ML20209J222
| ML20209J222 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 09/11/1986 |
| From: | Atomic Safety and Licensing Board Panel |
| To: | |
| References | |
| CON-#386-743 LRP, NUDOCS 8609160212 | |
| Download: ML20209J222 (182) | |
Text
O R/G WAL o
UNITED STATES NUCLEAR REGULATORY COMMISSION I
I I
IN THE MATTER OF:
DOCKET NO:
LRP INQUIRY INTO THREE MILE ISLAND UNIT 2 - LEAK RATE DATA FALSIFICATION O
I LOCATION:
BETHESDA, MARYLAND PAGES:
1006 - 1184 DATE:
THURSDAY, SEPTEMBER 11, 1986 O
ln(/
l 0(
ACE-FEDERAL REPORTERS, INC.
O CVfim Reporters l
444 North CapitolStreet Washington, D.C. 20001 8609160212 860911 PDR ADOCK 05000320 T
PDR NATIONWIDE COVERACE
1 CR28039.0 1006 BRT/sjg 1
UNITED STATES OF AMERICA 2
NUCLEAR REGULATORY COMMISSION 3
BEFORE THE ATOMIC SAFETY AND LICENSING BOARD 4
- -x 5
In the Matter of:
Docket No. LRP 6
INQUIRY INTO THREE MILE ISLAND UNIT 2 - LEAK RATE DATA FALSIFICATION 7
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _x 8
i 9
Nuclear Regulatory Commission Fifth Floor Hearing Room 10 East West Towers 4350 East-West Highway 11 Bethesda, Maryland 12 Thursday, September 11, 1986 s
,)
13 The hearing in the above-entitled matter convened'at 4
8:30 a.m.
15 16 BEFORE:
17 JUDGE JAMES L.
KELLEY, Chairman Atomic Safety and Licensing Board ig U.S.
Nuclear Regulatory Commission Washington, D.
C.
JUDGE JAMES H.
CARPENTER, Member Atomic Safety and Licensing Board 20 U.S.
Nuclear Regulatory Commission Washington, D.
C.
JUDGE GLENN O.
BRIGHT, Member 22 Atomic Safety and Licensing Board U.S.
Nuclear Regulatory Commission 23 Washington, D.
C.
24
(/
25 1
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
1007 1
APPEARANCES:
2 On behalf of GPU Nuclear Corporation:
3 ERNEST L.
- BLAKE, JR.,
ESQ.
JOHN NASSIKAS, ESQ.
4 Shaw, Pittman, Potts & Trowbridge 1800 M Street, N.W.
5 Washington, D.
C.
20036 6
On behalf of the Employees:
HARRY H.
VOIGT, ESQ.
7 MICHAEL McBRIDE, ESQ.
LeBoeuf, Lamb, Leiby & MacRae 8
1333 New Hampshire Avenue, N.W.
Suite 1100 9
Washington, D.
C.
20036 10 On behalf of Jack Herbein:
11 JAMES B.
BURNS, ESQ.
Isham, Lincoln & Beale 12 Three First National Plaza Chicago, Illinois 60602
/
13 CHRISTOPHER W.
FLYNN, ESQ.
~
Isham, Lincoln & Beale 73-1150 Connecticut Avenue, N.W.
Washington, D.
C.
20036 On behalf of Gary P.
Miller:
MICHAEL W.
MAUPIN, ESQ.
17 M.
CHRISTINA HENSLEY, ESQ.
Hunton & Williams 18 707 East Main Street Richmond, Virginia 23221 19 On behalf of Former Metropolitan Edison Employees:
20 SMTIH B.
GEPHART, ESQ.
21 Killian & Gephart 217-218 Pine Street 22 Box 886 Harrisburg, Pennsylvania 17108 On behalf of the NRC Staff:
24 JACK R.
GOLDBERG, ESQ.
,-m
(/
25 MARY E.
WAGNER, ESQ.
U.S.
Nuclear Regulatory Commission Washington, D.
C.
20555 ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80 4 336-6646
1008
$-1 C0NTENTS 2
RECESS:
3 NOON - 1125 4
5 llELEllE 6
NUMBER IDENTIFIED RECEIVED 7
Board Exhibits ll-A and 11-B - Effect 1025 of oscillations 8
Exhibits 10-A and 10-B 1027 9
Exhibit 9-G - Photograph 1183 1183 11 12
'V 13 14 15 16 l
17 18 19 20 21 22 23 l
24 im k) 25 l
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80 4 336-6646 j
28039.0 CRT 1009 U
1 PRO CEEDINGS 2
JUDGE KELLEY:
Good morning.
We indicated when we 3
stopped last night a number of items to begin with this 4
morning before returning to questioning, a couple of items.
5 Let me return to proposed Exhibit 6-B, a letter from Mr. Ben 6
Hayes, director, Office of Investigations, to the 7
Commissioners.
This exhibit was proposed by Staff counsel 8
and opposed by most of the other parties.
The Board is 9
excluding from the record Exhibit 6-B, proposed Exhibit 6-B, 10 for the reasons that I'll state.
11 First of all, we certainly can understand why 12 Staff counsel would offer this memorandum as an exhibit under 7-V 13 the circumstances that Mr. Goldberg described.
They Delieved 14 that it was in the record and described the circumstances 15 underlying that.
Furthermore, the record, as it stands, or s
16 as i t stood before yes terday, is misleading in the sense that 17 we have in evidence the OI report and supporting materials 18 and that IC report states, on page 1,
that it was approved by 19 Mr. Hayes. So one would infer, in the absence of proposed 20 Exhibit 6-B, that that report stood in its entirety and 21 reflected the opinion of the Office of Investigation.
22 It seems to us that the record we've made now is 23 at least not misleading in that sense.
We now know that 24 there is a separate document reflecting a disagreement in r
25 certain very narrow respects from some conclusions reached in t_'sJ ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
28039.0 e 7RT 1010 d
1 the report itself.
So at least that fact has been reflected 2
by this discussion.
3 Those considerations to one side, the following 4
factors are controlling, in our view.
First, the report, if 5
admitted, we think, could be prejudicial to certain parties.
6 We think that's apparent on the face of the report.
We are 7
not going to belabor it, so as to avoid the possibility of 8
compounding whatever prejudice there may be.
9 Secondly, and very important in our view, I think 10 really crucial, the memorandum adds absolutely nothing in the 11 way of new information on the issues before us.
The opinions 12 reflected in the memorandum are, apparently, based entirely CJ 13 on the report that is already in the record.
Further on that 14 point, the memorandum, we believe, is entirely conclusory as 15 distinguished from reports in the record, such as the Stier 16 report, the Faegre & Benson report, the NRR report, which 17 include conclusions but which are mostly factual or 18 analytical.
There's no facts, there's no analysis, there's j
19 just a couple of bottom lines in the memorandum.
l 20 We would like to note that we previously excluded 21 another document -- the exact title I'm not clear on, but I'll call it the GPU assessment panel report -- is that 22 l
23 close, Mr. Blake?
24 MR. BLAKE:
Yes.
p3 25 JUDGE KELLEY:
And the reason we kept it out,
'd ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmcrage RX)-33MM6
28039.0 r"RT 1011
()
I basically, is because there was no new information in it.
It 2
was just somebody else's analysis of facts derived from other 3
sources.
So, it being itself all conclusion and no new 4
information, we thought it appropriate to exclude it.
These 5
same reasons were largely the reasons that we do not believe 6
calling a sponsoring witness would materially assist the 7
record.
8 The board itself can assess the OI materials that 9
are in the record, with the assistance of Mr. Christopher, 10 the person who was principally responsible for developing the 11 report.
12 There is a separate element of possible either
(-
13 prejudice or delay, which we think weighs in the balance.
14 The memorandum did surface late, lor.g after the prehearing 15 record had been compiled and after the filing of prefiled 16 l testimony.
If we were to admit the report, fairness would 17 probably require that the affected employees at least be 18 given some further time to consider their positions, perhaps 19 file additional testimony, and that could delay, disrupt the 20 orderly progress of the hearing that we have underway.
21 The separate important consideration that we want 22 to speak to is that we do not believe that the mere fact that 23 this memorandum has been introduced, and we, the parties, 24 others, have read it, itself is a basis for significant i
(~g 25 prejudice.
And we believe that for several reasons.
We all
%)
r 1
i l
ACE-FEDERAL REPORTERS, INC.
202-347 37(U Nationmde Coserage 8i0-336-6646
28039.0 r7RT 1012
(
)
v 1
read it once.
We knew what was in it.
If you ask us 2
tomorrow, it seems to us -- frankly mine has faded a lot 3
already today.
4 One of the reasons is there aren't any faces to go 5
with these names, yet, for us.
These are names, some 25 6
names -- 24 parties, I believe, that are involved.
As we've 7
read their testimony we are going to have to focus closely on 8
their testimony before the individuals come in, because I 9
personally just don't have a fix on individuals such that, in 10 reading a memo like this, it is going to make any impression 11 on me at that point.
12 Perhaps more significantly, as I said before, this 7-( 'l 13 goes to our reaction to it, since there isn't any information 14 in this report, there's no information, no reasoning, there's 15 just a conclusion, that's just not the kind of thing likely 16 to make an impression on a reader.
If there were some new 17 fact, then I suppose that might stick, but there aren't any 18 new facts in this memo.
19 So, again, it's a circumstance that goes to the 20 fact that we believe that, having read this memorandum once 21 and now proposing to return it to the Staff, we don't believe 22 that it is going to have any effect on our analysis of the 23 record and our findings.
In short, we'll put it out of our 24 minds.
25 MR. MC BRIDE:
Judge Kelley?
u("%
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cos erage R&336-6646
l 28039.0 e-7 1013 l
N_]RT l
1 JUDGE KELLEY:
Just another comment, f
2 To sum up, an issue of this kind involves a 3
balancing of a number of relevant considerations, subsumed 4
under the rubric of possible benefit of admitting en the one 5
hand or prejudice or delay on the other.
Here we see no more 6
than a small benefit from admitting this exhibit, even 7
assuming it may be accompanied by a sponsoring witness.
We 8
weigh that against a real prospect of substantial harm and 9
reach our conclusion to exclude.
The proposition was put 10 yesterday about returning the copies, and perhaps we should lspeaktothat for a moment.
There may be some logistical 11 12 problems.
g-D 13 Mr. McBride, do you have a comment?
14 MR. MC BRIDE:
Not until you have said what you 15 had to say about the copies.
16 JUDGE KELLEY:
Let me ask for a counsel 17 proposition, given where we are now.
18 Mr. Goldberg, what do you propose?
19 MR. GOLDBERG:
I propose that the copies which 20 were distributed the day before yesterday be returned as was 21 requested, and some counsel and a few other individuals who 22 received it have already done that.
But not everyone, I 23 guess.
24 We are in the process of --
r~s 25 JUDGE KELLEY:
The day before yesterday -- just to V
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage M3 M-6t46
28039.0
(~7RT 1014 d
1 keep straight, that's a somewhat different version than was 2
offered yesterday.
3 MR. GOLDBERG:
That's right.
That has one 4
additional sentence in it, compared to the actual Exhibit S
6-B, which we proposed be included in the record, and I don't 6
see any reason to not have that copy returned in the interest 7
of the individuals who are named in the letter.
That's the 8
only reason for its return.
There is no reason that I see 9
why that information ought to be disseminated.
10 JUDGE KELLEY:
So you are asking that all copies 11 be returned to Staff counsel?
12 MR. GOLDBERG:
Yes.
13 JUDGE KELLEY:
Right.
Okay.
14 MR. GOLDB8RG:
Of the version which was 15 distributed the day before yesterday.
16 JUDGE KELLEY:
Okay.
17 MR. GOLDDERG:
Proposed Exhibit 6-Bcan be 18 retained.
19 JUDGE KELLEY:
Oh, it's only that sentence you are 20 really concerned about?
21 MR. GOLDBERG:
Yes.
Yes.
If they want to give 22 them all back I'll take them.
23 JUDGE KELLEY:
There's no discern over yesterday's 24 distribution as far as you are concerned, at least.
25 Mr. Gephart?
O l
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nanonwide Coserage fuo-33MM6
a 28039.0 1RT 1015 1
MR. GEPHART:
Judge, we would, of course, support 2
Mr. Goldberg, at least partially, as returning the day before 3
yesterday copy.
That means to be returned by all parties 4
with the exception of counsel for the affected individuals.
5 We would, as counsel for the affected individuals, however, 6
would retain our copy of the transcript.
I understand that 7
was inadvertently placed in the transcript two days ago, so 8
that our transcript would be identical to everybody else's.
9 JUDGE KELLEY:
Can we separate out the two 10 versions as to your request there.
The day before 11 yesterday's version --
12 MR. GEPHART:
Was the more complete version.
O 13 JUDGE KELLEY:
As to that one, do you have the 14 same proposition as to both?
l 15 MR. GEPHART:
Well, as to that one we would 16 support Mr. Goldberg, that everybody return it with one l
17 exception.
We are counsel for the affected individuals and l
18 we would propose that we retain our copy.
We see no reason 19 why we shouldn't.
20 JUDGE KELLEY:
I understand your proposition.
21 Okay.
22 MR. GEPHART:
As far as yesterday's copy, the more i
23 complete -- or less complete version, the additional sentence l
24 being deleted, we feel that also should be returned to 25 Mr. Goldberg by all the parties with the exception, again, ot O
t ACE-FEDERAL REPORTERS, INC.
202-347-3NU Nationmide Coserage W 14tM6
~
28039.0
.(~]RT 1016 V
1 the counsel for the affected parties.
2 JUDGE KELLEY:
You say put it back out of 3
circulation by giving it back to the Staff except you want to 4
keep a copy of each one?
5 MR. GEPHART:
Yes.
If the board is interested in 6
our reasoning, I would suggest that rather than do it on the 7
record I think it's really a matter between Mr. Goldberg, the 8
board, and us, as counsel.
I prefer to do that in camera, 9
and explain to the court our reasons for this.
10 JUDGE KELLEY:
We'll just go around the table and 11 get everybody's position on this.
12 Mr. Blake?
13 MR. BLAKE:
Judge Kelley, one, I think that the 14 record should clearly provide that this particular item never 15 even arrived at the " identified" stage as an exhibit in this 16 proceeding and, therefore, for that reason, ought not 17 accompany this record at all.
That is, there's no 18 obligation, no legal requirement it accompany it, even as an 19 identified exhibit.
It was a proposal.
There was going to 20 be a suggestion.
21 If you look at the transcript of the proceeding 22 two days ago, you find that my discussion of Exhibit I was 23 picked up in the index and is a modified exhibit, and 5, 24 similar.
But 6, while it was discussed as a name, never 25 became, even, an identified exhibit.
So there is, in my O
ACE-FEDERAL REPORTERS, INC.
20244747tt)
Nationaide Cos erage RO 33N
l 4
t 28039.0 (7RT r-1017
)'
1 view, no reason for this document to accompany this record in 2
any way, shape, or form.
3 JUDGE KELLEY:
That would leave us with just the 4
transcript discussien of some document.
More than "some"; we S
said it was a memo dated April 11 and so on.
6 MR. BLAKE:
That's right.
But there's no reason 7
for the document to accompany the record either as admitted 8
or as an identified exhibit.
I 2egard the incorporation of 9
the document in the transcript as no more than a " goof," and 10 the boa 2d certainly directs that, as a transcript correction, 11 that documes.t be removed f rom this transcript.
12 Beyond that, I really don't see the board's s]
13 involvement here.
I don't see a need for board involvement.
14 I don't see that you have yet a suggestion to Mr. Goldberg, 15 but 1 don't see hin requesting some board directive to the 16 parties or what not, what to do with the documents, some 17 oucht to keep them or what not.
Frankly, I don't see this i
18 document at this point plays a role in this proceeding and I 19 don't see that we need to entertain arguments and get some l sort of board direction or order about what we do now with 20 a
21 copies of a dccument which plays no role in this proceeding.
22 JULGE KELLEY:
Well, your idea would be that by 23 taking off the exhibit number, even, it never got beyond some 24 preliminary _ stage; is that right?
l 25 MR. BLtKE:
That's right.
It was proposed.
From I
ACE-FEDERAL REPORTERS, INC.
21C.W-3'm Na:ionw kie Coserage 800-336-6646
28039.0 r 1RT 1018 N]
1 the transcript it isn't even identified as a number in the 2
index of the transcript.
If it were, then I'd have trouble.
3 With identified exhibits, even though they are rejected, they 4
follow the record, and I'm suggesting that is not -- this one 5
never got to that stage and the record ought to clearly say 6
so.
7 JUDGE KELLEY:
Let me ask Mr. Goldberg's reaction 8
to that particular suggestion.
9 MR. GOLDBERG:
I don't object to that, for a 10 couple of reasons.
First of all, this is not a proceeding 11 under part 2.
There are no matters in this proceeding.
12 There's no appeal board.
The only thing tha t happens af ter f,
t1 13 this record is closed and the board reaches its findings, the 14 Staff makes recommendations to the Commission on enforcement 15 action.
We are the only proponent about that document.
We 16 are not going to complain about its exclusion.
So there's no 17 need to have it identified as an exhibit for any appellate 18 purposes in this proceeding.
So I don't object to 19 Mr. Blake's proposal.
And, as far as the return of the 20 documents from the other counsel, if they all agree to return 21 both versions I'll gladly accept them.
I am not going to 22 force it on anyone.
I don't object to counsel for the 23 numerous employees keeping a copy of the letter.
24 JUDGE KELLEY:
All right.
Do other counsel have 25 comment?
Mr. Burns?
(-)/
x_
ACE-FEDERAL REPORTERS, INC.
202-347 37(x)
Nationwide Coverage 8(XL336-6646
l f
28039.0
<-9RT 1019
/)
%)
1 MR. BURNS:
I believe we have already returned a 2
version of two days ago and we have no objection to returning 3
the other, 4
JUDGE KELLEY:
Mr. Maupin?
5 MR. MAUPIN:
I haven't anything to add to what 6
Mr. Blake said and I subscribe to what he said.
7 MR. MC BRIDE:
Judge Kelley, I do not subscribe to 8
what Mr. Blake said for this reason.
If Mr. Burns or 9
Mr. Maupin had handed this document around, I might very well 10 have'to agree with Mr. Blake that the board would have no 11 power over them to direct them to do something about a i
12 document that had never become part of this record.
- However,
',)
\\
13 the disseminating party was the Staff, which is not a party 14 to these proceedings, but an adjunct of the board in this 15 proceeding.
And in that role, as distinct from the rest of 16 us at this table, I think the board has some power over the 17 Staff.
And, where this matter has already worked prejudice 18 to our clients, simply because their employer has been made 19 aware of what that memorandum said by its dissemination to 20 its employer's counsel -- and that's not any criticism to 21 Mr. Blake, he's got a client to represent and his client was f 22.
made aware of this because he informs his client, presumably, 23.
of what goes on in these proceedings, as the rest of us do, j
24 too.
And it has worked that kind of prejudice, so I think fs 25 the Court ought to, as it has power over the Staff as an (J
ACE-FEDERAL REPORTERS, INC.
202 347-370)
Nanonwide Cos crage SE346M6 l
c l
28039.0 1020
(~)1RT 1
adjunct to the board in this proceeding, should do what 2
little it can, but it ought to do what it can, to some 3
extent, to ameliorate that prejudice.
The order ought to be 4
to direct all parties other than counsel for the affected 5
individuals to return all copies'of the document in whatever 6
form.
7 JUDGE KELLEY:
All right.
There's some agreement 8
here.
You can keep a copy.
I think that's established.
9 Just so I understand, or make sure I understand your 10 rationale for your approach, you are saying:
Take it out of 11 the case in an official way to the extent that you can and 12 that is because it's prejudicial and just shouldn't be in the O
13 case?
14 MR. BLAKE:
Yes, sir.
15 JUDGE KELLEY:
Well, we'll sort this out, 16 gentlemen.
And if it has to sit until tomorrow, I guess 17 that's all right.
We'll get to it.
It is clear that counsel 18 for the numerous employees can keep their copy.
19 MR. GOLDBERG:
One other comment, not on this 20 particular dispute, but one other comment on something you 21 said in the course of ruling on the exclusion of this 22 document.
You indicated that the GPU assessment panel report 23 was excluded for the same reasons.
That's generally 24 correct.
There was one page of that report, page 12, which
(-
25 was accepted.
v)
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage Mn336-6M6
23039.0 g-}RT 1021
\\~.)
1 JUDGE KELLEY:
Yes.
That we did with Mr. Herbein, 2
I think.
Nobody objected; is that right?
Everybody said 3
that was okay?
Okay.
Is there -- I'm sorry.
4 MR. GOLDBERG:
No -- just that.
5 JUDGE KELLEY:
Okay.
That covers that.
6 We had another point that arose yesterday 7
afternoon, in the evolving role of the Staff in the case, 8
what Staff counsel can or can't do.
The question in the 9
morning about objections, and we ruled, basically, that the 10 Staff could object to questions put to its witnesses but not 11 to the other witnesses for reasons I won't elaborate again.
12 Then we had an objection, in the afternoon, which 01
\\~#
13 essentially was that the Staff should not be allowed to 14 propose questions.
And that came up in the context of 15 several proposed questions from the Staff.
And we have 16 considered the points made about that by respective counsel 17 and we have decided on a symmetrical piece of guidance.
That 18 is to say, the Staff can propose questions to be put to Staff 19 witnesses, but the Staff, for much the same reasons that 20 apply to objections, should not be proposing questions to the 21 other members of the panel.
22 It seems to us that that by no means has the 23 effect of unduly shutting the Staff out of the process.
24 After all, we have these ongoing discussions among all 25 witnesses on all these questions.
The Staff witnesses have a g-u ACE-FEDERAL REPORTERS, INC.
202 4474700 Nationwide Coserage R33.1M486
28039.0
,e9RT 1022 N.]
1 fair opportunity to comment on other people's points.
At the 2
same time, Staff counsel may feel that the Staff perspective 3
on something didn't quite come out, and by proposing a 4
question to them, it allows them a way to bring that out.
So 5
we think that's a reasonable, practical approach.
It is 6
simple and there's some virtue in that.
So that's the 7
approach we are going to take.
8 MR. GOLDBERG:
Judge Kelley, I would like to 9
comment on that, if I may?
The Staff will have witnesses in 10 this proceeding for only a very small percentage of the 11 time.
For most of the proceeding there will not be any Staff 12 witnesses on the stand.
Based on your rulings, during the O
13 large majority of this hearing, then, the Staff doesn't have 14 a right to object to questions, doesn't have a right to 15 propose that the board ask the witnesses questions.
In light 16 of those rulings, I don't see any meaningful role that the 17 Staff can play in this proceeding, other than when Staff l
18 witnesses are on the stand.
19 JUDGE KELLEY:
Well, the ruling certainly seems to i
20 fit this context.
We need not r'each right now beyond -- by t
l 21 "this context," I mean this panel of experts.
It need not l
i 22 reach, now, beyond that.
We can give some further 23 consideration to comments you juct made in the light of a 24 different context.
25 MR. GOLDBERG:
Well, for example, one thing you n
ss I
i ACE-FEDERAL REPORTERS, INC.
I l
m,.3-
~_ m_,.
l I
28039.0 1023
(-)]RT l
\\.
l 1
ought to consider in deciding is when we move beyond this 2
panel. 'With the exception of Staff counsel, every counsel in 3
this proceeding has, basically, the same interest, and that 4
is exonerating the employees from any misconduct.
5 JUDGE KELLEY:
Why don't we get to the argument 6
when we get to the problem.
7 MR. BLAKE:
That's not a fair characterization.
8 JUDGE KELLEY:
Gentlemen, all I'm suggesting is we 9
have these people here, let's do business.
We have given you 10 a rule that will suffice for present purposes and I'm happy 11 to hear further argument at a later point.
But let's go 12 ahead for this morning with the rule that applies to panels n'
13 of expert witnesses.
14 The Staff might have had some comments on the 15 visual aid placards that we were looking at yesterday from 16 Mr. Moore.
The proposition was to admit them as exhibits.
17 Mr. Goldberg, were there any objections to those?
18 MR. GOLDBERG:
In fact it turns out there is no 19 objection from the Staff.
I guess, since it's not a Staff 20 proposed exhibit, maybe we --
21 JUDGE KELLEY:
We were going to take advisory 22 comments in those circumstances.
23 MR. GOLDBERG:
We don't have any objection to it.
24 JUDGE KELLEY:
Mr. Moore, there were a couple, in 25 particular, that were sort of saw-edged exhibits that I think p
V ACE-FEDERAL REPORTERS, INC.
202 347-37(U Nationwide Coserage 80k3346646
20039.0 r-1RT 1024 U
1 you were speaking about at some length yesterday.
Could we 2
focus on those?
Those two?
3 MR. MOORE:
Yes.
4 JUDGE KELLEY:
11 is next.
There are two pieces I 5
think those in particular would be useful.
Would it be 6
possible for you, at your convenience, to, either by some 7
machine or in some manner, reproduce 8-1/2 by 11 versions of 8
those?
I don't mean today or tomorrow.
I mean sometime 9
soon.
10 MR. ROCKWELL:
Your Honor, I wonder if we could 11 ask, maybe, Mr. Blake's office.
The problem is we would have 12 to transport them home.
7,
('
13 JUDGE KELLEY:
That's all right.
We are not in a 14 hurry.
15 MR. MC BRIDE:
Could I make a suggestion, Judge 16 Kelley, that might help everyone?
Mr. Moore is obviously a 17 man that appears to me to be of good faith and everyone has 18 observed him throughout these last three days.
I only had 19 the privilege yesterday, but I would have absolutely no doubt 20 if the board were ask him to take a crack at redoing those as 21 opposed to shrinking them onto 8-1/2 by 11 in precisely the 22 same fashion as he has, and then we could review them and see 23 if they resemble what these are.
That might simplify things 24 as opposed to trying to find some machine.
I 25 JUDGE KELLEY:
Any objection to that process?
l l
l l
ACE-FEDERAL REPORTERS, INC.
l 202 347 3700 Nationwide Coserage An336-6M6
28039.0 r-7RT 1025
\\s) t 1
MR. MOORE:
As an additional proposal.
These 2
exist.
I was copying from 8-1/2 by 11 when I made these 3
things.
Maybe the thing to do is just submit some copies 4
with some additional lines drawn on them to indicate exactly 5
what I've got here.
Then you have got the real 8-1/2 by 11, 6
with these additional markings added to them, Judge.
7 JUDGE KELLEY:
Any objection to that approach?
8 MR. MC BRIDE:
I think that makes perfect sense, 9
Judge Kelley.
10 JUDGE KELLEY:
Hearing none, at your earliest 11 convenience, we'll call them Board Exhibits 11-A and 11-B.
12 What is an appropriate caption, Mr. Moore?
For both of 13 them?
14 MR. MOORE:
Effect of oscillations.
15 JUDGE KELLEY:
Effect of oscillations, Board 16 Exhibits 11-A and 11-B.
r i
17 Does anybody object to the admission of the l
l 18 exhibits that Mr. Moore is going to produce for us?
Hearing l
19 none, they are identified and admitted.
20 (Exhibits 11-A and 11-B identified.)
1 21 JUDGE KELLEY:
They will be received later.
22 We had the licenses, I guess, Mr. Goldberg.
Do 23 you want to speak to them?
24 MR. GOLDBERG:
I guess we identified as Exhibit 10 25 the senior operator license of Mr. Zewe, effective date July 0
ACE-FEDERAL REPORTERS, INC.
3C-347 3700 Nationwide Coserage 800-33MM6
28039.0 f
f-8RT 1026 l
U
(
1 29, 1977.
I guess I would ask that that be identified as 2
10-A.
And we have distributed to the board and the parties l
3 an operator license, a so-called CRO license as opposed to an f
4 SRO license, of Mr. Craig Faust, effective date, October 20, l
5 1977.
6 These, then, are representative examples of the 7
licenses which are issued under Part 55 to senior operators 8
and operators at TMI-2 during the relevant time period.
They 9
contain the language to which Mr. Russell referred in his 10 testimony.
I would offer these for the record.
11 JUDGE KELLEY:
Do I understand correctly that the 12 supervisory people immediately over the operators, such as O
13 the shift foreman or shift supervisor, that people in those 14 positions may, in fact, hold an operators license but not by 15 virtue of their job?
I don't have to have a separate license 16 to be a shift supervisor, do I?
17 MR. GOLDBERG:
I'll let Mr. Russell answer that.
18 MR. RUSSELL:
The facility license requires that 19 shift foremen and shift supervisors be licensed as senior 20 reactor operators.
21 JUDGE KELLEY:
All right.
But there isn't a 22 separate license that comes into play when I become a shirt 23 foreman?
24 MR. RUSSELL:
tJ o, there is not.
It's the same 25 senior operators license whether you are shift foreman or O
ACE-FEDERAL REPORTERS, INC.
I I
- acom,
~_m~
28039.0 e7RT 1027 Iv: -
1 shift supervisor.
2 JUDGE KELLEY:
Thank you.
It is proposed to admit 3
these two samples of licenses for the sake of having the 4
language before us.
Any objection to that?
None.
All 5
right, fine.
So these are 10-A and 10-B, as styled by 6
Mr. Goldberg.
7 (Exhibits 10-A and 10-8 received.)
8 MR. GOLDBERG:
Also during the course of testimony 9
I believe the board requested a copy of administrative 10 procedure 1036 be made part of the record.
We have 11 distributed that to the board and the parties.
If the board 12 wants that a part of the record, it is available now.
~
13 JUDGE KELLEY:
Any objection to admitting that 14 document?
15 MR. MC BRIDE:
I'm sorry, what is the document?
16 MR. GEPHART:
What number would that be?
17 JUDGE KELLEY:
It would get us up to 11, I i
18 believe.
19 MR. MC BRIDE:
What's 11?
20 JUDGE KELLEY:
11 is the two exhibits from 21 Mr. Moore, i
22 MR. MC BRIDE:
The exhibit is a portion of the 23 Three Mile Island nuclear station administrative procedures?
24 JUDGE KELLEY:
1036, I believe it covers.
25 Any objection to Exhibit 12?
Hearing none, that's ACE-FEDERAL REPORTERS, INC.
202 347-3hD NJtionwide COserage
- n3346M6
..m.
28039.0
-,-7RT 1028 O
1 admitted as Board Exhibit 12.
2 MR. MC BRIDE:
Does the Staff know whether this 3
was the latest revision of this administrative procedure 4
during the time period in question?
5 JUDGE KELLEY:
Is that the case?
6 MR. GOLDBERG:
I would ask Mr. Russell.
I don't 7
know.
8 MR. MC BRIDE:
And I guess a further question of 9
clarification, whether these administrative procedures might 10 have been revised during the time period in question, and, if 11 so, it seems to me for completeness of the record we ought to 12 have all the revisions, because these procedures, like all O
13 procedures at nuclear plants, undergo fairly regular revision 14 and there could be some confusion in the record if we didn't 15 have all the revisions.
16 MR. CAPRA:
I cannot certify that this is in fact 17 the latest revision that was in effect at the time of the 18 accident.
However, I believe that that is the case.
We will i
19 have to get that verified, probably through the licensee, l
l 20 Judge.
l 21 The reason I say that is, after the accident we 22 had requested copies of all the administrative procedures and 23 operating procedures that were in effect at the time and I l
24 obtained this from our document control system in the series 1
25 of those procedures.
So I believe that it is, but it needs (E) l ACE-FEDERAL REPORTERS, INC.
202-347-37(1)
Nationwide Coserage MX)-346646
l 28039.0 r-7RT 1029 k_)
I to be verified.
2 MR. MC BRIDE:
Judge Kelley, I can just observe 3
from the first page of the exhibit that it underwent'a 4
revision on August 23, 1978.
It seems to me that we need to 5
know two things --
6 JUDGE KELLEY:
I think your point is well taken.
7 You want all versions in effect in the time period; correct?
8 MR. MC BRIDE:
Yes.
From March 1,
'78, which was 9
the first time this procedure was in effect, through March 10 28, 1979.
11 JUDGE KELLEY:
That will simply have to be checked 12 out; is that correct?
Whether this is the only one?
Whether 13 there are some others?
14 MR. CAPRA:
I don't believe that the Staff is in a 15 position to do that.
It may be more effective for the 16 licensee to do that.
17 JUDGE KELLEY:
Well, Staff, licensee, in Q
18 l conjunction, could you run that down?
Again, I don't think 19 time is of the essence.
We already talked about this 20 yesterday.
We are simply filling in the record.
If this 21 could be checked out in the course of the next week or so I 22 would think that would be _tdequate.
If there's more than one 23 version that we ought to add, we can add that.
But it has 24 been proposed.
It certainly appears from its face that it 25 was in effect some of the time.
We can let the ruling wait O
ACE-FEDERAL REPORTERS, INC.
202 347.jhy)
Na:ionwide Coserage
- Nk3 %fM6
28039.0
<-BRT 1030 k_)
1 or we can put this in, subject to check.
I don't think it 2
really matters.
3 MR. MC BRIDE:
I would prefer to let the ruling 4
wait because, if by chance we cannot satisfy ourselves that 5
we have all versions, then I would plan to oppose the 6
admission of this in evidence.
7 JUDGE KELLEY:
We'll let the ruling wait.
The 8
ruling is pending.
The exhibit is proposed Exhibit 1.
Then 9
we'll expect to hear back from Staff and licensee, when they 10 can run this down.
11 MR. CAPRA:
Judge Kelley, can I make one more 12 comment on this?
If it makes any difference, the Staff only 13 only utilizes this particular procedure during a specific 14 period of time during its testimony.
The time in question 15 was during October 1978 through January 1979.
16 JUDGE KELLEY:
You mean in connection with your 17 investigation?
18 MR. CAPRA:
In connection with our testimony, 19 that's the specific time period where this particular 20 procedure becomes important.
The rest of the time it is not 21 germane to our testimony.
22 JUDGE KELLEY:
Well, in any case, could you just 23 check out whether there are different versions and the 24 operative time.
Then it may or may not result that this is 25 the only one we are interested in, but let's check it out O
ACE-FEDERAL REPORTERS, INC.
202-347d71O Natiorluide C04CrJge P4Xk336-(M6
28039.0 r-SRT 1031 U
1 anyway.
2 MR. CAPRA:
Yes, sir.
3 JUDGE KELLEY:
Okay.
Anything else before we 4
resume questioning?
5 (Discussion off the record.)
6 JUDGE KELLEY:
Just for clarity in the record, the 7
status of the Exhibit 1 is as a proposed exhibit.
It stands 8
proposed.
It would be in the record as proposed and it will 9
be ruled in or out in the light of information, subsequently 10 to be received.
11 MR. GOLDBERG:
Judge Kelley, there was one other 12 matter pending with the board that I'm aware of but it may be O
13 that you are about to take that up.
There is a pending 14 proposed question from the Staff to the Staff witnesses.
15 JUDGE KELLEY:
Right.
Consistent with our prior 16 ruling, which was to the effect that at least in the present 17 format with a large panel ot experts, we would let the Staff 18 propose questions to Staff members, leaving open what we may 19 do later.
So, I will proceed to ask those questions.
20 I believe the objection arose when I read the 21 following question and we never got to whether it ought to be 22 answered.
I'll read it again.
23 Regardless of the magnitude of all errors 24 associated with the TMI-2 leak rate test and assuming all 25 these errors were known by the operators, did the TMI-2 tech O
ACE-FEDERAL REPORTERS, INC.
1.02,347 3ho Nationwide Coserage 801U%%86
28039.0
-3RT 1032 U
1 specs and the licensee's administrative procedures require 2
operator action if the test result was greater than 1 gpm 3
for identified leakage?
4 The board comment, it's a request for a legal 5
conclusion at least in part.
We had some discussion along 6
these lines the other day, nevertheless, and so we are going 7
to allow the question.
We understand Mr. McBride made an 8
objection yesterday.
We understand it.
We will allow the 9
question.
10 MR. GOLDBERG:
Judge Kelley, before the question 11 is answered I must confess that I did not hear the portion 12 yesterday about administrative procedures, and I would 13 respectfully request that the board reconsider at least that 14 portion of its ruling with respect to that portion of the 15 question.
That is to say, I do not believe it has been 16 established in any case law of this Commission or in the f
17 statutes that the administrative procedures are necessarily a 10 part of the license.
19 There is some reference to Regulatory Guide 1.33 20 in the technical specifications of this unit.
However, there i
21 is no document that is denominated Regulatory Guide 1.33, 22 They were even denominated differently when that came out.
l 23 It doesn't refer to any specific procedures.
And I think we 24 have a very serious issue of vagueness here and I think the 25 question at least ought to be rephrased to take out the O
I ACE-FEDERAL REPORTERS, INC.
I 202 447-37(O N.itioriwide roserare RXk33/*f646 L
28039.0 r-SRT 1033 b
1 portion about administrative procedures.
2 And I would renew my objection on the ground that i
3 the question is being put to a layman who is demonstrably not i
4 in the legal sense competent to answer this question.
But S
I'm not going to ask the board to rehash that issue.
I 6
simply want to note on the record that I had two objections 7
yesterday and I asked the board to reconsider that portion of 8
the question that I just referred to.
9 JUDGE KELLEY:
We have reconsidered and we are 10 going to overrule both objections.
The fact of the matter i
11 is, although Mr. Russell, as I understand it, is not a 12 lawyer, he did conduct this investigation.
He has reached fsb 13 certain conclusions about what people are supposed to do and 14 what they are not supposed to do.
We might disagree with him 15 as a matter of legal analysis.
That's a separate point.
But 16 we'd like to know what Mr. Russell thinks is the answer to i
17 the question, in his opinion.
I l
18 MR. GOLDBERG:
Judge Kelley, I believe you 19 misspoke when you read the question.
The last phrase was 20
I believe you said " identified I
21 leakage."
It is "a greater than 1 gpm unidentified 22 leakage."
l 23 JUDGE KELLEY:
I may have mumbled or misspoke.
24 Unidentified is the word; yes.
Go ahead and answer the i
25 question, if you can.
j ACE-FEDERAL REPORTERS, INC.
3c _
s _. o_,,
m u-n.
,- _,.n - --, - -,-,,.---..,.-----
,-,,--,-.--.-.c-.------
28039.0 f '3RT 1034 f
1 MR. RUSSELL:
The answer is that they are required 2
to implement the action statement.
I testified to this 3
earlier.
It is contained in the surveillance procedure 4
itself, in the acceptance criteria for identified leakage.
5 It is acceptance criteria 7.1 and it specifically states:
6 "If the identified reactor coolant leakage (line 30 of RCS 7
leakage data sheet 1 or line 32 on data sheet 2 exceeds 10 8
gpm, proceed with action statement 3.4.6.2-B."
9 The acceptance criteria which relates to 10 unidentified reactor coolant system leakage is criteria 7.2, 11 which states:
"If unidentified reactor coolant leakage (line 12 31 of data sheet 1, or line 33 on data sheet 2) exceeds 1 0
13 gpm, proceed with action statement 3.4.6.2-B."
14 It is the Staff's view that that procedure is very 15 specific, that if a calculated number on the data sheet is 16 greater than the acceptance criteria, that the operator must 17 proceed with the action statement required by the technical 18 specifications.
We need not rely on the administrative 19 procedures, but the administrative procedures would also 20 require the filling out of a deficiency report because a 21 surveillance test had exceeded the acceptance criteria.
This 22 is Administrative Procedure 1010.
23 JUDGE KELLEY:
Thank you.
24 JUDGE CARPENTER:
The question proposed by Staff l
l 25 counsel to Mr. Russell:
If the safety parameter (LCO) is l
ACE-FEDERAL REPORTERS, INC.
202 4 47-3700 Nationwide roverage Molk 3hfM6
28039.0 r-3RT 1035 k_)
1 determined to be independent of any instrument error, and the 2
instrument error analysis showed a plus-minus 0.7 random 3
normal error, would an LCO of 1.7 be acceptable --
4 MR. GOLDBERG:
Judge Carpenter, would you like 5
me --
6 JUDGE CARPENTER:
I'm having problems with 7
legibility.
8 MR. GOLDBERG:
That is very understandable.
I 9
would be glad to read the question for you, if you would like 10 me to.
11 JUDGE CARPENTER:
How about just the word --
12 MR. GOLDBERG:
If the parameters were determined O
13 to be 1,
by safety analysis not including any error analysis 14
-- you also missed that word.
And a random error analysis 15 shows that the random normal error for the instruments used 16 to determine whether that safety parameter is exceeded turns 17 out to be.7
-- The question goes on to say:
Would you set a 18 safety parameter at 1.7, and what the effects of that would 19 be.
That's the thrust of the question.
20 JUDGE CARPENTER:
I understand the thrust of the 1
21 question.
I'm having trouble with the specific words.
22 JUDGE KELLEY:
Well, have him read it.
We do ask 23 all counsel -- I have terrible handwriting myself, but do try 24 to write these things so that we can read them.
It has been 25 very difficult in several cases.
O ACE-FEDERAL REPORTERS, INC.
202-347 37tu)
N.itionwide Coser. ire Rtk 33MM
_ _ ~ _. _...
W 28039.0 r-SRT 1036
(_)
1 MR. GOLDBERG:
If a safety parameter (LCO) is 2
determined to be 1,
independent of any instrument error, and 3
the instrument error analysis showed a plus or minus
.7 4
random normal error, would an LCO of 1.7 be acceptable, 5
considering that the test readings near but less than 1.7 --
6 for example 1.6 -- also would have an expected plus or minus 7
.7 random normal error; so that the actual condition could be 8
1.6 plus or minus
.7 or as high as 2.3?
9 MR. MC BRIDE:
I object to that question, Judge 10 Kelley.
11 MR. GOLDBERG:
I'm not finished with the 12 question.
-- thus exceeding the safety limit.
O
13 Part two of the question is, therefore, would it 14 be incorrect to set the LCO at 1.7?
15 MR. MC BRIDE:
I object on two grounds.
The 16 question is indecipherable and its irrelevant.
The LCO is 17 not 1.7.
10 (Discussion off the record.)
19 JUDGE KELLEY:
The Court is not going to ask this 20 question for two reasons.
Frankly, it's quite confusing and 21 we have been over this whole area in great detail and we 22 don't believe it's worth 3t to resume it.
Let's move on.
23 JUDGE CARPENTER:
Resuming the questioning, where 24 we were yesterday afternoon, the next question is number l
25 249:
As a technical matter, would a water slug be more ACE-FEDERAL REPORTERS, INC.
n w.
~ ~ + ~ -,,,
, _ -...., _ _ _. _, _. ~,,, _. _ _ _ _ _
..__.__..__,.m.
28039.0 g-BRT 1037
)
-N/
1 likely to cause an increase or a decrease in the calculated 2
value of unidentified RCS leakage?
3 MR. KIRKPATRICK:
Barring any operator action, the 4
likelihood of an increase or decrease is about equal.
If 5
nothing happened to increase the makeup pressure during or 6
shortly before the beginning of the test, it would have no 7
effect because a downward trend in pressure would keep the 8
slug in the same place.
9 If the pressure were increased just prior to the i
10 beginning of the test, it would cause an increase in the 11 calculated leakage.
If the pressure were increased near the 12 end of the test, the reverse would occur.
O) s 13 MR. ROCKWELL:
To answer the question as written, i
14 it is really unanswerable, because it depends totally on your 15 assumptions.
And Mr. Kirkpatrick just gave some assumptions, 16 but really the true answer to the question is it depends on 17 what you assume about plant conditions, in terms of the three 18 factors that we have looked at.
19 JUDGE CARPENTER:
Mr. Stier?
l 20 MR. STIER:
I think we disagree with 21 Mr. Kirkpatrick.
We believe that the condition of the makeup 22 level transmitter would exaggerate an increase or a decrease 23 in the level of the makeup during the period of the test.
24 If it exaggerated a decrease, then it would give f-25 you a higher value for unidentified leakage.
If it U
ACE-FEDERAL REPORTERS, INC.
202 347 37W NJFionwidt Contrage 80lbhfM6
28039.0 1038 O^BRT 1
exaggerated an increase, then it would give you a lower value 2
for unidentified leakage.
If you had a decline in the level 3
of the makeup, as was the usual case, with no water addition 4
during the test, if you had a decline, the decline, rate of 5
decline during the test would be exaggerated by this 4
i 6
condition.
7 If you had an increase in pressure in the tank 8
during the test, either caused by water addition or a 9
hydrogen addition, then that would exaggerate -- that would 10 incorrectly show a higher level in the makeup, thus reducing 11 the calculated unidentified leakage.
And, in our judgment, 12 it doesn't make any difference whether the increase in N'
13 pressure occurs at the beginning or the end of the test.
The 14 effect on the calculated unidentified leakage is going to be 15 the same.
16 Now, there may be a misimpression, there may have 17 been a misimpression on the part of operators observing this 18 phenomena, because if you look at the strip chart trace 19 that's left by the hydrogen in the makeup when this condition l
20 is present, you see a bump in the trace.
There's a decline 21 caused by the decline in the level in the makeup.
There is a i
22 hydrogen addition.
And you see a deflection in the trace, t
23 and then it begins to decline again at about the same rate au 1
24 it was declining previously.
But the calculated unidentified t
25 leakage, or the difference in makeup level at the beginning ACE-FEDERAL REPORTERS, INC.
mm, s _ c _,.
,--,----,_,-.c-_-..
28039.0 1039 r-9RT l
U 1
and the end of the test, even tnough the end rate of decline 2
is the same as it was at the beginning, is at a higher point 3
than it would have been had there been no hydrogen addition.
4 So it doesn't make any difference, in our 5
judgment, and we can't think of any technical reason why the l
6 effect of a pressure increase would be any different at the 7
beginning or the end of the test.
8 Now, we can go into more detail on the basis for 9
that if you would like to hear some further explanation.
10 JUDGE CARPENTER:
Mr. Kirkpatrick, do you have a 11 comment?
12 MR. KIRKPATRICK:
Yes.
My own analysis of the 13 behavior of the slug indicates that it behaved essentially as 14 a bi-stable, by that I mean it either added a specific 15 amount to the level, compared to what it should have read, or 16 it subtracted a specific amount from the level compared to 17 what it should have read.
And that the time between one i
18 stable condition and the other stable condition was i
i 19 relatively brief.
1 l
20 This time could vary from about a half hour down 21 to a few minutes, but in all cases that I reviewed, I believe 22 the slug was either in one situation or another.
I 23 I
I believe the gas did bubble through this slug l
l 24 because the amount of pressure change that you would expect 25 to occur in the top of the tank was at least an order of (O
ACE-FEDERAL REPORTERS, INC.
2"2:"""
28039.0 g-9RT 1040 1
magnitude greater than the amount of pressure that would 2
cause this slug to switch.
3 In order to move a 1-inch slug from one position 4
to the other required only.036 inches of pressure.
A 1-inch 5
increase in the tank level, for example, with the water near 6
the middle of the tank, should have had an effect of over 7
half a pound.
8 Admittedly there was a compression effect due to 9
the fact that as the slug moved, the gas on the detector side 10 of the slug was compressed, and this would have had an effect 11 but because the pressure change in the tank was so much 12 greater than what it took to move the slug.
I believe this O
13 was a minor effect, so that you always had an addition or you 14 had a subtraction.
15 Now, if this addition -- for example, if the 16 pressure increased just prior to the beginning of the test, 17 you would have had the slug moved away from the -- you would 18 have the slug move toward the detector, just prior to the 19 beginning of the test, which would have increased the tank 20 level.
As the tank level reduced, due to identified leakage 21 going away, it would switch back, so that you would always, 22 in this case, have had an increase in the measured leak.
23 Conversely, if whatever it took to raise the 24 pressure, whether it was a level rise or pressure rise due to 25 hydrogen addition, if this occurred durino the end of the O
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nanonode Omcrage suWMM6
28039.0 l
9RT 1041 s
1 test, again you would switch the slug f' rom its former 2
position toward the tank over toward the detector, resulting 3
in a reduction in identified leakage.
4 JUDGE CARPENTER:
Mr. Stier?
5 MR. STIER:
It I could draw a picture it would 6
help me describe for you what I'm about to say.
I don't know 7
whether that's --
8 JUDGE KELLEY:
I think we have, if not a 9
blackboard, then -- in back of the charts there.
Off the 10 record for a minute.
11 (Discussion off the record.)
12 JUDGE KELLEY:
We'll take a 10-minute break while O
13 we get the artwork underway here.
14 (Recess.)
15 JUDGE CARPENTER:
Continuing on question 249, we its are about to see some artwork.
17 I would like to ask a question, a question more 10 focused.
The question as originally proposed said:
Would a 19 water slug be more likely to cause an increase or decrease in 20 a calculated value of unidentified RCS leakage?
I 21 Is it true that there are such a variety of 22 conditions that you can't identify the most likely set of 23 jassumptions, in the spirit of Mr. Rockwell's answer, that 24 would have prevailed at the plant; so you generally can't say 25
'nore likely, as a generalization, for an extended time period ACE-FEDERAL REPORTERS, INC.
- e mm, s m.n. m.,, m,,
,mo w-w...
28039.0 r~pRT 1042 U
1 on any particular day it might be one or another?
2 MR. STIER:
No.
I think you can make two very 3
basic assumptions.
And I think that the answer, then, I
4 becomes pretty clear.
If, during the course of the test, you l
5 don't increase the pressure in the makeup -- that is, you 6
don't add water or hydrogen -- then your unidentified leakage 7
will be overstated because of the loop seal condition.
8 If, during the test, you increase the pressure in 9
the makeup by adding water or hydrogen, then your 10 unidentified leakage will be understated.
With those two 11 assumptions, I think the effect of the loop seal can be 12 determined quite easily.
(
l 13 JUDGE CARPENTER:
Mr. Rockwell, would you agree i
14 with those two generalizations?
15 MR. ROCKWELL:
Yes.
And then, obviously, one has 16 to quantify.
Because you may have effects that are trending 17 in opposita directions at the same time and they net out, so 18 you have to quantity how much level you have lost from the 19 makeup over the course of the test as opposed to how much 20 addition has been made.
But the general thrust of the 21 response I agree with.
And those are the kinds of 22 assumptions I think you have to make in order to make any 23 l Intelligent evaluation of what's happening in a particular i
24 test.
h JUDGE CARPENTER:
Are they assumptions, or do you 25 (2)
ACE-FEDERAL REPORTERS, INC.
m m.3,e,
~._.-_
i L
J 20039.0 j
RT 1043 i
I have to look at the magnitude of several factors
)
2 simultaneously?
3 MR. MOORE:
As Mr. Rockwell just said, if there
)
4 are no additions to the tank, water or hydrogen, the only 5
parameter that's left to vary is the level dropping due to i
i 6
identified leakage of some type, probably identified.
4
{
7 Once you make an addition of water or hydrogen to l
8 the tank in the presence of a water slug, then it depends on l
9 whether you've added -- let's take the case of water l
t 30 addition.
That's the simplest one.
If you add more water i
11 then there's loss due to leakage over the period of the i
i 12 test.
We are adding more water than we lose, then you are O
i 13 going to understate the leak rate.
Let's see if I got it.
j j
14 Any water addition during the test is going to understate the l
15 leak rate.
l 16 JUDGE CARPENTER:
That seems to be fairly I
(
17 obvious.
MR. MOORE:
Hight.
In the case of hydrogen 18 j
l 19 addition, if that were to I guess, either one -- I hadn't l
i l
20 i thought the thing throuqh.
i i
21 i
MR. ROCKwELL-I auess the point is the magnitude
' of the increase or decrease in the amount of additional 22 j
i
- idered as well 23 hydrogen over-pressure created has to be m
i 24 an whether a: not thoro condi tionc 2: e ex; t' 1!a at all.
So l
t O
I
'"" ""'"' "' "" """ ' " *~' """ =*'" '"" """ """" '
1 i
i ACE-FEDERAL REPOR FERS, INC.
2H2 447 3'm N n a v H!e rostrage N t i 3 IfrNW i
28039.0
(-BRT 1044 V
1 some fairly specific assumptions in order to draw a 2
conclusion as to any given test.
And while one can 3
generalize, for example, in the time nearer the accident, 4
that there tended to be a downward trend in the strip chart 5
trace with respect to the makeup level, because of the 6
leakage, there are other generalizations which one has to be 7
cautious in making, in terms of trying to generalize about 8
the effect of the slug.
9 JUDGE CARPENTER:
When you say " assumption," I'm 10 trying to be sure I understand clearly.
Are you saying you 11 have to look at the strip chart over the whole time period 12 and establish in your mind what the characteristics of that O
13 period were?
And then, knowing those, one can do the 14 analysis?
Or is there some assumption that is not reflected 15 by evidence trom the strip chart that you might have to 16 make?
17 MR. ROCKWELL:
I believe, if you know the decline 18 in level of the makeup as reflected in the trace, if you look 19 for recorded water additions and you had that piece of 20 information. If you look -- I believe it's a different log, 21 for hydrogen additions, to determine whether there were 22 hydrogen additions, then with that information and assuming 23 that hydrogen additions and water additions were recorded so 24 that in fact the information you are getting from the logs is 25 accurate -- and as you may know we looked at the possibility ACE-FEDERAL REPORTERS, INC.
, um.,
s _ _
m,.
mo m
28039.0 cSRT 1045 U
1 of whether there were unrecorded water additions and 2
unrecorded hydrogen additions -- but once you have that data 3
before you then I think you can analyze it.
And I don't 4
believe -- and let me confirm this with Mr. Moore -- but I 5
don't believe there are other factors that one needs to know, 6
assuming the water plug is postulated to be there --
7 MR. MOORE:
That covers it.
3 MR. ROCKWELL:
-- that you need to know in order 9
to arrive at an answer.
10 JUDGE CARPENTER:
Thank you.
Mr. Stier?
11 MR. STIER:
Well, my answer previously, I think, 12 was correct, that the effect of the loop seal is going to be
_s 13 an exaggeration of decline or increase in the level of the 14 m%kepp.
?be magnitude of that, of course, depends on a whole 15 varlely'of conditions that exist in the reference leg of the 16
' level transmitter.
But the magnitude of the effect can be 17 determined fairly easi]y.
I say " fairly easily."
MPR did a 18 lot of work to do it, but the principles are, I think, fairly 19 clear.
20 That is, if you compare the log water addition, 21 the amount of water addition that is in the log, with the 22 s trip chart ' trace, the difference is the effect of this loop 23 seal condition.
That is, if 300 gallons are logged and you 24 look at the strip chart and you see that there appears to be p
25 a 350-gallon addition, then you know what the effect of the
?
%)
ACE-FEDERAL REPORTERS, INC.
202-347-3710 Nation %lde r0%erage muk3346646
?
28039.0 l
c '3R T 1046
/
(aI 1
loop seal is.
2 And if you find a number of these and it! snows a 3
consistent pattern, then you can demonstrate whether there is 4
a loop seal, or whether this condition exists -- wh" ether it's 5
a loop seal or not we discussed yesterday -- but you can see 6
whether the condition exists and you can measure its effect 7
on lesk rate tests.
And that has been done and the figures 8
that I cited in our report, figures IV-S and -6, depict the 9
effect of this condition as it was determined by MER, through 10 the entire period of operation of TMI-2.
11 We didn't go back and try to figure out what the 12
.Oop seal was doing and make assumptions about the underlying
,.s
(
i 13 condition.
We just observed its effects and showed them on 14 these figures.
15 Would you like me to get to the drawi.ng?
16 JUDGE CARPENTER:
Yes.
Go ahead.
17 MR. STIER:
The question I was dealing with before 18 the break is whether it makes any difference whether the 19 pressure is increased in the makeup at the beginning or the 20 end of the test.
Wnat I have drawn here is my representation 21 of a strip chart trace that depicts the effects of a hydrogen 22 addition.
23 I have tried to approximate the appearance of the 24 trace that is included in our report in figure 3 on page 94 25 of volume 1.
That is a picture of a strip chart in which
,f 3 ACE-FEDERAL REPORTERS, INC.
202-347-3700 Naiionwide Coverage 800-336-6646
9
)
i
<r 28039.0
(-}RT 1047 kJ 1
there has been a hydrogen addition made to the makeup during 2
the equrse of the test.
When this condition is present --
3 this is.as c1cce as I can come to depicting what is there.
4 This point (indicating) represents the beginning 5
of the test, the readings taken at the beginning of the test, 15 and this represents the readings taken at the end of the 7:
7 test.
You can see that if there had been no hydrogen 8
addition -- which is depicted by this bump in the strip chart 9
trace -- if there had been no hydrogen addition and the
'10 decline continued the rame as it was at the beginning of the 11 test, the ending reading would be down here.
So the 12 difference between the starting time and the ending time pAs
'> I N ]
13 would be the amount of decline in makeup level that would be 14 read by the test.
And would be included in the test 15 calculation.
16 With the hydrogen addition made to the makeup, 17 offsetting this trace, you can see that the ending reading 18 ends up higher than it would have had there been no hydrogen 19 addition.
So the test thinks that there was more water in thesystemduri$hthecourseof the test than there really l
20 i
21 was.
The effect of that is to reduce the calculated, 22 unidentified leakage.
23 Now, if your test had been started at this point 24
-- I'll move the line to the right -- and it ended here --
t
,e 25 and I'm moving the ending point to the right (indicating),
Q,,3 l
o-
?
I 1
l l
ACE-FEDERAL REPORTERS, INC.
I
_ x.y_m c_..
mm.m.
1 28039.0
<-7RT 1048
(_)
1 you can see that the effect is the same on the calculation.
1 2
That is, once the bump begins to decline on the 3
trace and the level of decline in the makeup continues at the 4
same rate as it had been previously -- and I say "at the same 5
rate," because in our review of strip chart traces, that's 6
what appears to occur, the rate goes back to the same rate it 7
had been previously -- you can see that the effect on the 8
leak rate test is the same.
The ending reading is higher 9
than it would have been had there been no hydrogen addition 10 to the makeup.
And that's why I said previously that, from 11 our observations, it doesn't make any difference whether the 12 hydrogen addition or the water addition is made at the V
13 beginning of the test or the end of the test.
14 Now, there's testimony.that operators believed 15 that they had to make the hydrogen addition at the end of the 16 test, close to the end of the test, to achie've the result 17 that they were looking for.
It is my belief that they were 18 focusing on this bump and hadn't recognized that it was the 19 offset that was ir.portant, rather than the rate of decline.
20 The offset is represented, of course, by the two red lines 21 that I have drawn on this figure.
And it is that offset that 22 you have to look at in order to determine the effect of a 23 hydrogen or uater addition on the leak rate test.
24 I hope that that helps explain our position, 25 JUDGE CARPENTER:
Apparently the hydrogen that is g-)
V ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
28039.0
< 'IRT 1049 k_)
1 added does not go into the circulating fluid at a rate that 2
is significant over a period of an hour?
3 MR. STIER:
That's correct.
4 MR. COLE:
That's correct.
5 JUDGE CARPENTER:
That would seem to me one of 6
Mr. Rockwell's " assumptions."
7 Mr. Kirkpatrick, do you agree with this 8
description?
It's simply the offset, no matter what time 9
during the test the offset occurred, that's important?
10 MR. KIRKPATRICK:
I agree with most of what he's 11 saying.
Indeed, I have seen bumps like that when hydrogen 12 was added, where the effect persisted for a considerable O
13 time.
However, I think in the more typical situation, the --
14 assuming there is, again, a downward trend in the level of 15 the makeup, that the bump that you get as a result of the 16 pressure increase tends to go away and I've done -- I've even 17 added up a whole bunch of these things to get an average and 18 I believe in a previous report that I made I indicated that 19 the longest time I had seen it persist was on the order of 45 20 minutes, and that in most cases it was gone In a half hour.
21 In some cases the bump goes away in like 10 minutes.
22 By " going away," J mean that the oscillating level 23 reduction, what you see is sort of an oscillation that is 24 trending downward, but it appears to go back to the original 25 line anywhere from, say, 45 minutes to 10 minutes, and O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 80fk336-6646
28039.0 7-pRT 1050 V
1 generally it is usually back within a half hour.
2 JUDGE CARPENTER:
Do you have any specific 3
knowledge of a mechanism that would cause that to occur?
4 MR. KIRKPATRICK:
Yes.
I believe that we do get 5
bubbling through.
You get caught up in the argument of, you 6
know, as you move the bubble, of course, you compress the air 7
on the detector side of the bubble as well.
This will have a 8
counteracting effect, which will tend to delay the rate at 9
which the bubble shifts over.
10 The weight of the water in the slug itself is very 11 small compared to the forces, with the pressure on the makeup 12 on the one hand and the pressure on the detector side on the 13 other hand.
So what you really have is a balancing here.
14 However, I believe that the volume you have on the side of 15 the detector compared to the volume that is changed due to 16 the movement of the slug, that in most cases -- not a.1 ways --
17 you will, in fact, cause the slug to assume an equilibrium 18 position.
19 I also have a picture which is un attachment to my 20 original prefiled testimony, that tries to depict this in a 21 schematic way.
22 JUDGE KELLEY:
Which attachment is this?
Does it 23 have a number or letter?
24 MR. KIRKPATRICK:
This is attachment 5 to my 25 prefiled testimony.
Om l
l l
ACE-FEDERAL REPORTERS, INC.
202-3>.1-37to Natkmwide Ceterage 800-316-6M6
28039.0
<MRT 1051
()
1 JUDGE KELLEY:
Thank you.
2 MR. KIRKPATRICK:
In the view at the top it shows 3
this side of the -- this is a depiction of the loop seal.
On 4
this side it shows the tube going to the makeup, this side to 5
the detector.
With the makeup pressure increasing, the 6
pressure on the makeup tends to shove it over here.
7 Once the loop seal reaches a point where the low 8
side of the slug is at the bottom point in the loop, whatever 9
-- or the low spot in the tube, whatever you like to call it, 10 I believe that at that point the air starts coming through.
11 From that point on the slug won't move any more and you'll 12 have no further effect on the level shift.
Admittedly, the 7-U 13 pressure on this side does increase, will tend to offset this 14 effect, but I believe it' does occur.
15 On the other hand, when you have a condition where 16 the makeup level is decreasing, the pressurn on the detector 17 side then overcomes the pressure on the makeup side, and it l
l 18 shifts back.
But again, when the other side of the slug l
19 reaches the low point in the loop, you again have the l
20 pressure bubbling back out, away from the detector side.
So 1
21 that it now assumes a new stable position.
t l
22 As I say, I think this shift back and forth I
l 23 depends on the size of the loop, depends on the pressure i
24 changes, depends on a lot of things, but in general will l
rs 25 happen within about a half hour.
l O ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8%336-6M6
28039;0
~3RT 1052
%)
1 JUDGE CARPENTER:
Mr. Stier, do you have any 2
comments on this description of the dynamics of the offset?
3 MR. STIER:
Well, no, because I don't completely 4
understand why all of the assumptions have been made about 5
the condition in the reference leg.
We just don't know 6
enough about the configuration of the reference leg and the 7
amount of water that might have been in it to make a detailed 8
analysis of how it behaved.
9 My figure that I just drew is based on our 10 observation of strip charts and, that picture, that is the 11 one that was taken from figure 3 of my report, we know to be 12 a strip chart trace showing a hydrogen addition.
And we know 13 during that period of time that there was an effect on the 14 indicated increase and decline in makeup level.
Whether it 15 was caused by the conditions that Mr. Kirkpatrick described, 16 I really don't know.
17 JUDGE CARPENTER:
Mr. Kirkpatrick also observed, 18 apparently for a number of strip chart records that he loaked 19 at, the curves don't look like the figure that's on there in 1
20 thic exhibit.
There's an offset -- that offset disappeared 21 with time, with the time being 10 to 45 minutes, if I 22 understood his testimony right.
23 MR. STIER:
This figure is kind of a baseline 24 because we know that that was a hydrogen addition because 25 somebody wrote on the strip chart "added hydrogen."
No gs
(_)
ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 80lk33MM6
28039.0
("3RT 1053
\\~)
1 question there's a hydrogen addition.
2 The other strip chart traces -- and of course I 3
don't know specifically which traces Mr. Kirkpatrick is 4
talking about, but I believe from our investigation that 5
there are no other strip chart traces -- there are very few 6
strip chart traces where you can say with any real certainty 7
that a hydrogen addition took place at the time you see the 8
bump on the strip chart.
There are a couple of times when 9
you were find a log entry in the auxiliary operator's log 10 indicating that there was a hydrogen addition at a certain 11 time, you go to the strip chart for that time and you can see 12 a deflection in the trace.
But there are very few of those 7.s
(
)
13 instances.
And the ones that we observed looked essentially 14 the same as the one that I have drawn for you.
15 JUDGE KELLEY:
Did you identify at an earlier 16 point your figure 3, exactly where that is?
17 MR. STIER:
Yes, volume IV, page 91 of my report.
18 And let me add something else, Judge.
There is an 19 extensive discussion of this subject in volume IV, IV-A, in 20 section VII.
A very extensive analysis of hydrogen additions 21 by MPR.
In table VII-1 of that section you will find a 22 survey of every known hydrogen addition that we can find from 23 any source; that is, every alleged hydrogen addition.
So you 24 can identify which specific test it relates to and you can
(~$
25 look at the strip chart yourself and see what you observe.
V ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage M 43364 M 6
28039.0 e'7RT 1054
%]
1 Let~me correct something I may have said.
This 2
chart shows the logged hydrogen additions and tells you where 3
to find them on the strip charts by time.
4 MR. HARRISON:
Just a second.
What we did was go 5
to CRO logs and AO logs.
Every time they had a logged 6
hydrogen addition we went to the strip chart to see if we 7
could find an effect on the makeup level chart.
That's what 8
is represented in here.
In those, the ones -- this is a 9
typical one of those few that showed us a very strong 10 indication.
That's all completely summarized in here.
11 MR. COLE:
The order of the magnitude of the tests 12 is we looked at over 100 logged additions.
It's not one or 7_O 13 two samples.
14 JUDGE KELLEY:
Mr. Russell?
15 MR. RUSSELL:
The analysis that was done by 16 Mr. Kirkpatrick, and Dr. Chung also looked at this.
NRR felt 17 was not significant to our evaluation of leak rate tests for 18 two reasons.
One, the trace that has just been shown, in 19 NRR's opinion, could either be a water addition or a hydrogen 20 addition.
That is, a water addition of about 90 gallons 21 during the particular test, number 10 in NRR's numbering, 22 which is the same test that Mr. Stier referred to, would give 23 the same effect.
And if it hadn't been for the fact that the 24 hydrogen addition was logged, we could have considered it 25 either a water addition that was unlogged or a hydrogen O
ACE-FEDERAL REPORTERS, INC.
202-347-37to Nationwide Coserage 800-336-6M6
I 28039.0
.RT 1055 1
addition.
Because they both have the same characteristics.
2 And there is the offset.
3 There are tests where hydrogen additions were 4
logged where they did not have this effect.
That could be 5
because there was not a loop seal or for some other reason.
6 It was not important for our evaluation to determine the 7
phenomena that was causing the effect.
It was only important 8
to know that if the pressure in the tank were increased 9
during the test, that that would have an effect on the 10 indicated level and have an effect on the calculated 11 unidentified leak rate.
12 Operator's did believe it was important to do it at f
l
'~'
13 the end and we've seen cases where hydrogen addition was done 14 toward the end.
So it appeared sort of irrelevant to NRR 15 whether it occurred early in the test or late in the test 16 because the computer data logger only took data over a 17 one-hour period.
18 Whether it was a hysteresis, and it was actually a 19 slope offset due to this water slug or for some phenomenon we 20 don't know for sure, the effect went away with time.
The 21 fact is that the operators thought it was important to do it 22 at the end and used tha t mechanism.
So we didn't spend a lot of time trying to resolve the issue of how the inner workings 23 l
lI and hidden mechanisms of this level detector worked.
We just 24 25 relied principally on the observation and most of our ACE-FEDERAL REPORTERS, INC.
2(C-347-371N)
Nationwide COs cf age MIk 3%6646
28039.0 1056
(-]7RT 1
questioning of operators was along the line of if they knew 2
they added a fixed amount of water to the tank and they 3
observed an increase that was greater than that amount, that 4
became significant because the amount they actually added was 5
less, and that would have the effect on the leak rate test.
6 It is independent of whether it is hydrogen or water that 7
causes that.
8 MR. MC BRIDE:
Judge Carpenter, may I ask out of 9
fairness to my clients that you ask Mr. Russell to clarify 10 whether his references to "the operators" in that prior 11 answer was to all of the operators or some of the operators?
12 MR. RUSSELL:
It is specifically to some of the O
13 operators.
14 MR. MC BRIDE:
Thank you.
15 JUDGE CARPENTER:
Mr. Russell, is it true in the 16 context of --
17 MR. ROCKWELL:
Were you addressing me, sir?
Did 18 you say Russell or Rockwell?
19 JUDGE CARPENTER:
Russell.
20 MR. ROCKWELL:
I'm sorry.
21 JUDGE CARPENTER:
Is it true in the context of 22 looking at a series of tests and trying to judge whether or 23 not a " valid" test was conducted, one of the criteria you 24 would use would be to look at the strip chart record and look 25 to see whether there were offsets, changes in slope, during O
ACE-FEDERAL REPORTERS, INC.
I 202-347-3700 Nar;onwide coserage 800-336-6 4 6 t.
28039.0 f ~ ORT 1057 t
-V 1
the test period, and whether those changes in slope 2
corresponded to logged additions of water or of hydrogen?
3 And, in the absence of any indication in the log that a known 4
addition of water had been made or a known addition of S
hydrogen had been made, that, therefore, ipso facto, it was 6
not a valid test, because a valid test would not show those 7
characteristics?
8 MR. RUSSELL:
That is generally what we did, but 9
we had to consider a lot of other factors, including plant 10 operating conditions at the time, that could cause changes in 11 the rate of makeup.
But, in general, the procedure 12 precautions, if they were followed, you would not observe an b,- s 13 offset in the trace.
That is, if you did not conduct a feed 14 and bleed operation, or do something else, as long as you 15 stayed within the precautions and limitations of the 16 procedure you should see a consistently declining level in 17 the makeup with level oscillations about a slope such that if 18 there was a slope offset, it would appear to us, if it 19 occurred during a test, to be a test that was invalid and in 20 our terminology that would be a test that was not done in 21 accordance with the test procedure.
Whether that was done l
22 willfully or not is a much more difficult question to 23 answer.
l 24 JUDGE CARPENTER:
Yes.
But the point is the 25 starting of that exploration of whether or not it was (s
t_J ACE-FEDERAL REPORTERS, INC.
- m. u,.,,.
8.u - + ~,,,
28039.0 e7RT 1058 O
1 willful, can begin simply with the facts.
2 MR. RUSSELL:
Yes, sir.
And that is exactly the 3
reason why we identified many more tests as being invalid 4
than those which we think were intentionally manipulated.
5 JUDGE CARPENTER:
Thank you for the 6
clarification.
7 Now, Mr. Rockwell, do you have any further 8
comment?
9 MR. ROCKWELL:
I would only add chapter 7 of our 10 investigation dealt with this extensively, volume II, in 11 which the strip chart traces were examined in areat detail, 12 looking for flattening or bowing in the trace.
To go to your 7
(_)
13 comment, if you want to start with the facts of the observed 14 phenomena, they are detailed quite extensively in that area 15 of our report.
16 JUDGE CARPENTER:
Would you agree that 17 understanding at the mechanistic level exactly what was 18 causing what we see in the strip chart, is, perhaps, not 19 necessary if one is simply looking at the end of the test, 20 whether a valid test was run?
In the sense that if there is 21 an unexplained blip offset in that trace for that one hour, 22 then one would question the validity of the test?
Is that a 23 fair statement for me to make?
24 MR. ROCKWELL:
Yes.
25 JUDGE CARPENTER:
Mr. Russell?
O ACE-FEDERAL REPORTERS, INC.
202-347-370)
Nationwide Cmcrage 8m336-%86
= ~.
28039.0 r-ORT 1059 b
1 MR. RUSSELL:
Yes.
2 JUDGE CARPENTER:
All of us trying to understand 3
is certainly proper and appropriate, but perhaps in the sense 4
of looking at validity of the test, no t crucial; is that 5
fair?
6 MR. ROCKWELL:
I would agree with that.
7 JUDGE CARPENTER:
Thank you.
8 MR. MC)RE:
One thing I might add to that, though, 9
is that I was not aware before, there is an opinion that the 10 effect r.ight disappear with time, so that, per the Staff's 11 testimcny, if some hydrogen addition were made at the 12 beginr ing of the test, if the presumed slug behaved as they g-
'~
13 have described it, it is pcssible that its effect would 14 disappear by the end of the test.
He's saying it's possible 15 it may have disappeared in 10 minutes.
If they put it in at 16 the end of the test and if it only was effective for 10 17 minutes, it might have disappeared by the end and would have 18 no effect on the leak rate, if I understand the testimony.
I 19 MR. ROCKWELL:
I would ~1ike to go to your l
20 observation, your Honor.
Even if that is true, that would be l
l 21 observable, too, from the trace.
1
(
22 JUDGE CARPENTER:
That's my point.
I 23 MR. RUSSELL:
That's the point I was making.
I 24 There were logged hydrogen additions that occurred during the 25 test that appeared to have no effect on the makeup strip ACE-FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coserage 8m3346M6
28039.0
-'9RT 1060 1
chart indicator and that might have been either because there 2
wasn't a water slug in the reference leg of that instrument 3
or for some other reason that we don't know.
It just was not 4
effective.
5 MR. CAPRA:
Judge, I don't want to complicate it 6
later.
I realize there are two level detectors.
The 7
instrument that's feeding the computer is different than the 8
detector that you are looking at on the makeup strip chart.
9 So you may add hydrogen and it could possibly effect the 10 level transmitter that you are looking at on the strip chart, 11 it may very well not affect the level transmitter that is 12 feeding the computer, and vice versa.
,_s t
)
13 Now, you do have some key to that in that you can 14 plot on the strip chart the computer printed values for 15 makeup level to see whether they match or not.
16 JUDGE CARPENTER:
You can look at the computer 17 data set -- as I understand it, there's only three readings 18 and three at the end, so you can't get the time course over 19 the hour out of the computer data set; is that correct?
You 20 can't look for offsets?
{
21 MR. CAPRA:
That's true.
You can only get the 22 beginning and ending points that the computer read and l
23 averaged.
For example, if the makeup strip chart shows a 24 beginning level of 60 inches and an ending level of 58 25 inches, that's the strip chart, and the beginning level ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8@336-6M6
28039.0 e'7RT 1061 b
1 printed out on the computer is 60 inches and the ending level 2
is 63 inches, the ending level between the level detectors, 3
there's a 5-inch difference.
Something happened to one level 4
detector that did not happen to the other.
5 Now, if during that period you had a logged 6
hydrogen addition, we made the assumption that the reason for 7
that disparity was the hydrogen addition.
8 JUDGE CARPENTER:
And that one detector had loop 9
seal and one didn't?
10 MR. CAPRA:
We didn't make a determination whether 11 there was or was not a loop seal but that would be a logical 12 conclusion.
f'\\
Y 13 JUDGE KELLEY:
I want to express a little concern 14' at a critical point.
We have chewed on it rather 15 thoroughly.
We've got a lot of other questions.
Are we 16 getting a little bit out of shape by getting into the 17 differential instrumentation at this point?
18 MR. CAPRA:
I don't think so, Judge.
It's my l
19 opinion -- the reason being it's going to become rather 20 critical.
We have an awful lot of questions posed by 21 numerous employees on why we judged certain tests the way we 22 did.
23 JUDGE KELLEY:
I understand.
They are coming 24 later, aren't they?
fg 25 MR. CAPRA:
Yes, sir.
But this is setting the u
ACE-FEDERAL REPORTERS, INC.
202-347-370n Nationwide Coserage 800-336-6646
28039.0 g RT 1062 V
1 ground rules that we won't have to explain as we go through 2
all the individual tests.
3 JUDGE KELLEY:
We have now spent an hour on 4
question 249 and I'm just getting a little concerned about 5
how this is all going to fit together.
6 JUDGE CARPENTER:
My assessment is that the hour, 7
plus the filed materials, perhaps will allow the board to 8
make findings.
9 Question 250 continues --
10 JUDGE KELLEY:
I want to just add an hour and one 11 minute.
Mr. Stier's drawing which you described in some 12 detail, I might propose the same approach that we had before,
(h 13 that this become an exhibit.
Perhaps Mr. Stier could go 14 home.
You might add a caption.
When I first looked at it I 15 thought it was a loop seal and then you explained it was a 16 strip chart.
In any event I understood your discussion.
17 Could we have this as the next numbered exhibit in order to 18 accompany the discussion in the transcript?
19 MR. GEPHART:
Judge Kelley, among Mr. Stier's 20 talents I believe is not included the fact that he's an 21 artist.
We also have studied all of the strip charts and, 22 although this particular one might -- as accurately as he 23 could -- represent the particular leak rate test that he was 24 trying to portray up there, the majority of these tests, I 25 think, are more saw-toothed and are not as clear.
I think C>
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 863364M6
28039.0 1063 V( 7RT 1
Mr. Stier drew that to show a principle.
2 JUDGE KELLEY:
He did.
That's my understanding, 3
to illustrate what he was talking about.
4 MR. GEPHART:
Since the actual strip chart that he 5
attempted to copy is part of the report, I don't see any 6
necessity for putting this drawing in.
7 JUDGE KELLEY:
Is that just your attempt to draw 8
that particular one you were referring to?
9 MR. STIER:
I used that as a model but this is 10 really a simplified version of it in order to illustrate the 11 point I was making.
The one in figure 3 does not have the 12 lines on it depicting the offset or the different points for 7-13 the beginning and end of the tests.
I suppose you could --
14 it's not that complicated a drawing.
You might be able to do 15 it yourself on a copy of what is in the report.
16 JUDGE KELLEY:
We'll get along without it.
Go 17 ahead.
18 JUDGE CARPENTER:
Question 250:
Decreases in the 19 makeup water level during leak rate tests resulted from 20 leakage during the tests.
Was there a significant amount of 21 identified leakage at TMI-2 in early 1979?
22 MR. KIRKPATRICK:
Yes.
I think our previous 23 testimony indicates this is clearly the case.
24 MR. ROCKWELL:
Yes.
There was a rising trend of 25 identified leakage.
O ACE-FEDERAL REPORTERS, INC.
202 347-3hW)
Nationwide Coserage 8%3364a4 i
28039.0 e'7RT 1064
~]
1 JUDGE CARPENTER:
Mr. Stier?
2 MR. STIER:
Yes.
3 JUDGE CARPENTER:
Question 251:
The MPR report 4
analyzes the impact of the loop seal error on a leak rate 5
test during which there is a decrease of 12 inches in the 6
makeup level.
The error results in "an overstatement of --
7 the unidentified leakage by about 2 gallons per minute."
Do 8
you agree with this analysis?
9 MR. KIRKPATRICK:
I haven't reviewed this 10 particular example.
However, I agree that a 2 gallon a 11 minute decrease is quite possible.
12 MR. ROCKWELL:
Our analysis is very similar.
Page 13 77 of volume II of our report, we conclude that a 200 gallon 14 water addition under normal plant conditions would produce a 15 4.4-psig increase in makeup pressure with the resulting i
16 displacement of the slug of 2.2 inches, yielding a 1.14 17 change in gpm rate.
18 The 200-gallon water addition is equivalent to l
19 about a 6-inch change in the makeup level, and if you double
(
20 that, our figures and the MPR figures are very close, l
21 MR. MC BRIDE:
Did he mean to say an increase or 22 decrease of 2 gallons per minute?
l 23 MR. KIRKPATRICK:
I meant to say an increase.
I I
l 24 misspoke in that case.
l l
S 25 JUDGE KELLEY:
Thank you.
)
(
i l
ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nanonaide Cmerage 800-336-6646
28039.0
~3 R T 1065 1
MR. ROCKWELL:
The only other thing I would offer 2
is that our Exhibit 33 offers the mathematical formulas that 3
we used to derive these numbers.
4 JUDGE CARPENTER:
Mr. Stier, do you have any 5
comment?
6 MR. STIER:
I would like to direct your attention 7
to the section of our report that deals with that analysis.
8 That is in volume IV-B, page A-31.
9 JUDGE CARPENTER:
Thank you.
10 Question 252:
Did the NRC review the design of 11 the TMI-2 makeup level transmitter.
Was such a review 12 required prior to the issuance by the NRC of the operating 6
1
%s) 13 license and technical specifications for TMI-2?
14 MR. KIRKPATRICK:
The NRC did not review the 15 design of the TMI-2 makeup level indication system until 16 after Hartman's appearance on WOR-TV.
Such review was 17 required for licensing TMI-2, 18 MR. ROCKWELL:
This was not a part of the Faegre &
19 Benson investigation.
20 MR. STIER:
This was not one of my questions.
21 JUDGE CARPENTER:
Question 253:
Was the state of 22 the art in RCS leakage detection in 1978 such that the cause 1
23 and impact of the loop seal error in the TMI-2 makeup level 24 transmitter should have been apparent to the NRC?
Second g-25 question, to Met Ed?
v ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cos erage 800-33M646
28039.0
<~9RT 1066 b
1 MR. KIRKPATRICK:
NRC inspections are audits, and 2
normally they don't get into the details of all such 3
installations.
NRC did, however, discover the loop seal 4
error after the problems with leak rate testing were publicly 5
identified by Mr. Hartman.
6 The licensee should have followed up on the 7
problems that the operators were having with makeup level 8
detectors.
Several operators and some shift foremen and 9
shift supervisors were aware that hydrogen did affect the 10 makeup level, even though it should not -- even though there 11 was no reason to believe that it should have done so.
12 Also, maintenance by instrument technicians p_
U 13 identified water in the dry loop reference leg as the source 14 of instrument inaccuracy.
Thus, there were several 15 indications of the loop seal problem which Met Ed should have 16 investigated but did not.
l 17 MR. MC BRIDE:
Judge Carpenter, I would like to 18 ask that you reask the first half of that question to i
l 19 Mr. Kirkpatrick, on the ground that the answar that he read 20 was not responsive with respect to the question as posed to 21 the NRC's actions.
He simply responded about what they do in 22 the course of their routine audits.
It seems to me quite
[
23 apparent that there are many people in this agency with an 24 enormous amount of education and experience in this industry 25 and the question is with respect to the state of knowledge in i
g ACE-FEDERAL REPORTERS, INC.
202 347-370' Nationwide Coserage 800-1 4 6646
28039,0 d(-7RT 1067 1
the industry, state of the art, and not what some inspector 2
does in the course of a routine audit.
i l
3 l
JUDGE KELLEY:
I'm going to interject here some 4
, points.
Just a few minutes ago, Mr. McBride, you stepped in 5
to point out that Mr. Kirkpatrick had misspoken.
I think 6
that's perfectly appropriate.
We did it right then and 7
there.
O j
'Our normal format here is follow-up, not 9
interjection, not that an answer wasn't responsive, and I 10 think if we stay on this road, the next thing you know we'll 11 be out of follow-up and into something rather different.
I 12 suggest you suggest a follow-up question along those lines, g-(_)
13 if you wish.
14 MR. MC BRIDE:
Very well.
15 JUDGE CARPENTER:
Mr. Rockwell?
Was this also a 16
, question for you?
17 MR. ROCKWELL:
Yes, it is, your Honor.
We have 18 come difficulties with the question, in the way it ic 19 l phrased.
Let me try rephrasing it and answer it as I think 20 it is possible for us to answer it.
I i
21 l
We did not investigate the state of the art of RCS I
22 leakage detection.
I think it would be fair to say that the 23 engineers on our investigation team felt that the potential 24 for a loop seal in the design of system piping was well I
25 recognized and that to the extent that the existence of a ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage RXL336-6646
28039.0
(~7RT 1068 U
1 loop seal created the potential -- excuse me -- the existence 2
of the "U" shape in the piping created the potential for a 3
loop seal was something well recognized and had it been 4
focused on, would have resulted in perhaps some changes.
5 JUDGE CARPENTER:
Well recognized by whom?
6 MR. ROCKWELL:
Either the designer, Metropalitan 7
Edison, or the NRC, I would presume.
In other words, any 8
engineer looking at that configuration would recognize the 9
potential for a problem.
10 JUDGE CARPENTER:
Well, I might feel -- I might 11 say I would hope so.
What I was asking:
Did you have any 12 evidence that there was any review, there was any real O
13 identification of the problem?
14 MR. ROCKWELL:
No.
And in fairness that's simply 15
-- that simply was not a part of our investigation.
We did 16 not go back to look at the design review by anyone of that 17 system.
18 JUDGE CARPENTER:
Mr. Stier?
19 MR. STIER:
I also did not research the state of 20 the art during that time period.
There was an awareness of 21 an error in makeup level readings.
That is, a difference 22 between the reading on the strip chart and batched water 23 additions, in that there was an effect of a hydrogen addition 24 which everybody understood should not have been there because 25 of the nature of the level instrumentation.
There were some gxV ACE-FEDERAL REPORTERS, INC.
Nationwide Coserage 800-336-6M6
._.-..~
28039.0 1069 f-}RT v
1 who recognized that in the operations department at TMI-2.
2 JUDGE CARPENTER:
Say that last, again?
3 MR. STIER:
There were some members of the 4
operations department of TMI-2 who understood that there was 5
something wrong with the instrumentation, creating the errors 6
in strip chart traces that I have described.
7 Now, whether -- I didn't find anybody who knew or 8
believed that that was caused by a loop seal condition that 9
we have been talking about here.
They just knew that there 10 was an error in the instrumentation.
11 MR. KIRKPATRICK:
I would like to add one comment 12 regarding the state of the art.
I have discussed the matter 13 with a number of instrumentation people.
This was a well 14 recognized problem.
15 JUDGE CARPENTER:
Yes?
16 MR. RUSSELL:
I don't want to further confuse the 17 record, but my own experience prior to this time was such 18 that it was more common to use a wet reference leg with a 19 system that also had water on the other side of the detector I
20 for measuring fluids; or use a dry variable leg and a dry 21 reference leg by putting some type of a bellows on the bottom 22 of the tank, such that you did not have a situation where you 23 had, typically, a wet variable leg and a dry reference leg.
l 24 Because it's very difficult, in doing DP cell calibrations, l
25 to keep, if it's a wet reference leg, air out of it when you
! O ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nanonwide Cos erage 8m336-6646
28039.0 e~3RT 1070 1
detach the DP cell tester, or in the case where it's a dry 2
reference leg, since the testers use water, to get the water 3
back out of the dry reference leg after you have done the 4
detector calibration.
5 So, I feel that the situation was a little more 6
unusual, to have a dry reference leg with a wet variable 7
leg.
But it is clear that either gas in a wet reference leg 8
or water in a dry reference leg would affect level indication 9
and that you should do something with design to preclude 10 that.
11 MR. ROCKWELL:
Your Honor, I just discovered'some 12 additional information that may be useful here.
Page 63, 13 volume II of our report, there's a footnote:
"B&W 14 anticipated the possibility of certain problems with the dry 15 reference leg system and suggested measures to minimize those 16 problems.
In particular, B&W suggested a reference leg with 17 a condensate loop and draining capabilities.
The suggestion 18 was not incorporated in the actual modification as identified 19 in the vendor's instruction manual."
20 There are a couple of references to GPU startup 21
{ problem reports.
Report 2459 and report 2836.
So there is a 22 little bit of information about that in our report.
23 JUDGE KELLEY:
I'll state 254, but first make a 24 contextual comment.
25 253 said:
Was the state of the art such that the O
ACE-FEDERAL REPORTERS, INC.
20244'-37m Nanonwide Coserage 800436-6M6
I I
20039.0 t'
1D71 V]RT 1
NRC should have discovered this earlier than it did?
254, it 2
seems to us, has a rather pejorative flavor and says, in 3
effect, why did it take you so long?
I don't really think 4
the two are different.
5 Mr. Kirkpatrick addressed the NRC part of 253.
Do 6
you want to comment on 254 in the same -- comment on 254, go 7
ahead.
8 MR. KIRKPATRICK:
No, I think my answer would be 9
the same.
Again, the NRC's inspection in this kind of area 10 is generally an audit and does not look at every single 11 installation.
12 JUDGE KELLEY:
When you say " audit," do you mean a 13
-- is " spot check" the right term?
14 MR. KIRKPATRICK:
I would call it a spot check.
15 We look at enough systems to convince ourselves that we think 16 the licensee is doing a proper job in this area and we let it 17 go at that.
18 JUDGE KELLEY:
And "this area" here means -- do 19 you mean instrumentation?
20 MR. KIRKPATRICK:
In the many hundreds of 21 instrumentation installations.
22 JUDGE KELLEY:
But you wouldn't look at each one, 23 one by one?
i 24 MR. KIRKPATRICK:
But we wouldn't look at each 25 one, no.
O, g
ACE-FEDERAL REPORTERS, INC.
ac.n,-
8 _ + c -,.
28039.0 1072 G( 7RT 1
JUDGE KELLEY:
Even though they all do have safety 2
significance in some sense?
3 MR. KIRKPATRICK:
That's right.
4 JUDGE KELLEY:
All right.
Thank you.
5 JUDGE CARPENTER:
Other follow-up questions?
6 JUDGE KELLEY:
If you want to have a little longer-1 7
to make up follow-up questions -- or are you all ready?
U MR. MC BRIDE:
We are ready.
9 JUDGE KELLEY:
Okay.
Good.
10
'i MR. GEPHART:
Judge, are you going to ask 255?
11 JUDGE KELLEY:
We struck 255, as previously 12 indicated, I believe.
13 (Discussion off the record.)
14 JUDGE CARPENTER:
The first follow-up question, 9
15 addressed to Mr. Stier:
Is it true that a " bump" such as you 16 describe, if it appeared on a strip chart, could be caused by 17 something other than a hydrogen addition?
For example, an 10 oscillation in the plant?
If so, would the result be a 19 reduction in the unidentified leak rate, not -- I'm sorry --
20 I'll reread the last part of it -- if so, would the result, 21 1.e.,
A reduction in the unidentified leak rate, not be the 22 same?
23 MR. STIER:
The answer to the first part is that 24 yes, traces can show some rather strange bumps and changes 25 for a variety of reasons, many of which we are not even O-s ACE-FEDERAL REPORTERS, INC.
202-3 A7-37tX1 Nationwide Coverage 800-336-6M6
N 28039.0 c'1R T 1073 U
1 certain of, that nay not be attributable to a hydrogen or 2
water addition.
We doa't think that there is any precision 3
in the way a hydrogen addition or -- let's take a hydrogen 4
addition.
A bump as I depicted on my drawing can be
,u 5
identified as a hydrogen addition, could be caused by 6
something else.
7 If there is, of course, other evidence, like a log 8
entry shows that a hydrogen addition was may at that time, we 1
9 think you can infer that it was a hydrogen addition.
But i
10 absent that, it may be caused by something else.
And as 11 you'll see from your examination of strip charts, as I'm sure
\\
12 you will get into, they they really aren't as regular and as 13 precise au I have drawn it.
I drew it the way I did in order 14 to illustrate a point.
But it's true that there are many y
15 fluctuations, many, changes in a strip chart trace, that are 16 I caused by a variety of factors.
i 17 I
In answer to the second part, yes.
If you happen y
18 to start a test at a low point in the fluctuations on the 19 strip chart trace and it ends up on a high point, regardless i
u 20
'of what causes the low point and the high point -- regardless i
21 of what causes the fluctuation, that fact alone will cause a 22 reduction in your calculated unidentified leakage.
l JUDGE CARPENTER:
The second follow-up question is 23 I
24 addressed to Mr. Kirkpatrick.
If hydrogen were added just 25 prior to atart of the leak rate test and the " bump" created ACE-FEDERAL REPORTERS, INC.
202 4 47-3700 Nationwide Coscrase 800 336e46 t
a
28039.0 d RT 1074 N)
I by the hydrogen in the makeup level were to disappear within 2
45 minutes, would the test depict greater than actual 3
unidentified leakage, assuming no hydrogen were added during 4
the test?
5 MR. KIRKPATRICK:
Yes.
The answer to that is 6
yes.
7 JUDGE CARPENTER:
Thank you.
+
8 MR. STIER:
Judge, may I amend my last answer to 9
one small degree?
If the bump were caused by something that 10 the leak rate test were taking into account, such'as a 11 compensating change in pressurizer, then the calculated 12 unidentified leakage would not be affected.
.I said that the
'~'
13 difference in readings would affect the unidentified 14 leakage.
I should have been more precise and said if the 15 test itself compensated for that because there was a change 16 of pressurizer level, then the -- it would not affect the 17 unidentified leak rate.
18 JUDGE CARPENTER:
That's clear.
19 The next question is also for you, Mr. Stier.
Are 20 you aware that after there was a change in the Unit 2 21 supervision, there was a more prevalent practice of adding 22 hydrogen to the makeup and that the operators were likely to 23 have added hydrogen prior to the tce t, to maintain adequate 24 hydrogen during the test?
25 JUDGE KELLEY:
Can I just note there's a slight 7S
(_/
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 80(L336 N4 '
20039.0 8RT 1075 9
0 1
modification to that question as submitted?
As submitted it 2
contained a reference to a particular party in the case.
We 3
are just reluctant to phrase a question in that way unless we 4
need to.
5 Can you give a date, about?
6 MR. MC. BRIDE:
Yes, your flonor.
The individual I 7
referred to in the question became the Unit 2 superintendent, 8
I believe in either Decemoer 1978 or January 1979.
9 JUDGE KELLEY:
All right.
10 MR. STIER:
I'm aware of what the reference is --
11 has been made to.
12 The individual to whom the question refers did
,l 13 express concern to the operations department that they
'~
14 maintained hydrogen overpressure within the bands that had 15 been established, and he seemed to be -- he seemed to show 16 some serious concern about that requirement.
17 JUDGE KELLEY:
We'll get into that more later.
It 18 just seems at this point we'd rather not focus that much on 19 it.
1 20 JUDGE CARPENTER:
The next question is for 21 Russell, Rockwell, and Stier.
22 In your investigation, did you not determine that, 23 for a substantial period of time, hydrogen additions were 24 made by auxiliary operators at a place remote from the
,- m 25 control room?
If your answer is yes, is it not true that the i
)
i l
ACE-FEDERAL REPORTERS, INC.
202-347-37141 Nationwide Cos erage Rxb3%-6646
29039.0 r~3RT 1076
()
1 accuracy of the logging times could not be verified?
2 MR. CAPRA:
The answer to the question is yes.
3 There was a period of time where hydrogen could not be added 4
from the control room.
I don't think it was just one time.
5 I think there were several times wnere the hydrogen addition 6
capability at times did exist and at other times did not 7
exist from the control room.
8 The only way that we had to indicate whether a 9
hydrogen addition -- at what time a hydrogen addition was 10 made was either by the control room operator's log or the 11 auxiliary operator's log.
12 It is true that in either case, whether it was
' ~ '
13 logged in the operator's log or the auxiliary operator's log, 14 that's not a definite indication that hydrogen was, in fact, 15 added at that exact moment.
There is even more question when 16 it happens to be added by the auxiliary operator.
- However, 17 that does not mean that it is always uncertain.
For example, 18 the auxiliary operator called back to the control room to 19 inform the control room operator that he added hydrogen --
20 this is speculation on my part -- he would inform the 21 operator at what time he logged the hydrogen addition.
22 JUDGE CARPENTER:
The question erds up saying, 23 "could not be verified."
24 MR. CAPRA:
That's correct.
25 JUDGE CARPENTER:
Mr. Rockwell?
O<
ACE-FEDERAL REPORTERS, INC.
1 nn-
~ _ c -,.
_L l
28039.0 r-BRT 1077 I%s) 1 MR. ROCKWELL:
I think the easiest way to answer 2
the question is that, yes, there is a degree of uncertainty 3
with respect to those times.
Hydrogen was being added by 4
auxiliary operators during certain periods.
For example, we 5
found a hydrogen addition, referring to page 70 of volume II 6
of our report, which appeared to have occurred at 59 minutes 7
into the test and, in fact, we noted in the footnote that we 8
were not certain that that in fact had occurred during the 9
test because of the very close margins on time; and our 10 inability to know how meticulous or fussy they were about 11 recording them at the exact minute that they were made.
12 JUDGE CARPENTER:
Mr. Stier?
13 MR. STIER:
Our experience is the same as 14 Mr. Capra and Mr. Rockwell.
15 JUDGE CARPENTER:
The next follow-up question is 16 for Mr. Russell.
Is there any requirement in the technical 17 specifications or procedures that an operator examine the 18 makeup strip chart before submitting a leak rate test for 19 approval?
20 MR. RUSSELL:
No.
21 MR, ROCKWELL:
Do you want an answer from me on 22 that, your Honor?
23 JUDGE CARPENTER:
It's only addressed to 24 Mr. Russell.
~g 25 JUDGE KELLEY:
You can comment, if you disagree, J
ACE-FEDERAL REPORTERS, INC.
c_
28039.0 RT 1078 1
particularly.
2 MR. ROCKWELL:
No, I do not disagree.
3 JUDGE KELLEY:
All right.
4 JUDGE CARPENTER:
There was a second part of the 5
question which said:
If the answer is yes, then this second 6
part applies.
Since you answered no the second part of the 7
question is not operative.
8 JbbGE KELLEY:
This is a Staff question for 9
Mr. Kirkpatrick.
It may have to be tied in somewhat but, 10 I'll read the question.
It's a homey question.
Does every 11 home with plumbing contain leak seals?
Please give an 12 example.
13 MR. KIRKPATRICK:
Yes, I would think it would.
14 Example, of course, would be the traps in the various 15 commodes and sinks.
16 JUDGE KELLEY:
Are there loop seals in home 17 plumbing that are analogous to the kind of loop seal we have 18 been talking about?
19 MR. KIRKPATRICK:
Yes, I believe they are 20 analogous.
Of course there are no enclosed systems, 21 generally, that would cause pressure that would move these 22 seals around, as the case we are talking about.
23 JUDGE KELLEY:
I'm not quite sure of the thrust of 24 this.
Is the suggestion, then, that any plumber should have 25 been able to figure out what this problem was and, es x
ACE-FEDERAL REPORTERS, INC.
Nationwide Coserage 800-33MM6
28039.0 RT 1079 1
therefore --
2 (Laughter.)
3 MR. KIRKPATRICK:
I think it's a little bit more 4
complicated than your home plumbing.
But I think it is true 5
that any competent instrument design engineer should have 6
realized that accumulation of water in this -- in a low spot 7
would have caused instrument problems.
8 MR. MC BRIDE:
Judge Kelley, may I observe, to the 9
best of our knowledge we don't represent any plunbers.
10 JUDGE KELLEY:
Okay.
Maybe it has a bearing, 11 actually, though.
I was asking you earlier about the kinds 12 of things you review and the kinds of things you don't
-s 13 review.
Now, here is something that has safety significance, 14 but is it of such an order, a design order that you just 15 assume that the licensee can handle this kind of thing?
16 MR. KIRKPATRICK:
I think the answer to that is 17 yes.
Normally the details of how you route the instrument 18 tubing from the instrument to an appropriate space that is 19 supposed to be measured is left up to the field 20 installation.
21 MR. RUSSELL:
I can give one example that I'm i
22 aware of, where the NRC is generically interested in actual 23 routing of reference legs and how it has a significant safety 24 bearing, to maybe put this in context.
25 The instrument is the water level detector for a 7
ACE-FEDERAL REPORTERS, INC.
W 147171al Nationwide Coserage 800-33466&6
28039.0 jes3RT 1080 b
1 boiling water reactor.
Because the reference leg has to come 2
from the reactor vessel out through the dry well of the 3
containment and subsequently out to the detector, there is a 4
potential for a difference in temperature or density in a 5
portion of the reference leg, namely that portion which is 6
inside the dry well, between the reactor vessel and the 7
inside wall of the dry well.
8 There are some installations where there is an 9
elevation change in that reference leg.
While that would not 10 be significant during normal operation if you are calibrated 11 for that tempera ture, during an accident, if you raise the 12 temperature in the containment, you would change the density O
V 13 and, therefore, have a potentially significant effect on the 14 indicated level.
15 As a result, we have required such routings to be 16 changed, such that there is no change in elevation in that 17 portion of the reference leg which is within the dry well 18 because of this effect.
19 Now, that's a wet reference leg but it's an 20 example of a case that I'm aware of where the Staff is 21 interested in actual routing of instrumentation lines.
22 JUDGE KELLEY:
But the example you cite where you 23 do scrutinize what is done is the exception to the rule, if I l
l 24 may put it that way?
l l
fs 25 MR. RUSSELL:
I believe that's the case.
The (J
l l
l ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Cos erage 80lV336-6M6
28039.0 M RT 1081 V
1 others are done pretty much by an audit, as Mr. Kirkpatrick 2
described.
3 JUDGE KELLEY:
Right, there but there's not 4
systematic pipe-by-pipe, system location review.
There is 5
the audit.
6 MR. RUSSELL:
That's correct.
7 JUDGE KELLEY:
Just one thing I'll raise before we 8
break here, off the record.
9 (Discussion off the record.)
10 JUDGE KELLEY:
We'll take a break for 10 minutes.
11 (Reces s. )
12 JUDGE KELLEY:
Back on the record.
O
13 JUDGE CARPENTER:
Mr. Russell, the board would 14 like to follow up on a follow-up quest 2on a little bit, as 15 part of our learning curve.
You were asked is there any 16 requirement in the tech specs or procedures that an operator 17 examine the makeup strip chart before submitting a leak rate 18 test for approval and yoer answer was no.
19 I believe the testimony is that the procedure l
20 calls for making such a test under steady state conditions.
21 It occurs to me that, inherently, one of the best ways of 22 judging whether steady state conditions prevailed during the 23 one hour in which the test was run,-is to examine the strip 24 chart, see whether there are changes in the slope of the 25 water level over that hour that are large enough to violate 7sU ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8(XA3364M6
28039.0 1062
[v%SRT
\\
I the common-sense criteria of what is called for in the 2
procedure, namely steady state.
3 Could that be interpreted that way or not?
4 MR. RUSSELL:
I interpreted the question much more 5
narrowly than that, as to whether I was aware of any 6
affirmative requirement that would direct an operator to 7
examine the makeup strip chart prior to submitting a test.
8 There are general references that would require 9
operators to monitor their instrumentation while they are 10 operating the facility.
There are generally standing 11 instructions on how you operate the plant.
There are 12 instructions in administrative procedure 1036 with respect to
(~)
13 instrumentation and how you use instrumentation.
And it is 14 true that they needed to have steady state conditions.
But 15 there are other ways of determining steady state conditions, 16 for example looking at other instrumentation, traces of 17 reactor power, et cetera.
18 So, I was not excluding that he would look at it, 19 and, in fact, many operators testified that they did use it.
20 I was stressing it only in the very narrow sense that it was 21 a written or specified requirement someplace.
22 JUDGE CARPENTER:
I took your answer that way.
In 23 a more technical sense --
24 MR. RUSSELL:
Operator practice is that it's a 25 very important instument that they do use because it gives O-
'T ACE-FEDERAL REPORTERS, INC.
202-347-37(x)
Nationwide Coserage 8(XL336-6M6
.28039.0
(~)yRT 1083
'n.
I them an indication of time and what is happening to the 2
3 JUDGE CARPENTER:
Does that change your answer to 4
a somewhat equivocal "yes"?
5 MR. RUSSELL:
It is not -- well, I would take 6
interpretation of existing operating instructions to say 7
explicitly that it is required.
I believe that it is 8
intended that the operators monitor their instrumentation 9
during the course of a watch and that they do review those 10 indications.
That's operating practice and operators did 11 testify that they observed the makeup strip chart trace.
I 12 took the question to be very narrow, in the context of:
Was lO 13 there any affirmative requirement to do that immediately 14 prior to turning in a test record that had been signed, for 15 instance?
For instance, the operator signing the tank, 16 should he review the makeup, review the pressurizer level, 17 review the reactor power and other indications prior to 18 signing it?
I think it's implied that the plant must be at 19 steady state and he must look at whatever he looks at to 20 determine that he had met that requirement.
21 JUDGE CARPENTER:
But isn't it true that a record 22 with a bump in it would imply nonsteady state?
23 MR. RUSSELL:
Not necessa...y.
We have seen bumps 24 that were cause by shifting main feedwater pump or other fx 25 things which were momentary and which were offset by changes e>
ACE-FEDERAL REPORTERS, INC.
202-347-37(n)
Nationwide Coserage 8(43%6M6
28039.0 r-RT 1084
(-
1 in pressurizer level that we did not feel invalidated a 2
test.
3 JUDGE CARPENTER:
Thank you.
I was hoping it was 4
this simple.
Apparently it is not.
5 MR. MC BRIDE:
Since there was some elaboration on 6
that answer I was about to write a follow-up question.
7 JUDGE KELLEY:
Go ahead.
8 MR. MC BRIDE:
Should I state it orally or write 9
it and bring it to you?
10 JUDGE KELLEY:
Just state it 01 ally.
11 MR. MC BRIDE:
I was going to ask the board to ask 12 Mr. Russell if, whether expressly required or simply implied O
13 or good operator practice or whatever he would want to 14 characterize it as, if the operator, by review of whatever 15 instruments in Mt. Russell's judgment he should have reviewed 16 concluded that the plant was not in steady state, would that 17 invalidate the test performed during that period?
t 18 MR. RUSSELL:
The answer to that is clearly yes.
19 We have examples of where that did occur.
And there were 20 several tests which we felt were invalid because of that.
We i
21 did not feel that that was something that was intentional.
22 We felt that was more an oversight.
So we were not able to 23 make the next step of a wi11 fulness, because there wasn't a i
24 pattern.
g 25 JUDGE KELLEY:
Just a small question.
The strip
%)
ACE-FEDERAL REPORTERS, INC.
2fCd47-37m Nationwide coserage 84n336-6646
28039.0 RT 1085 1
chart we have been speaking of, that is -- it is printed out 2
in the control room; is that correct?
3 MR. RUSSELL:
That's correct.
Since we entered 4
the photographs the other day we did obtain a photograph of 5
the strip chart recorder and where it is.
That was not on 6
the instruments.
But it is in the control room and it is 7
observable, as is pressurizer level, power, the other charts 8
that we are talking about.
9 JUDGE KELLEY:
You say it isn't in the photographs 10 we have?
11 MR. RUSSELL:
It's not -- it's there, but it's not 12 as clear.
We have an 8 by 10 enlargement because of 13 questions that came up about it.
14 MR. GOLDBERG:
Judge Kelley, we are having copies 15 made of a photograph which shows that and we'll offer it for 16 the record when we get the copies.
It's in the process of 17 being duplicated.
18 JUDGE KELLEY:
Okay.
I think that's helpful.
19 I was wondering -- it's follow-up, really -- the 20 ease at which an operator running a test could consult the 21 strip chart, even while the test is running.
Let's suppose c
22 i that I'm an operator who does want to manipulate the result t
23 i by adding some hydrocen, let's say.
So, halfway through the i
24 l test I do that.
Can I glance up at this needle and see the f3 25
! thing jump richt then and there?
If what I'm trying to do
(
ACE-FEDERAL REPORTERS, INC.
m_
x_.~.
(
28039.0 e
1086 f ')3RT 1
comes about, if there's a loop seal down there and it works, 2
so to speak, you could see that instantaneously?
3 MR. CAPRA:
Yes, sir.
You can.
May I refer you 4
to one of our pictures that we have submitted on the record?
5 JUDGE KELLEY:
Yes.
6 MR. CAPRA:
9-A, Exhibit 9-A.
7 JUDGE KELLEY:
That's the sort of panoramic view 8
of the control room; correct?
9 MR. CAPRA:
Yes, sir.
10 MR. STIER:
Isn't that 9-B?
11 MR. GOLDBERG:
Do it on both.
Point it out.
12 JUDGE KELLEY:
Off the record for a minute until A
~
13 we straighten this out.
14 (Discussion off the record.)
15 JUDGE KELLEY:
Back on the record.
Mr. Capra is 16 going to help us out by pointing to the large blowup of what 17 we have been calling -- he's looking at 9-A, I believe.
18 MR. CAPRA:
On 9-A, right here, to give you a 19 perspective of the control room, looking into the control 20 room, the area that we are talking about is over a little bit 21 to the left of center, this area I'm circling with the 22 pointer right now.
That's the controls for the makeup and j
23 purification system.
I 24 On the panel in the middle, which is the vertical
(
25 panel, there are two indicators first.
The first one is i
1 ACE-FEDERAL REPORTERS, INC.
202.147-370)
Nat sonwide Cose* age 8043%MW)
I 28039.0
-BRT 1087
.]
1 makeup pressure and makeup temperature.
The next one over is 2
the strip chart recorder for makeup level.
This is what we 3
have the picture of, that's coming this afternoon.
4 This is the same panel on Exhibit 9-B.
All the 5
way over to the right on this vertical panel here are makeup 6
pressure and temperature, and this is the strip chart 7
recorder for makeup level.
8 The switch underneath the recorder is the level 9
selector switch for selecting the level transmitter that is 10
-- provides input to the strip chart recorder.
11 For periods where hydrogen could be added to the 12 makeup from the control room, the button right down here to 13 the extreme right on the sloping section of the panel is 14 where you added hydrogen, with your indication of hydrogen 15 pressure right above it.
16 The controls for adding water to the makeup are 17 down in this section right here.
l 18 JUDGE KELLEY:
Thank you.
The makeup strip chart, 19 does that just continuously feed out a strip like an AP wire I
20 machine or an electrocardiogram or something like that?
l i
21 MR. CAPRA:
Sort of like a VCR.
It is 22 sprocketed.
There are two reels in there and one of them is l
23 a takeup reel.
j 24 JUDGE KELLEY:
But if I wanted to check what 25 happened in the past hour, I could just -- do what?
O l
l ACE-FEDERAL REPORTERS, INC.
l 202-347 3M)
Nationwide Coserage tok3346M6
28039.0
-4RT 1088 1
MR. CAERA:
You can pull the recorder out.
In 2
other words, you can drop this top part down and actually 3
pull the recorder out so that you can observe several hours.
4 JUDGE KELLEY:
Thank you, t
5 MR. MC BRIDE:
Judge Kelley, on that very last 6
point that Mr. Capra made, I wonder if you would ask him, as 7
a follow-up question, whether it was generally desirable or 8
undesirable for the operators to open the face of the strip 9
chart and pull out the paper necessary to observe a long-term 10 trend in the makeup level?
11 JUDGE KELLEY:
Okay.
You heard the question.
12 MR. CAPRA:
I did.
I don't know the answer to O
13 that.
I would think the answer was it was not desirable.
7 14 JUDGE KELLEY:
Why?
15 MR. CAPRA:
By pulling your recorder in and out 16 you would be putting perturbations on the system that could 17 affect the level indication.
I'm not positive on this but I 18 believe there were times, due to heating problems, whatever, 19 heating, ventilation and air conditioning problems, where 20 they did run with certain recorders pulled out, but not for 21 the purpose of observing long-term traces.
22 JUDGE KELLEY:
At the moment, if I don't pull it 23 out all I can see is the instantaneous reading taking place, 24 right, through the window?
25 MR. CAPRA:
I don't believe that's the case.
You ACE-FEDERAL REPORTERS, INC.
2112 347-370 )
Nationside Cmerage sn33MM6
,, ~ _, _
28039.0 r-SRT 1089 N]
1 can observe at least an hour to two-hour period of time.
2 JUDGE KELLEY:
Just by looking at it without 3
pulling it out?
4 MR. COLE:
I have been in many control rooms and I 5
have seen many good operators have those pulled out so he can 6
see a longer trend.
I have walked into the control rooms and 7-see them commonly pulled out so he can see that picture.
8 MR. MC BRIDE:
Would you, Judge Kelley, give me 9
leave then to ask whether, in doing so, it might cause a time 10 shift in the strip chart and require recalibration of the 1
11 paper?
12 MR. CAPRA:
I don't think that that's the case, b
\\~
13 the reason being that the strip chart is sprocketed such that 14 there is no slip in the paper.
It causes a pen deflection.
15 JUDGE KELLEY:
Mr. Russell, do you have any 16 comments?
17 MR. RUSSELL:
Yes.
As was indicated earlier, 18 there are a number of facilities that do, indeed, operate 19 with these out.
I'm aware of one that does it because of 20 problems with the pen sometimes sticking on the side so they 21 withdraw the drawer and operators do look at them on the 22 side.
23 To the best of my knowledge, that was not done for 24 long periods of time at TMI, and I testified earlier, when 25 the comment came up about looking at the rate of change of p
QJ ACE-FEDERAL REPORTERS, INC.
2tc-347-37m Nationwide Coverage NOL33MM6
28039.0 g-SRT 1090 O
1 the makeup over time, that that would have required an 2
operator action to open the instrument to review the side.
3 That might give him as much as six to eight hours.
Otherwise 4
you have to do an after-the-fact analysis of rolling out the 5
strip charts, which typically was not done by ( perators.
6 Mostly when they change the rolls, that just went 7
off to record storage someplace.
So I'm not aware that 8
operators looked at any significant periods of time on the 9
strip chart or did any analysis of that.
They would have, 10 typically, an hour, maybe a little more than an hour 11 displayed to them normally and they could open it to see up 12 to, maybe eight hours.
I just don't know how frequently that
_s
(
13 was done.
14 JUDGE KELLEY:
Do you know for a fact, or do you 15 recall, whether, during the time period that we are looking 16 at or the time frame of your investigation, whether this 17 instrument was run open or closed, I think you said?
10 MR. RUSSELL:
My recollection is there were times 19 it was, but I would suggest it would be better to direct that 20 question to operators, as to whether they did or did not pull 21 the instrument out.
22 JUDGE KELLEY:
All right.
All right.
I think the 23 whole point of this is to determine whether this really was a 24 readily available record one could quickly and easily see or x
25 whether it was not.
And the answer, I gather, is really kind ACE-FEDERAL REPORTERS, INC.
w2.mm
~ _ m o_,.
f 20039.0 SRT 1091
]
1 of mixed.
2 Anything else, Mr. Russell?
3 MR. RUSSELL:
There are Jome indications that it 4
had to be.
For example, if you are going to write or make an 5
annotation on the chart, you must open it up.
Whether they 6
also pulled it out at the time they opened it up is not 7
known.
And there was testimony we received that operators 8
did look for trends on this makeup strip chart.
But as I 9
indicated earlier, it would be better to ask the operators 10 that directly, as to whether that was a routine practice, to 11 operate with it open or not.
We did not ask that specific 12 question during our investigation.
13 JUDGE KELLEY:
Thank you.
That clarifies it for 14 us.
15 JUDGE BRIGHT:
Russell and Capra.
256, I l-16 believe.
We have a new section called " reactor coolant 17 system oscillations."
18 256 is:
What was the typical range of the 19 oscilla tions in RCS temperature, pressurizer level, and 20 makeup level at TMI-2?
21 NR. RUSSELL:
The change of o.cc111ations was not 22 independently calculated by by myself and Mr. Capra.
What we 23 did is we reviewed the work that had been done by 24 Mr. Kirkpatrick and also the work that was done by Faegre &
25 Benson at Exhibit 37, and in the Stier report, volune IV-A at O
i 1
l ACE-FEDERAL REPORTERS, INC.
l I
m,.3 m,
~m_ m m_,.
28039.0 RT 1092 1
111.13; then observed makeup level oscillations to confirm.
2 We found that oscillations in reactor coolant system 3
temperature were plus or minus a half degree Fahrenheit; 4
pressurizer level, plus or minus 2.5 inches, and about plus 5
or minus 1.5 inches in makeup level.
As I indicated, the 6
oscillation of the makeup level is consistent with the strip 7
charts that we reviewed.
8 JUDGE KELLEY:
Mr. Rockwell?
9 MR. ROCKWELL:
Since we are in a new section now, 10 your Honor, let me give you the references to our report.
11 Pages 25 and 26 of our summary, volume I; chapter 6 of volume 12 II and Exhibits 36 through 42, relate to the oscillation 13 iacue.
14 With respect to the particular question, it is 15 Exhibit 37 and we have plotted all of those parameters on 16 separate pages of Exhibit 37.
I'm going to be expressing 17 numbers as a range, not as a plus or minus.
18 For T av, page 1 of Exhibit 37,.75 degrees; for 19 pressurizer level we had a range ot 4 inches as reflected on 20 Exhibit 37, page 7; for pressure, I guess pressure wasn't 21 asked for, but it's a rance of 65 pounds; and for makeup 22 level, page 6 of Exhibit 37, a range of 2.5 inchec.
f JUDGE BRIGHT:
Mr. Stier?
23 i
24 l
MR. STIER:
I would like to offer one ather f reference to our report in addition to the one cited by 25 p
(_-
i ACE-FEDERAL REPORTERS, INC.
I mm.,
~_m~.
i l
I f
28039.0 RT 1093 l
I i
1 Mr. Russell.
It is in volume TV-R, page A-34.
I 2
MPR reviewed essentially the same data that was j
3 included in the Faegre & Benson report and came to 4
conclusions that were almost identical to the ones described 5
by Mr. Rockwell.
j l
6 MR. MOORE:
Just a clarification here that I don't i
l 7
believe was mentioned.
This data was for the few hours on I
8 January 4th and 5th, 1979, as the basis for this.
This is l
j l
9 not a summary over many months.
10 MR. ROCKWELL:
Yes, if I may add we were able to j
11 do the oscillation analysis because of the existence of i
12 reactimeter data for, I believe, the days of January 4 and l
,-s i
(
)
13 5.
Obviously we would have liked to have had that data for i
f l
14 the whole time period.
We didn't have it.
We analyzed the j
f 15 data we had available and then tried to draw some conclusions j
i 16 from that, generally, but recognizing that we are working l
l l
17 with a limited data base.
I i
i 18 JUDGE BRIGHT:
Question 257:
What caused the i
2 l
19 oscillations?
Were they attributable to the same source?
i 20 MR. RUSSELL:
The specific cause of the recorded i
21 oscillations were not investigated by NRC.
However, design 1
22 deficiencies or maintenance deficiencies with the condensate 1
23 and condensate polishing systems were reviewed and they are 24
! documented in a report by Mr. Lucien which is discussed in j
t I
,e m 25 l NUREG-680, supplement 5.
That is believed to be the source l
t l
j
(
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide rmerage N n).tMi646
28039.0 1094 g RT V
1 of the oscillations.
This occurs due to the close coupling 2
between the reactor coolant system temperature, pressure, and 3
steam generator level.
The small volume of water on the 4
secondary side of a Babcock & Wilcox designed plant makes the S
coupling between changes in the secondary side and the 6
primary side very quick.
It has a much smaller water volume 7
than is the case for a Combustion Engineering or Westinghouse 8
plant.
9 We believe as a result of the oscillations in 10 steam generator level, which were be caused by oscillations 11 in feedwater flow, which can be caused by oscillations in I
12 condensate, were a source of the oscillations in that the 7_
(_)
13 system tuning calibration as discussed in the Lucien report 14 may have contributed to that.
There are also entries in the 15 logs that indicate they were having problems with feedwater i
16 heater drains.
Sometimes the drains were directed forward to 17 the steam generator.
Other times the drains were directed l
18 back to the condenser, and there were difficulties in getting l
19 the system to stabilize.
20 So, in general we believe the oscillations were j
21 coming from the condensate and feedwater system on the l
22 secondary side, which were then being reflected into the 23 primary side.
i i
24 JUDGE BRIGHT:
Mr. Rockwell?
j 25 Mk. ROCKWELL:
We did not draw any particular l
}
ACE-FEDERAL REPORTERS, INC.
202-M7-370)
Nationwide roserage 80k 34N46
7.___.._.____.____.._______________________._____
i c
I i
28039.0 3RT 1095 I
1 conclusion as to where the oscillations derived.
In i
l l
2 particular, we did not look in depth at the condensate
(
3 polishers or the feedwater system.
Mr. Moore has already j
4 previously indicated in this hearing, briefly, a view that j
5 the instrumentation and control system and its level of l
l 6
sensitivity, in his view, had something to do with this.
I 7
think that's, perhaps, more a result of his base knowledge 8
acquired not through our investigation but as an engineer i
9 over his experience.
And he can address that in more detail, 10 if that would be helpful.
11 MR. MOORE:
I can give you a real brief summary 12 and then if you have some questions I'll be happy to answer
'O 13 them.
Basically the IC, which stands for integrated control i
14 system, is a necessity in a Babcock & Wilcox type plant in i
15 that you have an once-through steam generator, so there's no i
16 real level measurement to determine what -- how much water 17 there is in the steam generator.
What we have is feedwater j
i 18 comes in one end of the steam generator, it is heated by the l
19 primary water, it exits as steam.
There's no recirculation j
i 20 of water which occurs in other types of steam generators.
21 The result of this, one has this integrated 1'control system that controls the amount of steam going to the 22 23 turbine; controls the amount of feedwater being fed to the 24 I steam generators; it controls the reactor power and any l
25 perturbations that occur.
Any one of these variables will l
ACE-FEDERAL REPORTERS, INC.
2(c M7 37 tin Nationwide Coverage
- n334M6
28039.0 d RT 1096 V
1 affect the others.
So what we've got is a very close coupled 2
plant and all these automatic controls that are controlling l
3 it.
4 It's very difficult to determine exactly which one 5
of these variables is the - potentially the of f ending one, 6
because if one of them is varying it will affect the others 7
and it goes round and round.
8 What we saw in the charts I presented is the 9
effect on T av in the primary system.
I think this was based 10 on the data that we looked at for January 4th and 5th.
I 11 believe this was the driving parameter within the primary
-12 system.
As T av went up and down, the primary pressure 13 responded -- first of all, the level in the pressurizer 14 responded and then as a result of the level change in 15 pressurizer the primary system pressure responded.
16 As to the other side of the plant, the secondary 17 side, the turbine plant side, I did not examine that in 18 detail to try to determine what was the driving force over 19 there.
I suspect, without having looked at it in detail, 20 that it was variations in the steam and feedwater flow, 21 mismatches there, which would be likely to lead to the 22 observed effect on T av.
Just the feedback amongst all the 23 controls could set up this oscillatory behavior.
24 JUDGE BRIGHT:
Is that your answer?
25 MR. MOORE:
If that satisfies your needs, I could O
ACE-FEDERAL REPORTERS, INC.
202147-3RU Nationwide roserage mum 3MM6
20039.0
-QRT 1097 1
stop there.
2 JUDGE BRIGHT:
It sounds quite satisfactory to 3
me.
Mr. Stier?
4 MR. STIER:
We did not investigate the cause of 5
the oscillations.
We think that Mr. Moore's explanation 6
makes sense.
7 Just one note of caution, though.
As he 8
indicated, the reactimeter data that was studied, that formed 4
9 a basis for all..of this oscillation discussion, was for a 10 very short period of time.
And if, in fact, it was caused by 11 the ICS, that may very well are changed over the life of the 12 plant.
They may have fine tuned it so the oscillations might 13 have changed later on toward the latter -- toward March of 14 1979.
We just don't know.
We have no way of knowing.
15 JUDGE BRIGHT:
25C:
How did the oscillations 16 atfect the TMI-2 leak rate test?
17 MR. RUSSELL:
As we discussed earlier yesterday, 18 the oscillations caused high variability in the computed 19 unidentified leak rate.
I don't have anything further to add 20 from what we discussed at length yesterday.
l 21 JUDGE BRIGHT:
Mr. Rockwell?
22 MR. ROCKWELL:
Exhibit 38 is a plot of RCS j
23 leakage, calculated by our -- as a result of our running 24 these 5300 leak rate tests e'c e ry 3 seconds apart during the 25 January 4-5 time period when we had the reactimeter data 4
ACE-FEDERAL REPORTERS, INC.
202-347-3h W)
Nationode (:merage M4))MM6 J
28039.0 r~7RT 1098 k_)
1 available.
If I may, that's the plot.
That's Exhibit 38, 2
page 1 of 9.
(Indicating.)
3 As you can see from looking at the plot, it's 4
basically a sine type of oscillation, which is similar to the 5
oscillation that one is seeing in the other parameters that 6
are plotted and illustrated in these accompanied exhibits.
7 So the pattern is similar, so our conclusion was there was a 8
direct correlation between the oscillation in these plant 9
parameters and the oscillation in the leak rate test 10 results.
11 I should add a qualification.
You see a smoothly 12 oscillating set of results only when you run as many as we
'~
13 did; so that you have, in effect, filled in the points on the 14 curves.
15 If you are taking random tests, you get random 16 results because you don't know where on the oscillation you 17 are going to fall.
So, that qualification is important.
18 MR. MOORE:
Just a point of clarification.
There 19 was an oscillation of the primary plant parameters I 20 indicated, I think it was yesterday, and also in the 21 oscillation of the leak rate.
But there was that 22 interconnection between what I described as far as the 23 beginning of the test, as far as the plant oscillation, and 24 the end of the test, as to where you fell in those 25 oscillating parameters.
So this was a -- the frequency of O
ACE-FEDERAL REPORTERS, INC.
3c-347 3700 Nationaide roserage a n3 4 6646
~ -
l 28039.0 r~pRT 1099 L-)
1 this thing is not -- it's related to the frequency of the 2
primary plant oscillations, but it isn't necessarily the same 4
3 because of the -- when you pick the one-hour test -- if I may 4
go back to the chart again?
5 This would be T av versus time as representative 6
of the primary plant oscillations.
Just to refresh your 7
mind, say for a high leak rate, the test would start here and 8
end down here.
9 JUDGE KELLEY:
Can we show for the record that we 10 are now using what I think is an Exhibit called 11-A to 11 111astrate the point?
Yes.
12 MP. MC BRIDE:
Could you also suggest to him that O
13 when he said "here" and "here" he was referring to the 14 beginning and ending points of the high leak rate test, I 15 believe, just for clarity of the record?
16 JUDGE KELLEY:
Why don't you take it from the f
17 top.
18 MR. MOORE:
I'll start again, i
j 19 At this point we are looking at a high point in T 20 av, peak in the trace.
This line starts here, starting from 21 that high point, f or a period of one hour, ends at a low 22 poir.t in the trace for T av.
It is the fact that during this 23 one-hour period you are spanning from a high point to a low 24 point that I'm not going to refer on Exhibit 11-B.
25 Stepping back to Exhibit 11-A for a minute, you'll O
ACE-FEDERAL REPORTERS, INC.
2f C-347-3700 Nationwide Coserage NXk33M/4
28039.0 r-3RT 1100 U
1 note the time of this particular leak rate is 0314.
2 Now looking at Exhibit 11-B, at that time we 3
observed the high leak rate.
3 4
The point I'm trying to make is that, going back 5
to Exhibit 11-A, the leak rate does not go up and down 6
exactly as the primary paraneters are going up and down.
It 7
is dependent upon the one-hour time span relationship between 8
the peaks and valleys of this curve.
9 JUDGE KELLEY:
When you say " leak rate" do you 10 mean indicated leak rate?
11 MR. MOORE:
In this case the calculated gross, 12 gross leak rate, being shown here on Exhibit 11-B, which we O
13 did not have reactor coolant drain tank data with which we 14 could then go and calculate the unidentified leak rate or --
15 JUDGE KELLEY:
When I say " indicated," I meant 16
" indicated" as opposed to " actual."
17 MR. MOORE:
Yes, sir.
" Indicated" as opposed to, l
18
" actual."
19 JUDGE BRIGHT:
Is that your comment?
20 MR. MOORE:
Yes, that is.
21 JUDGE BRIGHT:
Mr. Stier?
22 MR. STIER:
Our answer wculd be essentially the 23 same as Mr. Moore's.
24 JUDGE DRIGHT:
Thank you.
l l
25 259:
What was the impact of the oscillations in C>
i 1
ACE-FEDERAL REPORTERS, INC.
x.m.,
- s. - u rm.m.
i
3 1 k i
\\
28039.0 J
g-3RT 1101 O
I gallons per minute on the typical leak rate test?
t*'
2 MR. KIRKPATRICK:
I think we have to recognize, as 3
,has already been implied, that the oscillatione varied from 4
time to time.
Sometio2es they were present to a larger degree N
5 than at others.
But I think a good, typical value would be 6
about 1 gpm, when the oscillations did exist, as an average s
7 impact.
.)
'r
'l.
8 MR. ROCKWELL:
We really didn't calculate an 9
average or typical impact.
Our analysis showed that there 10 w % a'rdnge of plus or minus 1.35 gpm, as a result of the a 11 oscillahion.
How one would derive a typical impact would be 12 more pryaerly within Mr. Moore's sphere, but we really didn't "O
13 do it.
14 MR. MOORE:
I(don't believe it is reflected in our 15 1 document but -- so in the past, it seems to me we might have 16 performed t,
but I don't have the data available other than 17 what we observed from here.
t 18 JUDGE BRIGHT:
Mr. Stier?
19 MR. STIER:
MPR has attempted to do a rough 20 -
calculation of the effect on the leak rate test.
Let me make
}
2' a couple of observatione before I tell you what the number is 22 that,they came up with.
First of all, MPR does not believe 23 l that you chn simply add this effect to instrument error.
We 7
24
tried to get a combined effect.
We don't think you can 25, directly add,the effect of oscillations to the effect of o
ACE-FEDERAL REPORTERS, INC.
Q 202147 DH Nationside Coserage 8(n 3)M4M s
t 28039.0 r'7,R T 1102 U
1 instrument error.
2 Secondly, MPR's best judgment is that this may 3
have been a period of high oscillations, and they have some 4
reasori to believe that, as you got toward the end of March, 5
oscillations in the plant were of a much lower level.
They 6
estimated the combined effect of these oscillations and 7
instrument error to be in the range of
.8 to 1 gpm.
8 MR. COLE:
And that is using the more extreme 9
oscillations of the January 4th period.
Other periods where 10 we have seen data, that error would even be smaller.
11 JUDGE CARPENTER:
Does this error always have a 12 positive sign, or can it be plus or minus?
13 MR. STIER:
Plus or minus.
14 JUDGE CARPENTER:
Mr. Kirkpatrick. when you 15 mentioned your number --
16 MR. KIRKPATRICK:
Yes, I'm talking about a plus or 17 minus 1 gpm.
18 MR. ROCKWELL:
I guess I would add that our plus 19 or minus 1.35 is the extreme.
To the extent that we were to 20 calculate a typical value, it would be less than that.
21 JUDGE KELLEY:
But thiaiks numbers are the effect 22 of instrument error plus oscillation.
23 MR. MOORE:
That's correct.
24 JUDGE KELLEY:
And there are other errors around 25 that would impact results.
O(~g ACE-FEDERAL REPORTERS, INC.
202447-3700 Nationwide Coserage 8M336-6M6
28039.0
-9RT 1103 1
MR. COLE:
There are method errors which we 2
discussed yesterday, which is a separate --
3 JUDGE KELLEY:
Right.
Right.
4 JUDGE BRIGHT:
260:
What was the maximum 5
potential impact of the oscillations, in gallons per minute?
6 MR. KIRKPATRICK:
Well, you know, you can "what 7
if" quite a bit.
In this case I assume what they mean is, if 8
you add up all the errors together, what would be the 9
impact?
It's highly unlikely that all of the errors would 10 occur, peak at the same time.
11 But, if you do that you could clearly get errors 12 of several gpm.
/']#
13 JUDGE KELLEY:
Are you talking about all errors or 14 are you still on oscillations?
I 15 MR. KIRKPATRICK:
I'm talking about all of the 16 errors that are caused by oscillation.
17 JUDGE KELLEY:
Okay.
18 MR. KIRKPATRICK:
You could wind up with an effect i
l 19 in an extreme case, I think, of several gpm.
i l
20 JUDGE BRIGHT:
Mr. Rockwell?
21 MR. ROCKWELL:
This is a difficult question to l
l 22 answer.
If the reference is to the oscillations, tc the 23 oscillations that we plotted in the January 4-5 time period, 24 we have already answered the question.
If it's to something 25 else, it depends on what you postulate.
v ACE-FEDERAL REPORTERS, INC.
202-347-37(4)
Nationwide Coserage Mn336-6M6
28039.0 r-7RT 1104 N_]
1 JUDGE BRIGHT:
Mr. Stier?
2 MR. STIER:
We think it's too speculative to try 3
to quantify.
4 MR. MOORE:
Since it is in terms of maximum 5
potential oscillations, really what you need to do to check 6
the maximum potential would be to look at the primary 7
parameters that are varying and determine at what point the 8
operation of the plant would cease.
How far -- how much 9
oscillation can you still have in T av and still operate, how 10 much pressure oscillation can we have -- this really sets the 11 upper bound for the maximum potential.
If we really mean 12 maximum potential, that's how far it could really go.
13 JUDGE KELLEY:
Can I comment mere?
I'd like to 14 have one more try at something I tried a couple of times in 15 the last couple of days, when I asked about the likely 16 effect, combined effect of all the errors.
Whether it is 17 starting within instruments, plus or minus
.7, I gather; j
18 adding oscillations, and getting up to some 1 gallon, maybe; 19 and then extreme oscillations going to several gallons, 20 perhaps.
Not to mention computer errors, I guess would be 21 another category of error that could be operating along with 22 an oscillation and an instrument error.
23 I think when I tried this before, and I don't know 24 if I fully understood how the instrument error had been 75 really derived, but I had the sense that there wasn't any pg G
ACE-FEDERAL REPORTERS, INC.
Nationwide Coserage 80lk33MM6
I 20039.0
-9RT 1105 1
really firm answer, therefore one really couldn't say much 2
beyond instrument error.
Perhaps instrument plus 3
oscillation.
So you are up around a gallon, give or take.
4 But I'm wondering about it from the standpoint of 5
the operators involved.
I think I'm just repeating myself, 6
but, having heard a little more, I'm going to try it once 7
more.
8 The operators threw away the bad tests, so we 9
don't know what those pieces of paper said about leak rates.
10 And I'm assuming -- I'll assume for a moment there wasn't any 11 leak rate problem.
That, if they had really known the truth, 12 leak rates would have been a half a gallon or something like
~s 13 that and there really never was a problem.
14 Still and all, they were getting these printouts.
15 The ones that were thrown away, I would guess -- well, I know 16 they showed more than a gallon or they would have been kept.
17 And I wonder if, given all these potential errors, what the 18 range might have been in your judgment, if you can make one?
19 Are we talking about tests that likely would have shown, in 20 light of all these errors, leak rates of upwards of 3 21 gallons, 4 gallons, 8 callons, 10 gallons?
22 Because, if so, even though such test results by 23 hypothesis were wrong, that's what the operator saw on the 24 piece of paper and I guess I can think of two ways to 25 evaluate tests having such high numbers, if there were any.
O ACE-FEDERAL REPORTERS, INC.
x.m..
s m _ c _,.
20039.0 RT 1106 1
I suppose an operator could get a test result that said 3 2
gallons, 7 gallons -- he might very reasonably say this is 3
obviously out of whack.
There's nothing else in the plant to 4
suggest this so it has to be the test, it's no good, so I'll 5
pitch it.
6 Maybe he shouldn't have done that, but his 7
attitude is a little more understandable if he has a pretty 8
whacky number on a piece of paper.
9 On the other hand, maybe if he's testing and he's 10 got a number of 5 or 6 gallons, all of a sudden, he should 11 have been extremely upset and alarmed and started to shut the 12 plant down and really taken vigorous action.
And, I suppose,
(
)
13 there are variants among these things.
14 But my question -- that's what I'm concerned 15 about.
Based on what you know about errors, and in conceding 16 that you can't give any very fixed number, what the range 17 might have been -- some cancel each other out and for various 18 reasons.
Still and all, can you say what those thrown away 19 tests might have said, in terms of ranges?
20 MR. RUSSELL:
I think the answer to that is yes.
21 Let me try a simple description.
22 JUDGE KELLEY:
We can ask the operators, too; they 23 read them.
24 MR. RUSSELL:
If we think of the readings causing 25 a normal or Gaussian value, you have a bell-shaped curvo.
Tf O
l ACE-FEDERAL REPORTERS, INC.
g
>c _
xme_m c_,
28039.0 1107 I '9RT
%J 1
you were to plot that on a graph with zero unidentified 2
leakage, you have a distribution, I think we used the number 3
.71 yesterday and you'd have a distribution that would be 4
Gaussian.
5 As the leakage increased to the reactor coolant 6
drain tank, when that leakage got up to be about 3 gallons a 7
minute, 2.5 gallons a minute, because of the error in 8
calculating the density, that Gaussian distribution curve 9
started moving to the right, increasing in value such that if 10 you had an identified leakage of about 2.5, you would have, 11 now, that Gaussian distribution of -- centered around 1 gpm 12 unidentified leakage.
('_)
\\
/
13 As leakage got greater, and we did see it get up 14 to, I think prior to March 16th, when they made that 15 correction, it got up to about 4 plus gallons per minute.
16 And the highest offset we identified from that effect was 17 about 1.5 gallons a minute.
So that the
.7 Gaussian 18 distribution would now be centered about 1.5 gallons per 19 minute.
That's with no operator-induced action by way of a 20 water addition.
21 A water addition tends to shift that bell-shaped 22 curve back to the left, towards zero.
So the ones that were 23 retained between zero and 1, several of them had the effect 24 of high leakage with water addition or something else that 25 I would cause the computer to think there had effectively been O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cos erage 800 336-6M6
20039.0 g-3RT 1108 U
1 a water addition, whether it was a hydrogen addition or 2
whether it was an actual water addition.
3 Now, during periods of time of higher oscillation, 4
that Gaussian distribution curve may be different than.71 5
plus or minus.
It could be narrouer because of the low 6
oscillation, a narrower distribution around the calculated 7
number; or it could be wider and flatter, if you had large 8
oscillations.
But I don't believe that they were seeing leak 9
rates during periods of time of high leakage that would have 10 been in excess of 2.5 gallons per minute or so, something in 11 that range.
And I believe that explanation is somewhat 12 consistent with Mr. Hartman's testimony which is in the
)
13 exhibits, which says that while they were varying he didn't 14 see real large numbers but he did see numbers on the order of 15 2,
2.5 gallons per minute and he did see some negative 16 numbers sometimes.
17 But, where that distribution occurs as a result of 18 oscillation, moves as a result of plant conditions, either 19 identified leakage to the drain tank shifting it to the right 20 or water additions during a test which are accounted for, 21 shifting it back to the left.
22 JUDGE KELLEY:
Would you say, if you can -- well, 23 I gather you are saying that readings in excess of 2.5 24 gallons per minute would have been rare, in your judgment?
If you had quantified identified ACE-FEDERAL REPORTERS, INC.
202-347-3Alo Nationwide Coverage 800-336-6646
28039.0 RT 1109 1
leakage.
If you had other identified leakage that you knew 2
about that wasn't going to the drain tank and hadn't factored 3
in you could nave had an effect from that, also.
But I would 4
say that above 3 gallons a minute would be rare.
5 JUDGE KELLEY:
6 MR. RUSSELL:
That's correct.
7 JUDGE KELLEY:
Okay.
But how about readings of 2,
8 given all these problems and sources of error?
9 MR. RUSSELL:
I think that did occur.
10 MR. MC BRIDE:
Judge Kelley, you stated an 11 assumption in your cuestion which I think I would like to ask 12 the board to keep an open mind about, and you stated all the
(
l 13 ones thrown away were over a gallon a minute.
I think you 14 will hear testimony that operators threw results away
- t. hat 15 were negative or in the event somebody added water and forgot 16 to account for it or otherwise made a mistake.
There may be 17 a number of tests in the category you assume, but I don't 18 think it's fair to assume they are all in that category.
19 JUDGE KELLEY:
I didn't mean to assume that.
The 20 record shows clearly that a lot of tests were thrown away and 21 that's all I meant to say.
Again, given all these errors, 22 assuming that some substantial number of tests were thrown 23 away, might half the tests have been 2 gallons or more?
Or 24 can you say?
25 l
MR. RUSSELL:
I don't think we can really say g ~ s.
J ACE-FEDERAL REPORTERS, INC.
mm-x _. c_.
m_
28039.0
<~3RT 1110 1
whether half were 2 gallons or more.
I think it is clear 2
that a large number were thrown away.
But since they were 3
throwing away based on a discriminator of 1,
it would really 4
be very speculative to say how many greater than 2 were 5
thrown away.
6 JUDGE KELLEY:
I'm asking whether one can look at 7
all these sources of error and make a generalization.
And I 8
gather you are saying no?
9 MR. RUSSELL:
That would would be difficult.
10 MR. ROCKWELL:
Mr. Moore has a qualification of 11 our oscillation study.
12 MR. MOORE:
Although we presented Exhibits 11-A
,,s 13 and 11-B as evidence of oscillatory conditions in the plant
~
14 and their effects on leak rate, really they represent a 15 combination of all the errors that existed during that time 16 span.
17 For example, this data was measured with in 18 instruments.
So, if there were random errors in the 19 instruments, that is reflected in this data.
They are not 20 exactly the same instruments that would be used by the l
l 21 computer, plant computer to do a leak rate, but they are l
22 similar instruments if not identical.
l 23 The method errors were built into the computer i-24 programs, so whatever method errors existed in the plant 25 computer were replicated in this.
So that what we are
\\_/
l l
l ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cosetage 8ak336-6M6
28039.0
~7RT 1111 Y
1 looking at, here, is what the plant computer would have 2
calculated and we are seeing the peaks and valleys that are 3
produced by that in running a one-hour leak rate.
4 So, for that period of time, this does represent, 5
withcut any -- if we rule out water additions, hydrogen 6
additions, things of that nature -- does represent the swings 7
that one would see in the calculated gross leak rate.
8 JUDGE KELLEY:
Your swings in gross leak rate were 9
about what?
10 MR. ROCKWELL:
0.5 to 3.2 gpm was the range of the 11 oscillation of the leak rate test results for this January 12 4-5 oscillation study.
7_s l,.
13 JUDGE KELLEY:
All right.
Having heard 14 Mr. Russell's response, do you agree?
Disagree?
Have any 15 comment on what he had to say?
16 MR. MOORE:
Not having examined the records in 17 detail, you know, on other sides of this -- before and after 18 that -- I really can't make a judgment as to what happened in 19 the other periods.
But I consider this probably wasn't too 20 far different from what we would see there.
21 JUDGE KELLEY:
He was generalizing.
You were 22 looking at a specific period.
23 MR. MOORE:
Yes, sir.
24 JUDGE KELLEY:
All right.
Mr. Stier?
25 MR. STIER:
We come at it from a slightly (g1 ACE-FEDERAL REPORTERS, INC.
202-347 37(1)
Nationwide Coverage 800-336-6M6
28039.0 f-3RT 1112 U
1 different approach, similar to something Mr. Russell 2
described.
If you examine figures, V-1, V-2, V-3 and V-4, in 3
volume IV-A of our report, you'll see histograms that are 4
prepared by MPR, showing the distribution of test results 5
that were filed.
If you draw a curve around those results 6
you will see a change for those that were discarded, that 7
probably is in the range that Mr. Russell described.
8 That is, you see half of a bell-shaped curve 9
ending at 1 gpm.
If you add the other half to that, it takes 10 you -- it centers around 1, indicating that the majority of 11 tests that were discarded were close to 1 but on the high 12 side, and would tail off at about 2.8 gpm.
13 JUDGE KELLEY:
That seems to be consistent with 14 what Mr. Russell said.
15 MR. STIER:
I want to emphasize, though, that this 16 is really an estinate.
This is not a pr ecise way to do this 17 but it does give you a picture of what the results may have 18 been and if you look at that in the framework of the 19 testimony you may be able to use it.
20 JUDGE KELLEY:
Right.
21 MR. RUSSELL:
I have a technical question that 22 could affect, potentially, the way the reactimeter data was 23 used.
That is that we know that one of the makeup level 24 instruments was out of service and had large oscillations 25 during the time period that the reactimeter data was taken.
%)
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 804336-6M6
l
[
l i
l l
28039.0 I
('"3 R T 1113
(_)
1 I don't know whether the makeup level instrumentation was 2
used as a part of the calculation or whether only reactimeter 3
data was used or what the source of that was, but that I
4 oscillation in the one level transmitter which was out of 5
service at that time could affect the calculation of the 6
unidentified leak rate.
7 JUDGE KELLEY:
Dy Mr. Moore.
8 MR. RUSSELL:
Right.
9 JUDGE KELLEY:
Can you respond to Mr. Russell's 10 question?
11 MR. MOORE:
In Exhibit 37 of the Faegre & Benson 12 report, looking for the particular page, specifically page 6
(~T
'~
13 of Exhibit 37, is a graph of makeup level as a function of 14 time as used in the reactimeter calculation.
As to what the 15 condition of the transmitter was, I cannot say.
But this is 16 the --
17 MR. RUSSELL.
Maybe I should have phrased the 18 question differently.
I believe that is LT-2, from the 19 character of the trace, and what we have seen from LT-2, 20 since it came from the tank strip chart, we think that's the 21 one not affected by the oscillation, but I thought it was 22 important to get that on the record because it was not 23 reactimeter data but also makeup level data that had to go 24 into the calculations.
S 25 MR. MOORE:
You misunderstood me.
This in (J
ACE-FEDERAL REPORTERS, INC.
202-347 1700 Nationwide Coverage 1W4336-6646
28039.0 e~3RT 1114
/
i
%.)
I reactimeter data.
Let me back up and verify that.
My 2
recollection is that that is reactimeter data, looking at 3
Exhibit 36, the channel 8 identifies this as makeup level 4
channel to the reactimeter.
It says " patch board number 96."
5 From that you may be able to identify which transmitter it 6
was.
But it is data that was logged on the reactimeter.
It 7
was not data that we extracted from a strip chart.
8 MR. RUSSELL:
In any event, the patch panel data 9
and the other data taken from either LT-1 or LT-2, there was 10 not a special measuring device installed.
So I just wanted 11 to make sure that we were aware that there was potential, if 12 the data were taken from the LT-1 transmitter, even though it t
4 13 was recorded on the reactimeter or the data logging computer 14 on some other source, there is a potential for oscillation on 15 that result.
I just don't know whether it was the LT-1 16 differential pressure detector that was the input, or whether 17 it was LT-2.
18 JUDGE KELLEY:
Is there any way to nail that 19 down?
20 MR. MOORE:
I don't know whether the data we have 21 available here is sufficient to identify it.
I don't know if 22 the patch board number, for example, is unique to a 23 particular transmitter or not.
We can check and see if that 24 can be verified.
25 JUDGE KELLEY:
Could you check into that?
73u ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 2336-6M6
28039.0 r~9RT 1115 b
1 MR. ROCKWELL:
Would you like us to nail it down, 2
your Honor, and give you a written clarification?
3 JUDGE KELLEY:
It seems like a significant point.
4 You run the risk of somebody wanting to cross-examine you, 5
but in any event it seems useful.
It's a pretty narrow 6
point.
Either it was or wasn't.
If you can narrow it down 7
it will be useful.
8 MR. ROCKWELL:
We'll either do it today or provide 9
a written response to you.
10 Jt:DGE KELLEY:
All right.
Shall we start the next 11 one?
12 MS. WAGNER:
Are you taking follow-ups on this
_s
'~]
13 section?
14 JUDGE KELLEY:
We thought we'd just go ahead.
If 15 you have some, hand them in, why don't you.
We can be 16 reviewing them.
17 JUDGE BRIGHT:
I would like to ask a question.
18 Can someone tell me what this reactimeter does?
19 MR. ROCKWELL:
Could you pose the question again?
20 JUDGE BRIGHT:
This reactimeter, I don't quite 21 understand why it was only available for two days.
22 MR. MOORE:
I could speak to that, your Honor.
23 JUDGE BRIGHT:
Please.
24 MR. MOORE:
The reactimeter, I believe the length f-25 of the tape is good for somewhere in the area of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
q)S ACE-FEDERAL REPORTERS, INC.
202-347-3701 NJtionwide Coserage 800-336-6M6
28039.0 r'3RT 1116
()
1 This was not a normal part of the plant, at least at that 2
point in time.
It was something special that was put in by 3
-- working with Babcock & Wilcox.
Its purpose was to, 4
primarily to be able to capture significant plant variables 5
in the event of some sort of a transient so that if we did 6
have a transient, that you would have captured a good number 7
of hours of data prior to the transient, and then throughout 8
the transient, and have that available to analyze that 9
10 The practice was that if you went through a 11 24-hour period without a transient, the tapes were reused.
12 So it was only 1.ne tapes that covered significant transients s
13 that were really maintained, except for this one.
I'm not 14 sure exactly why this one was kept, but this one was l
15 available with steady state operation for a significant l
l 16 period of time.
l
[
17 Others were available but they were transient 18 periods.
In fact, this reactimeter was one of the major i
19 sources of information for the day of the TMI-2 accident.
20 JUDGE BRIGHT:
I thought I heard someone say 21 something about a data recording system of some kind or other 22 that's operating all the time.
l 23 MR. MOORE:
This reactimeter operated all the i
l 24 time, recorded on a magnetic tape.
The problem of retrieval l
25 is that the tapes were reused to record subsequent days.
l ACE-FEDERAL REPORTERS, INC.
202 347-37(W)
Nationwide Coserage 800-33M646
l 28039.0 RT 1117 1
MR. COLE:
It's like an airplane recording for a 2
crash.
That's about equivalent to what it amounts to.
3 JUDGE BRIGHT:
I guess the thrust of my question 4
is the principal variable in the operation of the plant, 5
aren't those recorded all the time?
The power?
The 6
temperature?
The pressure?
All these good little things.
7 MR. MOORE:
I don't recall the specific charts we 8
have in the control room.
We'd have to check that out, which 9
ones are there.
I don't recall.
10 MR. COLE:
There's the normal recorded charts.
11 The thing that was unique about this reactimeter, it was a 12 high speed device and you could get a very thin cut.
We were
,~
4 1
13 running some special tests on steam generators that had 14 nothing to do with this, but we needed high speed reporters 15 that recorded transients when the steam generator would be 16 tripped off the line or things like that.
We were 17 specifically looking at steam generator performance.
It so 18 happened it was capturing this other data which people are 19 oow using for other purposes.
i 20 MR. MOORE:
But the response -- I think you have 21 an additional question if I understand it:
Do we have paper, 22 strip charts in the records that reflect these same 23 parameters?
And that we'd have to check to get an answer for 24 l you as to exactly which parameters we do have on file.
I
<~s 25 l
MR. RUSSELL:
There are several mechanisms, hard
)
n-ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cm erage 800-33M646
28039.0
(-}RT 1118
(/
1 copy records of plant data.
We have strip charts which are 2
of the type we have been looking at with the makeup.
There 3
is -- basically they are brush recorders and there are 4
several of those in the control room for which the records 5
are retained.
There are also data recordings that are done 6
by a computer data logging system, which takes, on a periodic 7
basis, and records the data and then prints it out on a piece 8
of computer paper.
You also have instruments in the control 9
room, which are your typical analog instruments, which 10 operators take readings on and write the data down.
That may 11 be another way of getting a record of what was happening at 12 the time.
But the description of the reactimeter, which is O)
\\'
13 high speed data retrieval and retention for a fixed period of 14 time for subsequent analysis, that's typically analysis done 15 like on a mainframe computer.
That's only available if you 16 take the tape off and analyze it.
If you continue to use it, 17 it overrecords with new data based upon where you are.
So l
18 it's a short history back in time of, like 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
l 19 JUDGE BRIGHT:
So the available information, then, i
20 aside from reactimeter, would not be adequate to do something l
l 21 like we see on the track --
l 22 MR. RUSSELL:
It could be done.
One could go and l
23 get pressurizer level traces and try to line up pressurizer l
l l
24 level traces and go through identifying timing events.
But 25 that would be very hard because pressurizer level is
-)
v ACE-FEDERAL REPORTERS, INC.
202 Nationside Coserage NG3346M6
-347-3700.-
28039.0 1119 p)pRT 1
essentially maintained and automatic.
Then you are just 2
going to be seeing the effect of the automatic system and 3
times, for instance, when you may have manually sprayed down 4
the pressuriser or done a startup or shutdown.
So it would 5
be very, very difficult to take some of the other hard copy 6
records to go back and reconstruct this.
That's why we used 7
the printed data from the retained leak rate tests, because 8
they did printout the values of makeup level, pressurizer 9
level, hot leg temperature, cold leg temperatures, et cetera, 10 which were to principal numeral values used -- they were the 11 numericals used in the calculation.
12 MR. KIRKPATRICK:
I would like to comment of the 13 you tried to do some of this and the data that is typically 14 available is on an hourly basis.
This reactimeter took a set 15 of readings every 3 seconds so you have 0 readings a minute.
16 So this is extremely voluminous compared to the typical data 17 that is recorded routinely that you had available for any 18 time you liked to look at it.
l 19 MR. COLE:
What I was trying to indicate, while we l
20 do not have reactimeter data for all the periods of time so l
21 you can get precisely the magnitude of these oscillations, l
22 you can get a good sense of whether the plant was oscillating 23
-- for example you can look at the trace of the hard copy of j
24 the pressure and you can see during periods of oscillation, 25 it's like a red disease, it is a very broad line.
And the I
i i
l ACE-FEDERAL REPORTERS, INC.
202-ut3700 Nationwide Coserage 80t936-f6M
28039.0 f-7RT 1120
()
1 basis for our saying that this may not be typical, 2
particularly during the month of March, when you look at the 3
pressure oscillations, while you can't get a precise, every 4
3-second picture, you get it clear that it's a very broad 5
band span of trace or pressure during the period that he's 6
talking about.
But in March that trace is very flat, which 7
would imply that there are very little oscillations.
8 So, there is some way of going back and getting a 9
judgmental view of.was the plant oscillating at various 10 times?
You can go back, even today, to check that.
11 JUDGE BRIGHT:
Thank you.
12 JUDGE KELLEY:
We are going to do a couple of 13 follow-ups and then quit for lunch.
14 MR. MOORE:
The Staff just passed down some 15 information to tell us that the, apparently, the makeup level 16 transmitter going to the computer -- I'm sorry.
After lunch 17 I'll come back.
I misinterpreted the data sheet --
18 JUDGE KELLEY:
All right.
Let's move along.
19 JUDGE BRIGHT:
I have a follow-up question.
It 20 covers the waterfront here for Stier, Moore, Rockwell, 21 Russell, Capra.
Isn't it a fact that the oscillations you 22 just discussed have a greater likelihood of occurring during 22 the early period of operation of a new nuclear power plant?
24 MR. RUSSELL:
Yes.
25 MR. ROCKWELL:
Yes.
O ACE-FEDERAL REPORTERS, INC.
.E347-3700 Nationwide Cos erage 800-336-6M6
1 28039.0 RT 1121 1
MR. STIER:
Yes.
2 JUDGE BRIGHT:
Was TMI-2 such a new plant?
3 MR. RUSSELL:
Based upon its operating history I 4
would characterize it as a new plant.
Although in the period 5
of time after the startup in January it did not appear that 6
the oscillations were as large.
So, in the context of the 7
question I would say, from mid-January to the day of the 8
accident in 1979, that the oscillations did not appear as 9
large.
But there clearly were large oscillations in the 10 September 30th through mid-January period for periods of time 11 when the plant was operating.
12 JUDGE BRIGHT:
Mr. Rockwell?
13 MR. ROCKWELL:
I think that's a fair description.
14 I might add that the qualifications that had been placed on 15 the degree which one can generalize from the reactimeter data 16 are appropriate and we don't disagree with those 17 qualifications.
18 JUDGE BRIGHT:
Mr. Stier?
19 MR. STIER:
My answer is yes, but it was newer in 20 January than it was in March.
21 JUDGE BRIGHT:
Thank you.
22 For Mr. Russell /Capra.
During your interview of 23 operators, were there any indications that the operators knew 24 the magnitudes of method, instrument, computer and other 25 errors inherent in the TMI-2 leak rate test?
-3, N
I ACE-FEDERAL REPORTERS, INC.
mmmm x _. c-1
28039.0 RT 1122 U
1 MR. RUSSELL:
No.
I don't believe the operators 2
understood why the effects were occurring.
I believe that 3
they understood, however, for instance increasing pressure on 4
a makeup for some operators would cause an increase in level 5
and that increase in level would result in getting a better 6
or a good leak rate test.
So they understood more the cause 7
and effect relationship but not the why, and did not 8
understand what was in the black box computer calculating 9
numbers.
10 JUDGE BRIGHT:
Thank you.
I think that would take 11 care of those.
12 JUDGE KELLEY:
Do you have any comment on that, 7
i 1
13 Mr. Stier?
Since you investigated that?
14 MR. STIER:
My assessment would be generally the 15 same as Mr. Russell.
There were some errors that were 16 sensed, observed by operators.
There were others that they 17 really didn't sense.
They understood in a general sense, 18 they had feelings about the variability of tests, the 19 accuracy of tests, that there were some errors someplace, but r
20 not with the kind of precision that the question seems to 21 call for.
22 JUDGE KELLEY:
Okay.
Let's break for lunch until l
23 1:30.
24 MR. GOLDBERG:
Judge Kelley, before we break for i
25 lunch, something that may make the board's lunch a little I
ACE-FEDERAL REPORTERS, INC.
202 4 47-3700 Nationwide Coverage 800-336-6M6
28039.0
-~pRT 1123
%J 1
more pleasant.
Mr. Gephart and I have had a conversation --
2 JUDGE KELLEY:
Are we on the record?
3 MR. GOLDBERG:
Yes. -- about the retrieval of the 4
first version of the April lith memo from Ben Hayes to the S
Commissioners.
Apparently, everyone is agreeing to return l
6 the copies.
There doesn't appear to be a need for a board 7
order directing the return of the copics.
And what I suggest 8
I is that, unless you hear further from Mr. Gephart or 9
Mr. McBride that someone is not returning their copy, the 10 matter can be left that there is no need for any further 11 board action on this matter.
12 JUDGE KELLEY:
How about the fact that the letter
,s
(
i
'~'
13 was bound in the transcript yesterday?
14 MR. GOLDBERG:
With the exception of the board's 15 copies, and I guesa Mr. McBride's --
16 MR. MC BRIDE:
We are going to take ours out of 17 our copy of the transcript so ours conforms.
We would ask 18 the board to do the same.
The only qualification on what l
19 Mr. Goldberg says is this of course excludes us because we l
l 20 have understood the whole discussion to have resulted in this 21 morning --
22 MR. GOLDBERG:
We either have all the copies from 23 the transcripts or they are being retrieved.
No one is 24 objecting to doing that.
With respect to the transcript, 25 even Mr. Gephart indicated so that their copy of the O
l I
l ACE-FEDERAL REPORTERS, INC.
202-34'-3700 Nationwide Coverage 800 336 6M6
l 20039.0 BRT 1124 G
1 transcript conforms with.everyone else's, they'd be willing 2
to return the copy that was appended to their transcript.
3 JUDGE KELLEY:
You are not keeping a copy?
4 MR. GOLDBERG:
Not one appended to the 5
transcript.
6 JUDGE KELLEY:
You are keeping a separate copy.
7 That's satisfactory.
8 Mr. Blake's proposition was that we should -- how 9
would one say -- make it not a proposed exhibit any more.
Is 10 that part of the proposition, too?
11 MR. MC BRIDE:
I think the proposition was it 12 never was.
It was only Mr. Goldberg had proposed a number 13 and the board had not assigned it such a number.
'~'
14 JUDGE KELLEY:
That could be, in the event that we 15 did, we countermand the assignment.
There seems, then, to be 16 consensus that the copies will be returned, taken out of the 17 transcript.
It is not a proposed exhibit.
It exists because 18 it is referred to in the transcript, but that's about it.
It 19 won't accompany the reco -d.
Is that satisfactory with 20 everybody?
21 MR. MC BRIDE:
Yes, Judge Kelley.
I wonder if we 22 could just for purposes of procedural matters down the line 23 here make things simpler for the court reporter, whose fault 24 I don't think this was, and that is that, until the board has 25 assigned an exhibit number to a piece of paper or a picture
,NI I
ACE-FEDERAL REPORTERS, INC.
-m s _ _-
~~
-_ ~.
1 28039.0 RT 1125 2
1 or whatever it nay be, that it would not be bound into the 2
transcript -- just in case we have this problem.
3 JUDGE KELLEY:
Fine.
We'll consider that.
Let's 4
eat lunch.
5 (Whereupon, at 12:30 p.m.,
the meeting was 6
recessed, to be reconvened at 1:30 p.m.
this same day) 7 8
9 10 11 12 0
13 14 15 16 17 18 19 20 21 22 23 24 25 O
F 4
ACE-FEDERAL REPORTERS, INC.
202 347 Nationwide Coserage 80tk3346M6
__.___.,.____ -.._. _ __. _..-.._ _37(ul_ _ _. _, _ _. _ _
28039.0 1
1126
{w) RT 1
AFTERNOON SESSION (1:30 p.m.)
2 JUDGE KELLEY:
We can go back on the record.
We 3
basically plan to go to the last group of questions and then 4
go to the Aamodt questions, for Mr. Rockwell and Mr. Moore, 5
so we can be sure we get finished with them before 3:30.
6 Then we'll return to the sequence.
7 I have one further follow-up question on 8
reflection about these test results that were likely to have 9
come out.
I think we established that they could have likely 10 ranged up to 2.5, maybe as high as 3.
Conversely, they could 11 they have gone to minus 2, minus 3?
Same swing in the other 12 direction?
13 MR. HARRISON:
No.
14 JUDGE KELLEY:
How about on the other idea?
15 MR. RUSSELL:
It would depend on whether a water 16 addition was made during the calculation.
Normally the 17 variation around a value, that value would be offset based 18 upon the error for the drain tank temperature composition.
19 So, if you had large leakage to the drain tank as we 20 indicated, you would have an oscillation about, in some 21 cases, plus or minus
.7 gallon a minute, around 1.5.
22 Early on, in January, when you didn't have such 23 hich leakage to the drain tank, the oscillation may have been 24 around say
.7 gallon, so the results would be, say from zero rs 25 up.
The negative numbers that we saw, large ones, were U
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646 1
28039.0 O'9RT 1127 c
1 generally associated with either an entry number, putting the 2
wrong number into the computer -- in one case I think it's 3
test 147 where you get a minus 6 gallons a minute identified, 4
or for some other occurrence that was not accounted for in 5
the calculation, such as a feed and bleed or something like 6
that.
7 JUDGE KELLEY:
It's not that critical, but I 8
assumed the errors we spoke about this morning, particularly 9
instrument errors and oscillations and built-in computer 10 programs, I heard it said that those errors could range up to 11 2.5, sometimes 3.
I thought they probably ranged into the 12 negative.
Are you suggesting that except for the factors you 13 just mentioned they didn't?
14 MR. RUSSELL:
That's correct.
Unless there were a 15 water addition that was made to compensate for the drain tank 16 error, an operator water addition that's logged -- say, for 17 instance, the errors would generally be offset.
So they 18 would vary about a positive number, say 1.5 to 2.5 or 3.
i 19 JUDGE KELLEY:
The point I want to make, I think, i
j 20 is pretty much independent of that.
I, just for 21 illustration, wondered about the negative side -- just drop I
I i
22 that for a moment.
23 MR. MOORE:
One thing I would like to clarify, 24 Judge, on Exhibit 11 we were talking about gross leak rate.
l 25 From that you must subtract the identified leak rate to 1
l l
l ACE-FEDERAL REPORTERS, INC.
202-347-3Xu)
Nationwide Coserage 800-336-6M6
28039.0 f--9RT 1128 U
1 determine the unidentified leak rate.
So that whereas the 2
gross leak rate that we saw here was always a positire value, 3
when we get to the unidentified leak rate, which is the real 4
crucial issue I think here, you have to subtract the 5
identified leak rate from that to arrive at the -- subtract 6
the identified leak rate from that to arrive at the 7
unidentified leak rate.
8 So if we had -- whatever we have here should be 9
cycling about some mean value which would be the identified 10 leak rate.
11 JUDGE KELLEY:
Let me ask my question.
I really 12 think my question is simple.
I think you are really more
.s 13 sophisticated than I need to be, 14 I understood the numbers, at least on the positive 15 side, ranged up to 2.5 or 3.
My question is this.
I'm an 16 operator and I run a leak rate test and it says 2.8.
So I 17 decide to run another leak rate test and it says
.1.
And 18 then I run even another one, I'm patient and I get 1.9.
It 19 is all over the map.
20 Now, with those kinds of fluctuations, sharp 21 fluctuations in result such as I see suggested by the 22 exhibits, Number 11 is it --
23 MR. MOORE:
Yes.
24 JUDGE KELLEY:
Were they likely?
Or, on the other 25 hand, was it more likely that if I ran a series of tests, ACE-FEDEIML REPORTERS, INC.
202-347-3700 Nationwide Coserage 8(4336-6M6
3 I
28039.0 BRT 1129 1
back to back, one hour after the other, I might -- if I got a 2
2.2 I'd next get either a 2.3 or a 2.1?
Would I get a 3
gradually trended change or a sharp fluctuatior., is my 4
question?
Whoever wants to speak to it.
Mr. Stier?
5 MR. STIER:
Well, putting together a combination
]4 6
of evidence that we gathered, including to a certain extent 7
what operators told us -- and I'm generalising -- and the 8
distributions that we see on the histograms that I described 9
to you before, I think a couple of things are clear.
10 First of all, there were sharp differences in test 11 results.
Testimony indicates that from test to test, back to 12 back, you would get widely varying results.
Maybe a gallon a e
i 13 minute unidentified leakage.
14 But there are other patterns, too, that you've got 15 to superimpose on that.
During certain periods of time, 16 depending on conditions -- like in, for example, December and 17 January, when actual unidentified leakage was very high and 18 exceeded 1 gallon a minute, successive tests would be run and 19 it became very, very difficult, if you ran a test using a 20 good makeup tank level transmitter, very difficult based on 21 this random distribution of results to get a result under 1 22 gallon T minute.
l Just as in March, late February and particularly 23 24 l in March of 1979, because of the col)ection in the reactor 3
25 coolant drain tank, it became very difficult, based on these ACE-FEDERAL REPORTERS, INC.
21C-347-3700 Nationw hie cowrage soix335m46
l l
l 28039.0 l
-SRT 1130
/
i
%/
1 random outcomes of test results, running them back to back, 2
to get a result under 1 gallon a minute.
3 So you had these wide variations but they moved in 4
relation to zero gpm of unidentified leakage, depending on 5
what the conditions were during the period of time when the 6
tests were run.
7 JUDGE KGLLEY:
But you still had the wide 8
fluctuations?
9 MR. STIER:
Yes.
10 JUDGE KELLEY:
And could they be irrational --
11 irrational is the wrong word -- but the business of going 12 from a test that showed zero to 1.5 to
.3, all inside three O
13 hours1.50463e-4 days <br />0.00361 hours <br />2.149471e-5 weeks <br />4.9465e-6 months <br />, back and forth like that?
14 MR. STIER:
I believe that they could.
15 MR. COLE:
There have been times when that 16 happened.
17 MR. STIER:
We do have a group of tests, however, 18 that you might want to examine specifically, where the 19 results were saved and weren't thrown away, in the period in 20 October 16th through the 18th.
We have a series of tests and 21 we can see that the tests were run back to back.
They were 22 presumably, all the test that were run in that period, and 23 you can see the spread of the results.
And they all showed 24 leakage that was high, unidentified leakage.
They all showed
(-
25 unidentified leakage over 1 gpm.
V)
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33MM6
28039.0 RT 1131 x_ -
1 So, although you had some variation, you didn't 2
have any during that period falling under 1 gpm because of 3
the random changes in test results.
4 JUDGE KELLEY:
I guess the point, perhaps obvious, 5
related to the point I was trying to make earlier.
If an 6
operator saw a very large leak rate, which was not 7
substantiated by anything else he could find out, goes out 8
and looks for leaks, can't find any, and concludes the 9
machine is out of whack for whatever reason, it would seem to 10 me that if you got seemingly random results back to back, 11 that might lend some justification to an operator's regarding 12 the machine as out of whack and unreliable and not to be 13 taken too seriously, as opposed to, if he got a series of 14 tests that read 1.5, 1.7, 1.9 -- then I would think he would 15 say:
My god, I got a growing crack, I'd better do 16 something.
At least there's some progression there in an 17 alarming direction.
Does that make any sense or am I being 18 too simplistic about this?
19 MR. ROCKWELL:
It does make sense.
Hartman speaks 20 directly to that issue, if I may read briefly because I think 21 it illustrates the point.
He says: "It was just the 22 erratieness.
I couldn't see how one day the plant could have 23 a leak of 2.5 and the next day half a gallon and the next day 24 6,
I'm saying like in a matter of hours.
We used to run g~s, 25 these things constantly during the midnight shift, in which
%.)
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 80fk33MM6
28039.0 r-BRT 1132 b
1 we didn't get a good one.
We'd have another one in right 2
away and they would always come out, but you would never get 3
the first one to be 2.5 and the next one 2.4 and the next one 4
make 2.3 again.
It was 2.5, 2.2,
.7 -- it was always a 5
significant amount, as I can recall -- you know, that is why 6
my faith was not there in what we got as results."
7 JUDGE KELLEY:
Thank you.
8 MR. ROCKWELL:
That's page 32 of our report.
9 MR. MOORE:
Exhibit 11-B shows in an 11 minute 10 period, if you could have run three leak rates to accomplish 11 that in an 11-minute period, at this point it would have been 12 3 gpm, would have been the mean here, and 10 minutes later it (s) 13 would have been about
.5, gross leak rate.
14 JUDGE KELLEY:
Thank you.
Mr. Kirkpatrick, do you 15 have something?
16 MR. KIRKPATRICK:
No, I didn't, your Honor.
17 JUDGE KELLEY:
All right.
Thank you.
18 Mr. Goldberg, I think, had a new photograph?
19 MR. GOLDBERG:
There was some questioning of the 20 witnesses before about whether the makeup strip chart could 21 be slid out from its position so th.>t more history could be 22 observed than could be observed when it was fully inserted in 23 the panel.
24 We have a photograph of that strip chart pulled r3 25 out, which Mr. Capra has now.
If the board wishes to have it e>
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800 336-(M6
28039.0 e~yRT 1133 j
L.)
~
1 a part of the record we can have it duplicated and 2
distributed to everyone and offer it.
3 JUDGE KELLEY:
Maybe Mr. Capra could just hold it 4
up and comment a bit.
Then counsel could take a look at it 5
and we can ask them again, later, whether anybody has any 6
objection to putting it in.
7 MR. CAPRA:
I can pass this around as soon as I'm 8
done.
Basically it does show the strip chart recorder pulled 9
out of its normal position.
I was aske arlier how m.uch 10 time you could see if you were looking at the instrument 11 straight on without the instrument pulled out.
It appears 12 from this chart that you could see approximately an hout and
\\
13 45 minutes.
If you pull the chart out it appears you can see 14 somewhere between six tb eight hours in addition to that, at 15 least on one side of the chart.
I'm not sure if you could 16 also see an equivalent amount on the other side, which hadn't 17 printed out yet.
18 JUDGE KELLEY:
Thank you.
Maybe you can just 19 circulate that and we can ask later if there's any problem 20 with our putting it in.
21 I'd ask, looking at that, the even hours are 22 printed out on the chart.
The odd hours are not.
But every 23 increment there is marked, a vertical line every 15 minutes.
24 JUDGE KELLEY:
Thank you.
25 JUDGE BRIGHT:
We are starting another section
{\\_)
S ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80lk336-6646
28039.0 r~9RT 1134 a
1 here, evaporative loss factor.
Beginning with Russell and 2
Capra, 261:
Did the procedure prescribed for the calculation 3
of RCS leakage at TMI-2 include an allowance, referred to as 4
an evaporate loss factor, for evaporative losses for the 5
6 MR. KIRKPATRICK:
No, it did not.
7 MR. ROCKWELL:
No, it did not.
8 MR. STIER:
No, it did not.
9 JUDGE KELLEY:
I don't believe we struck 262 10 earlier, that was an oversight.
We are going to strike it 11 now because it asked about TMI Unit 1 leak rate procedure.
12 We have already had a little testimony o'n that point.
We are
( y\\
\\
13 not going to pursue it any further.
14 MR. MC BRIDE:
Judge Kelley, would it be within 15 bounds to ask you to just take a moment to reflect on section 16 4.0 to confirm -- this is a comparative issue, not just with 17 TMI-1 but with other nuclear plants, that really bears on the 18 validity, from our point of view -- others may differ --
19 about the leak rate test itself.
20 JUDGE KELLEY:
Well, 263 makes a comparison.
We 21 left that in.
22 JUDGE BRIGHT:
I'm not sure what answer we are 23 going to get to that so it may or may not satisfy the point 24 we are trying to make.
Perhaps we can go to that and then gx 25 see if that's satisfactory.
O ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6M6
28039.0 g BRT 1135
'w) 1 JUDGE KELLEY:
We could consider it.
Go on to 2
263.
3 JUDGE BRIGHT:
263:
Were other nuclear power 4
reactors supplied by Babcock & Wilcox during the 1970s 5
permitted to include an evaporative loss factor in their 6
procedures for the calculation of RCS leakage?
7 MR. KIRKPATRICK:
Some were and some were not.
We 8
testified on this previously.
I believe the ones that had an 9
evaporative lost factor were TMI-1, Rancho Seco and one other 10 plant which escapes me --
11 MR. RUSSELL:
Oconee.
12 MR. KIRKPATRICK:
And Oconee.
( '#,)
\\-
13 I may have missed one.
TMI-2, Davis-Besse, and 14 Cryctal River did not have an evaporative loss factor.
15 MR. RUSSELL:
What about Arkansas?
16 MR. KIRKPATRICK:
I believe Arkansas 1 did have an 17 evaporative loss factor.
18 MR. ROCKWELL:
I read page 19, volume I of our 19 report which reports the units which had the evaporative loss j
20 factor and the amount of the evaporative loss factor, so I'll 21 just leave the reference, page 19 of our suinmary report, 22 volume I.
I don't it's necessary to read it again.
23 MR. STIER:
I don't have this question, your 24 Honor.
/g 25 JUDGE KtLLEY:
We also struck 264, which would Y
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationside Coverage MIK3364646
3 i
28039.0 r-7RT 1136
(_)
I bring us to 265.
2 I suggest we go through 265, through this section, 3
through 268, then we can hear from numerous employees ac to 4
whether they want to propose something further.
For now, 262 5
and 264 are out.
6 JUDGE BRIGHT:
265:
What evaporate loss factor 7
would have been included in the TMI-2 procedure if TMI-2 had 8
been permitted to have one?
9 MR. KIRKPATRICK:
The answer to that is none, 10 because as I previously testified, by the time TMI-2 was 11 licensed it was realized that the evaporative loss factor was 12 not appropriate for the reasons that I previously gave in the
,.s
/
4'
~'
13 earlier testimony.
14 JUDGE BRIGHT:
Mr. Rockwell?
15 MR. ROCKWELL:
We don't know the answer to the 16 question.
17 JUDGE BRIGHT:
Mr. Stier?
18 MR. STIER:
I don't know.
19 JUDGE BRIGHT:
266:
The NRC Staff has identified 20 certain leak rate tests performed and filed for TMI-2 that, 21 according to the Staff, would, if properly calculated and 22 corrected, show an unidentified leak rate in excess of 1 23 gpm.
How many of those leak rate tects would exceed 1 gallon 24 per minute if the appropriate evaporative loss factor were 25 included in the calculation?
i ACE-FEDERAL REPORTERS, INC.
202 Nationwide Coserage Rn3364646
-347-37(H
l 20039.0 RT 1137 1
MR. KIRKPATRICK:
Here again, I believe none, 2
because evaporative loss factor was shown to be inappropriate 3
by that time and was not permitted in the TMI-2 leak rate 4
calculation.
5 JUDGE BRIGHT:
Mr. Rockwelli 6
MR. ROCKWELL:
Our question is slightly different, 7
and asks us to respond as to whether it would have been 8
easier if an appropriate evaporative loss factor had been 9
included.
10 Without debating the word " appropriate" --
11 JUDGE KELLEY:
Doesn't that beg the question?
The 12 real question is, was there a factor or not?
,_s x
13 JUDGE BRIGHT:
You were not asked to reply to this 14 particular question, but the next one, which is the question 15 you were announcing.
Okay?
16 MR. ROCKWELL:
Okay.
17 JUDGE BRIGHT:
Mr. Stier?
18 MR. STIER:
I have that question.
I don't know 19 the answer.
20 JUDGE BRIGHT:
267:
Would it have been easier for 21 the TMI-2 operators to obtain valid leak rate test results of 22
. less than 1 gallon per minute if an appropriate evaporative 23 loss factor had been included in the TMI-2 procedure?
24 MR. KIRKPATRICK:
Yes, it would.
'f 25 MR. ROCKWELL:
Our answer would be yes, it would.
}
N
/
ACE-FEDERAL REPORTERS, INC.
202-347-37m Nationwide Cos erage M133MM6
28039.0 e'7RT 1138 b
1 Again without debating what the word " appropriate" means.
2 But if there had been a factor it would have been easier.
l 3
MR. STIER:
My answer is the same.
4 JUDGE KELLEY:
Let me ask about that.
It just 5
seems to me that this question is written in a 6
question-begging fashion that does flow from the word 7
" appropriate."
I thought the Staff at least was saying that 8
evaporative loss factor is not appropriate; therefore there 7
I 9
isn't any such thing as an appropriate evaporative loss 10 factor.
Am I correct about that?
11 MR. RUSSELL:
That's correct.
Let me add to 12 that.
,s
/\\
13 It would have been an error to include it, since 14 the maximum reactor coolant pressure boundary flaw size 15 assumed in the safety analysis is that associated with a 1 16 gpm leak, not a 1 gpm plus an evaporative loss factor.
So it 17 would have been inappropriate.
We focused on whether -- on 18 the " easier" part of the question.
That is, if you had a 19 subtraction number put in, wouldn't it have been physically 20 easier?
The answer to that is yes.
21 We answered the appropriateness in the two earlier 22 questions and we don't think it is appropriate and it was not 23 included.
An appropriate number would have been zero.
24 JUDGE KELLEY:
Put the question another way, it 25 would have been easier to get the number under 1 if you had r'%
ACE-FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Cmerage 800-3346M6 l
28039.0 RT 1139 1
been given an evaporative loss factor?
The answer is yes, 2
right?
3 MR. RUSSELL:
Yes.
4 JUDGE KELLEY:
Okay.
5 JUDGE BRIGHT:
268:
Is there any safety 6
significance to the inclusion of ar. appropriate evaporative 7
loss factor in a leak rate calculation procedure?
8 MR. KIRKPATRICK:
Yes, there is.
This factor 9
effectively increases the allowed unidentified leak rate and 10 thus the potential reactor coolant pressure boundary leak 11 rate from 1 gpm to 1 gpm plus whatever this evaporative loss 12 factor is.
It would permit operation of the plant with the 13 reactor coolant pressure boundary flaw greater in size than 14 that assumed in the safety analysis report.
15 See, this evaporative loss factor was allowed 16 because it was assumed that you could measure the the 17 evaporative loss at the start of the plant operation and that 18 this would not change.
In fact it was found that, as the 19 i plant operation went on, frequently the amount of evaporative 20 loss would be reduced so that you didn't know what it was, 21 and therefore it was nothing but another kind of unidentified 22 leak rate.
23 JUDGE KELLEY:
Coupled with your position that 24 there's no such thing as an appropriate evaporative loss
, cs,
25 factor for TMI-2, one could say:
Yes, there is safety L}
l ACE-FEDERAL REPORTERS, INC.
202 347-37(X)
Nationwide Coverage 8tXh33MM6 t
28039.0
- RT 1140 v
1 significance to including such a loss factor and it's 2
negative safety?
3 MR. RUSSELL:
Absolutely.
4 MR. KIRKPATRICK:
That's right.
5 JUDGE KELLEY:
Mr. Rockwell?
6 MR. ROCKWELL:
Perhaps another way of looking at 7
it is simply that in setting a limiting condition for 8
operation the use of an evaporative loss factor comes in 9
play, whether you included it or not.
To the extent that the 10 limiting condition for operation is a -- has safety 11 significance, then how you deal with the evaporative loss 12 factor is relevant.
13 JUDGE BRIGHT:
Mr. Stier?
14 MR. STIER:
I agree with the Staff's position.
15 JUDGE BRIGHT:
That concludes that partic'.lar 16 section.
17 JUDGE KELLEi*:
Follow-ups?
18 MR. MC BRIDE:
Yes, I believe so, your Honor, if I 19 could have the Roard's indulgence for a moment?
20 JUDGE KELLEY:
Yes, sure.
21 (Discussion off the record.)
22 MR. MC BRIDE:
Judge Kelley, it might be a more 23 efficient way to proceed if we could have the Board's 24 indulgence to come back to this at a later point.
I realize (g
25 we may lose Mr. Rockwell in the process, but rather than take
()
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage im 3M-6M6
20039.0 e-3RT 1141
(
)
w/
1 up everyone's time, there is a question or two on the 2
evaporative loss factor that may be governed by some 3
materials that are voluminous in the Stier report and buried 1
1 i
4 in prior documents of the Commission.
Perhaps we could just I
(
5 go forward.
6 JUDGE KELLEY:
What we have proposed to do at this l
7 point is to move over and ask a series of a dozen questions 8
from the Aamodt's to Mr. Rockwell and Mr. Moore, so we'd be 9
sure to cover their territory this afternoon.
Do you mean 10 later this afternoon or tomorrow?
11 MR. MC BRIDE:
I might have them within a half an 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
13 JUDGE KELLEY:
Why don't we do the Aamodt 14 questions, then, at this point.
15 If the other witnesses will bear with us, these 16 are questions that are directed specifically to Mr. Rockwell 17 and Mr. Moore.
I'll just add that they were, with the 18 exception of, I believe two of them, given to the witnesses 19 in advance.
Two of them we held because they seem more in 20 the nature of impeachment-type questions which -- the kind of 21 questions we agreed not to submit in advance.
22 Okay.
Beginning with number 1:
Provide your I
23 contract with GPU relative to your investigation of the 24 performance of leak rate tests at TMI-2.
Can you do that?
25 MR. BLAKE:
Objection.
I do not understand there j
l ACE-FEDERAL REPORTERS, INC.
l n w.~
nn-
28039.0 r~7RT 1142 i
)
v 1
to have been discovery allowed in this proceeding.
I do not 2
understand at this juncture what the materiality might be 3
from such a, what I might characterize as a fishing 4
expedition, for something in that report at this point.
I 5
think there might very well be appropriate types of questions 6
that could be eliciting information that might go either to 7
the credibility to be attached to their efforts or not, but I 8
regard that one as outside the scope here and unnecessary to 9
the bottom line in this proceeding.
10 JUDGE KELLEY:
Let me suggest this.
The question 11 asks for a particular piece of paper, not any quarrel 12 answering, necessarily.
Ms. Aamodt isn't here.
The Board
( ')
13 doesn't really want to make her argument for her.
14 What I would suggest is that we let that question 15 and objection pend, assuming that she'll be here fairly soon, 16 and then we could hear her on it and make a decision.
Is 17 that going to cause any problem?
18 MR. BLAKE:
No.
That's not a problem.
19 MR. ROCKWELL:
For the record, your Honor, my 20 recollection is thet the piece of paper does not exist.
21 There was no written contract.
22 JUDGE KELLEY:
There was no written contract.
23 MR. ROCKWELL:
There was one thing I didn't check 24 before I came out that I recalled but my guess is that that's 25 what my answer will be.
O ACE-FEDERAL REPORTERS, INC.
202d474700 Nationwide Coverage an 34m46
28039.0 (SRT 1143 V
1 JUDGE KELLEY:
Okay.
Then there might be a 2
follow-up, what's the oral contract?
But, in any event the 3
objection will pend --
4 MR. BLAKE:
That substantially moots the 5
argument.
6 JUDGE KELLEY:
In any case, we'll let that one 7
pend in the expectation that we'll be hearing from 8
Mrs. Aamodt.
I might just say that we are really striking a 9
balance here.
We don't know when Mrs. Aamodt is going to be 10 with us.
I suppose we would have, in an ideal world, tried 11 to schedule things so Mrs. Aamodt would be here when her 12 questions were being put to Mr. Rockwell, but she's not and 1
\\
13 he is so we are going to go ahead and put the questions now 14 and we think, on balance, that is fair enough.
15 Number 2:
Describe any constraints implied or 16 expressed that GPU, its subsidiaries or any employees or 17 agents of GPU or its rubsidiaries placed on you, 18 investigator, authors of your report or your agents relative 19 to the investigation of the performance of leak rate testing 20 at TMI-2.
21 MR. ROCKWELL:
In answering that question, your 22 Honor, I would like to try to answer it by considering what 23 someone else might regard to be a ccnstraint, even though I 24 may not.
Because we were asked to come in and do an 25 independent investigation and we made substantial efforts and O.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 80k334%4
28039.0 e-7RT 1144 h
1 gave a lot of thought to maintaining the independence of that 2
investigation.
I would like to try to address a couple of 3
those issues.
4 JUDGE KELLEY:
Sure.
5 MR. ROCKWELL:
First of all, when we arrived at 6
Three Mile Island., GPU Nuclear had assembled a team of 7
investigators, including Mr. Moore and the others who are 8
listed on the faces of all of the volumes of our report.
9 That team of investigators had been working for some period 10 of time; my recollection is on the order of five to six 11 days.
12 We were offered that team as the group of 13 engineering people who would assist us in conducting the 14 investigation.
And we accepted them as members of the team 15 after a careful meeting in which we described our objectives, 16 described our charge to conduct an independent investigation, 17 asked -- told them that they would be expected to have a "no 16 contact" rule in terms of any of the thinking or analysis or 19 discussion within the team; told them that there could be no 20 exceptions to that; told them that if anyone had any 21 reservation of any kind about that that they could depart the 22 team with no questions asked and no criticism.
And we also 23 told them that we owned their life for the next five weeks 24 and that they had to be willing to agree to that.
And none 25 of them dropped off the team.
O 4
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage an3346M6 m
,,. -, _,. _,. -. ~ _, -. _ _ _ _ _ -, - -
1 1
1 28039.0
-BRT 1145 1
So, one could look at that team and suggest that 2
there was an implied constraint by the choice of personnel.
3 After going through that process we were more than satisfied 4
that they were willing to live by the charge that we had been 5
given and, in fact, in the course of the investigation, we 6
were impressed with their dedication to that charge 7
throughout.
8 So, I do not f.elieve that, in fact, it was a 9
constraint.
10 I would think that the other principal constraint 11 was that of time.
GPU, actually Metropolitan Edison at that 12 time, was very concerned about obtaining an appreciation of i
13 the facts surrounding these allegations.
And one nust recall 14 that at that time they had almost no information.
We were 15 literally starting from scratch with a few statements from 16 Mr. Hartman.
So we recoqni::ed the legitimacy of their 17 concern, in terms of knowing:
Wha t are we dealing with?
18 What was going on and what is the siqnificance of all of 19 this, it any?
20 l
We conducted the investigation, therefore, under 21 substantial time pressure.
As I indicated eallier in my 22 testimony, the essence of the investigation wa conducted, 23 written and concluded in about 35 days.
We prepared dratt 24 reports, which are substantially what you see her",
and then i
c
'5 j over a period of about another eicht weeks we did cleanup;
(
I ACE-FEDERAL REPORTERS, INC.
I I
mmm
~ _ m-
28039.0
-9RT 1146 v
1 not so much cleanup investigation but just cleanup of the 2
writing and proofing and editing, making sure that there were 3
no inadvertent inaccuracies.
4 So, I would think one has, in fairness, to say 5
that that was a constraint.
6 Again, with respect to how significant it was, in 7
light of what we were able to accomplish and in light of, I 8
guess, our own perspective on that work as we sit here today, 9
we feel that it was not a constraint that seriously impaired 10 the validity of the analysis.
11 JUDGE KELLEY:
Thank you for that explanation.
12 Number 3:
Following your investigation in 1980 --
O 13 MR. ROCKWELL:
Your Honor, it's blanked out.
Is 14 this one of the questions you did not want to ask --
15 JUDGE KELLEY:
The next question and question 16 after it are questions we are going to put to you but you 17 didn't get it in advance.
I'm sorry, I should have 18 elaborated.
19 Number 3:
Following your investigation in 1980, 20 did GPU, subsidiaries or its agents contact anyone in your 21 company relative to the conduct of an investigation of the 22 responsibility of individuals for the performance of leak 23 rate testing at TMI-2?
If so, describe the contract -- I was 24 going to say contact -- contract.
If not, what, in your 25 opinion, caused GPU to seek other services to conclude the rsU i
ACE-FEDERAL REPORTERS, INC.
t a.w.~
~ ~- + c-,.
l 28039.0 f-3RT 1147 NY 1
leak rate performance investigation?
Would you have been 2
willing to continue the investigation?
3 MR. ROCKWELL:
We were not contacted.
I don't 4
know why.
5 JUDGE KELLEY:
Okay.
Would you have been willing 6
to continue the investigation if contacted?
7 MR. ROCKWELL:
It has some bearing, in terms of my 8
other obligations and commitments, and that would have been 9
the principal factor?
10 JUDGE KELLEY:
The work-load answer?
11 MR. ROCKWELL:
Correct.
If I had known it was 12 going to take two years --
7s
\\_]
13 (Laughter.)
14 MR. ROCKWELL:
-- I would have had a very long 15 discussions with my partnero.
16 JUDGE KELLEY:
Right.
Thank you.
17 Next, this was also withheld:
Why did you seek 18 access to the operators at TMI-2 in the course of your 19 investigation?
When did you first seek access to the 20 operators?
I'll stop there; there's a couple of more parts 21 but I'll just stop there.
22 MR. ROCKWELL:
The first part of the question was 23 why did we seek access?
24 JUDGE KELLEY:
Right.
s 25 MR. ROCKWELL:
Because we felt that having access U
ACE-FEDERAL REPORTERS, INC.
2(C-347-3700 Nationwide Coverage Sn3%fM6
28039.0 SRT 1148 O
1 to the operators would be cr itical and fundamental to any 2
thorough investigation, and in, fact we got some pretty nasty i
3 letters from counsel because we felt we had to go right to 4
the limit in terms of pushing for access and we did.
We 5
contacted operators up until -- we contacted every operator 6
that we hadn't beer. specifically told was represented by 7
counsel and we got a lot of nasty phone calls back saying:
8 They are represented by counsel, stay off.
At that point we 9
stopped.
)^
10 JUDGE KELLEY:
Okay.
This was during the course 11 of your five-week investigation?
12 MR. ROCKWELL:
I say that lightly in tone.
They O
13 weren't really nasty.
They were strong and firm and 14 appropriate but we wanted to make sure that there was no l
15 question that we had made the effort.
16 JUDGE KELLEY:
All right.
The next sentence of 17 this question I'm going to disallow.
I'll read it for th j
18 record:
" Provide copy of your first and all other requests 19 and the related responses."
It seems to me that you have 20 covered the substance of those transactions and we are not t
21 going to ask you to do that.
l 22 You've described verbal requests already.
The l
23 last part reads as follows:
Indicate whether notes or 24 transcripts were made by you or others of any verbal requests 25 and responses.
If so, provide the notes and transcripts.
O i
t ACE-FEDERAL [E' PORTERS, INC.
1
.n w-
F
~
28039.0 f-3 R T 1149
)
1 You have already said there were letters 2
exchanged?
3 MR. ROCKWELL:
Our requests to the emplovces were 4
verbal.
We would get their home phone numbers and get on the 5
phone and call them.
6 JUDGE KELLEY:
Right.
7 MR. ROCKWELL:
Then we would get calls and, in 8
some cases, these strongly worded letters back saying:
Stay 9
away.
And I believe we have -- I know in my correspondence 10 file, assuming it has survived, I would have copies of 11 letters from -- what was his name --
l 12 MR. MAUPIN:
Pidelle.
13 MR. MC BRIDE:
Eugene R.
F-i-d-e-1-1-e, a former 14 partner in our firm.
But I will not disclaim responsibility 15 for his letters, your Honor, if you have any questions for 16 me.
17 MR. ROCKWELL:
We could provide copies of those 18 letters if it's deemed appropriate.
19 JUDGE KELLEY:
Given your description I don't 1
20 l think it's necessary.
I'll go on to number 5.
And the rest i
21 of the questions I'll read were submitted in advance.
22 Identify the interviews and other evidence you 23 l used to cone to the conclusion that at least four cperators i
24 had knowledge or other involvement in the improper f performance of leak rate testing at TMI-2.
Name those 25 O
ACE-FEDERAL REPORTERS, INC.
-um x _ m m..
mm -
s
28039.0
(~yRT 1150 V
1 individuals.
2 MR. ROCKWELL:
I quarrel with the premise of the 3
question.
4 JUDGE KELLEY:
All right.
5 MR. ROCKWELL:
We did not come to that 6
conclusion.
We reported assertions of that nature, that.had 7
been gleaned from notes taken by Mr. John Wilson during I&E 8
interviene.
And we reported statements that Mr. Hartman had 9
made in his deposition.
10 Mr. Hartman had specifically refused to provide un 11 names.
We never have had names, and we did not draw the 12 conclusion.
13 JUDGE KELLEY:
I'm sorry.
You say you reported 14 names?
Or assertions without names?
15 MR. ROCKWELL:
Assertions without names.
We 16 simply said:
So-and-so has said, Hartman has said, that so 17 many CRos and so many supervisors have knowledge.
We 18 reported that as a fact.
Namely, he said it.
We didn't 19 endorse it as a finding or a conclusion.
20 The name would go as to information that we got 21 from these admittedly sketchy notes from the I&E interview.
22 JUDGE KELLEY:
Thank you.
That answers that.
23 6:
In your opinion was the improper performance 24 of leak rate tecting reporting limited to these individuals?
25 MR. ROCKWELL:
It really can't be answered.
We i
ACE-FEDERAL REPORTERS, INC.
202-347 37(X)
Nationwide Conerage 8(n336-N46 k
28039.0
(-9RT 1151 G
1 didn't conclude anything about any specific individuals.
2 JUDGE KELLEY:
Thank you.
Number 7:
Describe the 3
cooperative effort (in terms of operations personnel) needed 4
(a) to add water to the core inventory.
5 I'll stop there.
This is in two or three parts.
6 How many people does it take, working together, to 7
add water to the core inventory?
8 MR. M009E:
We did not investigate this, your 9
Honor.
But I can give you an answer, what I think it is, but i
10 I did not confirm it.
You have to use a switch that only 11 takes one individual to do it.
I didn't try to walk through 12 how they actually do it but my understanding is --
o m'f 13 understanding at would take one person.
14 JULGE MEMJdY:
The premisa of the n'.'stion is it L
15 has to be done bv more than one.
So you are disagreeing with 16 the premise; correct?
17 MR. ROCKWELL:
Yes.
18 JUDGE KELLEY:
From what we have heard here I 19 would assume that to be the case.
20 The second part:
Describe the cooperative effort 21 in terms of people that it takes to " jog water to the core 22 inventory."
Is the ancwer the same?
23 MR. ROCKWELL:
Yes, it is, and I'm not sure 24 whether it has ever been established that water was jogged to 25 the core inventory.
We examined that.
We refused to n%J ACE-FEDERAL REPORTERS, INC.
Nat uvimide Coserage Myk334664
^
.m
~ _ -..
28039.0 g-9RT 1152 V
1 include, in fact, it had been done, although we spotted some 2
instances where it might have been done.
3 JUDGE KELLEY:
Very well.
And the last point, the 4
number of people that would be required to add hydrogen to 5
the core inventory?
6 MR. ROCKWELL:
My understanding is one, again.
7 JUDGE KELLEY:
Right.
That's consistent with all 8
the testimony.
9 Number 8:
Was it possible for the operator in 10 charge of the leak rate test to perform the manipulations a, 11 b,
and c described in question 7 that I just read) while 12 conducting the test?
If your answer is yes, describe the O
13 rigors involved for a single individual to conduct the leak 14 rate test and make additions of water and hydrogen.
15 Has that been answered?
16 MR. ROCKWELL:
I think it has.
It would be 17 possible to add water to the core inventory during the test.
18 It would be possible to add hydrogen.
It would be possible 19 to jog water.
So in answer to the initial question, was it 20 possible, the answer is:
Yes, in all cases.
21 In terms of the rigors, or in terms of how it 22 would be done in terms of the control room, we didn't really 23 look at that.
I think others here are far more qualified to 24 address it.
Decause of having interviewed the operators.
I 25 think really our -- there is not a sufficient foundation for O
ACE-FEDERAL REPORTERS, INC.
wn,,
8-+<_,.
l l
28039.0
(-9RT 1153 l
k-)
I us to express a view on this.
2 JUDGE KELLEY:
Thank you.
Number 9:
Is it likely 3
that the manipulations (a, b, c, described in question 7) --
4 the same thing we referred to in the preceding question -- is 5
it likely the manipulations went unnoticed by other control 6
room personnel, from your experience with TMI and other 7
nuclear facilities; and if the answer is yes, provide your 8
rationalization.
9 MR. MC BRIDE:
I object to that question for lack 10 of foundation on the ground Mr. Rockwell has testified he is 11 personally unaware of any such manipulations, unless the 12 question is to be confined to his knowledge of Mr. Hartman's 7-V 13 testimony.
14 JUDGE KELLEY:
Well, I've read it partly for the 15 record.
In light of the prior answer, the fact that you 16 don't need a enoperative team, the premise of the question 17 seems to disappear.
I simply ask Mr. Rockwell if he has any 18
' comment to offer in addition?
19 MR. ROCKWELL:
I would respond to this in much the 20 same way as I responded to the second half of question 8, 21 which is that this was not an area that we investigated.
22 With respect to Mr. Hartman, Mr. Hartman says he was aware of 23 it and says that others were aware.
But that's Mr. Hartman 24 and that is the only information that we have.
25 JUDGE KELLEY:
All right.
I did not mean to O
ACE-FEDERAL REPORTERS, INC.
202-347."w Nationwide roserage m n334 446
28039.0
<-GRT 1154 NA l
1 ignore your objection.
I think the way it unfolded, in 2
effect, are you satisfied?
3 MR. MC BRIDE:
I'm satisfied, particularly because 4
a lawyer answered the question and I think he understood what 5
I was driving at and it all worked out, your Honor.
6 JUDGE KELLEY:
Fine.
10:
In view of the clarity 7
you found in NRC regulations (as summarized in your report) 8 would you consider the operators' consistent failure to 9
record the starting time of leak rate tests a deliberate 13 violation?
Let's stop there.
11 MR. ROCKWELL:
I'm not sure where we referred in 12 our report to the clarity of NRC regulations.
7_
13 JUDGE KELLEY:
Well, you never interviewed any 14 operators, did you?
15 MR. ROCKWELL:
Other than Mr. Hartman, no.
16 JUDGE KELLEY:
Other than Mr. Hartman, to the 17 extent this goes to state of mind, are you in a position to 18 answer?
19 MR. ROCKWELL:
No.
20 JUDGE KELLEY:
Okay.
The rest of 8 reads like 21 this:
Would you consider the discarding of leak rate tests 22 in excess of technical specifications a deliberate 23 violation?
If either answer is no, provide relevant 24 rationalization.
25 Again, to the extent it goes to " deliberate," are O
ACE-FEDERAL REPORTERS, INC.
202 347-374o Nationwide Coserage 800-33&6M6
28039.0 g-SRT 1155 V
1 you in a position to answer?
2 MR. ROCKWELL:
No.
3 JUDGE KELLEY:
All right.
Question 11:
How would 4
you agree or disagree with the attached (Aamodt) summary of 5
the conclusions of your report?
Attached two pages.
6 There were attached to the questions, two pages 7
which we supplied.
Have you had an opportunity to consider 8
that attachment?
9 MR. ROCKWELL:
Yes, I have.
10 JUDGE KELLEY:
Okay.
Do you' agree or disagree?
11 MR. ROCKWELL:
I disagree.
If that is intended to 12 be a sumnary of the conclusions of our body of work, I can't 13 even begin to tell you how I disagree, i
14 JUDGE KELLEY:
I should add that, perhaps that l
15 Mary, too -- we provided that to you.
It is not in the
[
16 record.
It is not served to the parties.
I could read the
(
17 page and a half in question, if Mr. Rockwell is saying --
18 well, you said you disagreed in a number of ways, I gather?
19 MR. ROCKWELL:
It is totally unrepresented of our 20 work.
If the question is what is a fair -- which is the 1
21 implied question, because it asks me to tell how I disagree l
22
-- what is a fair summary of our work, we'll come back 23 tomorrow and organize a summary of our work.
But it is a l
24 complex piece of work.
It's not fairly summarized by two 25 pages.
It's highly selective, it's unrepresentative.
l O I
ACE-FEDERAL REPORTERS, INC.
3c-347 3700 Nationwide Coverage m&33MM6
28039.0
(-yRT 1156
(_./
1 JUDGE KELLEY:
I guess I wasn't clear myself.
Do 2
you read this to purport to summarize your entire effort in a 3
page and a half?
4 MR. ROCKWELL:
All I can do, sir, is to read the 5
question.
It says, how would you agree or disagree with the 6
attached summary of the conclusions of your report, which 7
seems to me global in scope.
So I'm going to say it is not a 8
fair summary of our report.
9 JUDGE KELLEY:
We'll accept that as an answer 10 under the circumtances knowing the nature of your report and 11 the complexity of the issues.
12 MR. ROCKWELL:
In fact, if I may point out, we 13 wrote a summary which is 80 pages long.
We could summarize 14 it again in 8.
But that's exactly what our summary volume 15 was intended to do.
16 JUDGE KELLEY:
Okay.
Number 12:
Provide the 17 names of all persons, companies and other entities to whom 18 you provided your report and the dates of provision.
I think 19 that's probably rather broad, perhaps an unreasonable 20 request.
[
21 Can you give us in general an indication of who l
22 your report went to and around when?
23 MR. ROCKWELL:
We provided a copy to Metropolitan 24 Edison, a copy to Shaw Pittman, and a copy to each member of l
25 the investigatJve team.
And that's it.
It's a rather simple l
i l
l ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-(M6
28039.0 cSRT 1157 Y
1 answer.
2 JUDGE KELLEY:
It really just does ask who you 3
provided it to.
Where it has gone since, who knows?
But 4
that ansvers the question.
5 MR. ROCKWELL:
In fact, we made an inquiry because 6
we weren't sure of the scope of the question, to GPU Nuclear, 7
as to where they had disseminated it and the conclusion was 8
that we should limit ourselves to those to whom we 9
disseminated it.
1 don't know to whom they --
10 JUDGE KELLEY:
That's what the question says.
11 We'll go with that.
12 Number 13:
What contacts have you had with 13 Mr. Edmund Stier or his co-investigators, representatives or 14 other agents?
Provide the dates and substance of these 15 contacts, indicate whether notes or transcripts of these 16 contacts were made by anyone.
Indicate who initiated the 17 contacts.
Provide any relevant correspondence.
18 Strike at least the last sentence about relevant 19 correspondence.
Could you, at least, describe your contacts f
20 with Mr. Stier?
21 MR. ROCKWELL:
Yes.
I met Mr. Stier at 9:00 22 Monday morning this week.
I had never met him, seen him or 23 i talked to him before.
24 MR. MOORE:
I presume these are addressed to both 25 Mr. Rockwell and I?
s ACE-FEDERAL REPORTERS, INC.
202 4 474 700 Nationwide Coserage mn3h6646
l 28039.0 f
- RT 1158 x_/
1 JUDGE KELLEY:
Yes.
Technically it said 2
Mr. Rockwell, but I think it means the Rockwell team.
3 MR. MOORE:
I did have some interface with 4
Mr. Stier, I guess it was early in your investigation.
I 5
don't recall the dates or the exact detail of the meetings 6
with Mr. Stier, but it was basically to provide him'with some 7
information, I guess, on how we got started on the Faegre &
8 Benson report; is that correct?
Mr. Stier might have a 9
better recollection than what I have.
10 JUDGE KELLEY:
Do you want to comment, Mr. Stier, 11 from your end?
12 MR. STIER:
Yes.
Early in our investigation I met O
13 with Mr. Moore and others, other members of the team that 14 worked on the Faegre & Benson report.
They provided us with 15 background information, answered certain specific questions 16 about the technical analysis that had been done for the 17 report.
I don't remember precisely when those meetings took 18 place or how many there were.
They were rather informal 19 contacts.
20 JUDGE KELLEY:
Okay.
Thank you.
21 MR. HOCKWELL:
And I guess it should be, perhaps 22 helpful in rounding out the record, this is at a time after 23 our investigation team had been disbanded.
So it was not 24 consulting with the Faegre & Benson investigation team per 25 se, but people who had been on that team but were also g-)
V ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cowrage Nn33&M
28039.0 e-9RT 1159
(
)
1 employees of Metropolitan Edison or GPU Nuclear and were 2
available locally at that time.
3 JUDGE KELLEY:
Okay.
Give me just a moment.
4 Number 14 is cut down from the -- modified from 5
the original question.
The question as posed to 6
Mr. Rockwell, simply:
Identify the authors of the Faegre &
7 Benson report.
Are they the people on page 1 of the 8
overview?
9 MR. ROCKWELL:
Yes.
To provide some additional 10 clarity, volum'e II, which is the series of technical reports, 11 the author of each technical report is listed by chapter so 12 that one can tell who the principal author was.
With respect O
13 to volume I overview, I was the principal author of that; and 14 with respect to various exhibits, in most cases the author of 15 those exhibits is identified by name.
16 JUDGE KELLEY:
Thank you.
Just a moment.
17 (Discussion off the record.)
\\
18 JUDGE KELLEY:
We have one remaining question.
I 19 It's a question that, in our review of the questions as 20 submitted we debated about a bit and decided to go ahead and l
21 submit it to the witness.
I think it's the kind of question l
22 that is debatable.
Let's see whether any counsel have any 23 objection to it.
If not, if Mr. Rockwell doesn't mind 24 answering it, we'll get him to answer.
And if we hear 25 objection then we may just let it pend, pending the arrival O
ACE-FEDERAL REPORTERS, INC.
202-347-3hu)
Nationwide Coserage Mk3346646
28039.0 r-BRT 1160 r
1
\\_/
1 of Mr. Aamodt.
2 The question is this:
What were your fees in 3
total for the investigation of leak rate practices at TMI?
4 MR. BLAKE:
You have one objection.
5 JUDGE KELLEY:
Let's wait then on it.
I would 6
like to get Mrs. Aamodt here to have a chance to state the 7
other side.
It's a number we can get easily enough later if 8
the Board rules against your object on.
9 MR. ROCKWELL:
If I may just add some light 10 without answering the question, I have brought the numbers 11 with me.
We can provide it by letter.
I would add that we 12 did two investigations, as you know.
We did a reactor 13 start-up investigation at the same time as we did this one, 14 so any number I would provide would be on the high side 15 because of the time devoted to the ECP question.
16 JUDGE KELLEY:
Okay.
17 MR. ROCKWELL:
But pending advice from the Board 18 we'll provide this answer in writing if someone will let us 19 know.
l l
20 JUDGE KELLEY:
Fine.
One way or another we'll, if I
21 we decide not -- to drop the question, we'll let you know, i
22 If we need to ask the question, we'll ask you.
Thank you, i
23 MR. MC DRIDE:
Judge Kelley, lest there be any l
24 doubt, the nature of my objection is not that there cannot be 25 appropriate questions put to Mr. Rockwell, as I think the i
i I
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Natiotimide Coverage fkt.k34M46
I 28039.0
- r-SRT 1161 U
1 record will reflect, which would go to how much effort they 2
put into it, were they restrained by the company.
This one I 3
regard as inappropriate and immaterial to what the Board 4
needs out of Mr. Rockwell.
5 JUDGE KELLEY:
There are different ways of getting 6
at the kind of thing that is relevant.
Hence, our conceding 7
some ambivalence about the question, at least at this point, 8
so we are not going to pursue it any further right now.
9 We'll see later if Mrs. Aamodt wants to press her desire to 10 ask the question, or might have an alternative approach, we 1
11 will see.
12 JUDGE BRIGHT:
Mr. Rockwell, we have some
(~.\\
U 13 follow-up questions here, on the previ,ous section.
14 Reg Guide 1.45, and Faegre & Benson report tab 7, 15
- states, "a small amount of unidentified leakage may be 16 Impractical to eliminate but it should be reduced to a small 17 flow rate, preferably less than 1 gallon per minute to permit 18 the leakage detection systems to detect positively and 19 rapidly a small increase in flow rate."
20 Does the word " preferably" imply a degree of j
21 imprecision in the necessity of the choice of a limit of 1 22 gallon per minute for unidentified leakage?
23 MR. ROCKWELL:
I think it's an appropriate 24 question for Mr. Moore to answer, 25 MR. MOORE:
This goes back, if you recall, early p)
\\~
ACE-FEDERAL REPORTERS, INC.
28039.0 g-BRT 1162 0
1 in the hearing, I related my confusion over what I perceived 2
to be two 1 gpm limits in the reg guides one limit of 1 3
gpm for background leakage; another 1 gpm limit for an I
4 increase above that background leakage.
I'm not sure how 5
that -- could you read the question one more time?
6 JUDGE BRIGHT:
Well, the actual question says:
7 Does the word " preferably" in Reg Guide 1.45, imply a degree 8
in imprecision in the necessity of the choice of 1 gallon per 9
minute for unidentified leakage.
10 MR. ROCKWELL:
Let me make a response here, which 11 is that if one is -- I can't answer as to what was in the 12 mind of the drafter of the document.
Reading the language as
(~'#\\
13 quoted, that argument can be made.
14 JUDGE BRIGHT:
Very well, then, let's follow that 15 up.
If yes -- we will call it a qualified yes, anyway --
16 does that imprecision imply the appropriateness of an 17 evaporative loss factor in the leak rate test?
18 MR. ROCKWELL:
I think you are getting beyond what i
19 I'm comfortable addressing.
Mr. Moore was principally 20 involved in analyzing Reg Guide 1.45, and I would prefer that 21 he respond.
22 MR. MOORE:
With respect to that quettion, if you 23 read the reg guide, I believe the reg guide is addressing 24 provision for some background leakage and then sensing 25 increase above that background leakage.
The way it was --
(
ACE-FEDERAL REPORTERS, INC.
202 347.Dal Nationwide Cos erage N00.14&M
28039.0 e- !RT 1163
(
v 1
you go from this reg guide to the way it was implemented at 2
TMI-2, it is radically different from this.
Either -- it 3
looks like, in my mind what we have at TMI-2 was an either/or 4
situation.
You may have either background leakage or an 5
increase above background, as long as you don't -- the sum of 6
those two does not exceed 1 gpm, as reflected in prior 7
testimony.
I made reference to a standard review plan 8
wherein there was reference to background leakage.
So, also 9
of keeping a record of what this background leakage is, s o *g 10 that one can determine if there is an increase above that 11 background value.
12 So -- but as implemented in TMI-2, there is no i
/
13 direct provinion for background leakage.
14 MR. ROCKWELL:
I might add we observed in our 15 report that there was not an evaporative loss factor and that 16 we felt that one could have been provided for TMI-2.
That 17 was what we said in the report.
18 JUDGE DRIGHT:
One more.
Does the word " rapidly" 19 imply, in your judgment, a disagreement with 20 Mr. Kirkpatrick's testimony that the leak rate test should be 21 l run for more than 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />')
Referring back to the part of 1.45 1
22 which says " permit the leakage detection system to detect 23 positively and rapidly the small increased flow rate."
24 MR. ROCKWELL:
There seem to be two issues buried 25 there:
One is, how often do you run the test; because you qV ACE-FEDERAL REPORTERS, INC.
-n.,
8_
c-..
20039.0
(~3RT 1164 V
1 will discover it more rapidly the more often you run the 2
test.
That goes to the 72-hour interval.
Obviously if you 3
are hypothesizing a change in which we have gone from an 4
acceptable unidentified leakage to some greater value that is 1
5 unacceptable, if you run it for four hours you are not going 6
to have that unacceptable value until the end of four hours 7
as opposed to the end of one hour.
6 But gi/en the fact that the interval for 9
demonstrating compliance with the LCO is 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, that seems l
10 to me a relatively minor consideration.
11 MR. MOORE:
Also given the fact that they were 12 running, apparently running back to back tests.
If things C:')
13 had really gotten out of hand they would have had -- because 14 of the frequency they apparently were running them -- they 15 would have had a fairly rapid -- 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> indication, because 16 of the way they were running the tests.
17 JUDGE KELLEY:
I believe that takes us through the 10 questions directed specifically to Mr. Rockwell and 19 Mr. Moore.
i 20 We did say that we would do these follow-ups in 21 chunkc, and then there would be one last clear chance, one 1
22 bite at the apple -- two bites; this is the last one.
23 Does anybody else have any more questions for 24 either Mr. Rockwell or Mr. Moore?
25 The Board, I gather, han none.
i i
ACE-FEDERAL REPORTERS, INC.
202-347 1700 Nationmide rmerage mnJ4MM
.. _ _ _, _ _ _ - _ _...., _ _ _ _ _ _ _ _ _ _. _ _ _ _ _,~- _,,___ _ _..___._,_._ _ _. _ _ _.. _ _ _. _ _ _ _ _ _, _ _ _ _. _ _, -
p._._.._ _._______ _ _ _ _ _____ _ _ ~ - _ _ _._ _ _
1-I I
l f
L j.
I 28039.0 i
p)SRT 1165 4
1 That being so, gentlemen, you are welcome to sit j
)
1 2
in the rest of the day, but just on behalf of the Board, let 1
l 3
me say that you obviously have done a great deal of work on l
)
j 4
this pretty complicated set of issues in the past.
You have I
j S
also done a lot of work to got ready for the hearing, which 1
l 6
we very much appreciate.
Your presence here has been l
7 extremely helpful, I think not just for your answers but for f
8 the interchange that has taken place among all the experts 9
here.
We think we are in much better shape as a result of f
10 that process.
[
11 So, thank you very much.
There are one or two 12 things left hanging we may be back to you about, but O
13 basically we'll say thanks and excuse you.
Thank you very i
t 14 much.
15 MR. ROCKWELL:
Thank you, your Horbr.
It has been 16 a pleasure to work with all ot you and everyone else here in 17 the room we have had a chance to meet.
b 10 JUDGE KELLEY:
Pine.
We are going to quit at 3:30 19 today.
Let's take a five-minute stretch break and then sit 20 down and finish.
f i
i 21 (Recess.)
f e
22 JUDGE KELLEY:
The next question then would be the I
t 23 Russell /capra question 2 -- it's 269 f or Kirkpa trick and 324 f
I 24 for Mr. Stier.
TMI-2 adminis tra tive procedurec.
j 25 The question reads:
Were the administrative
) O i
l ACE-FEDERAL. REPORTERS, INC.
I M 14747at Nat onnte Cmerage an3%6M6
[
.n-,
28039.0 1166 p)3RT
(
1 procedures applicable to TMI-2 included in the legal 2
requirements imposed on the TMI operators?
3 I'll note the question is legal in nature.
4 MR. GOLDBERG:
Excuse me, that is not 269 to S
Mr. Kirkpatrick; that's 269 to Mr. Russell and Capra.
4 6
JUDGE KELLEY:
That's what I said.
The same 7
question is asked, is it not, to all the witnesses, minus O
Rockwell?
9 MR. MC DRIDE:
It's 324 to Mr. Stier --
10 MR. GOLDBERG:
Sorry.
11 JUDGE KELLEY:
Were the administrative procedures 12 applicable to TMI-2 included in the legal requirements O
13 imposed on the TMI-2 operators?
14 MR. KIRKPATRICK:
I believe they were but I think 15 I should deter that question to Mr. Russell, since he would 16 have a much more complete answer than I would.
17 JUDGE KELLEY:
All right.
10 MR. HUSSELL:
The answer is yes, they are.
Title 19 10, Code ot Federal Regulations, 50.36 requires aach 20 applicant for a facility license to include in its 21 application proposed technical specifications.
Included in 22 those technical specifications must be, first, surveillance 23 requirements relating to tests, calibration or inspection to 24 insure that the nececsary quality of systems and components 25 is maintained; and, two, administrative controls rela ting to I
ACE-FEDERAL REPORTERS, INC.
m,_
s m_. r_
..o, _
28039.0 r-9RT 1167 b
1 organization of management, procedures, record-keeping, 2
review and audit and reporting requirements necessary to 3
assure operation of the facility in a safe manner.
4 TMI-2 technical specification 6.8 entitled 5
" Procedures" requires written procedures be established, 6
implemented and maintained in areas of surveillance and test 7
activities of safety-related equipment and specifically 8
includes those procedures identified in Appendix A of 9
Regulatory Guide 1.33, November 1972.
10 The administrative procedures identified in 11 Appendix A of this guide includes:
A.0, log entries and 12 record retention; A.7, shift relief and turnover; and A.3, O
13 equipment control (for example, locking and tagging).
14 The relevant TMI-2 surveillance procedure is SP s
15 2301-3D1 "RCS inventory" and the relevant TMI-2 16 administrative procedures are AP-1010, technical 17 specification surveillance program"; AP 1012, " shift relief 18 and log entries"; and AP 1036, " instrumentation out of 19 service control."
i 20 Additionally, 10 CFR50.51 " Conditions of Licenses" 21 requires licensed operators to manipulate the controls of the 22 facility and be present at all times during the operation of 23 the faellity.
24 Operators are issued licenses under 10 CFR 55 that i
l 25 specify as follows: "The licensee shall observe the operating l
l ace-FEDERAL REPORTERS, INC.
mm.,
~ _m m,.
..o, _
e.-.---__-
5 J
i i
l l
28039.0 pBRT 1168 l
b I
procedures and other conditions specified in the facility i
l 2
license which authorize operation of the facility or i
3 facilities."
4 Finally, the utility was licensed to possess, use, i
5 and operate the facility in accordance with the procedures l
6 and limitatior.s set forth in the facility license, which l
f 7
includes the technical specifications.
Therefore, the TMI-2 4
8 operators were required through their operator and/or senior 1
9 operator licenses, and the facility license and technical
(
10 specifications, to comply with the licensee's administrative l
[
11 procedures, AP-1010, AP 1012, and AP 1036, and surveillance I
12 procedure SP 2301-3D1, which the Staff beleives are the O
1 13 relevant procedures to this proceeding.
14 JUDGE KELLEY:
Thank you.
i i
15 Mr. Stier?
i 16 MR. STIER:
The question calls for a legal 17 analysis that I was not called upon to make during the course i
10 of my investigation and which I did not make.
19 JUDGE KELLEY:
Okay.
Thank you.
20 Number 267:
If the administrative procedures were
[
l 21 not legal requirements imposed on the TMI-2 operators, do you l
i 22 agree with the several references in this proceeding to i
f 23 alleged violations at administrative procedures?
)
i 24 I gather you say they are legal requirements;
[
25 right?
(
! O i
ACE-FEDERAL REPORTERS, INC.
(
2tcM47 3'(t)
Nanonulc rmerage M4Xt)h(M6 l
20039.0 fqRT 1169 V
1 MR. RUSSELL:
That is correct.
2 JUDGE KELLEY:
Okay.
Note that.
Any further 3
comment on this question?
4 MR. RUSSELL:
The only additional comment would be 5
that there are many cases in NRC practice of enforcement 6
actions being taken for violations of licensee administrative 7
procedures, as well as other licensee procedures.
Failures 0
to tollow those procedures can be categorized as violations, 9
open items or other characterizations, depending upon the 10 severity.
11 So, in this case a violation of the procedure is 12 intended to be descriptive and not necessarily representative 13 of a severity classitication that we would use in an 14 enforcenent proceeding.
15 JUDGE KELLEY:
Mr. Stier, any comment?
16 MR. STIER:
My question is somewhat different.
It 17 reads, "If the administrative procedures were not legal i
15 requirements imposed on the TMI-2 operators, why does your 19 report refer in several places to alleged violations of 20 administrative procedures? "
21 The answer is I was retained by GPU Nuclear to 22 look for deficiencies in the performance of their employees 23 and to provide them within sight into how those deficiencies 24 occurred and to make -- help them make some assessment of 25 their severity.
ace-FEDERAL REPORTERS, INC.
x mw, s _ a.m.
,m n,,-
20039.0 f-~8 R T 1170 b
1 The administrative procedures are binding within 2
the framework of the company on their employees, and it is 3
against those standards that I measured the conduct of the 4
employees.
5 JUDGE KELLEY:
Were you using the term " violation" 6
as a term of art connoting a legal violation or merely that 7
somebody didn't comply with some procedure?
8 MR. STIER:
The latter.
9 JUDGE KELLEY:
Thank you.
10 268 of Russell /Capra, I guess it is.
271 11 Russell /Capra, and 268, Kirkpatrick.
12 Do you agree with the suggestion that the NRC O
13 should or might take action against an operator found to have 14 violated only an administrative procedure?
15 The Board is tempted to strike this one for 16 vagueness, but if you have a comment, Mr. Russell?
17 MR. RUSSELL:
For individuals who willfully fail 18 to comply with the requirements of important administrative l
19 procedures, enforcement actions may be appropriate and 20 necessary, e
21
" Willful," as I have used it, means to me i
22 deliberate failure to comply, such that the person had 23 l knowledge of the requirement and knowledge that it was not 24 followed.
" Willful" could also be a recklese disregard of 2S requirements.
ACE-FEDERAL REPORTERS, INC.
- e: m m, s_ um_,.
.. m_
20039.0 SRT 1171 1
For example, an operator would show reckless 2
disregard if he had knowledge of the requirement, had 3
indication that the requirement was not being followed, and 4
made a decision not to follow up on his indications.
5 In simple terms, an operator who looked the other 6
way when technical specification requirements were not being 7
followed would be demonstrating reckless disregard of his 8
license responsibilities.
Important administrative 9
procedures, to me, are those that are specified in the 10 license or in the technical specifications as they are in 11 this case.
12 Administrative procedures 1010, 1012 and 1036 are O
13 the administrative procedures I am discussing.
14 JUDGE KELLEY:
Maybe the key word in the question 15 is the word "only."
Where it says, "you have only violated 16 administrative procedure." That would suggest that they 17 really aren't very important.
10 Do you regard the administrative procedures 19 involved here as having significant safety implications?
20 MR. RUSSELL:
The administrative procedures here 21 are not only covered generally in the regulations, in 50.36, i
22 but they are covered explicitly in the facility technical 23 specifications and we require people to follow those 24 proceduren.
25 JUDGE KELLEY:
Okay.
O f
ACE-FEDERAL REPORTERS, INC.
i 202 347 37m Ntionanle rmerage mn3%N4
28039.0 f-~BRT 1172 O
1 MR. RUSSELL:
The answer is yes.
2 JUDGE KELLEY:
Mr. Stier, any comment on that?
3 MR. STIER:
My question asks whether 1 intend to 4
suggest that the NRC should or might take action against an 4
5 operator found to have violated only an administrative 6
procedure?
I don't intend to make any r2 commendations.
l 7
JUDGE KELLEY:
All right.
8 Follow-ups on just those three, we might just 9
ask?
10 MR. MC BRIDE:
Yes, your Honor.
And I have two i
11 documents here tha t are referred to in the f ollow-up 12 questions for the Board's convenience.
They are my only OO 13 copies and they are actually lent to me so I would appreciate 14 it if I could get them back.
15 (Discussion off the record.)
16 JUDGE KELLEY:
Back on the record, we have several 17 questions from the numerous employees growing out of these 1
18 last three Os and As.
We have been handed two documents that 19 relate to the questions.
20 Let me just be sure we are straight on these.
21 Mr. l'c B r i de, your first question cites Regulatory Guide 22 1.33.
One of these documents is called " Safety Guide 33."
23
- Is that the same thing?
24 MR. MC BRIDE:
That's precisely the point of the 25 first question, your Honor.
If you'll turn to the technical ace-FEDERAL REPORTERS, INC.
20244t37sti Nationwide Coscratte M A UMM6
28039.0 QRT 1173 i
1 specifications, 6.1.8.a, you will see the document referred 2
to as " Regulatory Guide 1.33, November, 1972."
3 JUDGE KELLEY:
All right.
I think I understand.
4 The question to Mr. Russell is as follows:
Is it 5
a fact that there is no document with the exact title, 6
" Regulatory Guide 1.33, November, 1972", as set forth in 7
TM; -2 tech specification 6.8.la?
I have a copy of that 8
portion of the tech specs.
It doea refer, among other 9
things, to the applicable proceduIes recommended in Appendix 10 A of Regulatory Guide 1.33, November, 1972.
11 I'll stop there.
Do you understand the question?
12 MR. RUSSELL:
Yes, I do.
This is a rather simple i
i 13 administrative convenience on the part of the Staff, that we 14 did not, after the AEC to NRC changeover, when we changed to 15 the termJnology " regulatory guide" from " safety guide," go 16 back and reissue all of the safety guides as regulatory 17
- guides, i
18 Safety Guide 1.33
- Safety Guide 33 is identical 19 to Regulatory Guide 1.33, and it just was not revised because 20 we had not changed that since 1972.
21 MR. MC BRIDE:
Judge Kelley, excuse me.
That 22 doesn't answer the question.
The question was:
Is it a fact 23 that there is no document with that exact title?
24 JUDGE KELLEY:
It seems to me he did answer-the 25 question.
That is a fact.
The exact title is different now; O
ACE-FEDERAL REPORTERS, INC.
202..u? 370tl Natsonwide Cmcrge Nnl%fM6
2e039.0 RT 1174 1
correct?
2 MR. RUSSELL:
The exact title is different.
The 3
content is the same.
And the Reg Guide 1.33, which is the 4
quality assurance program requirerr.ents for operation, 5
November 1972, is the document that is specified in the 6
facility technical specifications.
7 JUDGE KELLEY:
can you tell us how it becomes 1.33 8
as opposed to simply 33?
9 MR. RUSSELL:
1 is power reactor series of 10 regulatory guides.
Thoce regulatory quides issued by the NRC 11 after it was created, which relate to power reactor operation 12 start with the designator 1.
Those in the raa terials area, the
(
):
13 test or reactor area start with other designators.
The.33 14 would indicate this is the 33rd guide in the power reactor 15 series and it is the old safety guide numt.er 3 3.
16 JUDGE KELLEY:
Then J quess my follow-up question 17 would be:
Do you think operators, line people who have 10 occasion to refer to the document, might be confused as to 19 what document was intended if they were to look in the tech 1
20 l specs and see the reference to 1.33?
Would they have trouble 21 finding it?
Would they know which one it was?
t 22 MR. RUSSELL:
The issue, I think, that's 23 significant is that operators generally operate the plant by i
24 the procedures that are issued.
They don't generally operate i
25
[ by the tech specs unless the procedures refer them back to I
l i
l l
ACE-FEDERAL REPORTERS, INC.
i se me.,
~ _m -
- m.. _
28039.0
<~SRT 1175 N-1 the tech specs.
2 The operators are knowledgeable about the tech 3
specs; we examine them on them, but they in fact operate the 4
plant by the facility operating procedures, i
S MR. MC BRIDE:
Judge Kelley, I object to that 6
answer and move to strike it on the ground that the witness, j_
7 who is a nonlawyer began the answer with, "what I think is 8
significant" or words to that effect, which is, in essence, 9
the witners serving an evidentiary objection.
It seems to me 10 if the Board put the question to him, the witness ought to 11 answer the question.
12 JUDGE KELLEY:
I think we already said about two O
13 minutes ago as far as we are concerned, he did answer the 14 question.
I'm asking him a follow-up question.
I want to 15 know whether this would cause confusion.
He's attempting to 16 answer that point.
i 17 I don't know, too -- it's a matter of degree --
l 10 whether in the course of follow-ups we are going to get 19 continually into colloquies back and forth and we are just 20 going to get follow-ups submitted to us and we are going to 21 read them and that is going to be that, subject, perhaps, to 22 claritication.
But this is not, I don't think, going to be 23 an occasion for a sort or indirect cross-examination by 24 counsel.
That's not what this proceer ng is supposed to do.
25 So, where do you come out on this?
I can 1
ACE-FEDERAL REPORTERS, INC.
202,34719U Nat sonaide Coscrage
$11334-6M6
_ _, _ - -, ~ _
-.,m--------_._..._
20039.0
-s9RT 1176 s_)
1 understand, if I were doing legal research in a library 2
someplace and I looked up this set of tech specs and it cited 3
me to 1.33, I might, I suppose, take a while to find Safety 4
Guide 33.
I wouldn't know unless somebody told me it is 5
exactly the same thing.
6 You have given me some of the background here.
Do 7
you think that there is a basis, here, for contusion by 8
operators or not?
9 MR. RUSSELL:
No, I do not.
10 JUDGE KELLEY:
Is that because -- and if I'm i
11 restating your answer, bear with me -- you say they follow 12 the procedures in their day-to-day work; they don't, as a 13 day-to-day matter, look at the tech specs; is that correct?
14 MR. RUSSELL:
That's correct.
It's clear that the 15 licensee implemented procedures that were required by the 16 technical specification and the licensee procedures in this 17 case have been identified.
Operators use those procedures.
18 JUDGE KELLEY:
Would the operator, in the course i
19 of looking at the procedures, get the substance of the 20 guidance he might need that he would get if he were at Reg i
21 Guide 1.33?
Again as a general matter.
We are not asking 22 about specifica.
We haven't put specifics.
23 MR. RUSSELL:
All he would get out of reading Reg l
24 Guide 1.33 or, in this case, Safety Guide 33, would be that 25 there must be a procedure entitled, for instance, " Shift t
O ACE-FEDERAL REPORTERS, INC.
s. - + c -,.
28039.0 cSRT 1177 i
.I
'w/
1 Relief and Turnover."
It would give him no substance as to 2
what he is required to do as a part of shift relief and 3
turnover.
He must, in fact, go to the facility procedures to 4
find out what it is he has to do.
5 JUDGE KELLEY:
Thank you.
6 Next question:
Is it a fact that the facility 7
license for TMI-2 did not include any explicit reference to 8
administrative procedures?
9 MR. RUSSELL:
I would need some time to check that 10 because, in general, the facility license itself references 11 the regulations in effect and the regulations at 50.36 12 require administrative procedures.
The license, however,
,-)
13 does not reference, or the technical specifications, specific 14 administrative procedures of the facility.
It only requires 15 that they exist.
The numbering, for example, would not be in 16 the license or the technical specifications.
17 JUDGE KELLEY:
Do you need to check or are you 18 pretty well sa tisfied with the answer?
19 MR. RUGSELL:
I'm satisfied with the answer.
I 20 l don't believe that there is -- it would be very, very unusual 21 if it were.
22 JUDGE KELLEY:
All right.
A couple of -- three 23 more questions.
I haven't cross-referenced to Mr. Stier.
24 They were directed to Mr. Russell.
Perhaps I'll finish with es 25 Mr. Russell and then it you want to comment on any of them.
U ACE-FEDERAL REPORTERS, INC.
mm
, _ m..
28039.0 BRT 1178
]
1 Mr. Russell, do you know whether the TMI-2 2
operators were ever shown a copy of Safety Guide 33 during 3
the operation of Unit 2, whether during actual operation or 4
while in training?
5 MR. RUSSELL:
No, I do not.
But it would be very 6
unusual if they had been.
It is generally a licensing 7
document used by a licensing engineer or the group that is O
responsible for drafting procedures.
Not necessarily used 9
for training.
10 JUDGE KELLEY:
Okay.
The next part of it, related 11 question:
If not, then how would an operator have known what 12 Safety Guide 33 provided?
O 13 MR. RUSSELL:
As I indicated before, it is not 14 important that the operator know that Safety Guide 33 15 requires administrative procedures.
It is important that he 16 operate the plant in accordance with the procedures that are 17 given to him for operating a plant by the facility licensee.
I 18 And he, indeed, by the terms of his license, is obligated to 1
l 19 operate the facility in accordance with the operating i
20 procedures of the facility licensee.
21 JUDGE KELLEY:
Can I infer from your answers that i
22 the reg guides, the audience for reg guides is not the 1
23 hands-on operators but it is the people at the facility, or 24 maybe in Reading, Pennsylvania, who write specific procedures 25 for the plant?
ACE-FEDERAL REPORTERS, INC.
u,e. u, c.,m,,
m.
28039.0
-SRT 1179 (d
1 MR. RUSSELL:
That is correct.
The regulatory 2
guides are guidance to applicants, such that they can propose 3
programs to the Staff which would meet the requirements of 4
the regulation.
In this case, the regulation being 50.36, 5
that require certain administrative procedures; this is where 6
we spell out which procedures they must have.
It does not 7
give details on how they conduct or write the content of what 8
is within those procedures.
j 9
They are allowed to tailor that to their own needs 10 at their facility.
11 JUDGE KELLEY:
Finally, did you answer questions about administrative procedures with or without advice of 12 O
13 counsel?
14 MR. RUSSELL:
I answered them with advice of 15 counsel and based upon my own knowledge of representing the 16 NRC's view in operator licensing matters.
I'm the director 17 of the division that has responsibility for all operator 18 licensing in the NRC.
19 JUDGE KELLEY:
Thank you.
Mr. Stier on that 20 series, any comment?
21 MR. STIERt It seems to ne all of them have to do a
22 with the legal obligations of operators.
I have no comment.
23 JUDGE KELLEY:
Thank you.
24 MS. WAGNER:
Staff has one follow-up.
25 JUDGE KELLEY:
Okay.
ACE-FEDERAL REPORTERS, INC.
I 20214717m Nationwale rmerage
%n341M6
28039.0
~BRT 1180 1
Question from the Staff for Mr. Russell.
Did the 2
NRC publish a matrix or inform licensees of the NUREG guide 3
numbers and titles to uniquely identify reg guide documents 4
such as reg guide 1.33, November, 1972?
5 MR. MC BRIDE:
Your Honor, I object on the ground 6
this is an improper line as to whether line operators and 7
their supervisors outside the audience your Honor earlier 8
referred to did or did not participate in alleged 9
falsification of leak rate tests at TMI-2, and the Commission 10 specifically excluded the management of Metropolitan Edison 11 above the level of Mr. Herbein from the inquiry.
12 This question, despite the previous answers that O
13 your Honor just elicited, may be read to imply that the 14 operators had some knowledge that only someone in Reading, l
15 Pennsylvania or Sidney, New Jersey, may have had.
Therefore 16 it is irrelevant insofar as our clients are concerned.
I 17 JUDGE KELLEY:
Your question initially went to is 18 there such and such a document with such and such a title.
19 We then went down the road of how that all came about.
The 20 purpose of the question, as I understood it, was, might this 21 not have been some source of confusion?
And tried to sort 22 that out.
Now this question would bear on whether the NRC 23 actually told licensees and applicants:
We've got a new l
24 numbering system for reg guides.
That seems to me to fit in 5
with what we have been talking about.
ACE-FEDERAL REPORTERS, INC.
mm.,
, _ m..,.
__ ____-_ _ _ m,__ _ __.___,.
28039.0 e'G RT 1181 1
MR. MC BRIDE:
But the problem is that presumably 2
that matrix would go to the same audience that your Honor's 3
earlier question would refer to and that audience did not I
4 include the people who were the subject of the inquiry.
5 JUDGE KELLEY:
You are impeaching your own 6
question, then.
You are saying your original question is 7
irrelevant, i
8 MR. MC BRIDE:
No, your Honor.
Not at all.
If 9
your Honor please, the technical specifications that these 10 operators were obliged to follow, which I put before your i
11 Honor for convenience, refers to a document entitled 12
" Regulatory Guide 1.33, November, 1972." All my question was i
13 designed to establish was that there is no such document by 14 that title.
15 Now the Staff is saying oh, but there's a matrix 16 over here that goes to a different audience that tells you 17 they are the same.
I'm saying of what relevance is that to i
18 an operator?
He never had the matrix.
The technical 19 specifications didn't include it.
20 JUDGE KELLEY:
He probably never had the tech 21 specs.
Not "had" but that the testimony is that what the 22 operators looked to were the administra tive procedures.
23 I'm going to overrule the objection.
I think it's 24 trivial.
We have talked about this, we are trying to nail 25 down the sourco of the contusion over this title and we are O
i ACE-FEDERAL REPORTERS, INC.
mm.,
~ ~, m -
..a m
28039.0 1182
(~9RT 4
x_/
1 going to do so by finding whether or not there was such a 2
matrix.
3 Was there or was there not?
4 MR. RUSSELL:
There was and there still is and 5
we'll be happy to provide a copy for the record.
6 JUDGE KELLEY:
Is it short?
7 MR. RUSSELL:
The regulatory guide index and the 8
other information with i t, I don't know how many pages, 9
probably less than 10.
10 JUDGE KELLEY:
All right.
Thank you.
11 We are just a few minutes away from the time we 12 have to stop for this evening.
Let me ask if counsel have 7.).
(
\\
13 other things that we should raise now or think about for 14 tomorrow morning?
Mr. Maupin?
15 MR. MAUPIN:
I have a matter of major moment.
Did 16 you ever determine whether we could leave these boxes in the 17 room over the weekend?
18 JUDGE KELLEY:
Over the weekend?
That's a good 19 question.
I'd have to ask.
I would guess so.
I'll find 20 out.
21 Anybody else?
22 MR. GOLDBERG:
Yes, Judge Kelley.
We have 23 distributed another photograph of the makeup storage tank 24 level indicator, the makeup storage tank pressure indicator, 25 makeup storage tank temperature indicator, and would ask that L >s ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cos erage 80tk33MM6
I 28039.0 p'I3RT 1183
- V 1
this be identified as Exhibit 9-G, I believe.
And would 2
offer this into the record.
3 JUDGE KELLEY:
Is that the one that looks like.
4 this?
5 MR. GOLDBERG:
Yes.
Yes.
6 JUDGE KELLEY:
Any objection to this offer?
All 7
right.
It becomes 9-G.
8 (Exhibit 9-G identified and received.)
9 MR. MAUPIN:
Actually I have one other trivial 10 matter.
You read a proposed schedule for the rest of this 11 hearing on last Monday that I understood would be applicable 12 in general, beginning next week; is that (G
/
13 JUDGE KELLEY:
Right.
The main difference is you 14 don't have to come on Monday.
15 MR. MAUPIN:
But can we assume that the dates for 16 the lunch hour on Wednesday, Thursday and Friday will be from 17 12:00 to 1:00 as you announced, or 12:30 to 1:30, subject 18 always to some ad hoc change, or will they float?
It will be 19 helpful, having somebody at home we can establish 20 communication with, if we can establish with some certainty 21 when it will be.
22 JUDGE KELLEY:
I think every day we have had lunch 23 from 12:30 to 1:30, except Tuesday.
24 MR. MAUPIN:
12:30 to 1:30, Wednesday to Friday?
25 JUDGE KELLEY:
Does that seem all right?
If g-)
'u J ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336- % 86
l 28039.0 RT 1184 1
there's nothing else, then, we'll adjourn for today.
f 2
(Whereupon, at 3:25 p.m.,
the hearing was 3
adjourned, to reconvene at 8:30 a.m.,
on September 12, 1986.)
4 5
6 7
8 9
10 11 12 0
13 14 15 16 17 18 19 i
20 21 22 23 24 25 ACE-FEDERAL REPORTERS, INC.
i 202-347-37(W)
Nationwide Coserage MXA336-6646
________________________]
CERTIFICATE OF OFFICIAL REPORTER O
This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the matter of:
NAME OF PROCEEDING:
INQUIRY INTO THREE MILE ISLAND UNIT 2 - LEAK RATE DATA FALSIFICATION DOCKET NO.:
LRP
~
PLACE:
BETHESDA, MARYLAND DATE:
THURSDAY, SEPTEMBER 11, 1986 were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission.
(sigt (TY
)
[
JOEL BREITNER Official Reporter ACE-FEDERAL REPORTESS, INC.
Reporter's Affiliation O
.