 MR. CAMPBELL: I guess that makes sense. We can 8 include the recovery, and redo these calculations.
9 So, I think we could answer those questions, 10 certainly, at a later time.
11 MR. MICHELSON: You can appreciate, of course, 12 that it is desirable that one has an understanding of the 13 safety importance of motor-operated valves, that they are a i 14 minor contributor to core melt, and we simply, then, will 15 look at other things.
16 So we are kind of trying to put it in a basket:
17 Is it a minor contributor? Intermediate? Major? How 18 important are' they, relative to the core melt?
19 MR. CAMPBELL: I would say they are fairly ,
20 significant.
21 MR. MICHELSON: What does that mean?
22 MR. CAMPBELL: 20 percent of the core melt 23 frequency does involve failures of, say, motor-operated 24 valves. ,
25 MR. MICHELSON: Yes, but that statement in [
(2) i l
ACE. FEDERAL REPORTERS, INC.
l 202 347 3700 Nationwide Coverses 800 33H646
24594.0 41 BRT-( w) 1 itself_doesn't help me much. I have to see how you came up 2' with that because that doesn't -- I'd have to see what 3 failures you are talking about -- in other words, that same 4 core' melt frequency would occur because of the other 5 failures, even if all the valves worked.
6 MR. CAMPBELL: No, if the valve works it takes 7 away that contribution. Their failure alone won't cause it.
8 MR. MICHELSON: I thought you were pulling these 9 out of cut sets and I thought there were other members in 10 the cut sets and I wanted to see what they were to make 11 sure that, really, if I pull the valves out, whether it 12 would all go away. I'm not convinced that pulling the 13 valves out makes the numbers all go away. I'd have to see O 14 your work.
15 MR. CAMPBELL: I think at this point I have a 16 -lot better idea of what your real interest is. I think 17 we've got the information, at least on an ANO, and based on 18 the IREP, of course, that we can answer most of these 19 questions.
20 It turns out if you want to look at the dominant 21 contributions to core melts just in terms of cut sets, we 22 can boil that down to a few hundred, so it's not an 23 intractable job to go look at those one by one and really 24 see what's going on.
25 MR. MICHELSON: When you did the ANO-1 -- I I
(
ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 6646
- 24594.0 42 BRT
() 1 guess you did the ANO-1 work or were you just looking at it?
2 MR. CAMPBELL: We didn't do it --
3 MR. MICHELSON: You recall, the ANO-1 work, did 4 .they deal'with the loss of coolant accidents outside of 5 primary containment and the need for such events to isolate 6 the event?
7 MR. CAMPBELL: I'm sure they looked at that.
8 None of those sequences were dominant so, in terms of --
9 MR. MICHELSON: Of course they may not be 10 dominant if I had a highly reliable isolation arrangement.
11 MR. CAMPBELL: That's right.
12 MR. MICHELSON: And made some, you know -- it 13 could have disappeared simply because people used much more
(_ss ..
\# 14- optimistic. numbers of reliability of valves under dynamic 15 a conditions than was justified. If they had gone back and 16 put probabilities of one on the failure of the valve, it 17 would have come up with some interesting, perhaps 18 interestingly different results.
19 You know, I never can get a warm feeling about 20 how these things were looked at. That would make a 21 difference. Motor-operated valves may beccme very 22 important if it turns out they have a low probability of 23 success under dynamic conditions.
24 MR. CAMPBELL: I think one thing to keep in mind 25 is when you are doing a PRA, you have a whole lot of
/~N l V
ace FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coversee 800 336 6646
24594c0 43 BRT 1 information you have to sort through. There are a large 2 number of potential users of that PRA down the line.
3 You are very interested in valves, you probably 4 wish they spent a lot more time analyzing valve events and 5 sequences where valves were critical.
6 Somebody else might be more interested in pumps, 7 or human errors or this or that, and you just have to kind 8 of take a broad brush view, overview of everything, 9 particularly in a program like IREP, where the cost of the 10 PRA was limited. It was not a very, very large-scale PRA 11 effort. Certainly not -- I would not consider it the 12 present state of the art in terms of complete PRA.
13 MR. MICHELSON: I think perhaps by now you have 14 seen what we are worrying about. We are trying to 15 determine whether or not to even look at valves. And, if 16 so, which ones to look at. And we are looking for some 17 quantitative guidance along these lines as well as
') qualitative guidance. That's why you are here. And I'm 19 not quite sure in my mind I've gotten enough to help me in 20 that regard.
21 MR. CAMPBELL: I guess I came here -- I was a 22 little bit short on information about what you wanted. I 23 got a call that said can you calculate valve importance to 24 ANO, and I said yes.
25 Possibly, if Gary Burdick were here, he knew ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverase M 33 N
l
'24594.0 '44
-ERT .
1 C 1 pretty much what we had. And possibly his presentation 2 would have put things into a better perspective for you. I l 3 don't know. l 4 But I think we can -- I can certainly come back 5 some other time and answer some of the questions you have 6 here.
7 MR. MICHELSON: Let's put it on a qualitative 8 basis for a moment, and only on the basis of your own 9 judgment.
, 10 If you were asked to think about this problem, 11 would you even consider valves in, you know, in terms of i
12 the fact that there are many other components, turbines and 13 pumps and electronics, whatever -- if you were trying, with 14 your limited time and resources, to look at the most 15 important reliability questions relative to core melt, 16 would you be looking at valves at all? Or, if so, what 17 kind of valves would you be focusing on?
18 MR. CAMPBELL: If I were asked that question and 19 I had to pick a type of component, certainly I would look 20 at valves. I think those would be right -- very near the l
21 top of my list, if not the top. I guess, if were 22 concerned about reactor safety I might come at it from a
- 23 different approach though, in terms of taking a look at 24 various reliability assurance programs and options, and 25 what are the most important options there. Because those o
V l
ACE. FEDERAL REPORTERS, INC. l 202 347 3100 Nedanwide Coverage 800 33Hed6 l
_ _ - - _ _ _ .______ -.- . _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ . _ - _ _ _ . _ l
24594.0 45 BRT i
'I) _
I types of things affect not only valves but other types of 2 components in the plant at the same time. ,
t 3, I don't see that you could necessarily pull out 4 a type of component and examine that separate from the rest 5 of the plant, in terms of trying to improve safety.
6 MR. MICHELSON: I guess you were saying that you ,
7 would lock at it from the viewpoint: Are there suitable ;
8 programs to assure the delivery of reliable components?
9 MR. CAMPBELL: Yes. And when you get down then 10 to a very microscopic level, you would be looking at valves 11 and saying: Do they have -- are there programs for 12 maintaining these valves properly? Tecting them properly?
1 13 Verifying that they really do work? Maybe taking a look at 1
O 14 the testing program, see if -- if the tests aren't testing 15 the valves under actual conditions, I would be kind of i
16 interested in seeing some sort of different test. May be ,
17 certainly not monthly, but some sort of test, revealing --
18 to simulate actual conditions on valves. Things like that. l 19 MR. MICHELSON: You wouldn't select a particular 20 kind of valve as being a center of focus, then, but rather ,
21 just valves in general?
22 MR. CAMPBELL: Yes. If I were to pick a type, I 23 can see from these results that I would be interested in 24 the PORVs and the motor-operated valves. But there's 25 plenty of experience in terms of incidents that have Q
v ACE FEDERAt. REPORTERS, INC.
n2 347 33 Nationwide Coverage 300 344M6
24594.0 46 BRT (j 1 occurred at plants to indicate that valves are important, 2 ' including manual valves, to make sure they are in the right 3 position. So I do believe valves would be very important, 4 a very important part of the examination.
5 MR. MICHdLSON: If you are thinking in terms of 6 qualitative problems, instead of quantifications,'now you 7 can start dreaming a little more about common cause 8 potentials. ,.
i q
9 Electricity, of course to a number of valves, is 10 a common supply that can go to higher voltages or different 11 frequencies. But the quality of electric power supply can 12 be reasonably carefully controlled.
13 Let's take the case of air supplies to 4
" O 14 air-operated valves. We don't have the degree of control 15 over the contamination. It's hard to contaminate electric 16 power supplies but air supplies can be much more readily i
! 17 contaminated leading to common cause failure to a number of L valves at that time.
)
18 j 19 If you start thinking qualitatively, your ;
20 imagination could see that maybe air-operated valves are 21 something that, although they don't show up the same
- 22 perhaps on your quantitative analysis, that they have a '
23 potential which doesn't normally show up for -- getting 24 water into the line, for instance, and getting to all the 25 air operating lines at once. A very neat way to mess up a
.O ACE FEDERAL REPORTERS, INC.
I am.im somcc minee
l
~
24594.0 47
-BRT
() 1 lot of operations. They have strange failure modes also.
l 2 MR. CAMPBELL: I remember taking a look at j
3 air-operated data at one plant, and they had a long series . ;
- 4 of failures cause by oil in the air system that tended to
- 5 cause the valves to bind up. I agree.
6 MR. MICHELSON: I look for instantaneous 7 injection of a contaminant, for instance in an air system, t 8 like water: Shooting it into the airline. It happened in 9 a small way at TMI.
i 10 I look at that. It's a common cause challenge.
11 And I don't find it whenever people work out all their i
12 little numbers. I don't find these kinds of things.
13 MR. CAMPBELL: That's one thing I think, .not in l 14 the air system because air-operated valves didn't tend to 4
l 15 be that important at Arkansas, but one thing they did very I
t 16 well in this and some of the other IREP studies was to look l
l 17 at the system interaction type of things, where you --
1 18 where these dif ferent components have a common support 19 system, really go in and model those support system l 20 failures in some detail.
21 I think, again, the air system was not an 22 example of that but I think the other support systems were 1
23 modeled very well in Arkansas. In fact, those -- they were 24 taken -- you know, a very close look was taken at them and 25 they tend to dominate the core melt frequency. Loss of DC
(
ACE. FEDERAL REPORTERS, INC.
203 347 3M0 Nacionwide Cowess at&)M4H6
'24594.0 48 BRT 1 power is a dominant contributor. AC power is somewhat less 2 significant. Service water and these emergency cooling 3 systems are very important.
4 MR. EBERSOLE: They are roots.
5 MR. CAMPBELL: Yes.
MR. MICHELSON: So I guess in conclusion, then, '
~6 7 you feel that valves are a worthwhile endeavor if one has 8 limited time?
. 9 MR. CAMPBELL: Yes.
10 MR. EBERSOLE: Did you stop to think that the j 11 basic safety functions are nothing but pumps and valves and t'
- 12 pipes, and pipes are pretty reliable. That pretty well 13 localizes where the problem lies.
Okay.
14 MR. MICHELSON: Well, we thank you. I 15 guess your presentation is completed?
16 MR. CAMPBELL: Yes.
17 MR. MICHELSON: We thank you very much.
18 Did the staff have any other material they 19 wanted to present this morning? We are ahead of schedule 20 which is unusual. I'm sure we'll lose it somewhere along
, 21 the way but we appreciate having the extra time now. ;
22 Our next presentation is going to be about an
, 23 hour by Limitorque, so I think we'll take our break now.
24 We'll take a 15-minute break. Be Lack at 10:00.
l ,
25 (Recessed at 9:45.)
!O l
i ACE. FEDERAL REPORTERS, INC.
I 202 347 3700 Nationwide Coverage 300 336 4646
24594.0 49 BRT
() 1 MR. MICHELSON: The meeting will come back to 2 order and we'll proceed now with a presentation by 3 Limitorque, Mr. Dan Warsing, who will tell us something 4 about Limitorque's experiences. If you will, proceed.
5 MR. WARSING: Before we start, do you want me to 6 sit here or would you prefer I stand?
7 MR. MICHELSON: This is fine. Speak up loud 8 enough so the audience can hear is the main thing.
9 MR. WARSING: All right. My name is Dan Warsing 10 and I'm from Limitorque and I have been asked today to 11 speak a little about Limitorque's field experience to 12 improve reliability. I have to confess I haven't been that
.y 13 familiar with what this committee has been working on and
( )
14 for that reason I thought it would be better for me just to 15 cover some general areas and give you folks some time to 16 ask me some questions you would like to ask.
17 In general, our field experience began about 50 18 years ago. We have been making valve actuators that long.
19 By and large we do have a representation in the industry 20 for making rugged equipment that has long service life. We 21 have a lot of equipment that is still operating in power 22 plants and other facilities for 30 and 40 years with very 23 little or no maintenance effort required.
24 It has been our experience that in general, the 25 level of expertise and knowledge of the Limitorque product l )
v ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 33MM6
24594.0 50 BRT
) 1 in the industry is lacking. The questions that were asked 2 and the quality of the information we get on field 3 experiences is very poor. The descriptions of the problems 4 that were given, they are not logical. There's no way you 5 can explain what you are being told. And when you try to 6 go back and delve into it and get more information, the 7 people aren't able to get to the information that you need 8 to really make a proper analysis of what occurred.
9 MR. EBERSOLE: I take it you are saying people 10 don't even know how to operate the valves?
11 MR. WARSING: The operation is really pushing it, 12 but the setting up of it and even the effort to repair it 13 after it fails. They go in and, quote unquote, "fix it."
O 14 And they get it to work. But it's usually done in a 15 shotgun approach. We try this, that, and the other thing.
16 When it final 1y works okay it's fixed, we walk away, we 17 make a report, this is what we changed. But they don't 18 really know what they changed and what they really i 19 corrected because they never find the cause of the problem.
20 MR. EBERSOLE: An open set of valves, are there 21 any certification requirements on a person to do that?
22 Would you invalidate your warranties if an unqualified 23 person, untrained, didn't have a certificate of some sort
. 24 to do that?
25 f tR . WARSING: No, sir, we do not. We do offer ,
C) :
i i
ACE. FEDERAL REPORTERS, INC.
202 347 Nationwide Coverage 2 33 Mi646
_ _ ____ -_________________________-_________ ___ _ _ ____ 3700
24594.0 51 BRT
() 1 training programs. We have one fellow who spends a lot of 2 time going around the country to different power plants and 3 giving maintenance seminars. He also have people come into 4 our. facility in Lynchberg, Virginia, and also Chicago, 5 where we put them through training programs at least to 6 teach them how'to do the basic disassembly and reassembly 7 of the actitator.
8 MR. EBERSOLE: If you tighten up on that, you 9 lose customers, don't you, so you don't require
-10 verification or you require some pressure to be brought to
. bear on people who are maintaining your valves?
12 MR. WARSING: I'd say it's pretty true. A good 13 bit of the industry wouldn't appreciate it --
14 MR. EBERSOLE: Like the aviation industry, where 15 you don't take an engine apart unless you have a 16 certification to do it. You don't have that sort of thing.
17 MR. MICHELSON: How long does Limitorque warrant 18 an operator, generally?
19 MR. WARSING: Our stated warranty is one year 20 from date of shipment.
21 MR. MICHELSON: That's from date of shipment but 22 you know the date of installation is five years later or 23 something.
24 MR. WARSING: And date of startup is 10 years 25 later. As I said, that's our stated policy. In' actuality O
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 4646
l 1
24594.0 52 BRT l i
i
-(])_ 1 our working responsibilities are if it's a defect in 2 workmanship or manufacturing, it's covered practically 3 indefinitely unless there's abuse involved.
4 MR. MICHELSON: When you find it.
5 MR. WARSING: Yes.
6 MR. MICHELSON: Okay. I was just a little 7 curious how that works. Thank you.
8 MR. WARSING: As far as the methods that we have 9 to improve reliability, we do try to investigate the 10 reports we receive. We also -- there's a group of people 11 within the organization that review all of cur field 12 service reports, where our field people go out and do work 13 on equipment in power plants, and we do try to evaluate all O 14 those things and generate product improvements as we feel 15 them necessary.
16 In the past five years we have made a number of 17 product improvements, just to improve the reliability and 18 decrease the number of failures that can occur.
19 MR. REED: The field reports that you receive, 20 are they from your service engineers or from the utility?
21 MR. WARSING: Those are from our field service 22 engineers.
23 MR. REED: Are you ever concerned that they 24 might be subjective or biased? I mean, after all, the man 25 might be protecting his own -- or the company's own
^
r )
(_t .
ACE-FEDERAL REPORTERS, INC.
202-347 3700 . Nationwide Coverase 800 336 6646
'24594.0 '
53 BRT 7~s . .
- 4) 1 interests..
2 MR. WARSING: We certainly couldn't rule it out.
3 It has been my experience in working with the field
'4; -organization that-when they are filling out a report to 5- send in the office, they tend to be very direct about their 6 comments. Sometimes, in very explicit te rms . They are not 7- bashful. folks.
8 MR. REED: I'm sure some would. And then there are some that might be on the fast track, that wouldn't.
~
9 1
10 It might affect the future progress of the company.
'll I'm sitting here thinking of an incident, it's c I T 12 valve incident, some 20 years ago, where it was obvious to 13 !everyone except the field' engineer and the company that O 14 provided the valve that. the valve stems were. two inches too,
-15 short to be able to function. And yet it is a pretty
! . 16 simple thing, when the valve stems are too short; isn't 17 that true?
l 18 MR.'WARSING: ' Yes, sir.
19 MR. REED: Yet the company sat there maintaining 20 ' forever and anon that the valve stems were not two inches
- 21 too short. Except they wouldn't make the stroke.
22- So, I'm reflecting ~ -- Limitorque has been !
23 involved in this valve ' operator business for many years and 24 I'm sure they got the historic and all the:other service in 25 nuclear is probably a small fraction of'their business; is l
l
?
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverase 800 336 4646
1 1
I'594.0 54 !
'BRT-1 that'true?
'2 MR. WARSING: Certainly today; yes, sir.
3 MR.--REED: Yet, unfortunately, I have to think 4 that the performance, reliability of valves and operators 5 in nuclear is a -- is a different requirement, very
~
6 'different' requirement. And yet I suspect that your 71 Limitorque operators are still made for both services, the 8 nuclear and all the other services; is that true?
9- MR. WARSING: Certainly the two applications 10 .w ere -- the nuclear application is a much more critical one.
11 The. basic design --
-12 MR. REED: Do you really specialize, redesign,
- 13. -think about, concentrate more on field reports with respect O 14 to the -- a nuclear version of Limitorque?
15 MR. WARSING: Yes, sir. The design, of-course, 16 I'm sure you are aware, doesn't significantly enange, the
. l'7 - basic product design.
18 Components change, based L on the requirements for 19~ .the environmental conditions.
20 As far as reviewing service reports, I'm one of 21 the people that reviews those service reports, and, yes, 22 : sir, I definitely do look at them.
23 L MR. REED: I'm thinking of a recent case of a-24 . utility who, and this certainly is a utility's, largely 25 their fault ---because_of the gear arrangement,Lyou can put O
ACE-FEDERAL REPORTERS, INC.
-202347.noe- . Nationwide Commer son 366646
.24594.0 55
'BRT
-X 3_) I them in either way, put them in backwards. Some size this.
2 They go in one way and some size the other. They put them 3 in backwards and so the tooth engagement was very limited 4 but it -- and it resulted in a failure. )
5 MR. WARSING: Yes.
i 6 MR. REED: In reviewing that kind of situation, s 7 would you think about the responsibility of the designer to, 8 perhaps, recreate and make a little bit ~ idiot proof, that 9 aspect, that gear aspect?
10 MR. WARSING: In general terms, yes, we would.
11 In that particular case you would be talking about a major 12 design change in the basic product. In that case, it i 13 becomes cost prohibitive to the-utilities to really make O 14 that, type of a design change; to make it so that there was 15 no way he could install it backwards would mean you'd have 16 to. redesign motor and gearing end of the unit, which then 17 means that we have a major retrofit for all the utilities 18 to do to take advantage of it.
19 When it comes to something like that, generally 20 no, we don't do-it.
21 Our instruction maintenance manual does address 22 the correct installation position for the pinion, so the 23 information was available.
24 MR. REED: But you know, paper gets lost, 25 there's so much of it around the nuclear plant.
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coversee 800 336 6646
c
~G;24594.0 56 BRT ..
%.s
(~); i. 1 Instruction books ^and so forth are easily lost, and so
- Y i' s 2' forth.
.- 3- I'm not forgiving the utility for putting it c, 4 ,
wrong, but supposelyou used something even like -- color 5 d
'ce'ing. You are.saying it's a major problem changing the 6c ge'ar so it could 't be put in backwards. Is there such a 7 ~ thing -- we do color coding in cabling and all kinds of 8 things all the time to try to keep things separated. Is 9 there something in, let's-say identification, that would be 10 important? Coding -- something more administrative that 11 would help, if it's just this horrendous design change?
12 MR. MICHELSON: Like a label that says "this 13 side out," for instance.
3T O i 14 . MR. EBERSOLE: Or a breast tab that says 15 " Warning: I have a booby trap in this design. Hereby be 16 noted." -
t 17 MR. WARSING: That's an excellent point. We s-' l'8 really hadn't thought about doing that type of thing but 19 that's certainly something,we could do.
-.20 7 MR. MICHELSON: You were depending on relatively s ~ 21 , - skilled mechanics _who know what they were doing, to
. 4 22 . assemble and disassemble components.
,y-23 MR. WARSU!G:+ Yes, sir. In my experience going 24 .to nuclear plants, everything has to be' do,ne by procedure.
25 If it is properly done with th instructions we provide, it b,'
s_
I v
l ACE-FEDERAL REPORTERS, INC.
! 202 347 3700 Nationwide Coversee 800 33HM6
-24594.0 57 BRT
(,) 1 shouldn't occur. I agree color coding would certainly be 2 of value. It's a good point.
3 MR. EBERSOLE: The pride of some mechanics is to 4 be able to do the thing without reading it.
5 MR. MICHELSON: Do you remember, Glen, you keep 6 telling us we don't want too many procedures and 7 instructions for the mechanics.
8 MR. REED: Paper gets lost. In fact one of the 9 more serious accidents in the industry, or near accidents, 10 resulted from the turning of two pages of a manual at one 11 time.
12 I want to point out I think in high tech, in 13 nuclear, in the trend of technology and science and
.)
14 engineering, there is -- there should be an ever-increasing 15 burden on the designer to create his device, be it an 16 automobile or anything else, and there are laws now coming 17 in on automobiles, to create his device such that maybe 18 even the average person can drive it.
19 I won't comment on a particular car that the 20 average person could not drive.
21 It seems to me in all the product business, the 22 world in which we are moving into, the designer should 23 become very objective with respect to these field-reports 24 you are talking about, in saying: Did I make the valve 25 stem too short?
(O
\_)
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 80 M 36.# 46
24594.0 58 BRT
/~N i,. j 1 I really believe'that there are, in the long
[
2 run, these improvements bring credit on the company that 3 does . them, and on the reliability of his product. So I 4 really think you ought to think long and hard at Limitorque 5 about these kind of things that can booby trap the 6 maintenance worker.
7 You know, we keep harping on training, training, 8 training, and we are going to make superbeings in nuclear 9 plants. I don't believe that .
1 10 After all, they live in a society of much less 11 than superbeings and they go home at night to a society 12 that says "do your own thing."
13 Designers have .a tremendous responsibility to O 14 concentrate on creating a product that is not all that 15 difficult to drive, or maintain.
I 16 Thank you.
17 MR. MICHELSON: I would like to pursue with you 18 for a moment the importance of Limitorque becoming aware of 19 the field problems and field failures, whatever.
20 To what extent do you, for instance, routinely 21 attempt to obtain licensee event reports that relate to 22 valves? Do you make any effort to look at LERs, for 23 instance, that are issued by the utilities that relate to 24- valve failures?
f 25 MR. WARSING: We used to be on the mailing list h ,
. (~J L
' ACE. FEDERAL REPORTERS, INC.
, 202-347 3700 Nationwide Coverage 800 3 4 4646
24594.0 59 BRT:
( l for the LERs, and we were taken off and we tried to get 2 back.on, which we were not successful at. At this point we 3 do not get the LERs.
4 Also, as a matter of fact it's somewhat a 5 curious point, a lot of people, we get a lot of calls from 6 . utilities, frequently, asking for information or our 7 comments on different reports from INPO, EPRI, and various 8 other institutions within the' industry. And because we are 9 not a utility, we are a manufacturer, we can't be copied on 10 anything. So we never have the information to answer
.11 . peoples' questions.
12 MR. MICHELSON: This, I think, is a safety issue.
1 13 I believe the supplier of the component has certain O 14 obligations under, say part 21. And he somehow has to 15 maintain or be aware of what's going on and how it relates 16 to his product.
17 You -- are you aware that there's a so-called 18 compilation of LARs that's issued monthly, that has all the 19 LERs for the month in it, and -- in short abstract form?
20 MR. WARSING: I have seen copies of it, yes.
I 21 MR. MICHELSON: Doesn't your company get those?
22 MR. WARSING: No, sir.
23 MR. MICHELSON: There's an AEOD representative 24 here. Is that compilation sent free or do you have to buy ;
25 it?
O ACE FEDERAL REPORTERS, INC.
202 347 3700 ' Nationwide Coverage 800 33HN6
24594.0- 60 BRT l h 1 MR. SEYFRIT: It's sent free to. utilities. I 2 don't believe that there are any suppliers on the routine 3 mailing list.
4 MR. MICHELSON: You send it to the utilities 5 because you want to make sure that they, indeed, see it.
6 MR..SEYFRIT: Right.
7 MR. MICHELSON: Sort of a legal obligation. But 8 you don't send it to anybody else?
9 MR. SEYFRIT: I believe I'm correct, Carl. As 10 you know, Fred Hebdon handles that, and I don't believe 11 that there are any suppliers on that routine mailing list.
12 MR. MICHELSON: I don't recollect either. It's 13 a nominally-priced publication, as I recall, a few dollars --
O 14 MR. SEYFRIT: It can be purchased from.the 15 printing office.
16 MR. MICHELSON: On a subscription basis. It 17 would be very valuable I think for your company to buy a 18 subscription to it. It comes out routinely, then, to you.
19 It contains enough information on which you would then 20 solicit more information on a particular LER, if you wish.
21 I think we are willing to search with any 22 manufacturer who is willing to search for a particular kind 23 of information presently --
24 MR. SEYFRIT: Oh, yes. We do searches.
25 MR. MICHELSON: You do those right now free of C
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverase 8(53M-M
i 24594.0 61
.BRT ,
I) 1 charge, don't you?
2 MR. SEYFRIT: Yes.
3 MR. MICHELSON: In the future, maybe, when it 1
-4 gets to be a big - job --
5 MR. SEYFRIT: If we get too many requests, we'll 6 have to charge for it.
7 MR. MICHELSON: Right now, if Limitorque sees j 8 things they are worried about, they can ask for searches to 9 be done and AEOD will send you the searches and send you 10 the search results. So you need to contact them and work 11 with them. And you can get, right now, this information.
12 Yes?
13 MR. SEYFRIT: Carl, one point I think it would
( 14 be worth making, with the new LER reporting rule, you must 15 realize that it's not necessary to report every valve 16 failure. .
17 MR. MICHELSON: That's right. !
18 MR. SEYFRIT: We would only get those valve ,
19 failures that are involved in an event in which a complete 20 system is lost, or there are other --
21 MR. MICHELSON: I was going to get to that 22 because that was the next part of my question. So, from 23 the LER viewpoint, you can get the information. I think it 24 would be worthwhile for all major manuf acturers to be tied 25 in somehow, if they aren't already in the system, to be
(~h V
t ACE. FEDERAL REPORTERS, INC.
202 347 3700 Naionwide Coverses 800 344M4 L
'l 24594.0 62 BRT
() 1 aware. The simple way is to get a subscription to this 2 compilation. And then beyond that you make specific 3 information requests.
4 Yes?
5 MR. SEYFRIT: One more thing I might suggest, 6' that in terms of completeness of reporting, it might be 7 more worthwhile to contact INPO and get copies of the NPRDS 8 information. t 9 MR. MICHELSON: That's what I was leading up to.
10 Clearly the LER information is available to you.
11 The next step, though, is that the utilities are 12 also reporting, now, the single valve failures to the NPRDS 13 system. I was wondering, do you have any way of getting 14 NPRDS information. You are aware of what NPRDS is?
15 MR. WARSING: No, sir.
16 MR. MICHELSON: It's a data base. Nuclear plant 17 reliability data system.
18 MR. WARSING: Okay.
19 MR. MICHELSON: It's maintained by INPO. It 20 used to be maintained by the NRC but it's now maintained by 21 INPO. So your ability to get into it is going to be, 22 perhaps, nonexistent. I don't know. You have to talk to 23 INPO on that.
24 MR. REED: I wouldn't think it's nonexistent at 25 all. It's all a matter of finding the routine for signing ACE. FEDERAL REPORTERS, INC.
201447 3700 Nationwide Coverses soo 344H6 L
24594.0 63 BRT my 1 up and getting even a computer terminal.
2 MR. MICHELSON: As a nonmember of INPO, I'm not 3 sure that you can even sign up.
4 MR. REED: Oh, yes, manufacturers I am sure --
5 MR. WARSING: I know there was another 6 publication we tried to get and were told we couldn't get 7 on the list because we weren't a member.
8 MR. MICHELSON: This is a data base that you 9 call up and operate of f your own terminal. I'm not sure, 10 you'd have to pursue with INPO. I'll be very much 11 interested in finding out if you can get on it or not 12 because even NRC contractors in the past have had 73 13 difficulty. We are going to talk to one later. I'm going
~
14 to ask them again are you on the NPRD data base. They had 15 difficulty getting on it so it's not simple.
16 -
MR. EDERSOLE: Can you tell me wn> they have 17 this reticence, while they want to keep this highly 18 important information so close to their chest?
19 MR. MICHELSON: You have to ask them.
20 MR. SEYFRIT: I think there are a couple of 21 different kinds of reports that are put out by NPRDS.
22 They put out a summary kind of report which 23 includes all the failures and so forth but which does not 24 identify specific plants. That information, I think, can 25 be made available to the public without too much of a big 7
k l m
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwkle Coversse M3M
24594.0 64 BRT y 1 problem.
2 The difficulty is, when the work that they do 3 identifies specific plants.
4 MR. EBERSOLE: They don't want you to know who .
5 the culprits are, do they?
6 MR. SEYFRIT: That's correct.
7 MR. MICHELSON: Yes, Limitorque kind of needs to 8 know the plant we are talking about when they do their work.
9 It's not going to be so simple. But it's nothing I can do 10 more than say -- I was mainly wanting to know if you were 11 already doing this. As far as you know the company does 12 not have access to NPRDS?
13 MR. WARSING: That's correct.
7-
\ i 14 MR. MICHELSON: In addition, INPO produces 15 so-called SERs and SOERs. These are documents dealing with 16 significant events. Do you have access to any of that 17 information which is valve related?
18 MR. WARSING: No, sir.
19 MR. MICHELSON: I'm not surprised but I wanted 20 to ask.
21 So, you don't really get anything from INPO 22 presently, then, except indirectly, I guess, when a utility 23 comes to you they might give you something that they've 24 gotten from INPO.
25 MR. WARSING: Generally all the information we m
_.c ACE FEDERAL REPORTERS, INC.
202 347 3700 Nadonwide Coverage RX) 336 4M6
24594.0 65 BRT
()
I have gotten is from a third party.
2 MR. EBERSOLE: Are there any INPO 3 representatives here? I thought we might go to the heart 4 of this.
5 MR. MICHELSON: We tried once before and didn't 6 get to it.
7 MR. EBERSOLE: Let me --
8 MR. MICHELSON: Let me finish this one package.
9 As far as INPO is concerned, you are getting strictly third 10 party information.
11 MR. WARSING: Yes, sir.
12 MR. MICHELSON: The NRC produces large amounts
,_s 13 of valve-related information. Do you receive that or do
' ~ '
14 you have to be aware of it and buy it or ask for it?
15 MR. WARSING: You have to be aware of it to get 16 it.
17 MR. MICHELSON: How do you maintain an awareness 18 of NRC publications?
19 MR. WARSING: Again, it's always through third 20 parties.
21 MR. MICHELSON: I see. You don't have a direct 22 means. The third party brings you an NRC or laboratory 23 report or whatever, and then it's brought to your attention 24 and then you look at it?
25 MR. WARSING: The same thing applies, just to
,q G
ACE FEDERAL REPORTERS, INC.
202 347 3700 NationwWe coversee 2 336 4 646
1 24594.0 66 BRT
-! 1 make a point, to IE notices. We never get them directly
. 2 from the~NRC. We always get theirs through a third party.
3 MR. MICHELSON: Even if they were valve-related 4 on a. Limitorque operator, the NRC doesn't mail one to you 5 to bring it to your attention?
/
6 MR. WARSING: Not until very recently. There's 7 been a couple of them where, after they have issued them, 8 they have gone back and gotten a lot of comments on them, 9 and then come to us to get clarifications and issue an 10 amendment. Then we got copies of them.
11 MR. MICHELSON: So you are dependent upon the 12 licensees to bring to your attention whatever information g 13 they think you need to know.
14 MR. WARSING: That's correct. ,
15 MR. MICHELSON: Relating to their problem.
16 You of course sell to all the rest of the 17 industries, petroleum and whatever.
18 Do you maintain a data base? Or do you get 19 feedback of valve reliability and operability problems and' 7.0 so forth in some organized fashion or again happenstance?
21 MR. WARSING: All the data we get coming ther, ugh 22 our own field organization, and there is statistical data 23 we have been keeping for a year and half now on the entire 24 industry.
25 MR. MICHELSON: Do you have some sorts of O
ACE. FEDERAL REPORTERS, INC.
202 M1.M00 Nationwide Coverses aus344M4
n 24594.0 67 BRT
) 1 industry arrangement where they send this stuff to you 2 routinely to put into a data base?
3 MR. WARSING: It's only received when our field 4 organization is involved. If their own group does it and 5 doesn't report it to us, we have no way of knowing.
6 MR. MICHELSON: So you are depending on your 7 field engineers to compile this information for you?
8 MR. WARSING: Yes, sir.
9 MR. MICHELSON: And then you put it into an 10 electronic data base from which you can then retrieve the 11 information. Do you have any feel for how many records of 12 information are in that data base now?
-' 13 MR. WARSING: I'd have trouble quantifying it k~)3 now, sir.
14 15 MR. MICHELSON: Can it be characterized as a 16 very large one? Are there 10,000 ovents in it relating to 17 motor-operated valve problems? Or 100,0007 Or 10007 18 MR. WARSING: I would guess it would be more in 19 the order of 3- to 5000 at this point.
20 MR. WARD: Both general industry and nuclear 21 applications?
22 MR. WARSING: Yes, sir. It's all of the 23 industries that we service would be included in that data 24 base.
25 MR. WARD: Are there enough data there to --
p L) ~
l f
ACE FEDERAL REPORTERS, INC.
202 347.)700 Nationwide Coversee 800 3)MM4
i 24594.0 68 ,
BRT ;
i
< 1 have you reached any conclusion about whether the
- 2. performance of your valves -- valve operators is dif ferent ,
4 )
j 3 in'the nuclear industry than it is in the general industry?
I ,
, 4 MR. WARSING: Not to this point; no, sir.
- 5 MR. WARD: Will you look for that sort of thing? 1 6 Is that one of your interests?
5- 7 MR. WARSING: Our direction is more towards
! 8 product improvements and looking at what the weak links are, f l
l 9 where.we are having the most maintenance pi.oblems more than i
10 it is towards an analysis of the type ycu are referring to.
! 11 MR. MICHELSON: You are not oriented towards ,
j 12 reliability analysis, necessarily, are you?
13 MR. WARSING: Not in the sense that you are !
l O 14 referring to it, no. Product improvements are certainly a
i
- 15 for the purpose of improving reliability but not in the ,
i j 16 sense that you folks are discussing reliability analysis, ,
i :
17 no.
i 18 MR. WARD: It just seems you have a product
, 19 that's, you know, a fine, very popular -- I mean the ;
i !
20 marketplace, I think, has determined it's a very fine '
i -i 21 product for general industrial use. Obviously, you 22 dominate the market.
23 But, as Mr. Reed pointed out, the demands, at !
- t
, 24 least as we perceive them, the demands on this type of 25' equipment in a-nuclear plant are different than they are in O
i Acs.FeoaRAi, Raronians, INC. !
3st M7.M00 NationeWe Cove,ess 3804444W
24594.0 69 BRT I
) I the general industry, and it's not clear how the sane 2 product line can satisfy, you know, both types of service.
3 If you get all this information from LERs, and 4 NPRDS, and -- what are you going to do with it? What do 5 you see -- what commercial drive would there be toward a 6 modified product line, let's say,' for -- just for the 7 nucitar industry?
8 MR. WARSING: You are certainly correct in 9 saying that the application in a nuclear power plant is 10 very critical. No one would ever question that.
11 But what you are suggesting is that from our 12 position, that the people who own fossil-fired power plants 7S 13 or refineries don't find keeping their operations up and
\ )
14 running as equally critical to them; you know, it means a 15 lot of profit to them if a refinery goes down. Just like a 16 nuclear power plant goes down. So reliability is important 17 in our industry whether it's in the nuclear power industry 18 or whether it's in a refinery or a water reatment plant.
19 So, reliability is always an issue. We always 20 have attacked it from the standpoint of making product 21 improvements that improve the reliability and cut down the 22 incidence of breakdown. So I'm not sure that I see where 23 in that sense in torms of reliability nuclear power is 24 different.
25 It's more critical as a safety issue, but it's
(';
L)
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee 800 1)& 6444
c 24594.0 70 i BRT l C, 1 not necessarily more important.
2 MR. EBERSOLE: May I ask on that --
l
'3 MR. WARD: Well, wait, let me -- you know, as far as reliability and plant operability and that sort of I 4
5 thing, certainly they are similar demands. But I mean, you 6 know, the safety concerns at nuclear power plants are 7 dominated by the extremely rare, extremely high consequence 8 event.
9 You know, the normal standards of reliability, 10 at least in certain critical systems, just aren't good 11 enough in some cases. At least that's the perception we
, 12 hold, and sometimes I question it.
(
! 13 But, I mean, if you do get feedback from LERs l
O 14 and the NPRDS, that there are failures that are occurring 15 in nuclear plants which, perhaps, are redesign or some 16 change in equipment that could ef fect, but the rest of the l
17 industry is perfectly satisfied with the design, it's not 18 clear to me what's going to drive you to changing the l l
19 design in response to the nuclear ' experience.
I 20 MR. WARSING: I see your point. What drives us l 21 realistically is the marketplace.
22 MR. WARD: Yes.
23 MR. WARSING: If the nuclear power industry has 24 a perceived value in any design change that they feel would t i 25 benefit them, Limitorque would be glad to pursue it.
O :
l ace. FEDERAL REPORTERS, INC.
3sbM73700 Natiostw4de cow,ese NS))64N4
24594.0 71 ;
BRT
- pd 1 Generally, we haven't come across anything, yet, 2 that really has fit into that mold. More often the value ;
3 is not great enough to offset the problems with parts ,
i 4 interchangeability that the utility would be f aced with. l 5 But if there was a situation like that, !
6 certainly. I think maybe the one application that might 7 fit that is the qualification efforts towards IEEE382-80, ;
t 8 where we developed a new motor to meet those environmental 9 conditions.
10 Certainly, I think, that would fit the category .
11 you are talking about. i 12 MR. WARD: Yes. But there had to be a specific 13 requirement there and you were able to respond to it.
14 MR. WARS!NG: Yes.
15 MR. WARD: I think the problem with this sort of i
16 vague interest in, gee, let's make them more reliable, is 17 that you haven't been given any specific requirement.
}
18 If you are given some sort of new reliability 19 requirement and somebody in willing to pay for it, you'd i
20 work on a redesign or something to meet it.
21 MR. WARSING: Certainly.
22 MR. WARD: Okay. You don't have a target? ,
23 MR. WARSING: No, sir.
24 MR. EBERSOLE: I would like to ask you about, oh, l 25 just the nature of the valve as an integral system within a O
ACE.FauunAL Raronians, INC. I astm)* Naimw6deCewees 800-)M4644 '
24594.0 72 BRT f
V) 1 system rather than as a component. A valve, if you will, 2 is a system if you look at it by itself, it's an 3 electro-mechanical system.
4 What bothers me sometimes, as it does with 5 larger systems, I don't find anybody with what I call 6 parental responsibility for the whole range of valve 7 requirements, operating in a normal mode where it may just i e
8 be a sectionalizing valve with modest differentials, and if ,
9 it's an intercepter, maybe in the petrochemical industry, 10 I'm not sure, to avoid burning the whole plant if it has to [
11 do that, there's a union problem here in that you'll find 12 six people standing around one valve doing his little base 13 but there's no one with an overall responsibility except i 14 the operator who doesn't understand or have the guts anyway 15 to say whether the job has been done right. L 16 It was Carl himself who discovered in 1968 that 17 the 36-inch butterflies couldn't shut against the outflow 18 of a lot of coolant accident in all the reactors. There 19 was nothing in the specs that said they had to. They were 20 designed to be tight once they were shut but they weren't 21 designed to be shut, for gosh sakes. That initiated this 22 program all these years which I can't believe 'is going on, 23 can we intercept high flows in case of pipe breaks.
24 Whether this occurs in the petrochemical industry, I don't 25 know. I think it should. If I bust a pipe, I must not O
Acs.FEDERAi, RaronTans. INC.
24594.0 73 BRT s/ 1 burn down the whole damn petrochemical plant.
2 Could you comment on this sort of -- is there 3 some godfather, or parental interest in the valvo and its 4 integral function in your industry? Or is it a bunch of .
5 mechanical engineers? And before you say anything, one of 6 your valves in Browns Ferry sheared a 3-1/2-inch step. The 7 ongineer was to be damn sure he sealed the valvo, so ho put 8 in an extra 50 porcent. They hoped the valve would be 9 controlled by torque and limit switch, which it was not, so 10 I found there a lack of this parental responsibility. Do 11 you follow no?
12 HR. WARSING: I follow you completely. Thoto rm 13 are two things that happen in the industry that work
(_)
14 totally contrary to what -- to what you'd like to soo. And 15 the first is that wo supply the actuator to the valvo 16 manufacturer. Okay? This thoory, he is supplying a valvo 17 actuator package, valvo and actuator package to the utility 18 or the contractor, whomover. And ho is supposed to bo 19 taking responsibility for the package. Okay?
20 That tends to be a problem because he doesn't 21 know a whole lot about actuators, and unfortunately he 22 doesn't tell us a whole lot about his valvos and we don't 23 know a whole lot about his valves. And you combine that 24 with people who are writing specifications and 25 hypothesizing worst caso conditions, and we got very ACE. FEDERAL REPORTERS, INC.
_ _ 202 147.)?00 Natamekte Coversee Wn))4AM4 ___ _ _ _
24594.0 74 BRT !
(^'s
()_
1 conservativo and we got 80 percent voltage requirements and 2 10 second operating times and an 80 percent voltage 3 requirement produces, in an AC motor, in which you nood an 4 actuator that's capable of producing close to twice the 5 design requirement. And no one in the specifications says 6 you not only have to meet 80 percent but you have to bo 7 able to withstand a locked rotor condition at 110 porcent 8 voltage, so now you are looking at something on the order 9 of, round numbers, 225, 250 porcent of actual required 10 thrust values.
11 You havo got -- ossentially what you and up with 12 is a wholo bunch of ignorant people each looking at their
/
-s 13 part of it.
(s) 14 MR. EBERSOLE: Compartmentalization.
15 MR. WARSING: Yes. And the and result is 16 oxactly what you are talking about. You got a lot of bad 17 applications.
18 MR. EDERSOLE: Itow are we going to fix that? By 19 the way, that appilos to the whole nuclear business, wo 20 are too compartmontalized.
21 MR. WARSING: If I can sidostop that one, I 22 think you may be in a bottor position to answer that ono 23 than I am. I don't have a simple solution.
24 MR. EDERSOLE: Do you ask critical questions, 25 when you say you want me to put an operator on, the C'1 e
ACE FEDERAL REPORTERS, INC.
m )4 m oo Non.we coversee *n)Swa
24594.0 75 BRT 1 ultimate strength of your valve before it will disembowel 2 itself or sheer a shaft or whatovor? Because you can't 3 depend on the switches working all the time. I mean this 4 is just a minor aspect of a larger problem. But we have 5 cranos, you know, that can self-destruct.
6 MR. WARSING: We try to. The place we would 7 normally attack something like that is at the specification 8 writing stago. By the timo you got past that stage it is 9 really too late to generally correct the problem.
10 MR. EDERSOLE: I don't know of any other focal l 11 point to control this problem, other than your company or 12 one liko it has to ask very astuto questions about what's n 13 the nature of this thing wo are working on, what does it U 14 have to do under what conditions, and about the statistics 15 a while ago that you said you ero amassing. Is that 16 statistics under -- like a tablo, rod to green, or is it 17 really, in duress, doing its most critical thing? Are you 18 getting statistics of performanco in the noodod modo or aro 19 you just getting a blind not?
20 MR. WARSING: Wo don't have any ability in the 21 system that wo've put in to offect, to datormine whether or 22 not the system is in duross.
23 MR. EDERSOLE: I sometimos think a pony break 24 ought to be put on ovary valvo and overy timo you swing it, 25 you onergize it to make euro it will do what it has to do.
(3 t'
ace. FEDERAL REPORTERS, INC, m.m.nm 8. u.m com mmm ___
24594.0 76 BRT
() 1 MR. WAdSING: I'm not sure I follow you, sir.
2 MR. EBERSOLE: A friction drag or something, 3 that makos sure it will do what it has to do. A synthetic 4 load. But of course that brings the problem, maybe someone 5 will forget to take it off. All things come with a prico.
6 We are in a tank gol.
7 MR. REED: Carl?
8 MR. MICHELSON: Yos. Go ahead.
9 MR. REED: Early on you said something about, 10 words something to this offect: Little or no maintenanco 11 offort is required. You woro montioning you have been in 12 the valvo operator business for many years -- and I guess
- 13 that's probably true. In the other applications, little or 14 no maintenanco effort in required and probably Limitorquos 15 sat thero for 15, 20 years and the grease doesn't got 16 changed. They are going to got tended to and the torquo ,
17 switches probably don't got looked at and so on and so 18 forth.
19 Iloweve r , in the nucivar there suoms to be, we 20 scom to koop putting responsibility on the maintenance 21 crows and operators to drive'these things.
22 In the nuclear field more and more the drums aro 23 beating, more and more maintenance is required and even 24 analysis may be required. Aro you familiar with the MOVAT 25 system? Analyze Limitorque performance? '
O LJ i
ACE. FEDERAL REPORTERS, INC.
20M474700 Nationw6de Cowwsee 800414AM4
24594.0 77 BRT r^w
(_) 1 MR. WARSING: I am aware of it. I'm not that
- well-versed with it. I have soon a prosentation of MOVATS 3 and what it does. That's been my exposure to it in its 4 entirety.
5 MR. REED: Then maybe it's unfair to ask you but 6 I was wondering if you thought it had any application that 7 your company might considor some abbreviated version of 8 that as a built-in aspect to your motor operators.
9 MR. WARSING: Yes, sir, I think it has some 10 value, particularly in checking that the switch sottings 11 are proper. I think that's its primary value.
12 Past that I'm not suro -- I don't know the 13 product well enough to mako any comments on it that I fool
-)
14 confident in.
15 As far as Limitorque's interest in this aroa, 16 yes, wo would be willing to, to look into nomething liko 17 that. We have boon having some discussions about 18 developing that kind of a program. There's nothing, really, 19 in the works at the moment.
20 MR. REED: You know, ovorything that's designed, 21 I'm sure, can be improvod. I sort of take the opinion that 22 the automobile took about 80 to 90 years to reach a fair 23 optimization and I would expect that valvo operators would 24 tako many years to reach optimization for all the duties 25 that they might be called upon.
O
_j ACE. FEDERAL REPORTERS, INC.
202.)47 3700 Nationwide Courece MG))HM4
I l
24594.0 78 !
BRT
- p. s
(,) 1 MR. MICHELSON: Limitorques have been around for 2 many years. How many years, roughly?
3 MR. WARSING: About 50.
. 4 MR. MICHELSON: About 50.
5 MR. REED: Maybe it takes longer to optimize a 6 valve operator.
7 MR. MICHELSON: I would like to ask a couple of 8 follow-up questions relating to both what Jesse and what 9 Glen said. As long as we are on MOVATS, let me follow up a 10 little bit, there.
11 I had envisioned, of course, that it would be 12 nico if there were an extra electrical jack on a Limitorque
,3 13 operator that I could plug my little test device into and i !
'~'
14 it would toll me all thoso good things and I could watch it 15 when I did the surveillance on the valve. I guess from 16 what you are saying -- I was going to ask and I think 17 you've already answered, though -- presently you don't have 18 any R&D kind of activity, trying to develop such a devico 19 that could just be plugged right into your Limitorque 20 operator and it would give you the key information that 21 would assure that sottings are proper?
22 MR. WARSING: At this point, no, there really 23 isn't an ongoing effort. As I said, it's really in the 24 discussion stagoc as to what type of product we'd like to 25 develop in that area.
p '
(
ACE. FEDERAL REPORTERS, INC.
202 147 3700 Nathanwkle Coversee 800-))&AM4
24594.0 79 BRT
) 1 MR. MICHELSON: You do have an R&D laboratory of 2 some sort for product improvements though, I guess?
3 MR. WARSING: Yes, sir.
4 MR. MICHELSON: This would be a logical thing '
~
5 for them to do if they were so instructed by the company. ;
6 But presently, then, there's nothing to your knowledge?
7 MR. EBERSOLE: Carl, may I follow up on this 8 line? ,.
9 MR. MICHELSON: Fine. n I
10 MR. EBERSOLE: I want to follow up on this 11 design. You are talking about design.
12 It seems to me about 100 years ago the valve I 13 admired the most was the one that closed-the back end of 14 the 16-inch artillery piece so the operator would live when 15 it went off. Does it make sense to have the seats in liau 16 of what was used in that? What they had of course was some 17 sort of tough elastomer which would energize with pressure, !
18 it was an automatic servo, and it -didn't leak. I have 19 never really understood fully why stellate seats and torque 20 switches with virtually no motion allowed to come in on a 21 torque setting with no Bellville washers or anything to 22 give it a little degree of relaxation, why they made much o
23 sense. Do you all look at things like this? ;
24 MR. WARSING: That would really come under the 25 umbrella of a valve manufacturer. I wouldn't pretend to bo O
ACE. FEDERAL REPORTERS, INC.
202 347 3W10 Nationokte Coversee 2 336 4 646
24594.0 80 BRT
() 1 a valve expert.
2 MR. EBERSOLE: Oh, you are Limitorque. Carl, 3 we've got to talk with the valve folks.
4 MR. MICHELSON: We already have. You just
-5 didn't ask that question.
6 Let's go back, now, onto the question of, 7 essentially, matching an operator to a valve. The valve 8 vendor is the person who originates these orders generally 9 and they send an order to Limitorque, specifying certain 10 things.
11 Do the valve vendors of tentimes just buy so many 12 of this size operator and that size operator without even 13 telling you what valve it might go on? Or do they always O 14 tell you the valve application as a part of ordering the 15 operator?
16 MR. WARSING: In today's marketplace, no, they 17 tell us what valve they are putting it on.
18 If you go back not even 10 years ago, getting 19 that information was difficult.
20 MR. MICHELSCN: They just ordered so many 21 operators of various sizes?
22 MR. WARSING: Yes. Pretty much.
23 MR. MICHELSON: But now they do tell you the 24 application. Do they send you a copy of the purchasers' 25 specification or something? Or their own in-house
(~h L
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverses 800 336 4 646
t 24594.0 81
'BRT f
p)x_ 1 specification for that valve?
5 2 MR. WARSING: Usually nowadays, yes.
3 MR. MICHELSON: And that specification, includes 4 the dynamic loadings, if any, that might be. listed so you 5 are aware of them?-
6 MR. WERSING: Generally we get a peak load i 7 condition; whether it be an unseating requirement or a 8 dynamic condition or whatever. We are normally not given a 9 group of different conditions to operate under. Normally 10 just whatever, what is considered by the valve manufacturer l 11 the worst case condition.
12 MR. MICHELSON: Does the valve manufacturer ask 13 you for a recommendation as to what size operator it should 14 be?
15 MR. WARSING: That varies with valve 16 manufacturers. Some of them select their own. Some ask us 17 to do it.
18 MR. MICHELSON: So some of them do their own 19 calculations of thrust requirements or whatever and figure 20 out what size operator is needed by their own judgment and 21 they just happen to tell you about the valve but they 22 aren't asking you to look at the design, they are just 23 asking you to supply them an SBM-000 or whatever?
24 MR. WARSING: In some cases, yes. One point you 25 ought to be aware of in that area is that -- this goes back g-V ACE FEDERAL REPORTERS, INC.
, 202 347 3700 Nationwide Coverage 233H646
24594.0 82 BRT.
rm
(_) I to our comment earlier about the ignorance in the industry --
2 there are a lot of people, I've seen quito a number of 3 examples, where either the valve manufacturer has iG 4 indiscriminately taken an actuator and put it on a valve or 5 out in the field, a utility company says an SMB-000 is an s
6 SMB-000; it's off the shelf, it's like a box of Wheaties:
7 They are all alike. So we take it off of here and put it 8 on over here, that doesn't workt each' actuator, if done 9 correctly, is selected for the given valve application and 10 you just don't take it off of one valve and put it on
' l l' another.
'. 12 MR. MICHELSON: What do you do differently?
)
13 What's the difference between the two? I thought SMB-000s O 14 werec all alike. What do you do differently?
15 MR. WARSING: In selecting a valve ~ actuator, 1 2 16 there are a number of selections you have to make. Primary
. /,'j 17 I ones are you have to select the correct ratio, mechanical i\
18 ' ratio in the gearing. You have to select the correct motor I
19 size, and the correct torque spring assembly, so that the .
20 torque switch can be set in there properly.
21 That's probably the key ingredient.
22 MR. MICHELSON: So I can have two operators with 23 the same model designation that will have different
? t ' t'
,? 24 horsepower motors and different gear ratios and so forth?
25 MR. WARSING: Yes. -
I ts i
)
ACE. FEDERAL REPORTERS, INC.
202 347 3700 ' Nationwide Coversee 8043364M6
- -, . = .
24594.0 83 BRT
,~
(). 1 MR. MICHELSON: So you have to know more than 2 just that it's an SBM-000 or 00 or whatever.
3- MR. WARSING: Yes, sir.
4 MR. MICHELSON: That is interesting to know.
5 You supply a manual with each valve and that manual tells 6 the utility or the user exactly what gear ratio is in that 7 particular valve, then? That particular operator?
8 MR. WARSING: No, sir. We do not.
9 MR. MICHELSON: .How does the utility know that 10 he's got the right operator? Or if he wants to change one 11 out, what he has to change it -- what the exchange operator 12 should look like?
13 MR. WARSING: What we normally ask.them to do is 14 contact us with the nameplate information off the actuator 15 and we can make recommendations to him from there.
- 16 MR. MICHELSON
- From the nameplate of course you
-17 know the exact specification.
18 MR. WARSING: . We can retrieve it. But unless he 19- specifically asks the data be supplied to him, it's not 20 supplied as a matter of course.
21 MR. .MICHELSON: Once the operator has been wed 22 to a valve and the assembly has been shipped to a user, how,
~
23- does the user know what torque switch settings to set it up
~
24 for? Because the gear ratio is different, the horsepower --
25' you-have to have the exact torque switch setting --
()
ACE-FEDERAL REPORTERS, INC.
l 202 347 3700 Nationwide Coverage - 800 33H646
L Y.
.. '24594'.0 84
':BRT
.g
\_/ y 1 MR. WARSING: The-torque switch is factory set
/2 and it's limited to a maximum setti g.
3 MR. MICHELSON: The valv'e is set, not Limitorque. '
4 MR. WARSING: Set at Limitorque. Normally the
.5 valve manufacturer will typically do ,a hydro test and 6 adjust it as he feels necessary at a time.
7 MR. MICHELSON: Based on ability to open against 8 full pressure or something? .
9 MR. WARSING: Whatever he establishes is the r t 10 criteria for the valve.
11 MR. MICHELSON: It's up to tde valve vendor then 12 to tell the utility what torque settings should be used in 3
13 . operation.
s ,
14 MR. WARSING: Yes.
15 MR. MICHELSON: The little problem with torque 16 switch settings, apparently, is getting proper balancing of -
t zl7 .thb* torque' switch.ftself. That's. shipped from the factory, 18 ,I guess, prebalanced? ,
, '19 MR. WARSING: Yes, sir.
20 MR. MICHELSON: If nobody has diddled with it
~
21 along the way, it should still be in balance at the~ time of
- 22. . Instaliddion; it isn't something that's easy to jar loose?
e)-
23- MR. WARSING:' No, there are hex nuts that lock 24 it in position once it's balanced.
.25 MR.'MICHELSON: When we find these things are l.
t ._
r.
.i .. l; Ack: FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 33H646
. .~- - . . . .
24594.0 85 BRT
- ) l- way out of balance,.one or two possibilities
- Either 2 Limitorque didn't get it properly balanced to begin with or 3 somewhere along the_way somebody screwed it up. Because, 4 apparently,.you can get factors of two or three, in unbalance, 1
5 -and get to where these adjustments are that far off.
6 Numbers coming out of Davis-Besse for their 7 measurements are indicating some valves that far off. I
, 8 was somewhat surprised.
9 MR. WARSING: If they are that far off, my guess 10 is that it's probably a particular type of torque switch
- 11 that we use that is -- its adjustment can be damaged tar 12 handling. The balancing can be damaged by handling.
t 13 To be specific, it's the SMB-000 torque switch.
. Os 14 If you handle it incorrectly, you can affect its adjustment.
15 .MR. MICHELSON: By handling, you mean if you 16- take the cover plate off and_ work on the valve? That kind
- 17 of handling?
'18 MR. WARSING: Handling the torque switch. There 19 are two contact vendors that can be bent, effectively, and ,
20 if they are bent it can affect the adjustment of the double
- 21. O enlarger-units. In that unit, once it's balanced it has 22 a positive look.on it and doesn't affect the balance.
23' MR. MICHELSON: Socyou would only expect this to 24 appear in the triple 0 units?-
- 25. MR. WARSING: Yes, sir.
()
l - ACE-FEDERAL REPORTERS,~INC. '
202-347-3700 Nationwide Coverage 2 336 4646
__ _ - _ . . -. , , ~.
24594.0 86 BRT
() 1 MR. MICHELSON: That's a worrisome issue, if 2 they are that far off. If they have been damaged that way, 3 can they be rebalanced to zero without putting a new torque 4 switch in? ,
5 MR. WARSING: They can be, but you'd r,eed a 6- fixture to do it properly. That's what we do in the 7 factory.
8 MR. MICHELSON: But you have to bend something 9 back in position?
10 MR. WARSING: Yes.
11 MR. MICHELSON: And then you have to rebalance 12 the' system, but that's an acceptable technique; there's no 13' reason to worry about the future use of that switch, is
(
14 there?
15 MR. WARSING: No, sir.
16 MR. MICHELSON: There are other torque switches 17 that have turned out to be off.by a factor of 1.2 to 1.5; I 18 guess that's the kind of range that you could get into just 19 from people maladjusting. But when you see a factor of 2 20 to 3, you are saying that's got to be, probably, something 21 more than just maladjustment?
- 22. MR. WARSING: Yes, sir.
23 MR. EBERSOLE: Could you tell me in the ritual 24 of maintenance -- you know, like in elevators -- is there a 25 seal or other thing affixed to the valve that reflects the r
.(O
_/
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3364646
24594.0 87 BRT
(_/ 1 last opening and handling of the valve and provides a 2 permanent record of who did what?
3 MR. WARSING: No, sir.
4 MR. EBERSOLE: There's none of that discipline 5 in this business?
6 MR. WARSING: No, sir.
7 MR. EBERSOLE: So you never know if some jerk 8 hasn't just taken it off and fiddled with it?
9 MR. WARSING: That's correct.
10 MR. EBERSOLE: That's very poor.
11 MR. MICHELSON: Go ahead.
12 MR. WARSING: I was going to say I can see your
,g 13 point of it being very poor.- From our perspective, the
\)
14 torque switch has been used for years as the cure-all to 15 all sorts of problems: When in doubt, readjust the torque 16 switch and see if it will work.
17 If you look at it from that perspective, there 18 was never any -- you never really wanted to have a seal of 19 some type that said: Hey, we just adjusted the torque 20 switch. In the industry as we are talking about it today,
-21 I see your point; but looking at it from an historical 22 perspective, this was good reason for it being the way it 23 is.
24 MR.,EBERSOLE: Fossil plants, right, the usual 25 thing. Go in and play with it. I understand. We are O
s_J l
l l
ACE-FEDERAL REPORTERS, INC.
- 202-347-3700 Nationwide Coverage 800 33(Hl646
24594.0 88 BRT
( l still playing with it.
2 MR. WARSING: Yes, sir.
3 MR. EBERSOLE: We have never really adjusted to 4 this business in the utility industry.
5 MR. MICHELSON: In the design of your Limitorque 6 operator, have you adjusted all of the parameters of design 7 such that you could turn it to the maximum torque switch 8 setting, all the way up, without causing mechanical damage 9 to the operator itself?
10 MR. WARSING: Yes, sir ~.
l 11 1 MR. MICHELSON: That is a basis.
12 I noticed in your instruction manual, a couple f- 13 of places it said something about don't turn the torque -
u) 14 switch above a certain set, ting because it might cause 15 mechanical damage. It was never clear to me what that 16 could be. I thought you designed, mechanically, your 17 system, so that thero could be no damage other than, of 18 course if you went to a stalled rotor condition, you'd burn 19 a motor up, obviously.
20 MR. WARSING: There's a component placed on the 21 torque switch assembly, we call it a torque limitor-plate.
22 It is a mechanical piece that is installed to prevent you 23 -from setting the torque switch above its rating, above the ,
24 unit's rating or --
25 MR. MICHELSON: Is that limitor plate adjustable
($)
ACE-FEDERAL REPORTERS, INC.
202 347-3700 - NationwWe Coverage M3%4646
24594.0. 89
.BRT C 1 er eixed2 2 MR. WARSING: It 's pe rmanent . Well, it can be 3 removed. . It is held on with a screw. Someone can remove 4 it -- -
5 MR. MICHELSON: Anything held on with a screw is 6 adjustable; you could take it off.
7 MR. WARSING: It is certainly not meant to take
- 8 it off. You could.
9 MR. MICHELSON: You are saying if somebody did 10 take the limitor plate off and turned the torque switch up 11 higher, I guess it would be beyond its last number already?
I 12 or it could even be numbers on the dial yet, I guess?
- 13 MR. WARSING
- There are numbers on the dial. We 14 have limitor plates, the numbers are sequenced from 1 to 5.
15 MR. MICHELSON: If someone remcres a limitor 16 plate, and I weren't too smart, I would think I could turn 17 it on up as long as I was within the numbers on the dial?
18 MR. WARSING: Yes, sir.
19 MR. MICHELSON: If I did, now I could get into a 20 range where it would damage the operator mechanically and 21 not just burn the motor?
22 MR. WARSING: Yes.
23 MR. EBERSOLE: What's the responsibility; 24 shouldn't it be motor burnout and then put the 25 responsibility on the electrical --
(T
%-l i
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 33H446
24594.0 90 BRT
. /m Is_) -1 MR. WARSING: There's a dual criteria: The 2 first is the unit's actual rating. The second is to limit 3 you so you can't set it at a setting high enough to cause a 4 locked rotor condition on the motor.
5- MR. EBERSOLE: Suppose I put a wrench on the 6 spine shaft that comes out of your thing and I lock it up 7 and I have it in this maximum setting. What happens?
8 MR. WARSING: The torque switch will trip.
- 9 MR. EBERSOLE
- If it doesn't trip, then what 10 happens?
11 MR. WARSING: Then it locks.
12 MR. EBERSOLE: Then I must rely on electrical 13 overload?
'~'
14 MR. WARSING:- That's right.
15 MR.~ EBERSOLE: Do you ever define those loads in 16 any way?
'17 MR. WARSING: We have a procedure on how to size 18 overloads.
19 MR. MICHELSON: Is it safe to assume that there 20 will be no mechanical damage when you reach locked rotor 21 conditions? Is that a designed basis or could you already 22 cause mechanical damage, keeping in mind I have removed 23 that limitor plate now and I turn it on up and lock the 24 rotor. Could I get mechanical damage before the motor was 25 locked?
.(/
. ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide coverese m336 4646
.24594.0 91 BRT
( l MR. WARSING: Yes, sir, you can. Because 2 although we publish data that tells you what the maximum 3 overload capabilities of the unit are and we publish data 4 that tells you, with this particular motor size you don't 5 use a gear ratio above this, et cetera -- so the data is 6 all available.-- unfortunately those criteria are not 7 consistent with the 80 percent voltage criteria, so you 8 frequently have actuators which can't handle their own 9 locked rotor condition.
10 MR. MICHELSON: I see. That's interesting.
11 MR. EBERSOLE: I wonder if the electrical 12 . engineer who designs the circuit breaker knows you have 13 this gadget inside that limits torque. I presume he does, O 14 at the upper end.
15 MR. WARSING: He wouldn't size the circuit 16 breaker --
17 MR. EBERSOLE: There's an NRC standard for 18 setting these overloads, is there? Carl? Is there an NRC 19 standard for setting the valve overloads --
20 MR. MICHELSON: You have to bypass the thermal 21 overload by requirement and you have to size the wiring, 22 normally for locked rotor current. But I don't know of any 23 requirement in between, concerning the mechanical integrity 24 of the operator itself.
25 MR. EBERSOLE: One rational always is I will I T v'
i f ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646
24594.0- 92 BRT 1 ,) -1 destroy the valve to obtain the function. The second is 2 I won't quite. I'll let it get near burnout and save it so 3 I could use it later, depending on where you are.
4 MR. MICHELSON: Then you have to put in some 5 kind of magnetic overload device at four times instead of 6 six times -- locked rotor, wherever the ' limits are on the 7 mechanical damage. Some utilities bypass the torque switch, 8 for instance, and don't have torque protection.
9 Now, if you hang up at any point, your limitor-10 is locked rotor current, but you are telling re if I bypass 11 the torque, I bypass that little limiting plate then, too.
12 It's all gone.
13 MR. WARSING: You have defeated the function, 14 sure.
15 MR. MICHELSON: You are telling me now I might 16 have mechanical damage. What would be the likely l 17 mechanical damage? Is there a weak point that generally 18 fails first?
19 MR. WARSING: In the nuclear power industry, 20 gate and globe valves are predominant, in which case we 21 usually see that the thrust loads become excessive and you 22 break housings and housing covers. That's not uncommon in 23 the industry. ,
24 MR. MICHELSON: We are talking now just about 25 the operator,-of course. What breaks in the housing?
l I'T s/
l l
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336 6646
_______1___
1 l
'24594.0 93 BRT.
I s
(_) 1- MR. WARSING: The main casting.
2 MR. MICHELSON: You mean the thrust loads are so 3 great it just tears its support apart?
4 MR. EBERSOLE: Unless the valve comes apart.
5 MR. MICHELSON: I'm trying to localize it to the 6 operator for a moment.
7 MR. WARSING: You'll see loads in some of these 8 applications, four and five times the unit's rating, when 9 it locks rotor at 100 percent voltage. The units are all 10 capable of withstanding 2-1/2 times their rating and thrust, 11 but it's not uncommon to see those loads exceeded.
12 MR. EBERSOLE: Wouldn't it be prudent to put a 73 13 limiting sheer pin in before that total disintegration V 14 takes place?
15- MR. WARSING: I don't know where we'd put a 16 sheer sheer pin in the system. We have done work with 17 sheer pins before. For the type of accuracy you are 18 talking about, they are much too crude a device for what 19 you want to accomplish.
20 MR. EBERSOLE: Step plates would be better.
21 MR. WARSING: There is another option that we 22 discussed, but we haven't gotten a whole lot of interest 23 from the industry and that is we have a design, we know how 24 to design an actuator that controls output thrust instead 25 of output torque, which on a gate and globe valve O
G ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage M 33 H 646
24594.0 94 BRT
()
q I application could be very beneficial.
2 MR. MICHELSON: In your manual you have a i
3 recommended wiring diagram, or typical -- I think you use 4 the word " typical," or. something -- this is one way to wire 5 it up. Of course people, for various reasons, have wired 6 them up a number of other ways.
7 Once a utility decides to -- or architect 8 engineer decides how he's going to wire the operator in, 9 you aren't even aware of what he has decided, are you?
10 Does he come back to you, if he wants to do it differently .
11 than in the manual?
12 MR. WARSING: The normal process is that during 13 the initial purchase, that we submit drawings to him, based O 14 on information we have been given, what we think he wants l 15 wired, the way we believe he wants it wired. If he doesn't 16 agree.with that, he'll mark up our drawings and send them 17 back and we'll wire it the way he wants.
18 I have found that frequently there-is a stancard 19 wiring requirement in the specification, and everything is 20 wired that way. After it is out and installed in the 21 plants, then the systems people take over and all the 22 wiring gets changed.
23 MR. MICHELSON: That's traditionally what has 24 happened, I think. Of course at that point you are not
_25 .even aware of what they've done..
k ACE-FEDERAL REPORTERS, INC.
- 2o2-347 370o Nationwide covwase 800 3 4 6646
24594.0 95 ,
BRT
(~%
(_) 1 MR. WARSING: No, sir.
'2 MR. MICHELSON: If'they wire it such that they 3 don't-provide what are your recommanded protective actions, 4 like torque switch protection, then I guess they are off on 5 their own when the operator breaks apart. ,
9 MR. WARSING: Yes, sir.
7 MR. MICHELSON: That's your approach. Because b
8 you can't 'be policing how people wire things up.
9 MR. WARSING: That's right.
10 MR. MICHELSON: But there's nothing that says 11 that Limitorque ever reviewed the wiring -- nothing saying 12 you are even aware of it?
13 MR. WARSING: The only thing we have in our O 14
/
records wculd be that it was-inspected prior to shipment, 15 to a particular wiring diagram, our own Limitorque wiring
.16 diagram, so we know how it leaves the factory.
17 MR. MICHELSON: Yes.
18 This is your chance to ask Limitorque anything 19 you wish, I think.
20 MR. REED: Technical bulletins. You get the 21 field reports and make an analysis of something that 22 happened someplace, a particular model -- Limitorque 23 operator. Do you publish, let's say to people you think 24 have these valves, technical bulletins such as some 25 manufacturers do on their equipment, advising them of some e
(_n)
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336 6646 t _.
l 24594.0 96 BRT
/m
'( ) 1 anomaly or arrangement or some problem that might be in 2 there with their equipment?
3 MR. WARSING: We have done that. Normally 4 that's also accompanied by a notification to the NRC. We 5 have sent out some maintenance advisories of different 6 types to make sure that -- where we find there's a 7 maintenance problem, that there is widespread ignorance on 8 them, we try to make people aware of it.
9 MR. EBERSOLE: Are the motors you use on these 10 things -- they are for very short-term duty, aren't they?
11 MR. WARSING: Yes, sir.
12 MR. EBERSOLE: So they are light, they have 13 virtually no thermal storage capacity. You have to
() 14 interrupt current on them pretty quick or you'll burn them 15 up. Do the people who design these circuits get all this 16 information from you to protect the motor if they want to 17 . protect it? Not that they always want to.
f 18 MR. WARSING: Yes, sir. The data is available, 19 and generally there's a requirement that we submit motor 20 design curves, which gives them all of' the information on 21 the motor's thermal capabilities.
22- MR. EBERSOLE: Right.
23 Is there a cyclic duty limit? You can't operate 24 them any more than X times in X minutes?
25 MR. WARSING: The thermal rating is really done --
i LO ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nadonwide Coverage M3366646
24594.0' 97 BRT 7,
( l gosh, how do I explain this? Our motors, the typical 2 rating we use on an'AC motor is what we call a 20 percent 3 run, 15 minute duty motor.
4 What that means is that at 20 percent of .
5 startin,g torque, it can run continuously for 15 minutes.
6 The valve running load condition is generally between 20 to 7 40 percent run condition , and , typically your 30 to 60 seconds 8 operating time is on the high end. If you have a rapid 9 closure you may be at 10-second operating time. So that 10 generally you'd have to run a lot of cycles very quickly to 11 exceed the thermal capabilities, but those temperature rise 12 curves are a part of the information provided.
13 MR. EBERSOLE: It gets to the customer.
O 14 MR. WARSING: It gives him the allowable time at 15 locked rotor also.
16 MR. EBERSOLE: It's a possible notion that DC 17 valves should be kept out of hostile environments in 18 preference to three-phase AC valves. Is there any basis 19 for that?
20 MR. WARSING: The qualifications that we have 21 done on the RDC units are not as stringent or don't have 22 near as demanding an environment as the AC motor 23 qualifications have.
24 MR. EBERSOLE: So it's important to keep them 25 outdoors instead of inside of containments. Is that ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coversee 8004364H6
.24594.0 98 BRT
-() 1 because it's commutated?
2 MR. WARSING: Yes.
3 MR. EBERSOLE: In any case, aren't they 4 ' weather-tight or leak-tight?
5 MR. WARSING: No, sir. In fact, they are 6 definitely not. Our qualification has always been on the 7 basis that we will not try to prevent the atmosphere from 8 entering because we don't think that's practical in the 9 real world, to expect the equipment be maintained in that 10 condition. So wo allow the environment to enter. We 11 actually provide what we call motor T drains, which are 12 plugs that allow the moisture to drain back out of the 13 motors.
b
'- 14 MR. EBERSOLE: AC motors inside containment?
15 MR. WARSING: AC motors inside containment and 16 DC motors also.
17 MR. EBERSOLE: DC too? You don't can them?
18 MR. WARSING: No, sir.
19 MR. EBERSOLE: That's interesting.
20 MR. MICHELSON: Is that true of the torque 21 switch in the compar* ment too?
22 MR. WARSING: Those are allowed to breathe too.
23 MR. EBERSOLE: I can understand an AC motor 24 which has no internal switching unless it has some portion 25 of reduction switching, and I hope they don't.
p r ACE-FE 7ERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 33 H 646
24594.0 99 BRT
()
1 MR. WARSING: They don't.
2 MR. EBERSOLE: The motor will normally be rather 3 cool. compared to the environment I.C, which is going to be 4 steam. There will be virtually instantaneous condensation 5 on tho guts. Do you provide short circuit face to face 6 ground clearances that will tolerate that kind of thing?
7 I'm beginning now to know --
8 MR. WARSING: I see your point.
9 MR. EBERSOLE: It's going to be a condenser in 10 there because it's cold. It hasn't run yet. All you get 11 is a flash.
12 MR. WARSING: The clearances, really -- I don't 13 know that we do anything special on the clearances.
O 14 MR. EBERSOLE: In any case, they are -,
15 MR. WARSING: But they manage to go through the 16 environmental testing without any trouble.
17 MR. EBERSOLE: There's been some shady work 18 going on in the testing laboratories where they go ahead to 19 preheat the items tested because you don't get faults in 20 some of the items.
21 MR. MICHELSON: Or meet them very, very slowly 22 so you have equilibrium all the way up.
23 MR. EBERSOLE: It's immoral ( what?) what you 24 find. Do you do any of this in your own shop? Run it 25 through the real world, you know where it's a cold motor
- b J
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage MS344646
d 24594.0 100 i BRT~ {
(j' 1 and then you hit it with a Turkish bath and measure leakage !
2 curves?
r 3 MR. WARSING: I'm not a qualification expert. I I
4 may be getting a little out of my league but the - best I 5 -know about the test, the units were put in a chamber'and i 6 the first _ thing they do is they run the two transients. ;
?
7 There's no effort to bring the chamber up to temperature
- 8 ahead of time. !
? ?
j 9 MR. EBERSOLE: What I'm saying is simple. Do i
! 10' you happen to know any physical clearances on studies and i >
j 11 so forth on the terminal boards. By the way they pick up ;
i 12 dust, 16 of an inch on 440?
3 13 MR. WARSING: No . . Those are all -- we have done ,
j
(~ 14 all the testing on those. We have no problem with those at
- 15 all. [
16 MR. EBERSOLE: On the 440, it's usually wet and l 17 dirty -- well, that's an illuminating sort of. piece of [
t ;
18 knowledge, Carl. I didn't realize they were open. I l
19 thought they were hermetically sealed. - -l I . .
- j. 20 .R. MICHELSON:
M I thought-they were too. I l
i 21 thought they had early problems on the seal for the box, l t
I j 22 for the limit switch.
23 MR. EBERSOLE: This is the current flapf-- f f
24 MR. MICHELSON: They had to change - -all the ?
? :
25: Limitorque did, then, is change it out to a different kind i
, i l
ACE-FEDERAL REPORTERS, INC. - !
2n w.nco Neuenwide coverase soo n s. eses i
.. , _ _ _ - . . _._ _.. .__- _ _ _ .. _ ._... _ __ _.....___._~t
24594.0 101 BRT (j 1 of seal, they were happy. But maybe that wasn't the case.
2 MR. EBERSOLE: Interestingly enough, upstream of 3 that with the terminal boards and junction boxes and pull 4 boxes that's got the big hassle that has TVA shutdown all 5 over the place, they don't do what you feel they do.
6 MR. MICHELSON: They are having trouble under 7 these difficult conditions, so the problem has to apply to 8 the operator unless there's something unique to them and 9 the seal. We'll have a look at the environmental structure 10 of the contacts.
11 A little different question. In your choice of 12 motors, is there -- could you generalize as to how much 7_ 13 slip a motor can have before it stalls, with the types of
( ;
'~'
14 motors that you select? 5 percent? 10 percent? 20? 50?
15 MR. WARSING: I'm not an electrical engineer so 16 I'm not sure what a slip is. Our motors tend to have high 17 slip because we use very high starting torque motors.
18 MR. MICHELSON: In essence, the importance of 19 this question is: How slow might I expect a valve to 20 become before it finally quits moving altogether? If it 21 had only 5 percent slip, then it doesn't change its speed 22 much before it quits. If it's got 50 percent, then there's 23 quite a range.
24 MR. WARSING: Our AC motors have a very flat 25 curve. What we have -- our motors specifically designed to, i
a ACE-FEDERAL REPORTERS, INC.
r 202-347-3700 Nationwide Coverage 800-3364M6
24594.0 102 I BRT
_()_
1 essentially to take the knee action out of the speed torque 2 curve because we don't want that for our application. We 3 need a high starting torque from initiation without any
- 4 drop off in order to accelerate the load to pull the valve 5 out of seat. So you'll find our motor curves have very 6 little knee in them at all. They'll run very constantly 7 about the same speed, the AC motors, until you get up to 8 about 80 percent of the starting torque capability, and 9 then they drop off very quickly.
10 DC, of course, is very load sensitive. They are
.1 11 different. But the ACs are very flat.
12 MR. MICHELSON: One of the questions that 7s 13 concerned us is the ASME code, section 11, which talks 14 about surveillance tests for valves. I guess you are !
15 acquainted with that.
16 In there they talk about, if the closing time of 17 a valve changes by more than 25 percent, say, then you --
18 thus and so. The valve has become inoperable.
19 I'm wondering in the case of an AC operated 20 motor operator, whether 5 percent would ever be reached 21 before the valve would quit completely?
22 MR. WARSING 25 percent change in operating 23 time?
24 MR. MICHELSON: Yes. Change in operating time. ;
25 Is this even credible for an AC operator? Because the t
i (s.-)
, ACE FEDERAL REPORTERS, INC.'
202 347 3700 Nationwide Coverage 800 33H646
~ . . _ . _ . _
24594.0~ 103 4
~ BRT +
1 h 'l ~ motor has to do a lot of slipping, see, to drop that much 2 and still function. (
i 3 I think -- it's not realistic but I don't know. >
i 4 MR. WARSING: It's not impossible but you are l, 5 right, it's very unrealistic. If it does it, it's going to
~
6 go to locked rotor. That's the only way that will happen.
7 MR. MICHELSON: I was wondering if you had any :
8 view or numbers or anything on how much slip you do get :
9 before the motor simply quits. So I --
10 MR. WARSING: I can provide you --
t 11 MR. MICHELSON: No one has ever found one of 12 these 25 percent reductions, to my knowledge. I'm :
I 13 wondering if anyone is ever going to find them with,.at O 14 least, a Limitorque' operator because I don't know that your 15 motors can ever slip that much. :
16 MR. WARSING: They couldn' t.' I.can't see any way }
17 you'd ever find that.
- 18 MR. MICHELSON: Therefore the code is kind of a f 19 strange thing. It has picked a~ number.that has no reality 20 in it, perhaps. In which case the code ought to be fixed.
- 21- What is.a realistic number if people are going
{
22 to use section 11, which is 'what is being presently used.-
23 So everything is operable on the section -11 basis because-24 it either works ~or it doesn't work.- But they aren't ;
25' finding 20 percent reductions in-speed. And'I don't know _;
i ACE-FEDERAL REPORTERS, INC.-
2ns.3473?co_ . Nationw6de comese 800444p6 '
- . . , . , .. ~ - . _ _ . . . . . _ . , . _ . . __ . . u. - _.___.._:
24594.0 104 BRT
()
I that they ever will with Limitorque operators.
2 MR. WARSING: If you are using reduction in 3 speed as an indicator of an equipment problem, in the case 4 of a Limitorque valve actuator, it's not a good indicator 5 at all.
6 MR. MICHELSON: That's kind of what I'm trying 7 to say without letting -- by letting you say it.
8 Realistically, maybe 2, 3 percent speed change 9 might be observed, but not 25 percent.
10 MR. WARSING: In a real small valve, small valve j 11 with quick operating time, like 5-second operating time, 12 just by doing maintenance and changing the limit switch 13' settings you might change it 10 percent.
O 14 MR. MICHELSON: Will your operator still work?
15 MR. WARSING: Yes.
16 MR. MICHELSON: It depends a lot on the size of 17 the operator, also relative -- in other words, it depends 18 on the loading of the operator as to how much slip it can 19 tolerate before it stalls. It can tolerate more slip, you 20 know, if it's a big operator on a very small valve. But it 21 depends also on the motor. Some motors just slip so much 22 and then they quit; no matter -- even under zero load. Of 23 course they slip because there is some kind of loading.
24 You have to have a loading to make it slip.
25 MR. EBERSOLE: I want to ask an old question.
(O
_/
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Notiott@ W M3N
24594.0 105 BRT-p y 1 If you have a hammerblow feature to accommodate the fact 2 you can't come off a torque condition without some impact, 3 in the course of coming off the seat like that you lose the 4 voltage, and you recover it later after having a partial 5 withdrawal, it will.still come out, won't it, I hope?
6 ~ MR. WARSING: Yes, sir. The sizing procedures 7 do not take any credit for the hammerblow effect.
J 8 MR. EBERSOLE: It's kind of an add-on. Is there ,
9 an explicit and perfectly obvious place, tab someplace, i i
10 that says: Do not connect this in reverse phase on 11 three-phase designs? And what happens if you make that 12 mistake?
i 1
13 MR. WARSING: If you misphase it, you have 4
O 14 neither limit switch nor torque switch will be in the 15 control circuit and you have no control whatever.
4 16 MR. EBERSOLE: What's the thing you tell the 17 maintenance man to do? And is it put in red letters or 18 something?
19 MR. WARSING: It's not in red letters. It is 20 certainly a part of the instruction maintenance manual as 21 one of the first things you always do is put the valve in 22 position and check the phasing. That has to be done.
t i 23 MR. EBERSOLE: That's a ritual that everybody is 24 supposed to understand.
25 MR. WARSING: Yes, sir.
(3 4 V ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nanonwide Coverage m3%4646
24594.0 106 BRT
-( ) 1 MR. MICHELSON: Is it universally understood the 2 torque switch has to be bypassed by the time a hammerblow 3 is engaging? Or can you adjust the torque switch such that 4 you don't need to bypass it during the hammerblow engagement.
5 MR. WARSING: If it's a position seated valve, 6 you don't need to bypass it. That really depends on the 7 particular valve application and configuration of the 8 applicator. There are some applications where it would 9 never be practical to not bypass it.
10 Your question as far as the knowledge in the 11 industry, I expect there are a lot of people who don't 12 understand bypass circuitry at all from some of the 13 correspondence I see.
14 MR. EBERSOLE: In the matter of the torque ,
15 switch setting, do you take the view, since you don't look 16 at the valve itself, I guess your scope is your spline 17 shaft -- do you ask for any critical mechanical features of 18 the closure characteristics of the valve, where it is open 19 until it's torqued in in a 1/10,000th of an inch or 20 20/1000ths or whatever -- in short, do you have to 21 compensate for the resilience or lack of it in the valve 22 train?
23 A In a sense we do. We do it by the output speed.
24 The f' aster the output speed, we have adjustments factors we 25 use to back it off to compensate for that.
n
(_)
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide coverage 800 336 4646
., 24594.0- 107
-BRT 7-)
(_ 1 MR. EBERSOLE: Does it give you this resilience 2 or foot-pounds per micro-inch or whatever, from the valve 3 manufacturer?
4 MR. WARSING: No, sir. Those we have 5 incorporated by experience. ,
l 6 MR. MICHELSON: But it will change with respect i 7 to the different valve, won't it? L 8 MR. WARSING: Yes, sir.
9 MR. MICHELSON: How is that compensated for in 10 the final recommended torque switch setting that they send t
11 to the user?
12 MR. WARSING: That's really done by the valve 13 manufacturer.
( 14
-l MR. MICHELSON: He's supposed to know what 15 changes to make to that setting, if any?
4 16 MR. WARSING: That's right.
17 MR. MICHELSON: Let me postulate another 18 situation that would eliminate most of the concern about 19 knowing how to set torque switches and so forth, that might 20 raise some other concerns and I want your reaction.
21 If, for instance, you simply bypassed a torque 22 switch completely, so that it isn't -- you don't worry 23 about where it is set, even -- that's what it amounts to --
24 but you size the operator, now, to the particular 25 application, so you don't destroy the operator, if it hangs t'~T U
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee 2 33(Hidd6
. - - . . . . -. ~ .. .- - - - . - .- .. .. - .. .
k 24594.0 108 q l~ BRT l
,() .1 up -- what would be wrong with that arrangement? What i
2 would be the, you know, the shortcomings? I roelize it a ,
i 3 might_ burn a motor out once in a while or wha-ar r, but is there some fundamental reason why you should neve do that?j!
~
4
/ -
j 5' or is this a judgment call? It certainly gets I'
'E the f ,
6 question of setting the torque switch proper}y. it .;. ved i 7 you always the maximum thrust available to de your job. If 8 the motor stalls then'it would have stalled under any other l 9 conditions anyway because you used up everytning it has got.
r' 10 MR. WARSING: Considering that we are here 11 talking about reliability,.by defeating the torque switch 12 and limiting all your overload protection, you are i
13 certainly going to reduce the reliability. Because, unless l
.( )
14 you sized everything for the application so that 'you can --
l' 15 all the equipment is designed on a regular basis, it will i .
16' be capable of withstanding the maximum output values; you .
17 'are going to have a bad application. It's-abuse of the
[ 18 equipment.
l 19 MR. MICHELSON: The assumption is you design for i- ~ 20 the maximum output value of the operator, you design the 21 entire train for that.- Now, do you need any torque' switch 22 protection.other than to just keep from burning out a motor?
l 23 Mechanically I fixed everything else in the system -- I J
24 really, I did a proper matching of the operator to the ,
i 25 valve. And ILmade everything correct so that I can use the f
i
) r i
F ;
l . ACE-FEDERAL REPORTERS,'INC. I L m2 3c.3m sai m ca , asmen o
. ~ . . . .- , - . - -- . . . . , - ,. -,
[
l
- I1 24594.0 109 BRT.
. 1 maximum output of the motor.
2 If I ever stall the motor, I recognize if it 3 sits there very long it's going to burn up. I can put in 4 some fancy breakers to protect that even if I wish. But 5 now you don't worry about all these settings or 6 maladjustments or whatever and you always know you are 7 going to get the maximum thrust the valve is capable of 8 doing. You know now if it stalls it would have stalled 9 under any other condition.
10 Now, in terms of normal operation, though, you 11 haven't provided any protection but mechanically it's all 12 designed for that thrust to operate normally with that 13 thrust. So what's wrong with that approach?
14' MR. WARSING: What's wrongfwith that approach is 15 that your dynamic efficiencies of your equipment and the 16 inertia mean that your output load going into seat is 17 I greater-than you can generate from a static' condition to
, 18 pull it back out of seat. So even though you may design to 19 withstand those loads, once you seat the valve you may not 20 be able to electrically unseat it.
21 MR. MICHELSON: Yes. Of course you'll sti).1 2:2 have-to. seat it on a limit switch. But you' design such 23 that if the limit switch fails, you don't bust the valve in ,
-t /.
24 two.
25 But for valves which must perform only-in the
(
i ACE. FEDERAL REPORTERS, INC. '
202-347 3700 Nationwide Coverage 800 33 H 646
]
24594.0 110 CRT
() 1 closed direction, this wouldn't be an objection if it did 2~ hang up? It wouldn't be an objection --
3 MR. WARSING: It could be done that way.
' f4 MR. MICHELSON: ,
There's at least one utility 5 that does this all the time and they haven't had -- they 6 have had one -- few valve motor burnouts, but most 7 utilities have even when they have done it other says.
8 They haven't, to my knowledge, had a sticking 9 problem and they don't have an adjustment problem because 10 there's no adjustment to be concerned with.
?.
11 The safety function direction is adequately i
12 taken care of with the thrusts -- with the maximum thrust 13 available, always.
14 MR. WARSING:
Do you know what type of valves
.15 they use?
,, , 16 . MR. MICHELSON: These'are Limitorques?
17 MR.-WARSING: I mean the valves themselves, wedge 18 gates?
19 MR. MICHELSON: The utility is on all valves, as 20 far as I know. And has been doing it for over 10, 15 years 21 in operation.
22 MR. WARSING: I can see how it could be-done but
,.3 23 it would take an awful lot of care in the selection and 24 application.
pj j ,
25 MR..MICHELSON:- I don't know that in thoss days ACE-FEDERAL REPORTERS, INC.
-202-347 3700 - Nationwide Coversee 800 336 6646
24594.0 111 !
BRT. l l
rh is ,) 1 that was the case , even.
2 (Laughter.)
3 But that's the way they presently are wired up.
4- In five units, five different units in nuclear power plants.
5 And it is working fine and they don't have, then, the 6 adjustment question of.when does the bypass drop out and so 4 7 forth. It's a continuous bypass. It's a continuous "no 8 torque" protection.
9 On back seating they do have torque protection, 10- limit switch protection, really.
11 MR. WARSING: Okay.
12 MR. MICHELSON: You don't want to back seat with 13 all that thrust, obviously. But obviously it is performed O 14 with the full output capability of the valves limited only 15 by whatever that little plate is that's put in there, this 16 limitor plate, presumably protecting the mechanical 17 integrity.
18 MR. WARSING: But if you mechanically bypass i
19 that, you have eliminated the benefit.
20 MR. MICHELSON: I haven't heard they have_ever 21 busted an operator apart.
22 MR. EBERSOLE: They certainly sheared a 36-1/2 23 inch stem and they sheared it with -- sure did put some 24 strain on the casing.
25 MR. MICHELSON: Oh, yes.
r
()h l- ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800 336-6646
i
\
~
24594.0. 112 BRT' I r j 1
h- 1 MR.-EBERSOLE: ~ This brings'up an earlier
- 2 - question.' Since we'findsthere's no parental interest in g '#t.
- 3 the' valves or oversight or whatever, do you think your
-. { company might offer to integrate all this into one 5 overseeing function, and we have a national standards for 6- valve' design or;something like this? Or underwriters? I 7 see n'y hope of anybody anyplace,else. Everybody builds
,.8 valves,-goodness'knows --
9 1 MR. WARSING: I think we'd definitely be willing
( 10 to develop a standard. I havegajlittle trouble on the t .
11 commercial _ side seeing.how Limitorque, . being a supplier to 12 dalve manuf acturers, .could takel care of overseeing the
- 13_ entire package. The relationships are such that it doesn't
.O~ 14 lend itself.to that.
~
i 15 MR. EBERSOLE: Where would you see such an
'L 4 16 oversight fanction? 7I'm talking about integral valves.
17 MR. KARSING: I would say the standard is
+
g.
18 developed properly so the valve. manufacturer and actuator 19 manufacturer are supplying the propar information for a ,a 20 review --
21- MR. EBERSOLE: >You see it ASME --
'22 , MR. WARSING: -
that either the utility or i 23' '
' engineer could really do a proper evaluation 'and be, sure 124 .he's getting what he needs. ,
1 i $'
.i 25 MR.~EBERSOLE: There's no ASME standards on this A
j
- l. s s
{lr >
p ACE-FEDERAL REPORTERS, INC.
- Nationwide Coverage 800 336 6646
' _ _ -__--_-- __---- _ _ _ _ _ _ _ __ - _ _202 347-3700 - ._____n___
24594.0 113 BRT
- o. .
( _) 1. thing that we are talking about, are there?
2 MR. WARSING: No, sir.
3 MR. EBERSOLE: It's a wilderness.
4 MR. WARSING: It has been left to the 5 marketplace.
6 MR. MICHELSON: Henry, do you have any questions?
7 MR. JONES: No.
8 MR. EBERSOLE: Maybe it's the ASME, a void in 9 .their structure.
10 MR. MICHELSON: It's a standard 1641 on valves, 11 but it only touches on what we are talking about here. But 12 it did require, as I recall, and Earl Brown can correct me, 13 it has been a long time since I worked on it -- it did O 14 cover the question of wedding the operator to the valve, as 15 I recall; is that right?
16 MR. BROWN: Yes, the qualification was for a 17 valve assembly: Valve with operator.
18 MR. MICHELSON: But it really didn't delve into 19 these details that we are talking about here, in terms of 20 torque protection and so forth?
21 MR. BROWN: The only thing addressed by that 22- standard is to present a demonstration that the valve 23 assembly could operate under a prescribed set of severe 24 conditions.
25 MR. MICHELSON: As I recall, it required a A
V I
ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800 33HM6 t
_24594.0' 114 CRT
'Q l' stalled rotor -- that stalled rotor. conditions be a design j( j 2 requirement, wasn't it? Is that a vague -- just a vague
'3 recollection -
'4 MR. BROWN: I'm not sure.
5 MR. MICHELSON: I don't remember now. There 6 could have been several drafts earlier, too.
7 But it was dealt with, but that's the only place 8 I know of that the question of the assembly was ever dealt 9 with'and there it was still dealt with in a pretty general 10 way.
4 11 I don't think that standard would solve the 12 kinds of question we are dealing with today either.
13 MR. BROWN: The kinds of questions you are L f
() 14 raising now show that that by itself is not adequate.
15 MR. MICHELSON: That's right. That would be my 16 conclusion as well.
17 Any other questions? Because we are -- we have 18 now reached the end of the time allotted for this 19 particular discussion.
20 We certainly thank you very much. We have been 21 anxious to talk to somebody from Limitorque,-just to get 22 some of these answers because they are not readily 23 available. We have talked to valve vendors and others.
24 .Certainly we appreciate being able to talk directly to the.
25' manufacturer. I guess, of about 90 percent of all l
i
[
ACE-FEDERAL REPORTERS, INC.
202 347 3700 ' Nationwide Coverage 800 336 6646
24594.0 115 BRT O
(_) 1 operators, motor operators, are Limitorques in nuclear 2 power plants. It's an important --
3 MR. WARSING: I hope we gave you some insight 4 into what our experience has been. If there's anything 5 else we can do to help, feel free to call on us.
6 MR. MICHELSON: We certainly appreciate your 7 time. It has been a pleasure. Thank you.
8 Brian Curry is from Philadelphia Electric, who 9 has been using the MOVAT system at. the Limerick plant, I 10 believe, and is going to tell us some of his experiences ,
11 with the system as a further help to the Subcommittee.
12 If you will please, proceed.
13 MR. CURRY: Today I have been asked to discuss 7-)
14 the use of MOVATS testing at Limerick generating station 15 and its effect on Limitorque motor operator reliability. I 16 will review the torque switch, why MOVATS was used at 17 Limerick, the knowledge gained and the results of the 18 MOVATS testing performed, application of MOVATS in the 19 future at Limerick, use of MOVATS to enhance reliaLility, 20 limitations on MOVATS use at operating plants, and 21 -conclusions on the use of MOVATS to enhance reliability of 22 Limitorque motor operators.
.23 I'm a registered professional engineer, employed 24 .by the Philadelphia Electric. I have been involved with 25 Limitorque motor operators for 13 years. I participated in O
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646
24594.0 116 BRT r
1 the checkout, startup, and maintenance of Liinitorque motor 2 operators at Peach Bottom 2 and 3 and Limerick 1. I 3 developed and taught a course on Limitorque motor operator 4 operation and checkout. I participated in three quality 5 assurance audits of the Limitorque Corporation. I 6 participated in the development add demonstration testing 7 of MOVATS. I coordinated the MOVATS testing of 8 approximately 450 rising stem or torque seated butterfly 9 valves at Limerick. Approximately 250 of these were 10 safety related, and had been 100 percent reworked for 11 grease in compatibility.
12 During the rework process, numerous mechanical 7 13 problems were corrected. These problems at this point were
( .
'~'
14 primarily due to abuse of the operators during the 15 construction shifts. The long storage prior to 16 installation of these operators also contributed to the 17 problem.
18 MR. MICHELSON: Excuse me, you said you looked 19 at 450 butterfly valves?
20 MR. CURRY: No, 450 rising stem or torque seated 21 butterfly valves.
22 MR. MICHELSON: But only butterfly?
23 MR. CURRY: No. They were primarily rising stem --
24 MR. MICHELSON: You looked at globes and gates 25 and butterflies, in other words?
t ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage M336 6646
-24594.0. 117 BRT
~() 1 MR. CURRY: We looked at a limited number of 2 butterflies. There are not a lot in the industry.
3 MR. MICHELSON: That's what I wondered. Most 4 are globe and --
5 MR. CURRY: That's correct.
6 MR. MICHELSON: Thank you.
7 MR. CURRY: Limitorque motor operators are 8 controlled by torque --
.9 MR. WARD: Could you stand on the other side, do 10 you think?
11 MR. CURRY: That blocks out them. You can see.
12 Limitorque motor operators are used by torque and limit 13 switches. The torque switches control the oc.tput torque by O 14 sensing the compression of a spring. This spring right 15 here in the spring pack is the spring I'm talking about.
16- The motor -- the motor rotates a worm, located i
17 by the spring, driving the output gear.
18 (Slide.)
19 When the valve disc contacts the seat, the valve 20 stem itself stops. This stops the worm gear rotation 21 causing the worm to try to walk off of the worm gear itself.
i 22 .The worm movement is controlled by the spring 23 back right here and it's measured by the torque switch, 24 which is mechanically linked'to the worm.
t 25 If the spring constant is 1000 pounds per inch l ACE-FEDERAL REPORTERS, INC.
202 347 3700 ~ Nationwide Coverage 80S 33H646
24594.0 118 BRT v
is) 1 and the torque switch actuates at 1/10th of an inch 2 compression, the force on the worm itself is 100 pounds.
3 If the distant from the inner section of the gear teeth 4 between the valve stem -- worm and worm gear is 6 inches, 5 the resulting operator output torque then would be 50 foot 6 pounds.
7 The output torque --
8 (Slide.)
9 -- the output torque is converted to a thrust 10 load by the inclined plane action of the stem and stem nut 11 threads. This is an illustration of valve stem and you can 12 see it acts almost like an inclined plane.
13 (Slide.)
O 14 This conversion is effected by the friction 15 between-the stem and stem nut, and it's very sensitive. We 16 found certain valve manufacturers developed much more stem 17 thrust than we originally had expected. Okay.
18 MR. MICHELSON: And the reason was?
19 MR. CURRY: As near as we can figure, that the 20 friction between the bronze stem nut in the operator and 21 the steel valve stem was much lower than assumed by the 22 valve manufacturers.
23 MR. MICHELSON: You mean by the operator 24- amnufacturer?
25 MR. CURRY: Okay. At Limerick we had an item
(~')
v ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 6646
I 4
24594.0 119 BRT
() I called a valve motor operator compatibility form that had a 2 section filled out by the valve manufacturer and by 3 Limitorque.
4 MR. MICHELSON: Yes.
5 MR. CURRY: The valve manufacturer's section 6 specified the design stem thrust to operate the valve and 7 it also specified the maximum thrust that the valve could 8 withstand.
9 The section filled out by Limitorque also 10 specified the design stem thrust that they gave the torque 11 _ switch setting for, as well as the maximum stem thrust you 12 could expect out of the operator at its maximum torque 13 switch setting.
O 14 The design number by Limitorque and the design 15 number by the valve manufacturer agreed, one to the other.
16 However, when we tested the unit, it developed a lot more 17 thrust than we expected. We did an investigation of the 18 spring pack to determine whether it was the one that we 19 expected to see, per Limitorque's description. We also did 20 some limited testing to determine actual spring pack forces 21 at a given compression. Those numbers agreed with the 22 output torque number we expected from the operator. But we 23' got the high thrust. And the only thing we could ascribe 24 it to was the friction factor between the valve stem and 25 stem nut is much lower than was expected by the valve (jh
\
- ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336 4 646
24594.0' 120 BRT t ) 1 manufacturer.
2 MR. EBERSOLE: Isn't that a strong function of 3 the lubricant you use?
4 MR. JONES: And wear.
5 MR. CURRY: It's relatively insensitive to 6 ' lubricant application. When we found the problem, we said 7 this stem looks dry and we applied a lubricant to it and it 8 did not change the stem thrust significantly at all. It 9 may have changed it 1000 pounds in a 12,000-pound rating 10 and that was it.
11 I'll mention a little bit more about stem 12 lubrication later on in the presentation.
13 MR. EDERSOLE: You are telling me there's metal O 14 to metal contact? ,
15 MR. CURRY: Yes. There is some lubricant there 16 but unless there's a real lack of it you will get, you know, 17 a consistent stem thrust value each time you test the valve.
18 If the thing is so dry or so dirty that it 19 growls as it operates, you will get reduced stem thrust.
20 There's no doubt about it. But it's very clear -- if you 21 hear a valve that is growling -- we had this happen to it --
22 it was growling, we applied lubricant to it, stroked it two 23 times and retested it and bingo, we had our stem thrust 24 right back up where it was before.
25 MR. MICHELSON: Is this subject to the relative l I)
Lj l
SCE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 4646
24594.0 121 ERT 1 alignment of the operator to the valve?
2 MR. CURRY: We didn't do any testing in that 3 area; of checking the alignment of the tube.
4 Visually I would say they were aligned.. You did 5 not have an obviously measured, visual --
6 MR. MICHELSON: Your problem was you were ,
7 getting more thrust applied to the valve than you had 8 planned on; is that right?
9- MR. CURRY: That's correct.
10 MR. MICHELSON: So things were working better 11 than anybody thought.
12 MR. CURRY: That's correct.
13 MR. MICHELSON: So, lubrication isn't the answer O
1 14 because the lack of lubrication will make things work worse 15 than you had planned, not better?
16 MR. CURRY: Yes. On certain valve manufacturers.
17 MR. MICHELSON: So what could possibly make it i
18 work better?
19 MR. CURRY: The surface finish between the valve 20 stem and stem nut.
21 MR. MICHELSON: Even better than had been 22 predicted in the calculation?
23 MR. CURRY: Were assumed in the calculations.
24 MR. WARD: The valve manufacturer might have 25 thought he was being conservative by specifying high O
%.J
! ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 2 336 6646
24594.0 122 CRT
(')
'%/
I friction but it all depends on what you mean by 2 conservative.
3 MR. CURRY: That's correct. You can go off the 4 other end, so to speak.
- 5. MR. MICHELSON: Okay.
6 MR. EBERSOLE: Well, in the long run when the 7 stem goes down, there's another screw there someplace, 8 isn't there, a projecting screw to move the valve up and 9 down so you have another problem down below.
10 This shows me, the worm gear, you say, and the 11 lock and stem nut but then I have to go into whatever the 12 worm is that lif ts and lowers the gate or whatever. That's 13 got its own problems?
O k' 14 MR. CURRY: No, the -- that is the valve stem 15 thread that actually lifts the gate.
16 MR. EBERSOLE: That's what I mean.
17 MR. CURRY: In every -- most designs that we had, 18 you had a nonrotating valve stem so that the friction was 19 between the stem nut in the Limitorque operator and the 20 valve stem itself.
21 There is a certain valve design -- certain valve 22 designs that have rotating stems, where you have the stem 23 and stem nut is by the valve manufacturer, but it's a 24 relatively small number of valves.
25 MR. EBERSOLE: Are they dissimilar valves? The
()
(~m ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverses 800 336 6646 t
24594.0 123 BRT (m_) I worm and the --
2 MR. CURRY: What was that again?
3 MR. EBERSOLE: These are dissimilar materials?
4, MR. CURRY: In most cases these were dissimilar 5 materials. There were some valves where you had a bronze 6 or brass stem and the bronze stem nut.
7 Limit switches control the valve stem travel by 8 counting the number of turns of the worm gear.
9 (Slide.)
10 Since the valve stem thread gear and pitch is 11 fixed, it travels proportional to the revolution. It works 12 like the odometer of a car,-which is why I have this 1/10th, 13 1, 10, 100.
I 14 After a fixed number, the electric parts operate 15 placing the Limitorque switch into the control surface. So 16 if you have this driven from either the worm shaft gear or 17 drive sleeve, it drives in here similar to the odometer of 18 a car with a series of intermittent gears, and after a 19 sufficient number of revolutions this gear right here 20 rotates 90 degrees and you have your contacts operating.
21 At Limerick, MOVATS was used as a continuation 22 of other testing methods that we developed or tried for 23 limit problem resolutions. Prior to the use of MOVATS, 24 problems were investigated using various techniques, using
-25 torque wrenches and depth micrometers. In this case we (g >
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coversee 800 33 6 6646
24594.0 124
-BRT (a~j 1 found certain problems without using MOVATS and got them 2 corrected and resolved and agreed to by Limitorque when we 3 found the problem.
4 These problems were torque control and not 5 Limitorque control problems.
6 As a result of the testing, certain things were 7 learned about Limitorque motor operator problems. Many 8 torque switch problems are not torque switch problems. The
) 9 torque switch mis-operation is usually caused by a 10 mechanical proolem. The usual torque switch problem 11 ' correction, replacement of the torque switch, changes the 12 operation point of the operator and thus cures the problem.
13 Hydraulic locking of a spring pack can occur in 14 Limitorque operators that have been equipped with grease 15' release, which should have cured the problem.
16 Mechanical binding can occur in the spring pack, 17 causing the spring rate to increase and the motor to stall.
18 You effectively have a much higher spring rate spring pack 19 than you expected. This also causes high output torque and 20 high stem thrust.
21 Incorrect spring pack installation in the 22 operator can cause premature torque switch operation.
23 Since the torque switch senses spring pack 24 motion as spring pack compression, if the spring pack is 25 loose in the housing it sees that movement as force on the n
U ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 3364646
24594.0 125 BRT I 1 output gear when it really is not there.
2 Resetting the torque switch on SMB-000 operators 3 without placing the operator in manual will cause excess 4 output torque.
5 Rewiring of SMB-000 torque switches can 6 radically alter the operation point.
7 MR. EBERSOLE: In segmenting the spring pack, do 8 they measure the thrust forces that they achieve? Or do 9 they infer it from torque, torquing up the spring pack?
10 They want to measure the thrust, don't they? That's what 11 you are after eventually?
12 MR. CURRY: The valve wants to see thrust. It
, 13 doesn't care what the output torque is. As long as it sees
! /
'~'
14 sufficient thrust, by any means, it will achieve its 15 function.
16 MR. EBERSOLE: When you set the spring pack, is 17 it measured as a thrust when you pack it?
18 MR. CURRY: You are saying when I install the 19 spring pack how do I verify it's properly installed?
20 MR. EBERSOLE: Do you put a weight or lever and 21 measure the compression forces?
22 MR. CURRY: No.
23 MR. EBERSOLE: You infer it from winding it up?
24 MR. CURRY: No. You are in a different area 25 than I'm talking about here. The spring pack fits in a ACE. FEDERAL REPORTERS, INC, 202 347 3700 Nationwide Coverage 800 336 6646
24594.0 126 BRT
() I cavity. There's a shoulder --
2 MR. EBERSOLE: I see it.
3 MR. CURRY: There's a shoulder here in the 4 housing.
5 (Slide.)
6 And there's an adjustment shoulder on most 7 operators that has to be properly set so there's no excess 8 play there at the same time. If you compress it down, 9 you've in' creased the preload on this. Ilowever, the worm 10 can still rattle back and forth inside of there. So if you 11 leave it loose in the outwards direction you.have reduced 12 your closing stem thrust and if you depress it down this 7_,
23 way, you've decreased your -- or you've set it up so that
(_) 14 the -- it's more likely to open in the open direction than 15 the torque switch is likely to operate in the open 16 direction.
3 17 MR. MICHELSON: But that amount of compression 18 is specified by the number of turns on the nut, isn't it?
19 MR. CURRY: You mean the preload on the spring 20 pack?
21 MR. MICHELSON: Yes.
22 MR. CURRY: There is a specification from 23 Limitorque for what the preload is supposed to be.
24 MR. MICHELSON: But it's done by the number of -
25 turns on the nut, isn't it?
l
()
i 1
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee 800 336 4646
24594.0 127 l BRT l
c
(,), 1 MR. CURRY: In some cases.
2 MR. MICHELSON: How else is it done?
3 MR. CURRY: In some cases it's done by the 4 number of turns on the nut. In other cases it's done by 5 the gap between the torque limiting sleeve and the thrust 6 washer and in other cases the nut is tightened down hard ~
7 against the shoulder.
8 MR. EBERSOLE: But in this case do they do an 9 actual measurement of what you are after which is the 10 pounds force to compress?
11 MR. CURRY: Not to the best of my knowledge.
12 Mr. Warsing could answer that question much better than 13 myself.
14 MR. EBERSOLE: I gather you infer rather than 15 directly measure what the compression pressure is?
16 MR. CURRY: We are worried about what the stem 17 thrust was.
18 MR. EBERSOLE: Yes.
19 MR. CURRY: We really didn't concern ourselves, 20 what the actual forces were on the spring pack.
21 MR. EBERSOLE: But those would determine what 22 the forces were?
23 MR. CURRY: That's true. If it was grossly off 24 such that we needed a different spring pack, that's when we 25 would go and we changed the spring pack and that occurred b
a ACE. FEDERAL REPORTERS, INC 202 347 3700 Nationwide Coverage 800 336 4646
24594.0 128 BRT A
- is,) 1 in only two instances.
2 MR. EBERSOLE: Spring packs are calibrated, then?
3 MR. CURRY: They have fixed ranges.
4 MR. EBERSOLE: Like valve springs on a car?
5 MR. CURRY: Yes. They have fixed ranges.
6 MR. MICHELSON: If you replace the pack, is it 7 the entire pack or individual washers?
8 MR. CURRY: We replaced the entire pack as a 9 unit.
10 MR. MICHELSON: Have you ever had one of those 11 washers break?
12- MR. CURRY: We never saw it.
13 Any further questions here? Okay.
~#
14 The SMB-000 torque switch is much more sensitive 15 to abuse than the other style torque switches.
16 Some valve manufacturers' stem to stem nut 17 coefficient of friction is much lower than they assumed.
18 Due to this, the stem thrust is much higher than expected.
19 Valve stem thrust is not affected by lubrication 20 except by the extre,me lack of it. If the valve stem is so 21 dry or dirty it growls during operation, the stem thrust is
. 22 reduced.
23 As a result of the problem resolutions, 24 Limitorque has incorporated changes into their current 25 production. These changes are to an individual -- to A
V l
l ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646 L
~
24594.0 129 BRT f )) 1 individual components and in assembly techniques.
2 MOVATS at Limerick was not used to set the limit 3 switches. All the personnel involved in the checkout and 4 startup of the motor operator valves were instructed in the 5 proper techniques. As a result of this, only about 15 6 valves or operators were damaged due to all causes during 7 this period.
8 During this period, no motor burnouts occurred 9 at all, which I think is a significant improvement over 10 most plants in their checkout /startup phases.
11 Correct close limit switch setting can be 12 verified by observing motor current, observing torque
,, 13 switch operation, listening to the motor loading or I I '
14 listening to the disc pulling out of the seat.
15 The close limit switch must change state af ter 16 the disc pulls out. It cannot change before then or you'll 17 run the risk of torquing out the valve and having the valve 18 stop right at disc pull-out.
19 Correct open limit switch setting can be 20 verified by checking that there is some travel available 21 after the valve stops for motor operation in the open 22 direction. These methods are specifically called out in 23 the Limitorque motor operator valve maintenance procedures 24 at both Peach Bottom and Limerick.
25 MOVATS was valuable in the initial problem p 1 i
i ACE. FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 33H646
24594.0 130 BRT
() 1 determination by providing a dynamic record of the actuator 2 operation and before and af ter testing results. After the 3 initial problem of identification, it frequently could be 4 identified by other means. We frequently would see a 5 problem just by observation of the torque switch operation 6 or we would pull the spring pack out of the operator and by 7 examination of the spring pack or its insulation we could 8 see what the problem was', without necessarily involving 9 MOVATS at that point.
10 Thus, problem identification can be done without 11 MOVATS after the potential for the problem is identified.
12 MOVATS will be used at Limerick in the future if 13 a maintenance activity disturbs a torque switch mounting,
' 14 spring pack condition, or if the valve stem or stem nut is 15 replaced with a unit manufactured by other than the 16 original manufacturer. It will also be used if a problem 17 occurs that cannot be resolved by analysis. This is 18 subject to change as knowledge is gained of the affect of 19 changes to the valve and the operator assembly.
20 We feel that MOVATS is a developing technique 21 that will yield further knowledge of operator and valve 22 problems that have not been encountered yet. In one case, 23 valve stem to. bonnet interference had been recorded but not 24 identified. By comparison of the before and after data, 25 the identifying characteristics were noted.
(~)
L; ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 6646 L
24594.0 131 BRT
) 1 Valve manufacturer operate err sizing is 2 sometimes incorrect due to invalid design assumptions. If 3 a sample of the valve manufacturers production was tested, 4 the validity of its assumptions could be determined and 5 operator sizes could be reduced in certain cases.
6 I believe that many valve operator reliability 7 problems are due to the lack of understanding of Limitorque 8 operator operation and training on the part of plant 9 maintenance personnel. This problem is caused in part by 10 excess dependence on the Limitorque construction manual.
11 The present manual lacks necessary detail and is inadequate 12 in certain areas. In addition, the Limitorque operators 13 tender to be treated either as an item that is bought off
( )
14 the shelf ready to run, or as a device that.tries to 15 destroy itself on an irregular basis.
16 Improper operation by plant personnel 17 contributes to the problem. Operating the valve by pushing 18 the contactor after torque or limit switch operation 19 imposes excessive stress on both the valve and operator. A 20 Limitorque service representative has stated, and this is 21 for non-nuclear plant personnel he was referring to -- that 22 plant personnel often report doing this to seal a packing 23 leak or stop a valve from leaking through. An indicator of 24 this would be a tripped thermal overload with a valve open 25 against the back seat.
p
'J ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646 J
24594.0 132 BRT n
(_) 1 Operating the valve against excess differential 2 pressure also stresses the valve and operator. We have 3 seen this occur also.
4 If the thermal overloads are bypassed when this 5 occurs, a motor burnout may result. Improper manual 6 operation of the operator also causes corrective 7 maintenance to the Limitorque operators.
a 8 Present equipment problem reports on Limitorque 9 motor operators do not contain enough information for a i
10 proper analysis to be performed to determine root causes or 11 root cause trends. Those reporting the problems do not 12 understand the operators well enough to determine if the
_ 13 true cause was found, corrected, and reported.
\'
j 14 I've recently been involved with two valve 15 problems at Peach Bottom. In both cases, the MOVATS' 16 equipment was not able to be used as presently configured.
17 Modifications to the Limitorque operators may be required 18 in some cases to MOVATS test valves in older plants. This 19 increases material and time requirements for the testing.
20 In conclusion, MOVATS has identified 21 previously suspected but unverified problems in Limitorque <
22 motor operators. It has identified deficiencies in valve 23 manufacturers' sizing and setting calculations.
24 Use of MOVATS by the valve manufacturers should 25 allow more accurate sizing and setting of Limitorque motor
- ()
L
! ACE-FEDERAL REPORTERS, INC.
! 202 347 3700 Nationwide Coversee 800 33 H 646
'24594.0 133 BRT
_) _
(G 1 operators. Confirmation of these problems allows 2 identification and correction of the problems to be 3 performed much more easily.
4 In addition, since the true problem is corrected, 5 it does not reappear. ,
6 MOVATS' use will be continued at Limerick in the 7 conditions already noted. We are establishing a training 8 program based on the knowledge gained from MOVATS. This 9 training program will be used at Limerick and Peach Bottom.
10 By instruction, Peach Bottom bought many of the benefits of 11 MOVATS without the implementation of a testing program.
12 Use of MOVATS does enhance the reliability of 13 Limitorque motor' operators, but not only by the direct O . 14 ef fect of verifying the condition of the operators tested 15 but by the indirect effect of training personnel to 16 identify problems without it. Use of it enhances the overall 17 reliability, including operators that are not tested.
18 We have seen the training ef fect at Limerick 19 with our maintenance personnel at this point, where they 20 have supported us in our MOVATS testing.
21 MR. MICHELSON: Do you own your own equipment, 22 then?
23 MR. CURRY: Yes, we do.
24 MR. MICHELSON: So you trained your own 25 operators to interpret the CRT displays. and so forth?
0;
\_-
ACE FEDERAL REPORTERS, INC.
N ___ _ - __ _ . _ _ - _ _ . _-_ _ _
24594.0 134 BRT
(~T
\/ 1 MR. CURRY: We trained our people on what they 2 were looking at, how to operate the equipment, how to test 3 the operators, and have our own unique techniques to do 4 that. .
5 MR. REED: Interesting story. You make two or 6 three very interesting statements I would like to follow up 7 on.
I 8 I guess you'd ' have to agree that the whole valve 9 business, not just limiting it to the Limitorque operator 10- aspect, but the Limitorque business is a very fragmented 11 marketplace business, with the result, that I've heard from 12 you, is that the utility became very much in-depth involved 13 in a piece of their valves, to try to assure better 14 operability. Your testing program is the example.
15 Also, I think you said that some 250 safety
- 16 related valves had many problems from construction abuse 17 and storage. Are those the words you said?
1 18 MR. CURRY: That's correct.
19 MR. REED: You essentially went over these 20 valves. You bought a product. You thought you had a good 21 product. You thought it should operate but you had to go 22 over it with a fine-toothed comb to sort out many', many i
23 problems and perhaps you even found some things that the 24 manufacturer / supplier didn't even know about himself.
25 Now, this leads me to -- I, maybe, went beyond
(
ACE. FEDERAL REPORTERS, INC.
202-347 3700 hadonwide coverese soo 336 6646
[_ _
e 24594.0 135 BRT
/~%
(_) I what you were thinking. This leads me to wonder, again, 2 about this fragmented activity that creates a valve 3 assembly to do a job, proper application in a nuclear power 4 plant. .
5 Do you think all utilities can do this or should 6 do this? Or should there be some way in which there's 7 better assurance that this assembly, now fragmented, is 8 integrated to be more reliable and in the interests of 9 public safety?
10 MR. CURRY: From our Peach Bottom plant, we 11 developed the valve motor operator compatibility form that 12 I described earlier, in an effort to do just that, to 13 assure ourselves that the valve manufacturer had supplied O 14 the correct information to Limitorque and we had an 15 appropriately sized operator.
16 When we initially found certain problems, 17 because we reviewed all of our torque switch settings 18 initially, before we got into the checkout /startup phase, 19 we found discrepancies. And when we went back to the valve 20 manuf acturer, they would frequently tell us: Talk to 21 Limitorque, that's their part. And the valve manufacturer 22 tended to supply some information to Limitorque and assume 23 Limitorque would supply a unit that was correct.
24 There is a problem in the interface between the 25 two of them and I'm not sure that I have a good answer on O
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 80 4 336 4 646
24594.0 136 BRT
() I how to resolve the problem.
2 MR. REED: Do you think each individual utility 3 will resolve the problem? Doing the kinds of things that 4 you have been doing?
5 MR. CURRY: I can't speak for other utilities.
6 All I can speak for is my own utility and knowing what 7 we've done.
8 MR. REED: It sounds like a rather inef ficient 9 route that's not likely to lead to total coverage or 10 improvement, if each individual in a utility becomes a 11 volunteer to try to sort out the fragmented issues.
12 MR. CURRY: All I can say at the present time it
- 13 seems that is the approach being used by the utilities. We 14 are getting phone calls from other people at other plants.
15 They describe a problem to me and sometimes I can tell them 16 what the cause is and other times, it's similar to Dan's 17 experience. You get a description of what the problem is 18 and you say: That can't happen. Someone is not telling 19 you the entire story. Something else is happening.
20 I found this even in our own plant that I'm not 21 stationed at.
22 MR. REED: What, 25 years ago, as I said, I had 23 this occasion of stems that were too short. And everybody 24 denied they were too short but they couldn't stroke the 25 distance.
\ ,,-'
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
\
r 24594.0 137 BRT-1- It sounds to me like it's still out there, 2 complicated versions of this fragmented valve business.
3 Apparently you, as a utility, undertook to make 4 it.right for your station.
5 MR. CURRY: Yes..
6 MR. REED: But for the industry, what would you 7 -recommend?
8 MR. CURRY: I feel there should be a users' F
9 group on Limitorques.
10 MR. REED: How about the whole valve assembly?
-11 MR. CURRY: The valve assembly itself, or the 12 valve itself --
13 MR. REED: Or the application.
O 14 MR. CURRY: If you get a users' group and you 15 get people that are suf ficiently educated and interested in 16 resolving the problem, you can- come up with a procedure or r 17 an interface document that should give us better assurance 18 that we have a valve and operator assembly that will 19' fulfill the requirements as specified by the designer. But 20 there is a lack of a design document or an interface 21 mechanism between the. valve manufacturer and Limitorque.
22 MR. REED: Why I tried to extend your .
23 recommendation on a users' group from just the Limitorque, -
1 < ,
24 rising stem-type valve issue, is because I, quite frankly,
- l. 25 think that there are some valves, particularly-internal O
V I
! ACE FEDERAL REPORTERS, INC. '
202 347 3700 Nationwide Cowrage 000 336 4646
[2'459:4.0 138 LCRT
[ '
1 pilot. operated relief valves, that are misused, misapplied, 2 because of the environment in which they work.
! 3 The whole valve issue, somebody said here earlier, it maybe is 15 to 20 percent of the core melt 5 . risks, contributed to core melt risk. It is fragmented.
6 And it seems to me there ought to be some way to tie it 7 together, and not.have each individual utility as a 1
8 volunteer try to straighten it out.
9 MR. EBERSOLE: What's the foreign practice, do 10 you know? Like the French? Japanese? English? Whatever --
11 Germans?
! 12 , MR. CURRY: I haven't had a lot of experience 13 with them as far as what they do or do not do. I really
! (~)
~
'14 can't say what the foreign experience is with motor 15 operators.
t: 16 Speaking of something other than Limitorque i
17 bo' tor operators, I haven't dealt with them a lot to know-1,8 . , their problems in detail and really be able to comment on
- 19 them. It's the interface between the valve manufacturer 20 and the Limitorque manufacturer, is really on other types 21 of actuators -- but I can't say for a. fact that is the case.
22 I think problem resolution is a little bit
- 23 easier.say on a hydraulic or pneumatic basis because I took
- 24. it apart and the cylinder was scored or corroded or the
, 25 seal was blown out where there was no fluid lef t or I had
).
{ J.
ACE-FEDERAL REPORTERS,'INC.
! s, 202 347 noo Nationwide coverses an354646
T.
- 24594.0 139 BRT
() 1 it filled up with water. It is much more apparent what the 2 problem is than with the Limitorque actuators.
3 MR. REED: Jesse, to answer what's the 4 experience in other countries, I saw a little bit of the 5 experience in France this last May, where apparently on a 6 national, overall, because they have standardized plants 7 basis, they had developed a new type of PORV, and they got 8 around and put it,-backfitted it for two years in their 9 plants and then found out that it didn't work very well.
10 So, again, it shows the complexity of the overall 11 valve application and success issue. It is not just 12 operators. It has to do with internal pluggings and 13 environment and hydrogen formations and leakages-and all 14 kinds of things.
15 I wonder, you know in health physics they have 16 branch groups, health physics societies, meetings, forums, 17 symposiums, and I guess probably in reactor physic'. they do.
18 Do we have, in industry, such things as splinter vr a 19 specific industrial' organization for valve engineering and 20 application?
21 MR. CURRY: Not - to the best --
22 44R. REED: Where people like yourself would come 23 together to expose all these subtleties and issues?-
24 MR. CURRY: Not to the best of my knowledge.
25 I'm an electrical. engineer so, again, I don't normally get
. ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-6646
~ , - -
24594N: 140 lBRT n#
Q I
1- exposed to the mechanical side of it, which is where-the
- ,._ .q2 prime responsibility normally lies for valve procurement 3 and specification of actuators.
i
'4 'MR. REED: Do you know if INPO -- INPO, on 5 occasion, brings-together maintAnance people and some peoples operations people and CEOs and so on in various 7 forums or symposia. Do they ever concentrate on the valve 8 issue? Bring tcgether. valve people?
yj .
{ 9 MR. CURRY: The.only INPO effort that I know is
~
1 10 :the meeting they. held on Limitorque motor operators in '83,
'll .I believe it was.- But, as-far as other types of valves, I
{
i 12 have no. knowledge of it.
13 MR. MICHELSON: I think when one thinks of
- .O 14- operators, of course, you have.to consider.the problems of 15 their operators as well and its contribution to safety is 16 not' clear because, of course, what we heard earlier this f 17 morning related 'only this morning to a particular plant 18- where air operators were not important. In-Westinghouse 19 plants, air. operators are very important and serve most of I g ,- 20 the safety functions. So we shouldn't be slipping away and 21 think- if: we solve our relations with Limitorque that would 3
<22 'take J eare,of everything _because'Q on pressurized water
. e .. V.
-reactork,,.a'ir operation becomes exdremely important.
~
-23 7 ):&
your: boilers, it isn'd.. Most i everything on a GE-
.24 Where a on c .
a i- ,25 plant is electrical. ,! .-
lO .
c Y n'cs-FEDENL REPORTERS,'-INC.
p amArm ' . Namende Cemse f MD N MM
24594.0 141 BRT l- MR. EBERSOLE: Yes, it is, too. Come on, Carl --
2 MR. MICHELSON: I'm saying relatively speaking 3 it's much more important on a Westinghouse plant.
4 MR. EBERSOLE: The SRVs are --
5 MR. MICHELSON: Well, that's a very special case.
6 I'm talking about the generic class of valve problems and 7 not a particular valve problem.
8 MR. EBERSOLE: That's the life blood of safety 9 on a boiler, though.
10 MR. MICHELSON: So if you start owners' groups a 11 on the industry-wide basis, I think we would be remiss not-12 to look at air operations as well as electrical operations.
13 MR. REED: And internal pilot operation.
14 MR. MICHELSON: Yes. The whole spectrum.
15 MR. CURRY: Limitorque's big problem is people 16 just don't understand them and if you don't understand 17 what's going on inside the operator, you have a failure and 18 then you shotgun approach it until somehow we fixed it.
19 MR. REED: Do you think that perhaps the device 20 is too complicated? Could it be simplified? Could it-be 21 improved? Could it be made more idiot proof? Or.is it a 22 design ~ feedback resistance problem?
23 You know, if something is too complicated to 24 drive, maybe it needs to be improved?-
25 MR. CURRY: I think, my own personal. opinion is
/~
l- b l
l l
ACE-FEDERAL REPORTERS, INC.
'202-347-3700 Nationwide Coverage 800 336 4646
I 24594.0 142 BRT
/~T
() I that it is a simple device and it -- once you understand it 2 and realize what the device is telling you by its external 3 indications, you can understand what's going on inside of 4 it and it's easy to check out and set up and verify that 5 it's correct.
6 MR. REED: Can the mechanic understand it? You 7 are not going to do the overhaul on all of them and you are 8 not going to be in all companies.
9- MR. CURRY: I believe that he can if he has 10 proper training. That is the most critical part of it.
11 Proper training and proper instruction will' carry you 12 through a lot of-rough parts.
13 MR. REED: You wouldn't recommend any 14 improvements on the Limitorque operator?
15 MR. CURRY: I'm not saying they are not subject 16 to improvement. They may have had improvements based on 17 some of the results of what we found at Limerick. And I 18 see that as a good sign. It's not that Limitorque takes 19 the position that this is our design, take it or leave it, 20 that's your problem, you fix it at your plant, we are not
- 21. going-to send anything new out like that.
22 You know, anything can be improved but you've 23 got to provide a justification for. improving.
24 MR. MICHELSON: Could we pursue for a moment --
25 where do you get your information about other valve l [a
,3 ACE-FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coverage 2 336 6646
-24594.0. 143
~BRT 1 ' failures? What are the sources you use? Which are your 2 best sources about other peoples' experiences with valves?
3 MR. CURRY: I have gotten phone calls from other 4 people in other plants. I have also reviewed, at least 5 'some of ' the AEOD LER summaries . And I have also, reviewed 6 one of the NPRDS summaries, of Limitorque problems.
7 I have reviewed them to see what they said and 8 what the problem was ascribed to, how they corrected it.
9 And in certain cases I found that what the person making 10 the entry -- I have no idea who that person was --
11 described as a problem could not have occurred.
12 An example would be an NPRDS entry. There was a 13 case where someone said that the torque switch setting had 14 drifted low. A torque switch setting can drift.high but it 15 can't drift low unless there-is some mechanical problem 16 inside. But not by the torque switch itself.
17 The thing that-I saw in the AEOD LER summary was 18 in a number of cases it took a utility-two or three tries 19 before they got the valve to, apparently, work correctly 20 finally.
21 MR. MICHELSON:- From what you are saying, though, 22 you depend upon the detailed analysis performed by others 23 to kind of focus on problems and then they are conveyed to 24 .you as finished reports. To what extent do you analyze raw 25 data? -You know, to what extent- --
t f(_-)s l
ACE. FEDERAL REPORTERS, INC.
202-347 3700 - Nationwide Coverage ' 800 336 6646
24594.0 144 BRT
)- l MR. CURRY: At Limerick or all the way around.
2 MR. MICHELSON: Well, Brian, to what extent do 3 you use the MPRE data base in the process of doing this?
4 MR. CURRY: Okay. At other plants it's mostly 5 on a casual basis, if I know someone at the other plant or 6 they've called me for some reason about an entry that I 7 made on a note pad, a question related to it. Or called 8 them, they have made an entry saying: What do I do now?
9 In this case it's a casual basis. If they call 10 me up with a problem I'll assist them as best I can. But I 11 work for the Philadelphia Electric, I don't work for the 12 other utility.
13 Within the Philadelphia Electric Company, when a
-bs 14 problem comes up, I am likely to be the person that's 15 called about it to say: What do we do now? And come up, 16 with a resolution to it at that point.
17 After looking at the NPRDS summary, I was not 18 terribly impressed and I really don't have an active use 19 for it.
20 MR. MICHELSON: Do you use the NPRDS data base 21 itself on line?
22 MR. CURRY: No. I do not personally.
23 MR.-MICHELSON: You never go on line and try, 24 because you can, with a PC, of course, or any other simple 25 terminal.
n
( -)
m i
l ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336 4 646
l 24594.0' 145 BRT
-k/
[h 1 MR. CURRY: I have never done it. In looking at 2 the data in it I don't have a real interest to do it.
3 MR. MICHELSON: So you don't go into the NRC's l 4 sequence coded data base either to look for valve 5 experience?
6 MR. CURRY: No. One of the things when a 7 document shows up on my desk that references another 8 document I will normally go back and get the base. document
, 9 and find out what the true source is, what caused it to be 10 generated.
11 MR. MICHELSON: You react , though, mainly from 12 summary-type reports that you get indicating -- showing
{} 13 14 descriptions of problems and whatever?
MR. CURRY: That's correct.
15 MR. REED: It looks to me like' Philadelphia 16 Electric has made this kitchen sink issue of valves which 17 are all over the place, maybe 10,000 in the plant, they 18 have-made it a specialty investigation and looked into it.
19 But I'm not so sure that many utilities have, and most 20 utilities, I think, rely on rank and file maintenance 21 groups to ferret out the electrical mechanical problems of 22 the' valves, without a specialty engineering, I guess 23 perhaps a testing corporation that you have; a number of 24 people that are specialized in it.
25 Maybe it's the case that the valve integration,
~
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646
'24594.0 146 BRT-(-) -1 coordination out there by the suppliers in the marketplace 2 will always demand that the utility provide the people to, 3 on an individual -- and again a fragmented basis -- search 4 the problem and come up with answers. .
5 ,
I noticed something you just said. You said you 6 didn't look very much at NPRDS, and I suspect you know why.
7 You probably feel with your concentration on the issue you 8 know better than anything you are going to get out of it; 9 is that true?
10 MR. CURRY: That's true. I didn't dismiss it j 11- out of mind before' I looked at it at least once.
12 MR. REED: But I suspect that you have t'N 13 confidence that you have looked into this thing and you are
-g 14 .further ahead than what they are going to be able to 15 provide from their data base. But I just know that most 16 utilities aren't going to be in that position, so the valve 17 -- solving the valve issue and its relationship to nuclear 18 plant issue -- I hate to'use Jesse's word, he said 19 " wilderness," it is a little like a wilderness.
20 MR. MICHELSON: Your experiences which you out 21 liked this morning I' guess were Peach Bottom experience of 22 the, is_that right?
23 MR. CURRY: No.
24 MR. MICHELSON: Your same generalizations apply-L 25 to_ Limerick?
s_
-)
ACE-FEDERAL REPORTERS, INC.
202-347-3700~ Nationwide Coverage 800 336-6646
24594.0~ 147 BRT
's 1 MR. CURRY: I was at Peach Bottom from '72 to '75.
2 MR. MICHELSON: Your same generalizations, do 3- they apply to Limerick? What I'm driving at is this:
4 Peach Bottom was built in a different time frame than 5 Limerick and I was going to ask were there any significant 6' differences in experiences between what you saw for these 7 much older valves, which I think the situation was 8 different, than perhaps for the newer generation when 9' perhaps the valve vendor was doing a better job of matching
_10 the operator to the valves? Perhaps.
11 MR. CURRY: The thing that we saw, and let me 12 make one point right now, I found out that NPRDS is only 13' q'} available in our corporate office at 23rd Street. I am 14 situated at Limerick with occasional trips to Peach Bottom.
15 MR. MICHELSON: That means you have to go down 16 there to get on the terminal to play with it?
17 MR. CURRY: That's correct.
18 MR. MICHELSON: That isn't a part of the secrecy 19 problem with it, I hope?
20 FROM THE FLOOR: It's just available right there 21 now.
22 MR. MICHELSON: You don't have a terminal at 23 Limerick that can go on line with the data base? I see.
24 Go ahead.
25 MR. CURRY: Okay. What was your question --
o V
ACE-FEDERAL REPORTERS,' INC.
202-347 3700 Nationwide Coverage 800-336 6646 ,
24594.0 148 BRT (dD s 1 okay, the difference between Peach Bottom and Limerick.
2 One of the differences that we've seen is that 3 the valves at Limerick are much stiffer. I describe them
,4 - as being seismic valves.
'S As a result of this when they hit the seat, the 6 loads- build up much more rapidly than in what I'll call a 7 nonseismic valve, that doesn't have the really wide yoke 8 supports or anything like that, that the loads build up 9 much more slower. The spring pack compression rate is 10 slower in a nonseismic than a seismic valve.
11 MR. MICHELSON: Of course Peach Bottom is 12 seismically designed, different standards, perhaps.
13 MR. CURRY: Yes, you can see it in the valve CT) 14 design. .
15 MR. MICHELSON: You are saying these are more 16 ruggedly built valves from the standpoint of seismic 17 resilience.
18 MR. CURRY: They are stiffer and they operate 19 slightly differently in their operating characteristics.
20 MR. REED: Better or worse?
21 MR. CURRY: I can only say they operate
'22 differently _as far as characteristics go. We have had many 23 fewer problems at Limerick with valves during our -- we are 24 in a sense power testing right now -- than I remember 25 experiencing at Peach Bottom in the same phase; that we are L ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverase 800 336 4646
24594.0 149 BRT
-- 1 relatively problem free with valves.
2 One example is that we've had a total of three 3 motor burnouts in this power testing. Two of them were on 4 control valves, where we think we have a controller problem.
5 And one of them was, apparently an improperly 6 manually-operated valve, jammed it.
7 MR. MICHELSON: So, for whatever reason, 8 Limerick is behaving better than the Peach Bottom 9 experience?
10 MR. CURRY: It's very excellent in my opinion.
11 MR. MICHELSON: So, perhaps there has been --
12 one could speculate there has been, perhaps, a better
{} 13 14 matching of operators to valves?
MR. CURRY: One would hope so. The experience 15 that we've had, sometimes in'the -- in certain cases when 16 we were working with the MOVATS equipment initially, we 17 experienced operator stalls or we saw, in identifying a 18 problem, the operator stalled. It went.to motor stall 19 torque. In those cases we did not have valve failures with 20 the exception of one that I know of which was a control 21 valve with a very small stem size. But in most cases the
.22 valve would tolerate motor stall.
23 MR. MICHELSON: The valve vendors are somewhat 24 different for Limerick, probably, aren't they?
25 MR. CURRY: -Absolutely.
[
a(')
ACE-FEDERAL REPORTERS, INC.
202-34773700 . NationwideCoverase_ 800 33Mi646
24594.0~ 150 BRT
[
\ l MR. MICHELSON: Can you get a large number, say, 2 Wallworth at Peach Bottom. I don't think there are any 3 more of those at Limerick.
4 MR. CURRY: At Limerick there's Anchorage Bell 5- and Valve, who are our chief or prime people that had much 6 higher than expected stem problems.
7 MR. MICHELSON: That kind of clouds the problem 8 too.
9 MR. CURRY: Yes. And then we had Crane and 10 Pacific, who were right on.
11 MR. MICHELSON: Yes.-
12 Are there any-other questions? According to our 13 schedule, it's 12:20,_ time for lunch. Before we leave I
. 14 'would like to thank Brian for his fine presentation and 15 information. We appreciate it_very much. It'sinice to 16 hear from those who use the equipment as well as those who 17 manufacture it. Thank you very much and we'll adjourn 18 until 1:20.
19 (Whereupon, at 12:20 p.m., the meeting was 20 recessed, to reconvene at 1:20 p.m., this same day.)
21 :
22 23 24 L 25 I c]
l N.
i
~
ACE-FEDERAL REPORTERS, INC.
202-347-3700 - Nationwide Coverase 800-336-6646
-24594.0 151 ERT
() .1 AFTERNOON SESSION (1:20 p.m.)
2 MR. MICHELSON: I believe we are ready to get 3 started again. Before we get started on Virginia Power, I 4 would like to give Brian Curry just a moment to give us 5 some further information to make sure the record is 6 sufficiently clear, so Brian, if you will.
7 MR. CURRY: I'm Brian Curry, again. The point 8 we definitely wanted to make is that when we reworked the 9 0-listed operators, we said that they were due to 10 construction abuse. That was not construction abuse on the 11 part of Limitorque in their fabrication of the units but 12 actual construction activities where they were doing a
(}' 13 hydro, or they had to open or close the valve for some 14 reason and the construction personnel would actually abuse 15 the operator and either damage or destroy the clutch in the 16 thing or break the handwheel or something like this.
17 The other thing was the long storage requirement, 18 in some cases we had valves that were 10 years out of 19 operators -- operators that were 10 years.out of date 20 manufacturing when we put them into service.
21 MR. MICHELSON: Any questions on that?
22 MR. EBERSOLE: One thing. We learned a while 23 ago that these supposedly environmentally qualified valves 24 'are open construction.
- 25 MR. CURRY
- Yes.
! r~S
%)
i ACE-FEDERAL REPORTERS, INC.
I 202-347 370D Nanonwule Coverase M3366646
-24594.0 152 BRT
'ks 1- MR. EBERSOLE: Do you have any personal 2 curiosity about whether or not that will produce surface 3 condensation on the terminal boards and so forth and cause 4 unfortunate performance? Are you going to look into that?
5 I'm talking about just from a private utility standpoint.
6 MR. CURRY: It could happen. I have seen the 7 Limitorque test facilities and how they set them up but
- 8. they are wide open to the environment.
9 MR. EBERSOLE: But they can be preheated.
10 There's a variety of ways to get around these things.
11 MR. CURRY: True. All I can say is maybe we'll 12 have to take a look at it.
/~% 13 MR. EBERSOLE: Why don't you chill one of them I
Q
'14 and give it a steam bath and see what happens.
15 MR. MICHELSON: The other thing that's important, 16 by the way, in doing that test'is make sure you monitor the 17' condensation during the injection of steam into the 18 autoclave or whatever. Don't monitor it after you are all 19 done to see-if it s'.ill works. The real problems occur as
-20 a transitional problem, as the condensation starts to occur, 21 the arcing, breakdown occurs. If there's no voltage at the .
22 - ~ time, nothing happens. When you are all done' it will work 23 fine.
24 MR. CURRY: One thing on Limerick, I can't 25 remember now for Peach Bottom, but'all of our-isolation
(~\
V l
l ACE-FEDERAL REPORTERS, INC.
202-347 3700 - Nationwide Coverage 800 3364646
24594.0 153 BRT 1 valves in the harsh environments are AC units, they are not 2 DC units.
3 The other thing is that for us we removed all 4 the terminal blocks installed by Limitorque, including 5 those for the motor terminations, and we made the motor 6 terminations using environmental qualified splices. In 7 fact when we had a split voltage motor, 220/440, where you 8 had pigtails brought together with a high hat connection, 9 we removed those and put an environmentally qualified 10 splice on it so the motor will not have a problem shorting 11 out. We may have degradation in the control circuit due to 12 loading, of moisture, condensation, but the motor
(' ; 13 connections themselves are, you know, protected.
14 MR. EBERSOLE: You had to do that as your own 15 modification to the standard product, didn't you?
16 MR. CURRY: That's correct.
17 MR. EBERSOLE: But it came to you as a qualified 18 product. So we have this weird thing; what's happening 19 elsewhere?
20 MR. MICHELSON: Not everybody has done that but 21 they all claim to be environmentally qualified.
22 MR. EBERSOLE: Of course your buddies will be 23 mad at you when they find you did that.
24 MR. CURRY: Philadelphia Electric as a company 25 has taken a lead in certain areas. We had a fire at a v
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage C
24594.0 154 3RT-
1 Peach Bottom 1 plant, many years ago now --
2 MR. EBERSOLE: I remember.
3 MR. CURRY: And as a result we have upgraded and 4- have very tight cable standards as far as fire resistance 5 goes, probably much better than others in the industry, 6 just for that reason. Something happens, we are sensitized 7 to it and we take corrective action.
8 MR. EBERSOLE: Well, good. You can do it with a 9 steam genny, in situ.
10 MR. MICHELSON: Mr. Hardwick is going to give 11 the introduction. If you'll give your introductio.. we'll 12 go into the presentation. Thank you.
(} 13 MR. HARDWICK: I'm Ray Hardwick, manager of 14 nuclear programs, from the nuclear operations department of 15 Virginia Power. We are pleased to talk with you this 16 afternoon about a predictive maintenance program and 17 especially the role of MOVATS and our experience using the 18 system. ,
19 In way of a brief introduction I would like to 20 mention to you that Virginia Power is the 14th largest 21 investor-owned electric utility in terms of assets, and the 22 fourth largest investor-owned supplier of nuclear generated 23 electricity. In the first quarter of '85 we produced more 24 nuclear-generated electricity than any other utility in the 25 U .S . To date, approximately 46 percent of our system b
I ACE-FEDERAL REPORTERS, INC.
M SiZD ~ Nationwide Coverage 800-336 4646
24594.0 155 BRT r'
~- 1 generated power is with nuclear.
2 We do have four nuclear plants, Surrey 1 and 2, 3 that.came on line commercially in '72 and '73, and North 4 Anna 1 and 2, commercial operation in '78 and '80.
5 As your addenda showed, Bob Nicholls is with me 6 here to discuss the details of our predictive maintenance 7 programs with you. Bob has been with Virginia Power for 27 8 years, with experience in electrical maintenance in both-9 fossil and nuclear operations. He is a nuclear operations 10 and maintenance electrical equipment specialist and is 11 currently on cur corporate staff, he's a program director 12 for predictive maintenance. He.has worked with MOVATS in
{} 13 reviewing our motor-operated valves at Surrey, and he's 14 prepared to point out some of the problems that.we have 15 identified using the system and - discuss details and will 16 respond to whatever questions you all have.
17 MR. MICHELSON: What. fraction of your larger 18 va1ves are motor-operated in pressurized water reactors?
19 Do you have any feeling? I know they use many more 20 air-activated in the boilers, but do you have some kind of 21 a feel?
22 MR. NICHOLLS: I'm sorry. I don't have a 23 personal feel.
24 MR. MICHELSON: -It's much less than boilers, I 25 know.
l l
()
ACE-FEDERAL REPORTERS, INC.
rmaarm sre r~w- ew sm
i 24594.0 156 BRT
(~T Nl 1 Okay, go ahead, proceed.
2 (Slide.)
3 MR. NICHOLLS: Let me review a little bit the 4 presentation outline just to put in perspective what the 5 main thrust of our presentation is today.
6 In the experience in the use of the MOVATS 7 system is four basic areas which we'll try to address in 8 the presentation to the committee.
9 One is what attracted Virginia Power to the 10 MOVATS system.
11 Number'2 is the motor-operated valve problems 12 detected with the use of the MOVATS system in the area of
{} 13 possible constructive actions and improvements on MOVs from 14 our viewpoints; and the consideration of the use of the 15- MOVATS system in the predictive maintenance program.
16 Virginia Power was introduced to the MOVATS 17 system in 1983. During that time we were doing some tests 18 on the MOV-rerate issues, and we had MOVATS test a small 11 9 -sample of our valves to provide field data in support of an 20 engineering study on valve performance.
21 This data'went to our AE engineers, also went to 22 Limitorque,-and we wanted to know. exactly what was going on l
23 in the field versus, you know, the calculations.
24 We saw something we liked very much with MOVATS 25 and at that time in '83 we established a service contract r3 k_/ .
ACE-FEDERAL REPORTERS,'INC.
202-347-3700_ _ _ NationwideCoverage 800 336 6646 _ ,.
'24594.0 157
~BRT 1 with MOVATS for the Surrey units and the North Anna units.
2 That's still active.
3 MR. MICHELSON: Excuse me, when you say you have 4 a service contract, does that mean they come in -- you 5 don't own the equipmen,t. They come in with the equipment 6 and the people?
7 MR. NICHOLLS: That's correct. What we 8 established in '83 was an on-call type of program.
9 MR. MICHELSON: Thank you.
10 MR. NICHOLLS: To where, within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, MOVATS 11 could be on site with the equipment, with the expertise, to 12 address the problem-identified MOVs, to help the
~
13 maintenance forces to pinpoint the problems, develop the 14 corrective action, recommendations and --
15 .MR. MICHELSON: Have you since bougnt your own 16 equipment? -
17 MR. NICHOLLS: It's budgeted. We are putting it 18 into our program in '86, yes, sir, for both Surrey and 19 North Anna.
20 Just to mention it, we have had in place all 21 along with Surrey and North Anna, preventive, corrective 22 maintenance programs, construction manual guidelines, 23- construction procedures, vendor training, and all of those 24 things. And we saw an added feature in doing signature 25 analysis that we followed up on when we saw this in '83.
O L)
ACE-FEDERAL REPORTERS, INC.
202-347-3700 ~ NationwideCoverase 800 336-6646 .3
24594.0- 158 BRT-
-1 (Slide.)
2 The major attraction to us, quite simplistically, 3 was the as-found conditions could be documented. I heard 4 some good discussions from the committee this morning on-5 the valve married to the operator married to the motor
- 6 control center married to the cable, et cetera, et cetera.
7 This was, to us, kind of a bottom line effort. It goes 8 through all the gyrations from the valve supplier to the 1
9 field. We can see how it's actually performing in a-live, 10 installed condition in which to measure its performance.
11 The as-found conditions can be documented.
I 12 After maintenance conditions are as left,_ documentation is 13 now available for independent review outside the ;
- {
i 14 maintenance department by performance engineers, which 15 gives a kind of redundant rev'iew , from those who are Ii 16 actually doing the work. Electrical and. mechanical
' 17 degradations could be detected by performance documentation i 18 and reviewed for corrective action.
19 MR. EBERSOLE: .Let-me ask you a question. How 20 did you-motivate' management to do-this?-'I tried it 20-21 ; years- agoiand it didn't work, , so I'd like to hearisomebody 22 tellJmechow he made a success out of it.
123 Was it;on the basis you might-improve 24 availability which satisfies-the masters <of business L 25- 1 administration? Or was it on the safety end? .or composite.
l Oi
{
4: ACE-FEDERAL REPORTERS, INC.
.. L M**3M
- W W -.__ - . # 3
, 24594.0 159 l BRT 1
1 of both of these? How did you manage to make that sale?
2 MR. NICHOLLS: All of those. I was in a unique 3 . position, personally, because I had been with the company a 4 long time. Hopefully the integrity of my comments stood up.
5 And I believe in signature analyses.
6 I had tried to start a program such as this in 7 the early '70s and it fell flat on its face because we 8 didn't have the support at that time because of other 9 priorities. Since then we have committed ourselves to 10 predictive analyses, predictive maintenance. And, as we go 11 on in the presentation, you'll see that credibility in our 12 predictive maintenance program for rotating equipment, such
(} 13 4s charging pumps, et cetera, just fell right in line.
~
14 For the first time we can see signature analysis, 15- what this slide represents, to where we can document what 16 .the valve is telling us for independent review.
17 MR. EBERSOLE: Predictive maintenance. In other 18 words you say: I am going to anticipate I'm going ~to have 19 to maintain this thing yet. It ain't broke yet but it's 20 coming down.
21 - MR. NICHOLLS: That's correct.
22 MR. EBERSOLE: The Japanese do that.
23 -MR. NICHOLLS: To answer our original question,_
24 our corporate folks, or vice-president in particular, 25 Mr. W.L. Stuart, has come up through the ranks as a shift (m.
Aj-I ACE-FEDERAL REPORTERS,' INC.
u_ F%19.EQ .ha 800 336 6646
'24594.0 160 BRT
(')
\/ 1 supervisor, et' cetera, and he has good insight on what's 2 really needed in the field and he saw the merits of this 3 program.
4 MR. EBERSOLE: That was the void in our program.
5 There was no such background. You can see through the 6 newspapers what has happened to-TVA.
7 MR. NICHOLLS: Mr. Michelson, the reason for 8 showing this slide is just to illustrate what I'm sure the 9 subcommittee here is already aware of and that is the 10 documentation features of running loads.
11 MR. MICHELSON: I have a little problem with 12 that slide. I have seen a similar one before and I'm
{} 13 14 always puzzled on your bottom chart there showing the torque switch eventually dropping out and just left with 15 the limit switch bypass. I guess that's what it's showing, 16 isn't it?
17 MR. NICHOLLS: Yes, it is.
18 MR. MICHELSON: While the torque switch is in --
19 what are they measuring that gives us higher amplitude?
20 And then the narrow amplitude after just the limit switch 21 is in?
22 MR. NICHOLLS: Being responsible for the data 23 gathering part of it and not totally responsible for the 24 interpretation of it -- there . ire others in the room, like l 25 Mr. Dobbin now and others, who can address that more
(
ACE-FEDERAL REPORTERS, INC.
, 202-347-3700 Nationwide Coverase . 800 336-6646 , _
24594.0 161 BRT l
,, l 1 correctly than I could. Would you like to comment, Art?
2 MR. CHARBONNEAU: That's a signature used in the 3 old current sensing technique.
4 MR. MICHELSON: That's really looking at an AC 5 wave then, isn't it. Then why, if you have a limit switch 6 bypass -- okay, you are using the old technique and you had 7 transformers on both contacts, didn't you?
8 MR. CHARBONNEAU: Right.
9 MR. MICHELSON: In other words, if you had 10 . bypass you wouldn't see any pip on the torque switch -- the r
j 11 hammerblow, rather, because it's bypassed by a short 12 circuit. But this is a different arrangement where you 13 were monitoring the two and trying -- and putting them on
{'}
14 one CRT. .
15 Okay. That explains it. Thank you.
16 MR. EBERSOLE: Are the instrument's response 17 characteristics clearly fast enough to get you a good 18 picture?
19 MR. NICHOLLS: We think so; yes, sir.
20 This series of slides that we are going to 21 address now are of the -- addressing the problems that were 22- detected with the use of the MOVATS system.
'23 The close-to-open bypass switch requirement r 24 adjustment. Our corrective action at that time.
t 25 Procedure reviews had more detailed instructions G
O ACE. FEDERAL REPORTERS, INC.
m M P3 -3700 - Nationwideh 800 336 6646
24594.0 162 BRT
%' 1 on the adjustment of the limit switch.
2 Operator design thrust exceeded; was the as-found 3 condition. And the action,that we took was the operator 4 thrust design and proper torque switch set points were 5 reviewed as part of the corrective action. Torque switch 6' adjustments were corrected as necessary.
7 When I say this design review, and things of 8 that sort, it's not only the Virginia Power engineering 9 in-house reviews but it's also the AE review, in our case 10 would be Stone and Webster, and when necessary, Limitorque 11 would be involved. So you get the whole team together to 12 look at what's going on.
13 MR. EBERSOLE: Was the operator designed thrust --
14 was it just inferred? Because you are doing current 15 measurement?
16 MR. NICHOLLS: No. It was measured to my 17 knowledge.
18 MR. EBERSOLE:- It was measured?
19 MR. NICHOLLS: Yes.
20 MR. EBERSOLE: How did you do that?
21 MR. NICHOLLS: With the monitoring equipment 22 with the TMD, and the thrust curves that can be obtained 23 from the signature, these things can be calculated.
24 MR. EBERSOLE: Well, it was inferred, then.
25' MR. NICHOLLS: Yes.
<~
.L)3 l
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-33(Hl646
24594.0 163 BRT b
s- 1 MR. MICHELSON: The design thrust you are 2 referring to there is . really the needed thrust, by design?
3 Not the thrust capability of the operator, of course; isn't ,
4 that right?
5 MR. NICHOLLS: I think we are maybe in two areas.
6 One area is the design thrust characteristics of all the 7 members that are given by the calculations to the valve 8 manufacturer, to the AE, et cetera.
9 MR. MICHELSON: Yes.
10 MR. NICHOLLS: As I understand what we are doing, 11 the MOVATS equipment can look at the literal thrust values.
12 MR. MICHELSON: The actual, yes.
13 MR. NICHOLLS: The actual values of what this 14 valve is seen, regardless of what the design is. The valve 15 is saying to the MOVATS system: This is what you get at 16 the valve seat. These are your foot pound forces. Le't ' s 17 take those field data as found, actual performance values, 18 and bounce it against the calculations and system design, 19 et cetera, et cetera, and see what comes out.
20 MR. MICHELSON: Yes.
21 MR. EBERSOLE: How do you do that without strain 22 damage?
23 MR. NICHOLLS: I'd rely on Art and his 24 technology there.
.25 MR. EBERSOLE: It sounds like you are inferring I
(~~)
s-l ACE-FEDERAL REPORTERS, INC.
. 202-347-3700 Nationwide Coversee 800 33(Hl646
324594.0- 164 BRT-1 it.
2 MR. CHARBONNEAU: The MOVAT is based on the more 3 'the spring part moves the -- that's what you are gauging.
4 MR. MICHELSON: You are usingfa load cell to 5 determine the trust?
6 MR. CHARBONNEAU: We calculate the displacement 7 of the spring pack to the actual load being delivered. All 8 subsequent telling just measures the --
9 MR. MICHELSON: But you calibrate it with the 10 load cell.
11 MR. CHARBONNEAU: 'The load cell is used to 12 calibrate. If I-may, your discussion on-operator design
{) 13 14 thrust being exceeded is in fact the finding that-the operator designed for that type operator is being exceeded.
15' MR. MICHELSON: That's the number that's meant 16 here?
17 MR. CHARBONNEAU: That's the one meant here.
- 18 Not the design thrust calculation to open or close the 19- value is either excess or' inadequate.
I 20 MR. MICHELSON: You -are saying here like we 21 ' discussed this morning, the operator is actually delivering
~
22 more thrust than it ought to be able to deliver by nominal 23 calculation?
24 MR. CHARBONNEAU: By Limitorque's design 25 parameters. They place a limit although you can exceed lO y
ACE. FEDERAL REPORTERS, INC.
- i. _
, m347 3700 N.iion.id.co r soo.33 w 646 ._
24594.0 165 BRT q\~#
1 that for a certain number of cycles, but an SMB-000 is 2 designed for a maximum thrust loading of 8000 pounds. We 3 scw the operator delivering 9000 pounds for whatever reason, 4 you have that finding. <
5 MR. EBERSOLE: Before you sit down, that load is 6 obtained when you compress the spring pack on the torque 7 switch setting, isn't it?
8 MR. CHARBONNEAU: Correct.
9 MR. EBERSOLE: What's the ratio of that loading 10 on a nominal basis, to the loading in the completely 11 duressed load, like attempting to intercept a full 12 hydraulic flow. It must be the torque switch setting is 13 substantially higher than the load it would get when it's
(}
14 trying to intercept a mass flow, which is the design basis?
15 MR. CHARBONNEAU: No. That's a good point ? to 16 bring up because it's more of a technical point. It's one 17 Bob hasn't specifically addressed but it comes up very 18 often. If you or the valve manufacturer tells us you need
~ 19 10,000 pounds to close that' valve under the designed 20 conditions, then the MOVAT system can test the operator and 21 see when the torque switch is tripping and is it tripping 22 at 10,000 pounds. It doesn't make a difference whether 23 there's load or flow or pressure. 9 24 MR. EBERSOLE: If it's tripping at 10,000 pounds 25 it won't work?
'i.
ACE-FEDERAL REPORTERS, INC.
fmJn7-riET) - N 800 33(HiM6 j
'24594.0 166 BRT i 1
rm l 1 MR. CHARBONNEAU: Excuse me?
2 MR. EBERSOLE: If it is tripping at 10,000 and 3 you needed 10,000 --
4- MR. CHARBONNEAU: Oh, you are right. We'd 5 typically put an additional 25 percent and match it up 6 against-the valve to make sure you don't damage the valve.
7 MR. EBERSOLE: That's what I wanted to hear you 8 say. But you will validate the torque switch tri'p is 9 higher than the full load requirement?
10 MR. CHARBONNEAU: You tell me what load you want 11 and we'll tell you if you are getting it; that's right.
12 MR. EBERSOLE: That will be a nominal 25 percent?
13 Approaching the torque switch setting.
14 MR. MICHELSON: Is that the typical margin 15 between needed thrust and available thrust?
16 MR. CHARBONNEAU: Between needed and --
17 Oh, no. The required thrust to open or close 18 the valve typically is much less than the capability of the --
ry 19 MR. MICHELSON: I'll expand it. The thrust 20 required for the worst case --
21 MR. CHARBONNEAU: Is that a known number?
22 MR. MICHELSON:~ That's a number and you provide j'
2'3 an extra.25 percent,. typically there's an extra 25 percent 24 margin available from the operator; is that what you were
( 25 saying?
(~')
l \_/
, 1 ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 80ft3364646
-44 24594.0 167 l BRT-
^[ -) 1 MR. CHARBONNEAU: We typically recommend. We 2 aould set it at 10,001 pound and assuming with dead nuts on 3 it, it should operate each time.
4 MR. EBERSOLE: Is this to say, then, there's an l
5 in-built test that if you can obtain a torque switch trip 6 at X number of pounds and that number of pounds is greater l
7 than the full load requirement, every time you seat you 8 i have a full load test; is that correct? Because you-l 9 illustrate --
10 MR. CHARBONNEAU: I think what you are asking me 11 could this substitute full pressure testing?
12 MR. EBERSOLE: Yes.
l 13 MR. CHARBONNEAU: My response is "yes."
14 _ Absolutely.
15 If you can't verify the manufacturer's number 16 then you better check the containment stress calculation, i
- 17 size calculation,.they are all OA calcs.
t 18 MR. MICHELSON: The key. problem is the needed l
19 thrust you can'=t measure. You can only do that in a 20 laboratory with the pipe broken.and so forth.
l 21 MIL EBERSOLE: You mean to specify. -
22 14R. MICHELSOM' There's no easy way to account 23 for. aging effects, 00 rc ton and so forth, and the torque 24 that might have be,n a e g;ted to the stem packing nut, all i
j 25 ~ this stuff is outside because he's'got a number. You gave
(~~')
\J l
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 4646
.py .
24594.0- ,
168 BRT p-
'\ -)- I him a number initially, 10,000. Maybd the number today to 1
'2' do that job would be 14,000. He doesn't know that. Nobody v,
'3 3 knows that.
4 MR. EBERSOLE: But in the end he's going to
~'
- S' -
5- compress the spring pack at 12,500. ,
6 MR.'MICHELSON: But that's enough. I need 14,000.
7' MR. CHARBONNEAU: If that valve doesn't work,
~8 yourinumber was wrong. I' don't know why it was wrong.
9 If it will make you all get a nice warm feeling, 10 in the very near future we will be doing some full pressure
. J11 testing where we are going to take the calculated numbers,
+
12 okay,1we are going to watch the thrust' signature and find 13 out what was that unseating load. Okay? Now we are going
{}_
14 to match.4t up to the calculated.
15 MR. EBERSOLE: I'was attempted to ride in on the 16 torque switch competence, the competence to get to the 17 torque switch trip as being evidence of carrying a full 18 . load. ( ,
-19 MR. CHARBONNEAU: We can dolthat today. We can
- t-20 do'that today.
21 MR. NICHOLLS: Art, I may retrace you for just a 22' minute to run through the rest of the problems. This :
23- wasn't by design but it's working out as kind of an l 24 illustration for us.
l l 25- The utility, when we get into motor-operated O :'-
ACE-FEDERAL REPORTERS, INC.
. 202 347-3700__ .NationwideCoverase_ _ _, '800 336 6646 __.
24594.0 169 BRT
'l valve. problems, we may have, in our case, in the predictive 2 analysis, troubleshooting mode, we may not be involved in 3 the everyday problem-solving of valves like Limitorque is, 4 like Art is, and we don't become the experts. 12 months of 5 "everything is-happy," then all of a sudden something is 6 unhappy and we have to be instantaneously professionals to 7 figure it out. I think that's almost impossible.
8 So our philosophy is let's get data, let's share 9 data, let's get the professionals to give us input so we 10 are on the right track and not going in circles. This is 11 my' opinion of a common illustration of that.
12 The last one on my list is the valve was s 13 inadvertently backseated. The reply was valve coats 14 allowances were removed and limit switch reset for proper 15 operation at the valve open position.
16 MR. MICHELSON: While we are on this question of 17 sharing data which you indicated, do you have access to the 18 data that MOVATS might have taken from other utilities who 19 have the identical valve and identical application, to see 20 how their results look by comparison with your results? Do 21- you do that sort of thing?
22 MR. NICHOLLS: I'll answer the. question, yes, we 23 do. As a matter _of fact the relationship we have with 24 MOVAT is we do have a sharing situation where valves on the 25 Davis-Besse valves, at other places, have surfaced as a
/
l ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 2 336 6646
24594.0 170 BRT 1.- problem, we are in tune with the MOVATS folks and see how 2 it might apply to corrective actions on our plant.
3 MR. MICHELSON: What I was a little curious 4 about is whether or not utilities when they have MOVATS 5 come in and do this work, consider the information i 6 proprietary and not necessarily available to anybody else.
7 Is there a sharing of this information in the industry is 8 what I'm wondering about. If there isn't, you know, that's 9 a little different consideration.
10 Your experience has been you were able to get 11 information for comparable applications?
12 MR. NICHOLLS: Engineering reviews, discussions --
13 no problem with that as far as MOVATS is concerned, no.
14 MR. EBERSOLE: You must 'be developing more or 15 less proprietary techniques or --
16 MR. NICHOLLS: I'm not sure it's proprietary 17 techniques or a common sharing of common concern of the 18 utility and what the folks going from plant *to plant to 19 plant are deciphering every day that we want to latch onto.
20 MR. EBERSOLE: It's an expert service.
4 21 MR. MICHELSON: But whether or not that service 22 is necessarily shared, generally a service supplier 23 supplies the information to his customer, not to other 24 customers. And if the other customers want it they have to 25- work it out with what:will pay the bill.
b O
- l. ACE-FEDERAL REPORTERS, INC.
I. 202 347 3700 Nationwide Coverage 800 33H646
24594.0 171 BRT' 7:
( l MR. NICHOLLS:- One of the subcommittee members 2 mentioned this morning the need for parenting or the 3 possibility of parenting problems as it relates to valves 4 in general.
5 In my personal opinion, that is very much needed.
6 I think the utilities of course are clannish and we share 7- our problems to try to help each other to get out of the 8 problem before we'get into it. But it has -- whoever the 9 parent will be, whether it's INPO, whether it's this or
-10 whether it's that, has to take in the whole brood, so to 11 speak: The valve, the operator, the electrical aspects of 12 the motor controller, thermal overload design -- the flow 13 characteristics of it, how it is performing in what' system, 1
O 14 et cetera, et cetera.
15 MR. MICHELSON: From your exposure to INPO, had l 16 they seemed to be interested in this area? Taking a lead 17 in it?
18 MR. NICHOLLS: Yes, sir. They did. I'm not ,
19 sure it's a lead, but we are involved with INPO on that 20 very same thing.
i '21 Plus we~use the NPRDS a little bit more ,than, i 22 perhaps, Brian does, since I'm a corporate-person. When-23 -these things come over we circulate it right away to our :
-24 maintenance Staff and also share information with our t
25 subsidiaries.-
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 6646
l 24594.0 172 CRT O)
(_ 1 MR. MICHELSON: Do you search the base directly 2 on the terminal, the NPRDS data base? Do you do your own 3 searching? It's a big data base. You have searched --
4 MR. NICHOLLS: The answer to that is we do 5 search it. We have a group, I'm not sure if they work for
- 6. Ray or not, but they are assigned to maintain the NPRDS 7 system and do various searches for problem areas that are 4
8 the concern of the moment, so to speak.
9 MR. MICHELSON: So you go to them and ask: Who 10 else has seen this kind of a valve problem? And they go
- 11 into the data base to see who else might have reported such.
12 MR. NICHOLLS: That's correct. But the cry for 13 help, MOVATS is frecuently used as well.
14 MR. EBERSOLE: The inpug power to these things, 15 more or less do they not come from standard pieces in 16 another room somewhere which may be overseated or chilled 17 or whatever, so it's a calibration problem. And the time 18 constants on these, I don't know how carefully this is done 19 I for coordination or whatever -- do you look at all this?
20 MR. NICHOLLS: Yes, sir. As a matter of fact 21 the trip current constants, the sizing of the' thermal 22 overloads within.the thermal overload relay, the cable 23 sizing, the EQ' factors as far as the environmental factors 24 for all that stuff --
( 25 MR. EBERSOLE: So it gets involved.
ACE-FEDERAL REPORTERS,'INC.
202-347 3700 Nationwidc Coversee 800-336 4646
.i.
)
124594.0 173 l
-BRT -
/~T A_f- .1 MR. NICHOLLS: Yes, sir.
2 MR. REED: Your. parenting comment, overall i 3 parent.for this fragmented operation and tying it together 4
4 and you mentioned INPO -- I'm thinking of something five 5 years ago and I'm just wondering if we can get a good set 6- of parents.
7 I'm thinking when EPRI did some tests on valves 8- some five years'ago, pilot operator leak valves, and that 9 was going to solve an immediate problem. They were working 10 on relief valves, and the tests -- I don't think they were 11 very good parents because they didn't do tests appropriate 12 to the field conditions. They didn't use borated water, 13 'didn't use hydrogenated. water, and then they just jockeyed 14 the valves back and forth.- That's not the way-they work.
15 They sit there for months: and then they have to work and 16 that's when you want them to work.
17 I was intrigued by the -idea that maybe you need 18 a users 1 group, an owners group. I'd go on to say you need 19 an owner group, supportive group of real genuine, involved
- 20 people from'the workplace.
21 MR. NICHOLLS: . Absolutely.
22- MR. REED: You can't go to the INPO hierarchies 23 .and manipulators of data banks, and come up with'the real 24 Lanswers that.are.needed.
I 25 Be careful. -We should be careful about how do
.1 h
l ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage M336 4646
24594.0 174 BRT-1 you solve this fragmented industrial scene. We don't want 2 more EPRI tests that are not testing the thing according to
.3' the workplace and environment.
4 MR. NICHOLLS: Brian did have a good point. I 5 was a part of the users group reactor trip breakers, and 6 the aftermath to that was very good because it got the nuts 7 and bolts and maintenance-type time people and 8 engineering-type people that are interested in the INPO 9 community to address that issue with common interests. I 10 think that should be addressed, I think, as well in this MOV 11 situation.
I 12 These are additional problems we found:- Torque l 13 switch setting excessive, torque switch setting inadequate.
.O 14 There's a common answer to those two problems that were 15 detected. Based on valve and operator manufacturers' j 16 supplied specifications, and MOVATS' as found data, an ,
l 17 engineering review process was started. After proper ]
18 engineering and setpoint safety review, adjustments were l l
19 made to assure vendor-recommended thrusts were attained.
20 The second thing is something the subcommittee j 1
21 brought out earlier on Brian's talk on torque switch found ;
22 unbalanced. This condition for us is still under review 23 based'on equal torque delivered from setpoint positions in 24 both close and open operations for the same torque' switch 25 setting.
f)/
s_
ACE-FEDERAL REPORTERS, INC. i 202-347-3700 Nationwide Coverage 800-336-66M .
24594.0 175
.BRT
[\
(_/ l~ MR. MICHELSON: Were you able to ascertain any 2 root cause for imbalances in that switch?
3 (Slide.)
4 In other words, I heard this morning that 5 'perhaps you are talking about imbalances of a factor of two
.6 or three, then it has got to be some kind of mechanical 7 damage. Were you seeing imbalances of two or three? And
'8 did you find mechanical damage?-
9 MR. NICHOLLS: We did not find mechanical damage.
10 MR. MICHELSON: Did you find that large an 1
11 imbalance?
12 MR. NICHOLLS: As close as I can recollect, the 13 imbalances were not that significant to affect the
- bf-s 14 operation of the valve, damage the valve.
15 MR. MICHELSON: Were they large?
16 MR. NICHOLLS: They were small. But, still and 17 all, it was there and root causes.is what we were after.
18 If it's something in our maintenance-practice, 19 if it's something that some other valve manufacturer, the 20 ' tolerances, the OA program as it finds its way through the 21 community up to where it's installed -- I'm not sure of the 22 root cause.
23 MR. EBERSOLE: From a physical standpoint I
.24 don't see what the concept of having equivalent torques is.
25 I can imagine -it must take a higher . torque to go in and
,~
. h.
ACE. FEDERAL REPORTERS, INC.
202 347-3700 NationwideCoverase - 800 336 4646
24594.0 176
. B RT -
(~%.
\_) 1 grind into the seat than it does to come off of it?
2 MR. NICHOLLS: Some of our valves do have the 3 same torque setpoint for open and close.
4 MR. EBERSOLE: Why?
5 MR. NICHOLLS: I can't answer that.
~
6 MR. EBERSOLE: That's what I'm trying to get at.
7 I don't see unbalance per se --
8 MR. MICHELSON: It really indicates a lack of 9 calibration. If one in one direction produces 8000 thrust 10 and one in the other direction produces 3000, there's 11 something wrong with the calibration of the torque switch.
12 MR. EBERSOLE:' There's hysteresis in everything.
13 MR. MICHELSON: It isn't hysteresis. There's 14 something wrong --
4 15 MR. NICHOLLS: Personally, I'm not -- from an 16 engineering standpoint '- realize the significance of the 17 unbalance. Do you?
18 MR. CURRY: We set it the same for the closing 19 and opening balances. The torque switch has no. active 20 function, it's a protective function.- And we see no reason 21 'why it should be higher in the open direction because the 22 torque switch bypass gets it out of the seat without having 23 an active function.
24 Once it is somewhere in mid-stroke, if something 25 should happen that the valve binds up, let's have it O
l
- ACE-FEDERAL REPORTERS, INC.-
! 202-347 3700 Nationwide Coverage 800 33H646
24594.0 177 B RT -
1 stopped and, you know, conditions may. change where we would 2 be able to operate and we will not damage any equipment
- 3 that way.
4 MR. EBERSOLE: That's an interesting 5 philosophical point. I can hear someone say if it binds 6 mid-stream I'll burn it out to make it go there. So you 7 have to make that resolution. I'll literally burn it up to 8 try to make it go. You know, like the landing gear on an 9 airplane, at one time, which had fuses in the circuits to 10 get the gear down.
11 (Laughter.)
12 MR. EBERSOLE: You may want to burn it up.
13 MR. NICHOLLS: MOVATS brought this to our 14 attention. Can I let Art address his viewpoint of this 15 -issue on spring packs?
16 MR. CHARBONNEAU: Do you mean that or on the 17 unbalance?
18 MR. MICHELSON: Unbalance.
19- MR. CHARBONNEAU: Couple of ways you can 20 unbalance it, the 3 to 1 you were talking about,_we have 21 seen them 10 to 1, a massive unbalance. There's a real 22 slick way to do it. All you have to do is shut certain 23 . types of Limitorque operators, close them into the seat, 24 and then decide you want to-check the torque switch to make 25 sure the gears are not broken, so you take the fork switch
, !"h
! U l-i ACE. FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 33H646 l
-24594.0 178 CRT.
p ls 1 out.and you check it and you put it back in.
2 If you had taken that torque switch out and it 3 was loaded exactly the way it was supposed to be, you took 4- it out and unloaded it and you put it back in relaxed so 5 the next time the spring pack relaxes, it's way out of I6 whack.
7 The reason we do it at, let's say, 1 and 1, we 8- determine the motion - that you get for the torque switch in 9 each direction at an equal setting. Once you know that you 10 get equal motion, then you can believe your readings. If ,
11 somebody puts -- unbalances a torque switch and then goes 12 to Limitorque and says, tell me how much load I get at a i
13 torque: switch setting of 2, and Limitorque dutifully tells 14 them 10,000 pounds, they go back and set it at 2 and bend 15 the stem. Then they.go back to Limitorque and say, I did 16 _what you said, I bent my stem, and Limitorque looked at the 17 -stress calculations and says, you can't, if you have 10,000 18 pounds.
19 That's one way you can do it. You can have rack 20 wear. Thst's why they have the balancing capabilities on 21 certain torque switches,'so it's_there to make adjustments 22 as time goes on or if you are replacing a torque switch and 23 on the SMB-000s and most double Os, it has a little leak, l.
24 so inadvertently if something does'get bent one way or the 25 other, that will throw a torque switch easily out of .
(~s')
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336 6646
i 24594.0 179
- BRT:
/~'T - 1
'V 1 balance.
2 MR. MICHELSON: So you can get this way from 3 damage but you can also get it from an improper maintenance 4 operation, namely removing the switch when it's in the 5 fully seated-position.
6 MR. CHARBONNEAU: That's right.
7 MR. MICHELSON: That would also throw it off, 8 then, even though there's no physical damage.
9 MR. CHARBONNEAU: I could give you five other
'10 ways, too.
11' MR. MICHELSON: I expect, thank you.
12 MR. NICHOLLS: Apparent excessive spring pack 13 preload. Engineering and maintenance were made aware of 14 the importance of improper preload and the problem that 15 could develop with the spring pack preload misadjusted 16 during maintenance. The proper corrective action was taken.
17 The next problem that was uncovered was 18 potential valve seating concerns. The signature revealed 19 non-smooth seating of valve components. Maintenance was 20 - performed and the seating problems -were corrected.
21 MR. EBERSOLE: Are these spring packs sealed or 22 heavily lubricated so they don't rust solid?
23 MR. NICHOLLS: To my knowledge they are lightly 24 lubricated and they are sealed.
25 The other problem was high motor current, 150 g
V ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 3364646
.w. . - . . - . - . - ~ .-- . - . --.. .. ...-. , .
-24594.0 180 1BRT
' f }
percent 1of' rated.' To'us this gave us clues to possible ;
n D _
1 j.
2' gear problems, motor: bearings, and a possible electrical
, , ~3- degradation. The problem was found and. corrected.-
i 4- MR. EBERSOLE: Something occurs to me, I might 1 t'
4
'5' as well mention'here as well as anywhere. There's quite a I
6 few; places where everythingf looked good except the disc.or
. ^ 7- gate was not' moving. Do you have something'in your program i.
' It'was detached, in other
- 8- . where you are looking at that?
d P
! 9 words.
f v 10 .MR..NICHOLLS: Yes. . I think part of our.out' age :
11- requirements are type C testing of valves. I
).
y-
! =12 ' Spring gap readjusted to proper clearance after 13 verification of properipreload setting.
l
~
[
14 (Discussion off-the record.). t 15 MR. NICHOLLS: Spring pack, high thrust running .
l 16 load,' operator. gear wear 1-- I did not 'research to find out l
i -
4' 17 which one it was on this particular problem. -
l- 18 MR. MICHELSON:. Do you know of any cases where
\ -
l 19 _ there was damage to the valve ' that you, for someLreason, l
20 didn't pick up? What kind of d,amage might that --be?
.' . 21 MR. NICHOLLS:
You mean by moving the MOVAT 22 system?
p 23. .MR. MICHELSON: .In-your'MOVATS experience to
-:24' date,.have you foundLthat something -- that a-valve.seemed 25 to pass the test and.yet later you found--it damaged?
o 1 I ACE. FEDERAL. REPORTERS, INC. .
_ _ _-_ i _____ MMM m_ __N M c semew__ !
24594.0 181
.BRT
_p
= (_/. 1 MR. NICHOLLS: To this date, no. To this date, 2 no.
3 MR. MICHELSON: So it spotted all the known 4 problems to date?
5 MR. NICHOLLS: To date. .I might address 6 something that'i;s in the presentation a little later, is 4
7 that we've only tested 35 valves with MOVATS.
8 MR. MICHELSON: 35. The sample is still small.
9 MR. NICHOLLS: It's a sampling performance of 10 MOVATS. The other problem was operator gear wear, that was 11 noted, and the signature showed that there was operator 12 gear problems and this condition was investigated and noted 13 for future reference; and comparison of the in situ l
14 signature analysis.
15 MR. MICHELSON: How does the gear wear show up 16 in the signature? .
17 MR. NICHOLLS: Erratic indications in the thrust 18 indications. On the signature itself.
19 MR. MICHELSON: Does this really say " gear wear" 20 or does it say "Here's the possibilities."
21 MR. NICHOLLS: I think it's the clues, the 22 possibility. You look at the parameters of the 23 possibilities'and do a probability number.
24- MR. MICHELSON: At least it tells you you better 25 tear it apart and take a look.
l'~1 U
l ACE. FEDERAL REPORTERS, INC.'
202 347 3700 Nationwide Coverage 800 33H646
24594.0 182 BRT
,~m
(- 1- MR. NICHOLLS: At least. Yes, sir.
2 Kind of a summary.
3 (Slide.)
4 All the as found problems were corrected in a 5 timely manner. Which is good, because you do have all --
6 you have the mechanical group, the operations group, you've
'7 got.the electrical group, and then you've got the operating 8 test group all involved with the different departments.
9 Now we could have a signature that could be shared with 10 these folks without going word for word. The timely manner 11 was interesting to us, in using the MOVATS system.
12 Post maintenance data on these valves provided gg 13 as lef t performance conditions for. engineering review. And V
> 14 for future comparisons.
15 From '83 to '85, the MOVATS folks assisted in 16 corrective maintenance efforts on a total of 35 valves.
17 And the last comment we appreciated as a utility, 18 and it's not taking anything away from our Limitorque 19 friends either, but the MOVATS experience provided added 20 insight on motor-operated valve maintenance and additional 21 details to our maintenance procedures. We enjoyed that 22 because they are doing that every day. Brian and I -- I 23 can't speak for Brian, but I don't get involved with 24 valaves until there's a problem. Then I have to become an 25 expert and I can't, so I rely on the experts to help.
O ACE. FEDERAL REPORTERS, INC.
[. 2 347 3700 Nationwide Coverase M336 4646
1 1
i 24594.0 183 BRT O.,
V 1 (Slide.)
2 This is'just two brief areas that perhaps could 3 be considered possible corrective action and improvements.
4 Follow-up studies on MOV performance 5 characteristic changes that may occur during the MOV's life 6 cycle. From the cradle to the grave, so to speak.
7 Detailed instructional guideline improvements 8 from the operator and the valve manufacturer to produce a 9 more detailed maintenance procedure.
10 'Quite frankly, I'm not sure if this is generic 11 or what, but a lot of times the vendor folks do leave a lot 12 of in-between-the-line interpretation in trying to decipher g~s 13 the meaning of their manual and to try to.get detailed O 14 instructions for the performer of the maintenance, the ,
15 fellow that's down in the trenches going through the
. 16 guidelines to where he's got guidelines plus he should have 17 that educational tool right in front of him when he's doing 18 the work so he can provide a better job.
19 (Slide.)
20 These were just a few things we listed that may 21 affect the life cycle of the valve Operator manufacturer 22- tolerances, operator application to valve type and size, 23 transportation, storage, installation during construction, 24 maintenance practices, proper engineering applications, 25 environmental factors, frequency of use factors -- just to rn ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverses 800 336 4646 !
24594.0 184 BRT
-(> 1 name a few of the things the valve has to attend to as it's 2 on its way through its life cycle in the plant system.
3 (Slide.)
4 The possibility of future use of the MOVATS 5 system, the initial operation's predictive maintenance 6 program for rotating equipment. I would like to share with 7 you just a little bit of what we are doing with the program 8 that's about 18 months old.
9 (Slide.)
10 In predictive analysis for Virginia Power 11 totaling four operating units.
12 We are trending over 2000 data points each month.
g 13 We are involved with approximately 400 rotating pieces of 4
s /
14 equipment, components we are looking at each month. We are 15 also doing research and development to improve our ability 16 to identify equipment component problems using signature 17 trend data.
18 We were doing this before the MOVATS system came 19 and that, again, is what attracted us to the MOVATS system, 20 the documentation of signature analyses.
21 MR. MICHELSON: What signatures were you 22 trending?
23 MR. NICHOLLS: We were trending reactor coolant 24 pump --
25 MR. MICHELSON: By signature, were you trending x_-
)
ACE FEDERAL REPORTERS, INC.
, 202 347 3700 Nationwide Coverage 800 336 6646
'24594.0 185 BRT
/
~ _)
(T I the motor current?
2 MR. NICHOLLS: Sure. We are trending the 3 discrete frequency analyses in vibration amplitudes over a 4 spectrum from the accelerometer data, from oil analyses 5~ data --
6 MR. MICHELSON: Oil?
7 MR. NICHOLLS: Yes, sir. From temperature data, 8 from observations made by the collectors of data who are 9 also the interpreters of the' data, so you have three or ,
10 four parameters.
11 You have discrete frequency as-far as vibration i
12 information; you have oil analyses, temperature, and-
~
13 observations.that are routinely done on all these' pieces of 14 equip, ment.
15 The research and development part of our program 16 is to be able to grow to a point where we can identify the 17 discrete components within a component. _For example, on 18 reactor coolant pumps, we are able, now,-to pick out the 19 rotor bar frequencies of the rotor, and identify what they 20 are. We can track those if they are getting into problems 21 and we are learning every day how to identify other 22 components within an element of a piece of equipment and 23 trend it and'know exactly what that is. So if it starts 24 showing up as a bad guy, we can check for spare parts,-
25 trend on when its failure mode might be critical, so forth c
q
%j l
[
ACE-FEDERAL REPORTERS, INC.-
202-347 3700 Nationwide Coverage 800 336 6646
24594.0 186 BRT
(~h '
() 1 and so on. !
2 (Slide.)
3 This is just to share with you the rotating
- 4 equipment lab philosophy. It has been my philosophy over 5 the years that when you are doing trend analyses and you 6 are doing-that type of thing, you need a proper atmosphere 7 to work in. It can't be in the electric shop or in the 8 maintenance shop where 50,000 other things are going on.
9 This is just a sample of our two young men setting up data.
10 The headphones they are wearing are radio headphones 11 because we believe if there's good communications in the 12 field when the fellows are out getting the data off the 13 D~j equipment, whether it's the MOVATS equipment, whether it's 14 the charging pump, et cetera, good communications between 15 the data collectors is important. Plus we take the i
16 approach of taping all this discrete data. We do it on a 17 tape recorder.
L 18 This is an example of one of our research 19 projects where we are taking frequency measurements on the 20 performance of this circulating oil pump plus the oil 21 samples that we take every month inside the transformer 22 itself.
, 23 (Slide.)
24 The whole purpose of mentioning this is to share 25 with you that the MOVATS system in our opinion has trend A
V ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 6646
24594.0 187 ,
BRT.
1 signature capabilities that I think we can learn from and 2 identify a lot of problems from.
3 This is just a simple approach to our data 4 collecting with accelerometers and so forth on this 5 particular motor.
6 (Slide.)
7 This is a typical lab setup with' computer data 8 . trending devices, tape recorder for the original data, so 9 forth and so on. This'is located, by the way, at North 10 Anna power station.
11 (Slide.)
12 This is a young man processing the data similar i
p 13 to what Art does.in his lab when he gets this signature O 14 data, and processes it through XY plotters and has his 15 techicians get down and then sits in with us on 16 consultation on what the possible problems are..
.17 (Slide.)
18 We did win "INPO Good Practice" this year on our 19 predictive rotating equipment program for signature 20 analysis. We are quite proud of that.
21 (Slide.)
22 The MOVATS possibilities in the predictive 23 maintenance program: Documented signature trending; post 24 maintenance testing; problem solving aid; document for 25 engineering performance review.
O ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 4 646
24594.0 188 BRT
(~)
\_) 1 This last statement is not meant to be 2 melodramatic at all.
3 (Slide.)
4 It's something that came out of my heart when I 5 was writing a conclusion to this. That's the human factor 6 that's involved in all of the things we have discussed this 7 morning.
8 The industry training programs, procedures, 9 vendor instructions and proper motor-operated valve 10 application and operation still is shadowed by the human 11 factor. Documented signature analysis to measure the 12 performance of equipment with independent engineering 13 review will assist the nuclear industry in achieving the 14 excellence I think we are after.
15 Everyone in the room has different opinions 16 about everything all the speakers have said, I'm sure, but 17' when you can document, on a repeated way, signatures from-18 equipment and let the experts. argue and discuss and debate 19 and finalize trending to identify problems, no matter where 20 they are created, we think this mechanism is -- has promise.
21 MR. MICHELSON: It might also be promising in 22 that you can presumably go to other plants for comparable 23 valve and comparable application, get a signature which 24 then can be compared with the signature in your plant. And-25 then of course you sometimes -- sometimes there will be O
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33641646
h
.24594.0 189 BRT
'(p[.
l differences that will be explainable. Sometimes they will 2 indicate there's something funny about your valve or the 3 other valve that needs to be looked into, because of 4 differences in the characteristic signatures.
5 MR. NICHOLLS: The compiling of a data base 6 interpretation, fr6m the things-Brian is doing and MOVATS 7 is doing around the country, I think, is important. And it 8 will help escalate the root cause detection.
9 MR. MICHELSON: Do you have any personal views 10 on this question of: Well, what's the best way, now, to 11 benefit from the lessons that we think we can learn and 12 also to formulate some reasonable proposals on what ought fs 13 to be done about the situation, if anything. Of course, t
14 first what we have to do is kind of decide if it's a 15 problem and if so, how big a problem, and then what should
-16 be done about it? Do you have any views on how this could 17 best be done?
18 MR. NICHOLLS: How we.can best approach as an 19 industry the motor-operated valve problems in general?
20 MR. MICHELSON: Yes.
21 MR. NICHOLLS: First thing -- this is my opinion, 22 again -- it puts us in a position where the child can 23 perhaps choose the proper parent so we can get the proper 24 results. I think that's the first step.
25 Then when we have elected the proper parent, O
t i ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 4646
24594.0 190 BRT.
f
(.}./ 1 whoever that is, then that parent can then start collecting 2 'this data base in order to make the contributions I think 3 are necessary.
4 MR.'MICHELSON: One suggestion this morning of S course was that you gather toqether as children and don't 6 select a parent but rather have a working group.
7 Anot.her proposal I think you have just made is 8 you ought to gather together and see to it'that an 9' appropriate parent is selected to do it, to do the work.
10 That's a little different approach, I think.
11 MR. NICHOLLS: It has been my opinion that the 12 arms do a good job and the legs do a good job and the feet 13 do a good job and so forth, but unless it has a head to put 14 it all together, it's difficult to walk correctly.
15 MR. MICHELSON: The head is the key question. <
16 You don't think a working group, as such, would be a good 17 head?
18 MR. NICHOLLS: The working group of the proper 19 selected parenting types would be a good approach. At 20 least a good start. But it would have to be almost like a 71 commission-type thing, or like a -- like the working group 22 that has a sole person who is responsible.
i 23 MR. MICHELSON: If you didn't do that, then the 24 electives are not -- the alternatives are not too many.
25 INPO might be an ' alternative, or EPRI might ce an I
($).
- ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 4100 336 4646
i 24594.0 191 l BRT l n
(_) 1 alternative. Have I missed any?
2 MR. NICHOLLS: I don't think so.
3 MR. HARDWICK: This could be an issue that the 4 Newmark subcommittee on technical issues might address in 5 some form after they have some experience in setting 6 themselves up.
7 MR. MICHELSON: Which subcommittee is that?
8 MR. HARDWICK: Newmark has one subcommittee. It 9 was referred to as a hardware issues subcommittee but now 10 it'has the name of a technical issues subcommittee.
11 MR. REED: One thing that brothers me about that 12 is how parents get to set the way in which things will 13 unfold without the input of children, to use the words.that O 14 are being thrown around. .
15 Newmark is a group of vice-presidents. This is 16 an issue close to hardware and the workplace and the 17 wrenches, nuts and bolts. It almost seems to me that 18 somehow the children that are out there, we'll call them 19 the children out in the workplace, should -- I'm sure they 20 know what the problem is. They recognize the coordination 21 issue, they recognize the fragmented industry valve 22 production situation. Perhaps they recognize most of the 2:3 problems but perhaps not some of the subtle engineering 24 problems. I say that because I think of pilot operator 25 relief valves have a very subtle hydrogen boron release l
l l
l l dCE-FEDERAL REPORTERS, INC.
2 347 3700 Narionwide Coversee 2 336-6646 l
24594.0 192 BRT 1(_j 1 problem, which parents aren't addressing, not even the 2 French in their socialized electric utility industry.
3 I'm intrigued by Mr. Curry's statement. Maybe 4 if you got the children together that really knew what they 4
5 were talking about for a while and let them try to scope -
6 out_the problems, and I'm sure they'd come out with about 7 90. percent of the things, issues and problems; then maybe 8 the parent selection could come along to who is going to 9 push it, who is going to put it together, how is it going s 10 to be financed, supported, and the emphasis and so on and 11 so forth.
l 12 You know, some of these areas like health 13 physics have all kinds of meetings and support and this, O 14 that and the other. But valves just!never got any 4
l 15 attention.
16 .MR. NICHOLLS: I agree. In my opinion, it
, 17 should start off with the_ users' steering committee to 18 decide what kind of direction we should take as far as the 19 parenting group is concerned. 'La getL the field input, you 20 get the guts and the nuts and the bolts input of what 21 everyday problems are that face MOVs.in the application of 22 their performance in the system; of their application of 23 the-periodic tests they go through every month.
24 It seems like, to me, that the field folks 2
25 should be part of that thrust, to get what's going on in 1
, sj ACE-FEDERAL REPORTERS, INC.
202-347 3700 - ' Naionwide cowee: 800 336 4646
-24594.0 193-CRT (O_j 1 the field and not what's going on in the chairman's of fice, 2- so to speak -- or an area that's not where the hands get 3 dirty.and knuckles get busted. The real world is where 4 problems get identified.
5 MR. WARD: How good do these valves have to be?
6- How do you know when you are finished with your improvement 7 program? Have you -- what are the goals, and why a I
8 specific goal?
9 MR. NICHOLLS: We really haven't set performance 10 goals and standards as yet. It's too early for us in this 11 program.
12 MR. WARD: Do you think you might? .Are you 13 aiming for that?
O 14 ,
MR. NICHOLLS: I would think so. There should 15 be goals. You have to have goals to shoot for.
16 Particular types of valves, how many problems 17 reoccurred over a period of years on that particular type 18 or valve; was it a common maintenance problem, ,
19 manufacturing tolerance problem -- I think those things 20 have to be tracked, traced, analyzed. ,
21 The only way to get, in my opinion, good data 22 like that is to get good field-based data that can be- ;
23 looked at from the industry standpoint to make the proper 24 improvements-for valves and operators and electrical stuff s
25 that supplies those things.
O ACE FEDERAL REPORTERS, INC.
. 202 347 3700 Nationwide Coverage 800-336 6646
24594.0 '194 BRT O
v 1 MR. WARD: But how good do they have to be? Are 2 you going to be influenced by the sort of thing we heard 3 this morning, about what the risk assessments are showing 4' about the need for approaching perfectability in valve 5 performance?
6 MR. NICHOLLS: .The word " excellence" popped into i 7 my mind. In my mind, if we are dealing with thes,e safety 8 systems and the protection of the public and workers, that 9 system has to perform. How bad can it operate or how bad 10j b can-we make it' operate and say well, it still operates 11 already -- to me that's not acceptable. You have to say 12 it's the best it can be and if it has a little glitch 13 somewhere along the line, let's identify it, let's take 14 care of that so it can be better than it can be.
\ 15~- '
MR. WARD: You say if there's some Gort of 16 'lperformance goal, you sort of see it developing from maybe lh \
the inherent limitations of the equipment but right up to 18 that? Rather than a performance goal based on what a PRA 19 tells you is needed in a given system? That's the sort of 20 thing you are thinking of?
21 'i [' MR. NICHOLLS: Yes, sir. Right.
22 > The safety-related function of motor-operated 23 valves, it seems like the field data that can come into 24' signature analyses for the engineers, both in the 25 Limitorque camp, and the valve camp, and the design v _,
lM ! s ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverses 800-336 6646
24594.0 195 '
RT 1 engineers, system design engineers' camp, can take a closer 2 look at these valves on how they really are performing in 3 the field to make the assessment on whether their 4 assumptions'and calculations were correct, whether the 5 valve application was correct.
6 Someone made a mention this morning, I thought 7 it was interesting, that one utility doesn't have torque 8 switches, doesn't have this, doesn't have t. hat. It's 9 designed to operate no matter what in the safety direction.
10 You know, there are things to be considered too.
11 MR. MICHELSON: Any other questions? Well, we 12 thank you very much. It gives us, I think, even more i
O 13 insight to work with and we appreciate your time. Thank 14 you.
15 I believe the -- is Research going to have 16 somebody that wants to make a speech first?. Jeff, did you 17 want to say a few words first?
18 Jed is going to introduce the subject of NRC 19 research-related activities on valves. I think you are 20 going to bring us up to date a -little more on what's 21 happening and then I believe we will follow.that with some 22 presentation of Oak Ridge people.
23 Jed, you can kind of watch the clock.- We have 24 time but our break was scheduled for about 3:50. 'I'd just-25 as soon take a break a little earlier than that. If
{~s').
~
l l- ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-33H646
.i e
24594.0 196 BRT
'\
f_/
x_ ,
1- there's a convenient place along the way to break your 2 presentation, or just beforeLOak Ridge starts or whatever, s: 3 'let me know.
'4 MR. VORA: My presentation will last, 5 Mr. Chairman, a. couple of minutes.
6 My name is Jed Vora and I work for Dr. Bill 7 Morris in the division of engineering technology of the 8 . Office of Research. My responsibility is with the planned 9 aging research program, which includes the electrical and 10 mechanicals.
11 During the last meeting we. had an opportunity to 12 present to you and discuss _with you the various areas' 13 programs related to the va'1've operability and reliability.
14 Since then, we actually added two research 15 activities, the one with the PORV, and the other is the 16 valve operability for. pipe isolation.,
17 MR. MICHELSON: You said those were added since 18 the last presentation? ,
19 ,
MR. VORA: .This'is correct.
20 ^ MR. MICHELSON: Are you going to tell us a 21- _little about those?
22 y /MR.~VORA: Dr.'Weidenhamer will give us some 23 '(nformation about the qualification prcgrams, and there's a 24' .part of the~ aging research program- where you are looking g- 25 into the diff'erent valves, the SOVs, MOVs, PORV and BVs.
U ACE-FEDERAL REPORTERS, INC.
Xi2 347-3700 Nationwide Coverage 800 336 6646
24594.0 197 BRT Ny) 1- The evaluation of MOVATS as a signature analysis 2 technique to detect the aging degradations and other 3 abnormalities ,in motor-operated valves will be' discussed by 4 Dr. Eissenberg in much detail this afternoon.
5 The various research activities within the EEICB, 6 I think there was some interest, Richard mentioned to me 7 there might be interest about the funding levels. It's 8 approximately $400,000 for all the valva-related activities 9 in the EEICB, as part of the aging program.
10 MR. MICHELSON: That's about the same as it was 11 before.
12 MR. VORA: Approximately. This is correct.
13 MR. MICHELSON: But you added the PORVs since 14 then. After you tell me.what that program is, tell me what 15 else will suffer when it gets done.
16 MR. VORA: After 1986 that's approximately 17 $500,000 and approximately $650,000 for 1987.
18 This thing is primarily detected by, according 19 to the aging program strategy, we'll finish the Phasc I 20 assessment of motor-operated valves and we'll issue a 21 technical report. We are also about to finish the report 22 on~ check valves. So, from Phase I, actually on a 23 transition . stage going from Phase II, where we are doing 24 some in situ monitoring and assessment and also the
(
e .25 evaluation of advanced monitoring and detecting techniques.
(m)
Ace-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage ~ 800-33H646
24594.0 198 BRT
(+
\~)
1 This is just an overview of the various 2 valve-related research activities.
3 MR. EBERSOLE: Let me ask you a question about 4 integration of various efforts. Valve operability for pipe 5 isolation, I take it that's when the pipe is split or 6 brdken. Off in the shadows is a thesis coming forth that 7 we know Jots about pipes, 106-A and -B perhaps that inn 8 didn't used to know, and through fatigue analysis and a 9 number of sophisticated techniques we are about to enter 10 into a regime where we'll invoke the fact that pipes will 11 never break before they leak first.
12 What's your own view about whether this
\
13 requirement will persist or not? That we have to consider 14 breaks or --
15 MR. WEIDENHAMER: We are trying to address the 16 concern whether pipe isolation valves will grow against the 17 dynamic heads when we do have a downstream break, and also 18 whether or not the valve will open if we have to inject 19 Luto the piping system.
20 MR. EBERSOLE:- I'm saying, though, this 21 hypothesis that we'll have to treat the dynamic flows from 22 fully broken types may disappear. I don't know.- I would 23 like to keep this-logic here for a long while yet.
24 MR. MICHELSON: If they disappear, we won't need 25 any more ECCFs either. A little job key pump we have in l
ACE-FEDERAL REPORTERS, INC.
I 202-347-3700, Nationwide Coverage 800-33(Hi646
24594.0- 199 BRT O
1 the yard will be plenty enough for that little leak, if you
~
2 have'that trend. Man, we just --
3 MR. EBERSOLE: You have to keep it in mind. It 4 may disappear. I don't know.
5 MR. MICHELSON: I'm not sure.
6 MR. REED: Jesse likes to talk that way because 7 he's a boiling water reactor advocate, but I'm not going to 8 let it disappear for boiling water reactors.
9 MR. MICHELSON: You are right, though. It is 10 something that has to be thought about.
11- MR. REED: You are damn right.
12 MR. MICHELSON: Perhaps the full guillotine 13 break isn't the rate rationale outside a container either 14 but we have had too much experience with broken pipes 15 outside of containment to believe that they will never 16 again break.
17 MR. EBERSOLE: We are going to assume for valves 18 they will break?
. 19 MR.'WEIDENHAMER: We believe right now. But
- 20. it's in the planning stages. I'll discuss that a little 21 more.
22 MR. VORA: As far as the evaluation of MOVATS is 23 concerned, the objective of this program will be to learn 24 exactly what the equipment can provide about safety-related l r3 25 MOV operational readiness above and beyond the currently l
i
(/
l ACE-FEDERAL REPORTERS, INC.
.202-347-3700 Nationwide Coverage 800 336 4646
. . ~ _.
1 l
24594.0 200 l l
,._B RT t \
\_/
1 used ASME section 11 methods.
2- Dr. Eissenberg will discuss this particular 3 issue in detail in his presentation.
4 MR. MICHELSON: One thing that perhaps you can 5 clarify, or somebody else, and that is that there's nothing 6 in the MOVATS system that prevents me from setting it up 7 and operating the valve with full flow through it and going
~8 through its normal -- while we are monitoring -- there are 9 no operational limitations as to what the pressure, 10 temperature is, is there? It can be done under fully --
11 fully operational conditions in the plant? Thank you.-
12 MR. VORA: We reviewed two months ago all the (3
\/ 13 operating data, the nuclear plant experiences -- and we are 14 just preparing a report on it and we are beginning two NRRs 15 toward one of the. tasks, towards one of the issues, generic 16 issue 70 --
17 MR. MICHELSON: When_will that report be.
18 available?
19 MR. VORA: In two to three weeks.
20 MR. MICHELSON: Okay. I guess we'll be sent a 21 copy of it.
22 MR. VORA: Just to give highlights, there was 23 much discussion this morning ti.at we found out about:230 24 reportable events associated with PORV blocked valves. 10 25 involved mechanical failures,_91 involved PORV failures, g3
, N/
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage - 800 336 4 646
24594.0 201 RT 1 30 involved the blocked valve failures and about si'x of 2 them were design-related events.
3 The report will further identify the failure 4 modes, mechanism, failure severity, and how many PR and 5 blocred valve failures were involved, type of failures --
6 we'll have nine or 10 specific points that have been 7 discussed in this report. We should be available -- it 8 should be available, I think Gary Harmer and Frank's groups 9 are monitoring this program with us.
10 MR..EBERSOLE: I think I saight mention something 11 since you talked about PORVs, we just came back from 12 Davis-Besse yesterday and one of their defenses was, oh, we
('- ')
13 could have invoked lead B, when they lost the boiling water.
14 I want to call out while these valve people are here 15 because I think as you all know there's a flaw in the logic.
16 They thought they would prop open PORVs, pump in primary 17 water, and keep on cooling that way without secondaries.
18 As you all know, the PORVs have never been environmentally 19 qualified for the very atmosphere they produce themselves 20 and I think this is a significant obvious flaw in this 21 logic that ought to be fixed.
22 MR. VORA: I think this is part of the program 23 we are going to address; is that right? George? You are 24 going to discuss that?
,~' 25 MR. EBERSOLE: The PORVs themselves, for that L
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
24594.0 202 BRT l'~
\-))
1 matter, are' lousy valves when used for this purpose and we 2 need new and better valves that will handle two phases, and 3 do so for equivalent reliability in both open and closed 4 corrections.
- 5 MR. CHURNEY: Maybe I could make a comment on 6 that. Frank Churney, NRR. I think the environmental 7 qualifications for the conditions you are talking about are 8 being looked at under unresolved safety issue 845.
9 MR. EBERSOLE: That's the common pot that 10 everything is being looked at under. It's so lost in that 11 vast pot of considerations that, you know, it's like a part 12 of a --
13 MR. CHURNEY: If it's not looked at under that, 14 it will be looked at under generic issue 70. One of the 15' two.
16 MR. EBERSOLE: You know, there's 10,000 items in 17 it and tends to be a scrap of meat in a Mulligan stew.
18 MR. VORA: I guess with this brief introduction, 19 I would like Dr. Weidenhamer --
20 MR. MICHELSON: Let me ask you a couple of 21 questions because I think that's all the discussion of the 22 budget as I understand it, the amount of money available 23 now is the same as when we discussed it back in the summer?
, 24- June?
25 MR. VORA: As of today, that's-what I know. I fS O
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 33H646
24594.0 203 BRT
(~)x
\_
1 guess that's my understanding.
2 MR. MICHELSON: The final budget cuts didn't go 3 any deeper into your area?
4 MR. VORA: Not so far, but there's a possibility.
5 MR. MICHELSON: We can't speak of the future as ;
6 a whole, but I just wondered if anything had changed since 7 we last heard about it.
8 Okay, that clarified my question.
9 MR. REED: _ Addressing, again, this fragmented 10 scene and how to best bring it together. I just would like 11 to know, and I suspect, your report that you are putting 12 together is based on all the LDR information, NPRDS 13 information and other facts you can draw together? You are 14 not going to try to predict or direct or -- come out with 15 any conclusions? You are going to draw together facts.
16 Because it seems to me that we are a long way 17 from conclusions of how to cure, improve and so on, the 18 whole valve scene; even to the point of how do you 19 structure it and who should be the children' and who should s
- 20. be the parents?
21 MR. VORA: I think to some extent, if you look 22_ into'the aging program as applied to a valve, a 23 . motor-operated. component or subcomponent and you.do the 24 phase assessment offoof it according to the planned aging
- 25
~
-strategy, some of the technical elements should be ACE-FEDERAL' REPORTERS, INC.
202 347-3700 Nationwide Coverage 80 4 336 6646
24594.0 204
.BRT 4 l
tb'
\~
I addressed in the Phase II of the program for that component.
2 When Dr. Eissenberg works on his motor-operated l 3 valves, he.will try to interact with the standards 4 committee, talk with the experts, talk with the utilities, ;
E and those kinds of elements are built into-the program for 6 every component. Did I answer your question?
7 MR. REED: Somewhat. But it seems to me, it is 8 a very complicated and subtle issue, if you take the valve 9 issue across the board, various operators and valves and 10 applications, utilization and performance. And, of course, 11 very fragmented.
12 It seems to me that the, really, the utilities --
() 13 you listen to stories like the Philadelphia people just 14 told and so on, and the motor operator is one part or piece 15 of the valve that a particular kind of valve that makes up 16 the scene, it seems to me that the utilities and MOVATS and 17 Limitorque and perhaps some other air-operated people.or 18 pilot-operated valve people have a loot of the pieces of 19 knowledge which haven't been integrated. It's your work --
20 is your work, your reporting going to try.to address this 21 integration and see if somebody doesn't become a volunteer 22 to try to package the thing?
23 MR. VORA: This is, I think, one of the thrusts 24 of the program at the component level. For a minute I must l
25 mention this to you. For any component with the motor, ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646
24594.0 205 r
b, RT 1 whether it's a motor-operated valve or check valve -- this 2 is what we are trying to achieve.
3 (Slide.) .
4 These are the major goals, to identify and 5 characterize what are the aging and system wear ef fects.
6 When we talk about motor-operated valve we talk about the
.7 boundary of the motor-operated valve which includes the 8 electrical segment, the mechanical segment, and talk about 9 all the interfaces which are associated with it.
10 We are also approaching this, also, from the 11 systems point of view as well as from the component point
, 12 of view, actually establishing what are the temperatures, 13 the pressures, the radiation involved and stresses involved 14 and then identify the aging and characterizing.
15 Then we want to identify what are the methods 16 for inspection, surveillance and monitoring.
17 When we have done the report, we'll have an 18 opportunity to talk to the MOVATS, get into the data base, 19 get all this information into it and come out with a Phase 20 I report and then (y) into the Phase II report for 21 development.
22 MR. REED: In view of the complexity of the 23 overall issue, do you think that you and. Oak Ridge are 24 going to come out with the bible?
25 MR. VORA: But what we are trying to do,
(
} No.
l I ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage . .
800-336 4646 . _ _
i 24594.0 206 BRT
[h w/
.1 sir, is that we are going to take a motor-operated valve, a 2 few motor-operated valves, and demonstrate an approach that 3 how, for-a given environment, for a given design, for a 4 given application, how you can do this characterization for 5 the motor-operated valve.
6 MR. REED: Let me tell you what I'm worried 7 about. Some 10 or 15 years ago in the regulatory -- I'll 8 say "we" -- started to come out with a rule, regulation or 9 findings with respect to the fact that phosphate water ,
10 chemistry ought to be used in all secondary site or steam 11 generators. Fortunately, before that was issued, somehow 12 it got turned around. It would have been a disaster on the 13 reputation of the regulatory to have ever put it out 14 because again there's.a very complicated issue, subtle.
15 Water chemistry and all the witch doctoring that goes with 16 it.
' l'7 I'm not sure the valve thing's not witch doctor.
E
)- 18 It can be boiled down to hard mechanical issues and f acts 19 and electrical issues and facts. But I'm just. worried that 20 you might go beyond your f acts and try to point directions 21 and make conclusiona that you would be embarassed about s
l 22 later without a lot of other input, l
l .23 MR. VORA: I think you are absolutely right. We l
L 24 are very, very cautious. 'That's why we are going through .
i 25 the phased approach, get the-discussion, get the review ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 4 646
r 24594.0 207 RT 1 from the utilities, with the participating organizations, 2 and go very, very systematically. I think you are 3 absolutely right.
4~ MR. MICHELSON: It appears to me that their work 5 is just another input to this owners group or whatever, 6 further information. Certainly it's not the answer --
7 further input-information.
8 MR. VORA: I think Dr. Eissenberg will talk a 9 little bit about the approach as specifically applied to 10 motor-operated valves.
11 If there's no other questions, I would like to 12 have Dr. Weidenhamer talk about the program related to the 13 valve research.
14 MR. WEIDENHAMER: I will be talking about the 15 work associated with valve operability and reliability-16 being conducted in the mechanical /s'tructural engineering 17 ' branch.
18 (slide.)
19 It consists, as you see, . of perhaps five blocks 20 here.- Three of them are under this " qualifications 21- program," which I'll discuss last. There are others -- two 22 others sticking off here by themselves.
23 -This one over here,.which we talked about 24 previously,.is an attempt to -- I believe it's also a
'25 qualification problem, in that how do we qualify valves
-[ }-
! . ACE-FEDERAL REPORTERS, INC.
202-347-3700 ' - Nationwide Coversee 800-336-6646
[_
l 24594.0 208
_BRT
./ \
C/
1 that are to close against these dynamic pressure heads, or 2 to open against these large pressures.
3 The reason I show this off ,by itself is this 4 qualification program is to stop in '87. This is when that 5 begins and is to go in '88. I'm showing it by itself even 6 though it may go under an equipment qualification program.
7 Incidentally, I'll talk about the funding levels 8 at the end of this.
9 Just to finish up this one here, again, it's an 10 attempt to develop the qualification criteria, or identify 11 the kinds of load we should qualify these kinds of valves 12 to under these accident conditions.
13 You identified that about three or four months 14 ago. We recently received a proposal on that. I would 15 like to submit it to you for your review and comment.
16 Without getting any further into that, it's a
- 17 test program and it would consist of doing some of these 18 tests.
19 MR. MICHELSON: It will establish the loading 20 conditions that the valve must meet.
21 MR. WEIDENHAMER: Yes.
I 22 MR. MICHELSON: Where are you going to address, 23 .now, the adjustments to the valve operators and so forth, 24 that must be properly made in order to -meet these load --
(~3 25 once you build a valve for these loading conditions, how do U
I ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverase 800 336 6646
l l
l 24594.0 209 RT 1 you assure yourself the valve will be able to function 2 properly? In other words, how do you assure that it's 3 maintained and adjusted at the correct state? Keeping in 4- mind that this state is far beyond what it normally has to 5 do to pass a surveillance test.
6 Is that a part of your study?
7 MR. WEIDENHAMER: No.
8 MR. MICHELSON: It's sort of a human factors 9 problem, I guess, in a way, or is it maintenance -- is your 10 maintenance program going to do that? Where is that part 11 going to be handled?
12 Because a part of that is how during maintenance 0..s 13 do you know you have adjusted it right and how do you know .
l 14 the valve is still functioning .a such a way as would meet 15 these loads if it ever saw them, which you may not normally 16 see.
17 MR. VORA: One of our strategies would be to 18 coordinate with the work Dr. Eissenberg has done already on 19 motor-operated valves. We have worked together as a team.
20 In fact we are working on the ETEC program for the leak 21 test, we are bringing our-three experts into the test site 22 and then we can coordinate the program so he can provide 23 some of his input and expertise to exactly what your 24-
~
concerns are.
25 MR. MICHELSON: Thank you.
O(~N l
l- ACE-FEDERAL REPORTERS, INC.
202-347-3700 - Nationwide Coverase 800 33HM6
1 24594.0 210 i BRT l C) 1 MR. EBERSOLE: Let me ask a question. The 2 business of starting an aging program must certainly start 3 at the establishment of initial conditions that are 4 satisfactory. Then you are going to look at departures 5 from initial conditions -- that's what aging is.
6 MR. WEIDENHAMER: Yes.
7 MR. EBERSOLE: I just learned a while ago that 8 Virginia found that valves -- they put in straight tubing 9 or something on theirs to isolate them from the presence of 10 condensation one would get in containment. If I were to go 11 out in the field at large today, would I find these 12 so-called qualified valves open, study-type terminal boards
() 13 with the anticipated -- the thesis that this would be not 14 so much condensation and leakage current as would affect 15 the operation of the valve? Or would I find, in fact, 16 hermetic sealing?
17 MR. WEIDENHAMER: You'd find a little of both.
18 MR. EBERSOLE: So it's a rather loose picture 19 we've got, isn't it? I then urge you before you start on 20 the aging program to find out what you have to start with.
21 MR. WEIDENHAMER: _ Yes. Part of this program 22 would be to identify what kind of valves are we getting to 23 look'at -- should we begin to look at.
24 MR. EBERSOLE: Some of the aging effects would g_g 25 be the accumulation of dust and grime and other things that i %)
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverase 800-33MWi46
1
/
24594.0 211 BRT O
1 make a rather adequate terminal board from the condensation 2 standpoint become so dirty as to short it out.
3 MR. MICHELSON: Just for the record, I think 4 Philadelphia Electric was the one that said they had gone 5 back and put the slices ~in and so forth. I don't think we 6 actually asked the people from Virginia what they had. I 7 had thought about it on a couple of occasions.
8 MR. EBERSOLE: Why don't we do that now? What 9 does Virginia Power do? Do they have open wiring on their 10 hostile environment valves?
11 FROM THE FLOOR: I don't believe they are here 12 right now.
> 13 FROM THE FLOOR: There's a point a lot of people 14 are overlooking here. Whether you talk about terminal 15 blocks, and it was Philadelphia Electric talked about them 16 and as Limitorque would speak, you are really talking-about 17 the terminal blocks that are other than the limit switch 18 and torque switch. The limit switch and torque switch are 19 in themselves a terminal block. Okay?
20 MR. EBERSOLE: Are they hermetically sealed?
21 MR. WARSING: They are hermetically sealed. The 22 unit performs throughout the test. I think there's a 23 little misunderstanding here.
24 MR. EDERSOLE: They may well do that.
25 MR. WARSING: That's a fact.- Not conjecture.
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Covernee 800 336 4646
24594.0 212 RT 1 Pardon me.
2 MR. MICHELSON: The problem applies to both 3 types. The problem is more aggravated as you go to 40 4 volts _so'it depends on what your control power is -- now 5 the questions get into, well, did they monitor? Did they 6- apply full voltage to look at breakdown at the terms and 7 . torque switches during the test, and so forth? If they did,
- 8. you have a good answer.
9 MR. WARSING: They monitored through the entire 10 test.
11 MR. MICHELSON: Were they also~ energized?
12 MR. WARSING: Yes, they were.
O 13 MR. .MICHELSON: Then you have a good test.
14 MR. WARD: Was the equipment preheated in the 15 test or was it cold?
16 MR. WARSING: I don't think it was but I can't 17 say positively at the moment.
18 MR. WARD: That's one of the points under 19 contention and that's one of the things Mr. Ebersole is 20- concerned about.
21 MR. MICHELSON: What the Sandia people found is 22 the real problem during the' heat-up cycle,in which_the 23 condensation is occurring, and once they. were up to i
24 temperature, even, the problem disappeared generally. But 25 we have'to be concerned about it throughout the exposure.-
}
i ACE-FEDERAL REPORTERS, INC.
[ 202-347-3700 ' Nationwide Coverage 800 3HWi646
24594.0 213 l BRT y)
\m 1 MR. WEIDENHAMER: This block here, again, is a 2 testing program. It's primarily to -- it deals with 3 checking out some of the code requirements. Perhaps Frank 4 Churney can correct me here after I review what it's really 5 trying to do.
6 We are testing some check and gate valves to
'7 determine whether or not leakage is a way of detecting 8 degradation. If it is, then this testing will provide a 9 data base to determine whether current licensing 10 requirements regarding leak rates can be relaxed.
11 Right now section 11 -- is it section ll? The 12 code allowed a certain amount of leaks.
.f')
k/ 13 MR. SEYFRIT: There are requirements in the 14 standard technical specifications that valves that isolate 15 reactor and pressure boundary from interfacing low pressure 16 systems have to be periodically leak tested to be sure that 17 -the disc is in fact.in place. This is an NRC Staff 18 requirement, not a code requirement.
19 .The code. aspects of that is that most people 20 perform these tests at somewhat considerably lower 21- pressores than what are present when the plants are in
'22 operation and there is a formula in the code that can be j 23 used to extrapolate leakage rates obtained at these lower 24 testing pressures up to what they would be at higher 25 reactor operating pressures. In that sense we are doing ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336 4646
1 24594.0 214
(.BRT U
1 some verification on code formula.
2 MR. WEIDENHAMER: It would confirm whether or 3 not that formula is acceptable for testing leak rate for 4 real operating projections.
5- And then the fourth one is to evaluate the 6 ability of acoustic emission, first to detect the leak and 7 then quantify it.
8 So' there are four pieces of information that 9 they are planning to get out of this testing, at least for 10 check and gate valves.
11 MR. EBERSOLE:- Speaking of acoustics, would it 12 not be at least worth thinking about on every valve 13 operator, having something like a sensitive microphone with i
14 appropriate manipulation of the output to tell you what a 15 valve was doing in a comprehensive se1se? I heard someone
- 16 say a while ago if the valve was dry a nd groaning, it l 17 wasn't in too good shape. It certainly suggests if you had
.18 extended ears on these valves so you could hear what each I 19 one did when you wanted to, it might be worth something.
20 MR. WEIDENHAMER: Yes. I suspect you could 21 learn something from that.
22 MR. EBERSOLE: It was used years ago at the
! 23 laboratory at Oak Ridge, the equipment monitoring technique 24 for machinery that had distinct audio signatures, some g-) 25- complex cyclic pattern, over a matter of'some seconds.
V ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverase 800 33Mi646 .
-24594.0 215 BRT O
1 There's nothing new about it but in valves it might well be 2 worthwhile.
3 MR. WEIDENHAMER: I suspect this is an extension 4 of a pipe crack -- so I think it's an extension --
5 MR. CURRY: Philadelphia Electric is using 6 acoustics extensions to detect pilot valve leakage. If we 7 have one that's going to leak on us, we'll take a look at 8 it and possibly replace it. And know which one.
9 MR. EBERSOLE: Each piece of equipment has a 10 notices signature. I don't know many that are silent. It 11 can be used just like a trace --
12 MR. WEIDENHAMER: This leaves us with the three i
P) k/ 13 boxes here under the qualification program and that's the 14 one I'm involved with.
15 Two, actually one has been completed, the purge 16 and vent valve. I gave you a presentation on that about 17 six months ago. But the motivation for this testing of 18 large butterfly valves, actually some 8- and 24-inch valves, 19 and this containment isolation system valve test. program --
20 I'll get into a little more detail on that -- was really 21 motivated because of the containment integrity people.
22 They wanted to know -- they wanted to get a better handle 23 on what a containment leakage model would be under severe 24 accident conditions..
I ; 25 They believed that instead of the containment l
l
{
ACE-FEDERAL REPORTERS, INC. r 202 347-3700 - Nationwide Coverage 800-336 4646
24594.0 216 BRT-
.g
\~
l building actually breaking, being the leakage problem, they 2 believed that the leakages would most likely occur on 3 penetration.
4 So these obviously are at the penetration area
'S and we wanted to get some ideas o'f how -- what are the 6 leakage characteristics of these kinds of valves under 7 severe accident conditions.
8 So the test program, first of all -- it's under 9 the qualification program. I always have problems trying 10 to get people to understand this ,
11 MR. EBERSOLE: Let ne ask you about old plants 12 now. Prior to 1967, 1967, I was a party to the writing of
( 13 the first safety guide. It had to do with MPSH on pumps.
14 It was found in that particular design, if one lost the 15 partial pressure of the containment atmosphere due to the 16 intransient less while the valve'was trying to get shut, 17 you wouldn't have enough over pressure to have suction 18 pressure on the suction pump uptakes for recirculation of 19 cooling. This implies that some older pipes, older than 20 that vintage, might have a valve problem of critical 21 character because they would lose a fraction of the 22 atmospheric over pressure in the containments and therefore 23 lose the MPSH. Has that been checked out? Do you follow 24 me?
25 MR. WEIDENHAMER: Not really, no.
ACE-FEDERAL REPORTERS, INC.
312-347 3700 Nadonwide Coverage . 800-33H646
_ _ _ - . _. - . - . _ ._ . = . _ - _ - _ ___- .. . _ _ . _
,'I .
j I
2459'4.0 217 BRT-LO 1 MR. EBERSOLE: The - valve would either be slow in
~
f 2 closing and you lose-this fraction of the noncondensables l'
- 3' and lose a critical fra'ction of.the MPSH contribution.
{ .
- 24. MR. WEIDENHAMER: I believe these valves are- ,
i We ;
5 closing very rapidly so you.probably are not losing it. _
i.
- 6 did not consider your -- ,
e l
7~ 'MR. EBERSOLE: S'o one can safely hypothesize-you j
[;- don't lose the noncondensables in older containments. t i
! 9 MR. MICHELSON: Would you refresh my memory for _
j ;
j 10' .a minute. On your qualification program, was that a ,
t .
j -11 performance-qualification as well as environmental?-
i i j 12 MR. WEIDENHAMER: This was primarily a leakage 13 test program. It was to,.one", determine' what are the i
- j. 14 amounts of-leakage underHthe severe-accidents. .
{L 15 MR. MICHELSON:~ It was in the already-closed
!. 16 position ' they were doing the : testing.
And then they were-
! 17- looking.--
) 18 MR. WEIDENHAMER: Before we did that check we >
~.
i 19 actually took it to the - design basis.-level to see if we.
l 20 could'close the valve against the1 design basis pressure ;
21 increase and ' temperatures. And then --
[-
22- MR. MICHELSON . That's.the part ILdon't
, 23 ' recollect you;did, though. .
Because"what facility did you.
1' .
. 24 use!for'50-inch valves,.40-inch valves?- ;
- 25 MR. WEIDENHAMER: We didn't test those I nger -
1
.O.. :
i-t t
? ACE-FEDERAL REPORTERS,-INC.
m m .3 m s u w.ca , m3m -
24594.0- 218 BRT J l valves. We tested up to 24 inches.
2 MR. MICHELSON: You did get up to 24 inch on it?
3 MR. WEIDENHAMER: Yes.
4 MR. MICHELSON: Did you produce a report?
5 MR. WEIDENHAMER: Just been published. I'll 6 make sure you get it.
7 There were a number of other qualification 8 issues that we were looking at in here to really describe,
- 9. this work. It takesmetwohourstodescribethisprogrSb.
10 I can't give you all the details of it but I think the i
11 report would help. If you have any questions I'll be 12 willing to answer them.
() 13 This one here we call the containment isolation 14 system valve test program. It's a continuation of this 15 containment integrity issue: What are the leakage models 16 for these kinds of penetrations.
17 We are beginning-this year, 1986, to perform
-18 what we call the dynamic test part of'this, where we are 19 going to subject the typical containment isolation system 20 valves to severe accident dynamic loads, and then also test 21 the closing, up to the design basis level, then carry it on 22 to the severe accident, to check out the leakage 23 characteristics due to.the dynamic loads.
24 In FY-87, we will use another fixture to
,_ 25 displace a replicated containment building with a piping
\_)
l l
' ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 4646
24594.0 219 BRT:
J l system attached to these valves to determine whether or not
- 2 this displacement, when the force is acted through the 3 piping onto the valve housing will create a problem closing 4 these valves. And, at the same time -- I believe that's 5 all we are going to do.
6 MR. MICHELSON: One of the things you might want 7 to watch carefully, some of these containment isolation 8 valves have been put in rather odd geometries. They have 9 elbows just upstream, things of that sort. That creates a 10 significant displacement of velocity vector across the 11 cross section of the height, thus the loading on the 12 butterfly may be a whole lot different than you would- have 1
() 13 gotten looking at straight runs. It's very sensitive to 14 that and I just want to make sure somehow that can be 15- accounted for, at least in one test, to see how important
~
16 it is relative to thrust required to close.
17 MR. WEIDENHAMER: Okay. That essentially covers 18 the essentials of the containment isolation.
19 This last block maybe should have been, when we 20 went through this six months ago, it was dashed because we 21 didn't know precisely what we were going to be doing in 22 certain areas. Since that time we have identified the-23 major work here related to our related tests so we probably 24 should make this a solid line.
25 We are coordinating some efforts with the HDR t
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
24594.0 220 BRT p
v 1 facility, that's the decommissioned German reactor, and we 2 are planning to insert a typical gate valve into this 3 system and add supports to make it -- to replicate a 4 typical U.S. system. piping system.
5 Here, this is our first chance to get some ideas 6 of what these loads look like from the operation of a valve
~
7 in a real system. What are these vibrations, these 8 flow-induced vibration loads as we try to close this valve?
9 We have no ideas of what those are.
10 So, this is the major part of this related test 11 effort is related to that. As it stands right now we are 12 planning to take part in that test program to begin next
() 13 May, in Germany.. And then there will be some additional 14 testing ' in FY-87.
15 MR. MICHELSON: The test will be in May of next 16 year?
d 17 MR. WEIDENHAMER: May of '86. Then there will 18 be.some additional testing in FY-87. -Budgeting or funding 19 for this work is approximately a little over $1 million per 20 year, from last year FY-85'through FY-88; a little over $1 21 million.
~
22 MR. MICHELSON: Okay.
23 You heard the question earlier,.perhaps, that a 24 couple of the members raised about foreign experience.
- 25- How does research -- to what extent does J
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coversee 900 336 6646
-2 24594.0. 221 BRT O
1 research keep up with foreign valve-related experience of 2 any sort? Or is this an NRR function? Or where in the 3 Staff do people keep up with what the foreign governments 4 are doing in terms of valve programs? Maybe it's AEOD that 5 does it, I don't know.
6 MR. WEIDENHAMER: Well, we attempt to -- well, 7 every year we have this water reactor safety research 8 meeting and the foreigners are invited and we have-the 9 opportunity to present our programs. We discuss our 10 programs with individuals. They present the i rs .'.
11 As far as we know, the Japanese are the only 12 people right now that are really doing anything, or have 13 _done anything in valve testing. We are attempting to get '
14 that information from them in exchange for some of our 15 information. So we are looking into foreign test results.
16 MR. MICHELSON: I was under the impression that 17 the French had an extensive program from talking to at 18 least one person who works on the. program. Have you heard 19 anything about what they are doing?
20 MR. WEIDENHAMER: Yes. As a matter of fact --
21 MR. MICHELSON: It was a long time ago I talked 22 to-them and I don't know whether it fell through or moved.
23 along.
24 MR. WEIDENHAMER: No, I had an opportunity to i
i g) 25 talk to one of them a couple of months ago and I told him I kJ ;
I l
ACE-FEDERAL REPORTERS, INC. .;
l 202-347-3700 Nationwide Coverage 800 336 6646
'Mygg i . !
l :q :
24594.0 222
- BRT
() '
I would send him a copy of my program if he could supply us 2 with something --
n 3 i MR. MICHELSON: They are doing something along 8
4j thes linas, aren't they?
5l MR. WEIDENHAMER: I'm not sure they are doing it.
?
6 j Thay may hav9 done it some time ago. I'm not sure that i
7- they are doing any real valve research right now.
8 MR. MICHELSON: The Germans, are they doing any 9 that you are aware of? Apparently not. Japan is the only 10 one active at the moment?
11 MR. WEIDENHAMER: Japan has done some work over 12 the past three or four years. Some of it has been
() 13 concluded. We may have trouble getting it. You know how 14 they are, the foreigners. We are attempting to, 15 essentially, make a trade with them for some of our 16 equipment qualification information for the results of some 17 of their tests.
18 MR. MICHELSON: Karl Seyfrit may have a comment 19 on this.
-1 20 MR. SEYFRIT: The French have recently done a 21 rather extensive study on some main stream isolation valves.
22 That report has been issued. I think-we have a copy of 23 that, don't we, Earl? We have a copy of.that.
24 MR. WEIDENHAMER: -Could you send me a copy of
, 25 that? I'll~ call --
,V -
",\.
,- ACE-FEDERAL REPORTERS, INC.
202-347 3700 ' Nationwide Coverage 800 336-6646
24594.0 223 BRT
( J-N/
1 MR. MICHELSON:. That's the only thing you are
.2 aware of?
3 MR. SEYFRIT: That's the only thing I'm aware of 4 that we have any particular study. You have to recognize 5 that-foreign operating information, data, is very difficult 6 to come by. We do have an exchange of information from the 7 NEA, and some small exchange from IAEA.
8 But this is in the nature of reporting specific .
9 events. They report what they want us to have and that's 10 all we get.
l 11 MR. MICHELSON: Yes.
12 MR. SEYFRIT: That data is collected and put in 1
() 13 a foreign events file in Oak Ridge. But that's really ,
14 about the only information we get routinely.
15 MR. MICHELSON: I see. .
. 16 MR. WEIDENHAMER:- Part of the Japanese agreement 17 on exchanging this qualification information is- they . supply 18- it in native tongue, so'we may-have to have-that
. +
19 information translated.
20 MR. MICHELSON: If it isn't translated by a
-21 fvalve expert it will come out very strange.
22 Yes, Pete?
Pete Wohld? -
? ~
23 MR.'WOHLD: It might be interesting to note that 24 Davis-Besse'is going to be having some deferential pressure' I 25 testing on approximately 20 valves.-
It's their program to
. ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336 6646
24594.0 224 BRT
,m r )
1 do that before they start up. They haven't provided the 2 details but they may be doing some closure testing as well 3 as opening testing under flow and differential pressure 4 conditions.
5 MR. MICHELSON: They will have the MOVATS 6 equipment attached during these tests? \
7 - MR. WOHLD: So they'll know the actual thrust 8 , conditions under the dynamic loads.
9 MR. MICHELSON: As well as under the static 10 condition. They have already dond those measurements.
+ 11 MR. WOHLD: They have done a few static tests i
12 too.
o
() '3 MR. MICHELSON: That would be most interesting 14 to see.
15 It's a good time to have a break now, isn't it?
16 You are coming up next, David. Let me just declare a break 17 until -- let's make it 25 after. -
?
- 18 (Recess.)
, 19 MR. MICHELSON: Can we get going here? _
We are 20 running just a little behind, now. We don't want to"got 4 21 too far behind.
22 The next speaker will be David Eissenberg from 23 Oak Ridge National Laboratory to tell us about their work 24 on the evaluation of the MOVATS system.
25 If you will start.<
4',.s l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage ,
800-336-6646
24594.0 225 BRT p
G 1 MR. EISSENBERG: I'm David Eissenberg, from Oak
'2 Ridge National Laboratory, and I'm the manager of the 3 nuclear plant aging research activities at Oak Ridge. I'm 4 . going to describe part of our activities which deal with 5 motor-operated valves.
6 The aging program at Oak Ridge has been studying 7 motor-operated valves for about two years, and the 8 objectives were the objectives of the age program, that is
-9 to demonstrate technology for maintaining operational 10 readiness of MOV, or other components, important to safety.
11 And the approach was to do a formal study to characterize 12 how and why motor-operated valves fail in nuclear plant
()~ 13 service, to identify and evaluate methods for detecting and 14 trending time-dependent degradation, i.e. aging, and other
- 15. abnormalities that can lead to valve failures; and to 16 develop the technical basis for recommending when and whdt 17 maintenance actions should be taken based on the 18 diagnostics.
19 Just for the record, because there is some 20 differeice of opinion on definition, J have noted our 21 version of definitions which is that failure mode, I think l 22 this is pretty universal,-is how it failed. Valve failed l
l 23 to-open, failed to close.
24 The failure cause, in the aging program or from 25 the aging program point of view, is why it failed, the p
n ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
I 24594.0 226 CRT
. b(\
1 approximate cause of failure, not the root cause, because 2- by and large trending is done on the basis of the 3 approximate cause, i.e., the bent stem or switch trip 4 during hammerblow, worm gears, et cetera. That's the 5 immediate cause that caused the failure. We do not look, 6 in general, at the root cause.
7 We can identify three types of abnormalities 8 which lead to failure, and I think this is important in our 9 reference to the MOVATS study. That is, there is 10 time-dependent degradations due to the cumulative effect of 11 stressors, and an example is a bent stem. It doesn't in 12 general bend at one time, it has been overloaded for a long
() 13 time.
14 Then there are abnormalities which lead to the 15 degradation, that is the -- now we are talking about the 16 high torque switch setting, then causes the bent stem. So 17 this is one step back from the aging effect, but it causes 18 the aging effect and this abnormality, then we are 19 interes'ted in knowing that it exists. We would like to 20 trend this abnormality but this is one that we might be
. 21 able to find also. And finally there's a third type which 22 is an' interesting variety, and these are abnormalities 23 which lead to failure only under some anticipated operating 24 conditions.
1 l 25 I think it's important that the incorrect bypass l
O) v ACE FEDERAL REPORTERS, INC. -
202-347 3700 Nationwide Coversee 800 33Md46
[
l l
l 24594.0 227 BRT
( t U
1 switch settings which have been popular recently are really 2 r.ot aging, and they are not an abnormality which leads.to
.3 aging. It is simply an on/off switch. If the conditions 4 are such that the torque switch trips, you have a failure 5 - because of this abnormality. If the torque switch doesn't 6 trip you don't have a failure and you can go on for the 7 40-year-life of the plant.
8 So this third abnormality is clearly not an 9 aging effect. .There's no time dependence associated with 10 it. However, it could be picked up by the same types of 11 monitoring that we would use in monitoring this type of 12 aging effect.
() 13 MR. EBERSOLE:
~
Is that last one equivalent to 14 saying: Well, it never would have worked in the first 15 place? It was maldesigned or inadequately specified or 16 didn't even start with the right valve?
17 MR. EISSENBERG: On you adjusted it wrong but 18 you don't know it until you get to t'he operating condition ,
19 in which it shows.
20 MR. EBERSOLE: Even that is preferative, a valve 21 which was not adjusted rather than a valve maldesigned in l
22 the first place because it didn't meet the conditions it 23 was expected to. All of this, at one time it was good --
l 24 I'm saying at no time, perhaps, was it ever good.
l
( g- 25 MR. EISSENBERG: At no. time was it good for that b
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide coverage 800-336 4646
24594.0 228 BRT i
1 operating condition. It's a switch; either it's good or 2- not good based on the operating condition.
3 This review today will cov r three activities 4 and I'm going to go very briefly through this first 5 activity, which was the preparation of an aging assessment.
i 6 (Slide.)
7 Phase I of motor-operated valves. I'll tell you 8 what's in it but I'll go through it fast. It has been 9 published and is available.
10 I will spend much more time on the in situ 11 testing of motor-operated valves using MOVATS, which will 12 be the subject of another NUREG, which is going to be
() 13 published in the near future.
14 And finally I'll tell you something of the 15 testing of MOVs to evaluate diagnostic methods, which is 16 currently going on at Oak Ridge.
17 (Slide.)
~
18 . The aging assessment Phase I was published 19 recently, and there's its formal title. It contains this 20 kind of information. It reviews the MOV design and 21 subassemblies; it reviews.the mate, rials of construction, 22 the stressors expected in normal and accident conditions; 23 reviews the technical surveillance requirements; it then 24 looks at the failure experience from the LER data base, l ,
25 NPRDS and other publications and reports; it reviews with k.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646 ,
24594.0 229 BRT
+
<v U 1 the manufacturers; and we have visited Limitorque and we 2 visited three valve manufacturers, and discussed their 3 maintenance practices and also discussed with them the 4 types of aging-related failures that they are familiar with.
5 As you'll see later, those failures haven't
-6 really occurred in nuclear plant service but they occurred 7 in other applications because plants haven't been operating 8 that long.
9 And then we developed a list of measurable 10 parameters that could be used for trending aging type 11 degradation, aging effects.
12 The result of this,.this bottom line, then would
() 13 carry us into a Phase II where we would say what devices 14 could we install that could then trend these particular 15 parameters. And I'll give you examples of the parameters 16 and devices as we go on. But the bottom line of this is to 17 lead us into a Phase II, to see what -- what devices are 18 commercially available that could then monitor the health 19 of the valve.
20 (Slide.)
21 Some other results, particular results of the 22 phasing assessment are, in our motor-operated valve 23 experience we noted that maintenance frequencies of MOVs 24 are higher than for other actuated types. I think this 25 came out of earlier discussion, too. We used the same data OU l
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwidc Coverage 800-336 6646
24594.0 230
.BRT O' 1 base.
2 The predominant causes of failure were torque or 3 limit-switch'misadjustments or failures and motor failures.
4 No reported cases of operator failure due to 5 normal aging were noted. .That's what I mentioned earlier.
6 That may be too strong a statement, "no," but there's no 7 significant cases of operator failure that you could pin 8 down and say it just wore out.
9 Maintenance practices tended to be aimed at 10 restoring operability rather than identifying and 11 correcting the degradation. This also has been mentioned 12 earlier.
) 13 (Slide.)
14 Some other results of the Phase I assessment, 15 and this came out of discussion with Limitorque, that is..
16 that th.ese -- these were the subassemblies and.we. discussed 17 these with the manufacturer. They came up with this, kind
' 18 of rating or ranking of.importance of this kind of failure 19 -for the gearbox, motor, switch assembly, and limit switch. .
20 I don't know-if you have any comments or object' ions, but 21 those 'are the listings that we got. .
22 This gave us a feel for what we should be 23 monitoring for over the life of the valve.
24 These are the parameters that we identified in
. 25 our study that could be used for trending abnormalities,
(
l ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee 800-336 4646 i
w __ _ , . - _ - _ _ . _ _ . . . _ _ . . .
24594.0 231 CRT
~
1 and it's a pretty long shopping list. And I think we've 2 covered what everybody brought up, including putting on 3 microphones on the outside of an operator.
4 These are a possible list of continuous 5 monitoring devices that could be installed. And, in 6 addition, we did not neglect the use of periodic 7 disassembly and inspections and measurements taken during 8 that period, and trending of those measurements.
9 These would be done under longer time intervals 10 but this could be as valuable as this type for certain 11 kinds of degradation.
12 (Slide.)
( 13 Now I'd like to discuss the task involving the 14 testing of valves using the MOVATS method.
15 We have two objectives. The first objective was 16 the one that Jed described earlier, which was the request 17 from NRR to evaluate the capability of.MOVATS to provide 18 useful diagnostic information for determining operational 19 readiness of MOVs, beyond that currently provi'ded by ASME 20 section 11.
21 Our second objective was the aging program 22 objective, which was to identify and characterize the types 23 of abnormalities to be found in MOVs which were at that 24 time operational. That is, they had passed their section
_ 25 11 requirements, the utility considered them operational.
(' -
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 6646
24594.0 232 BRT 1 (Slide.)
2 Some of the pages on the handout that I have are 3 out of place and you could probably find them. They should 4 all be in there but they got slipped around.
5 The easy part was the question of the ASME 6 section 11, and it was mentioned earlier. I heartily agree.
7 The section 11 test includes these four types of tests:
8 Valve exercising to observe disk movement, exercising full 9 stroke to measure stroke time, then for some valves leak 10 rate determination, and position indicator verification.
11 The MOVATS type test would apply to basically 12 these two. (Indicating.) This test, if anything, is
() 13 ,
better than this test, because, as was pointed out, a motor 14 cannot slow down enough to be able to fail this test.
. 15 There's a 25 percent decrease in the full stroke time 16 requirement and their motor will do it.
17 In fact, in our diagnostics, and this will 4
18 support'it, we measured with an accelerometer the
-19 characteristic vibration f requency . of the motor and we took c
. .. ., j 20 pieces of it during.the startup of the motor, and then 21 during the period when the motor was operating" but the ;
22 valve was seated and it was torquing up. And we noticed [
23 slowdown. The slowdown was about 2 or 3 percent, right at ,
24 the last segment, the last fraction of a second of sampling 25 that we took to get a frequency analysis, the dominant 7_
(_)
l l
ACE FEDERAL REPORTERS, INC.
- 202 347 3700 Nationwide Coverage 800 336 4 646
i 24594.0 233 BRT 1 frequency had just changed just a few points, so there is 2 no significant slowdown of the stroke time. That wouldn't 3 even show up in the stroke time because it occurs right at
-4 the end stroke.
5 So the first question is answered: Anything is 6- better than the section 11 method. And MOVATS clearly, and 7 -I'll point that out later, is better than section 11. But 8 that's not difficult to achieve, in terms of telling us 9 something about operational readiness.
10 (Slide.)
11 In that context, in order to answer the question 12 of what does MOVATS provide about operational readiness, I
() 13 needed a definition of operational readiness. And so I 14 invented a pair of definitions which is mine but I hope is 15 a fairly universal. set. That is, I made the distinction
~
16 between an operable MOV and an operationally ready MOV, in 17 the sense that an operable piece of equipment has been 18 demonstrated by test to have met certain performance e
19 requirements as indicated in the test under the conditions 20 of the test. That is, it operated now, or did operate just 21 before.
22 Operational readiness I would define as the
, 23 ability in its present condition to meet the performance i
24 requirements -in the future under all anticipated operating
< 25 conditions. The two key words are "as it degrades in the
, I)
(_/
l ACE FEDERAL REPORTERS, INC. .
l
( 202-347 3700 Nationwide Coverage 2 336 6646
a am _ w w ,a- v: 4+
I l
24594.0- 234 '
BRT
(~T V
1 future" from degradation, from where; and, "under all --
2 under conditions that it may see during an accident."
3 And, so, my conclusion is that operational 4 readiness can be inferred, or only inferred from 5 operability tests. If it has been operating, if it has 6 j been passing tests, then the chances are it will pass the 7 next test, if the conditions aren't very different and if 8 not much time has passed. But the key to measuring 9 operational readictss, it seems to me, should be more from 10 monitoring and trending the abnormalities that lead to 11 failure.
12 You really have to know how things are going to
() 13 change in the future before you can predict operational 14 readiness. You can infer it but in order to pin itdown 15 l you need to do trending.
16 So my first conclusion, again, pertaining to the 17 section 11 test, is that it measure operability. It does 18 really not measure operatibnal re,adiness.
19 (Slide.)
20 We then gave MOVATS an incorporated subcontract 21 to provide test data and analysis according. to this brief 22 description of a statement of work.
23 They were to do the tests themselves. We were 24 not involved. They would obtain the cooperation of one or l 25 more utilities.
()
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
24594.0 235
-BRT O"'
1 They would develop and submit a test procedure.
2 They would use_the test procedure to obtain valve 3 signatures for.a total of about 40 MOVs which were 4 considered op,erationally ready and also to be, insofar as
- 5 possible, part of a safety system.
6- And they would analyze the data to obtain a 7 diagnosis of the defects found and identify the remedial 8 actions to be recommended. They ultimately obtained test 9 results for a total of 36 valves at four nuclear plants.
10 MR. MICHELSON:. Were any of those boiling water 11 reactors?
12 MR. EISSENBERG: Well, we are not identifying
() 13 any of the nuclear plants.
14 ,MR._MICHELSON: That's not much of an 15 identification. I want to know because I'd like to get a 16 feel for whether that type of valve operation was included
, 17 in the observations.
18 MR. CHARBONNEAU: It was Brunswick accounting, 19 TMI --
. 20, MR. MICHELSON: TMI-l? It would have to be.
21 MR. CHARBONNEAU: Sequoia, I be,lieve it was --
22 I'm sorry, it was. It was Surrey.
23 MR. MICHELSON: I looked at the 36 valves 24 examined and only one of them had BWR written beside it.
25 MR. CHARBONNEAU: That was kind of embarrassing n
U ACE. FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-3364646
l 24594.0 236 BRT x .
-l because it was a coordination difficulty at the Brunswick 2 station. I don't want to tell you --
3 MR. MICHELSON: Does that mean there was only 4 one boiling water reactor valve in that sample; is that 5 correct?. Sample of 36?
6 MR. CHARBONNEAU: That's right.
7 MR. MICHELSON: So I really can't draw any 8 conclusions about boiling water applications from it.
9 MR. CHARBONNEAU: I wouldn't anyhow, frcm our 10 experience, draw any difference.
- 11 MR. MICHELSON
- The rest was pressurized water.
12 MR. EISSENBERG: Brian Curry explained how the
() 13 MOVATS system operates with his very good viewgraph. This 14 is a much more complicated picture. It also-shows a t 15 different arrangement for the linking in of the torque
- 16 switch. This shows a rack and pinion. Here is the spring -
17 pack and the spring pack, again, can slide it back and a
1 18 forth on its spline. -
19 As it slides back and forth, thip rack moves 20 back and forth, and rotates this gear which then rotates 21 the torque switch in either direction, and when the torque 22 switch rotation is sufficiently great,.it'will open the 23 ~ switch and turn of f the motor.
24 As was pointed out, the limit switch.is a g- 25 counter. It just simply counts revolutions. It looks l'ike i
(/
ACE. FEDERAL REPORTERS, INC.-
i 202 347 3700. Nationwide Coverase 23M i .
24594.0 237 BRT 1 another rack and pinion that it has, but this is actually a 2 screw, a screw and a gear and the -- as the motor turns, it 3 turns the screw which then winds up this limit switch and 4 eventually this switch is set for so many, basically for so 5 many turns. Just like an odometer.
6 If there's any questions about the operation of 7 the system, I'd go into it in more detail but I don't know 8 how many of you are familiar with it.
9 Okay.
10 (Slide.)
11 The MOVATS device, also as Brian and others 12 already noted, measures these three parameters: The
()
13 position of the worm, that is the axial position of the 14 worm is directly measured by MOVATS. It measures the 15 switch actuation times during ,the stroke; and the motor 16 current signature during stroke. .
17 (Slide.) ".
18 ,
If there's any amplification of these I could do that, also. #^
, 19 20 This is the MOVATS device. I guess it's from a 21 patent application, with all those numbers on it.
22 FROM THE FLOOR: If it's an application you are 23 not supposed to be showing it, but we received the patent a 24 week ago.
25 MR. EISSENBERG: It didn't have any stamps on it.
/
1 ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336 6646
24594.0 238 BRT
~
1 This is the device. Basically the worm shaft protrudes out 2 of the motor-operated valve and this device, this little 3 piston, is butted up against it, spring loaded, in such a 4 way that when it operates the -- this little LVT system 5 moves and'then senses the very small motion of the -- axial 6 motion of the worn.
7 MR. EBERSOLE: It's a load cell?
8 MR. EISSENBERG: It's not. It 's moving this .
9 It's loaded up on --
10 MR. EBERSOLE: It's an armature then.
11 MR. EISSENBERG: It's following and it senses 12 where it is at all times.
c.
,) 13 MR. EBERSOLE: Yes.
l'4 MR. EISSENBERG: We've seen this picture also, a 15 slight different version of it.
16 (Slide.)
17 This is a close to open cycle. Is this -- this 18 is an idealized version of a close to open cycle and in it.
, 19 there's two hammerblows. .In each case the.hammerblow was 20 of sufficient thrust or force that it' fed back ipto the 21 torque switch and tripped the torque switch, and in each 22 case the torque switch trip is noted but because the limit 23 switen, the bypass of the limit switch was still in place, 24 it didn't shut off the valve actuation. But in each case 25 it did trip it.
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336 6646
24594.0_ 239 BRT (O'h 1 That may not be the normal situation but it 2 shows the significance of knowing where the -- where these 3 hammerblows occur with reference to this point where the 4 bypass switch is actuated.
5 The other thing it did, it did show -- this is a 6 valve in the opening direction and it shows that the valve 7 backseated because in the open direction, if it had been a 8 limit controlled valve, which they are in general, and the 9 limit switch was set right, it probably would have just 10 ground down to a halt and you would have seen no increase 11 in thrust at the end of stroke.
12 Now I've got a couple of examples from a MOVATS
() 13 plot.
14 (Slide.)
15 This is the kind of information MOVATS produces.
16 On it it does show the two, this particular one shows two 17 hammerblows on the thrust signature. It shows two This is the ' running load during the valve 18 hammerblows.
19 actuation. This is the off position, this'is the running And there's no particular features that we see on "*
, 20 load.
21 this running load that we could use for diagnostics except 22 the two hammerblows.
23 I would like to point out that this is a kind of 24 a summary plot and that by means of the electronics, MOVATS
,_s 25 can amplify all of the -- all of this signal, amplify any
() .
ACE FEDERAL REPORTERS, INC.
2)2 347 3700 Nationwide Coversee 800 336 4646 t_
24594.0 240 RT 1 portion of it to any magnitude and get a fairly good 2 representation of what is there .
3 So that this is not intended to tell you the 4 precision with which the signal is acquired. But it does 5 show the two hammerblows and it shows it is protected.
6 I think in this case the valve initially was not 7 set right. This is an after, from before and after. If 8 you notice, the af ter gives a bypass time of 3 seconds out 9 of a total stroke time of 27 seconds. So it's more than 10 10 percent of the stroke, something like 12 percent or 11 whatever, of this stroke is now bypassed. This is an 12 after -- this is after it was found that it was much 7
13 shorter, and that by agreement with the utility it was 14 changed. ,
15 MR. MICHELSON: The squiggles on,the voltage 16 curve, what's the significance?
17 MR. EISSENBERG: These? -
18 MR. MICHELSON: No, the voltage curve.
19 MR. EISSENBERG: I think that':s an artifact of 20 the way at that time'he was taking.-- it's the switch 21 position rather than a voltage, in that sense.
.9 22 MR. MICHELSON: Could we get a clarification?
23 MR. CHARBONNEAU: We had talked earlier, current 24 sensing as we call it. This is done with voltage sensing 25 so the darkened areas are just, you can just classify it as ACE-FEDERAL REPORTERS, INC.
1 202 347-3700 Nationwide Coierage 800 136-6646
24594.0- 241 BRT 1 noise. We are not interested in that for a while.
2 MR. MICHELSON: It goes noiseless and then 3 starts to generate noise.
4 MR. CHARBONNEAU: On the far left, there's no 5 power. Then you start the cycle, energize the circuit, the 6 voltage goes up. Then you go along as Dave is showing you 7 and you reach your next drop. That is the bypass switch 8 dropping out. You continue on out --
9 MR. MICHELSON: Why no squiggles at that point?
10 MR. CHARBONNEAU: I yield.
11 MR. EISSENBERG: As far as we are concerned --
12 MR. MICHELSON: That is a real reproduction of
'l 13 what you saw on the scope? Is that fair?
- 14 MR. CHARBONNEAU: That's a real reproduction of 15 what you see when you transfer it to an XY plotter. You l 16 don't get all the details.
17 MR. MICHELSON: But it's the same thing I',d see i 18 on a scope roughly. That strange look. Okay. ,
19 MR. EIS'SENBERG: I would like to note the MOVATS 20 signature obtained for motor current and switch position ..
. 21 for the same valve, and you can see the same two squiggles.
22 So one statement that could be made, which 23 Limitorque made this morning was that one could, in 24 ' principle if not in fact, set the switches on the basis of i
3 75 motor current and switch position and that, in a thrust v l ACE-FEDERAL REPORTERS, INC.
202 347 3700 Natlor; wide Coverage 800 336 4646
24594.0 242 BRT
\ql 1 measurement -- a thrust measurement is not a necessary 2 ingredient in principle.
3 (Slide.)
4 Now,,what is it in fact? I guess what I've done 5 is looked over the 36 valves and found that 20 of them I 6 could clearly identify a motor current blip. Another six 7 or so I would not want to bet a lot of money that I could 8 identify it or not. And the rest I could not.
9 However, these were not amplified. If I had my 10 hands on the MOVATS amplifier I might or might not be able 11 to pick up motor cu'c rent. And I'll get to that in some of 12 the Oak Ridge work in which we are monitoring motor current.
() 13 But it looks like to me that most of the valves 14 could be set on the basis of motor current and switch 15 position.
16' (Slide.)
17 Switch position, incidentallyr could be read in 18 the control room for the so-called two-rotor valv,es, and I 19 guess I'm asking Brian Curry, and the fellow from VEPCO, do 20 you have three-rotor system switches? .
21 MR. CURRY: We have virtually everything but 22 four rotors. ,
23 MR. EISSENBERG What do you use the other two 24 rotors for?
25 MR. CURRY: Spares.
ACE. FEDERAL REPORTERS, INC.
202-347 3700 NaionwideCoversee 800 3M 6646
i 24594.0 243 BRT ,
(~
V 1 MR. EISSENBERG: Your light is on the same rotor 2 as the switch?
3 MR. CURRY: That is correct.
4 MR. EISSENBERG: The light is on the same rotor 5 as the bypass-switch. That means we could tell exactly 6 where this occurs in the control room by noticing when the 7 light goes, I guess it's off, or on. This is the open 8 light.
9 When the open light goes of or the open light 10 goes on we know exactly where in the stroke we are. Then 11 if we could monitor motor current and observe this blip, 12 then we can set the switches, the bypass switch and know
() 13 that it's set correctly, rather simply.
. 14 MR. MICHELSON: Since you named the plants you 15 took your sata from and I think you named Sequoia as one of 16 them; was that,right?
17 MR. EISSENBERG Well, we decided not to know 18 any of -- where the plants were.
19 MR. CHARBONNEAU: That's right. You are not .
20 supposed to know. .
21 MR. MICHELSON: Because if you name Sequoia then 22 I have a real problem on some of your results which 23 wouldn't match anything on Sequoia.
- 24 MR. CHARBONNEAU: Sir, Sequoia was definitely 25 one of them.
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee 800 336 4 646
24594.0 244 BRT 1 MR. MICHELSON: Well, Sequoia definitely isn't 2 wired up that way. They run the bypass throughout the 3 travel. They don't come in with bypass on opening.
4 MR. JONES: On opening.
5 MR. MICHELSON: Yet the results I looked at in 6 your draft report, every one of them seemed to indicate 7 that there -- they were always looking for -- did you ever 8 on the results you looked at, did you see any that had 9 bypassed throughout travel?
10 MR. EISSENBERG: We saw only what we got from 11 MOVATS. And the other thing is we saw mainly after, not 12 before settings.
[) 13 MR. MICHELSON: Maybe MOVATS needs to tell me.
14 Are they aware that's the way TVA wired their valves?
15 MR. CHARBONNEAU: If we had seen the closed ohm 16 switch set for 90 percent of the valve cycle, we would have 17 looked at it with a wrinkled brow. It's not acceptable but 18 the more unusual. I assure you the valve tested at Sequioa, 19 the close/open switch, was not set for 90 percent. It was 20 not set for 90 percent. -
21 MR. MICHELSON: That's odd.
22 MR. EISSENBERG: I can say I have seen the TVA 23 maintenance manuals and I don't recall that in the l
24 maintenance manual.
25 MR. CHARBONNEAU: When you go to a four-rotor ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
24594.0 245 BRT 1 system it is not unusual for them to use the number four 2 rotor, which is the number 13 switch for the close to open 3 on the bypass for that torque switch to get away from 4 messing up your lights.
5 MR. MICHELSON: I realize that. The point is, 6 though, as a matter of policy on opening the valve, they do 7 not use the torque switch for any of the travel.
8 MR. CHARBONNEAU: Oh. Now that's --
9 MR. MICHELSON: All the results I looked at, not 10 knowing what plants they were, looked like every one of the 11 results had watching the bypass kick the torque switch in 12 and out. I'm confused, then.
() 13 MR. CHARBONNEAU: We said two different things.
14 One was whether the bypass was closed for most of the cycle.
15 That bypass from those signatures shows clearly it is not 16 closed for most of the cycle. The voltage' sense also told 17 you the ohm torque switch did not open.
18 I'd have to go back and look at the reports 19 because it should say on the MOVATS reports whether the 20 open torque switch was even in the circuit or not. They 21 may have listed the lead on.the open torque switch, and yet 22 because they are still using that motor for light 23 indication, and we clip onto 5, we are showing a close to 24 open bypass switch opening up, which is not doing anything.
25 It's just wired in there. That's what they are seeing.
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336MA6
24594.0 246 BRT 1 MR. JONES: You may have been monitoring the 2 wrong limit switch. There's two switches. There's two 3 limit switches in there, one of which is bypassed. The 4 other is actually the 90 percent limit switch, taken all 5 the way open. So you may have been monitoring the one --
6 MR. CHARBONNEAU: I'd be happy to check but I 7 know the folks that did that testing. Let me also say on 8 Sequoia, that was one where the electricians, I'm sorry to 9 say, followed the procedures exactly and every one of their 10 switches are set correctly by their procedure, 11 MR. MICHELSON: Which switches are you referring 12 to?
[ _
13 MR. CHARBONNEAU: Close to open bypass. I know 14 the torque switch is in there because we had to meet --
15 MR. MICHELSON: The point is that the open 16 bypass isn't in there. It's bypassed all the time. It 17 does not use torque for protection on opening. That's what I
18 confused me.
19 MR. CHARBONNEAU: The valves we tested were.
20 MR. EISSENBERG: We'll have to get some data 21 exchange.
22 MR. CHARBONNEAU: We'll be happy to recheck it.
23 MR. MICHELSON: Yes, recheck it.
24 MR. CURRY: One comment is the way Bart hooks up 25 his equipment, he can tell you where the limit switch is l
l ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 3364M6
T 24594.0 247 BRT
('T LJ l bypassing the torque switch if it's wired in there. You 2 can still have the torque switch itself bypassed using a 3 jumper, or it could be wired completely out of the circuit.
4 I know two different plants that do that at least on some 5 valves. -
6 MR. MICHELSON: That's a possibility.
4 7 MR. CURRY: It looks like it is being bypassed 8 but in reality it's either not in the circuit or it's 9 jumped out.
10 MR. MICHELSON: That's probably what happens in 11 which case it shows you've got to read the wiring diagram 12 before you connect up the equipment then.
() 13 MR. CURRY: It comes characteristically with the 14 voltage switch sensing rather than the current switch 15 sensing. The current switch sensing involves a lot more 16 OC/0A problems with lifting leads and breaking them, where 17 with the voltage switch sensing you can Vo in and not alter 18 the circuit. configuration. -
19 MR. MICHELSpN: It just shows that you can be 20 misled by the results if you don't read the wiring diagram 21- in the process. ,
22 MR. EISSENBERG: The results of the 36 tests are 23 summarized statistically on this chart.
24 (Slide.)
- 25 It is very clear to see that a very large -
i x-) .
(
ACE. FEDERAL REPORTERS, INC.
202 347 3700 ' Nationwkle Coverrgs 2 3364646
~24594.0 248 BRT l'h V
1 fraction of the 36 valves that we tested had improperly set 2 bypass switches. By and large that means they were set not 3 to cover the torque switch trip due to the second 4 hammerblow. There were also a very large percentage of 5 incorrect torque switch calibrations, either too high or 6 too low. And the rest of these degradations were noted.
7 (Slide.)
8 MR. MICHELSON: I would suggest, of course, you 9 find out the answer to the question we raised because it 10 affects the results significantly. If the torque switch is 11 continuously bypassed people don't bother where it is set 12 or how it is set relative to the switch operation, which is
() 13 also not being used, by the way.
The contacts are there 14 ~but they are not using it.
15 MR. EISSENBERG: That's a good point.
16 MR. MICHELSON: It could affect your answers.
17 Pete, did you have a comment?
18 MR. WOHLD: On the previous slide where you had 19 .$he blips, I wonder if,you had any diffe,renti'la pressure 20 test. The one thing that happened at Davis-Besse, under 21 differential pressure, that pulse, the spike becomes a .
22 pulse, depending on the differential pressure at different ,
23 heights, and the bypass time. It was started with 5 ,
24 percent and then they thought maybe 10 percent was okay.
25 By the time they add the lost motion of the ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coveraos 800 336 66d6
o 24594.0 248 RT 1 fraction of the 36 valves that we tested had improperly set 2 bypass switches. By and large that means they were set not 3 to cover the torque switch trip due to the sccond 4 hammerblow. There were also a very large percentage of 5 incorrect torque switch calibrations, either too high or 6 too low. And the rest of these degradations were noted.
7 (Slide.)
8 MR. MICHELSON: I would suggest, of course, you 9 find out the answer to the question we raised because it 10 affects the results significantly. If the torque switch is 11 continuously bypassed people don't bother where it is set 12 or how it is set relative to the switch operation, which is k' 13 also not being used, by the way. The contacts are there 14 but they are not using it.
15 MR. EISSENBERG: That's a good point.
16 MR. MICHELSON: It could affect your answers.
17 Pete, did you have a comment?
18 MR. WOHLD: On the previous slide where you had 19 the blips, I wonder if you had any differential pressure 20 test. The one thing that happened at Davis-Besse, under 21 differential pressure, that pulse, the spike bec'omes a 22 pulse, depending on the different'ial pressure at different 23 heights, and the bypass time. It was started with 5 24 percent and then they thought maybe 10 percent was okay.
r~w 25 By the time they add the lost motion of the t ,
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
[
24594.0 249
'BRT- '
.(~N.
u,e .
'l motor and then a 9 percent pulse, they ended up having to 2 have at least 14 and then out to 20 percent to have margin.
-3 I was wondering if that --
l 4 MR. EISSENBERG: You are right. One of the 5 limitations of MOVATS that I indicate is that it does not 6 test at this time on anything but ambient conditions. I
$ 7 think it has the capabilities but that's -- well, as I got 8 the results that was true. You didn't tell me of any tests 9 that were run at other than ambient.
'10 MR. CHARBONNEAU: You are absolutely right,
'll these were all run at no pressure condition. But that 75 12 percent,-Carl, _in light of your comment, if we took out all 13 of this one plant, Sequoia, all of theirs out and we had 14 made.this point that you ought to take them all out because 15- they were all being set by an incorrect procedure.
16 MR. MICHELSON: They are jumped permanently. It 17 makes no difference. -
18 MR. CHARBONNEAU: Whether the torque switch is 19 jumped or not we can get that clarified. The bypass a
^
)
20 switches for the l'adingo conditions weren't right. I don' t 21 believe it's. fair to quote-75 percent improperly set bypass-
{ 22 owitches.. It's more'like half that. r 23 MR. EISSENBERG: You mean the other half was --
, 24 MR. CHARBONNEAU: The other plants that either p g 25 had procedures that said you should set it here and it
)
l u
[
- J te 4 ACE-FEDERAL REPORTERS, INC.
UT N" , 202-347 3700 Nationwide Coverage 800 336 6646 i
m _ __ _ , . _ _ . - _
l 1
1 24594.0 250 I RT 1 wasn't set here and it was dropping out too early.
2 It's a subjective comment. You can do whatever 3 you like with it, because our industry study of over 350 4 valves, similar tests, show that particular degradation at 5 36 percent.
6 MR. EISSENBERG: But I guess I missed the point.
7 Why is the number higher because of the TVA? Because they 8 were all wrong?
9 MR. CHARBONNEAU: So many of the 36 valves that 10 they had that degradation --
11 MR. EISSENBERG: It's a procedural problem. I 12 see your point. The statistics were correct but the sample p
13 was a poor sample.
14 (Slide.) s 15 We then classified the abnorrcialities according 16 to these three schemes: time dependent degradations, those 17 which were incorrect adjustments, and of those two, which 18 were those which would lead to a time dependent degradation 19 and which was simply the switch that would cause failure 20 under some anticipated operating conditions.
21 And this is the way we ended up, that there were 22 a number of degradations that were picked up by MOVATS.
23 There were a number of other incorrect adjustments. There 24 are other incorrect adjustments and some of those could be 25
^
classified as the type that could cause an instant failure
(' /')
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80 4 336-6646
y I
24594.0!'5 '
251 n( O
, U,BRT #
1 if the operating conditions were sufficiently severe. And 2 I think the point is well taken that the, Davis-Besso i
3 incident demonstrated something which I guess we know, and 4 that is that when there's pressure on the valve then the 5 forces required to lif t and close the valve differ from
']e t 6* each other. And when there's a delta.p across the valve, 7 the force required to move the obturator, that is the 8 beginning cf the stroke, that the hammer stroke is greater s- 9 than when there's no delta p across the stroke. So there's t
10 a need to get more realistic operating data under operating 11' conditions than what we've got.
12 (Slide.) '
O 13 Conclusions about MOVATS' capabilities. MOVATS 14 can provide information regarding operational readiness
- 15. ,.beyond that obtained by section 11 tests, by identifying s i -
16' , and quantifying incorrect adjustments which can cause
, ' \
17 Ldailure. That is, these are the nonaging related, but that 18 MOVATS is very good at correcting or detecting this kind of N ' i-19 incorrect adjustment. And degradations which can lead to
, [20 failure if allowed to continue, i.e., the bent stem.
21 I guess my statement'there, without detracting A 's 22 from MOVATS, is'that it's very easy to be better than
-23 section 11 tests.
c:
24 (Slide.)
i i g-)_ 25 There are these limitations that we would
'J
/>
ACE-FEDERAL REPORTERS,'INC.
202 347 3700 Nationwide Coverage 800 336 4646
-24594.0 252 !
BRT 1 identify with regart to MOVATS. Some of the degradations 2 cannot'be detected by MOVATS signatures. Signatures at 3 ambient conditions may not provide information about 4 performance under operating conditions, and I think the 5 case of Davis-Besse and the torque -- the hammerblow 6 magnitude is a case in point.
7 Periodic testing at shutdewns may not be 8 sufficiently frequent. Degradation can occur between
- 9 shutdowns. That is, we like -- it would be nicer to have a 10 continuous monitoring.
11 And that manual installation and removal can 12 lead to excess radiation exposure. At this time, or as far
() 13 as I know, MOVATS'was installed one valve at a time, the 14 test done, and the installation removed.
15 So that these two are implying that a device 16 permanently installed on the valve, a MOVATS type or other 17 device permanently installed on the valve could eliminate 18 these problems. And also, could also allow for operation 19 at conditions -- operating conditions.
20 (Slide.)
21
- MR. MICHELSON: For a clarification on that 22 slide, you talk about the signature at ambient conditions.
23 MR. EISSENBERG: Shutdown.
24 MR. MICHELSON: As I understand it there's 25 nothing'to prevent this from being done at operating ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 2 3364646
. _ _ - , _ . _ . . ~
24594.0 253 RT 1 conditions; is that right?
2 MR. EISSENBERG: I just made the point, the way 3 that MOVATS was used for our tests was not that. But 4 there's nothing conceptually wrong.
5 MR. MICHELSON: The utility didn't allow you to 6 go in under operating conditions and measure? Just under 7 shutdown conditions; is that the case?
8 MR. CHARBONNEAU: That's correct. It's very 9 difficult to ask a plant to set up the actual dba --
10- condition. The 36 valves we tested we've probably only had 11 the opportunity of testing 10 or 12 of them at full delta p, 12 of which four were at Davis-Besse.
13 So we can.see the signature, you can set the 14 valve up in the ambient condition and we'd know what to
-15 expect.
16 MR. MICHELSON: A good valve, someday, to choose 17 to get these kinds of information would be the reactor 18 water cleanup on a boiling water reactor which runs at full 19 temperature and pressure and full flow routinely, and which 20 also can be shut off at any time without interfering with 21 reactor operation and can then be reopened again.
22 This gives you an opportunity to operate a valve 23 under hot' flowing conditions to find out how it might l
24 differ from cold conditions that you could go back and l
25 check again later on the same valve.
l ACE-FEDERAL REPORTERS, INC. -
202-347-3700 Nationwide Coverage 800 336 4646
I 24594.0 254 f_BRT 1
Q/
1 It's a real good one to choose. There aren't 2 too many in a nuclear plant where you can do that but 3 that's one of them.
4 MR. CHARBONNEAU: That's why the 16-valve test 5 that we have been talking about that Pete alluded to was 6 going to be quite important. We think we can predict 7 what's going to happen. We are quite confident. We 8 already did it on the four valves at Davis-Besse.
9 MR. MICHELSON: I'm quite anxious to see 10 somebody test a reactor water cleanup valve under both cold 11 and hot conditions and so forth, sooner or later. It's one 12 of those we are interested in from the viewpoint of b 13 isolation under-break conditions as well. This is'as close 14 as we can get.
15 MR. CHARBONNEAU: We'll make mention of that at 16 a BWR somewhere.
17 MR. MICHELSON: Talk to Glen about it.
18 One question, you use a term in this draf t 19 report that you used today also, I'm just not up with all 20 the words, what is an "obturator"?
21 MR. EISSENBERG: That's the term that refers to 22 the gate of a gate valve, the plug of a globe valve. The 23 thing that blocks the globe --
- 24. MR. MICHELSON: I will always talk about plugs
- x. 25 and disks. I'm old-fashioned. This is a new term --
l l
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 2 336-6646
I l
24594.0 255 BRT
(~ ~ \
1 MR. EISSENBERG: That's a professional term, I 2 guess.
3 MR. MICHELSON: Is this something everyone 4 understands but me? Does the code use it? Is it starting 5 to use it?
6 MR. EISSENBERG: They don't want to say gates 7 and globes and plugs --
8 MR. MICHELSON: So they talk about obturators; 9 is that pronounced right?
10 MR. EISSENBERG: Yes.
11 (Slide.)
12 I would like to flip over to the tests running 13 at Oak Ridge now with regard to the aging program, although 14 it overlaps a lot with what we were talking about with 15 operating plants and the MOVATS' test. We are conducting 16 . tests to identify a set of useful operating tests for 17 detecting abnormalities and trending degradation. We are 18 carrying out baseline tests at Oak Ridge. We will then 19 carry out, and we are starting this week, our first ambient 20 accelerated test using a mounted valve. Then we'll go on to 21 implanted defects in the valve and see how-we can trend
-22 those defects. And then hopefully we'll carry-out tests at 23 operating conditions and compare operating conditions and
. 24 ambient, which is what'you were referring to.
l r' 25 Right now we had earmarked a good place to do it l
l
(_) '
ACE-FEDERAL- REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646
~
l l
l 24594.0 256 ;
BRT !
1- as ETEC. That may not come through because of limitations l 2 in'their program, but we certainly would like to work with 3 utilities and try to get data showing, as Art is talking 4 about, operating conditions versus ambient conditions data.
5 Finally, we'll carry out field tests using some 6 k'ind of package of diagnostics, a MOVATS-type package.
7 And I would like to now go into some of that work that we 8 have been doing.
9- (Slide.)
10 These are the parameters that we are currently 11 monitoring: Strike time, motor running current, surge 12 current, switch actuacion' times, worm shaft translation, 13 stem thrust, stem velocity, stem position; motor, gearbox 14 and valve vibrations, using an accelerometer; and motor 15 temp 3rature. This includes the winding, stator winding and 16 the outside motor temperature.
17 The current we are -- we are currently using a 18 Limitorque operator 2, next to a motors. These.are the 19 nearest we could get to check out these techniques. We 20 also have two SMB operators connected to valves. We are 21 going to install one of those in the very near future but 22 the purpose of the the tests using these valves was to 23 check out whether these had the~ser.sitivity and selectivity 24 that we needed to look for defects.
25 (Slide.)
ON.
i ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide coverase 800 336-6646
24594.0 257 BRT O
\~)
1 I guess that's what I said here. The initial 2 tests focused on establishing capability of instrumentation 3 to measure the parameters. We need to understand the 4 normal signals. We need to determine reproducibility and 5 determine the response to known imposed changes. So we'll 6 first be taking data of a normal valve and then we will be 7 accelerating, aging them, to see if we can see trends and 8 then see if we can determine what they really mean.
9 The rest of the viewgraphs are samples of some 10 of our normal valve data and a few cases where there seemed 11 to be some abnormalities. This is data that was taken from 12 the sensors, put on a digitizer, and then these were the 13 results after they come off the computer. So this is a 14 nicely arranged set compared to some later sets which were 15 done right at the valve using on-line equipment. But this l
16 is the beginning.
17 This is a close to open stroke. This is the 18 beginning of the stroke. This is the first four seconds of 19 the stroke. And there are some very interesting features.
20 First you see the motor's current, starting 21 current peak, the surge , and then it-leveling off.
22- .The stem velocity does some peculiar things.
23 Initially you have some stem motion then"the stem motion 24- goes away and comes back.
l 25 It turns out this valve had been closed and.
l I
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 6646
24594.0 258
, B
,RT .
(-)
1 seated by means of torque switch and it had a load on it.
2 So, as soon as the motor starting turning the 3 first thing that happened was the stem moved up. Basically 4 the two -- I'm now addressing people that know what's 5 inside the valves -- the lugs on the worm gear came off the 6 seat and were able to move somewhat around. This is just 7 part of the half a revolution of the worm gear and it just 8 simply relaxed the obturator bearing on the seat.
9 Then it stopped until the lugs picked up on the 10 other side and then started the stem movement.
11 We have a strain gauge mounted. This is the 12 compression strain during the seating that it saw. This is O)
(- 13 a dropping to zero. And that's -- that coincides with this 14 stem velocity. And then zero strain, and then here is the 15- running load in the closed /open direction.
16 There is when the limit switch came on af ter 17 beginning of stroke. This is'the protection, limit switch 18 protection. An accelerometer reading, of which we have 19 done some frequency analysis on the accelerometer, we have 20 three of them. One is shown here.
21 This is the torque switch _ angle. What we've 22 done is installed an angle indicator directly on the torque 23 switch in a more or less permanent way so that we could l
l 24 track the_ rotation of the torque switch continuously.
25 We are using a torque switch that only has one ACE-FEDERAL REPORTERS, INC. -
202-347-3700 Nationwide Coverage 800-336-6646
l 24594.0 259 j
O~7RT 4
1 direction so that this doesn't mean anything in this j 2 particular case. It's at the other end that it does.
3 (Slide.)
4 This is the other end of the stroke when the 5 valve seats. Th~is is an open to close stroke, and this is 6 the tail end of the open to close stroke. I'm just giving 7 you some of the pieces of the data we got.
8 Here is something that's interesting and it ties 9 in with a remark that Brian made. Brian, you noted when 10 the valve seats the velocity goes to zero while the worm 11 advances, that is while you wind up the spring. Well, in 12 fact it doesn't go to zero.-
O- 13 MR. CURRY: It is closed.
14 MR. EISSENBERG: No, it goes only to about 50 15 percent of its travel and then levels off, and here is the 16 strain building up on the valve stem with a strain gauge, 17 and here is the torque switch angle following the strain 18 gauge.- And during that period of time when you have a 19 linear change you find that the stem velocity is going at 20 half its speed. You didn't know that.
21 We show limit switch actuations also, and 22 accelerations.
23 Here is the motor current. And you can see that 24 the motor current rises-as the. stem strain builds up. It 25 is not a very sharp rise, but it's reproducible and
(')N
\_
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646
24594.0 . 260
. ERT 1 ' detectable. So that we could use motor current to tell us ;
i 2 something about the condition of the valve as well as 3 torque switch angle and strain gauge.
4 (Slide.)
S We recognize there's some redundancy in these ;
1 i 6 readings, and we realize the idea is to come up with a 7 ' package at some point'that is nonredundant but will tell us e 8 enough about diagnostics.
9 Now, some miscellaneous mysteries that we have 10 comerup with to illustrate the kind of sensitivity we have. ,
11' This is a plot, a complete stroke of the 12 operator, 90 seconds. And this i;s valve stem velocity.
( 13 This.is highly amplified. You saw that it was some 14 fraction of an inch per second and this is down to the 15 hundredths of an inch per he could so it's highly amplified 16' and we get a very clear oscillation,_and that frequency is 17 precisely the stem nut rotation frequency. l 18- So we are seeing in the' stem velocity the nut 19 rotation ~ frequency, which shouldn't be there because it's a 20 continuous process. It suggests that there is some rough 21 spot on the stem or. stem nut which is causing the valve to 22' work itself-upyin jumps and we don't;know why but th'e
-23 sensitivity is there'to detect the stem nut rotation l . .
24~ ' frequency on the_ velocity-indicator.-
i i
25 It didn't" occur in the'open to close. direction
- O l
I
.. ACE-FEDERAL REPORTERS,' INC.
~
J 202-347 3100 . ; Nanonwidecoversee 800 33H646
l l
l I
i i
24594.0- 261 "RT v
1 anywhere near as much as it occurred in the close to open 2 direction and we don't know why there either. But again we 3 are at the point where we are determining sensitivities as 4 what these signals show us.
5 MR. MICHELSON: Was this a valve in the shop, in 6 the lab?
7 MR. EISSENBERG: Just a random valve we got our 8 hands on.
9 MR. MICHELSON: This was a real field installed 10 valve?
i 11 MR. EISSENBERG: It was used in the gas cooled 12 reactor program about 15 years ago.
13 MR. MICHELSON: This is one in your laboratory?
14 MR. EISSENBERG: Yes.
15 MR. MICHELSON: Thank you.
16 (Slide.)
17 MR. EISSENBERG: We are going to have a nuclear 18 qualified valve installed, probably in about.a month.
19 This is a case where we made a change and we 20 observed our change. on our sensors.
21' In this case we tightened the packing. We 22 relaxed the packing to zero and then using a torque wrench 23 we tightened the packing ambit temporarily to various 24 readings of the torque wrench. This is the-strain gauge j
1
{} 25 reading, but we'd-get a similar one if we used the strain.
ACE-FEDERAL REPORTERS, INC.
~ 8347 3700 NationwideM 8043364646
I 24594.0 262
, ,_BRT 1 Angle the stem loadings up in a realistic way, except we 2 see these oscillations and we see this kind of a step 3 change.
4 We are not sure of what these mean. We think 5 that the step change has to do with the valve design and 6 that is this is a globe valve and as the obturator, the
. 7 plug, goes by the opening of the valve and then goes by the 8 section of the valve internals where it's rubbing on all 9 sides versus where it's rubbing on only one side, the 10 loading changes. So we think this may be a valve 11 characteristic but again we are not sure. This is also 12 highly amplified.
13 This, I believe, is the stem note rotation 14 frequency that you see on this.
15 MR. EBERSOLE: Can you pull the packing down 16 hard enough to make it act as a break and synthesize the 17 worst low that the valve will ever see?
18 MR. EISSENBERG: We are currently doing that.
19 That's our accelerated aging test. We are going to set the 20 torque switch very-high and pack the packing very. tight and
- 21. see if we can't cause degradation in the valve. We have 22 just started that series of tests and we are . going to i
23 tighten it as tight as we can without knowing the pipe will 24 break.
l i
(-)
%)
25 (Slide.)
l l
- ACE-FEDERAL REPORTERS, INC.
202-347-3700- Nationwide Coverage 800 336-6646
l 24594.0 263 RT l
1 A couple of more. This is applied to motor 2 current, instantaneous motor current versus valve stem 3 -loading with strain gauge. This is attainable during an 4 entire cycle. The valve starts off where there is a stem 5 loading in the opposite direction. This is, again, because 6 the valve had been actuated in the other direction. In 7 this case it's open to close so the amount of strain is 8 very.small at the beginning of the stroke because it had 9 just opened. This is just running load. And then here is 10 where it goes through the row on the strain and the motor 11 current builds up as the motor starts.
12 Then the operation of the valve, most of the 90 13 seconds occurs right in here. As the valve opens the 14 running load varies, and you saw that in the last graph 15 where you get this oscillation. The oscillation appears as 16 this thicker line.
17 Then when the valve seats, the stem loading 18 increases sharply and the motor current increases. We can 19 draw this on an XY plotter over and over again, very 20 reproducible data on motor current versus stem load.
21 Although it doesn't change very much, we can 22 very accurately track it.
23 (Slide.)
24 We have done the same thing here with the torque
~
25 switch rotation device. Again we are just simply rotating, f-)3
\-
i' i
ACE-FEDERAL REPORTERS, INC.
l l 202-347-3700 - Nationwide Coverage _ 80 4 336-6646
24594.0 264 BRT 1 following the angle of the torque switch which is basically 2- reading the same thing that the MOVATS reads, but again 3 using a different device. And we get the same curve. They 4 should be linear. We have seen it in the past that they 5 are linear. And so we get the same kind of curve. We can 6 get motor current versus torque switch rotation and they 7 trend each other very closely. If we would tighten the 8 packing, this would simply move up. If we use the same 9 plot the running load would go up and we'd get this 10 variation in here and then it would take off.
11 (Slide.)
12 Finally, here's something that was a r
t_y
/ 13 serendipitous type operation. We had some calibration of 14 stem loading with a strain gauge and torque switch angle on 15 a MOVATS equivalent device. We had gotten enough data to 16 define this line.
17 When we went -- then we did other tests and 18 cycled the valve a number of times. When we came back 19 after a while we tried to get that same curve again and we 20 got it displaced and we reasoned that we had lost 21 lubrication, and so we relubricated and when we did it came 22' almost back to the same original line.
23 So our statement is that we may have seen 24 degradation of stem lubrication during the 160-odd cycles.
I 25- It turns out that that had been observed at the ACE FEDERAL REPORTERS, INC.
202-347-3700 . Nationwide Coverage '800 336 6646
24594.0 265 l
1 Oak Ridge gaseous diffusion operating plant operating 2 valves, that at 150 cycles they started to see the noise 3 and degradation that results from a loss of lubrication.
4 What they had done was build a continuous lubricating device.
5 They got away from the problem. I don't know whether that 6 would be needed but I believe this is realistic and it is a 7 large degradation.
8 This is not what Brian said. There's probably 9 some rationale we need to understand why some valves seemed 10 to degrade and some not.
11 MR. WARD: Yes. Go ahead?
- 12. MR. CURRY: One comment on that is that you are
('_# 13 at 65 cycles there; correct?
14 MR. EISSENBERG: Yes.
15 MR. CURRY: In the industry they normally assume 16 that the valve under worst case conditions would be 17 stroking once a week for its 40-year life. So at that 18 point you are at three years in effective life of the valve.
19 The chances of, in an operating plant, the valve is not 20 lubricated either once a year or once an operating cycle 21 are probably slim or none.
22- So while it's not lubricated for thre9 years you 23 are-going to see degradation but the chances of it 24 happening in the plant are probably very slim.
(} 25 MR. EISSENBERG: I think that's true with this I ACE-FEDERAL REPORTERS, INC.
~
202-347-3700 _ Nationwide Coverage 800.336-6646
24594.> 266 1 .one proviso, that they were ambient conditions in a room.
2 Under -- some valves may see more severe service under 3 which the lubricant might heat up or dirt get in there. I 4 think in general what you are saying is right. Your valves 5 are not really cycled as many times as even we cycled them 6 in this test. Again, these were reproducible data sets.
7 Well, that's all I-have in our current program.
, 8 It's currently in progress so I hope to have information 9 collected as we see how the valve -- the diagnostics 10 respond to degradation.
11 MR. MICHELSON: We'll be looking forward to 12 following your progress. Thank you very much, David.
7-13 We'll have to move on right away to Joel Page, who will 14 discuss the Davis-Besse event, valve-related problems.
15 MR. EISSENBERG: Here are some additional copies 16 of my report. Some of the pages in this are out of place.
17 MR. PAGE: My name is Joel Page, I'm with the 18 mechanical engineering branch of NRC, or NRR, anyway. I 1
19 think I'll be around for about five more weeks. Today I'd 20 like to discuss some Davis-Besse valve failures.
21 Primarily I would like to emphasize the MOV 22 failures which happened in the off-speed water system and i
23 in the main steam system. If we have time, I'd also like l
f 24 to discuss PORV failure, the main steam safety failure, the
~S 25 atmospheric vent failure and the turbine bypass valve.
(G ACE-FEDERAL REPORTERS,-INC.
202-347-3700 , Nationwide Coverage _ 800 336 4646
, .24594.0 267
-BRT O
1 I didn't really evaluate the latter of those 2 myself but we do have MEB members here who can handle the 3 discussion on that okay.
4 (Slide.)
5 With respect to the motor-operated valve 6 failures, the' valve functions in question were the AF-599 7 and AF-608, which are off feedwater isolation valves. To 8 the best of my estimation these valves have dual functions 4
9 to open and to close. They are wet-seating valves and the 10 close is from a steam feedwater rupture control system 11 signal. They must close when they get a signal that there 12 either is a break in the steam line or there is a break in ,
k/O -13~ the feedwater line.
14- They must open on operator demand, once you 15 determine what the situation is.
16 The MS-106 valve is the main steam valve which 17 is described as an isolation valve, but it supplies steam 18 to the auxiliary feedwater - turbine. It also receives SFR CS 19 system.
20 MR. MICHELSON: What are its safety directions?
-21. MR. PAGE:- It is to open.
22 MR. MICHELSON: Doesn't it have to close? What 23 happens if the steam line breaks?
24 MR. PAGE: I'm working strictly on something I l
25 got a. couple of days ago. I must state this is very k-,es]
i l
r l
ACE-FEDERAL REPORTERS, INC.
. 202-347-3700 Nationwide Coverage #533H646
l 24594.0 268 BRT k
1 preliminary. We are in the preliminary stages of 2 evaluating what's going on there.
3 MR. MICHELSON: Right now the records show they 4 were only worried about it opening.
5 MR. PAGE: That's correct. With respect to the
, 6 two previous valves they were described as passing, which 7 really was of concern to us.
P The utilities' investigation indicated that the 9 root causes of the failures had to do with the switches :
10 being set incorrectly. Now, these were both torque 11 switches and the torque switch bypass limit switches. They 12 were all indicated on all three valves to be incorrectly O
N/ 13 sat.
14 With respect to the 599 and 608 valve , they also 15 indicated that the lack of delta p testing was a cause 16 there, one of the root causes. They did no delta p testing 17 on these valves. However, for the MS--106 valve this was 18 not indicated as a problem. - And the reason this wasn't 19 indicated as a problem is that they had done delta p
~20 testing on this valve and it had passed successfully. So i
21 there is some unique divergence there. They indicate 22 possibly scale or something 21se or manual closure of the 23 valva had somehow wedged it in, but they still, at this 24 point, there's no known phenomena that actually caused that
<x 25 failure.
U l
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Covera 3e 800 336-6645
24594.0 269 7 >;RT O.
1 MR. MICHELSON: Davis-Besse people had actually 2 opened that valve under delta p conditions then? Full 3 delta p?
4 MR. PAGE: That's correct.
5 The utility proposed several corrective actions ,
-6 with respect to these valves. Their first corrective 7 action was to set the bypass limit switches on basically 8 what's considered a rule of thumb, which I believe they got 9 from the MOVATS testing, which is 20 percent of full disk, 10 or I guess the new word today is obturator travel, and 11 using the MOVATS system to verify all settings were correct.
12 Additionally they were to increase the open torque switch O 13 settings to a maximum which precludes valve damage.
14 I think this is something we'll definitely want 15 to follow up on, on what that exactly means. Because in 16 their evaluation of stem stresses it appears tSat they are 17 only looking at P over A, which is strictly one stress on 18 the valve stem. That's something I'm sure we'll want to 19 evaluate in a little more depth.
20 The Staff's findings, which, as we just 21 mentioned, are preliminary at this point, and they are 22 . covered in a memo dated 10/3/85. Number 1, with respect to 23 the Torrey Pines study which the utility had been using to 24 set their switches by, was that it should not be used by
{J 25 itself. It is very theoretical in nature. It is strictly ACE-FEDERAL REPORTERS, INC.
202-347 3700 .NationwideM 800-33H646
24594.0 270 RT 1 based on design drawings. With respect to the 599 and 608 2 valve it appears that some of their dimensions were in fact 3 incorrect, even though their own procedures indicate that 4 they do a field verification of these dimensions.'
5 So there was something that fell through the 6 cracks here, we are not quite sure what happened. But in 7 any case the bottom line is it's strictly theoretical and 8 nice as a baseline of what you should be seeing, but then 9 you should also look at that more closely.
10 It requires excellent preventative maintenance.
11 The Torrey Pines study also indicated that on the valves 12 they looked at at Davis-Besse, 25 percent of the valves had 13 baron precipitation on the stem. And they said that this 14 is something that just couldn't be allowed if you are going i
15 to use"this study.
16 The second staff finding was that they needed to 17 determine all the operational conditions to which these 18 valves were going to be subjected. Now, that will include 19 normal conditions, test conditions, which very often 20 involves no delta p of any sort, and also design basis 21 accident. It appears that these valves are dedicated for 22 just numerous ' varying conditions and we are not sure if 23 they are really suitable for all these various conditions,
- 24 because that is a wet gauge valve and could in fact damage 25 the valve in ome cases if you are closing it with no delta V}
('
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverase . 800-3M6W
24594.0 271 RT 1 p and they are in fact set up for design basis conditions.
2 The report indicates additionally that they encountered 3 greater than 1050 Psig, and to this point they haven't 4 identified just what the pressure was that they saw.
5 Additionally, we are asking them to use 6 confirmatory tests at available plant conditions. What 7 this means is to vary the temperature and pressure with 8 whatever they have available to compare it to their 9 theoretical settings, try to get an idea of: This is where 10 we should be and this is what we are seeing at the 11 conditions that we can reproduce. Maybe that will give you
,, 12 an idea of the margin as far as what you should see, i \
13 considering the valve degradation as it is. And maybe then 14 you'll know where to set it, if you are going to use that 15 valve for all the conditions.
16 Additionally, one other thing that came out is 17 that the thermal overloads must be in place, we feel, 18 during testing.
19 The utility has used regulation 1.106 and has 20 basically bypassed all those thermal loads for safety 21 related valves. We are a little concerned about that. I 22 guess there's a lot of discussion about that regulations 23 guide. It seems to have been misused somewhat.
24 That covers the ifrst set of valves. I don't
( ~ ') 25 know whether we should stop here and discuss questions -- I U
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 6646
24594.0 272 RT 1 think it would be better to discuss questions on these f
2 first valves independently, because the persons on these 3 PORVs and safeties are different than myself.
4 MR. MICHELSON: Are there any questions?
5 MR. REED: Well, just the point that I don't 6 know how much boric acid chrystalization, boron 7 crystallization was found but you are also going to have on
'8 a packed stem valve, some boron scum and some boron 9 crystals on the stem. I never felt it was a factor with 10 respect to stem stroking.
11 MR. PAGE: That it wouldn't affect the road on 12 the stem? ~
CE) 13 MR. MICHELSON: There isn't any boron involved 14 in these.
15 MR. PAGE: Not in these particular ones. But 16 the study-itself brought out the fact that the study is 17 only good if you have an excellent preventative maintenance 18 program and it's noted very clearly this is one thing they 19 came across which they felt a concern for.
20 MR. REED: You'll always have some with a packed 21 valve and that's what most of the valves in.the tree are, 22 PWR, some scum or minute crystallization ~in the very best 23 and par excellence maintained condition, and I don't think 24 it's a factor.
{) 25 MR. MICHELSON. Other questions?
l ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 33H646
24594.0 273 g- RT.
v 1 Why don't you proceed onto the next part.
2 MR. PAGE: This has to do with the PORV. It !
-3 appears the PORV was actuated three times and during the 4 third actuation it remained open. It's a Crosby valve, 5 licensee's investigation did not conclusively determine why 6 it failed and, again,- they indicated that possibly there
-7 was some foreign material within the disk.
8 One of the problems that we the staff felt was a
, 9 consicerable problem was that the thing had not been stroke 10 tested since September of 1982, which appears to be 4
11 contrary to the licensee's proposal in their own IST 12 program, which we don't understand exactly. I think that O 13 they had proposed shutdown testing and they had not been 14 doing that.
15 MR. REED: This PORV was backed up with a motor 16- operator?
17 MR. PAGE: I think all of them are.
l 18 MR. REED: Was the motor operator in the open l 19 position, do you know? Just holing against a PORV?
l 20 MR. HAMMER: The plant was operated with the l
21 block valve open.
l 22 MR. REED: Open. And there was no excess I
j 23 leakage?
24 MR. HAMMER: Through the PORV, you mean?
(~} 25 MR. REED: Right. Single valve against pressure v
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverase. 800-336 4 646
.- - ~. _
24594.0 274 B
r, RT d
1 on the pressurizer, and there was no excessive leakage 2 because they hadn't closed the block valve, obviously.
3 MR. HAMMER: I would imagine not because you'd be 4 limited -- on RCS leakage. 1 5 MR. REED: That brings into focus the 6 complicated situation of hydrogen separating out in the 7 riser line and the valve holding against' hydrogen, and 8 probably some complication with respect to boron scum and
! 9 it's an internal pilot operated valve, probably impacts on 10 the fraction factors of the piston and cylinder wall. It 11 becomes a very complicated scene and they hadn't exercised 12 it.
13 Frankly, if I had a pilot operated relief valve 14 I would find some way on a monthly program to close my 15 block valve and exercise my pilot operator and give myself 16 a little push of the hydrogen gas through.
t 17 MR. HAMMER: The staff in general _has generally 18 taken the position that PORV'should be stroked, at a 19 shutdown interval, at a minimum, but generally not to 20 ' exercise power because of the chance of a --
21 MR. REED: It can be exercise power.
22 MR..PAGE: To further enhance what we just 23 . discussed there, I guess Robinson Un: t '2 a few years ago 24 .had an-incident with their PORV, whedd they had gotten into 25 a situation where they wanted to relieve pressure but.their l
l l ACE-FEDERAL REPORTERS, INC.
- j. 202-347-3700 Nationwide coversee soo.33H646
24594.0 275 BRT O
1 PORV was quite a leaker, my understanding was, and so they 2 just opened the block valve. That took care of the spike 3 they were seeing coming. Then they couldn't close tha 4 block valve.
~
5 As a result of that we did a study of at least, 6 I don't know, 10 or 12 plants, I forget what it was, and it 7 appeared that 70 percent of them run with the block valve 8 closed. That's just off the top of my head, but I think it 9 was about 7 out of 10.
10 MR. REED: That all happened because they allow
- 11. concentrated hydrogen to build up in the riser pipe. If 12 you exercise you won't get concentrated hydrogen gas in the 13 riser pipe.
14 MR. PAGE: Your proposal is to close the block, 15 bring down the pilot --
16 MR. REED: Open the pilot, you get a gush, fill 17 the space, you purge the little bit of gas you are 18 accumulating and then cycle it back.
19 MR. PAGE: You are just concerned about getting 20 rid of the hydrogen than stroking the valve.
21 MR. REED: If you fill up the hydrogen you are 22 going to get leakage and it's a domino effect that's going 23 to bring you into the next problem. .
24 MR..MICHELSON: I think we'll have to move on.
I
! 25 MR.-PAGE: Main steam safety valves. Apparently l
p)3
'w l
i i
I ACE-FEDERAL REPORTERS, INC.
L 202-347-3700 Nationwide Coverage 800 336 4646
24594.0 276 f-[RT
.U 1 there were 18 of these valves, all listed during the event.
2 Some of the valves were more out of setpoint than 1 percent, 3 which was a requirement of their tech specs, and they were 4 all in the low side, apparently. Some as high -- as low as 5 3 percent.
6 There are dresser spring actuated valves. Eight 7 of them were sent to Wiley for full flow steam testing to 8 determine what setpoint they were at and collected -- all 9 the other stuff. The inward points are slightly smaller 10 than specified by the manufacturer of the valves but nobody l 11 seems to know whether that's affecting the operability of s 12 the valves yet. It's still under study.
13 MR. REED: But the valves did open and did close.
1-1 14 I wouldn't worry that much about the percent lowdown.
15 MR. PAGE: It appeared that the ones that were 16 out were on the low side.
17 (Slide.)
18 The atmospheric vent valves. Apparently both of 19 these valves opened several_ times during the event. One 20 did not reclose at the correct setpoint, and the testing ,
21 apparently showed that there was some sort of electrical 22 problem. The valve which didn't close at the proper 23 setpoint also apparently was still experiencing leakage i
,\
24 after closure. And that's about all that can be said about,
{} 25 that. They are air-operated.
l ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6646
m.- .
24594.0 277
(' RT l N
1 (Slide.)
2 MR. REED: Another thing you might look at, you 3 can get a problem with air-operated valves based on 4 .icrient'ation, whether they are installed in a vertical pipe, 5[ c horizontal pipe, or something like that.
ye 6 ,q ,
- I would check into whether they are in a 7 vertical pipe or not. That may very well have something to l
8 dolwith the reliability of function.
9 MR. PAGE: Just the run, not the actual mounting 10 of the valve?
, , , 11 1 MR. REED: The mounting of the valve.
12 4 MR. PAGE: Whether it's mounted horizontally or (3. ) '
13 vertically?
14 MR. REED: It has to do with the very heavy 15 weight on a very flexib'Le stem, . and the weights and the way 16 the springs are arranged. If you mount it in the 17 horizontal on a pipe they are generally much, much more 18 reliable.
19 MR. PAGE: The horizontal is better?
20' MR. REED: In a horizontal run rather than 21 vertical run. The valve itself would be mounted vertically, 22 90 degrees from that.
23 / MR. PAGE: The last was'the determined bypass
. 2:4 valve.. The valve had failed prior to the actual incident.
l
(])' 25 That was one of the concerns, of. course, whyfwas it not l
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 6646
24594.0 278 r~" RT .
(_)
1 noted prior to the incident. The disk was separated from
^
2 the stem and about the -- I guess what happened during the 3 incident was the breakage of the ygke. The best thing th a't 4 I remember from reading the write-up was that they figured 5 a slug of water slammed into the disk and pushed it into 6 the bottom of the stem. This was a result of steam traps 7 that couldn't drain for a couple of reasons. fr 8 It appeared that the isolation valve on the 9 steam traps was locked closed, and it also appeared that 10 some of the steam traps themselves were a little clogged up.
11 And So, this isn't a safety related system, by the way. /
12 the safety related concerns, however, that came out from5 13 this review was that the loose parts from the valve were 3 14 never accounted for. It pointed to poor maintenarice, poor 15 preventative maintenance for sure, and the possibility of 16 generating missiles in the area, immediate area of the s
17 valve.
4 18 So, it could be a systems interaction-type of 19 concern.
20 MR. REED: Just a question of clarification.
21- This was one valve in a bank of, say, four or eight valves 22 for steam dump?
23 MR. PAGE: This is turbine bypass.
-24 MR. REED: Turbine bypass or steam dump.
25 .MR. ROTHBERG: What the system is supposed to do
() 1
- ACE-FEDERAL REPORTERS, INC.
202-347 3i70 NM Coverage 800-T R " 3
N 24594.0 279 r"RT (_) 1 is condensate, instead of dumping to atmo' sphere it dumps to 2 the condenser through this system. l , 3 MR. REED: Like the atmospheric dump, only this 4 dumps it into the condenser and it's in a bank of valves. 5 MR. ROTHBERG: Yes.- I think there are six. 6 MR. REED: II m wondering if there's one of these 7 in trouble and there should be an upstream isolating valve
, . , +,
.it' 8 why they didn't take it out of-service?
9 MR. ROTHBERG: As a matter of fact they were 3 l j -10 only using two valves in this whole system because they 11 found that they got better -- they say that they got better 12 service out of using two valves so they used these two O' 13 valves. i It turned out this one was one of the ones that
, 14 was broken. /
15 MR. MICHELSON: Does that take care of it? 16- Thank you very much. We'll have to move right onto Jim 17 Jeffries. He's going to discuss some things that he found 18 in looking through some information that he had. 19 Jim, do you want to tell us what you got? 20 .MR. JEFFRIES: Yes. Okay. I recently issued a f qq ,, 21 draft report investigating the' nuclear industry valve y J22 qualification process. This was generated as a result of a
, _ p (>, , 23 concern Mr. Reed had regabding the fragmented nuclear valve
.,g l 24 program in the iddustry. Basically what this report does
'25 is to examine some'ch the causes of the fragmented picture.
((}
~
1 ACE-FEDERAL REPORTERS, INC.
, $ gam ' , [ I Nationwideh 800-336-6646
24594.0 280 RT i 1 MR. MICHELSON: Be sure to speak up loud enough 2 so the people in the back can hear us. 3 MR. JEFFRIES: The bottom line is a nuclear 4 valve qualification process is fragmented in part because 5 of the following reasons: Number 1, potentially inadequate 6 valve specifications; 2, the absence of a valve problem 7 evaluation center; 3, the absence of an industry oversight 8 work group responsible for directing a coordinated 9 engineering effort examining valve problems; 4, 10 insufficient intercommunication; 5, absence of a full scale 11 test of facilities; 6, the lack of an adequate valve 12 maintenance criteria; and, 7, the inability or unwillingness ( \~)' 13 of any one organization, namely utility, regulator, or 14 valve manufacturer, to build and maintain a testing 15 facility. 16 In addition to these efforts, I talked to 17 individuals outside of the nuclear industry to see if there 18 was a process that could be used, or that could be examined, 19 to see -- to compare that process with the nuclear industry 20 process to determine whether there was anything different 21 or anything additional that should be added to the process. 22 So, what I basically ended up doing was talking 23 to people at Morgantown Energy Technology Center. One of 24 the key differences between their process, which examined (~3 25 lock copper valves which are used in coal gasification, was q ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
24594.0 281 RT 1 that they had a full scale testing facility. And they were
? able to get the cooperation of the manufacturers, the plant 3 people, in terms of reviewing the valves, establishing 4 whether they operated, and the appropriate -- in the 5 appropriate environment, and the individuals that I talked -
6 to indicated that that was a key plus for them in terms of 7 getting cooperation, in terms of addressing qualification 8 of problems. 9 That's basically the conclusions that I had in 10 the report. 11 There was a section that looked at delta p 12 problems for valve closure in high delta p service.
~
;3 One of the conclusions that I came up with in 14 that particular area was the fact that it appeared that 15 many of the problems are attributed to. torque switch 16 failures. Okay? As opposed to actual calculation, ability 17 to calculate the delta p. So that, in essence, is the 18 conclusions that I came up with for this report.
19 MR. MICHELSON: I'm not sure I understand the 20 torque switch problem on high delta p. Tell me more. 21 MR. JEFFRIES: Okay. Basically, what I did, I 22 called several valve manufacturers and -- to get an idea of 23 how they go about calculating delta p. Basically, it's l 24 upstream -- downstream of the valve and across the valve. p 25 Okay?
- v. ;
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
l l 24594.0 282
,/*RT
-]
l 1 In essence, they can calculate that with no 2 problem, all three components. But when you have to 3 calculate the individual aspects it becomes very 4 complicated and as a matter of fact one of the points that 5 was indicated to me was the fact that they can -- the 6 manufacturers and designers can do that but because of the 7 time and the amount of money involved -- okay? It becomes 8- very critical. 9 So, what I did, I tried to get an indication, 10 well, okay, is the methodology that you use f airly 11 conservative? Or is there a question of accuracy? And 12 from what I gather, there is no problem in terms of O 13 accuracy. Okay? So then I said: Well, okay, fine, is 14 there a problem elsewhere? You know, is there a 15 misapplication problem? Is there a maintenance problem, et 16 cetera? 17 And in my investigation _I started going through 18 some of the old AEOD reports and one of the indications was 19 that the valve operating characteristics changed with time. 20 And as a result of that, it leads to torque cwitch 21 adjustments and as a matter of fact it was identified.that 22 about 25 percent of the motor-operated events were 23 attributed to torque switch malfunctions. And from what I' 24' can gather, it's a torque switch _or a maintenance problem 25 as opposed to really a design problem. The crux of the (]) ! ACE. FEDERAL REPORTERS, INC. t RFAPJ7 TED Nationwide Ctri"] 800-336 4646
'24594.0 283 f -9RT I NY l 1 problem.comes into play when you have to make an adjustment, 2 how do you go back and assure that you have the proper 3 torque switch settings, and that's where MOVATS can be a
-4 key player in that regard.
5 - MR. MICHELSON: Because unless you applied full 6 delta p and then watched the valve function under that 7 condition, you wouldn't know if_that was the appropriate 8 setting for present day conditions or not? 9 MR. JEFFRIES: Right. 10 MR. MICHELSON:- Very often they do not apply the 11 full delta p test, they use what I call a nominal load test.
~12 MR. REED: Jim, your conclusions seem to jibe
(~#) 13 with a lot of what has been happening here today; such as 14 the absence of a valve problem evaluation center, and 15 you've heard about how do we find the parents and the 16 children here and get this thing together and get some 17 unfragmenting, or coordination of the overall valve problem. 18 One conclusion you didn't mention in your 19 conclusions here is prohibitive cost. 20 Well, I can understand that to tie it all 21' together, it is going to be costly. Maybe it wouldn't be 22 so costly if we could get all the coordination and 23 intercommunication, which you mention,-to happen. 24 I don't know. How would you.think -- you've (} 25 heard the whole story, have been looking at it, what would ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 6646 l
24594.0 284 RT 1 you think would be a practical and reasonable way to get I l 2 this fragmented situation coupled? I 3 MR. JEFFRIES: I guess my first approach would 4 be to contact INPO, because of all of the organizations 5 that I talked to, INPO comes closest to emulating a valve 6 evaluation center. And, in essence what I would do then, 7 once I had INPO at least committed to pulling together the 8 manufacturers, the utility members, the regulators as far 9 as principal organizations, establish a program plan. Once 10 that has been taken care of I would then have to focus on, 11 okay, fine, how can I -- how can the industry, per se,
,, 12 assure that valves are going to operate as intended?
()- 13 I~would then have to ask, or find some way of 14 either getting funding through utilities, manufacturers, 15 government -- either all government or some type of 16 incentive in terms of funding a facility that can provide 17 full scale testing capability that to me is critical. That 18 seemed to be cropping up all over the place. 19 So the key things would be having an 20 organization that would take the lead, namely INPO. And 21 the second aspect would be having full scale testing 22 facility. I think as far as funding, that can be handled, 23 again, on a collaborative scope. 24 MR. REED: Something like what you just said was I (*
'/
m 25 mentioned because I guess Newmark was mentioned as a leader. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
i l l 24594.0 285 BRT g
%).
1 I of course threw a little cold water on that because I 2 said they were all vice-presidents. This is a working man's 3 issue. This is a "right down in the trenches" issue. Dig 4 this one out and get it put together and have it be less 5 prohibitive in cost and not windmill around and waste all 6 kind of money. 7 Hey, there are a lot of smart foremen out there 8 in the industry and workmen in the industry who probably 9 know a lot a';out the tricks cf this situation. If they 10 could be brought into the scene early, it may very much 11 limit a lot of the evaluation center just building 12 monuments. , 13 MR. MICHELSON: This shouldn't be a regulatory 14 issue. It should be a very practical plant issue. But 15 there are some safety implications that a regulator would 16 like to assure are being taken care of. It's basically, 17 you know, the plants ought to be doing this for themselves 18 simply for their own self protection. 19 MR. JEFFRIES: My investigation seems to bear 20 that out. I talked to people at Trojan, I talked to people 21 at Boston Edison, and I talked to people down at Duke Power. 22 And they all~have instituted what they call valve 23 betterment or valve improvement programs in general. Okay? 24 In essence, the thing that generated the need 25 for this was availability. Okay? Because valves {~j)
~
L ACE-FEDERAL REPORTERS, INC. i 202-347-3700 Nationwide Coverage 800 336-6646
. l 24594.0 286 RT 1 contribute heavily to plant unavailability, they, in 2 essence, felt that, hey, we've got to do something to 3 improve it, therefore we are going to -- we've identified 4 that valves are the key complaint within the two programs --
5 we are going to institute prog' rams to correct the problems. 6 What I want to point out is, okay, the industry 7 is fragmented but therefare indications that some plants 8 are :taking the initiative to address these problems. 9 MR. MICHELSON: Without regulatory pressure. 10 MR. JEFFRIES: Right. 11 MR. MICHELSON: Are there other questions? 12 MR. REED: I would like to ask Mr. Curry if you 13 would like to modify his recommendation with respect to a 14 users' group or owners' group to try'to kick this off or 15 tie it together or get it coupled. Do you think it ought 16 to go the INPO route first? i 17 MR. CURRY: I'm not sure. I'm going to ask John 18 to answer that one. 19 MR. MILITO: I'think they could put it out on a 20 not- to get interested matters together but I don't know if 21 INFO is the one to do it because it's an auditing group 22 rather than research -- what you'll be doing here is more 23 research and development testing and they are really an 24 auditing agent.
. 25 I would use'their note pad possibly to get l
l ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide coverase 800 336 6646
1 1 24594.0- 287
- ( ,RT U
1 people interested. 2 MR. REED: I don't know what this Subcommittee 3 is going to do or what the full committee is going to do. 4 I would like to think there would be some sort of a letter 5 coming out that there are claims to be coordinated, coupled -- 6 and it looks like an industry issue which is complicated 7 and that somebody ought to find a leader, a parent, a way 8 to do-it. 9 MR. MICHELSON: I believe that does it. Thank 10 .you, Jim, for your time. I think we'll need to take a few
'll minutes of Subcommittee time in-terms of where we go from 12 here.
O 13 When I was at Davis-Besse last week I brought 14 back a ton of data packages on the MOVATS testing they have 15 done to date. I. think Richard now has these. I didn't 16 make copies of them. I thought maybe you'd look at them 17 and decide if you wanted to look in depth. Then you could 18 get copies of these typical ~ data sheets. They are very 19 interesting to go.through, at least, you know, pick out a 20 few typical valves and look at them but they are available 21 in case you wish. 22 The other question I would like to raise for 23' Subcommittee investigation, which we have just kicked 24 around and finished talking about several times today, is 25 we have talked INPO on two occasions to come in and chat {} i I~ 'l ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 6646 _
l l
\
l 4 24594.0 288 RT 1 with us. On both occasions they essentially didn't show up. 2 We have asked them before, for interest -- come 3 in and tell us what they are doing on valves. You can 4 check your record and make sure. We asked them before. 5 Should we ask them again? 6 MR. WARD: INPO? We also asked EPRI to come in. 7 MR. MICHELSON: EPRI did not come. They are not 8 prepared yet. But INPO has turned us down before. I'm 9 sure of it. But check your record. 10 The question is, do we want to talk to INPO at 11 all? It is not -- there's not a good opportunity unless we 12 should for some reason have another meeting. I' ' l 13 MR. WARD: Have they issued any reports on valve -- 14 MR. MICHELSON: Yes, they sent to me some SER 15 and SOERs relating to valves. They also sent the report -- 16 I think Richard has them. Do they send them to you? They 17 sent them to me directly -- 18 MR. WARD: But just individual incident reports. 19 MR. MICHELSON: Yes. One of them is a 20 compilation after-survey they did of work earlier. If 21 Richard doesn't have it, I'll get it to him, although I -- 22 it has to be distributed under the usual confidence rule 23 for EPRI reports -- INPO reports. 24 MR. WARD: The problem is maybe they didn't do e'3 25 much except to look at individual incidents.
%)
ACE-FEDERAL REPORTERS, INC. , 2 7 -347-3700 Nationwide Coverage 800 336 4646
24594.0 289 RT 1 MR. MICHELSON: They have looked at a number of 2 incidents in which valves were involved to try to determine 3 some thoughts about root cause and so forth. They have not 4 been looking at valves in these reports, at least, as a 5 category problem area on what to do about the valves. I 6 haven't seen that type of report. 7 I talked to Olson briefly when we were in 8 Atlanta last March, I guess it was now. At that time they 9 didn't -- I didn't sense a very strong interest in what we 10 were talking about today. However, they may have a lot 11 different interests now. I would not -- in other years 12 gone by, maybe other pressures have become applied or
,7_ ) i \
13 whatever. So we presently have no further Subcommittee 14 meetings in the near future. So we don't have their input. 15 MR. JEFFRIES: Can I raise a point? I talked to 16 Dick Baker down at INPO and I'm presently trying to get 17 ahold of their valve maintenance workshop program. 18 Hopefully, in the near term, I will be getting a copy of 19 that program. Maybe that will be something that could be 20 included. 21 MR. MICHELSON: It will help. They have 22 conducted a number of valve workshops was my understanding 23 in the past. One of them about two years ago, something 24 like that. That's a little different sort of thing than {) 25 what we are talking about here today. l ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 336-6646
24594.0 290
RT s
1 MR. BROWN: Three years ago. 2 MR. MICHELSON: I would think by now the flavor 3 of their workshop could be shifted more towards what kind 4 of R&D can we do to get at the root cause of some of these 5 happenings. 6 MR. WARD: But that's not the sort o'f work INPO 7 is going to do, would be interested in doing. 8 MR. MICHELSON: Normally they don' t seem to be 9 interested in doing that type of work, that's right. They 10 like to stick to events, what happens, how to fix it. But 11 not to get too far into -- they would like to turn it over 7- 12 to EPRI if it's going to be deep, a lot of testing or ()3 13 whatever. They are just not prepared for it. 14 MR. REED: Carl, I think the first thing we have 15 to do is define where the Subcommittee has arrived and what 16 we agree on. If we agree on the problem, as we've sort of 17 heard it, that there is a coordination, coupling, 18 manufacture, utility problem, it's in a state of 19 volunteerism, it's fragmented, it's a very important issue 20 to safety and reliability -- if we agree to all these 21 things then it seems to me that the Subcommittee has to 22 decide, or the full committee, on an approach to get the 23 leadership, or find the leadership or get it acted upon. 24 Quite frankly it seems to me if we all agree on l' )
~-
25 all these things, perhaps we would go to EPRI or INPO or ACE-FEDERAL REPORTERS, INC. 202-347J670 N Coverage 800-3364646
g I 24594.0 291 l 1 ~both those: people at-the top level and say: We, the ; 2 Subcommittee, feel this way about the valve scene in the ? 3 . United States. What would you suggest doing about it? Is > 4 -- it going to be an industry activity and .something that 5 . falls to INPO or something? Or users' groups or owners' 6 groups? There must be somebody in EPRI, some committee or 7 something that looks at new projects. 8 MR. MICHELSON: One of the questions in my mind , 9 is does the Subcommittee merely want to make a 10 recommendation to the Chairman that this should be done and 11 let the staff do the chasing ~or should the committee do the 12 chasing or what? Normally this committee doesn't= chase O 13 that sort'of thing through the industry but rather makes a 14 recommendation to the chairman who then decides whether-he 15 wants to do it or not. 16 MR. REED: You mean this chairman? 17 MR. MICHELSON: No, no. The chairman of the NRC. 18 I think all we do.is write a letter supporting our case and 19 that's what I thought we should do, next, is prepare a
, 20' summary paper identifying what we have done to date and 21 some of the kinds of things we have observed-and then 22 making some firm recommendations which would go in the form 23 of a letter, then, to the chairman.of the-NRC. 'Oine of -
-24 those ' recommendations can . very well be - We should go out --
25 ILdon't know who you'll decido yet, we could either be the {'} ACE FEDERAL. REPORTERS, INC. 202.W-M00 NanonwWe Cowage 20 N MM ,
24594.0 292 F RT O 1 ACRS or ask the staff be charged to go out and try to put 2 such a program together, if that's our recommendation. 3 That's one of the things in my mind was do we 4 want to hear anything more. I think we've pretty well beat. 5 MOVs to death now. We did not in the process, though, hear 6 from INPO in the sense of finding out how they might feel 7 about this. I don't think we want to call another meeting 8 just for that purpose but I did want to highlight the fact 9 that we had not really gotten their view and we should get 10 a recent view, if you feel it's necessary. 11 MR. REED: On the other hand, of course -- 12 MR. MICHELSON: The first question on the table 13 is: Do we think it's necessary to talk to INPO about this? 14 MR. REED: On the other hand, of course, some 15 alert company person from Philadelphia Electric might go 16 home and warn his officers that something might be coming 17 out of this. 18 MR. MICHELSON: They don't have to do that. 19 INPO is going to get the transcript, I'm reasonably sure, 20 and read it. I think they do fairly regularly because I 21 get sometimes questions that indicate they do. 22 MR. MILITO: I have already made the notes to go 23 to my managemer.t when ,I,get home. , 24 MR. EBERSOLE: The transcript shows industry is () 25 offered an opportunity to straighten itself out or if it ACE-FEDERAL REPORTERS, INC. 212 147 3700 Nationwide co verans 2 336 4 646
24594.0 293 RT 1 doesn't it will come down from another direction. 2 MR. MICHELSON: I think industry has indicated a 3 desire that perhaps something of this sort should be done. 4 I don't know that anybody made any offers to do it but 5 industry has indicated that perhaps there would be an 6 interest on the part of at least a few utilities to get 7 something like this going. 8 MR. EBERSOLE: I thought it was interesting to 9 find that it had both an operational advantage to take this 10 thing on as well as a safety advantage. They get paid both 11 times. 12 MR. MICHELSON: I have reason to believe several ,q k/ 13 utilities are already doing a great deal of work along 14 these lines, including those we heard from today. I don't 15 think it's that big a deal, now, to start to get a 16 coordinated effort out of that nucleus. 17 MR. EBERSOLE: How about the others? Carl, just 18 the other day I read in LER where that backup safety valve 19 on the boilers failed because of some air pressure 20 transient -- it was a leaking system -- 21 MR. MICHELSON: This is another question wo have 22 to ask ourselves, now. Do we want to go beyond 23 motor-operated valves, which are the lion's share of 24 boilers but not necessarily on PWRs, and if so, what else 7 25 do we want to look at, keeping in mind that there are other ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-336-6646
_.y _ . _._ ~ . _ _ _ . _ . _ _ . _ _ . . _ . - . _ _ . - _ . . _ . . _ . . . - _ _ ___ ._ i L 24594'.0 294 l t 5, PRT- :i (_ . a 5 El things in this world.besides valves that are pretty j
- 2 .important, too. Perhaps like auxiliary feedwater systems.
l j [3' 'MR. EBERSOLE: There's valves and there's pumps. i. L 4- MR. MICHELSON: And turbine drives and motor i I . . I. 5 ' drives. What are the wishes of the' Subcommittee in what , 6 other. valves do we wantEto give some attention to?
-i i- 7 MR. EBERSOLE:. .What about all'the rest as a ,
8 family?: i j 9 MR. MICHELSON: How much time do you have? '[ i f .10 MR. WARD: Do you want to keep all-this on the : I 11: ' record in a formal meeting? 1 i ! 12' MR. MICHELSON: I think so, for this purpose. i
- -(:) 13 Otherwise I would have released the reporter. I think you 1
i I- 14 want a-report of the wishes of the Subcommittee. l l ! { 15 'MR. REED: I think we ought to cover the design ;
.16 applications, proof testing, performance, reliability of i t 17 valves as an overall package. .Ought to be in that package. -
i 18 I wouldn't go at this time on-into pump and this other .
- j. 19 thing. I'm more concerned'about valves as an:overall issue. i 4-h r
120 .I think some valves are being Lused . bottom-side up or; l I I
- 21' ~ horizontal orevertical and allot of these things, the. wrong l
l 22 applica'tions, and I think the indus.try .has to move forward f ! '23 andEcollect this information, ferret it out and get l !~ .24 : improved. reliability in the whole thing. j'(]} : ~25 MR. MICHELSON: What should be the function of L i f ACE-FEDERAL REPORTERS, INC.' I ., . . . _ , . . _ . . .. -. N N J _ L NM Cmemse, ,_,MM,___,,_,,_,,_,
24594.0 295 ,~~RT v 1 the ACRS? 2 MR. WARD: I guess we heard two things, neither 3 of which we are greatly surprised at. One is that valve 4 failures are an important contributor to risk and we know 5 that from PRAs that have been done. We also know that from 6 anecdotal experience. And we haven't heard anything that's 7 altered that perception. 8 Then we've also heard that while there are 9 activities going on in the industry -- well, first of all 10 there seems to be potential for improving valve performance 11 rather markedly. There are activities going on in the 73 12 industry, but they are fragmented, and it's not at all r 13 clear that any very dramatic improvements all across the 14 board are going to come from this very soon. We suspect, 15 if there were some sort of a more organized effort to 16 attack the problem, it might come sooner. 17 Now then, the question is: Is this something 18 the NRC should do? In terms of, you know, developing 19 regulations or reg guides or is it something that most of 20 us -- most of us would rather see the industry develop some 21 sort of a more coordinated program. 22 MR. MICHE T. CON : What should be the role of the 23 ACRS in attempting to foster such development and how f ar 24 should we go and how much further -- more time should we 25 spend looking at valves, for instance?
})
ACE. FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-fiM6 . . . . . . . _ . . _ _ _
24594.0 296 a" b;RT. 1 MR. WARD: I'm not sure if we heurd about 2 air-operated-valves, and other valves, we'd probably get 3 about the'same sort of story I suspect. So I'm not sure 4 there's much point;in looking -- 5 MR. MICHELSON: One of my concerns in those 6 areas'would be: Is there something equivalent of MOVATS 7 being developed to get the information on air operators, 8 for instance, versus motor operators. 9 MR. WARD: Sven for motor-operated valves,: 10 although MOVAT seems to be a useful technique for, you know,
-11 looking.into problems, it doosn' t appear that everybody is 12 'using it in that there still isn't a big umbrella ,()
13 coordinated program. 14 So I think you need that sort of approach for 15 all types of valves. 16 MR. EBERSOLE: You know I suspect air-operated 17 valves are even more susceptible to be enabled to' meet 18 emergency loads than motor valves are because they tend to. 19 have flat torque characteristics, the' motor drives -- 20 MR. MICHELSON: Okay. I think the feeling, the 21 consensus of ' the committee is that we would like to 22 continue to look at> valves. Do you have a feeling for how 23 much time we should reasonably devote? We have'been 24 averaging about every two to three months, I think.
's 25 MR. WARD:- That isn't what I said, Carl. I'm.
{~^J
- . ACE-FEDERAL REPORTERS, INC.
- m.347 37ao , .
N.iion.ia. cover . _ . soin364646 _
24594.0 297 (',RT V 1 not sure we are going to learn any more about the situation 7 by looking at alfferent types of valves. I suspect the 3 story is going to be pretty much the same. 4 MR. REED: I agree with what I thought Dave was 5- saying, and how he just said it now. 6 I don't think that the ACRS Subcommittee ought 7 to. continue to pursue the details of this valve issue. I 8 think our job is to make sure that it gets launched 9 somewhere and moved out upon by somebody. 10 I do think the NRC trying to get involved in 11 regulations and guides and these kind of things at this 12 point in time is very premature, and that they can't do it. 13 Until you've got the workplace input that needs 14 to be gotten from the children, and given to the parent, 15 you are only going to make mistakes, spin your wh 1s. 16 MR. MICHELSON: I have a little problem with 17 what you have said, though, because I thought you wanted to 18 start hearing about PORVs in detail like we have MOVs. Do 19 I misunderstand? 20 MR. REED: I'd be glad to do.it but I think if 21 we got the leadership established and decided to move out 22 on it,' they would find out these subtleties about PORVs. 23 MR. MICHELSON: I feel we have looked at a large 24 group of valves. We have found characteristically how the I (') s_e 25 problem is being handled. I think that's as far as we need l l l ACE-FEDERAL REPORTERS, INC. - FM5'463 Nationwideh 800 336 6646
24594.0 298 RT 1 to go. We don't have to look at all other valves to prove 2 that the same problems exist. Clearly, though, a program 3 is needed that will look at all valves and not just MOVs. 4 MR. WARD: Where should that program be? It 5 doesn't look to me that it's appropriate -- 6 MR. MICHELSON: To accomplish that, I would 7 suggest that I try to sit down and write a few pages to 8 cover the background material, what we've done today for 9 the full committee to know, you know, the basis for our 10 determination. 11 MR. WARD: I've got a proposal for what the 7s 12 committee might do but I'll make it to you privately; all ()
'~'
13 right? 14 MR. MICHELSON: All right. But I need to know 15 right now, what do you want me as Subcommittee chairman to 16 prepare? I can prepare a report. Or I can just simply 17 prepare a letter. Or I can prepare both. You know, what 18 do you think is the next step? 19 MR. WARD: The letter -- 20 MR. MICHELSON: The letter would be for the full 21 committee to send to the chairman. 22 MR. WARD: I think the full committee is going 23 to have to hear something. 24 MF. MICHELSON: I think they might even want to f') 25 read something before that. So I thought I would try to ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverste 800 336-6646
24594.0 299
'RT v
1 put together a few pages of background material. 2 MR. WARD: Probably a good idea. 3 MR. MICHELSON: And then make a present,ation to 4 the full committee and we will have to talk about, later, 5 what the full committee might want to hear. 6 MR. WARD: Yes. 7 MR. MICHELSON: This, I think, would be 8 reasonably scheduled for about January 5th; does that sound 9 about right? 10 MR. PAGE: I think we have some time in November. 11 MR. MICHELSON: That's early. g- 12 MR. WARD: Carl is not going to have his report G 13 written by then. 14 MR. MICHELSON: I have to put the material 15 together. Then I have to send it to the Subcommittee 16 members, along with what I think a proposed letter might 17 look like. 18 MR. WARD: We can postpone it. 19 MR. MICHELSON: Then they have to comment on it. 20 I think realistically it's January. I would say at that 21 time, probably two hours. 22 MR. WARD: If you want the committee to write a 23 letter I would say two hours. 24 MR. MICHELSON: They have to hear a lot and if f~ 25 they aren't satisfied, we have to come back the month after i ACE. FEDERAL REPORTERS, INC. 20244717')0 Nationwide Coverage an.336 6646
24594.0 300 BRT l') (J 1 that with more speakers and whatever they want. 2 Would that be the reasonable way to wrap up the 3 valve issues now? In the process we are going to blanket 4 all valves even though we only looked at one type. That 5 would be -- is that the consensus, then, of the 6 Subcommittee of the method and direction? 7 MR. EBERSOLE: Are we talking about valves just
, 8 as modules and components? We are not really talking about 9 systems, because there are some system questions, one thing 10 being that the historical tendency has been when you get in 11 trouble, shut off all outgoing flows of radioactivity.
12 When you do that, invariably you shut off the heat transfer, 13 and that brings about the problem. 14 MR. MICHELSON: That's an entirely different 15 problem. 16 Does that take care of our Subcommittee' meeting 17 then? Henry, did you have any comments or questions? 18 MR. JONES: No. 19 MR. MICHELSON: That finishes it. 20 (Whereupon, at 5:30 p.m., the meeting was 21 adjourned.)
, 22 -
23 24 i l 25 ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverap 800 336 4 46
-i CERTIFICATE OF-OFFICIAL REPORTER m
b This is to certify that the attached proceedings before the UNITED' STATES NUCLEAR REGULATORY COMMISSION in the matter of: NAME OF PROCEEDING: ADVISORY COMMITTEE ON REACTOR SAFEGUARDS , ! SUBCOMMITTEE ON RELIABILITY ASSURANCE (VALVES) I i DOCKET NO.: , PLACE:. WASHINGTON, D. C..
. o. DATE:
TUESDAY, OCTOBER 8, 1985 were held as herein appears, and that this is the original ) transcript thereof for the file of the United States Nuclear
- Regulatory Commission.
(sigt) f AM' I (TYPED)[p
- JOEL BREITNER ," '
Official Reporter ,
- R h b~rkkS D f [b N N .
- O
m R. MAJOR RELIABILITY ASSURANCE (VALVES)
/) ACRS SUBCOMMITTEE MEETING ON LOCATION Room 1167.1717 H St. NW.. Washington, D.C.
DATE _0ctober 8.1985 ATTENDANCE LIST , I PLEASE PRINT:NAME AFFILI ATION C. p:/ci/e/Ss N' _ c /MhM 4 & A4 . kry D u 12D 6,. /?ow/ ru ,wos -
- s .
a.
/J. TOWrS aon ,..i n w r pJ &
Q. MJJof 5 d u A b>ici>1 l 6O4 vvL-
/
EAAL T. 6 Ao -rd vra m /AEoD 4 W weiocuusne swo /ogicrop t.m p bm.ii, ~l, fer ri />;, .e // JJ i / b , , , b* f,3,, srm /oFL 4 Es, a.. 5>J %u. , D o,.s m . L z c. # o. . .
- m. s.,. u,, . .
% %rv 9 Rt_
L, v se no, e a a e, /un n
/ /M/ TdA?Qtt f Gn/2A b3 f)AP3'/A/G ELAlymm /3/menck i Ei,tew A C.
I r/. J 4 caw flee %7r
. Rf 144A// ~
irae/srzt
/
C* , M . Pl rad. 7) a_ 6 e c c.h. y 0echVel Poswr Gtr1
- s. ). bd bmL 6.. c ,
eene < . mam -Teaoo e oico,a / tu,c.re.ssc b Lnur F:L 4 ccH oiv e p US Co c.P. (]) Fl . K h t= Rc R S ,IN R C-
~~~?
R. MAJOR
^
ACRS SUBCOMMITTEE MEETING ON RELIABILITY ASSURANCE (VALVES) (v\ LOCATION Room 1167.1717 H St. NW., Washington, D.C. DATE october 8.1985 ATTENDANCE LIST PLEASE PRINT: NAME AFFILIATION , O A/ /f/ /fA- EC 0. m' co. h m. cum 3 ~ M ,
)?ie$n,w/ 'cr- Kiaaeal Atac))cJzein m/o/mr il no , f /1uhr' vso 9-u son n ac lcrsincan t r>arr k,ho~ QQ' Afb J475 N eerfWnrub t
$ $ 0 H n cRosi ne n t0
' ' ' #" * " " ' ""^" " ~ ~ " ~ " ' - "
O l?6TWA F /YtEI lM KMC kts,J D . S wus.- o Tatcra dotset/
/ L2 h id E H A J % Lcho Ghtssc bYSff4n h MJen B J <-a c L' C !ai/l c o y llenarnT f. wise Bascocx + vILcs y 8te.cwM oDh 16 m -n /L c-c
}.3.NAvvtuIch.& \bNiuis Run O
e V L
INTRODUCTORY STATEMENT BY ACRS SUBCOMMITTEE CHAIRMAN ON RELIABILITY ASSURANCE (VALVES) OCTOBER 8, 1985 WASHINGTON, D. C. The meeting will now come to order. This is a meeting of the ACRS Subcomittee on Reliability Assurance (Valves).. I am C. Michelson, Subcomittee Chairman of this meeting. The other ACRS Members in attendance are J. Ebersole, W. Kerr, G. Reed and D. Ward. The ACRS Consulta6t is H. Jones. The Subcomittee will continue discussions of valve reliability. A risk-perspective on valve performance will be sought. Also to be studied is the importance of valves from a safety standpoint. Richard Major is the the ACRS Staff Member for this meeting. O The rules for participation in today's meeting have been announced as part of the notice of this meeting that was published in the Federal Register on September 24, 1985. It is requested that each speaker first identify himself or herself and speak with sufficient clarity and volume so that he or she can be readily heard. We have received no written coments or requests for time to make oral statements from members of the public. We will now proceed with the meeting. Do any subcommittee members or consultants have any coments at this time? [ Opening statements, if any;.
',I will now call upon Gary Burdick of Research to begin:
W
__ ___ _._ _ _ . . . . _ ~ _ _ . . _ . _ _ __ _ ___ . _ _ _ . _ J t 1 -i .. 4 [ k l I j L 1 I i
- AGING CHARACTERIZATION t .
- l. [ }
jf ' AND DETECTION OF DEFECTS ! i IN ) 4 VALVES
- O (A PHASED APPROACH TO ASSESSMENT)-
i I i
'i -
i, 4 i
, A PRESENTATION T0:
. fs ACRS SUBCOMMITTEE ON RELIABILITY ASSURANCE l "'" 8' '85 O
- J. VORA, EEICB/DET/RES 1
Y U l
- .__._....-_.-__,_.,-._..,.,__,..-.____..._......:...,.-..._._,_.,.._...-~,_..-._-.-_,__
i Program Goals i: e To identify and characterize aging and service ll ) wear effects which,if unchecked could cause ' degradation of structures, components, and systems and thereby impair plant safety , l
- e To identify methods of inspection, surveillance and monitoring, or of evaluating residual life l of structures, components, and systems, which l will assuis timely detection of significant l
aging effects prior to loss of safety function e 'To evaluate the effectiveness of storage, l maintenance, repair, and replacement practices l ' in mitigating the effects of aging and diminishing the rate and extent of degradation caused by I aging and service wear
O O Q~ ~ 'l l l Working Definition of " Aging" i , The term, " aging" is used to represent the cumulative l f l degradation occurring within a component, structure l or system which, if unchecked, may result in loss of ~ function and impairment of safety. Factors causing aging / degradation may include: , t e Natural internal chemical or physical processes ( f i e External stressors and environment I ! e Service wear (cycling; vibrations)
~
! e Testing i t ! j e Improper installation, application, maintenance
?
i i
0 0 . G t 3.0 ReSearCh Approach I Phas's 11 PhaseI l e Operating Experience Verification of Improved Recommendations for l' i e Review and Analysis I,S,and M M Select I, S, and MM' o Systems e Review of Methods e Tests of Natured e Evaluation of "Mothbelled" and Technology for and g Aged Components, and $ Equipment I, S, and MM Components @ e Screening Type Components, Models, e Modification of Codes, go, Samples with Simulated Standards and Guides Examination and Tests Degradation e Lifetime Extensions I Assessment e Interim Recommendations e Cost / Benefit Study i for Engineering Tests I in Phans 11 ! Figure 1 k
- l 1
- 1, S, and MM - Inspection, Surveillance and Monitoring Methods i
i 1 i i _ . - - - , ___ _ . _ - _ _ - . . _ _ _ . _ _ _.- _ . _ . . - - _ _ - _
_u --,...4 - - - - - - - - - - .-s .a _. _4 . .,_m,,. _sa_ p a a ___,___2__ ,_ , _ gw i
.o e lidl
=
Il li! I,1 , m . , l [ bl : o=f N! l N hl31; i-a I i s
]
! < l ldi ipi
= '
11 l,l,l,l,I 4} llll! 'N jl1lllIl s i.g .. , s N 11,11 i
, O o. O i
i Utilization of Research Results (General Benefits): I
! e Better understanding of aging processes i
i e improved confidence in methods for detecting and mitigating aging and service wear effects o Provide a basis for timely and sound regulatory decisions regarding continued operation of nuclear plants of all ages i 4 1 e Avoid unplanned and costly plant shut downs 1 I e Reduce excessive testing i . i ! e improve reliability and availability 4 i .. -. . _.
l
~
I O O O' Utilization of Research Results(Specific Benefits) e Support the NRR " Maintenance and Surveillance Program" o Support implementation of 10 CFR 50.49 e Support and/or endorse IEEE Standards 317,323,344,382, l i 501,535,572,649,650 , l e Supoprt and/or endorse ASME-O&M Standards OM-01,04, 06,08,10,13,15,16 i - : e Support NRR in resolution of generic issue ll.E.6.1 "In Situ l Testing of Valves" l e Support NRR in evaluating aging and service wear reliability of diesel generators e Development of criteria for evaluating in-service performance of '.' Snubbers" as piping restraints e Support NRR in development of criteria for evaluating plans ! for "Mothballing" plants and reactivation e Support implementation of 10 CFR 50.51 while considering plant license extensions e SUPPORT ilRR Ill RESOLUTION OF GENERIC ISSUE N0.70 "PORV AiiD BLOCK VALVE RELIABILITY"
10 e o I d ' f r AGING CHARACTERIZATION , e AND-
- DETECTION OF DEFECTS IN .
4
^
- G-1 VALVES (A PHASED APPROACH T0-ASSESSMENT) 9 1 i 1
! l 3 i l . r
.i . i
! ' A: PRESENTATION T0: ; ACRS SUBCOMMITTEE ON RELIABILITY ASSURANCE :
~ 0CTOBER 8, 1985- i L 0- <J. VORA, EEICB/DET/RES 9 "E*7F"- T $ =$ 9 F F w % ? y'*F= *w =l 7 -y Wt q*' .sP- .++grya tw 1Tse'*+-i--9 .up- ._ $g.#*pte pq --t --
p> -$e gire**-- g -y 9 ep- w f wg=-erAq .,4y4- g NMye--4
Program Goals e To identify and characterize aging and service wear effects which, if unchecked could cause ' degradation of structures, components, and systems and thereby impair plant safety - l e To identify methods of inspection, surveillance and monitoring, or of evaluating residual life of structures, components, and systems, which will assure timely detection of significant i aging effects prior to loss of safety function e To evaluate the effectiveness of storage, maintenance, repair, and replacement practices l in mitigating the effects of aging and diminishing the rate and extent of degradation caused by ! aging and service wear
g g .
- g. ..
Working Definition of " Aging" The term, " aging" is used to represent the cumulative degradation occurring within a component, structure
- or system which, if unchecked, may result in loss of' function and impairment of safety. Factors causing aging / degradation may include: .
' e Natural internal chemical or physical processes e External stressors and environment . e Service wear (cycling; vibrations) e Testing e improper installation, application, maintenance l t
o O on
.I t
3.0 Research Approach Phasei Phas' e il e Operating Experience e Verification of Improved Recommendations for Review and Analysis Select I, S, and MM* e I, S, and MM Systems e Review of Methods e Tests of Naturally and Technology for I e Evaluation of Mothballed'* and Aged Components, and Equipment Components @ I, S, and MM g Components, Models,
$ e Modification of Codes, e Screening Type i for Samples with Simulated Standards and Guides Assessment Examination and Tests Degradation e Lifetime Extensions e Interim Recommendations for Engineering Tests e Cost / Benefit Study i
i ' in Phase 11 i Figure 1
- I, S, and MM - Inspection, Surveillance and Monitoring Methods 1
! l' i
a -e u + -- ~' --wa .6--,ea--
- - - .nm . eA_
lg @ p% Y@e : 11 - v 1I {1{}I lin 13 111 u g [ }lll O N! \ N l;}
)
=
e h.Ib 1 11,
= -
l11 11,111 . au 4j,i l11l)'N nl }}}} s i.i~t N l ' llin s, o 11
= -
li,,i k g 1111 ,
d
~
O O . Utilization of Research Results (General Benefits): i e Better understanding of aging processes e Improved confidence in methods for detecting and mitigating aging and service wear effects ' e Provide a basis for timely and sound regulatory decisions l regarding continued operation of nuclear plants of all ages e Avoid unplanned and costly plant shut downs 4 > e Reduce excessive testing i I e Improve reliability and availability i l w --$-gMg--wee,g w,--e _ _m ee es ,- ww w w w w-we ww mv-e eiw.wseme s-asem+w' + we wes m e w w w-ewww ee- g -m- w-s-v-we* --w-e e e-m -ewee- - =-wv ve----- - e w ----+e -=-
O O O ~l. Utilization of Research Results(Specific Benefits) . o Support the NRR " Maintenance and Surveillance Program" e Support implementation of 10 CFR 50.49 e Support and/or endorse IEEE Standards 317,323,344,382, I 501,535,572,649,650 . o Supoprt and/or endorse ASME-O&M Standards OM-01,04, i 06,08,10,13,15,16 i l = e Support NRR in resolution of generic issue ll.E.6.1 "In Situ ! Testing of Valves" ! e Support NRR in evaluating aging and service wear reliability of diesel generators i e Development of criteria for evaluating in-service performance of " Snubbers" as piping restraints e Support NRR in development of criteria for evaluating plans for "Mothballing" plants and reactivation l l e Support implementation of 10 CFR 50.51 while considering plant license extensions l
- e SUPPORT URR Ill RESOLUTION OF fiENERIC ISSUE N0.70 "PORV AND BLOCK VALVE RELIABILITY"
.. O o. o i
VALVE IMPORTANCES FROM THE ANO-1 1 REP DAVID J. CAMPBELL JBF ASSOCIATES, INC-l ! OCTOBER 8, 1985 l l , l
O O O i l l , BACKGROUND
- REDUCED CORE MELT FREQUENCY MODEL
- BASED ON IREP STUDY OF ANO l
- BROAD DEFINITIONS OF VALVE TYPES i
i l l l
O O O IMPORTANCES FUSSELL-VESELY THE FRACTION OF THE CORE MELT FREQUENCY THAT INVOLVES FAILURE OF ANY COMPONENT IN THE SPECIFIED GROUP BIRNBAUM THE SENSITIVITY OF THE CORE MELT FREQUENCY TO CHANGES IN THE FAILURE PROBABILITY OF COMPONENTS IN THE SPECIFIED GROUP s ,,,, ,n n e oc,, s rove
,-,e--,-., -, , -- , , , - . - , , _ , - - _ - - - - - r - - - -. - - - - - - , . -_______-
O O O IMPORTANCES OF MOTOR-0PERATED VALVES IN ESF SYSTEMS FUSSELL-VESELY BIRNBAUM SYSTEM IMPORTANCE IMPORTANCE EMERGENCY 5 E-2 P~ 8 /##'
,* ?
. y# 5 E-2
. c ~ s,A FEEDWATER HIGH PRESSURE 3 E-2 2 E-2
'lNJECTION i Low PRESSURE 3 E 2 E-2 INJECTION 1
- REACTOR 7 E-5 5 E-6 BUILDING SPRAY SERVICE WATER 8 E-2 5 E-2 REACTOR COOLANT NA NA l
. EMERGENCY ROO M 7 E-2 2 E-2 j COOLING s % (,,, J) (,n- a it - > A<A)
O ~ O O IMPORTANCES OF CHECK VALVES IN.ESF SYSTEMS FUSSELL-VESELY BIRNBAUM SYSTEM IMPORTANCE IMPORTANCE - EMERGENCY 3 E-3 8 E-3 . FEEDWATER
- HIGH PRESSURE 7 E-4 2 E-3 '
i INJECTION low PRESSURE 1 E-5 9 E-4 INJECTION , ! REACTOR 4 E-6 1 E-5
- BUILDING SPRAY l SERVICE WATER 2 E-3 3 E-2 REACTOR COOLANT NA NA l EMERGENCY ROO M 2 E-3 2 E-2 i COOLING
O O O IMPORTANCES OF AIR-0PERATED VALVES IN ESF SYSTEMS FUSSELL-VESELY BIRNBAUM , SYSTEM IMPORTANCE IMPORTANCE ! EMERGENCY NA NA I FEEDWATER HIGH PRESSURE NA NA INJECTION L0w PRESSURE 7 E-4 4 E-3 ! INJECTION i REACTOR NA NA. BUILDING SPRAY j < SERVICE WATER 9 E-5 5 E-6 REACTOR COOLANT NA NA ! EMERGENCY ROO M NA NA COOLING
l . O O O IMPORTANCES OF MANUAL VALVES IN ESF SYSTEMS 1 FUSSELL-VESELY BIRNBAUM SYSTEM IMPORTANCE IMPORTANCE EMERGENCY 5 E-4 1 E-3 l FEEDwATER HIGH PRESSURE 1 E-3 2 E-3 ) INJECTION Low PRESSURE 7 2 E-2 INJECTION
~ '
REACTOR 5 E-6 5 E-6
- BUILDING SPRAY
! SERVICE WATER 7 E-4 3 E-3
- REACTOR COOLANT NA NA EMERGENCY ROOM 2 E-2 9 E-2
- COOLING i . - - - .. -- _ .. _ -_
O O O IMPORTANCES OF RELIEF VALVES IN . ESF SYSTEMS (FM'to) FUSSELL-VESELY BIRNBAUM SYSTEM IMPORTANCE. IMPORTANCE EMERGENCY NA NA
-FEEDWATER HIGH PRESSURE NA NA l
INJECTION L0w PRESSURE NA NA
- INJECTION '
REACTOR NA NA BUILDING SPRAY
- SERVICE WATER NA NA
- REACTOR COOLANT 1 E-1 8 E-2
- EMERGENCY ROO M NA -
NA
- COOLING h . .-. _ _ _ _ - - _ - _ _ _ _ _ _ _
O
+-Worm Crear St
%,e nut o))- Spring Pack Motor -
~
r7 8 $,,,,,,, W _ _ _ _ 7 _ _ r. _ , W- /- (\h(\f\Q _ il J _f gg%
~
Drive 6 ears-+ ggggg _ Tripper Cams 4 W "" . , wainan A\\\\\\\\\\\'N Tong,uc _ Shoulder Locknut Switch Limib Switch O
O IFI. L
- 100 FL. =
MA = 100 t.e l ik. o '
= 50 ft. =
MA = 50 to l l IfI.
.L
= 100 ft. "
with friction utual MA=45teI i O
. . - - - - - - - - - , --- e.--,-,,,y. ,,,.--,,,,,e,,,.y,,--..-,,. a, ,,---%m-#w.,-e,- ,, y# -%-, - m,w,-g -
Stem Spur Pinion intermittent Pinions O .. .. .. ,
, : i i : s i Rot , ii , ii - - -
. ii , i
; ; ; i i
i 'i 'i l
'00hl0h.__-h___-_fy I
M i : i 'i Intermittent hrs
\/ l 6
Itimit Switch Drive From Wormshafb or O
.orivesiecys t
g ( Intermitten:, Intermibbent Pinion
& ear er Rotor l
l : - L turn of Intermitient Gear = l : W burn of Lniermihnt Pinion 1 - ! c= l - Intermibtent & ear
= l l m 1 O =
l Limit switch nrive. eear
l EXPERIENCE IN THE USE OF THE MOVATS SYSTEM Presented to: Advisory Comittee On R'eactor Safeguards Subcomittee On Reliability Assurance (Valves) October 8, 1985-Presented by: 1 Raymond J. Hardwick, Jr. l Robert E. Nicholls l Virginia Power ; Nuclear Operations Department-
' Richmond, Virginia O: .
l
.. . l I
l l O I EXPERIENCE IN THE USE OF THE MOVATS SYSTEM O - PRESENTER'S COMMENTS l l O
Advisory Committee On Reactor Safeguards Subcommittee' Presentation October 8, 1985 EXPERIENCE IN THE USE OF THE MOVATS SYSTEM PRESENTATION OUTLINE
- 1. What attracted Virginia Power to the M0 VATS System
.i 2
- 2. Motor Operated Valve Problems Detected with the Use of the MOVATS System
- 3. Area of Possible Corrective Action and Improvements on MOVs
+
- 4. Consideration of the Use of the M0 VATS System in a Predictive Maintenance Program O
t l l l
~O ADVISORY COMMITTEE ON REACTOR SAFEGUARDS !
SUBCOMMITTEE PRESENTATION t OCTOBER 8, 1985 PRESENTER'S COMMENTS SLIDE 1 VIRGINIA POWER LOGO
~
VIRGINIA POWER-14TH LARGEST INVESTOR OWNED ELECTRIC UTILITY IN TERMS OF ASSETS-4TH LARGEST INVESTOR OWNED SUPPLIER OF NUCLEAR GENERATED ELECTRICITY IN U.S. IN IST QUARTER 1985, VIRGINIA POWER PRODUCED MORE NUCLEAR GENERATED ELECTRICITY THAN ANY OTHER UTILITY IN U.S. i VIRGINIA POWER
- O RATING (MwE) COMMERCIAL OPERATION DATE SURRY 1 775 12/22/72
, SURRY 2 775 05/01/73 NORTH ANNA 1 877 06/06/78 NORTH ANNA 2 890 12/14/80
9 [')
'~'
SLIDE 2 PRESENTER'S BACKGROUND - ROBERT E. NICHOLLS 27 YEARS WITH VIRGINIA POWER ELECTRICAL MAINTENANCE FOSSIL AND NUCLEAR OPERATIONS NUCLEAR OPERATIONS AND MAINTENANCE ELECTRIC EQUIPMENT SPECIALIST. PROGRAM DIRECTOR FOR PREDICTIVE MAINTENANCE SLIDE 3 THE M0 VATS SYSTEM WAS PRESENTED TO VIREINIA POWER IN 1983 THE FIRST SYSTEM OF ITS TYPE WE WERE AWARE OF IN THE INDUSTRY TO ADDRESS THE PERFORMANCE OF MOVs OUR CONTRACTS DEPARTMENT PLACED A SERVICE CONTRACT WITH MOVATS TO ASSIST OUR MAINTENANCE (])
- DEPARTMENT WITH MOV PROBLEMS MOVATS TESTED A SMALL SAMPLE OF VALVES TO PROVIDE FIELD DATA IN SUPPORT OF AN ENGINEERING STUDY OF VALVE PERFORMANCE VIRGINIA POWER HAS IN PLACE TO MAINTAIN OUR MOVs 1, PREVENTIVE AND CORRECTIVE MAINTENANCE PROGRAMS
- 2. INSTRUCTION MANUAL GUIDANCE
- 3. MAINTENANCE PROCEDURES
- 4. VEND 0R TRAINING TO OUR KNOWLEDGE WE WERE THE FIRST -UTILITY TO ALLOW THE' MOVATS SYSTEM TO ASSIST -0N MOTOR OPERATED VALVE MAINTENANCE O
4
O stiDE u - M0v OeeN-Ct0Se-Ct0SE OeeN DOCUMENTATION wAS 1Ne ATTRACTION AS FOUND CONDITIONS COULD BE DOCUMENTATION AFTER- MAINTENANCE CONDTIONS OR "AS LEFT" DOCUMENTATION WAS NOW AVAILABLE FOR A INDEPENDENT REVIEW OUTSIDE THE MAINTENANCE DEPARTMENT BY PERFORMANCE ENGINEERS ELECTRICAL AND MECHANICAL DEGRADATION ~ COULD BE DETECTED BY PERFORMANCE DOCUMENTATION AND REVIEWED FOR CORRECTIVE ACTION . 4 t O O .
.e
O
- SLIDE 5 A SCHEMATIC REPRESENTATION OF TYPICAL THRUST / SWITCH ACTUATION SIGNATURES, FROM THESE AND SIMILAR MOTOR CURRENT SIGNATURES THE FOLLOWIhG PARAMETERS CAN BE DETERMINED:
LOAD TO UNSEAT VALVE (HAMMERBLOW) RUNNING LOAD LOAD AT TOROUE SWITCH TRIP AVAILABLE LOAD (FINAL LOAD MINUS RUNNING LOAD) VALVE CYCLE TIME TIME OF HAMMERBLOW TIME AT WHICH CLOSE TO OPEN BYPASS SWITCH OPENS-OPERATOR INERTIA INDUCED STEM LOAD FINAL LOAD STARTING MOTOR CURRENT-RUNNING MOTOR CURRENT FINAL MOTOR CURRENT WITH THE AB0VE SPECIFIC DATA AND SIGNATURES AVAILABLE, THE FOLLOWING TYPICAL DEGRADATIONS OR COMBINATIONS CAN BE READILY IDENTIFIED IN MOST CASES:- OhERTbR EHNC DEGRADATION OPERATOR GEAR WEAR MOTOR ELECTRICAL DEGRADATION IMPROPERLY. SET BYPASS SWITCHES IMPROPERLY SET TOROUE SWITCHES BACKSEATING CONDITIONS-EXCESSIVE STROKE TIME INADEQUATE LOAD AVAILABLE TO OPERATE VALVE UNDER ACCIDENT OR DESIGN CONDITIONS IMPROPER ALIGNMENT OF OPERATOR-T0 VALVE EXCESSIVE PACKING LOADS THERMAL OVERLOAD TRIP SETPOINTS SUCH AS MOTOR CURRENT, CYCLE TIME, PARAMETERS RUNNING LOADS, HAMMERBLOW PEAKS, AVAILABLE LOAD, INERTIA, ETC. CAN NOW BE COMPARED TO PREVIOUS DATA TO DETERMINE LONG TERM DEGRADING OF 0PERATOR CONDITIONS. O ,
O SLIDE 6 MOTOR OPERATOR VALVE PROBLEMS DETECTED WITH THE USE OF THE M0 VATS SYSTEM CLOSE-TO-0 PEN BYPASS SWITCH REQUIRED ADJUSTMENT PROCEDURE REVIEWED TO ADD MORE DETAILED INSTRUCTIONS ON ADJUSTMENT OF THE LIMIT SWITCH OPERATOR DESIGN' THRUST EXCEEDED OPERATOR THRUST DESIGN AND PROPER TOROUES SWITCH SETPOINTS-WERE. REVIEWED AS PART OF THE CORRECTIVE ACTION, TORQUE SWITCH ADJUSTMENTS WERE CORRECTED AS NECESSARY VALVE WAS INADVERTENTLY BACKSEATING VALVE C0AST ALLOWANCES WERE REVIEWED AND LIMIT SWITCH RESET FOR PROPER OPERATION AT VALVE OPEN POSITION 1 SLIDE 7 MOTOR OPERATOR VALVE PROBLEMS DETECTED WITH THE USE OF O THE M0 VATS SYSTEM TORQUE SWITCH SETTING EXCESSIVE TOROUE SWITCH SETTING INADEQUATE BASED ON VALVE AND OPERATOR MANFACTURERS SUPPLIED SPECIFICATIONS AND M0 VATS AS FOUND DATA AN ENGINEERING REVIEW -PROCESS WAS STARTED, AFTER PROPER ENGINEERING AND SETPOINT SAFETY REVIEW, ADJUSTMENTS-WERE MADE TO ASSURE VENDOR RECOMMENDED , THRUST WERE ATTAINED. TORQUE SWITCH FOUND UNBALANCED THIS CONDITION IS STILL UNDER REVIEW BASED ON EQUAL'TOROUE DELIVERED FROM SETP0 INT POSITIONS FOR BOTH CLOSE AND OPEN OPERATIONS FOR THE-SAME TOROUE SWITCH SETTING, - [ *4 V
O' SLIDE 8 MOTOR OPERATOR VALVE PROBLEMS DETECTED WITH THE USE OF THE M0 VATS SYSTEM APPARENT EXCESSIVE SPRING PACK PRELOAD ENGINEERING AND MAINTENANCE WERE MADE AWARE OF THE IMPORTANCE OF IMPROPER PRELOAD AND THE PROBLEM THAT COULD DEVELOP WITH THE SPRING PACK PRELOAD MISADJUSTED DURING MAINTENANCE. PROPER CORRECTIVE
. ACTION WAS TAKEN POTENTIAL VALVE SEATING CONCERN
-SIGNATURE REVEALED NON-SMOOTH SEATING OF VALVE COMPONENTS. MAINTENANCE WAS PERFORMED AND SEATING PROBLEM CORRECTED HIGH MOTOR CURRENT >150% OF RATED CLUE TO POSSIBLE GEAR PROBLEMS, MOTOR BEARINGS,-
AND POSSIBLE ELECTRICAL DEGRADATIONS. PROBLEM WAS FOUND AND CORRECTED O SLIDE 9- MOTOR OPERATOR VALVE PROBLEMS DETECTED WITH THE USE OF THE M0 VATS SYSTEM SPRING PACK GAP SPRING GAP READJUSTEDDTO PROPER CLEARANCE AFTER VERIFICATION 0F' PROPER PRELOAD SETTING -- THIS PERMITTED SPRING PACK TO BE IN CORRECT RELAXED POSITION HIGH THRUST RUNNING LOAD VALVE PACKING LOAD EXCESSIVE AND/0R BENT STEM PROBLEMS WERE INVESTIGATED AND PROBLEM CORRECTED OPERATOR GEAR WEAR THE SIGNATURE . SHOWED OPERATOR GEAR PROBLEM. CONDITION WAS INVESTIGATED AND NOTED FOR FUTURE REFERENCE I b l l l
("l SLIDE 10 PROBLEM DETECTION
SUMMARY
ALL AS FOUND PROBLEMS WERE CORRECTED IN A TIMELY MANNER POST MAINTENANCE DATA PROVIDED AS LEFT PERFORMANCE CONDITIONS FOR ENGINEERING REVIEW FROM 1983-85 M0 VATS ASSISTED IN THE CORRECTIVE MAINTENANCE EFFORT ON A TOTAL 0F 35 MOVs THE M0 VATS EXPERIENCE PROVIDED ADDED INSITE ON MOTOR OPERATED VALVE MAINTENANCE AND ADDITIONAL DETAILS TO OUR MAINTENANCE PROCEDURES SLIDE 11 AREA 0F POSSIBLE CORRECTIVE ACTION AND IMPROVEMENTS-FOLLOW-UP STUDIES ON' MOV PERFORMANCE CHARACTERISTIC CHANGES THAT MAY OCCUR DURING THE MOVs LIFE CYCLE DETAILED INSTRUCTIONAL GUIDELINE IMPROVEMENTS FROM e- THE OPERATOR AND VALVE MANUFACTURERS TO PRODUCE A MORE DETAILED MAINTENANCE PROCEDURE NOTE: DOCUMENTED SIGNATURE TREND DATA 0F DIFFERENT TYPES AND SIZES OF VALVES WOULD BE VALUABLE IN TRENDING MOV PERFORMANCE CHARACTERISTIC CHANGES AND PROVIDE A DATA BASE TO IDENTIFY THE PROBLEM AREAS
O SLIDE 12 POSSIBLE EFFECTS ON THE LIFE CYCLE OPERATURE-MANUFACTURERS TOLERANCES OPERATURE APPLICATION TO VALVE TYPE AND SIZE TRANSPORTATION-STORAGE INSTALLATION DURING CONSTRUCTION MAINTENANCE PRACTICES PROPER ENGINEERING APPLICATION ENVIRONMENTAL FACTORS FREQUENCY OF USE FACTORS (~') NOTE: BY SIGNATURE ANALYSIS TRENDING AND DOCUMENTED
'- MOV PERFORMANCE, A REVIEW OF PROBLEM AREAS COULD BE IDENTIFIED NO MATTER HOW THE PROBLEM WAS DEVELOPED, SLIDE 13 THE POSSIBILITIES OF FUTURE USE OF THE MOVATS-SYSTEM NUCLEAR OPERATIONS PREDICTIVE MAINTENANCE PROGRAM FOR ROTATING EQUIPMENT PROGRAM . STARTED IN 1984 AT VIRGINI A POWER'S NORTH ANNA POWER STATION' 1 SURRY POWER STATION'S PREDICTIVE PROGRAM WILL 4
START THIS YEAR, 1985 e
k 10~ SLIDEL14 ROTATING EQUIPMENT SIGNATURE ANALYSIS PROGRAM FOR NORTH ANNA AND THE PROGRAM START-UP TRENDING NOW UNDERWAY FOR THE SURRY STATION SLIDE 15 THIS SLIDE SHOWS THE SIGN ON THE NORTH ANNA PREDICTIVE LAB - " PREDICTIVE ANALYSIS LAB" -- TO SET ASSIDE THE PROPER ENVIRONMENT FOR ANALYSIS WORK SLIDE 16 THE DEDICATED PREDICTIVE TEAM CHECKING OUT EQUIPMENT PRIOR TO TAKING. DATA RADIO HEAD PHONES ARE- USED TO FR0 VIDE GOOD COMMUNICATIONS IN HIGH NOISE AREAS 1 SLIDE.17 PRESENTATION OF THE DEDICATED PREDICTIVE TEAM MEMBER , GATHERING DATA
^t\/ ")
- SLIDE 18 DISCRETE FREQUENCY DATA, Oil ANALYSIS AND TEMPERATURE TRENDING IDENTIFIES DEVELOPING PROBLEMS AT A EARLY STAGE THIS SLIDE SHOWS ACCELER0 METERS IN -PLACE AS PART OF THE DATA GATHERING PROCESS
' SLIDE 19 PRESENTATION OF THE LAB.AND TEST COMPONENTS USE IN 1
PREDICTIVE ANALYSIS TAPE-RECORDERS
*~
COMPUTERS REAL TIME ANALY7ERS SLIDE 20 PREDICTIVE TEAM MEMBER PROCESSING TREND DATA O e e g e y-- - .----r.
O SLIDE 21 INP0 GOOD PRACTICE OUR. PREDICTIVE PROGRAM IS 18 MONTHS OLD AND WE ARE PROUD OF OUR INP0 GOOD PRACTICE SLIDE 22 M0 VATS POSSIBILITIES IN THE PREDICTIVE MAINTENANCE PROGRAM DOCUMENTATED SIGNATURE TRENDING POST MAINTENANCE TESTING PROBLEM S0LVING AID DOCUMENT FOR ENGINEERING PERFORMANCE REVIEW IT APPEARS THAT THE MOVATS SYSTEM WOULD BE A VI ABLE TOOL IN TREND ANALYSIS FOR MOTOR OPERATED VALVES SIGNATURE ANALYSIS ON THE DIFFERENT TYPES AND SIZES OF MOTOR OPERATOR VALVES WOULD SEEK OUT O HIDDEN PROBLEM DEVELOPMENT FROM ANY SOURCE SLIDE 23 THE HUMAN FACTOR THE INDUSTRY TRAINING PROGRAMS, PROCEDURES, VENDOR INSTRUCTIONS AND PROPER MOTOR OPERATED VALVE APPLICATION AND OPERATION STILL IS SHADOWED BY THE HUMAN FACTOR. DOCUMENTATED SIGNATURE ANALYSIS TO MEASURE- THE PERFORMANCE OF EQUIPMENT WITH INDEPENDENT ENGINEERING REVIEW WILL AS."IST THE NUCLEAR INDUSTRY IN ACHIEVING EXCELLENCE
2 - 2. a ---
--.-4 .- , -JA- a u I
O
'l i
EXPERIENCE IN THE USE OF THE MOVATS SYSTEM O SLIDE PRESENTATION P O.
-.-~- . . .- --- -
- !O ;
t 5 EXPERIENCE IN THE USE--0F THE MOVATS SYSTEM i-4 .- Presented to: , 1 ,
. Advisory Committee On Reactor Safeguards Subcomittee l _
On Reliability Assurance (Valves) )- October 8, 1985 !~ e t j- Presented by: t Raymond J. Hardwick, Jr. !- Robert E. Nicholls l Virginia Power Nuclear Operations Department ' j Richmond, Virginia-I s 1 10 . 5 ! I e
. . - - - , . ,,.a,,~ - , , - . , , - - , . .er.. , . , - - , . -.,e- -.yr,.,,,.,.,,.v.4..,_.,-_,_,,,,ew,,_,n.m.
,,,,~,..+,.,,_,m-,n,, .c,,.,,,,+ne ,,.,-,v.,_ , _ ,,,
TO SLIDE 1 (VIRGINIA POWER LOG 0) . SLIDE 2 VIRGINIA POWER NUCLEAR 0PERATIONS DEPARTMENT LOGO SLIDE 3 -VIRGINIA POWER INTR 0 DUCTED-T0 MATS.1983 SERVICE CONTRACT WAS PLACED FOR THE SURRY AND
, -NORTH-ANNA UNITS ,
L- -* MOVATS TESTED A SMALL SAMPLE OF VALVES TO PROVIDE FIELD DATA IN SUPPORT OF AN ENGINEERING STUDY 0N , VALVE PERFORMANCE SLIDE.4- -THE MAJOR ATTRACTION OF M0 VATS ~ DOCUMENTATION OF THE MOVs FIELD'PERF0,RMANCE-O a f O .
Slide 5 O l l N THRUST ! SIGNATURE g L m' l backseat f I running load peak load delivered L second hammer blow peak, when stem picka up disc first hammerblow, when operator picks up stem I g lost r. notion action, zero load on spring pack 8
- start of valve cycle when switch is tumed to open I
i torque switch trip t U^ I limit, bypass and torque switches are all closed a bypass switch opens i a g cycle, torque or limit switch opens
\\ \ \ \ \\ \\ \ \\ \\ \ \ \
e-
- SWITCH SIGNATURE TIME cuitusscoNDS) &
l l l O
O' SLIDE 6 MOTOR OPERATED VALVE PROBLEMS DETECTED WITH THE USE OF THE M0 VATS SYSTEM CLOSE-T0-0 PEN B'l PASS SWITCH REQUIRED ADJUSTMENT OPERATOR DESIGN THRUST EXCEEDED VALVE WAS INADVERTENTLY BACK-SEATING SLIDE 7 MOTOR OPERATED VALVE PROBLEMS DETECTED WITH THE USE OF
-THE.M0 VATS SYSTEM TOROUE SWITCH SETTING EXCESSIVE
- TORQUE SWITCH SETTING INADEQUATE TORQUE SWITCH FOUNP UNBALANCED SLIDE 8 MOTOR-OPERATOR VALVE PROBLEMS DETECTED WITH THE USE OF THE MOVATS SYSTEM
()
- APPARENT. EXCESSIVE SPRING PACK PRELOAD POTENTIAL VALVE SEATING CONCERN HIGH MOTOR CURRENT >150% OF RATED SLIDE 9 MOTOR OPERATOR VALVE PROBLEMS DETECTED WITH THE USE OF THE M0 VATS SYSTEM SPRING PACK GAP HIGH THRUST RUNNING LOAD OPERATOR GEAR WEAR.
5
-O SLIDE 10 ' PROBLEM DETECTION
SUMMARY
ALL AS FOUND PROBLEMS WERE CORRECTED IN A TIMELY MANNER POST MAINTENANCE DATA PROVIDED AS LEFT PERFORMANCE CONDITIONS FOR ENGINEERING REVIEW FROM 1983-85 M0 VATS ASSISTED THE CORRECTIVE MAINTENANCE EFFORT ON A TOTAL OF 35 MOVs THE M0 VATS EXPERIENCE P.ROVIDED ADDED INSITE ON MOTOR OPERATED VALVE MAINTENANCE AND ADDITIONAL
~ DETAILS TO OUR MAINTENANCE PROCEDURES +
SLIDE 11 AREA 0F POSSIBLE CORRECTIVE ACTION AND IMPROVEMENTS
' FOLLOW-UP STUDIES ON MOV PERFORMANCE
-CHARACTERISTIC CHANGES THAT MAY OCCUR DURING THE MOVs LIFE CYCLE A DETAILED INSTRUCTIONAL GUIDELINE IMPROVEMENTS-FROM 4
kJ THE OPERATOR AND VALVE-MANUFACTURER TO PRODUCE A MORE DETAILED MAINTENANCE PROCEDURE SLIDE 12 POSSIBLE EFFECTS ON THE LIFE CYCLE OPERATOR MANUFACTURER TOLERANCES OPERATOR APPLICATION T0. VALVE TYPE AND SIZE TRANSPORTATION STORAGE INSTALLATION DURING CONSTRUCTION MAINTENANCE PRACTICES
-PROPER ENGINEERING APPLICATION ENVIRIONMENTALFACTORS FREQUENCY OF USE FACTORS ;
i
O SLIDE 13 THE POSSIBILITIES OF FUTURE USE OF THE MOVATS SYSTEM
*~
NUCLEAR OPERATIONS PREDICTIVE MAINTENANCE PROGRAM FOR ROTATING EQUIPMENT SLIDE 14 ROTATING EQUIPMENT SIGNATURE ANALYSIS TREND OF OVER 2000 DATA POINTS EACH MONTH INVOLVING APPR0XIMATELY 400 ROTATING EQUIPMENT COMPONENTS EACH MONTH ' RESEARCH AND DEVELOPMENT TO IMPROVE OUR ABILITY TO IDENTIFY EQUIPMENT COMPONENT PROBLEMS USING SIGNATURE TREND DATA 1 SLIDE 15 PICTURE OF SIGN TO IDENTIFY ROTATING EQUIPMENT LAB ([) SLIDE 16 PREDICTIVE TEAM MEMBERS CHECKING OUT EQUIPMENT SLIDE 17 PREDICTIVE TEAM GATHERING DATA 4 SLIDE 18 SHOWS METHOD OF GATHERING DATA SLIDE 19 SHOWS PREDICTIVE LAB AND EQUIPMENT SLIDE 20 SHOWS PREDICTIVE TEAM GATHERING PROCESSING TREND DATA SLIDE 21 INP0 GOOD PRACTICE - 1985 PREDICTIVE ROTATING EQUIPMENT PROGRAM " PREP" ($)
(]) SLIDE 22 M0 VATS POSSIBILITIES IN THE PREDICTIVE MAINTENANCE PROGRAM DOCUMENTED SIGNATURE TRENDING , POST MAINTENANCE TESTING l PROBLEM SOLVING AID DOCUMENT FOR ENGINEERING PERFORMANCE REVIEW SLIDE 23 THE HUMAN FACTOR - THE INDUSTRY TRAINING PROGRAMS, PROCEDURES. VENDOR INSTRUCTIONS AND PROPER MOTOR OPERATED VALVE APPLICATION AND OPERATION STILL IS SHADOWED BY THE HUMAN FACTOR. DOCUMENTATED SIGNATURE ANALYSIS TO MEASURE THE PERFORMANCE 0F EQUIPMENT WITH INDEPENDENT ENGINEERING REVIEW WILL ASSIST THE NUCLEAR INDUSTRY IN ACHIEVING EXCELLENCE , () i I +
,, .- ,-n -n~ -w-
. . . . - - . - , . . - . . . . . . ~ ~ . . . . . . . . . . . - . . . - . . . - . - . . - - . .- . - . .
- 1. .s . ,
i
- g ..
, j
- h. l I
.\
i i - o^ i- NUCLEAR-PLANT-AGING RESEARCH (NPAR) PROGRAM ;
~
- MOTOR-0PERATED VALVE RESEARCH :
i j i i
- i. , t j'
i ! D. M. EISSENBERG l !. 0AK RIDGE NATIONAL LABORATORY
- i lJO.
4 i l d 6 t , 1
- OCTOBER 8, 1985 ACRS !
i - WASHINGTON. DC i 4 L i e b' a-t' i .t l Lo . .L D k
THE ORNL/NPAR. PROGRAM HAS BEEN STUDYING THE MOTOR-0PERATED VALVE FOR ABOUT TWO YEARS. OBJECTIVE
- DEMONSTRATE TECHNOLOGY FOR MAINTAINING OPERATIONAL READINESS OF MOTOR-OPERATED VALVES IMPORTANT TO SAFETY APPROACH f
CHARACTERIZE HOW AND WHY MOVS FAIL IN NUCLEAR PLANT-O SERVICE
~*
IDENTIFY AND EVALUATE METHODS FOR DETECTING AND TRENDING h TIME-DEPENDENT DEGRADATION (AGING) AND OTHER 4 _ARdORMALITIES THAT CAN LEAD TO MOV FAILURES DEVELOF THE TECHNICAL BASIS _FOR RECOMMENDING WHEN AND ,. WHAT _ MAINTENANCE ACTIONS SHOULD BE - .TAKEN BASED ON l DIAGNOSTICS h t LO 0
t
.O SOME DEFINITIONS AND EXAMPLES
' FAILURE MODE: HOW IT FAILED (FAILED TO CLOSE) :
FAILURE CAUSE: WHY IT. FAILED. THE ABNORMALITY THAT WAS THE PROXIMATE ~
'CAUSE .0F- FAILURE (BENT STEM, TORQUE ' SWITCH TRIP, TORQUE SWITCH TRIP'DURING HAMMERBLOW) ;
TYPES OF: ABNORMALITIES ~: TIME DEPENDENT DEGRADATIONS (AGING)-DUE TO'THE CUMULATIVE EFFECTS OF STRESSORS..(BENT STEM) . ABNORMALITIES WHICH LEAD' TO DEGRADATIN (HIGH TOROUE-SWITCH SETTING, LACK.0F STEM LUBRICANT) ABNORMALITIES WHICH ~ LEAD TO FAILURE UNDER SOME~
-ANTICIPATED' .0PERATING CONDITIONS '(INCORRECT BYPASS SETTING, LOW TORQUE SWITCH SETTING)
O
. - - . - - - . - - .. . _ . .- . . -, -. . _ . . - . . - . ~ - - . .. -
I O F : i THISiREVIEW WILL COVER THREE ACTIVITIES
- a .
2
- AGING' ASSESSMENT.OF.' MOTOR-0PERATED VALVES j :(NUREGICR-4234 VI) 1 e
f ! i
< = IN-SITU TESTING 0F MOVS USING M0 VATS i
'(NUREG/CR-4380 - TO BE PUBLISHED) .
t l CONTROLLED TESTS OF MOVS TO EVALUATE DIAGNOSTIC METHODS !g 1
- i i
i ). . k' I l i-I 4 +; , ! 'I l f'. l i t '. t u
,.,.---,,4- ,y , . ~ , , 4 e a w# . ~ . - , + - , - , ,- ,,a,c-. - - , ,,, .'we
- . - -.,-y,.,w.-. e , wh,ww%# ,v,~e-e---,y,y+-y-w,..m-
S O. LA' PHASE-I AGING ASSESSMENT'0F MOVS WAS PUBLISHED RECENTLY* A FORMAL-REVIEW T0 CHARACTERIZE MOV FAILURES MOV DESIGN / SUBASSEMBLIES s MATERIALS OF CONSTRUCTION STRESSORS.IN NUCLEAR PLANT SERVICE TECHNICAL SPECIFICATION SURVEILLANCE REQUIREMENTS 4 MOV FAILURE EXPERIENCE IN NUCLEAR PLANTS 4 RECOMMENDED MAINTENANCE PRACTICES i
*~ IDENTIFICATION AND CLASSIFICATION OF AGING-RELATED
, FAILURES MEASURABLE. PARAMETER IDENTIFICATION
*NUREGICR-4234. VOLUME 1. AGING AND SERVICE WEAR OF ELECTRIC MOTOR-OPERATED VALVES USED IN ENGINEERED SAFETY FEAT.;RE SYSTEMS OFNUCLEAR POWER-PLANTS..
0 !O - _. _ = - - - - _ - - _ - _ _ _ _ _ _ - - _ _ _ _ - _ _ _ _ - _ _ _
.c '
g .-
' ~
TOL SOME RESULTS OF THE. PHASE I ASSESSMENT
--MOV:FAIEURE-EXPERIENCE-INNUCLEARPLANTS MAINTENANCE FREQUENCIES OF MOVS ARE HIGHER THAN FOR OTHER .
ACTUATOR-TYPES (PG 33)-
- THE PREDOMINANT CAUSES OF FAILURES ARE TOROUE OR LIMIT-SWITCH MISADJUSTMENTS OR FAILURES, AND. MOTOR FAILURES
-(PG 27) ;
LO
- NO REPORTED CASES OF OPERATOR FAILURE DUE. T0 (NORMAL)
AGING WERE NOTED 4 MAINTENANCE PRACTICES TENDED TO BE- AIMED AT RESTORING OPERABILITY. RATHER THAN IDENTIFYING- AND CORRECTING THE- ! DEGRADATION OR ABNORMALITY V
$I v- ; ...---..,_-,,..-..-...-,,,_..-.-_2.._,,,.-,u..,_-,_....--_.',,,.;-.~,,..~...,,_..,_.u.....-,,..._.__.._a.,_,
._. A. . m ,.w. m J 4 O h
.SOME RESULTS OF THE PHASE I ASSESSMENT
(]' ). RANKING OF FAILURE CAUSES 1AS IDENTIFIED BY MOTOR-0PERATED MANUFACTURER (TABLE 6.5) SUBCOMPONENT RANKING FAILURE CAUSE , GEARB0X ASSEMBLY 1 STEM NUT WEAR, GALLING, BREAKAGE 2 VALVE OBTURATOR/ SEAT SEIZURE DUE TO INERTIA 3 CLUTCH MECHANISM WEAR,- BREAKAGE 4 LUBRICANT DEGRADATION, HARDENING ELECTRIC MOTOR ASSEMBLY 1 MOTOR FAILURE (E.G., STATOR BURNOUT) ,, 2 TOROUE-SWITCH RECLOSING DUE TO SPRING-BACK OF OPERATOR
. () 3 INSULATION (ELECTRICAL)
BREAKDOWN SWITCH ASSEMBLY 1 FASTENER LOOSENING, BREAKAGE TOROUE SWITCH 2 CAM WEAR, FRACTURE 3 -CONTACT CORROSION LIMIT SWITCH i FASTENER LOOSENING, BREAKAGE 2 NONMESHING GEARS' 3 LACK OF GREASE / GREASE HARDENING IN GEARED LIMIT SWITCH. 4 CONTACT CORROSION () - l
O SOME RESULTS OF THE PHASE I ASSESSMENT PARAMETERS USEFUL FOR TRENDING MOTOR OPERATOR DEGRADATION HAVE BEEN IDENTIFIED DURING ACTUATIONS
- STROKE TIME MOTOR RUNNING CURRENT J
SURGE CURRENT AND DELAY TIME SWITCH' ACTUATION TIMES - WORM SHAFT TRANSLATION OUTPUT THRUST VIBRATION MOTOR TEMPERATURE
- STEM VELOCITY, POSITION ACOUSTIC EMISSION INSPECTIONS
. DIMENSIONS e ALIGNMENT e CRACKING e APPEARANCE e PLAY e SPRING PACK ELASTICITY e LUBRICANT COMPOSITION
- ELECTRICAL RESISTANCES
Q , IN-SITU TESTING OF MOVS IN NUCLEAR POWER PLANTS USING MOVATS SIGNATURE ANALYSIS
- OBJECTIVE EVALUATE CAPABILITY OF MOVATS TO PROVIDE USEFUL
- DIAGNOSTIC INFORMATION FOR- DETERMINING OPERATIONAL READINESS OF MOVS BEYOND THAT CURRENTLY PROVIDED BY ASME l- SECTION XI 1
lQ i
- IDENTIFY AND CHARACTERIZE TYPES OF (DEGRADATIONS, INCORRECT ADJUSTMENTS, ETC.) FOUND IN SOME ABNORMALITIES i SELECTED OPERATIONAL-MOVS J
i I ( O
e M0 VATS INCORPORATED WAS GIVEN A SUBCONTRACT TO PROVIDE TEST DATA q AND ANALYSIS ACCORDING-TO THIS. STATEMENT OF WORK !
. 0BTAIN THE COOPERATION OF ONE OR MORE UTILITIES
~
- * . DEVELOP AND SUBMIT A TEST PROCEDURE BASED ON M0 VATS TECHNOLOGY
- USING THE TEST PROCEDURE OBTAIN VALVE SIGNATURES FOR A i
TOTAL OF_40 MOVS CONSIDERED OPERATIONAL BY THE UTILITIES ([) -- ANALYZE. DATA T0 CBTAIN DIAGNOSIS OF CEFECTS FOUND, IDENTIFY REMEDIAL ACTIONS TO BE RECOMMENDED TEST RESULTS WERE ULTIMATELY OBTAINED FOR A TOTAL OF 3G VALVES AT , FOUR NUCLEAR PLANTS 4 0
-_ . . . . . . . = - . --
4 T ~O i THE REQUIREMENTS OF THE ASME BOILER AND PRESSURE VESSEL CODE - ARTICLE IWV-3000 SECTION XI DIVISION 1 WERE IDENTIFIED k 4 4 4 4 VALVE EXERCISING TEST TO OBSERVE DISK MOVEMENT .
- EXERCISING FULL STROKE TO MEASURE STROKE TIME
!O * ! LEAK RATE ! POSITION INDICATOR VERIFICATION i k i 1 . i O
O DEFINING VALVE OPERABILITY VS OPERATIONAL READINESS IS NECESSARY TO RESPOND TO NRR CONCERNS OPERABLE: DEMONSTRATED BY TEST TO HAVE MET PERFORMANCE REQUIREMENTS UNDER SPECIFIED CONDITIONS (E.G., BY ASME TEST) Q OPERATIONALLY READY: ABILITY (IN ITS PRESENT CONDITION) TO MEET PERFORMANCE REQUIREMENTS IN THE FUTURE UNDER ALL ANTICIPATED
, OPERATING CONDITIONS
-OPERATIONAL READINESS CAN BE INFERRED FROM OPERABILITY TESTS TO SOME EXTENT l
OPERATIONAL READINESS CAN BE PREDICTED MORE DIRECTLY FROM MONITORING AND TRENDING ABNORMALITIES LEADING TO FAILURE 0
.O
- 6 A REVIEW OF A TYPICAL (LIMITOROUE) MOTOR OPERATOR WILL ILLUSTRATE WHAT MOVATS DOES l O ..
B,1 g e O o t y l . w h .
,' N R o -s o -
4- s 9x e mu. 1 1
/
1 vs h \'
' ~'
q
$ - < ,7 y g
<9 ; / ;
n '4k A l' 6...e g ) = s:
&8j fB k -
1.3E" co ,
' ~
I E' __ m-ga- k. 06 408-0001 4
i i. i' LO 1 i i .r
~
- MOVATS IS A SIGNATURE ANALYSIS SYSTEM UTILIZING THREE' SENSORS
4 i: I i [ . WORM POSITION i
- j. -
- i - . SWITCH' ACTUATION TIMES i
[
. -MOTOR CURRENT lO 1
l - i i I i I l I i r O i l t
WORM POSITION IS MEASURED BY MECHANICALLY LINKING TliE WORM TO AN O LVDT SYSTEM (THRUST MEASURING DEVICE)
/ / / /////
-h-
?= ;!
Ie , b l .... ,
; 2
=
s .
, , u.
P
, . y y -
, .s (
- ( -s
- i . ;~.. ,
0 '" lm-s- ,- -ss ( s . d 5 ' -m - j s' tm e < l m/ THE INSTALLATION AND REMOVAL OF THE TMD IS DONE AT SHUTDOWN AN ONE VALVE AT A TIME O
DIAGRAM OF A TYPICAL.MOV STEM THRUdT AND SWITCH SIGNATURE O ""'"^'S"' """ ' '" A THRUST SIGNATURE A t ^ , , t p valve hits backseat r,n
@ running load peak load delivered '
I
- 6 L. second hammer blow peak, when stem picks up disc first hammerblow, when operator picks up stem lost motion action, zero load on spring pack '
, - start of valve cycle when switch is turned to open
[ p torque switch trip . limit, bypass and torque switches are all closed A bypass switch opens or limit switch opens ( l
$ z N \ \\\\ \ \\ \\\ \ \ \
7 SWITCH SIGNATURE TIME (MILUSEOONOS) t
. r THE MOVATS SYSTEM CAN ANSWER YOUR QUESTIONS CONCERNING YOUR MOTOR OPERATED VALVES FOR ADDITIONAL INFORMATION, CALL OR WRITE: h 5
x
. M, -
N O MOVATS, INCORPORATED < 2995 JOHNSON FERRY ROAD MARIETTA, GEORGIA 30062 PHONE (404) 998-3550 .
,n G O TEST IDENTIFICATION: hE N!~l V STORAGE LOCATI0tJ : 01-13 PLOT NO. 2 ~
TMD L SNITCHES: CLOSE TO OPEN STROKE TIME : 27.67 ' SECONDS AVE Rt.N THRUST : <PRELOAD MAX RLN THRUST : <PRELOAD BYPASS TIME : 3.36 SECONDS UNSEATING THRUST' : 39763 N LNSEATING TIME : 1.3 SECCT4DS THRUST AT CST : N/A PROTECTION MARGIN : 2.06 SECONDS TOROUE SN. SETTING: 3.0/2.5 V/O 5.000e-1 Vy O.e-7 i T/0 5.00000 TL -4.*109a1 V/O 2.50000' ,
"'"'3'lilllililllililllilllililillilli' Vy O.e-6 SWITCH T/0 5.00000 -
SIGNATURE TL -4.110o1 Vy - . j .l . . . . , num THRUST , SICNATURE
-_j !
l
. . . . . . . . l Tt. vx . .- ~
A
o
' O C\
TEST IDENTIFICATION: VALVE M11 STORAGE LOCATION : 01-15 - PLO'T NO. 4 MOTOR CURRENT 8e SHITCHES: CLOSE TO OPEN STROKE TIME : 27.71 SECONDS PEAK INRUSH : 190.3 AMPERES AVE. RUN CURRENT : 14.3 AMPERES MAX. RUN CURRENT : 14.3 AMPERES END CURRENT : 13.8 AMPERES
.q V/O 5.0000-1 Vy 0, e-7 T/0 5.00000 TL -4.11001 V/O 5.000e-1 ,
-
- I Vy 0.e-7 SWITC11 _
T/0 5.00000 TL -4.109e1 i , ,_ MR I IRRENT - CNATURE w ! i - l I TL I Yx
~
C-
- MOVATS AS A DIAGNOSTIC FOR DETECTING AND TRENDING DEGRADATIONS HAS THESE LIMITATIONS SOME DEGRADATIONS CANNOT BE DETECTED BY MOVATS SIGNATURES SIGNATURES AT AMBIENT CONDITIONS MAY NOT PROVIDE INFORMATION ABOUT PERFORMANCE UNDER OPERATING CONDITIONS O -
- PERIODIC TESTING AT SHUTDOWNS MAY NOT BE SUFFICIENTLY FREQUENT MANUAL INSTALLATION / REMOVAL CAN LEAD TO EXCESSIVE RADIATION EXPOSURE O
1
?
i LO
-CONCLUSIONS ABOUT HOVATS CAPABILITIES :
1 I
- MOVATS CAN PROVIDE INFORMATION REGARDING OPERATIONAL.
READINESS BEYOND THAT OBTAINED BY SECTION XI TESTS BY ! IDENTIFYING AND QUANTIFYING: l t O - INCORRECT ADJUSTMENTS WHICH CAN CAUSE FAILURE f UNDER SOME ANTICIPATED OPERATING CONDITIONS: I.E., INCORRECT SETTING OF LIMIT SWITCH {
- DEGRADATIONS WHICH CAN LEAD TO FAILURE IF' ALLOWED TO CONTINUE: I.E., BENT STEM- i l- , ;.
= .
LO : l i t i
. i O
SUMMARY
OF ABNORMALITIES CLASSIFIED BY TYPE i TIME DEPENDENT INCORRECT ADJUSTMENTS AND DEGRADATIONS OTHER ABNORMALITIES BENT STEM EXCESSIVE INERTI i GEAR WEAR INADEQUATE STEM LUBRICATION ([) MOTOR PINION BINDING IMPROPER SEATING STEM WEAR VALVE BACKSEATING GREASE HARDENING INCORRECT TOR 00E SWITCH CALIBRATION MOTOR DEGRADATION UNBALANCED TOR 00E SWITCH
- EXCESSIVE SPRING PACK GAP
- EXCESSIVE PACKING TIGHTNESS
- IMPROPERLY SET BYPASS SWITCH
- LOOSE STEM NUT LOCKNUT
- I:
' ABNORMALITIES WHICH CAN CAUSE VALVE FAILURE UNDER SOME ANTICIPATED OPERATING CONDITIONS.
O
O ABNORMALITIES DETECTED BY MOVATS WERE CLASSIFIED BY TYPE
. TIME DEPENDENT DEGRADATIONS (PROXIMATE FAILURE CAUSES)
. INCORRECT ADJUSTMENTS OR OTHER ABNORMALITIES
- LEADING TO TIME DEPENDENT DEGRADATIONS (ROOT CAUSES)
- LEADING TO FAILURE UNDER SOME ANTICIPATED OPERATING CONDITIONS (PR0XIMATE FAILURE CAUSES)
O
-- i a
1 1 t i ! STATISTICAL RESULTS OF MOVATS TESTS :
, t
- 1. IMPROPERLY SET BYPASS SWITCH 751 9. STEM WEAR 81 j 2. INCORRECT TORQUE SWITCH 10. GREASE HARDENING 8 ;
CALIBRATION 50 11. GEAR WEAR 6 ,
- 3. UNBALANCED TORQUE SWITCH 33 12. MOTOR DEGRADATION 3
- 9. EXCESSIVE SPRING PACK GAP 17 13. MISCELLANEOUS !
.l j O 5. EXCESSIVE PACKING TIGHTNESS 8 ABNORMALITIES 35 l 6. EXCESSIVE INERTIA i j 7. LOOSE-STEM-NUT LOCKNUT 4 8. VALVE BACKSEATING 1 i i I t i t 7 k i f, f i i O i i ! __ 1
O: CONTROLLED TESTING OF MOVS TESTS ARE BEING CONDUCTED TO IDENTIFY A. SET OF USEFUL DIAGNOSTIC , TECHNIQUES FOR DETECTING ABNORMALITIES AND TRENDING DEGRADATION >
- AMBIENT BASELINE TESTS (ORNL) !
AMBIENT ACCELERATED AGING (ORNL) F t AMBIENT IMPLANTED DEFECTS (ORNL) ' O OPERATING CONDITIONS / AMBIENT COMPARISON TESTS (ETEC) FIELD TESTS (UTILITIES) :
. l 0 l 9
'\; . 1
. t
't l
- -TESTING AT AMBIENT CONDITIONS IS,IN PROGRESS
- STROKE' TIME l MOTOR-RUNNING CURRENT'
! I SURGE' CURRENT / DELAY TIME I 4 l; .* SWITCH ACTUATION TIMES [ c . i i WORM SHAFT TRANSLATION !O c i [
- STEM THRUST l r i
STEM VELOCITY !; STEM POSITION ! l-MOTOR GEARBOX VALVE VIBRATIONS i : MOTOR TEMPERATURES i li TESTS CURRENTLY INVOLVE A LIMITOROVE SMA-2 OPERATING. AT 4-
- IN. GLOBE VALVE.
1 < l [
,. .. . . _ - . . . . .. - -. ... -. - - .. - -. - . - . ~_
1 L 1 j t INITIAL TESTS -FOCUSSED ON ESTABLISHING THE CAPABILITY OF THE >
- INSTRUMENTATION TO MEASURE.THE PARAMETERS OF INTEREST :
i I e-f UNDERSTAND NORMAL SIGNAL CHARACTERISTICS i i I i
- DETERMINE REPRODUCIBILITY !
l DETERMINE RESPONSES TO KNOWN IMPOSED CHANGES 3 [O I , 4 I [ :
- i. i f I 5
i j i k IO 4 l 1 2 e 8
Morus. is ween Ge4A cues woen 64*A Los s usar caiva, s aus.44, pe o seamaars en. = oeivc stud wss( oama sye) c Awa
\
scu 4 toes crea usr wang Sns Sww is ge: eye 3 w _. . weirsw erc4 .
'f uno is " " M8f d*tuA 25 -
to[upweg y mams MOTOR CURRENT
- : v :
e : : ', : - i
,l -
8 .____;_ __ . .;. . _ _- i : ' g,1-- : , : STEM E OCI W
,g
.E. .
8
-;_{_-. ""} - 4_ _ _! __- I t.GE+85 - " *
- :oevv44 ton, t."ts E4su.y : NEL* STEM STRAIN
)
; oP* Seat-(**o 1 ^r-9 .
3 ( i
.P: : : -
t : : n : m$ L. r.l___r _ __! r ez.,:_ _ _ _ _ _ _ _ _ 1 _ _ _ _ _ ,_ _ _ _ _:_ _ _ _ _ _ _ _ _ _ _ O - s.1--
! i
- mu=r I
.!trarT.Tcs0/c
- camas ed .
l : : i : : 2-1 , MTOR ACCELEROMETER
-e.[
jtfy ; w g y. p g. M g,
],
I
- n, ,.
4 g : .dm1 ss s=. na= - i
.a i i : -
8 1 2 3 O', . TIME (s) 4
, MOY1198.0AT
, Movies, RUNE 18 (C-TO-Cl,S/24/85 , Q$ $ lN e l - L
i l l l l UMar SWifet OGTv44ter /*MKES TbRoot switch =- CowTACT Wita sE Al* ActvKT65 Amp morog
- LoM AoTod. SI'vTS opp ,
Tus e .
..._.u ______- ____ --------- g
- u sTu E 0c!TY
)
. ........ . .. ..i., .. . . .
- , ,,,, e y y ,,, .
..i m.i.. i ... . .........L . . . ...... ;, ; .
/ '
-g.1_ ,
g, JOEM -
- i i i ML. STM STMEN .d
- : '. i
. 5 e . . .
a : : : S..____. .__._ . _ _ _ _ - . . 1.5 f l TestE mtTO4 0/C 1 { .
. . i r,
-1.5 opp uent P w s oee :
3.1 . LlutT SWT04 O/C 4 .
~B.1 :
5'F : N EVE R 0 8 8 EB8T i V. . i
- g. _
2-- / j MITWI ACEELMOPETH
. a_LhL_n,nLu,hi..nl L 'i Nal Ma : -
8
' "F1 T"l
' 7 in '3 ' '
nr .
- 'PI T. V .
U. '
. e
-2. :
+h: .
- . : : tasur m. mots 5.. . .
(_
... + .
w . . g : 28 ~I~ I I I 88- 99 98 91 92 . N0Y1184.DAT MOYles,MUNe18 (0-T0-C),S/24/85 TMeus, swsftg e 2 Mena4 e too pr4
e. _ ,r x s O. 4 O i j r
,3 . -
W
. E. .
I
,% -#o s2 -
" ~~ ~'~ ~~~~
. . - _ _ _ . . _ _ e ,
1 '
>n ' *t
~
s
~
4,
\
f 0 . J M
._. .e A-------- n _ _ _ _ . . . __
-x 32 -1 m -
W a ;.,
- wr8 o* -
S .,
. 4.-
-T
- a s
i . . . . ~Q I. 1 g v
$ w ' - -
i 3 5 w
- t ,t t _ .g 1 F
~
- , g 3 i
$$g ,
u % - tE5 - _ .g a i .
\ . . . - . .-
i i i _ ,, . l
,i .. .. .
f . . - . . . . . r 1, . . . - . .' . . . . - . . ,
- . ~. - ---- -
- ,o . ?
! , g li1eieiiiiiiiieii tiitiieiii11iiniii !
" 5 # 3* 3*=5 1
- r. i t t *s "s E. q ~Si 33,8 3 3 ( I. .-
W b I
l
\
1
-1 l
1 I I I 1 t s .-. Io -hCH Glo8E. VALVE [LT.m1Tonaus 5mA-2 moTon, orsaAvost, (mov-loo) .
.,_ L _ T 1 5_._., . - . ., ei e +
l ._ ./q t {.!.1,., , 3, ago,e - g , i t9 L s=== ", ' . .. n.. ' . _ . _J' 5 see se t.u _ -4 . # 4 .- . s , . -
, ,a ,4 h -
' ' ' F.Pwl
,, . s...
, L _. -
w - { . '. ...1-,_-,* i i f o
- I . ... ....'
28mme l,. i .- .._.i 9 .._ ! l
. . .. ...-. l..X --.. ]
anos.
-s - <
' 1 i t
_N . 4 e . . ,_ 3, , . 3 a looto
.~ T us 1
W. . VI _ _w 150isks . g . N ioo Fr% :
- es amm.
y , hkh
~
t
-
- O - " ' O # Le oh4 2-u g ,
- ^ _ _ ._. -
^"
- vyF ""'"* Se # 64
% Q^ <
- . . . . . - . . . . . .. ^:.::::; -
ss : .
-5664 k . . .... H. 4 %
-viruvery ,
=l4444 -t a a a a a a a i2 m a 1 i i i i ia a n i ia a O 10 2b .h go 50 h lo h 1e leo o m h > w m 6e m se 1e %
TIME (sec) Timt (Sec) opt.M ---+ CLOSE. C LOS E. : OPE.N i b1
L -- - s - % 2. _
, l
/
o k x a s.4 b - Ei-o ,[' Y 8 i _ w ;6 $
- t. .
~
~
- s . _ _ _ __ ___
1 g ; -w is B s 1 - 1na-I
.a s, ). .-
Y ln. 5 i e 3 O 1i a g - 1 ,5 5 3 . v -
~
E{ m 1 ~id > l k 4,
)
~
no ? W a '
- Cl Na 1
J....... ................ .. . . . . . . . _ o i t i t b i
.= 8 to F M e4 0 (w]wswnavat.w
- i
~
i
O % l
)
l l
/
/
Toeeus $wne.u Tit w . m . m . Seves Osa . I 3a - . g:-
#N Lam l ~ PL Du % o.,e 5 Stones, 1 2.5 -
l [ p , l j l - ; I qu - M ou t e Swee ' g Ras b h Assewdl
- a. R. veri (4. L c.nw.7) D e T.
o i" o -"' l cJ
.1 .
l . r" I I
- 1 l t l I
o.5 - ! P1.+.e EnscolasJ #4 Fvil Oped MIW. l O i i Z i i i e i e i i i , I,. ! J 1 5 6 7 3 d g p 33, g g g g n gg g Tbeeps sulTc4 SoffFried [W 4 Mioo Tows Giel Renmed
- w. &m CuaAssy r.a. opes T.
~
j Ct.se Staest r swnto @ 1.15 b.4e e 100 R4 l l i 25-ts _ - __ /
O O O . 40405T 5, I'lf5 w AGAY 30.1145 AFTER PAOPE4. ST8M Lu3anc4riod Cou0lTlod #E.5To4ED samma r 16est - A0 Gust 5,1475 y - AFTER ~ lb$ C1CLES y ,; g [5 TEM LUDAlCATiod DEGAADED) l 3mAB8 1 m - w l A M858
- I E aseen -
l l 4 bM g Itho l
< h - l 8 Plte -
l . l T isen. - I_ I leece -
. 1 AT 4 Te A&vt (wlTCil SETTidG cP l.6 ggg _. l (1.15 15 MW) , beg R A D&D STSA l LuBAsc4Tiod CA4 4Esotr 14 4 AEnucTiod M
- l lal TR&e5rieTTse ETEA TH4UST oe 6900 ILF
% ,. l W4tCM 45 17 'le er THE T MosT TRANset ettEb I USl46 MerEA- STRA LUSaiCAT1ord 3*** l e I I I I ' I I I - I i ' ' I I ' I I 8 o s a s 4 5 6 1 # 1 io si is is ig is #6 17 it n' 2o TbRica. SW tTCM 4eTArled (degrees)
AtoploO STEM toAbidi V5. TD4005. SWTTCH SWAT AeTATwyd AT* oieFataar STE.st 1,us444rkwd ceanstleds HoM i-st,-ts
** , l I
DET/RES PROGRAMS RELATED TO VALVE OPERABILITY AND RELIABILITY I l MSEB EEICB I AGI pH : TION VALVE TESTS VALV P R BILITY o QUAh0FJA{ ION (Illt (ETEC) ISOLATION . I CIV l fE CHECK VALVES S0V MOV PORV/BV l
'Pg\LVE VENT I !'
'-1 r-kb I
l bhkY_hES - mVATS u _ _ _ _ _ _1 O O O
-. p#
DAVIS-BESSE NTOR OPERATED VALVE FAILUES I. VALVE FUNCTIONS A. AF-599, AF-608 - AUXILIARY FEEDWATER ISOLATION VALVES B. MS'-106 - MAIN STEAM SYSTEM SUPPLY TO AUXILIARY FEEDWATER TURBIE II. UTILITY INVESTIGATION A. ROOT CAUSES
- 1. SWITCES SET INCORRECTLY 2.- LACK OF DP TESTING FOR AF-599 AND AF-608 O 3. UNKNOWN PHENDE NA CAUSED INCREASED OPEN TORQUE EQUIREENT FOR E-106 B. CORRECTIVE ACTIONS
- 1. SET BYPASS LIMIT SWITCES AT 20% OF FULL DISK TRAVEL USING N VATS
- 2. INCREASE OPEN TOROUE SWITCH SETTINGS TO MAXIPtN WHICH PECLUDES VALVE DAMAGE III. STAFF FINDINGS A. TOREY PINES STUDY SHOULD NOT BE USED ALOE B. DETERMIE ALL CONDITIONS THAT VALVES NST OKRATE UNDER C. .USE CO EIRMATORY TESTS AT AVAILABLE PLANT CO EITIONS O
i
(vY i TURBIE BYPASS VALVE SP13A2 (ACTION ITEM 9A/9B). f' o VALVE FAILED (DISK SEPARATED FROM STEM)
. PRIOR TO ACCIDENT. DURING TE ACCIDENT TE VALVE WAS DAMAGED FURTER (VALVE YOKE F , CASTING BROKEN)
I o NOT A SAFETY RELATED SYSTEM !O 9 o SAFB Y REMTED CONCERNS
- LOOSE PAPTS IN SYSTEM THAT EE NOT ACCOUNTED FOR INADEQUATE MAINTENfNCE AND INSKCTION PROCEDUES
- FUTENTIAL FOR GEERATING MISSILES HAS NUT
- BEEN INVESTIGATED i
l 1 k lO i
j , O DAVIS-ESSE PORV
-o PORV STUCK OPEN DURING THIRD ACTUATION o PORV IS A CROSBY PILOT OPERATED VALVE o LICENSEE'S INVESTIGATION DID NOT CONCLUSIVELY
, DETERMIE FAILURE CAUSE HOWEVER, COULD E TE i RESULT OF mREIGN MATERIAL IN TE PILOT DISK AND 1 SEAT ^ o PORV HAD NOT BEEN STR0KE TESTE) SINCE SEPIDEER 1,1982. STAFF BELIEVES THIS CDULD HAVE CONTRIB(RED TO DEGRADATION Af0 LACK OF KNOWLEDGE TEREDF i - l i t i O l. I
O DAVIS-ESSE MAIN STEAM SAFETY VALVES (ITGV) o ALL 18 LIFTED DURING 6-9-85 EVENT. o SOE VALVES WEE OUT OF 1% SETFOINT TOLERANCE. SOE MAY HAVE OiATTERED (EVIDENCED BY PRESSURE OCILLATI0fG). o MSSVS AE DRESSER SPRING-ACTUATED VALVES. 4 o 8 MSSVS SENT TO WYLE FOR FULL FLOW STEAM TESTING i 4 TO DETERMIE SETPOINT, BLOWDOWN AND SEAT LEAKAE. ESULTS WILL DETERMINE WHETER ADDITIONAL MSSVS WILL BE Itsitu. i b o MSSV INLET PIPES SLIGHTLY SMALLER THAN AS SRCIFIED BY DESSER. LICENSEE A!O DRESSER INVESTIGATING TE - = EFFECT ON VALVE OPERABILITY. ~' 9 4 o l l 1 V
._ s.-
O DAVIS BESSE ATM)SPERIC VENT VALVES (AW) o EDTH AW'S OPEED SEVERAL TIES DURING 6-9-85 EVENT. o OE DID NOT CLOSE AT PROPER SETPOINT.
' TESTS SHOW ELECTRICAL CONTROL SYSTEM PROBLEM. }l
-o TIE SAE AW LEAKED AFTER CLOSING.
o- VALVES AE AIR OPEPATED Willi VARIABLE PELNATIC ! CONTROL. O 1 i r 4 I ' l -
,. , . - _ - . , . . - - . , - . - , - , . . - . . . - , -__. ..r. - - - - - _ _ , - , -}}