ML20053A768
| ML20053A768 | |
| Person / Time | |
|---|---|
| Issue date: | 05/20/1982 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1088, NUDOCS 8205270261 | |
| Download: ML20053A768 (60) | |
Text
s... _
,..,e
- U~b~ 4..
i l " p l9 T l '
w
/ j (bk b
' d-
- p NUCI.ZAR REGULATORY COM!CSSICN
~
O 1
9 O
l In the Mattar of:
ADVISORY COMMITTEE 0N REACTOR SAFEGUARDS SUBCOMMITTEE MEETING ON THE QUALIFICATION PROGRAM FOR SAFETY RELATED EQUIPMENT DATE: May 20, 1982 PA m : 142 thru 201 AT:
Albuquerque, New Mexico
/
/l,) / /
j\\, ;,
/
I,, [' '
)
[
/
> }; f i l
3 w
O
/
.HDERSOX REPORT 1XG G
/' - t 400 Virginia Ave.,
S.W. Washing: n, D.
C.
20024 h
Telachc=e: (202) 554-2345 8205:2's, % l
()
I UNITED STATES OF AMERICA 2
NUCLEAR REGULATORY COMMISSION 3
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
SUBCOMMITTEE MEETING ON THE 5
QUALIFICATION PROGRAM FOR SAFETY RELATED EQUIPMENT 6
7 Valle Grande Room 8
Amfac Hotel 2910 Yale Boulevard 9
Albuquerque, New Mexico 10 Thursday, May 20, 1982 11 12 The subcommittee was convened at 9:00 a.m.
13
()
PRESENT FOR THE ACRS:
J.
J.
RAY, Member and Subcommittee Chairman 1
15 DAVID A. WARD, Member ANTHONY CAPPUCCI, Staff 16 DESIGNATED FEDERAL EMPLOYEE:
RICHARD SAVIO ALSO PRESENT:
19 Present for Sandia National Laboratories:
L.
BONZON l
II L. CROPP l
l D.
JEPPESEN 22 W.
MC CULLOCH W.
SNYDER 23 Present for Ebasco Services Incorporated :
L. GRADIN O
i
143 1
P_ R Q g E E p I_ E g g 2
9:00 a.m.
3 MR. RAY:
We will be on the record.
O 4
This is a continuation of a meeting started yester-5 day of the Advisory Committee on Reactor Safeguards subcommittee 6
on Qualification Program for Safety Related Equipment.
7 I am J.
J.
Ray, subcommittee chairman.
8 I would like to open today's session by calling first 9
on Mr. Lawrence Gradin, Manager of Equipment Qualification 10 for Ebasco Services.
11 MR. GRADIN:
Again, my name is Larry Gradin.
I am 12 the Equipment Qualification Program Manager at Ebasco Services.
13 We in industry appreciate having the opportunity
()
14 to provide input to these valuable hearings.
Before providing 15 specific comments, we would like to commend Sandia for their 16 excellent presentations.
We have been reviewing and making 17 use of Sandia information for several years now, and find 18 that data very very valuable.
19 Unfortunately, the Sandia data and input has not 20 been given the attention we believe it should in the regula-21 tory process.
For example, we have assisted the industry and 22 Sandia in defining source terms and realistic accident scenarios 23 which still remain unaccepted by the NRC staff.
Furthermore,
(])
24 as stated yesterday by Mr. Jeppesen, no IEEE 323 1974 test 25 has been endorsed by the NRC.
O
144
- b 1
The question should be asked why the ACRS, after 2
such protracted effort, why this is so.
3 Other matters we wish to bring to your attention 4
are, one, our concern that the proposed rule 10 CFR 50.49 5
and REG Guide 1.89 revision 1 are not being planned for issu-6 ance at the same time.
7 Indeed, you, Mr. Ray, indicate on May 5th in your 3
ACRS subcommittee hearing that you are also concerned with 9
this lack of sychronization.
The last three years of evolv-10 ing and changing NRC staff guidance and interpretation of 11 equipment qualification must be resolved once and for all.
12 More than 20 bulletins, letters, EQ meetings with 13 industry have led to the definition and redefinition of mild
()
14 and harsh environments, NRC requitsments for submittals, 15 clarification upon clarification upon clarification, all of 16 which provides a variety of interpretations of the same re-17 quirement.
18 The new proposed rulo and regulatory guide must 19 envelope and make void all of these previous guidance document s.
20 Number two, recent NRC positions -- for example, i
21 attendance and presentations of the 5-5-82 ACRS subcommittee 12 and proposed REG Guide 1.89, revision 1, implies that the I
l 23 requirements of IEEE 323 1974 is not applicable to mild
[}
24 environment equipment.
By implication, IEE 649, 650, 387, l
25 535 -- by the way, these deal with motor control centers,
145
[}
1 battery chargers and verters, deisel generators, and batterien 2
-- are all located with mild environments.
Therefore these 3
standards themselves are not required.
It must be clearly O
4 and unambiguously stated by the NRC that they do not intend 5
to have these standards envoked on mild environment equipment, 6
or redefine mild environment equipment requirements.
7 We cannot let the IEEE standards be issued, the 8
standard review plan reference these documents, and then the 9
NRC comes out and says nothing need be done with mild envir-10 onment equipment.
11 Number three, the Sandia presentations include much 12 mention of the need to identify true functionality.
This is 13 one of the keys to the actual success of equipment qualifica-()
14 tion.
We endorse this position, as our own experience shows 15 that arbitrary generic criteria must not be applied to specif-16 ic cases.
Thank you.
17 MR. RAY:
Thank you, Mr. Gradin.
18 Is there any discussion that anyone wants to present 19 on Mr. Gradin's remarks?
20 Dick, how do we get a copy of Mr. Gradin's document.
I 21 is it in the transcript?
12 MR. SAVIO:
He can give us that hand-written.
23 MR. RAY:
You will give us that this morning?
(])
24 MR. GRADIN:
Yes.
25 PR. RAY:
Fine, thank you.
i
146
(])
1 For the balance of the meeting, it was my intent 2
to make it free-wheeling, questions and general discussion 3
on the part of anyone in the organization of Sandia, or the 4
two participants from Ebasco.
They should feel free to offer 5
suggestions or comments.
6 I would like to start with asking whether or not 7
there were any residual comments on the part of the Sandia 8
representatives that they would like to make.
9 MR. CROPP:
Yes, Mr. Ray, I would like to make a 10 comment about in the instance of the Hydrogen Burn Survival 11 Program.
We pointed out yesterday that the hydrogen burn 12 problem is neither trivial nor insurmountable.
I am not sure 13 we indicated th~at we have seen utilities who fully recognize 14 this and seem to have taken care of the problem.
15 We don't mean to imply that every utility has a 16 problem that they are going to have to look at in tremendous 17 additional amounts of detail.
Some of them have done a great 18 deal of work already.
All we meant to imply in those state-19 ments is that you cannot just brush the problem aside.
You 20 have to consider the parameters, the variables, and prove 21 that you have taken care of those that could cause you problens.
22 MR. RAY:
Mr. Cropp, that came through fairly clear:.y 23 to me.
And I am fairly dense.
So I suspect that when this
()
24 record is published, if the utilities look for it, they will 25 get that message.
O
147 1
I do want to commend you and Mr. McCulloch for the ggsb 2
presentation.
I think it was quite clear and productive of 3
better concepts and better understanding.
p,
(' ')
4 Any other comments?
5 In the absence of any other comments, I would like 6
to open the floor to questions on the part of my associate, 7
Mr. Ward, or the two staff members who are here.
Did you want 8
to lead off, Dave?
9 MR. WARD:
Yes, I have got a couple of things.
On 10 the hydrogen burn survival, could we hear -- you said that 11 you were going to develop an algorithm which would be an aid 12 to the staff in evaluating the proposals from licensees.
And 13 I know you haven't developed that yet.
But could you tell
_/
14 us a little bit more about what you have in mind?
Is this 15 going to be some sort of a complex computer program?
I am 16 haunted by appendix K.
Or is it going to be a set of quanti-17 tative guidelines, or a set of qualitative guidelines, or just 18 what sort of thing do you have in mind?
19 MR. MC CULLCCH:
Fundamentally we try to look at 20 the analytical capabililties that we think the NRR can --
21 MR. WARD:
Excuse me, we have an interrupting back-22 ground noise.
23 MR. MC CULLOCH:
We are trying to take into account
( })
24 what we think the NRR can maintain, and what kind of analytical 25 capability they can maintain.
And included in our considerat:.on j
148 1
there is that we do not expect them to have a very elaborate V('ab 2
computer code, to the types that we often use, for instance, 3
which would require one or two or three people just to keep O
4 operational.
i 5
On the other hand, we feel like that we will be 6
supplyina them with information that is somewhat complex, 7
covers a good number of physical mechanisms and so forth.
.So 8
we think we are talking about having to evaluate each one of 9
the physical mechanisms in some way.
10 We would like to be able to supply a set of what i
11 we would call nomograms, or simple graphical techniques so 12 that they can cover a variety or a range of parameters.
Per-13 haps we may have to be as complicated as programable calculator-()
14 kinds of analysis techniques.
Some of them may be very simple.
15 The evaluation of a maximum gas temperature for instance, may 16 be just a simple chart-look-up sort of presentation.
But we 17 will have to evaluate each one, each parameter, or each vehicl e 18 mechanism that we are trying to address individually.
But 19 our goal is to provide something to them that would be reason-20 ably expected for a technically competent individual to main-21 tain on his own at his desk and not require a large analytical 22 support staff.
23 MR. WARD:
Does that mean for example that they
(}
24 could decide and input these nomograms or whatever, for exampl e, 25 the amount of hydrogen that they would assume to be generated O
149
N 1
cd by a given plant?
I mean, if --
2 MR. MC CULLOCH:
Yes.
3 MR. WARD:
Okay, it would.
You aren't predetermining O
4 that all of the zirc in the core, or two thirds of the zirc, 5
or something like that is input --
6 MR. MC CULLOCH:
No.
We would try to present the 7
material in such a way that the NRR would have a range of 8
parameters.
Let me address one, ignition criterion.
9 If they want to consider 10 percent as the safety 10 analysis basis, then that number would be there.
But if they 11 wanted to check the analysis of someone who had submitted an 12 analysis of the basis of an 8 percent criterion, they could 13 check the analysis at least, even though they might not agree
()
14 that that is the proper input data to use.
15 However, if the utility company or the licensee 16 makes a valid case that his ignition system will indeed oper-17 ate at 8 percent, then that data would be there and available.
18 It would be included within the range.
19 Now the range of parameters that we will cover in 20
-- and I will use generically the nomograms -- would be 21 developed in consultation with NRR and others in the field 12 to determine what range we want to cover with each one.
23 Also the first cut of this algorithm will be specif--
()
24 ically addressed to ice condenser PWRs.
One of the very next 25 things we want to do is to expand that to look at other types 0v
150
(]}
1 of containments.
And in the process of doing that we may 2
find that by increasing thesparameters, let's say of contain-3 ment volume, for instance, by just increasing the range on 4
that parameter, that we can cover those containments as well 5
as the ones we have already addressed.
6 MR. ilARD:
Will they be able to develop from that, 7
then, a sort of a pressure and temperature profile in the 8
containment atmosphere?
Or will there be information on 9
the temperature profile that would be expected in a given 10 type of instrumentation at some given location in the contain-11 ment?
How is that aspect going to be handled?
12 MR. MC CULLOCH:
Fundamentally, I think, what they 13 will be able to get directly will be what we call that worst-O 14 case hypothesis.
And that is taking the worst of all of the 15 contingencies, and then we will be able to modify that envir-16 onment to take into account those threat reductions that are 17 justified for the particular application that is of concern.
18 And those wil'.
take into account both where you are in the 19 containment, and the design of the instrument that we are 20 concerned with and how it is mounted.
All of those things 21 will be taken into account.
And the result of that will be 22 a closely approximate description of the environment in that 23 location for that component.
/G 24
(_/
With that you will be able to arrive at the compon-25 ent temperature.
O
151 1
MR. RAY:
Could I interject at this point, since
(]
2 you are on the algorithm.
Is there general agreement between 3
your organization and NRC that this idea is desirable?
O 4
MR. MC CULLOCH:
Yes.
5 MR. RAY:
They expect this, in other words?
6 MR. MC CULLOCH:
Yes.
7 MR. RAY:
Thank you.
8 MR. MC CULLOCH:
I might say one other thing a1ong 9
that line.
The agreement exists very solidly with the program to managers for our project.
Obviously this information will Il be of use to a lot of other people as well.
For instance, 12 the utility companies, I am sure, when they know how they are 13 going to be evaluated, would be very interested in knowing O
i4 end checxing ehe informetion ehee we w111 sugg1y.
15 But in addition to that, there are other areas in 16 NRC who are interested in this kind of information.
Specif-17 ica1ly the Equipment Qua1ification branch of NRR has approached 18 us and asked us questions that take a little different slant.
19 You will see from what I presented yesterday that primarily 20 what I am taking about is a design tool, that the a1gorithm 21 can develop the appropriate environment in which an instru-22 ment is to operate.
And then we can look at test data on those 23 instruments, or on the candidate instruments, to find one which
(]
is capable of surviving in that environment.
That is fine.
24 25 Now what do you do about all the existing plants and existing O
152 sb 1
designs of plants where the instrument has already been 2
specified, or in some cases already installed?
Now the 3
question is I have this instrument, do I, the NRC, have to GV 4
require a retrofit or a redesign?
The ones that are there, 5
are they really that bad?
What is the degree of threat or 6
danger of failure that existing systems have?
7 By looking at tests that we can do in our radiant 8
heat facility, we can assess the damage that might be expected 9
for components, instruments, which may not on the surface 10 appear to be fully qualified for a particular application.
11 But we can evaluate the degree of threat that is there.
We 12 can do that by repeatedly testing these things to what we 13 think are representative heat flux profiles.
Again, as I
()
14 said yesterday, those representative heat flux profiles have 15 to be done with some kind of a test apparatus.
They can't 16 be done -- we can't get a hydrogen burn environment where the 17 energy is coming from the burning hydrogen to really test 18 these things.
We can do that at the radiant heat test facility.
19 And so by testing these components which are already 20 installed or are already designed to be installed, we can 21 evaluate what they are likelihood of failure is.
And in that 22 case we would be particularly interested in functional failures, 23 not just material property changes.
That is one area of future
({}
24 work that we intend to propose to the NRC as a follow-on to 25 our work.
b-s l
153 c'l 1
MR. RAY:
In your presentation yesterday, you made V
2 it clear that while the applicant, the industry, the user 3
industry, the utilities, would like to brush the hydrogen burn J
4 probem away and treat it as being incidental or not important 5
or not urgent, and not critical, I got your point that you 6
made was that this shouldn't be done, that it is a serious 7
situation.
And it requries analysis at least.
I got the 8
import of your conclusions to be -- and don't take this as 9
necessarily all-inclusive and final -- that while it is a 10 serious problem and has to be considered in significant detai l, 11 wholesale replacement of equipment is not necessarily implied 12 at this time in your results, that the really significant 13 thing is that protection is needed as of now.
14 MR. MC CULLOCH:
That is right.
15 MR. RAY:
And I think that message to them should 16 be encouraging.
17 MR. MC CULLOCH:
I think so as well.
We, in the 18 last couple of weeks, have had some interaction with utility 19 companies.
And they understand our position and are reasonab Ly 20 responsive to it.
It is a situation of the regulatory side 21 of the business pushing on the other side to do something 22 that is going to take some effort, take time and money.
23 And I can understand their hesitance.
They really want us
()
24 to do something that is right, so that we don't ask them to 25 do something and then some months from now ask them to do
154 egg 1
something else.
And generally with that kind of a preamble, q
2 the utility companies have been very perceptive of the infor-3 mation that we have given them.
(
)
4 MR. RAY:
Recognizing the import of Mr. Gradin's 5
comments this morning, have you had any reaction yet from 6
the NRC as to their responsiveness to this avenue of qualif-7 ication?
8 MR. MC CULLOCH:
Yes.
The people that have talked 9
to us about this area generally are looking for ways of being 10 responsible to the regulatory function, and yet not put un-Il reasonable and severe requirements upon utility companies.
12 Being flip about it, you can say they are looking for a way 13 out.
And they see that what we are saying does make sense.
,,(/
14 Their experience indicates that it is reasonable to expect 15 that protection will be available.
And this is a sort of a 16 middle-of-the-road position which can be responsible both 17 to the rate payers and to the regulatory function.
18 MR. RAY:
So that in this area at least one would 19 expect their response to be objective?
20 MR. MC CULLOCH:
Yes, relatively so, sure.
21 MR. WARD:
What did you mean, Jerry, by equipment 22 won't have to be replaced, but will have to be protected?
23 MR. RAY:
I am thinking in terms particularly, l
/~
. ()
24 for instance, of the cables.
The cables don't have to be 1
25 qualified to sustain necessarily a burn right at the cable
,a
%J f
155 sb 1
sources.
If they are tucked away in recesses and that sort 2
of thing, and with selected locations within the containment, 3
and routing, and so on, and the equipment to which they 4
supply energy or from which they take intelligence be located 5
appropriately so that this sort of thing could be a practical 6
arrangement.
And maybe components such as fans and so on 7
within containment can be located in cognizance of the results 8
of the burn investigation, so that they won't be submitted 9
directly to the effects of the burn.
Or a shield would be 10 put up near and around them, this sort of thing, rather than 11 have to replace components with something that still has to 12 be qualified -- a design that still has to be qualified to 13 sustain direct impact of high temperatures, this kind of thing.
14 Maybe I am reading too optimistic a tone into what 15 your preliminary conclusions are, so far.
16 MR. MC CULLOCH:
As long as you keep it within the 17 perspective that it is the responsibility of the using organ-18 ization to substantiate tnat the protection is in place and 19 will be appropriate in an accident environment, I don't think 20 you have overstated it.
21 MR. GRADIN:
May I make a comment on that?
22 MR. RAY:
Yes.
23 MR. GRADIN:
One of the slides that Lloyd Bonzon 24 did not show yesterday, which he has in his reports, or he l
25 has shown in July, 1981 meetings with the NRC, the four-day
156
(}
intensive meetings, was the actual Catawba penetration test I
2 which demonstrated very graphically the thermal lag capability 3
of a metalic enclosure around the penetration terminations.
4 Within industry there is a significant effort --and industry 5
is addressing it -- of looking at main steam line breaks and 6
pressurized water reactors and showing the very sort duration 7
transient is in fact enveloped by the LOCA qualification tem-8 peratures which are substantially less.
The numbers we saw 9
yesterday which were 36S degrees F are not that substantial, 10 at least for the newer plants.
The newer plants which address 11 appendix A of 323-74 have numbers like 340 degree F for a 12 LOCA environment.
A thin metal enclosure would easily provide 13 sufficient thermal lag barrier to protect everything that is
/~
(_)
14 within that enclosure -- almost everything.
I am generalizing 15 a little bit.
But I am trying to bring some real world to 16 this.
17 For cables, there is other things.
Industry, as 18 an example, has always designed cables to withstand short 19 circuit capability.
A cable can withstand a short circuit 20 for -- I think the temperature rises 250 degrees Centigrade 21 on the insulation.
A very short duration can be reviewed.
22 Now I haven't done this personally.
But I am sure that to 23 a great extent most of these things would not be of major
(]')
24 So I am really in support of what Sandia is saying.
concern.
25 Because when we have looked at almost every component for the OV
157
('*
1 main steam line break events, the one-minute-type transient O
2 at something like 420 degrees F for an item that was qualified 3
for 340 F, we found little problems except for very, very sma;,1 O
4 mass components which have very small, very thin, sheet metal 5
enclosures.
I would think that Sandia could bear me out on 6
that because they have done extensive testing in this regard.
7 MR. WARD:
Is there any -- well, go ahead, were you 8
going to comment on that?
I am drifting off a little bit 9
from this point.
10 MR. MC CULLOCH:
All right.
Let's stay with this 11 one.
12 While I think the atmosphere is very agreeable 13 between us and the utilities relative to the protection, I
(
14 think there is still a fairly wide disagreement as to what 15 the basis of a safety analysis should be in terms of the, 16 let's say, worst credible environment.
Our numbers, in com-17 parison to the analyses that have been done by the utility 18 companies are still quite a bit higher for gas temperature 19 and the duration at temperature.
And I don't want to forget 20 that aspect of it.
Right now, in terms of numerical differences, 21 that is where the biggest part is.
22 MR. WARD:
Is there any potential at all for cables 23 igniting in a hydrogen burn, or in what could be a series of
,-.()
24 connected hydrogen burns?
25 MR. MC CULLOCH:
Well, yes, the potential is there,
'f O
158 1
if they are exposed.
The cable manufacturers are doing things qgg 2
to retard that possibility.
3 MR. WARD:
I mean with the types of cables that are
(
)
4 in service now.
5 MR. MC CULLOCH:
Yes.
The ones that are in -- some 6
of the cables that are in service now are really quite old 7
in design.
That is one area where we have difficulty in get-8 ting test specimens.
The ones that now exist are all in the 9
possession of people who need them for ba:kup, for stores, 10 supplies.
And it is difficult to get a sizable piece of cable Il for testing.
12 But we feel like if we make the case to utility 13 companies, we can get a few feet to do some experiments k/
14 anyway.
So the potential for fire on existing cables, I 15 think, that is of concern.
We need to evaluate that with the 16 specific cables that are installed.
17 Lloyd made the point over and over again that there 18 is different -- even though the material on its first specifi-19 cation, the specification may be the same, that there is a 20 difference from batch to batch, and from manufacturer to 21 manufacturer.
And some of these differences are important.
22 The thing that you are into with cables and with 23 all the organics, whether for seals, lubricants, gaskets,
(')
24 adhesives that might be used, all of those things you are in 25 a very complex chemistry.
And nearly every one of them has
159 c'}
1 to be evaluated on its own.
2 MR. WARD:
When I go through a plant now being buil t 3
I seem them smearing all this flamastic or whatever it is in r'%
%)
i 4
all the cable trays.
I didn't hear any discussion of whether 5
that is in the materials test program.
I guess that is more 6
than one thing too, chemcially, I am sure.
Is that part of 7
the materials test program evaluation?
8 MR. MC CULLOCH:
I don't know that.
Lloyd?
9 MR. BONZON:
Yes.
If I can -- there is another 10 program which we didn't review this time, essentially a fire 11 aging program -- a fire program, I am sorry.
That is handled 12 by another division.
That is the group that has been looking 13 into the separation distances and the coatings and the like.
()
14 If your question was does that stuff age or do other 15 things, we have not actually looked at the aging of the 16 flamastic or whatever coating material.
We have looked at 17 the fire-retardant aging a little bit in terms of what is in 18 the cable itself, and I showed you actually two, a small-scale 19 and a large-scale thing.
And we don't think that particularly 20 ages.
Now that doesn't address the question of whether the 21 coatings themselves might age, especially if you applying it l
i 22 over older style cables which are themselves not as fire-retar-23 dant, maybe, as some of the newer ones.
The reasons for the
(])
24 coatings was to protect the older style cables.
I don't know l
j 25 of any work that is being done in that regard right at the l ()
160 1
moment.
ggp 2
MR. RAY:
In the latest plants cable with fire 3
retardant characteristics have been designed and presumably
(
)
4 tested by the manufacturers for uirect impingement of flame.
5
.d if it survivies that kind of thing, I would fully expect 6
it to survive a deflagration of hydrogen, which is a transient 7
condition.
So I would expect the latest plants to be fairly 8
secure from the viewpoint of flammability of cable in a hydro--
9 gen burn.
10 MR. WARD:
Yes, but the latest plants, Jerry, are 11 putting in this stuff, apparently because they need it for 12 protection against the exposure fires.
So I am wondering 13 how effective is that going to be in 20 years, I guess, is
(.,
(m) 14 my question.
15 MR. MC CULLOCH:
Another thing to remember is that 16 the plants which are now being built, the cables were specified 17 in design several years ago.
And in many cases the companies 18 the utility companies in knowing that and knowing that the 19 cable manufacturerr __e changing almost on a monthly basis, 20 updating and improving their products, that many times before 21 a foot of cable is installed, that utility company or the 22 construction company has bought up the supply that they will 23 need through the life of that plant.
And that cable is no longer available, before it is ever installed.
And so some
']
24 v
25 of the things that are going in now are really several years V
161 s
1 old.
2 MR. RAY:
So the availability of test samples is 3
definitely a: problem?here.
O 4
MR. MC CULLOCH:
Yes.
5 MR. WARD:
I would like to come back to that point 6
later in a more general way, if we could.
7 MR. RAY:
Sure.
8 MR. GRADIN: If you don't mind, I just borrowed this 9
from Lloyd Bonzon.
This was the actual slide he had utilized to in July, 1981.
What it does is demonstrate very graphically 11 actual tests that prove that the thermal lag concept is 12 indeed true, which we use in industry.
In fact, this partic-13 ular slide is in a submittal or plant we are working on, to O
(_)
14 the NRC's first category 1 NUREG 5-8-A plant.
After all the 15 theoretical calculations we have utilized this to demnstrate 16 an actual National Laboratory-proven validity of the calcu-17 lational methods.
What you see h re is this is the transient 18 outside.
This is the metal junction box temperature.
And 19 here is the components inside.
This is time in minutes.
It 20 is very substantial before you reach equilibrium.
21 So a short transient hydrogen burn should not be 22 a substantial problem for many components.
But some, yes, 23 indeed it will.
I agree on gaskets and things like that.
() 24 MR. RAY: In the earlier parts of that transient, 25 Mr. Gradin, where there is a major differential between the A%/
162 1 qgg temperature seen by the occupant of the shielded area, and 2 the outside environmental transient. How many degrees dif-3 ferntial is that? I can't read it from here. ,_s l ) 4 MR. GRADIN: This is in centigrade. It appears to 5 be as much as 75 degrees centigrade. 6 MR. RAY: That is fairly sustained through that 7 period, it looks like, the ordinant. 8 MR.GRADIN: This is time interval from here to here 9 This is 10 minutes to 40 minutes. It is a half an hour. 10 MR. RAY: And it is 40 degrees cene.igrade differ-11 ential sustained for a half-hour period? 12 MR. GRADIN: On this graph, 40 minutes into the 13 graph, they have a difference between -- it is still 25 degrees i._) 14 centigrade difference. That is very substantial. We are 15 talking about a typical equipment that might be qualified 16 for 340 degree F. And the accident number they talk about 17 was 365 F. We should have, with some equipment, not too much 18 problem. 19 MR. RAY: Yes, you would be within the max temper-l l 20 ature qualified. l l 21 MR. GRADIN: This was one of the examples of the l l 22 three-dimentional smoothing (ph) about some of the good data l 1 23 Sandia does have. I think you should use it. l [', 24 MR. WARD: We know, though, Larry, that when we m.- 25 looked some of the stuff that Mr. McCulloch showed yesterday, c i, 1
163 3b 1 we see for the analysis for 10 percent single burns, Sequoyah 2 upper compartment, that the temperature is still about out 3 maybe eight minutes or so after the initiation of the burn. O 4 So we aren't talking about -- 5 MR. GRADIN: Well, they have an ice condenser. 6 Sequoyah is a little different than most PWRs. However -- 7 MR. WARD: I know, but it is the ice condensers I 8 that, I guess, we are most worried about. 9 MR. GRADIN: So that is another point. The ice 10 condenser is the one concern. The majority of pressurized-11 water reactors in this country, dry containments, in fact the 12 NRC has issued in several documents their belief that it shou Ld 13 not be a substantial concern in larger-volume containments. 14 And in that upper compartment we are talking about way up on 1 l 15 top of -- we are talking about a typical PWR. We might have 16 a 400-foot high containment building. The highest level of 17 the receptible equipment, the electrical equipment, the 18 organics, might be at elevations of 60, 70, 80. Above that e] e-19 vation you have' containment spray piping. You have the 20 metalic duct work for ventilation. You don't really have 21 the equipment as susceptible. I would think that is why they 22 take that number, 365 for equipment. 23 MR. MC CULLOCH. That number, 365, is just indicative l 24 of a number which is somewhat -- a little bit higher than 25 j qualification temperatures that have been achieved so far. 1 O
164 1 But in our experiment that was done in a test vessel, and so 2 that the exposure time was very short, compared to a full-3 size containment. The plot that is up there right now does O 4 show that a system like a metal valve or something or a metal 5 box around electronics very well will miss that short-term 6 transient. But in this case that short-term transient ic 7 a threat of only up to that chamber temperature which is on 8 the order of 175 C. That is like 350 or so F. To put it in 9 perspective with yesterday. In that temperature range, the 10 dominent mechanism of convection, heat transfer from air to 11 the component which is not a very effective heat transfer 12 mechanism compared to radiation from gas which is at 2,000 13 degrees. So there is a benefit to be realized from thermal () 14 lag, but it has to be put in a right perspective of what is 15 the environment that the component is exposed to. 16 MR. RAY: Excuse me. Before you leave this diagram 17 do I understand correctly that the internals were completely 18 enclosed in a box in this incident? 19 MR. BONZON: Yes. I was going to try to clarify 20 that, show you which thermocouples were measured. This 21 basically shows the device being loaded into our test chamber 22 This is a metal box, which is not a sealed box, but it essen-23 tially encloses. What it encloses is the bottom part of the (]}; 24 penetration. Which looks almost identical -- this would be 25 the outside of the containment and there would be an identica: O
165 1 version of connectors inside the containment. That encloses c( } 2 the box. What you are seeing is for *he environments we are 3 essentially taking the average of these kinds of readings O 4 which are outside the box but inside the test chamber. And 5 then this set of readings is inside the box, but not attached 6 to the connectors directly. This is attached to one of the 7 connectors directly, in other words, we have it physically 8 touching one of the connectors. All this demonstrates is 9 this is the environment temperature surrounding the box, 10 this is what the air inside the box sees, and this is simply 11 what the connector sees. And there is nothing very magic 12 about this. This is not a wonderous National Lab development 13 It is simply a measurement of reality. And things lag. I ) 14 don't know what other demonstration you need. 15 MR. WARD: Yes, it is not a landmark measurement, 16 or anything. 17 MR. BONZON: Unfortuntely it appears to be somewhat 18 unique on the things that are measured in this business. 19 I would encourage all my utilitly friends to do these 20 kinds of measurements when they do tests, just for the heck of l 21 it. So that is simply an explanation of what it looks like. 12 MR. RAY: Thank you. 23 MR. WARD: Let's see. I guess still on the subject () 24 of the hydrogen curr, if we could. I think we discussed this 25 yesterday, but aoparently I didn't quite get the point. r~g (/ l l l
166 {5 1 What sort of environmental qualification requirements are the re 2 going to be for the hydrogen igniter systems? I understand 3 they are to be safety-grade. Are they going to be designed O 4 or qualified so they will survive a large break LOCA? Are 5 they going to be designed and qualified to survive a first bu rn 6 or a series of burns, or all the burns that could conceivably 7 come in a series? And I don't just mean the glow plug, but 8 all the circuitry in the containment. 9 MR. MC CULLOCH: I am not aware yet of any even 10 preliminary rule that is taking shape, that directly addresses 11 that problem yet. In general, the kinds of things that are 12 being said right now is we need a deliberate ignition system. 13 Tell us what you have got, and how it is going to live. And () 14 the burden has been given to the licensee to say what his 15 system is capable of doing. Some of them have done some 16 preliminary testing, cursory testing. And we are in the pro-17 cess, NRC and NRR is now in the process of reviewing some of 18 those. But I guess it is just too early to start to define 19 a qualifications test. Now I feel like because of the impor-20 tance of these igniters to safety, that that is the end goal. 21 I don't know yet of a rule or qualification test scenario or 22 description that is taking shape. 23 MR. WARD: I guess it is really a question we need () 24 to ask of the staff. But an igniter system that is destroyed 25 or disabled by the first burn isn't going to do us much good
167 O I in an accident. \\-) 2 MR. MC CULLOCH: Exactly. 3 MR. CROPP: I think a number of utilities have done \\# 4 a number of tests in the small-burn facilities showing that 5 their igniters do work, time after time after time. That is 6 the intent. 7 MR. WARD: But is that with the entire igniter 8 circuit, or just the plug end of thing? 9 MR. CROPP: To the best of my knowledge, it is with 10 the entire igniter circuit. The one we saw was in a box. II And all the electronics was in its 12 MR. WARD: Okay. Because I know yesterday, I guess 13 it was Mr. Bonzon said something about some of the utilities O'(,) 14 are using' spare or lighting circuits or something to power i 15 they system of igniters. I am not sure the lighting circuits 16 are expected to survive a hydrogen burn. 17 MR. MC CULLOCH: We are not really in a position to 18 say that they have done that completely. They do address 19 things like multiple igniters in areas where there might be 20 a problem. Our experience has been that by putting everything 21 that is needed for the igniter, a transformer and certain 12 controls and shields and that sort of thing, can be put into 23 this box. And they do not seem to be threatened at all by ({} 24 the environment. The wiring, the controls and power wiring 25 to that system, I don't know that we have really addressed O
168 I that as such. There is a lot of experience with how you protect 2 wire, and various kinds of circuits. And that technology 3 or that expertise, that experience with the utility companies O (/ 4 would certainly be available to this. 5 Now I made the statement about displacing some of 6 the lighting to do this. And they have selected redundant 7 circuits, that sort of thing, to do that with. And in genera L, 8 I think the utility companies are moving pretty well in that 9 area, because those plants that need this kind of protection to are deeply committed to that thing working. And if they 11 can't demonstrate that that is a reliable safety system, 12 then that gives them all kinds of problems that they have 13 got to treat otherwise, not the least of which is the possible () 14 rupture of the containment. 15 MR. WARD: Well, I appreciate that they are concerned. 16 But I guess the question is whether they are addressing the 17 concern fully and properly. 18 MR. MC CULLOCH: I have little doubt that before it 19 is over with that the requirements placed by NRC will be 20 somewhat in excess of what they think is necessary. But that 21 is the nature of beast. And right now I think it is a fairly 22 healthy atmosphere of cooperation working toward eventually a 23 fully-qualified hydrogen ignition system. /N 24 MR. WARD: Thank you. 9G 25 MR. RAY: Thank you. I would like to ask a questior l l [
169 1 The point has been made several times by you, Mr. McCulloch, 2 and by Mr. Bonzon in his presentations yesterday and today 3 that your reports and the results of your research, partic-O_s 4 ularly from the viewpoint of the characteristics of the res-5 ponse of materials to the accident environment are.quite useful 6 to the utilities for them to anticipte, and perhaps from a 7 design viewpoint qualify in advance in what the utilities 8 are being guided to put in, to be better off when these 9 permutations and combinations of regulatory requirements and 10 changes in regulatory requirements become a real exposure. 11 I wonder, are your reports to the commission or to 12 the staff in any way proprietary, or are they in the public 13 record so that the utilities have complete access to them () 14 for whateve'r guidance they see fit to take from them? 15 MR. CROPP. Wherever we can we make all reports 16 general in nature and generic in nature so that they are com-17 pletely free to anyone. There are instances where we may get 18 into proprietary information on a specific vendor's product. ~ 19 At that point what we do is issue a complete draft of that i 20 report on limited distribution to the NRC. And then it is 21 up to the NRC and that vendor to decide which parts, if any, 22 are proprietary or not. So there is the potential that some 23 parts of some reports may never make it to the general public. (]) 24 MR. RAY: In cases like that, are those parts deleted, 25 and the body of the report otherwise made available? I mean, i
170 sO' do you expurgate from the report the parts that are not free? I 2 MR. CROPP: We have not gotten to that point yet. 3 I assume that we will. If we find that the vendors make a O 4 claim for proprietary information, and that claim is upheld, 7 f (< webandooneoftwothings--certainlythecompletereport 'f will remain proprietary. And we could issue another report ,/; 7 that gives as much information as we possibly can without in-8 cluding the proprietary information, 9 Of course, the NRC always takes and gives the general c I '10 results of a test in bulletins and that form of information 11 anyway. So the general results are always'known. i 4, 12 MR. RAY: Thank you. 13 MR. BONZON: Could I expand upon that? ( 14 This is one of the thorns which always stick you 15 in the side. The reports on the QT program that I described 16 up until very recently were all available to the public. / 17 And we had established a very large distribution list because / 18 of the interest of the utilities and people working in the 19 area. 20 Two things have happend which are absolutely frustrat-21 ing. One is the limited distribution report that Lou Cropp 22 was just talking about, the probb m t proprietary nature. 23 We have had one report of thac AO , coming out of the QT (]) 24 program, in particular the one describing the Catawba pene-25 tration, of which I just showed a viewgraph. That should not s [ \\ O
171 ) I have gone under a limited distribution report. But because 2 we are in this very uncertain stage, it did. That is not as 3 bad as it sounds, because if people know about the report, 4 they can in fact get the raport. But thateisvsomewhat of a 5 road block. 6 The second thing ' hat has happened is very interest-- 7 ing to me. We had a large dittribution for these reports 8 because people had requested to be on the distribution. And 9 in the interest of saving money, someone in the government 10 decided that they would more or less arbitrarily cut about 11 half of those people out. That happened without anyone asking 12 me or anyone else whether these are the right people to cut 13 out or the wrong people to cut out. But in fact half of them 14 were cut out, so that we could demonstrate that paper is being 15 saved in the government, I assume. 16 We have tried desparately to reinstate people who 17 are interested in this program. I think we will accomplish that by sending l'tters to all who were on previous distri-18 e 19 bution to see whether they wish to remain on distribution. 20 But the rules have changed. And the rules say that unless 21 you have a direct interest and reason to be on a distribution 22 you will no longer be on the distribution and you have to buy 23 the reports like everyone else. (7) 24 To me that is a little bit of false economy. Because 25 in fact the: people who are submitting things to the licensing
172 1 process should have access to this. My experience has been 2 the way new ideas come into being in reality is that industry 3 generally leads. In other words, they take our reports and 4 they do other kinds of testing. It is not so much that NRC 5 regulates and then things follow. 6 And so I believe our work has been better used 7 perhaps by the industry because they have access to the report s, 8 than perhaps by the NRC telling the industry to use these 9 reports, if you see my point. I think it is a terrible false 10 aconomy to cut these people out at this time. 11 MR. RAY: In your efforts to correct that situation, 12 have the powers that be been responsive? 13 MR. BONZON: I have suggested that the best way to () 14 do it is for me to send that letter. But of course that can't 15 be done under the rules of the system. Whoever distributes 16 these reports back in Washington will send a unit standard 17 form letter out to everyone. I am told it will be in June. 18 But has happened then, you have a year's worth of time when 19 no one knew that this cutting had been going on. And for a 20 year at least there will be reports missing in people's files. 21 And they will have to go back and update the files. The 22 answer is they have promised to be responsive. I haven't 23 seen the letter. (]) 24 MR. RAY: But you haven't seen it yet. 25 MR. BONZON: I would push on that. I beg on you O
173 (]p 1 to push on them, because I think it is a very important issue, 2 Why do the work if you don't let people have access to the 3 work? b J 4 MR. WARD: It sounds to me like it is a conspiracy 5 by the quick copy industry. 6 MR. BONZON: Well, there is a longer history. We 7 used to, in fact, publish and distribute these reports. But 8 there was a contract given to someone to distribute reports. 9 And we no longer can distribute them directly. So we send 10 them back to Washington, who in turn redistribute. Let the 11 record show I am shrugging my shoulders. 12 MR. RAY: But you would like to have a prerogative 13 of direct distribution from here, from your organization? } 14 MR.BONZON: That is not the point. The point is 15 get them distributed. It seems to me a lot easier just to 16 send them directly from here. 17 MR. RAY: That is my point. 18 MR. BONZON: But if it can't be done, as long as 19 they are distributed, I don't care. 20 MR. RAY: My own reaction is that if they are kept 21 out of -- from the viewpoint of distribution costs -- if they 22 are kept out of the hands of a governmental agency, you are 23 going to do it more economically. (]) 24 MR. BONZON: I have no comment. 25 MR. MC CULLOCH: There is a related issue to the _m
174 (~T 1 proprietary thing. And that is that when we go out to try 'J 2 to purchase nuclear-grade equipment for the purpose of our 3 tests, particularly if that component has already passed 7-O 4 somebody's qualification test, there is a hesitancy on the 5 part of those companies to provide that to some other testing 6 If they feel like there is any possibility that that component 7 may fail a preliminary test which is not really a qualification 8 test or anything like that, but the word might get out, and 9 then their components would not be regarded as being compet-10 itive with others which were not tested. So in order to keep 11 an open-door policy with the component manufacturers, we do 12 need the ability or the process of protecting information 13 which is specifically related to particular components which 14 might damage prematurely their market position. Companies 15 have a great deal of investment in nuclear-grade equipment. 16 And they don't want that compromised in any way. They are 17 not very cooperative if they think it might be compromised. 18 MR. WARD: That is the point I want to bring up, 19 Jerry. I was really disturbed by that. And then you mentioned 20 again this morning that some cable materials aren't readily 21 available to you. This seems like a terrible situation to 22 me, kind of irresponsible. 23 MR. RAY: That is right. (]) 24 MR. WARD: I agree there has to be some protection 25 to prevent the alarmist reaction syndrome from developing O
J 175 rl } every time you run a test and something breaks or burns or 1 2 something like that. That has to be controlled. But to attelnpt 3 to control that by impeding the availability of materials you O 4 need for research testing or for evaluation testing, is terrible. 5 Jerry, I don't know if this is something we need 6 to -- I mean, if this is really the situation, and if it is 7 really impeding the program, I think the staff needs to give 8 it some attention to develop some way that the material will 9 be available as it is needed. Like, I find it pretty hard to to believe that the guy tells you that he has got a lot of 11 cable in his warehouse. They aren't making it any more. And 12 he can't give you a thousand feet of it because he might need 13 it. I just can't believe that is a good excuse. () 14 MR. RAY: I can supplement your feeling, David, 15 in that respect. Because I know of specific instances where 16 construction of a plant is in progress, and all of a sudden 17 it is realized by either the AE or the principal contractor 18 that they underestimated the cable requirements. But they 19 knew that utility ABC in Credunkville had that cable in 20 stock. They had overbought or they had bought anticipating 21 future needs and what not. And they can get help from that 22 utility to continue their construction project. They will 23 tap their resources to give them cable to finish the job. (]} 24 And if that is the case, then it seems to me that it is even 25 more important from the viewpoint of the general welfare of O /
176 {} the public that samples for test purposes could be made avail-1 2 able. And in general that would not require as much a sac-3 rifice from your stock as is the case to complete another O 4 fellow's project. But that has happened. I know it is a 5 matter of recrod. So I think it is a false situation. 6 MR. WARD: This may be something we need to take 7 to the commission in some way. I don't know. 8 MR. MC CULLOCH: That may come from the same source 9 as the component manufacturer. They are worried about research 10 tests being done, and the impact that that might have. And 11 whether they think about it on the surface, or it is subcon-12 scious, they are not inclined to be cooperative, because they 13 are afraid of the results. O(_/ 14 MR. WARD: Right. And I think that they have got 15 a valid concern. You get, as I said, into this alarmist 16 syndrome. But I think it is up to the staff, to the commission 17 it has to be, to develop some procedure that this whole thing 18 can be resolved. 19 MR. RAY: It seems to me that there should be cir-20 cumstances that the staff, by direction of the commission, 11 if necessary, would clarify to the utilitites that this is 12 a desirable situation that the samples be made available, 23 and so on, complete cooperation be involved, and that the () 24 results involved in those tests will very definitely not be 25 misinterpreted or deliberately misapplied, and so that there O
177 sn I is some reassurance to the utilities. There is reason for %J 2 them to be alarmed, and certainly on the part of the manufac-3 turers because of the tremendous investments they have. There ,,s 4 are too many examples in the past where misinterpretation or 5 let me say too fast reactions or results on the part of the 6 staff have imposed excessive requirements on the industry. 7 So the precidents are there to support this alarm. 8 It is my intention to prepare a report in response 9 to the things that have been brought up at these two sessions 10 at this meeting as a subcommittee report to the ACRS. Now II what action we can persuade them to take therefrom to the 12 Commission is something else. But that subcommittee report 13 will be in the public record. And the commission certainly nU 14 reads some of these things. So whether it is an official 15 IDC(ph) on the part of the full ACRS, or it is just a record 16 of what has surfaced in this meeting and therefore might stir 17 up the commission itself remains to be seen. But some of the 18 points that have been brought out, such as this one, will 19 definitely be in the report. And you will see the draft 20 before it is read to the committee at the next meeting. 21 Do you want to continue with your questions? 22 MR. WARD: I have got some more. Why don't we go 23 to somebody else's for a while. /] 24 MR. CROPP: Excuse me. I wonder if I might make 25 one more comment on that subject. We are working with the O \\,_/
178 slll 1 NRC right now to develop a set of procedures to do just what 2 you mentioned, to try and create an atmosphere where we can 3 get the information and assure that it won't be mistreated 7-) 'wJ 4 and yet get under way with these sorts of tests. So we are 5 working with the NRC. And I hope that something will be 6 established in the near term. We have made agreements with 7 some component manufacturers that essentially tell them, 8 look, we will give you the opportunity to see what we are 9 saying about your component before we say it-to the whole 10 world. If you have serious disagreements, whether we agree 11 with those disagreements or not, we will include comments 12 that you will provide in the report. And some of that begins 13 to work very well. p 'u / 14 MR. RAY: Good. Are you permitted to respond to 15 requests from industry, manufacturing industry, equipment com-16 ponent industry, materials, to make tests for them? 17 MR. CROPP: No. 18 MR. RAY: No, what? 'They don't come, or you are 19 not permitted? 20 MR. CROPP: No, we are not permitted to. We can't 21 do tests for a manufacturer. We do our work for the NRC. 22 MR. RAY: How about utilities? 23 MR. CROPP: No, the same for utilities. (x 't,) 24 MR. RAY: Same exclusion there? 25 MR. CROPP: We are not able to do tests for utilities. (3 ( )
179 1 MR. RAY: I would like to just take a couple of 7") v 2 minutes to summarize in a very brief manner, and maybe too 3 brief a manner, Mr. Bonzon, the story we got yesterday. And 4 in some respects it is emphasized again today. The impression 5 I have personally is that Sandia's present research in support 6 of safety-related equipment qualification programs have the 7 overall objective of assessing testing methodology for qual-8 ification by first testing materials used in equipment com-9 ponents for safety-related systems to establish the character-- 10 istics of response to the accident environments; and b, test-11 ing equipment components of safety-related systems that use 12 these materials. 13 In the second case, the equipment component testing O(_j 14 the project is in the analysis and planning stage and for 15 both electrical and mechanical equipment, sensors and so on. 16 General questions here -- perhaps you can't tell me, but I 17 wonder -- can you tell me how near completion is the materials-18 testing aspect of your program? 19 MR. Br.'IZON: I had a couple of comments that relate 20 to that. One thing I would like to preface it by saying l 21 research takes time. I don't think that is commonly understood l 22 an.ong the NRC staff or the industry,or anyone else. When you 23 do research you open other avenues. To directly answer your l (~)N question, we have not projected an end. What we have done 24 ( l 25 is said look, here are priorities we need to address. We have O l l
180 {p started on elastomeric materials because they are common, they 1 2 are ubiquitous, use in instruments and cables. We are moving 3 into things like seals and gaskets, which are still materials, LJ 4 but you add other stress factors, and they appear in other 5 kinds of equipment, and so forth. We are trying to look 6 into the aging of electronics, for example. 7 Your statement was generally correct. And I would 8 agree that you are right. We have attacked materials first, 9 to understand the methodology. I have referred to it many 10 times as we are testing the test, to distinguish from testing 11 the component or material. We are trying to develop method-12 ology. And we started as simply as we could because we didn't'. 13 understand the simplest systems. mV 14 Your second question had to do, I think, with equip-- 15 ment specifically. There I think what you saw yesterday, wheri 16 Dave Jeppesen described the EQRT program is that equipment 17 tests have been emphasized a little more in his program. 18 The NRC has said we need tests done on full pieces of equip-19 ment. And so our smaller subtask has sort of been we try to 20 do it in a companion way. We are using the same facilities. 21 We are using roughly the same people. So there is much overlap. 22 But we have kind of let his program handle the equipment 23 aspects, and let us concentrate on material aspects, just 24 because we have so many people and we need to divide the l 25 work. And all of it is important.
181 ejgg i One of the points I wanted to make is that there 2 is this kind of tie between the programs. And he is now mov-3 ing more strongly in equipment than we probably would have, 4 given our other priorities. So there is a definite need for 5 that kind of work. And it has been taken over in that progran. 6 MR. RAY: One of the supplementary questions I had 7 was in your area, that is the materials aspects, where do we 8 go from here. Now you have indicated your effort has been 9 to establish the priorities. Do you have a program outlined 10 to be subsequently followed in this sense? 11 MR. BONZON: The program that I gave you testerday 12 has all the tasks that we are currently pursuing. 13 MR. RAY: That have been identified to date, pre- ,(j 14 sumably? 15 MR. BONZON: Identified to date. There are a number 16 of other programs which are being talked about. One is a 17 program called -- we used to call it time-related degradation, 18 I believe Dave Okrent probably coined those words a few sub 19 commi. tees ago, a few committees ago -- where the purpose is 20 to take a broader look at the aging aspects of not just elast-- l l 21 omers, but corrosion -- I assume, pipe stresses. I am not 22 sure exactly what he had in mind. But we have been asked 23 very recently to do a broader look at the aging aspects of r[~') 24 time-related degradation. We are in the process of formulat-25 ing that program. Now that certainly is going to feed into b,n l I l
~ 182 (h I the existing programs in terms of priority and schedules. 2 And it is much broader than just electro equipment or just 3 elastomer materials. So that is another program which will 4 assist all of our programs in developing priorities. 5 MR. RAY: Do you have any specifically identified 6 seismic projects under way? We didn't get to talk too much 7 yesterday about this. 8 MR. BONZON: Seismic was a rather new invention for 9 us because we have had so many other problems to tackle. In 10 two ways I think we are moving into the seismic area. In 11 terms of Dave Jeppesen's program, on EQRT, it will be included 12 in the sequence of tests as you saw yesterday in terms of 13 components. 14 We are going to look at it more from a research ~ 15 point of view in terms of the effect of things like aging 16 ahead of the seismic event, and what effect does that have. 17 Now there has been a program under way sponsored by EPRI at 18 Wiley Laboratories to look at the correlation between aging 19 and seismic events. They have done quite a lot of work. The 20 first stage of that was to take components which they felt 21 probably would have no correlation and demonstrate that. Tha 22 report is in progress. They are going into a phase 2, as I 4 23 understand it, where they are looking at other components () 24 which it is not as clear whether there is a correlation. 25 We are moving in that direction. We do not have O 1
183 ) I (V a very firmly established program at this time. We have been ~ 2 asked by our NRC sponsors to think about it this year. It 3 is one of those essentially no-money kind of things. And 4 perhaps move into it next year. It has not been ignored, 5 it has been, like all other things, it has been said that that 6 is not the first priority we have to tackle. 7 MR. RAY: Except for preliminary contemplation, you 8 haven't specifically addressed it, as I see it. 9 MR. BONZON: That is right. No test where we have 10 included seismic, yet, at Sandia. Of course, as you know, 11 there are many programs under way in seismic. I think South-12 west Research has had a program with the NRC for a number of 13 years on seismic. () 14 MR. RAY: Yes, but I meant at Sandia. 15 MR. BONZON: None. 16 MR. WARD: Could I ask a related question? 17 MR. RAY: Sure. 18 MR. WARD: This question is probably coming partly 19 from my own ignorance of the situation. But, in developing 20 the testing methodologies and the sequences and the testing 21 scenarios, is there really a clear picture? Has the NRC de-22 fined a clear set of, you might call it, exposure scenarios 23 to which equipment should be qualified? I mean, this is com-(]') eletely independent of how you test it. Is it really clear 24 25 l what exposure scenarios we are taking about? Let's say for AV
1 184 gb 1 example all of them ought to start out with 30 or 40 years 2 of normal aging. And then you have a large-break LOCA expo-3 sure, perhaps, and then a hydrogen burn exposure. Another n(_) 4 leg of that might be the forty years of aging, and then an 5 exposure fire. Another one might be the 40 years of aging 6 and then seismic. And I don't know whether that then leads 7 to a LOCA and a hydrogen burn or not. 8 Is there a clear picture of what those exposure 9 scenarios are? 10 MR. BONZON: Let me say that the NRC has a definte 11 stand, and that is that it needs to be justified on a plant-12 by-plant basis. The problem in this country is that all 13 supposed standards or REG Guides which allude to these () 14 issues are generally tutorial in nature. Instead of saying 15 this is the sequence, these are the environments, it says 16 consider whether aging is important. Consider whether dose 17 rate is important, consider whether this is the right sequence. 18 You cannot find a prescripition. It has to be done. It has 19 to be justified on a plant-by-plant basis. Since it is on 20 the record, I hate to say this. But it is a bring-me-another-21 rock kind of problem, without telling you what color the rock 22 is, or what shape you would like. That is why in this country 23 we have tested and retested and rethought about the problem. 24 We have not a clear prescription. There is a danger, of course, 25 in clear prescription. That is, everyone does it wrong. AV
185 (])b 1 You can write it down. But unless you have basis for it, it 2 may be the wrong one. So everyone is doing it wrong, possibly. 3 But no, you cannot go and find numbers. What you 7ks 4 find for example -- 5 MR. WARD: That is the random concept of regulation. 6 MR. BONZON: Everyone is consistently wrong. 7 MR. WARD: Everybody can't be wrong, everybody 8 can't be right. 9 MR. BONZON: Let me give you an example of the 10 kind of things that are done. In the revision to REG Guide 11 1.89, in the source term -- accident source term, the numbers 12 are prescribed in terms of fission product releases to the 13 containment and to the coolant water and so forth. And they 14 are described as things like 100 percent of noble gasses, 15 and halogen percentages, and cesium percentages. And then 16 on a plant-by-plant basis you theoretically have to convert 17 that into doses and dose rates that you are given at equip-18 ment location and the like. Now, they give you an example, 19 but you cannot use that example without justifying that that 20 example holds for your plant. So there is no envelope. There 21 is no specific numbers without you going through the justifi.- 22 cation. I am not sure which side of the fence I sit on here. 23 On one hand, I understand that all plants are dif-(G 24 ferent. On the other hand, it would be a awfully nice if _/ 25 you could give very specific guidance as to how to do these (Z)
186 I things. I would say that the IEEE, who have been writing {) 2 standards in this area for years, are no better. They don't 3 provide you with any more guidance, and perhaps even less 4 guidance. They have, for example, tried to rewrite 323-1974 5 since 1977. And it is up to draft 13 or 14, depending on 6 whether you count 11(a) and 11(b) and 12(c). But it has been 7 through so many drafts, and yet it is not out. So it is 8 difficult to write a prescription. 9 I was commenting yesterday about the ease at which 10 foreign countries can do this kind of thing. They do write 11 prescriptions. And they do say this is exactly how it is 12 done -- not arbitrarily, they have done some work. But they 13 do write prescriptions. There you have a nationalized indus-O) (_ 14 try. You have a single reactor supplier in most cases, in 15 France in particular, you have a single reactor supplier. 16 So you can do that kind of prescription. 17 I am not sure you can write the prescription. But 18 it seems to me we could do a better job of being a little more 19 specific. 20 MR. WARD: This is a problem in more than this 21 context, but I am always, I guess, a little bemused by the 22 argument that we can't write a prescription, but we expect 23 the individual reviewer of an individual plant to, in effect, (]) 24 develop a prescription. It seems to me, Jerry, that that is 25 -- I tend to be simple minded an awful lot, but it seems to O
187 I me that is a fundamental flaw in this process and it continues sggg 2 confusion in the development of the whole issue. 3 MR. RAY: Yes, that is certainly worthy of comment. (,,'l \\ 4 Of course to realize the best of all worlds without impacting 5 the liberties under which we live, both in the conduct of 6 business and so on, is a difficult thing to do. 7 I might say somewhat in defense of the NRC staff, 8 that ACRS is on record quite frequently accusing them of being 9 too prescriptive, particularly from the viewpoint of impacting 10 design prerogatives of the utilities and the AEs and so on. 11 I would, however, concede that this is an area where perhaps 12 some degree or prescription would be justified. 13 MR. BONZON: I would also defend the staff, because (_/ 14 if I were forced to write down a prescription, I would have 15 a very difficult time of doing that. It is not simple. It 16 is plant specific. It is an individual problem. Plus, the 17 other thing, of course, that impacts this, in many areas you 18 cannot write it down because the research isn't there to say 19 hcw to do things. It just does not exist. The state of the 20 art is not -- you don't know how to humidity age or what the 21 sequences are, always conservative, if that is your goal. 22 It is impossible to do that. 23 The comment I made yesterday was we have to stay 24 with the state of the art if we are going to license plants. ) 25 I am sorry, but the state of the art exists now. You have ,-ms Y
188 I gb to go forward. You can't just say well, I will wait for the 2 research, or I don't know what to do. You must do something 3 and that something is do the state of the art -- recognizing ( ) 4 that as new information comes along, and research programs, 5 that might be changed. But that is the way life is. 6 MR. RAY: I don't envy the staff the position they 7 have, or the chore they have. On the one hand, being restricted 8 by Congress and so on and the needs to safeguard the public's 9 welfare, and on the other hand to be practical from the view-10 point of construction and design of nuclear power. They are II between the devil and the deep blue sea. 12 MR. BONZON: I would say our objective in life in 13 these research programs is to assist them in that area. What ( ) 14 we are trying to do is do the research and then hopefully trar s-15 late it into something that could be used. There, again, it 16 will never come out of, you know, do the sequence and you are 17 always right, but it will say, hopefully, how to judge what 18 sequence is appropriate, you know, this methodology handbook 19 I was talking about yesterday, that kind of approach, that 20 kind of guideline structure which is based on research would II then be useful. And people would not necessarily be following 22 the same sequence, but they would be following the same guidan ce 23 in how to develop the sequence. It is one step removed, but v) it is an important step. i MR. RAY: It at least emphasizes the factors that L.,
189 qb i should be considered. 2 MR. BONZON: Yes. 3 MR. RAY: And I am just as new at this activity as (~>3 's 4 David Ward. And my own reaction is -- and maybe this reveals 5 the lack of familiarity on my part -- I felt that your con-6 cepts in this respect are really trail blazing. And what I 7 am concerned with is that the information that is so potentia lly ~ 8 useful is made available to industry without restrictions. 9 That is why the question I asked earlier. 10 It would seem to me that in the materials area, in 33 the sort of trail blazing you were talking about that would 12 lead to the methodology handbook, there should be a minimum 13 of proprietary rights concerns. (,) 14 MR. BONZON: That is not a problem. 15 MR. RAY: And therefore this might very well be 16 universally available. 17 MR. BONZON: Yes, if I might comment on that. We 13 do not have the problem of proprietary information in terms 19 of materials research so much. Yesterday when I said we 20 divided our program between equipment and materials exactly i 21 for that reason, because on one hand it is easy to publish t 22 research things, especially when you are not dealing with 23 a component which passes for fails. On the other hand, when (' } 24 you deal with equipment you have this pass or fail problem. 25 So that has not been the problem at all. O
190 1 I have been in this business maybe too long, in 2 the sense that I become somewhat of a mother hen. When we 3 first did research which proved to me that things like dose O 4 rate and synergism is important, it became very frustrating 5 if people didn't jump on the bandwagon and say, boy, I agree 6 with that. Every test is now going to have an end. I have 7 since mellowed a little. And I am convinced that it isn' 8 truely an evolutionary process. You gradually pick this up, 9 and you gradually understand it a little more. So I am now 10 content with this evolutionary process. It wouldn't bother 11 me if we speeded the process a little, but it is still 12 evolutionary. 13 MR. WARD: You are just getting old. () 14 MR. BONZON: I may retire in this work, I am not 15 sure. 16 MR. RAY: The point was made in response to a. 17 question earlier this morning as to whether or not you are 18 restrained in providing service to manufacturing organization s 19 and utilities. I would feel -- maybe this should be a 20 question. Would this include responding to them by way of 21 advice? Now, that does not mean conduct tests. But should 22 they ask for advice on the procedure they should follow, or 23 the development they should pursue in preparing qualification (]) 24 tests in response to regulatory requirements? Would you be 25 restricted from providing that kind of advice? It is advisor t, () .--e
191 sb y now, and not specific response or conduct of tests. O 2 MR. CROPP: We answer phone calls and we try to be 3 as helpful as we can to whoever calls. We are not restricted in providing advice over the phone and those sorts of things. 4 5 We do it quite often. 6 MR. RAY: Very good. 7 MR. BONZON: I would like to respond to that too. 8 Our charter as a National Lab says we can't compete in the 9 commercial sector, it simply can't be done. I would hope 10 that that would get me a lot of friends in the commercial yy business because I am not competing with them. Unfortunately 12 they don't read that same line in our charter. 13 The second thing I would like to say about helping O i4 taauserv-eniax the dieseet ne1e we neve beea ia taauserv 15 is making these reports as widely available as possible, so 16 they have direct access to everything we know. As a matter 17 of fact, I think they probably know it better than I know 18 because they have time to read the reports, and I don't becaus e 19 I am busy answering their phone calls to say what shall I do 20 in this next test program. 21 We have in fact tried to somewhat discourage this 22 kind of consulting effort. About six months ago to a year 23 ago we were getting phone calls so often that we couldn't do 24 the work. Just literally people would call up and say, well, 25 what do you know about X material? You know, just the time O
192 (-5 1 factor alone prohibits us to do that kind of thing. G 2 But essentially what we have done is said, look, 3 here are the reports that you should be interested in. Either 1> 4 obtain them, or if we can help you obtain them, we will do 5 that. 6 But if we walk on both sides of the fence, I think 7 we are going to get in real trouble. And we have tried 8 desparately to avoid that. On the one hand telling NRC their 9 program is not good, and on the other hand telling them how 10 to do the program doesn't make much sense to me at all. 11 MR. RAY: I can understand the bind you are in. 12 But even just advising them of reports and the potential import 13 of those reports is a major service. And I would hope your' (m q,) 14 overhead includes the cost of such consultations. 15 MR. BONZON: Every year I put a little bit in the 16 budget which says general assistance. 17 MR. RAY: The last question that I have come here 18 this morning with design to ask, I would want to go into 19 executive session and close the record. So I would like to 20 go back to any other questions that can be discussed in open 21 session. 22 MR. BONZON: I have written down on a piece of paper, i 23 if you are running out of things to talk about, just some 24 l (} thoughts I have had over the years, and I would be happy to 25 throw them out on the floor, or whatever, if you see it as O v
193 1 appropriate. 2 MR. RAY: Lloyd, we are willing for more consulta-3 tion or more commentary on the part of Sandia, go ahead. 4 MR. BONZON: These are primarily philosophical in 5 nature and have no answer. But I would like you to consider 6 them. 7 On thing I think again I would like to emphasize 8 is that this is an evolutionary process, and research takes 9 time. You cannot throw an infinite amount of money at this 10 problem and solve it instantly. It is not clear to me that 11 is universally understood. 12 Another point I would like to make is that what I 13 see come out in terms or studies and work is there have been 14 too much emphasis on word rewording. There have been too 15 many studies which try to take the same data and massage it 16 some way or another. And there has not been enough emphasis 17 on the research which backs up the words. 38 I would say there is probably an insufficient amount 19 of coordination among the NRC groups even within research, 20 but certainly across research NRR. I think this has been a 21 universal problem. It has been there a long time. We are l 22 not going to solve it. As examples of that, some of the 23 things we have encountered at Sandia that we have not very {} 24 often been consulted by NRR or, in some cases, research. 25 One classical example is when we had the test of the penetrat Lon O
194 (5 1 which I showed you, the D.G. O'Brien and Duke Power penetration. %) 2 As a result of our reporting and as a result of our test, where 1 3 we had some failures, there was a meeting between Duke Power 4 and NRC staff on what to do next. Somehow it would seem that 5 that would be a reasonable place for Sandia to be involved, 6 at least to explain what we did in our test, so that one 7 could then see how to proceed. We weren't involved in that. 8 I think maybe that would have been a place to get involved. 9 We have some trouble with direct access to NRC infor-10 mation even though we work for NRC. It is pretty hard for 11 us to get directly things like 79-01-B responses that are 12 coming through in terms of what equipment is there and what 13 is being qualified and how it is being done. () 14 We have had some trouble in getting access directly 15 to the data bank that NRC has established on equipment qual-b ification through Franklin Labs. They have helped us out on 17 occasion. But I am thinking that maybe some kind of a tele-18 phone link to that would be appropriate. 19 Our discussions with NRR have been primarily the 20 formal route. By that I mean we write a report. Perhaps a 21 a real research information letter is written in a very 22 formal way. But it seems to me maybe we should have a little 23 more informal discussion. Because what happens is when a (]} 24 research information letter comes over, it is a very blunt 25 instrument. Research says you shall be concerned about this. O
195 1 {} And sometimes it takes our friends at NRR by surprise, in 2 effect. And that creates very bad feelings. 3 This terminal block issues was a classic example. 4 They were suddenly faced with -- the terminal blocks of the 5 Three-Mile Island that I showed yesterday, with the probability 6 curves. 7 MR. RAY: May I interrupt? 8 MR. BONZON: Sure. 9 MR. RAY: What about the project managers on the 10 NRC staff? Don't they stay in current status with you as your 11 research projects progress? 12 MR. BONZON: Another point I was going to make along 13 those lines is our research project manager has established ("% (_) 14 a, quote, " review group," composed of people from various 15 organizations which should be interested in this to serve 16 as liason between research and NRR or IE and whatever organ-17 ization they come from. Work assignments and pressures and 18 the like don't make that group as effective as they might 19 overwise be. Plus I think the people involved in their review 20 group are not at a high enough level. In other words, you 21 have staff members talking to each other. And then of course 22 they talk to their management. But they are not in a decision-l 23 making process. They have other work assignment. So the (]') review group does not work as well as it should. In principle 24 25 it is a very good idea, because that should serve as a liason I (-sJ
196 sb i so that everyone understands what Sandia is doing, or any 0 2 other group, understands the results that might come and 3 affect them, whether it is terminal blocks or O'Brien penetra-() tions or whatever -- make them aware before it happens, 4 5 rather than come as a surprise and force our friends at NRR 6 into a defense posture rather than a sharing-of-information 7 posture. 8 I am not sure that answers your question directly, 9 but there is this group formed to do the kind of liason I 10 think you were alluding to. gy MR. RAY: Yes. I think what you are saying, 12 without being explicit is that they aren't doing their job. 13 MR. BONZON: No, I am saying that there are many () 14 jobs to be done. And this one maybe doesn't take -- doesn't 15 have as much priority as it should, because what happens is 16 when something comes up, which is difficult to handle, then 17 it comes as a surprise. If the review group functioned 18 with more fluidity, they would be aware that we are doing 19 work in this area, and that they maybe should get their ducks 20 in a row ahead of the report coming out. 21 MR. RAY: Would you agree with me if I said that 12 they are ineffective? 23 MR. BONZON: I would agree they could be more S 24 effective. No, I would not agree they are ineffective. I (V 25 think that is too strong a statement, because the group does O m w-
197 1 work and it does function. It does not function as effectively (('T / 2 as -- 3 MR. RAY: Okay, we will banter with semantics for 4 a few minutes. If I were to say that could be more effec-5 tive, you would be in a position, should that elicit a 6 response, and they were to come to you to say what can we do 7 to be more effective, you would be in a position to respond? 8 MR. BONZON: I would agree with that statement, yes 9 And I think it is to everyone's advantage. This 10 is the point I want to make, is that we need to provide infor-11 mation to our sponsors. And our sponsors are not only research 12 from which we get our money directly, but ultimately the 13 people in NRR are requesting this. work to be done in other () 14 groups. And so they need to know ahead of time that some 15 important issues are coming by, or some new research is com-16 ing by which might affect their decision-making process. 17 This is all I am trying to say. I am just trying to make 18 this transition not so abrupt. Because the abruptness gets 19 us in trouble. Sandia gets in trouble because NRR doesn't 20 understand that we are even working in this area. And they 21 don't understand the implications. And the problem is when 22 the report is written it is for the general public. The 23 intervenors read it. NRR reads it, but at the same time. So you have the immediate confrontation problem.- So I'a,m 24 25 just suggesting that it could be more effective.- O
198 s 1 MR. RAY: Lloyd, I realize by asking the typ; of 2 question I have been asking I may be putting you people on f-3 the spot. But in the American system the ACRS is chartered ( 4 to advise the Commission. And of course there are many very ' -/ 5 important aspects of that advice and areas of advice that 6 have higher priority, perhaps, than what w' are talking about 7 But nevertheless, this is something that it seems to me is 8 worthy of being addressed and corrected to whatever extent 9 one can implement that correction. 10 So sometimes addressing the thing that requires II advice or merits advice will open up a next of rats. And 12 the irritation that that results in can be a problem. It 13 is not really intended to create those problems that we ask 7 (_) 14 the leading questions we ask. But I would say that the response 15 you made is the kind of response that needs. 16 MR. BONZON: My job is to disseminate this infor-17 mation, I think. This is the principal thing. It seems to 18 me that in any setting the people here should have a least 19 first access or immediate access to the people that are spon-20 soring the work. That is my point. And I would like to 21 create that situation by a freer exchange among all parties. 22 I think this review group process in principle is 23 a good idea, and it is a mechanism that we can use perhaps 24 (} more effectively. 25 The other problem I have noticed is the words, t \\_/
199 b 1 " equipment qualification" are very exciting just now, as 2 seismic was a few years ago, and so on. There is always a 3 hot issue. Because of that, there are an awful lot of studies ()# 4 going on in the United States sponsored by different groups. 5 The NRR has certain amounts of money that they can use in 6 terms of technical support and the like. Research is doing 7 some work. Various branches within Research are doing some 8 work. There seems to be some artificial distinctions between 9 mechanical and electrical equipment, you know, and that sort to of thing, a certain branch handles one. Another branch 11 handles another. One of the things I worry about is is there 12 enough coordination among these groups. 13 Aging, as an example, is a problem which transcends () 14 all these boundaries. It is not unique to electrical equip-15 ment or mechanical equipment. And yet we somehow have 16 artifically said mechanical equipment is in this domain, 17 electrical equipment is in this domain, and so forth. And 18 I worry about that kind of coordination. Someone needs to 19 look a little more strongly. I spend some of my time reading 20 reports by other national labs, other organizations who are 21 doing work for the,NRC in branches which are not research. 22 They come from different organizations. And I look and them, 23 and I say, gee, that is kind of what we are doing. And I () wonder what the difference is. And the differences primarily 24 25 are they are funded by another organization, and they are O
200 1 designed to answer, in most cases, very immediate questions (-) V 2 -- what am I going to do about a source term? That kind of 3 very spectilc thing. O 4 I get a call from Oakridge Laboratory,or whoever 5 is doing the work, saying~, gee, what do you know about source 6 terms? And it seems funny that this is happening to me. I 7 think a little more coordination there might help. 8 So I throw these out. I have no answers for most 9 of these. But these are some of the things I have observed to over the years in this area. Perhaps you can just take them 11 with a grain of salt, and see if any of them appeal to you. 12 MR. RAY: I thank you for your comments, Lloyd, 13 and they are the sort that we need. And they won't be () 14 dismissed lightly. 15 MR. CROPP: I wanted to make one other statement 16 concerning a question that was asked yesterday by Mr. Ward. 17 MR. RAY: Mr. Cropp, you feel free. 18 MR. CROPP: The question was asked concerning the 19 igniter, some of the Sandia concerns about igniter placement 20 and whether or not our concerns had been fed back to NRR 21 and then therefore back to the licensees. And at the time 22 we indicated that we weren't certain whether that had happened 23 or not. 24 (]) I did some checking with some of our other people, 25 and that certainly has happened, on a rather timely basis, i
201 ~sb 1 as a matter of fact, in two instances. So we have had a very 2 close coupling with NRR and the licensees concerned in 3 those areas. 4 MR. RAY: Thank you. 5 I think for the record we can terminate it. We 6 will go into executive session to continue the discussion. 7 And we would like it not trnascribed. Thank you. 8 Off the record. 9 (Thereupon, at 10:40 a.m., May 20, 1982, this 10 record was closed.) i 12 13 () 14 15 16 17 18 19 20 l 21 l l 12 23 ) 25 ( l l l l
NUCLEAR REGULATORY COMMISSION This is to certify that the attached proceedings before the () ACRS subcommittee meeting on Qualification Program for Safety Related Equipment in the matter of: Date of Proceeding: May 20, 1982 Docket llumber: Place of Proceeding: Albucueraue, New Mexico were held as herein appears, and that this is the original transcript thereof for the file of the Commission. Sheila Bjornlie Official Reporter (Typed) s v f Official Repor Signature) . -.}}