ML20072J060
| ML20072J060 | |
| Person / Time | |
|---|---|
| Site: | Three Mile Island  |
| Issue date: | 01/08/1982 |
| From: | Keaten R GENERAL PUBLIC UTILITIES CORP. |
| To: | |
| References | |
| TASK-*, TASK-01, TASK-02, TASK-03, TASK-06, TASK-08, TASK-1, TASK-2, TASK-3, TASK-6, TASK-8, TASK-GB NUDOCS 8306290900 | |
| Download: ML20072J060 (137) | |
Text
._ _________ - _____ _ __ _ ________________
ik 488 I
UNITED STATES DISTRICT COURT
(]
- SOUTHERN DISTRICT OF NEW YORK V -------------------------------------------x GENERAL PUBLIC UTILITIES CORPORATION, JERS2Y CENTRAL POWER & LIGHT COMPANY, a METROPOLITAN EDISON COMPANY and PENNSYLVANIA ELECTRIC COMPANY, a Ce Plaintiffs,
-against- :
THE BABCOCK & WILCOX COMPANY and :
J. RAY McDERMOTT & CO., INC.,
a Defendants.
x Continued deposition of ROBERT KEATEN, taken by De fendants , pursuant to adjournment, at the offices of Davis, Polk & Wardwell, Esqs., One Chase Manhattan Plaza, New Yo rk , New York, on January 8, 1982 at 9:50 o' clock a.m.,
before Catherine Cook, a Shorthand Reporter and Notary Public within and for the State of New York.
i l 0306290900 820108 PDR ADOCK 05000 T
DOYLE REPORTING, INC.
CERTIFIED STENOTYPE REPORTERS 369 LExtN@ rON AVENut WALTER SH APIRO, C.S.R. New Yonx. N.Y. too17 CHARLE3 SHAPIRO, C.S.R. TELaPMoNE 212 - 867 6220
489 2 Appe a rance s :
3 KAYE, SCHOLER, FIERMAN, HAYS & HANDLER, ESQS.
Attorneys for Plaintiffs 4 425 Park Avenue New York , New York BY: STEVEN GLASSMAN, ESQ.,
6 .
of Counsel 7
8 DAVIS, POLK & WARDWELL, ESQS.
9 Attorneys for Defendants One Chase Manhattan Plaza 10 New York, New York 11 BY: ROBERT F; WISE, ESQ.,
of Counsel 12 14 ALSO PRESENT:
15 Nina Ruffini 16 17 18 oOo 19 20 l
21 llI 22 ROBE RT KE ATEN, having been previously 23 duly sworn, was examined and continued to 24 testify furthe r as follows: .
l f, ~ ,5 EXAMINATION (continued) l^
490 1 Kont6n 2 BY MR. WISE:
3 Q Mr. Keaten, you are aware that you 4 continue to be unde r oath today?
h 5 A Yes, I a,m .
6 Q Yesterday we were discussing B&W 7 Exhibit 347 which is a copy of your initial draf t of 8 the interim summary report of the Investigative 9 Task Force looking into the accident of TMI-2 on 10 March 28, 1979. Particularly we had been looking 11 at pages 42 and 43 of your report which contained 12 the conclusions section.
13 Turning now to page 43, which is a 14 continuation of the conclusions portion of your 15 initial draft, and looking at the second full 16 sentence on the page which reads, " Operating and l 17 surveillance procedures were incomplete, con tradi cto ry ,
1 18 and at least in one case, in violation of the 19 technical specifications."
20 How did you reach that conclusion as 21 of September 1979?
l 22 A As of September 1979 I helieve that 23 that was my opinion although that opinion later E4 changed at least in one regard.
25 Q What was the one case to which you
l 1 Kocton 491 l
2 referred in your sentence where the procedures were 3 in violation of th e te chnical specifications?
4 A The procedure re ferred to there was (h 5 the procedure which was used to perform required 6 surveillance tests of portions of the emergency 7 feedwater system. At the time that this was 8 written, I think I believed based upon work done 9 by others, not by me personally, th at the way 10 in which that surveillance te s tinct was done was 11 centradictory to the requirements of the technical 12 specifications. I should add that a later and 13 more care ful inve stigation of that contradicts 14 what is said here.
15 Q Did the surveillance procedures with 16 respect to the emergency feedwater system call 17 for the closure of block valves with respect to 18 emergency feedwater during the course of surveillance?
19 MR. GLASSMAN: At what point in time?
20 MR. WISE: The surveillance procedures 21 that were in existrince ar of March 28, 1979.
22 A Yes. The surveillanco procedure called 23 for closure of the block valves as part of the 24 testing. The issue that was raised by the early i O ,5
. investigation and which resulted in this comment
wv o, a =- . p.. _p 1 Konton 492 i
~
2 at this point in time was whether it was permissible 1
3 . under the technical specifications to have block
' ^
4 valves in both trains of emergency feedwater closed
- ijll 5 simultaneously for testing purposes.
i 6 Q Did your investigation find that tha ,
7 surveillance procedure in issue permitted that?
8 A To the best of my recollection, it 9 called for that.
10 Q Did your inves tigation dete rmine whether 11 on the day of the accident both of those block 12 valves were, in fact, closed during the initial 13 eight minutes of the event?
, O 14 A I don't recall that our investigation 15 really pursued that since that had already been 16 determined by previous investigations.
! 17 Q Your task force took it as a given fact j 18 -that that had occurred?
19 A We took it as a given fact that at the 20 start of the transient -- I should be very careful i
i
! 21 on how I exp re s s this. We took it as a given l
22 fact that about eight minutes after the re acto r l
23 trip, those two block valves were found te be 24- closed.
0)
(_ 25 Q Did your task force reach any conclusions e w
1 Konton 493
() 2 as to whethe r the operation of the plant with both 3 block valves for emergency feedwater closed was in 4 accord with proper operating procedures?
jll 5 A I don't know what you mean by the 6 term " proper operating procedures. " .
7 Q Let's remove the word " proper" and 8 simply say, was it in accord with the operating 9 p ro ce dure s of the plant as your task force fo un d 10 them?
11 A The problem I am having with the 12 question is that no one was ever able to determine 13 for sure when and under what' circumstances the 14 valves were closed the last time before they were F
15 found to be closed.
16 Let me ask the question this way, Mr.
Q 17 Keaten: Your task force concluded, did it not, that 18 those blocked valves should not have been closed on 19 the morning of March 28, 19797 l
20 MR. GLASSMAN: I don't know what you 21 mean by "should not have been closed." We 22 'are' talking compared to particular procedures l
23 or what?
l i l 24 Did your task force reach any conclusions Q
I '
25 as to whether or not it w as correct to have those l
1 Konton 494 2
) valves closed at the time that the transient began?
3 A The task fo rce neve r concluded whethe r 4 or not the valves were closed at the time the llh 5 transient began. ,
6 Q Did your task force even look into that 7 issue?
8 A We had one individual who did some 9 investigation into it, but our activities were 10 frankly affected by the fact that we were aware 11 that the NRC's I&E investigation had gone into 12 that in considerable detail and that concurrently 13 with the time frame in which we were doing work, 14 that investigators for the Kemeny Commission were 15 delving into that in great depth, and I personally 16 w as in contact with some of the Kemeny Commission 17 investigators who were purcuing this issue, so as J
18 we did in other areas, we didn't try to duplicate 19 work that looked like it was going on with already 20 thoroughness.
I 21 In the actual pursuit of the question 22 of how the valves came to be c1cs ed ,when they were 23 closed and why, we really ended up relying on the 24 other investigations.
1
[\
\J
' 05 Q So your testimony is that GPU and Met
i . ;
i l 1 Kocton 495 1
2 Ed never made an investigation of its own as to 3 whether or not and how, if so, those valves were 4 closed on the morning of March 28, 1979; is that jll 5 your testimony?
l 6 A o the r th an the actions that we took l 7 in cooperating with the other investigations. For 8 e x amp le , one of the Kemeny Commission investigators 9 wondered whether thero could have been a sneak i
10 circuit that had caused the valves to be closed, 11 and at their request we carried out the investigation 12 of the possible sneak circuit.
13 Similarly, I believe before the task 14 force really initiated its activities, I believe 15 that the Met Ed personnel had cooperated with the 16 NRC IEE Division in carrying out their investigation.
17 Finally, I should recall that I mentioned, i
18 I believe yesterday,that there was an earlier 19 GPU investigation of how those valves came to be 20 closed that predated the task force's activities.
21 Q Who conducted that investigation?
22 A That was Mr. John Miller.
23 Q Did you ever re ceive a written report 24 l from Mr. Miller on that s ubj e ct?
( 25 A Yes, I believe I did.
?
9 1 Konton 496
() 2 Q What did Mr. Miller conclude?
3 A To the best of my recollection, his 4 conclusion was the s ame as that reached by late r lll 5 investigators, which was that it was not possible 6 to determine with complete certainty when and under l l
7 what circumstances the valves were closed.
S Was Mr. Miller able to determine whether Q
9 they were closed at any point during the first I
10 eight minutes of the transient on March 28, 19797 !
11 A I don't believe that there has been 12 any controversy on the fact that at approximately 13 eight minutes into the accident they were found 14 to be closed. The issue was when we re they closed.
15 That is the point of my question. I Q
16 am trying to find out now whether we have some 17 issue in this case as to whether or not there was 18 eme rge ncy feediaevr"to the ste am gene rators during 19 the first eight minutes of the accident. I must 20 say, until this moment I had always thought that 21 that was a settled issue. But you apparently seem 22 to be introducing some doubt as to whether those 23 block valves were, in fact, closed at the beginning 24 of the transient, and if they were closed at the O- 25 beginning of the transient, how long they remained
10 1 Konton 497 2 closed before the accident when it was established 3 the operators opened them.
4 MR. GLASSMAN: They didn't look into h 5 the particular issue. I think that is what 6 he stated.
7 Q Let me ask my next question. Did your 8 task force ever attempt to determine whether or 9 not emergency feedwater had been provided to the 10 steam generators during the firs t eight minutes of 11 the accident?
12 A The task fo rce and its supporting 13 activities ce rtainly did look into' the plant parameters 14 for the first eight minutes of the accident. I am 15 aware of no indication that there was any emergency 16 feedwater flow to the steam gene rators during the 17 first eight minutes.
18 Let me try to explain where the confusion 19 is arising. There has been one. hypothesis submitted 20 that the valves were erroneously and perhaps 21 unknowingly closed by an operator in the very early 22 stages of the transient. While I do not know of 23 any data which supports that hypothesis, I don't 24 1 know also of any data that refutes it.
J 25 Q Who submitted that hypothesis?
- - ._ . _______-_______A
1 Kocton 498 gg 2 A I frankly don't re me mb e r.
U '
3 Q Do you know of any report embodying 4 that hypothesis?
Jg) 5 A I just don't know.
6 Q Do you recall whether that hypothesis 7 postulates any timing with respect to the closure 8
of the valves?
9 A 'I do not recall having heard or read 10 anything that specific.
11 Q Do you know of any analysis that was 4
12 performed with respect to that hypothesis?
13 A I guess I don ' t know what analysis O 14 you would perform.
15 Has anyone asked the operators whether Q
16 or not they took any s teps to close the valves 17 during the early part of the transient?
IO MR. GLASSMAN: As far as Mr. Keaten is 19 aware?
20 MR. WISE: As far as Mr. Keaten is aware.
21 A I just don't re call at this point in 22 time.
- 23 Q Th e n ext s en te nce in your conclusions l
24 section of the draft report readL, "At critical
()
1 25 moments during the accident, important and significant l
I Konton 499 l l
2 p re cautions to procedures were not observed with 3 resultant actions in codflict with the procedural 4 requirements."
h 5 Had you reached that conclusion as of 6 September 28, 19797 7 A To the best of my recollection, that 8 sentence is an example of the type of thing that 9 we talked about yesterday wherein in preparing this 10 first draft of the report, I deliberately put in 11 things that I wasn' t sure whether they were correct 12
, o,r not,; but whe re they encompassed. ideas t h a,t I 13 thought that the report needed to address. As I 14 also stated yesterday, in some cases I have 15 an awfully hard time remembering exactly where 16 my understanding was in September 1979.
17 The b est of my recollection is that that 18 was not a conclusion on my part but merely a point 19 that I felt needed discussion.
20 Did ycu have anything in mind when you Q
21 wrote that?
22 A The issue of the operating procedures 23 that were used and how they were used and the 24 adequacy of the procedures themselves has been O
25 discussed at such great length that I don't know
l l
1 KQCton 500 )
2 how to get back to what I knew in September 1979 on
(
3 that s ub j e c t.
4 Q Pe rh aps it would help if we examined j llh 5 your initial draf t with respect to " 3h e 3mtionale 6 fo r th e Control Room and Staff Pe rs onnel Re spons e " ,
7 which begins at page 13 of your report and continues 8 through to page 26 of your report. I would like 9 to concentrate particularly on item 6, which is a 10 portion of that section and report that begins at
! 11 page 23 and is titled "Us e of Procedure s " The 12 first paragraph reads , " Deficiencies were also ;
i 13 foun4 in the use of procedures. Specific examples i
14 are the PORV tailpipe tempe rature , continued 15 operation with low RCS pressures, the actions 16 associated with turning off the RCS pumps, an d th e 17 attempt to transfer to natural circulation."
18 As of the time you prepared this 19 initial draft, Mr. Ke aten , isn't it a fact that 20 you had completed your initial interviews with the 1
i 21 operating staff?
22 MR. GLASSMAN: Objection. That has i
j 23 been asked and answered several times. We 24 had Mr. Keaten's answer not only.to questions 25 with regard to initial interviews but the l
- - - . .. -- .n --,e, - <- , , -.e-- - - - - - -.----.---r-,..---e- , . -
1 KootOn 501
() 2 possibility of subsequent interviews. We 3 have been through this quite a bit.
4 MR. WISE: Could I have an answer to lll 5 the question?
6 A As I s aid ye ste rday , I couldn't 7 specifically remember which interviews fell where 8 with respect to the issuing of the report, and I 9 haven't looked at anything since then that would 10 re fresh my recollection.
11 Q Mr. Keaten, do you know today whether 12 when you wrote this you had completed any interviews 13 of the operating staff?
14 MR. GLASSMAN: Obviously the answer was 15 yes and we went through this, some before and 16 some after. And Mr. Keaten indicated three 17 or four times in the s ame area. It is a 18 waste of time.
19 MR. WISE: It is not. The colloquy is l
l 20 a was te of time. With respect to this 21 testimony that the witness had completed his 22 initial staff interviews .
23 MR. GLASSMAN: What do you mean by l
24 " completed"? Obviously the witness testified 25 that there were some. We all know there were l
1 Konton 502 2 some and we will stipulate there were some.
( .
3 You are trying to get implications out of a 4 ques tion that really are inappropriate.
lll 5 Q When you wrote this, did you have any 6 basis for it?
7 A when I prepared this firs t draft of the 8 report, as I indicated earlier, the task force had 9 carried out a portion of its investigation including 10 some interviews. I elected to start preparation of 11 the report recognizing that the investigation 12 was not complete, but as I s aid y este rday, feeling.
13 that we had learned enough in some areas to make 14 it profitable to start the process of getting it 15 into a report.
16 (continued on next page) 17 18 19 20 lI 22 23 24 25
k 1 1 Kocton 503 2 Q Is there a procedure with respect to what 3 operators should do when PORV tailpipe temperature 4 is elevated above its normal operating range?
lll 5 A There is a procedure which addresses what 6 the operator should do in the event of leakage through 7 the PORV and that procedure has in it th e information 8 that an elevated tailpipe temperature may be an 9 indication of PORV leakage.
10 Q What is the procedure required to be done 11 in the event that tailpipe temperatures are elevated 12 above the range set forth in the procedure?
13 MR. GLASSMAN: kre we asking the witness' 14 understanding today of a procedure or are we 15 asking the results of a task force view or a 16 preliminary finding at the time of this draft 17 interim report? I am not sure what we are asking 18 for here. Procedure is obviously subject to 19 change.
20 MR. WISE: Mr. Glassman, it's becoming 21 pretty obvious that you are obstructing this. All 22 these questions have been directed at Mr.
23 Keaten's work on the task force. I don't think 24 there is any question about that.
O< 25 MR. GLASSMAN: Why don't you restrict your
i i Kooton 504 2 questions and make them appropriate to the area 3 of inquiry instead of making them so general 4 that one can later misread the record? ,
1 dh 5 I also want to make sure the witness 6 understands the time frame of each question so 7 we don't jump back and forth.
8 BY MR. WISE:
9 Q What did your task force find, if anything, 10 the procedure as it existed on March 28, 1979 required 11 operators do in the event of an elevated temperature 12 in the PORV tailpipe?
13 A The eventual finding of the, task force was
- 14 that elevation of the tailpipe temperature above 130 15 degrees did not in itself warrant any action with 16 respect to the block valve. This paragraph was 17 removed from the final version of the report because 18 it was found to be incorrect.
19 Q Who reached that conclusion, all the members 20 of the task force?
21 A To the best of my knowledge, all of the 22 task force members concurred in all of the findi ngs of 23 the task force.
24 Q Did you actually review the procedure 25 that covers what to dc 'in the event of an elevated
l I 1 Konton 505
() 2 tail pipe temperature?
To the best of my recollection, yes, I 3 A 4 did, h 5 Q Had you reviewed it at the time you wrote 6 this? .
7 A I do not believe that I had, to the best 8 of my. memory. Again recognizing my difficulty in 9 knowing just when I did what.
10 Q You wrote the following: "The (blank) 11 procedure," then you have a paren with a blank and 12 close paren, " requires that the PORV block valve 13 be closed if the indicated ' tailpipe temperature 14 exceeds 130 degrees Fahrenheit."
15 Did you make that up out of thin air 16 when you wrote it?
17 A No. That was based upon information that 18 was given to me by others.
19 Q You wrote after that: "The plant had 20 operated in violation of this requirement for an l
21 extended period prior to the accident."
22 Did you make that up out of thin air?
23 A No. It was again based upon information 24 given to me by others, but again I must remind you O)
\- 25 that in preparing this first draft, I did not strive
1 Kooton 2 for high accuracy but rather to get ideas down on 3
paper that people could use as the basis of preparing 4 a more accurate and clearer report.
dlh 5 Q You wrote following that sentence:
6 "Although testimony has indicated this was a conscious 7 decision by the plant management, the fact that the 8 procedure was not revised, nor was there a special 9 action statement in effect indicating a lack of 10 understanding of the importance of procedures."
11 Did your task force find that the tail 12 pipe temperature procedure for the PCRV had 13 been revised following the time that temperatures 14 were found to be elevated above 130 degrees 15 Fahrenheit?
16 MR. GLASSMAN: Objection. There are some 17 assumptions here as well as to what was found 18 and when such temperatures were found. There 19 are assumptions in this question which are 20 inappropriate.
21 MR. WISE: Could I have the question 22 read back.
23 (Question read.)
24 MR. WISE: My understanding was from the
($)- 23 ear 11er testimony that Mr. xeaten has a1 ready
N : .
x 's l 1 Kontran '
507 o s t 1 C ,,
2 testified that they were found'to N'clevated e ,
above 130 degrees Fahrenheit. My recollection 3 -
11 4 may be poor on that.- I will askLthe 'oundation 1h 5 question first to clear it up.
- s 3
I 6 Q Did your tack force find th$: prl,dr to .
t.
7 March 28, 1979 the temperature in the tai [ pi[e4 ,
8 from the PORV was in fact elevated above 130.de7rees
'i ,,
I g Fahrenheit? , s, ,
10 Q Yes, that was car finding.
11 Q Had'it been so for some period of time 12 prior to March 28, 19797
(~T 13 A My memory was that it had been, although V
14 I don't remember the exact' period of time.
~ '
15 Q Did your task force find that any 16 revision had been made to the procedure following the 17 time that the PORV tail pipe temperature had become 18 elevated over 130 degrees Fahrenheit?
19 A The task force did not find that any 20 revision had been made and it furthermore found 21 that no revision was required to allow the plant to 22 continue to operate.
23 Q Did your task force ever'inve'stigate 3
24 I the symptoms by which an operator would be able to 25 determine whether or not he had a failed open PORV. "i l ,
- \ ,
' ' ~
s i ,,
v' 508 1 Konton
(~h 2 A To some degree we investigated that, yes.
L J' 3 Q To what degree did you investigate it?
4 A Enough to understand that that tail pipe r
h 5 temperature elevated above 130 degrees is not a 6 unique symptom associated with leakage through the 7 PORV but also may be associated with leakage through 8 one of the code safety valves.
9 Q Isn't it a fact, Mr. Keaten, that your 10 investigation found and that you were awara in 11 September of 1978 that one of the symptoms of a failed 12 open PORV is high temperature in the tail pipe?
13 MR. GLASSMAN: Compound question. First 14 you are talking about what the task force 15 found and the task force obviously concluded 16 its work sometime after September 1978. Your 17 question also related to what Mr. Keaten 18 understood as of September 1979, I should say.
19 MR. WISE: That's an easy thing to fix.
20 Q Let's take it first as of September 21 1979. Isn't it a fact that you knew in Septmeber 22 of 1979 that one of the symptoms of a failed open 23 PORV was a high or elevated temperature in the PORV 24 tail pipe?
(h
\' 25 A I think that in September 1979 we
1 Kooton 509 2 understood that if there is leakage through the PORV 3
that that might result in elevated temperatures in 4 the downstream tail pipe.
lll 5 g I am not talking about leakage now, 6 Mr. Keaten. I am talking about how an operator was 7 supposed to determine that he had a failed open 8 PORV. Had you become aware in September of 1979 9 that one of the means by which an operator could 10 determine that fact was from the PORV tail pipe 11 temperature?
12 A I need to*make sure that I understand your 13 question because I think you are asking something 14 different from what you did earlier. The procedure 15 that we have been talking about, which talks about the 16 use of the downstream thermocouples, addresses leakage 17 through the PORV. It does not address a PORV which 18 is failed open. It was and is the understanding of 19 the task force that those thermocouples were not 20 intended to be used for a PORV that had failed open 21 in the sense I believe you are now asking the question 22 and that in fact the operators had not been given 23 guidance how to use them for that purpose.
24 Q Isn't it a fact, Mr. Keaten, that you 25 '
determined during the course of your invostigation
1 l
1 Konton 510
() 2 that the oeprators said that if they had seen the 3 correct temperatures on the tail pipe, they would have ,
4 recognized that they had a failed open PORV?
lll 5 A I recall the discussions that we had 6 with the operating crew in which there was apparently 7 a miscommunication of information regarding the 8 current temperature on the PORV tail pipe at, I 9 believe, the first time that it was called up. I 10 don't specifically recollect a comment that 11 'if I had known it was the other valve, I would have 12 known that the PROV had failed open. I don't recall 13 one way or the other.
14 Q Was that a comment made to you by Mr.
15 zowe?
16 A Which comment?
17 The one you just mentioned about the Q
18 plant operating staff.
19 MR. GLASSMAN: You are talking about the 20 one he recollected?
21 MR. WISE: I am talking about the 22 com=ent that you recollect being made during 23 your discussions with the plant operating staff.
24 A As best as I can recollect, the comment j O 25 that I was referring to was a discussion by Mr. Zewe
I 1 Konton 54J 2 in which hc indicated that he remembered hearing the --
3 I believe the first tail pipe temperature reading that 4 he asked for being transmitted to him was around 230 degrees, and my recollection is that a subsequent llh 5 ,
6 review of the computer chart on which that 7 readout appeared indicated that the actual number 8 was something like 280 degrees. I don't remember 9 whether we discuscad with the operator who called up 10 that temperature, what he remembered it to be.
11 Q That would be Mr. Kunder?
12 A I frankly don't remember right now.
13 Q At the time that you had your discussion 14 with Mr. Zewe and the other members of the plant 15 operating staff, had you personally reviewed their 16 prior testimony given to the various investigating 17 bodies with respect to their actions during the 18 transient?
19 A The best of my recollection is that I had 20 reviewed some testimony that they had given.
21 Q You in fact knew as of the time of your 22 discussion with Mr. Zewe, did you not, that the 1 23 operators had called up the tail pipe temperatures i 24 as part of an effort to determine whether or not
' \_
O) 25 they had a leak or failed open valve at the top of
_ - - - - - - ----w- *- --wa- - - - 9 --
m- m 'V-
1 Konton 542 2 the pressurizer.
3 MR. GLASSMAN: Which operators are we 4 talking about?
lll 5 MR. WISE: The operators on duty on 6 March 28, 1979.
7 MP. GLASSMAN: We are not talking about 8 all of them. You seem to be talking about l 9 particular individuals. I don't know if you 10 are talking in general now or some individuals.
11 Mr. Keaten may answer if he understands the 12 question.
13 A Again it is very difficult for me to 14 recollect just when I learned something.
15 Q Perhaps, Mr. Keaten, it would help if 16 we referred you to B&W 346, which is a transcript 17 of your interview of the operating staff, and in 18 particular I would like you to look at the page 19 which has been marked for purposes of this litigation 20 as 2889.
21 MR. GLASSMAN: Transcript of a particular
'22 interview that was marked in yesterday's 23 sessions is that correct? .
24 MR. WISE i Yes. Marked yesterday as
/' % ",
k 25 B&W 346.
i 'i ' ' -
1 Kocton 513
, 2 Q If you would look at the bottom of the 3
page where there is an indication that you were 1
4 speaking and you say: " Bill, I want to go back to a llh 5 point which has been discussed several times and I 6 just simply want to ask if you know any more than 4
7 what I heard you testify before the ACRS."
8 Does that help refresh your recollection 9 as to whether you were familiar with Mr. Zewe's i
10 testimony and the testimony of other operators before 11 the time that you took this interview?
i 12 A This does refresh my recollection on this 13 specific point. I do now generally recall that I 14 heard Bill say we ma de this point in an ACRS meeting 1
15 which occurred prior to the time this discussion j 16 occurred.
! 17 Q Do you recall anything as to what Mr.
I 18 Zewe had said at that time about why he was looking i
19 at tail pipe temperatures from the PORV?
20 A I believe that he was attempting to use 21 those temperatures as an indication of whether the r
9 PORV was open or leaking, but I would like to point
- 22 l
23 out that that does not contradict what I said earlier.
24 Q I didn't say it did, Mr. Keaten.
25 You didn't write the procedures for TMI-2, f
1 Kooten .54 4 l
2 did you?
3 A No, I did not, 4 Q Did you talk to'those who did?
5 MR. GLASSMAN: Are we talking about ~~
6 O Do you remember the course of your work 7 on the task force?
8 MR. GLASSMAN: Those within GPU. or BAW7 9 Q Did you talk to anybody who was involved 10 in the actual writing of the procedures?
11 A Yes, we talked to people who had been 12 involved in writing or approving procedures, 13 Q Who did you talk to, that you can now 14 recall?
15 A We talked to a variety of different people 16 and I am not sure I can recall all of their names 17 but, for example, we talked to Mr. George Knnder, 18 Mr. Joe Logan, we talked to Mr. Jim Floyd, we talked 19 to Mr. Gary Miller, all of whom had some involvement 20 with the procedures.
I 21 Q Did you talk to anybody from R&W7 ID .
t
-22 A I am not sure. l 1
23 Q Mr. Keaten, do you know whether an alarm 1
24 was provided at TMI-2 with respect to the tail pipe 25 temperature, that is, the tail pipe leading away from-
1 Konton 535 r
2 the PORV?
3 MR. GLASSMAN: Can I have that read back.
4 (Question read.)
l lh 5 MR. GLASSMAN: The question ralates to 6 before the accident, I assume.
7 MR. WISE: Yes. I taks it not an awful 8 lot of operation has gone on at TMI-2 since the i
9 accident. ;
10 MR. GLASSMAN: I just want to keep it 11 straight.
12 A Sitting here today, I am not sure.
2 l
- 13 Q Do you know whether your task force 14 investigated that question?
15 A I am sure that the task force became-16 aware of whether or not there was an alarm, and if 17 there was one, what the alarm setting was.
18 Q Did it consider whatever it had found out 19 in that investigation in reaching its conclusion that 20 the'PORV tail pipe temperr.ture procedure had not 21 been violated?
h i 22 A Yes, it_did. l l
23 Mr. Keaten, did your task force reach
- Q I- .
24 any conclusion as to whether there was any limit which 25 could be tolerated on the tail pipe temperature at I
l
k
- 1 Kontcn 536 2 TMI-2 before the accident?
3 A I don't really understand your question 4 since it is my understanding that the tail pipe is jll 5 capable of accommodating full system temperature.
6 Q What do you mean by full system temperature?
7 A The normal operating temperature of the 8 reactor cooling system.
9 Q So was it the finding of the task force 10 that it made no difference what the tail pipe 11 temperature was?
12 A No, that's going too far. The point 13 about the tail pipe temperature is that it in itself O 14 is not a limit. The usefulness of it as it was c
15 installed, it's my understanding, was to provide 16 assirtance in determining the degree of leakage 17 that might be occurring from the PORV, or alternately 18 from the safety valves. The real limit is en the 19 amount of leakage from the system, which is not l
20 determined from the tail pipe temperature, and the 21 tail pipe temperatura was intended, as I understand, ED !
' 09 as an aid in diagnosing where leakage was occurring, l l l
l 23 if it was occurring.
24 Q Who told you that? l 2S A That's based upon the discussions that i
1 Konton 527 l- 2 we had as a result of the task force investigation, 3 and I don't remember specifically where that came 4 from. It very likely came from more than one source.
I 5 Q Did anyone at B&W tell you that?
6 A I don't recall.
7 Q Did you discuss' with anybody at B&W 8 whether the tail tipe temperature was supposed to do 9 more than simply relate to how much leakage from the 10 uode safeties and pressurizer -- and PORV was 11 occurring?
12 A I don't recall.
1 13 Q You testified earlier that you understood 14 that Mr. Zewe during his testimony had indicated that 15 he looked at the tail pipe temperature in an effort 16 to determine whether or not he had an opening at the 17 top of the pressurizer. Did you attempt to determine 18 how Mr. Zewe came to take that course of action?
l 19 MR. GLASSMAN: Objection. I believe that l
l 20 Mr. Keaten testified as to an opening or leakage i
l 21 at the top of the pressurizer and did not limit l 22 it in the way that counsel just limited it to.
23 It is a mischaracterization of Mr. Keaten's 94 recollection of Mr. Zewe's testimony.
^
\
l s/ 25 MR. WISE: I didn't catch the difference.
1
1 Kocton 538
() 2 I am sure it's there. I am not sure what it is.
3 Q Whatever Mr. Zewe's understanding was 4 with respect to the tail pip e temperature, did you 1 5 make any effort as part of the task force work to 6 determine how he had gained such an understanding? ,
7 MR. GLASSMAN: Objection as to form.
8 A The only discussion which I recall with 9 respect to that point is I believe that I recall that 10 Mr. Zewe told the task force that it was his 11 understanding that in a previous incident in which 12 the PORV was open, that the tail pipe temperatures 13 ~ had been substantially higher than the values which 14 occurred on March 28, 1979.
15 Q Did Mr. Zewe tell you that as a result of 16 that occurrence, he looked to the tail pipe temperatures 17 as a symptom of a failed open PORV7 18 A I believe that Mr. Zewe told us that he l 19 was attempting to use the information regarding the 20 tail pipe temperature as an indication, I believe, 1
21 along with other things, as to whether the PORV was 22 open or leaking or closed.
23 Q What other things did he tell you that 4
24 he was looking at?
O 25 A I frankly don't remember.
, e
1 Kcaten 519 2 Q Isn't it true, Mr. Keatsn, that you were 3 aware in september 1979 that by continuing to operate 4 the plant with tail pipe temperatures in excess of 5 130 degrees, one of the symptoms which the operators 6 during the course of the March 28, 1979 accident 7 looked'to for-an indication of whether the PORV had 8 failed open had thereby been masked?
9 MR. GLASSMAN: Could I have that read 10 back.
11 (Question read.)
12 MR. GLASSMAN: Objection. There is an 13 assumption there that the operators looked to
(
14 a particular item that we have been discussing, 15 the symptom of failed open PORV, and Mr. Keaten 16 has testified that his awareness related to 17 leakage and other items. I think it is a 18 mischaracterization of what has been said.
19 MR. WISE: I think the foundation is 20 adequate and I will take my chances on the 21 question.
22 MR. GLASSMAN: The objection remains.
23 Objection as to form. Mr. Keaten can answer i
24 if he understands it.
g-s
(/ 95 A I am not completely sure I do understand
a- * - - . ,
-V- -
'l Koaten 520 1
2 It. Again I have great difficulty with your tying 3 your question specifically to September 1979 because 4 I believe that much and maybe essentially all of the
$h 5 discussion of confusion in symptoms in fact occurred 6 later. To the best of my recollection, my own 7 knowledge in this particular area of the tail pipe 8 temperatures and the significance of them and the 9 role that they may or may not have played was very 10 fuzzy in September of 197S.
i 11 Q Can you testify, as a result of all the 12 task force work, whether it in fact is true that the e
13 operators during the couras of the March 28, 1979 14 transient did look to the PORV tail pipe temperature 15 in an effort to determine whether the PORV had failed 16 open?
I7 MR. GLASSMAN: Objection as to whether ;
18 Mr. Keaten can testify as to what was or was not 19 true. That of course is one of the things we
. 20 are trying to find out here. Mr. Keaten can 21 of course testify as to the conclusions reached 22 by his task force and the information on which I
- 23 it was based. But he is not the ultimate 24 arbiter of truth, as we all know.
95
- MR.. WISE: I agree. You are posing a
1 Konton 523 2 problem. You won't lot me ask questions about
(
3 conclusions unless I establish the underlying 4 facts, and then you object when I try to 5 establish the underlying facts. The result of 6 this is we will have to suspend this deposition 7 until all the facts have been completsd and 8 all the operators have been deposed and call 9 Mr. Keaten back and go t,hrough it at that time.
10 I am trying to focus my questions at thi s time 11 on the conclusions of the task force 12 with the facts that they had available to them 13 at the tiue. Obviously the ultimate 14 determination of fact will be for the trier of
! 15 the facts in this case.
16 MR. GLASSMAN: In that context I have no 17 objection to the question.
18 MR. WISE: May we have the question read 19 back.
20 (Question read.)
21 MR. WISE: I will amend my question in ggg 22 an effort to take care of Mr. Glassman's f l
23 , objection to the point of saying whether l .
24 the task force reached that conclusion.
- 25 A To the best of my recollection, on the
l 1 Konten 522 2 basis of the information which it gained during its 3 investigation, the task force concluded that one --
4 at least one of the operators in the control room I 5 did request information on the reading from the, tail 6 pipe temperature detectors and that he wanted this 7 information in conjunction with an attempt to determine 8 whether the PORV was fully closed. As I mentioned 9 earlier, the task force also concluded on the basis 10 of the information which it was given that to the 11 knowledge of the people that the task force dealt with, 12 that had not been the intent of the installation of ;
13 thqsa thermocouples, and further, that the operators 14 had not been given what the task force would regard 15 or did regard as adequate guidance on how to use 16 them for that purpose.
j 17 Q Mr. Keaten, the task force found, did it 18 not, that there was a procedure that mentioned the 19 temperature 130 degrees Fahrenheit for the tail pipe?
20 A Yes, it did.
21 Q It found that that temperature had been 22 exceeded for some period of time before the accident?
23 MR. G LAS SMAN : Are we talking about the 24 raw figure of 130 --
25 MR. WISE: The 130 degree figure mentioned l
l 1 Konton 523 2 in the proceduro.
3 MR. GLASSMAN: Just so the record is 4 clear, I object to the form of the question h 5 because it assumes that somehow there was 6 a violation of a procedure. Mr. Keaten has 7 testified at length here today regarding the 8 application of that procedure. I don't think 9 the question is therefore appropriate.
10 MR. WISE: Mr. Glassman, obviously one 11 of the questions in this litigation is whether 12 , or not that procedure was violated. Mr. Keaten's 13 task force reached one conclusion, others 14 may reach another conclusion. Whether one 15 chooses to characterize what happened as a 16 violation or not is a question that will have l
l 17 to be decided later. I am simply asking now l
18 whether that temperature of 130 degrees set l .
19 forth in the procedure was exceeded for some 20 period of time before the accident.
21 MR. GLASSMAN: I think he has already 22 testified that the figure of 130 degrees was 23 exceeded. He has also testified at greater f
24 length as to the relationship of that temperature q
A) 25 to the particular procedure with which you are j'
l
1 Kanton 524 2 concerned. I am not sure what else you are 3 asking here, That's wh.y I an objecting to l 4 the attempt to get a simple answer to a lll 5 confused question, 6 MR. WISE: I thought it was a simple 7 question.
8 Q Can you answer it?
, 9 A The task force came to the understanding 10 that the tail pipe temperature had been higher than 11 130 degrees Fahrenheit for some period or periods of 12 time prior to March 28, 1979.
13 MR. GLASSMAN: I believe that's far earlier.
14 Q Did the task force ever consider whether 15 the fact that the plant was operating with temperatures 16 in excess of 130 degrees Fahrenheit in the tail pipe l 17 leading from the PORV had any effect on the ability 1
18 of the one operator whom you have mentioned to 19 determine by reference to the PORV tai-1 pipe 20 temperatures whether the PORV was fully closed on the 21 morning of March 28, 19797 22 A It is my recollection that the task 23 force did have some discussions which were at least 24 related to that question, yes.
25 Q- Did the task force conclude it was helpful l
I
1 Konton 525 2 to the operator that the temperatures in the PORV 3 tail pipe had been elevated over 130 degrees 4 Fahrenheit?
lll 5 Q I do not recall that the' task force 6 concluded that it was helpful.
7 Q Did the task force conclude that in any 8 way adversely affected the ability of the operator 9 to reach a conclusion as to whether the PORV was 10 fully closed?
11 A I don't believe, to the best of my 12 recollection, that the task force reached a conclusion 13 on that subject. The proilem is that the task force 14 did not have, to the best of my recollection, enough 15 information to determine what the temperature would 16 have been at the time that it was called up, had 17 the initial temperature not been elevated compared 18 to what it was in fact, and the reason that that's an 19 issue is simply that the temperature would be expected 20 to be elevated, I think, substantially as a result of 21 the momentary opening of the PORV which was known to 22 have occurred. And the task force, to my knowledge, 23 never had information that it would allow it to 24 distinguish between that effect and what effect, if O 25 any, there was from the initially elevated temperature.
I
1 Konton 526 2 Q Do you know whether any analysis was done 3 within GPU to determine what temperatures might be 4 expected from a normal lifting and receding of the PORV?
lll 5 MR. GLASSMAN: I think we are looking for 6 an analysis after the accident, part of the 7 task force group --
8 MR. WISE: I think I would like to ask the 9 question whether Mr. Keaten knows whether anybody 10 ever did that at GPO up to today.
11 MR. GLASSMAN: I would like to emphasize 12 that this is precisely the reason that I have 13 made objections to clarify the time frame of 14 the question. We have been dealing with the 15 task force and the question, as I suspect, is 16 dealing with a different time frame.
17 The witness may answer.
18 A I do not recall that I have ever seen 19 such an analysis.
20 Q Did your task force make any investigation 21 concerning the knowledge that the operators had on 22 March 28, 1979 concerning how to detect a failed 23 or malfunctioning PORV?
24 A I don't recall anything in this regard 25 beyond items that we have already discussed.
i 1 Kooten 527 i 2 Q During the course of the task force 3 investigation, did you become aware that there had )
4 been a failed open PORV at TMI-2 on March 29, 19787 h 5 A I became aware of that fact either during 6 the task force investigation or prior to that time.
7 I am not sure which.
O Did your task force make any effort to Q
9 determine what, if anything, had been done as a 10 result of that incident to enable the operators to 11 diagnose such a casualty?
12 MR. GLASSMAN: Objection. The question l 13 is unclear insofar as it relates to "such a
' 14 casualty."
15 MR. WISE: Failed open PORV.
16 MR. GLASSMAN: You are not just referring i 17 to the March 29, 1978 incident?
18 MR. WISE: That's correct.
19 A We discussed earlier, I believe, about i
20 the addition of the signal in the control room which 21 indicated the status of the control signal to the 22 PORV. We have discussed this morning what at least one g f
l 23 operator has stated he thought would be the temperature l
24 on the tail pipe as a result of the PORY being open, O
V 25 and it is, I believe, my recollection that it stems, l
l
I 1 Kcaton 528 2 at least in part frcm the March 1978 event that you 3 referred to. I don't remember that the task force 1
4 dug into the 1978 incident in any greater detail than l l
g 5 that.
6 MR. WISE: Let's take a mid-morning 7 break. -
8 (Recess taken.)
9 (continued on the following page) 10 11 12 13 ,
14 15 16 17 18 19 20 22 -
, e 24 25 O
,e - , - =
.k ' 1 1 Konton 529 2 BY MR. WISE:
3 Q Mr. Keaten, did you ever personally 4 review the emergency procedure in existence on ggg 5 March 28, 1979 for pressurized sys tem f ailure?
6 A Yes, I believe that I did review at 7 least parts of it.
8 Q Let me show you what has, been marked 9 previously in this litigation as Esw 305. Do you 10 recognize that exhibit as the TMI-2 emergency 11 procedure for pressurizer sys tem f ailure? I 12 believe it carries the number emergency procedure 13 2202-1.5.
i O 14 MR. GL AS SM AN : As of what date?
i 15 MR. WISE: I think th e on'e we are 16 looking at there is a composite which 17 indicates that its last revision was done j
18 June 22, 1977.
I 19 I am sorry, Septembe r 29, 1978, 20 revision 3.
l l 21 A I b eli eve that I recognize at raast part S 22 of this.
23 I would like you to turn to the part
, Q
[
! 24 o f the procedure labeled "Section B. Inoperative t
25 l .
Pilo t . Ope rate d (ele ctromatic) Relief Valve (RC-R2)."
l
1 530 2 Koston 2 Do you see that?
3 A Yes.
4 Listed under that section are four items Q
g 5 unde r the heading " Symptoms . "
6 Do you see that?
7 A Yes, I do.
O Q Would you read symptom 3?
9 A Yes.
10 That reads, "RC-R2 discharge line Q
11 temperature is above the 200 degrees Farenheit 12 i alarm. Computer Point (402)."
i 13 Is the RC-R2 discharge line the same 14 thing as we have been referring to earlier as the 15 t ailpipe leading away from the PORV?
16 A I believe that is, yes.
17 Q Did you ever inquire of any of the IO operating staff who were on duty on March 28, 19 1979 whether they determined that this symptom was 20 present during the course of the transient?
21 A I remember discussing the tailpipe 22 temperatures with members of the operating crew 23 who were present.
I 24 MR. -T,LAS S MAN : Is.there a further question?
25 MR. WISE: I didn't think he was finished.
l
,_ w - , - - 'e. m m A.m - m h . - w-t-
m i
3 1 Keaten E3J p 2 Are you finished with your answer?
( Q 3 A Yes, I think so.
4 Q Did the operators,in fact, confirm to
& jll 5 you that this syrptom was present on the date of F
6 the accident?
7 A To the best of my recollection, at 8 least one of the operators indicated that the 9 temperatures were elevated above 200 degrees 10 Fa re nhe i t .
E 11 Q Did you ever ask that ope rator why he
[ 12 did not conclude that he had a failed open PORV as a 1 13 result of the presence of this s ys tem?
p -
- . 14 A I don't recall that that discussion --
5 15 the dis cussion that I am referring to of the tailpipe
- f 16 tempe ra ture s --w as in the context of these procedures 17 at all. We certainly did discuss with the operator 18 what he had done and what he had concluded, but I 19 don't remember ever discussing it in the context -
y 20 of this particular procedure.
=
21 It was part of the charter of : car task Q
i dBi 22 force, w as it not, to look into rationale for 23 control room and s taf f personnel response to plan 24 upset conditions as they did during the first few 1 '
' o5 hours?
E
4 1 Koctan 532 2 A Yes.
3 Q Did you lock at the symptom labeled t
4 "4? which reads, "The RC drain tank pressure and jg) 5 temperature are above normal on the control room 6 radwaste' dispos al control panel 8A."
7 Did your task force determine whether 8 during the course of the transient on March 28, 9 1979 the RC drain tank pre ssure went above normal?
10 A Yes, we did.
11 Q Had it?
12 A At some point in time, yes, the pressure ,
13 had gone above normal. The difficulty is that it 14 is not clear that the reactor operators knew that 15 it had gone above no rmal.
16 Q Was $here an alarm hooked up to the 17 RC drain tank?
i 18 A I don't today remember what alarm 19 configuration associated with that is.
20 Q Did your task force determine during i
- 21 the course of the transient on March 28 the RC l dBi 22 drain tank temperature had gone above normal?
23 A We formed an opinion on that.
l l
l 24 Q What _was the opinion that was fo rme d?
O)
\ 25 A The opinion was similar to the pressure ,
l
5 1 Kocton 533 2 that an opinion of the task force, the temperature 3 had gone above normal, but it was less clear that l
4 4 the operators -- it was less clear how well the l lll 5 operators knew about this . In the particular 6 case of the temperature, the situation was further 7 complicated by the f act that since the PORV was 8 known to have opened, although it was believed to 9 have only opened momentarily, that the tempe rature 10 in the drain tank would have been expected to be
- 11 somewhat elevated.
12 Did your task force dete rmine whether Q
13 the re was an alarm with respect to the drain tank 14 temperature?
15 A I am sure that the task force became 16 aware of the alarm situation. My problem is that 17 I don't remember what we found out.
18 The second symptom listed under this Q
19 section is, "RC System pressure is below 2205 psig 20 and RC-R2 fails to close."
21' Did your task force L dete rmine whether I .
22 during the transient on March 28, 1979, RC System [
23 pressure fell beluw 2205 psig?
24 A Yes, we did.
25 Had it?
Q
! I i
6 1 Konton 534 2 A Yes.
3 Q Do you s ee any other symptoms in this 4 p ro ce dure for a failed open PORV other than the lll 5 three I have just read to you?
. 6 A I can simply look at the mate rial here 7 that is in front of me, but as I indicated earlier, 8 I h ave reviewed only a portion of this procedure, 9 to the best of my recollection, and as a total 10 , procedure, I can't claim that I know everything 11 that is in here.' And also I would like to comment 12 that it is possible to be misled by looking only 13 at one individual procedure and my comment isn't 14 specifically directed to your questions here, but 15 that generally the procedures fit together as a 16 system and one procedure is not the sys tem. Secondly, 17 that there has never intended to be an implication 18 that an untrained individual could take the
- 19. procedures and correctly operate a nuclear power 20 plant. The procedures are only one element that is 21 required.
9 22 Q Did your task force look into the 23 f act whether the operators received training in 24 general?
o5- A Yes.
i
l.
7 1 Koston 535 2 0 Did your task force find out that your 3 operators had received training as to emergency 4 procedures?
g 5 A Yes, some training.
6 Q Is n' t it true that your task force 7 learned that operators were expected to know the 8 emergency procedures?
9 A The operators were expected to know 10 some portions of the emergency procedures.
11 Q Were those portions to include the 12 symptoms and immediate actions?
13 MR. GLASSMAN: Of which procedure?
14 MR. WISE: Of the emergency procedures ,
15 all of them.
16 A I recall b eing aware that the operators 17 were expected to have some knowledge with respect 18 to symptoms and immediate actions, but I don't 19 think that I am in a position to state exactly what 20 that was.
21 Let me clarify that. My earlier comments 9 22 with respect to the procedure system and with
.3 training were not directed at how a trained operator 24 might interpret this', but was directed on how I
)
O ,5 might inte rp ret this or misinte rpret it, in answe r
l l
8 1 Keaton 536 2 to your question.
3 Q Did you eve r, as part of the task force, l
4 work to determine that ope rato rs interpreted the lll 5 symptoms lis te d for a failed open PORV in this 6 p ro ce dure in s ome way other than what appears to 7 be the plain English meaning of these terms? -
8 MR. GLASSMAN: Are we asking for 9 whether the task force learned of the 10 operators' interpretation of the date of 11 the accident or s ome other time, or what?
12 It is a very unclear. question.
13 MR. WISE: At any time.
14 Q Did the tisk force eve r learn th at the 15 operators were interpreting Section B.1 of 16 emergency proce dure. 2202-1.5 in some way other .
17 than the plain English meaning of the words that 18 appear there?
19 MR. GLASSMAN: Objection as to lack 20 of fo un d ation . It is not de te rmine d 21 th at the task force looked into this particular 22 area that you are inquiring into now.
23 MR. WISE: That is what I am asking.
24 Did they ever look into it and if so, did 25 the operators interpret them differently l
. l t 1 Koaton 53 7 l
2 than anyone reading the plain English?
(
3 MR. GLASSMAN: Ob j e c tion also to the 4 implication in counsel's question that lll 5 information contained in a procedure which 6- Ihas obviousliTheen a' subj ect of. investigation 7 and litigation, is crys tal clear to anyone 8 who can read the English language. Obviously 9 we are. de aling with ques tions of procedures 10 and training and a complex situation here 11 whick has not been ' clouded in this fashion.
12 M R '. WISE: We can do it step by step.
13 Q Did you ever learn that the operators 14 interpreted the phrase "RC System pressure is 15 below 220 5 psig" to mean anything other than RC 16 System pressure was below 2205 psig?
17 A The answer to your question is no, we 18 did not learn that they interpreted that particular 19 phrase differently, but I point out that you only 20 read part of a s entence.
21 Q The sentence continues, "RC-R2 fails 22 to close."
23 A Th at 's corre ct.
24 Q In fact, on the day of the accident, O- 25 the RC-R2 failed to close; is that corre ct?
10 1 K3aten 538 2 MR. GLASSMAN: I don't think you let 3 the witness answer his ques tion. I think 4 you should let him finish.
g 5 A The issue in applying symptoms, it 6 would seem to me, is whether the operator knows 7 those symptoms, not whether an after-the-fact 8 investigation found that they existed. As we have 9 discussed earlier, the task force's unde rs tan ding 10 based on the discussions that it had with the 11 operating crew who was on duty at that time was 12 that they did not conclude that RC-R2 had f ailed 13 to close.
14 Q Did you ask them what symptoms they 15 believed existed for a failed open RC-R2, which 16 is the PORV?
17 A We discussed already, I believe, that one 18 operator stated that he believed that a symptom 19 for the PORV sticking would be of much higher 20 tempe rature on the tailpipe temperature than existed.
l 21 The other thing which I recall we 22 discussed with them in that regard was the drain 23 tank parameters that you were asking about earlier.
24 What my recollection is that we were told is that 25 the paramete rs , th e tempe r ature and pressure
11 1 Koston 539 2 spe ci fi cally , were on meters, but not recorders.
(
3 They were on a panel that was not immediately l
4 visible to the operator, and that at the time an lll 5 operator did have an opportunity to go and look at 6 the parameters , the rupture disk on the drain tank 7 had already burst, so the pressure returned to 8 normal ambient pre ssure. The temperatures were 9 somewhat elevated, but it was my understanding 10 based on what I was told that the operators 11 inte rp rete d th at caused by a morentary opening 12 of a PORV which was believed to have occurred.
13 Q You may have misunderstood my question.
14 MR. WISE: Could we have the reporter 15 read it back, please?
16 (Question read) 17 Q You see, Mr. Ke aten , it is not what you 18 discussed because we have gone over that at great 19 length. I am now directing my attention to whether 20 you eve r made .mm investigation to find out what 21 those operators who were on duty the night of the 22 accident thought they would see if they had a f ailed 23 open PORV.
24 A I thought it was implicit in my answer, b
\/ 25 but pe rhaps not.
l l
l
! I Kosten 540 2 Q Mr. Ke aten , you have dis cussed at 3 length the various things that you talked about i 4 with the operators and what their excuses were lll 5 for this, that and other things, 6 MR. GLASSMAN: Objection.
7 Q I am directing my question at this 8 point specifically to the issue as to whether 9 or not your investigation ever dete rmined what 10 the operators thought on the morning of March 28, 11 1979 were the symptoms for a failed open PORV.
12 MR. GLASSMAN: Objection as not even 13 being determined that the task force looked O I4 into a question of whether the operators 15 thought about this,and also I think aghin, 16 so far as th e re has been an investigation of 17 what the task force knew, it has been 18 discussed this morning. I am not sure what 19 else is new here.
20 MR. WISE: We have discussed earlier 21 what the purpos e of the task force was. I S 22 read from Mr. Arnold's charter specifica11y 23 the second paragraph of it. Mr. Keaten 24 testified that he understood that they were lCE) 25 ,,1ook 1,t, the ra,1,na1e ,,, cent,,1 ,o,m l
i
13 1 Kotton 54J 2 response d'uring the course of the events 3 during the first few hours. Anybody familiar 4 with this accident knows that the failed open PORV was a significant contributor to the ll) 5 6 accident. I think it is highly relevant to
. 7 find out what the operators during two hours 8 of this accident missed,the fact that they 9 had a failed open PORV. I am trying to 10 explore in some depth with this witness 11 what his task force which was set up to investigate 12 the rationale for what the ope rato rs did, did 13 MR. GLASSMAN: We haven't prevented w
14 you from going into this area at length. You 15 c an try again.
16 MR. WISE: I want to find out what 17 Mr. Keaten's task force did in terms of 18 determining what the operators on the morning 19 of the accident knew concerning how to 20 diagnose a failed open PORV and what l 21 specific systems they thought they would I
22 see if they had such a casualty.
23 MR. GLASSMAN: My confusion, so it is 24 clear, in terms of your question of what the 25 operators knew on the morning of March 28, 1979,
l
'l I Konton 542 2
("T
\/
and I am not clear whether your question is 3 directed to the task force's conclusion as 4 to what entered their mind on that morning 5 or alte rnatively, the state of their ll) 6 gene ral experience as of that date regardless f
7 of whether or not it came to their mind, 8 that is what I am trying to clear up.
9 MR. WISZ: I don' t know whethe r this 10 clarifies it any, but what I am trying to 11 determine is whether or not the task force 12 looked into the state of mind of the operators 13 on the mo rni'ng o f March 2 8, 1979, based upon O 14 their training, their availability of
, 15 pro ce dure s , their life experiences and whatnot 16 as to what they thought they were going to 17 see in the ways of symptoms in the event 10 l of a failed open PORV.
I 19 A The task force did attempt-tu determine l 20 the rationale for the actions that were taken or 21 not taken by the operators, and I think that the 9 22 best written statement of the findings of the task 23 force with respect to that is given in the final 24 version of the task force report in the appropriate s 25 section.
1 Kocton 543
. 2 with this particular question, I do not l 3 recall that the task force asked the specific 4 ques tion of wh at symptoms would you look for in ggg 5 the event of a failed open PORV, but I believe 6 that that information was implicit in some of the 7 discussions that we had with the operators, ih 8 fact, is implicit in some of the actions that 9 we re taken by the operators in an attempt to determine 10 whether the PORV was open or not.
11 My recollection is that the principal 12 types of things that they would expe ct to see 13 would be some or all of the following symptoms: One, s
14 an indication from the signal light that the PORV 15 was not closed because the control system was not 16 telling it to close. Two, an elevated temperature 17 on the tailpipe which was as high as the operator 18 felt it would be if the PORV was open. Third.would 19 be elevated temperature and pressure in the reactant 20 coolant drain tank, and frankly I can't remember 21 whether there are others that we dis cus s e d.
G 22 Q Do you see anywhere in the emergency 23 procedure for a pressurizer system failure mention 24 of the indicator light?
O
(/ 25 A As I testified earlier, I am' not familiar l
1 Konten 544 ,
2 with all of the details of this procedure. I do 3 not see it in the part.
4 Q As a result of your investigation, did g 5 you become aware as to whether or not that particular 6 symptom was at any time added to the emergency 7 procedure for pressurizer sys tem f ailure?
8 A I don't reeall ever exploring that 9 point.
10 Didn' t your inves tigation dete rmine ,
Q 11 Mr. Keaten, th at , in fact, every single symptom 12 listed in this procedure for a failed open PORV 13 w as , in fact, present on the morning of March 28, O 14 1979 between four a.m. and six a.m. ?
15 MR. GLASSMAN: Objection. The witness 16 has testified, unless I 'miisunde rs tood him, 17 that there was no particular determination
' 10 of these particular symptoms as related to 19 this particular procedure. There were 20 determinations such as were made in a 21 dif f e ren t context. There is an assumption, 9 2b ' h erefore, in the ques tion which is in app rop ri ate .
23 Regardless o f whether your Inves tigating Q
24 Committee ever linked up the fact th at these various l l
25 symptoms in the. eme'rgency procedure which was prepared
l 1
Konton 545 I 2 to cover a f ailed open PORV with the actual parameters 3 as they existed on the morning of the accident, are 4 you now able to state knowing what you know about g 5 the actual parameters as a result of your 6 investigation, and looking at the symptoms as 7 we have gone over them here in Section B of the 8 emergency procedure , that, in fact, every single 9 one of those v.ymptoms was present on the morning 10 of March 28, 1979 between the hours of four a.m.
11 and six a.m.?
12 MR. GLASSMAN: Objection. We have 13
, been over this material s everal times. You
\
14 have already examined the witness with regard 15 to each of the four points s ep a rate ly . He 16 has already testified that th ere w as an 17 investigation made regarding the symptoms and 18 regarding where the operators were aware of the 19 symptoms and stressed that as well, so we 20 have been over this several different times.
21 i
g I object to your trying to go over it again 22 and again.
23 MR. WISE: This will be my last question 24 in this area. If I can get an answer, we will 25 move on to something else.
L 18 1 K00 ton 546 2 MR. GLASSMAN: He can try again if 3 he w ants , but let's get on with it.
4 A My e arlie r comments a re ve ry relevant, ggg 5 and that is that I feel unwilling to give testimony 6 that might in any sens e be regarde'd as expert 7 testimony having to do with how this or any other 8 procedure should have been used by the operators 9 be caus e I have not had the type of training 10 program and the f amiliarity with these procedures 11 that I would need to have in order to be able 12 to answer that in a f air f ashion.
13 Q I am not directing my que stion at w,
14 whether the operators followed this procedure 15 correctly or not. My only question at this point 16 is whether, in fact, e ach of th e symptoms listed 17 in items 2, 3 and 4 under the s ection labeled 18 " Inoperative Pilot Ope rated Relie f Valve" were ,
19 in fact, present on the morning of March 28, 1979 20 between four and six a.m. in the morning.
21 MR. GLASSMAN: Objection. I am not 9 22 sure what you mean by "present." Present 23 in terms of some after-the-fact investigation 24 or present in the sens e that the operators O
\/ I 25 were made aware of them?
l l
f p
1 f Kocton 547
'T 2 MR. WISE: My question has nothing to
%}
3 do with whether the operators were made 4 aware of them.
jgg 5 Q My question is whether or not as a 6 result of the investigation you determined, Mr.
7 Keaten, that, in fact, between four a.m. and six a.m.
8 on March 28, 1979, the following symptoms were 9 present, RC System pressure at'some point du ring 10 the course of that period fell below 2205 psig, 11 that the discharge line tempe rature went above 12 'the 20 0. degree Fahrenheit alarm, - and that the RC 13 drain tank pressure and temperature were above 14 normal? Isn't it a f act that you learned as a 15 result of your investigation that all of those 16 symptoms were present concurrently at some point 17 during the course of the morning of March 28, 18 1979 between four and six a.m.?
19 A The only truthful answer to your 20 question'is "I don't know," and let me explain.
21 You s aid " concurrently," but I don't know that G 22 we have data that indicates that all of those 23 symptoms were present concurrently.
24 Did you ever make any investigation to Q
25 find out whether they were?
1 Keaton 548 2 A As I have stated earlier, we , of course, 3 as part of our investigation tried to make ourselves 4 aware of all of the available data. But let me 5 illustrate one difficulty, and that is the pressure ggg 6 was elevated, we believe, very e arly and only very t
7 early in the transient, and at that point in time 8 I am aware of no data that exists on what the tailpipe 9 temperature was.
10 Q You are speaking of a point in the 11 accident where the pressure elevated and caused 12 the PORV to lift initially?
13 A I am talking about the drain tank 14 pressure.
15 Q Not the RC7 16 A Right. Drain tank pressure. It returned 17 to normal when th e rupture dispe rsed.
18 Did you ever determine af ter the Q
19 rupture dispersed with respect to other parameters
{
20 of the plant?
l 21 A What other parameters?
O 22 Q That was not a clear way of putting j 23 the que s tion. Is that correct, that after the 24 rupture disk blew on the drain tank, that containment O
(_) 25 pressure began to rise?
-- _ ~-
1 Konton 549 1
2 A At some point in time, yes, I believe 3 there was some elevation in the containment pre ssure .
4 Q Isn't it also true that following the ggg 5 rupture disk blowing, the building temperature began 6 to rise?
7 A . At some point in time, I believe that .
8 there was an elevation in the building temperature.
9 Q Did your task force ever determine 10 whe ther a rise in building tempe rature and pressure 11 was a symptom of a loss of coolant acci de nt?
12 A The task force, I believe , determined 13 that that is one of the events that it might be O
14 a symptom for.
15 Q What other events might it be a symptom 16 for as dete rmined by the task force?
17 A We became aware that it can also be 18 a symptom of a leak or a break in the steam line, 19 on the se condary side .
20 Q Would a leak or break in the steam line 21 result in cooling in the primary system?
l O 22 MR. GLASSMAN: You are asking a l
23 hypothetical question.
24 MR. WISE: He will premise the question 25 that way.
1 Kocton 550 2 Q Did your task force determine or reach 3 any conclusion what ef fect a break in the se condary 4 system or the steam line would have on primary g 5 system temperature?
6 A The task force, I believe , determined 7 that that would depend upon the size of the leak 8 or bruak.
9 Q Is there a size leak or break in the 10 secondary side that would result in a rise in 11 primary system temperature as determined by your 12 task force?
13 MR. WISE: I' don't want to ask a 14 hypothetical.
15 Isn't it true, Mr. Keaten, that as a Q
16 result of the task force. work, it was determined 17 that a steam line break on the secondary side 18 would have resulted in a lowering of tempe rature 19 in the primary sys tem?
20 A I't-Asponds on the size of the break.
21 For a break that is smaller than the amount of O
22 steam that is gene rated by removing the energy 23 from the prinary sys tem, I would not expect to see
! 24 a decrease in temperature.
25 MR. WISE: Could I have th at read back?
l i
l_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . . . . . _ . _ _ _ _ _ _ _ . , . , . _ _ . . . _ - - - , - . . - __ ._,__ . . - - - - - - , _ . - - , - - - - -
1 Konton 551 2 ( Answer read) d 3 Q I don't know what you mean when you 4 say "a break smaller than the amount of steam lll 5 generated." I am-thinking.& bout break size in 6 terms of inches, feet, centime te rs , something of 7 that type rathe r than quantities of steam.
8 A And I was describing break size in 9 te rms of quantities of steam. My point is simply 10 that under normal circums tances there is a certain 11 quantity of ste am that must be removed from the 12 steam generator as it is generated by removing 13 energy from the primary sys tem. This quantity O 14 o f s te am is the steam flow,I should say is maintained 15 by a controller which se rves to maintain pressure 16 in the ste am generator by letting out just the 17 right amount of steam.
18 If there is a steam leak which results 19 in releasing steam through that leak in a quantity .
20 th at is less than would normally be removed through i
21 the action of the control system, then the control 22 system will simply balance the amount that it is 23 releasing to the point where the total flow out of 24 the steam generators is the s ame as before.
O)
(_ 25 Under those circumstances I would not l
_. _ , , _ . . - . _ _ - . _ y _
1 Konton 552 2 expect to see any change in the primary sys tem 3 temperature.
4 Q In the event of a steam line break g 5 of the size you mentioned where you would not 6 _ expect a decrease in primary sys tem tempe rature ,
7 is it true that you would also not expe ct an increase 8 in primary system temperature?
9 A From that effect alone, I believe that 10 is correct. .
11 Q Did your task force determine whether 12 there was an alarm system provided in the containment 13 building with respect to radiation?
14 A The task force became aware that there 15 were some radiation monitors in the containment 16 building.
17 Did the task force make any attempt Q
18 to determine whether they were functioning prope rly 19 on the date of the accident?
20 A I don't recall any particular 21 investigation of specifically whether they were O 22 functioning properly- I 23 Did your task force ever dete rmine Q
24 whether following the blowing of the rupture 25 disk on the reactor coolant drain tank , any i
i
1 Kocton 553 2 containment building radiation alarms were triggered?
3 A The task force was aware of and ;
4 supported the work that was done in attempting-lll 5 to develop a complete sequen ce of events which 6 .would include that kind of information to the 4
7 extent that it is correct and made use of that in i
j 8 its deliberations.
9 My somewhat ro undab out w ay o f 10 answering your question is simply that myself sitting 11 here:today, I am having trouble remembering just 12 what elarms came in when. But the task force i
13 in addition to convers ations with individuals 14 made substantial use of the so-called GPU sequence 15 o f. events which was developed.
16 Q With respect to the task force work i
17 in determining the rationale for control room l
i 18 operator res-ponse- Amring the transient, did you 19 over ask the operators whether they had seen a 20 building radiation alarm..euri'ag the course of I 21 the accidenu for the firs t two hours?
I 22 A At this moment I just don't re c a l,1.
23 Q Mr. Keaten, did the task force ever 24 attempt to dete rmine what s ymp toms , if any, the ,
([) 25 o,e,ato,s.ha ,that.txe ro,tu,ed disk had h , ken s
._a . . , - . - - . . - _ _ . - , _ , _.---.-...-,--,---c.mv, ,-.---e. _...-.c+. ,,rr c * - -- - - _ . - - . . . - . , - - - . , , . . . . . - - .-,iq . - . .
1 Konton 554
/- 2 on the reactor coolant drain tank?
Ng]
3 MR. GLASSMAN: Is the question directed 4 toward what symptoms the ope rators s aw , or j lll 5 what was available on the date of the accident?
6 MR. WISE: What was available to them i
7 on the date of the accident.
8 A The task force, as I indicated earlier, 9 made an attempt to be aware of all of the data
- 10 insofar as we could that was available to the operators ,
! 11 and we discussed w'ith the operators how they had 12 used that data as the basis for actions which 13 they took or did not take. I don't recall any 14 specific analysis of what were all the different 15 symptoms which might have been used to determine 16 whether the rupture disk had broken. And pe rhaps 17 it will simplify later questioning if I remind you 18 that the charter of the task force was not to try 10 to establish blame for th e accident. It was not 20 to try to determine why the ope rators didn '.t 21 do something. It was to try to unde rs tand the 22 rationale for what they did.
23 Did the task force determine whether Q
24 the operators had become aware during the course 25 of the first two hours of the accident of high-building l
, . - - - - n --- , -
- - . . , e-.. - - -
1 Kocton 555 2 tempe rature and pressure?
f']
\/
3 A Yes, I believe the task force had that 4 information.
ggg 5 Q wh at did the ope rators s ay ?
6 A My recollection is that at least one 7 operator was aware of the change in the building 8 parameters.
9 Q Did he tak e any action based on that i 10 in f o rma tion?
11 A My re colle ction is that those parameters 12 in conjunction with other parameters that were 13 available led at least one operator to conclude 14 that there was a s team leak occdrring, and my 15 recolle ction is also that as a re sult of this ,
16 actions were taken to isolate one of the steam 17 generators in an attempt to stop or limit that 18 leak.
19 Which operator was this?
Q 20 A I frankly don't remember.
21 Did the task fo rce attempt to probe Q
O 22 that operator's rationale for doing what he did?
23 A We did discuss the symptoms that he 24 interpreted -- he or they interpreted in this 25 fashion. I do not want to leave the impression on 1
l
1 Konton 556 2 the re cord that it was nece ssarily just one operator.
3 I simply don't remember whether it was one or more.
4 Q Did the task force make any effort to lll 5 determine whether that rationale was correct or not?
6 A I don't understand the use of your .
7 word *? corre ct. "
8 Q Did the task force make any attempt 9 to determine whether the rationale that the operator 10 gave for doing what he did stood up?
11 MR. GLASSMAN: I think the same 12 obj e ction would ho'id true. Either the 13 rationale was -- let me rephrase my 14 objection. Pe rhaps you are seeking whether 15 or not the rationale was one that the 16 operator really believed in. I am not sure.
17 Q Did the task force make any attempt 18 based upon- the dnive rse of .information , the knowledge 19 available to the task force, to gauge whether the l
20 ope rators ' rationale as he gave it was correct?
21 MR. GLASSMAN: I think the witness '
O 22 objection is well fo un de d . I don't know 23 what is correct or incorrect about a rationale.
24 MR. WISE: I w an t to find out whether 25 the task force took whatever rationale the
1 Konton 557
( 2 operators gave, noted it down and s topped 3 its investigation at that point, or whether 4
~
they probed the rationale to determine whether h 5 or not the rationale mado logical sense and 6 w as consistent with the plant parameters as 7 they existed, the training and the procedures 8 and whatever else might be out there.
9 A My ques tion is now, are you asking that 10 as a general question or specifically on this point?
11 Let's tak e it as a gene ral que stion first, Q
12 then we will ask it on the specific point.
13 MR. GLASSMAN: Now I am a bit confused 14 as to what the question is.
15 MR. WISE: I am trying to find out 16 what it was the task force was trying to do.
17 Q Were you simply acting as a reporting
- 18 agency and you went out and asked the operator 19 what was your rationale for doing such and such 20 and they would tell you X,Y,Z, and that would be 21 the and of it, and you would find out what the 22 rationale was and the task force won't go further 23 with it? or was it the function or the mission 24 of the task force to go along with the operators l 25 and determine whether the rationale and judge
I Koston 558 2 against the information that the task force had 3 made sense and was correct?
4 MR. GLASSMAN: Object insofar as it g 5 tries to pigeon the task force into two .
6 possibilities here.
7 Feel free, if there is some other Q
0 possibility to de s crib a i t , Mr. Keaten.
9 MR. WISE: All I am trying to do is 10 figure out what Mr. Keaten saw as the mission 11 and role of the task force with regard to 12 operator rationale.
13 A I think I un de rs tand the question. The O 14 answer is that it wasn't black or white in terms of 15 your two possibilities. Where the operators 16 described a rationale, I think the task force 17 did generally try to understand whether that 18 rationale was consis tent with the ope rators ' training, 10 p ro ce dure s , but not necessarily in every case was 20 that pursued in excruciating detail. It was pursued 21 on a more general basis.
9 22 In the particular case you are asking 23 ab ou t , my recollection is that the operator or 24 ope rato rs , as the case may have been, in te rp re te d 25 a combination of somewhat elevated building temperature
1 Konton 559 f
2 and pre ssure in conjunction with a f alling steam 3 pressure in one steam generator as being an 4 indication of a possible steam leak. Although I g 5 d'on't remember specific discussions on this point, 6 my general recollection is that the task force 7 agreed that those symptoms might be an indication 8 of a steam leak.
9 Q Did the task force ask the operator 10 whether he was aware that the s team generator had 11 earlier been isolated?
12 I probably got that wrong. Did the 13 task fo rce determine whether the operator knew 14 at the time that he was going through his rationale 15 to reach the conclusion that he might have a s te am 16 le ak b reak , that emergency feedwater to that steam 17 gene rator had ba'en. . reduced or -te rminated?
18 A I don't recall.
19 Do you recall whether the task force Q
20 learned that during the cours e of the transient at 21 some point the operators reduced emergency feedwater 22 to one of the steam generators in the belief that 23 they had a tube rupture?
24 A At one point during the transient -- excuse 25 me. I believe at two dif ferent points during the
1 Kaaton 560 I~h 2 transient or sequence of events, mo re accurately, V
3 one of the steam generators was isolated in the 4 belief that there had been at least a tube leak.
5 In fact, later s amples of water taken from the 6 secondary side tend to confirm that.
7 Q Did anyone on the task force attempt e
8 to determine whether the operator who~ concluded that 9 he might have a steam leak break took into account 10 the fact that emergency feedwater to th e s te am 11 generator had been reduced at an earlie r time?
12 MR. GLASSMAN: I note an objection. I 13 don't know if it has been established by this 14 witness that there was a time relationship of 15 the two events that is in the question.
16 Q Were you aware at the time that you 17 questioned the operator concerning his rationale 18 for interpreting the containment building symptoms 19 the way he did that emergency feedwater had been 20 reduced to the steam generators at a previous time?
21 MR. GLAS S MAN : Could I hear it again?
22 (Question read) l 23 MR. GLASSMAN: The question is whether 24 Mr. Ke aten w as aware of this supposed fact
/ 25 at the time that he qisstioned the operator ?
l
1 Koston 561 2 MR. WISE: Ri gh t.
~J 3 A The sequence of events on the secondary 4 ' side is fairly complicated. I am having trouble
$ 5 sitting here now remembe ring the relative timing of 6 some of the things which occurred. I am also 7 having my continual problem of remembering exactly 8 what I knew on one particular day.
9 Q All I am trying to do he re , Mr. Keaten, 10 is to probe and test the extent to which the task 11 force looked into operator rationale for their 12 actions during the morning of March 28, 1979. I 13 recognize that we a re covering a lot of materials 14 and the re were a number of interviews and a lot of 15 testimony that was available to the task force and 16 it is very difficult to separate out one particular 17 interview from another and what you knew at the 18 time of preparation of one draft f rom what you 19 knew at the time of preparation of a later draf t.
20 I do, however, want your overall 21 testimony as best you can give it today as to 22 whether the task force ever probed with the operators 23 the conclusion that they reached at some point during 24 the morning that they had a steam or they may have 25 a steam leak break, and in particular, did you ever
~ . - - - _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ ~ _ -_.. _ _ _ _ _
1 \
1 Kanton 562 l 2 probe with them or discuss with them the effect, 3 if any, of a termination of emergency feedwater to 4 the steam generators at an e arlie,r point in the g 5 transient?
6 A I have a general recollection of 7 discussing with the operators why they felt that
~
8 it was possible that they had a steam line break.
9 But I' honestly don't remember the specific ques tions 10 that we asked them and I don't remember today whether 11 we specifically asked them abo ut earlier changes 12 in eme rgency feedwater flow.
13 MR. WISE: It is a good point to break.
14 for lunch.
15 (Luncheon recess taken at 1:40 p.m.)
16 17 18 19 T
20 21 9 22 23 24 25 I j
l 563 1 1 2 AFTERNOON SESSION 3 2:05 p.m.
4 ROBERT KEATEN, resumed.
ig 5 EXAMINATION (continued) 6 BY MR. WISE:
7 Q Mr. Keaten, did you ever become aware .
I 8 during your work on the task force that a study had l
9 been performed in 1974 by MPR Associates with respect 10 to the expected temperatures in the PORV tail pipe 11 as a result of the valve opening?
12 A I believe that I became aware that some 13 analysis of that type had been performed by MPR.
14 Q Did you ever see their report?
15 A I believe that I did, although I am not 16 certain.
17 Q Did the task force make any investigation 18 to determine whether the results of their report 19 had been incorporated into operating procedures 20 by Met Ed?
21 A I believe the task force did in some 9 22 fashion ask that question.
50 Q What was the answer?
24 A I don't believe that the task force 25 over received any indication that it was incorporated.
i
564 1 Kocton 2 Q Did the task force attempt to determine O- 3 why not?
4 A I frankly don't remember that that question j 5 was asked.
6 Let me go back to B&W 347. Earlier this Q
I 7 morning we had begun our seri*es of questions concerning 8 the PORV procedures and the tail pipe with reference 9 to item 6 in Section B of your initial draft of the 10 interim summary report prepared at the end of 11 September 1979. I would like to refer back to thit.
12 We were on page 23 of tlie initial draft which ha.s s 13 , been marked 1878 for purposes of this litigation in the O 14 copy of the exhibit that you have before you. I would 15 like you now to refer to the third full paragraph 16 on that page which reads: "The LOCA procedure which 17 w a's1 pulled out and reviewed by the operators early in I the operation requires that in the case where HPI' f
IO [ is throttled the RCS' pressure be returned to normal s
20 (2200 psi)." - -
21 Had you determined as of the end of'<
e~ 22 u
September 1979 'that the LOCA procedure available' to 23 the operators on the morning ol March 28, 1979 l s 24 required in cases where HPI was throttled that RCS s '
. gg i
'\j ' 25 pressure be returned to' normal?
~\
g i
%s t ,.
ay\_
g '. 7 '
w
. . \
a :. . .,
y Koston 565 2 MR. GLASSMAN: I think it would be 3 appropriate to either call to the witness's 4 attention, or read into the record, the lll 5 remaining portion of the paragraph.
6 MR. WISE: I may have questions about the 7 remaining portion, and of course on your 8 direct examination and at the conclusion of 9 the examination you may bring out any additional 10 portions you would like to examine about. I 11 would like to concentrate on that first sentence.
12 The witness may read whatever else on that page l 13 that he wishes. I don't mean to restrict him.
14 A Without referring to some other things, 15 I can't be sure, but I believe that the wording of 16 that sentence is misleading. As far as what I had 17 determined as of September 1979, I don't remember where we were in the process of analyzing the 18 19 procedures and how the operators had used or attempted 20 to use those procedures on March the 28th, 1979.
1 l
21 Frankly, I just don't recollect exactly what I meant 22 by that sentence.
23 Q The rest of the paragraph continues:
24 "While this procedure was not really adequate to 25 cover the conditions which existed during the accident,
)
i
1 Konton 566 2 the operator should have recognized that his i
O 3 continued inability to restore RCS pressure to the ;
l 4 required level was an indication that the plant was (Jgg 5 not operating correctly." Let me continue with 6 the last sentence: "This could have led to a 7 reinitiation of HPI flow."
8 Upon what did you base your statement in 4
9 this initial draft that the operator should have 10 recognized that his inability to restore RCS pressure 11 to the required level was an indication that the plant 12 was not operating correctly?
13 A I don't recall that I had any strong basis O 14 for particularly the portion of the sentence that 15 refers to an indication that the plant was not operating 16 correctly.
17 Let me reiterate again that this first 18 rough draft was dictated on my part without any 19 attempt to be very careful or very precise in the 20 wording since I recognized that there was a great l
l 21 deal of work that would be required in order to 22 convert this draft into one which was really clear i
23 and accurate. My recollection is that this paragraph l 24 was clari,fied and made more accurate and precise in
() 25 subsequent versions of the report.
l ., .-
1 Keaton 567 i l
l 2 Q As of the end of September 1979, had you 3 determined that saturation conditions existed in the 4 primary system during the course of the first, say, h 5 hour5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> and 40 minutes of the transient?
6 A Saturation conditions in the primary 7 system as determined by the task force were reached 8 at some point during that time.
9 Q Did the task force determine how soon 10 after the beginning of the transient saturation 11 conditions had been reached, approximately?
12 A Yes, I believe the task force did determine 13 that. I don't remember the exact time. It was in the 14 first few minutes.
15 Q Did those saturation conditions persist 16 from a few minutes into the accident until the time 17 the reactor coolant pumps were chut off approximately 18 an hour and 40 minutes into the event as determined 19 by the task force?
20 A I believe that it's correct that the 21 data from the hot leg RTD coupled with the data on 22 the reactor coolant system would indicate that 23 saturation conditions existed at the location of that 24 detector from the time that it was initially reached 25 in the first few minutes until the reactor coolant
f 1 Konton 568 pumps were turned off.
2 3 Q Did you know as of September 1979 that I i
4 saturation conditions aristing in the primary system i
lll 5 outside the pressurizer were an indication that a 6 pressurized water reactor was not operating correctly?
7 A The phrase "not operating correctly" 8 I think needs some clarification.
9 Q I am referring, by the way, to the 10 terminology that you used in your report, so any 11 interpretation you may have based upon what you wrote l
12 at that time would be helpful.
I 13 A I understand. I won't reiterate that this 14 first rough draft -- the wording is not always 15 carefully selected and not always clear and accurate 16 and sometimes plain wrong. But in answer to your 17 question, in view of what I understood in September 18 of 1979, to the best of my memory,'there are certain 9
19 events which can occur in'a pressurized water reactor 20 for which the normal anticipated response of the l 21 system is to include saturation conditions in the l
22 primary loops. That is not a condition which is i-I I 23 normally associated with normal operation. It can 24 represent the correct response of the system to the -
l O- 25 particular event in question. That's the problem i
i
_ _ _ - . - . . _ . _ - _ _ ~ --
_ . . _ _ _ - . . _ . - _ - . - - - . - . - . . . _ _ - _ - . _ . . __ _ - -. . ... C _ ..__
1 Kooton 569
(} 2 that I have with this phrase " operating correctly."
3 Q Let me ask you this. Given what you knew
. 4 as of September 1979 about the events on the morning lll 5 of March 28, 1979, had you formed any impression at 6 that point as to whether the existence of saturation i 7 conditions in the primary system outside the 8 . pressurizer constituted an indication that the plant 9 was not operating correctly?
10 A Still have the same problem.
11 Q I understood your last answer to be that 12 there were certain circumstances within the realm of 13 all possible hypothetical conditions of the plant
. 14 where saturation conditions might be the normal 15 response of the system. I am now attempting to find 4
16 out whether you believed any of those circumstances 17 in fact existed at TMI-2 on March 28, 1979, as of 18 the time you wrote this.
19 A Yes, one of those conditions did exist.
20 Q which one was that?
21 A Namely, there was a leak in the system.
22 Q I may have been confused by your answer.
23 You do not believe, do you, Mr. Keaten, that the l
24 way the plant was operating at TMI-2 on the morning 25 of March 28, 1979 between four a.m. and six a.m. was l
570 1 Konton 2 a correct operation of the plant, do you?
3 MR. GLASSMAN: Objection. He is not here 4 to tell you his beliefs.
h 5 MR. WISE: I will fix the question.
. 6 Q You didn't believe in September of 1979, 7 did you, Mr. Keaten, that the plant had been operated 8 correctly between four a.m. and six a.m. on March 9 28, 1979, did you?
10 MR. GLASSMAN: I object again, in terms 11 of we are talking about beliefs or disbeliefs.
12 I think we want the understanding of the task 13 force at that point in time. I think Mr.
14 Keaten has already indicated a significant 15 problem with understanding or clarifying what 16 is meant by the word " correct" in these 17 circumstances. The question is very difficult 18 to follow.
19 MR. WISE: So what is the objection?
20 MR. GLASSMAN: I don't think I really 21 understand exactly what it is that you are 22 driving at right now. If the witness understands 23 the question, he can answer it.
We have had a 24 problem with this wording before. I think 25 perhaps he can straighten you out. He can try
1 Kocton 571
() 2 to answer if he understands it.
3 Q Do you understand the question? Do
. 4 you want me to put it to you again?
5 .A Repeating it won't help. I am trying to 6 untangle it a little bit, to the best of my ability, 7 to remember what I meant by this sentence, and I am 8 going to try and answer " operating correctly" in that 9 context. To start with at least, what I believe I 10 was trying to express here, is that the continued 11 existence of a reactor coolant system pressure that 12 was lower than the normal pressure that is maintained 13 following a normal reactor trip could have alerted 14 the operators to the fact that this was not the i
15 usual plant response to a reactor trip.
16 The reason the sentence gives me problems 17 is that the operators, to my understanding and to the 18 understanding of the task force, in fact did recognize 19 that this was not the usual response.to a reactor trip 20 and they were trying to figure out what was wrong i
21 and trying to take corrective action. That's one 22 reason why this paragraph was modified as a result >
23 of the subsequent discussions of the task force, L 24 was that as written here, it is rather misleading.
25 Did your task force'ever attempt to l Q l
t
1 K30tCn l
2 determine whether the operators on March 28, 1979 3 understood the phenomenon as saturation?
4 MR. GLASSMAN: So I understand the question, lll 5 is it directed to whether they determined 6 whether the operators had some understanding?
7 I am not sure that we have -- certainly the -
8 testimony has not established that or that 9 there is some unique understanding of a 10 phenomenon.
11 MR. WISE: This is an opening preliminary 12 question. I suppose w'e can find out what the 13 task force did. It is just an opening question 14 as to whether the~y looked into this issue at all.
15 A I don't believe that the task force, to my 16 recollection, ever specifically asked any of the 17 operators whether they understood what saturation was, 18 or, even more accurately, if they had understood 19 what saturation was on March the 28th. We had, I 20 think, one or more discussions in which the term 21 saturation was used, but I don't think that precise 22 question was asked. Further, I don't zecall that the 23 task force ever went back and looked at examples 24 that were given to see whether the concept of O 25
(_/ saturation was present on the examples.
- - , - - -. .- _ , - . - . . , , _ , . _ . . , ,,, -. - . ..~ ,.
1 Kooten 573 l
2 Q To your knowledge, did anybody on the l 3 task force ever ask any of the operating staff 4 during the conduct of the investigation whether they jgg 5 had recognized that they had saturation in the 6 primary system outside the pressurizer.during the 7 first hour and 40 minutes of the transient?
8 A To the best of my recollection, that 9 question was asked.
10 Q What was the response?
11 A My recollection is that the individual 12 who answered indicated that he did not recognize 13 that saturated conditions existed.
14 Q Which individual answered?
15 A My recollection is that it was Bill Zewe.
16 Q Did you ask the other operators who were 17 on duty as to whether they had recognized saturation 18 in the primary system outside the pressurizer during
- 19. the first hour and 40 minutes of the transient?
20 A I don't recall.
21 Q Did Mr. Zewe indicate to you in his S 22 response that he had any difficulty understanding j 23 the concept of saturation?
24 MR. GLASSMAN: At what point in time?
k) 25 MR. WISE: In March 1979.
a 1 Konten 574 l
l 2 MR. GLASSMAN: On the day of the accident?
3 MR. WISE: On the day of the accident.
4 Q In other words, did he say to you he ll 'S didn't understand as of March 28, 1979 what saturation 6 was, or words to that effect?
7 Q To the best of my recollection, his 8 response did not address his understanding of 9 saturation.
10 Q Did you explore as part of the task force 11 whether the operators had received information 12 concerning prior events at TMI-2 in which saturation
. 13 conditions had been reached in the primary system 14 outside the pressurizer?
15 MR. GLASSMAN: Can I have that read back.
16 (Question read.)
17 MR. GLASSMAN: The question I assume is 18 directed as to.whether the oporators had 19 received such information prior to the accident:
20 is that correct?
21 MR. WISE: Prior to the accident, yes.
22 A I have a general recollection of some 23 discussions with the operators with respect to their 24 knowledge of prior events at TMI-2.
l I do not recall l
f~T s_) 25 questions being put in the terms that'you used in your I
1 Konton 575 2 question. I just don't remember.
3 Q What response did you receive from the 4 operators with respect to their knowledge of prior h 5 operating events at TMI-27 .
6 A The operator or operators with whom we 7 had the discussion, I recall, seemed to be very 8 familiar with the so-called overcooling transient 9 which had occurred during April of 1978.
10 Did you ask whether they were familiar Q
11 with the conclusion reached by GPU Service that j l
12 flashing had occurred in the primary system during the 13 course of that transient?
l 14 A I frankly don't remember exactly what we l
15 asked them.
16 Do you recall whether you asked them if Q
17 they were familiar with the November 7, 1978 transient m
18 at TMI-27 19 A I don't recall.
l 20 Q Do you recall asking them whether they 21 were familiar with the GPU Service conclusion that 22 some flashing had occurred during the course of that 23 transient?
24 A Again, I don't recall.
25
K/1 1 Koston 576 2 Q Did you ever receive any explanation from the 3
operators as to why they believed RCS pressure was as 4
low as it was during the course of the March 28, 1978 transient, and I am speaking about the first hour and llh 5 6 forty minutes of it?
7 MR. GLASSMAN: Are we searching in this g question whether the operator provided information 4
1 9 of what they thought at the time of the transient 10 itself, rather than the afterthoughts?
11 MR. WISE: Yes.
12 A The explanation which I recall is that the 13 operators believed that there were problems with the I
14 Pressurizer heaters which were' making it im'ossible p to 15 restore the pressure to its normal value.
16 Q Did your investigation examine whether the 17 Operators had done anything during the course of the 18 transient to determine whether the heaters were working 19 Properly? -
20 A My recollection was that the operators had 21 attempted to have checks made to determine whether the 22 heaters were in fact operating properly.
23 Q You learned from them, did you not, that 24 after performing those chokes,-it was reported back to 1 25 the operators that the heaters appeared to be working
I 1 Koston 577 2 properly?
3 A My recollection is a little fuzzy on that 4 in that I think it is correct that they led the task lll 5 force to believe that the information they got back is
- 6. that at least some of the heaters were operating --
let 7 me try to phrase it a little more carefully -- that 8 for some of the heater banks there had been no problems 9 detected by the check.
10 I can't remember for sure but I have a vague 11 memory that there was at least one heater bank that 12 there might have been a problem with.
13 Q Did you ask the operators whether they 14 explored any other possibilities that might account 15 for the low RCS pressure that they saw during the 16 first. hour and forty minutes of the transient?
17 A I don't remember a question in those terms.
18 We talked to the operators about the various actions that 19 they were taking, the various things they were looking 20 at to the extent that we could about their thought 21 processes --
to the extent that they could remember it j 22 at that point in time.
23 I don't remember the specific question that 24 you asked.
=
25 Q To the best of your recollection today,-is i
l l
l i
1 Konton 578 2 that the only explanation the task force received from i
O' the operators concerning low RCS pressure, that they 3
4 believed at the time that there might be a problem with lh 5 the pressurizer heaters?
6 MR. GLASSMAN: I object insofar as it 7 implies that there were any explanations given at 8 all. We have been addressing ourselves to a 9 particular question. I don't know if anyone tried l
10 to explain actions in some general sense or not.
11 MR. WISE: I asked earlier whether they 12 asked the question and he said they did. We 13 had one response. I am simply trying to determine O 14 if there is anything else that the operators said 15 at the time.
16 MR. GLASSMAN: I think that is a fair 17 question.
18 A The operators, I think, indicated that they 19 at least explored the possibility that there might be 20 a relief or safety valve open. That is the reason why 21 they called up the temperatures and did other things 22 that we discussed earlier this morning.
23 Q You mean on the primary side?
24 A Yes, on the primary side. They were also, 25 as we discussed this morning, concerned about the
Konten 579 1
possibility of the steam leak. I think the conclusion of the task force was, as best as I can state it, that 3
4 they were in an environment where I believe that they recognized that they were in an unusual situation.
6 I don't think they were focasing just on RCS pressure. I think they were trying to deal with 7
8 all the things that they had facing them, which included 9- a high level in the pressurizer. It included building 10 conditions that we talked about this morning.
t gg I never, to my recollection, heard them offert 12 an alternative explanation as to what they had thought 13 was specifically the reason for the low RCS pressure.
O g4 Q Did you ever ask whether they considered 15 taking action to bring pressure up during the course of 16 the first hour and forty minutes of the transient on 37 March 287 18 A I don't recall asking that particular 19 question because, in fact, they were trying to take 20 action to bring pressure up. They were trying to find 9
~1 faults in the heaters if they existed so they could be 9 22 fixed.
33 I believe I recollect that they indicated 24 _
that they turned on all of the operable heater banks 25 and that is the normal way to raise pressure.
l Koston 580 1
- 'N 2 Q Did you find out from them what happened after they turned the heater banks on and tried to raise 3 ,
4 pressure through that means?
A I d n't know that we ever asked them that lll 5 6
question because the data is available as to what 7
happened.
8 (Recess taken.)
g Q Do you know if anybody at GPU at any time has attempted to evaluate the operator action to 10 11 derermine whether it was correct or incorrect?
12 A There certainly have been evaluations of the operator actions. When you start using words like 13
" correct" or " incorrect" then it .is necessary to put 14 15 it in the context of correct or incorrect about what 16 standard.
17 Q Whether anybody at GPU has attempted to 18 assess the performance of the Met Ed operators in the 19 morning of March 28, 1979, to arrive at a conclusion 20 whether based on their training and experience, and the i 21 procedures that they had available and the information 22 they had available to them during the course of the 23 accident, their actions were correct.
24 MR. GLASSMAN: I think it is the same 25 question. The witness can try again.
l l
l
g Kooten 58J l
l A The task force spent considerable effort with 1 C) Support from others in trying to understand the way in 3
4 which the operators' total training program, the I
procedures, their perception of those procedures and ll) 5 6 their understanding of what was happening to them at 7 the time led to the actions that they did and did not g take on March 28, 1979.
9 If my mem ry serves me correctly, there is 10 a discussion of exactly those points in the final 11 task force report 12 Again, if my memory serves me correctly, I 13 believe the final report concludes that given their 14 training, the procedures, their previous experience and 15 so forth, the decision that they made to throttle back 16 on high pressure injection flow,and circumstances now 17 with today's training and procedures, they would be 18 expected not to throttle back on it, that that decision 19 on their part was in fact understandable.
20 Q I understood you earlier to testify that 21 the task force did not consider the assessment of G 22 blame as part of its charter; is that correct?
23 A That's correct.
24 Q Do you know whether anybody within GPU
) 25 attempted to examine the operator actions on the
1 Kosten 582 2 morning of March 28, 1979 to determine in fact whether 3
the operators had erred, given the training and 4 information that was available to them at the time?
I think that has been asked lll 5 MR. GLASSMAN:
6 and answered. I don't know whether you have some 7 different meaning, the words " lack of error" as 8 opposed to " correct" or " incorrect" action, or 9 whatever, but I think Mr. Keaten has told you 10 what has been determined several times.
11 MR. WISE: I guess my problem is I am not 12 understanding how he can have it both ways. On 13 the one hand, when I ask him questions concerning*
O 14 whether the operators were pressed on one 15 particular point or another during the course of 16 the investigation, I have repeatedly been told 1
17 that it was not the function of the task force 18 to as'sess blame or attempt to determine whether 19 the operators acted correctly or not.
l 20 Then when I asked him whether a conclusion 21 was reached, I have received an explanation that 22 the task force concluded that the operators' 23 actions were, I believe using Mr. Keaten's words, 24 " understandable."
) 25 I am having difficulty in seeing how they
1 Koston 583 2 got from one place to the other without having 3 made an investigation as to whether or not, given 4 all the factors that were available to the ll 5 operators during the accident, they had reacted 6 correctly.
7 So that my last question is directed to 8 whether or not, aside from the task force, which 9 I understand from Mr. Keaten's testimony was not 10 attempting to make a critical evaluation of 11 whether the operators had acted correctly or 12 incorrectly, whether anybody else at GPU, to 13 his knowledge, had made such an analysis or 14 investigation.
15 A It is true that the task force charter was 16 not to attempt to place blame and that the task force 17 in fact did not try to do that. But it is true that the 18 task force attempted to evaluate the factors which led 19 to the operators' actions or inactions on the day of 20 March 28, 1979; the intent not to place blame, but 21 rather to understand where were areas where improvements 22 or changes were either necessary or desirable in order 23 to reduce, as far as possible, the probability of this 24 type of event occurring again. ,
'- 25 To that extent, the task force tried to
1 Kosten 584 1
2 understand what were the factors that led the operators 3
to take actions which are inconsistent with what we 4 today are telling them to do.
So we did attempt to answer the question, lll 5 6 was the problem in the training program or was it in the 7 procedures, or was it that the training program and 8 procedures were all right but for some other reasons 9 the operators did not take the actions that today we 10 would like for them to take. And the conclusion that 11 we came to is the one I referred to earlier, that given 12 what we learned about their training program, given 13 what we learned about their previous experience, their 1
14 knowledge or lack of knowledge of previous experience 15 in general, given the condition of the procedures 16 as they existed on March 28, 1979, that we felt that 17 the decision that was made to terminate HPI flow, 18 which today are instructed not to terminate it,that the i 19 decision on their part was understandable.
20 Q You are talking about the termination of
! 21 the HPI flow during the first eight minutes of the 22 accident?
23 A Yes, I am.
24 (Continued on following page) 25
~ -
1 1 Kooten 585
() 2 Q Mr. Keaten, are you aware that the 3 operators at the time they shut down the reactor 4 coolant pumps or the last of the reactor coolant
! h 5 pumps, an hour and 40 minutes into the event, 6 reinitiated HPI?
l 7 A I am aware that there is some, wha't I 8 guess would be best described as confusion over 9 exactly when HPI flow was reinitiated. The task 10 force discussed this item with the operators. The 11 task force also reviewed what plant data was 12 available that would tend to help the task force 13 understand when HPI flow was reinitiated and at what
(
14 level, and it is my understanding based on that total 15 effort, that one, the data isn't really adequate as 16 to when and exactly at what level HPI flow was 17 initiated and the memories of people involved are 18 not consistent with each other as to exactly what 19 happened.
20 Q Let's explore that for a moment.
21 Mr. Keaten, are you today aware that Met 22 Ed has filed its final LER, that is licensee event 23 report, for the accident on March 28, 19797 t 24 A I don't believe I am aware one way or the 25 other.
l.
- , , , , - g- wy -- 1 y-.-,, c- --.y.-- e. ---.y.- -%
1 Keaton 586 2 Q Did you ever see the product of Mr.
3 Wallace's work on arriving at the complete sequence 4 of events for the March 28, 1979 event?
llI 5 Do I have the wrong person? I thought it 6 was Mr. Wallace who was to work on the sequence of 7 events.
8 A No, somehow we got that one confused and 9 it may have been what I said that has misled you.
10 Mr. Wallace at one time was assigned to work on the 11 total GPU story of the accident, of which the sequence 12 would have been one part, of which the task force 13 work would have been one part. The work on the O 14 sequence of events, as we talked about earlier, I think 15 was done by Energy Incorporated, subcontractors who 16 were stationed at TMI. To the best of my memory, the 17 early portion of that work was done under -- " guidance" i
18 may be too strong a word, but the GPU person who was 19 following and overseeing that work initially I believe 20 was Dr. Long.
l
- 21 Mr. Wallace, of course, wcs aware of it h
22 in the same sense that he was aware of all the i
23 different activities going on.
24- At some point in time, and I don't recall l
() 25 exactly when, I became the sponsor of the' sequence of l
g Keaton 587 2 events work, although, again, it continued to be done 3 actually by the Energy Incorporated personnel.
4 Q Regardless of the shifting responsibilities lll 5 for overseeing this work, did there come sometime 6 when you eventually saw a final sequence of events 7 for the March 28, 1979 incident?
8 A Yes, I did.
9 Q Isn't it a fact that that sequence of 10 events includes a notation indicating that at 5:40 a.m.
11 on the morning of the accident, HPI was reinitiated 12 at or about the same time that the reactor coolant 13 pumps were tripped? I should say the last of the O 14 reactor coolant pumps were tripped.
15 A I don't remember exactly the time that is 16 shown in that sequence of events. I am aware and I 17 was aware at the time that I reviewed that before it 18 went out, that there was considerable uncertainty over l 19 that item.
I I
20 Q Regardless of whether there was 1
I l 21 uncertainty, isn't it true that it was put in?
l 22 A It was put into the sequence of events 23 because we know that at some time high pressure 24 injection flow was reinitiated.
t'h
(_) 25 Q Did anyone on your task force ever ask the A
1 Kosten 588 r^x 2 operators whether they had reinitiated high pressure l
\ '
3 injection flow at or about the time they had 4 terminated the last of the reactor coolant pumps?
lh 5 A I don't know that we asked that question 6 but in fact my memory is that at least one of the 7 operators volunteered that information.
8 Q Which operator?
9 A To the best of my recollection, it was 10 Bill Zewe, I'm not sure.
11 Q Did any operator contradict him?
12 A It was my understanding that others who
-s 13 were in the control room at that time disagreed 14 with the time at which that occurred.
15 Q who else disagreed, or rather, who 16 disagreed?
17 A Standing here right now, I don't remember.
18 Since we are on this point, let me also 19 tell you that the task force has been aware for 20 sometime that there is a problem with Mr. Zewe's 21 recollectior., if it is that he in fact turned on the 22 high pressure injection at the time the reactor 23 coolant pumps were turned off, and that is, although l 24 I recollect the portion of the computer record that i
()
(> 25 would allow us to reconstruct exactly when those 1
1 589 KGaton
~
2 pumps were turned back on was last because of actions 3 that were taken in the design of the computer system I
4 combined, in looking at the inventory in the supply i I 5 of water for the high pressure injection system, the 6 change in inventory during this time interval is such 7 that high pressure injection could not have been 8 turned on and left on at the time that the reactor 9 coolant pumps were turned off. l 10 Q Has there been any determination within l 11 GPU to your knowledge as to wheth.r high pressure -
12 injection could have been turned on at any time 13 following the termination or the shutdown of the 14 last of the reactor coolant pumps, which I believe in 15 the sequence of events shows that it occurred at 16 about 5:40 a.m., and the time that the printer 17 went back on line, if you will, or that the data is 18 available, again which I believe was 6:28 a.m.?
19 A Your memory on that point is better than 20 mine.
21 Q I may be off by a few minutes. I think 22 those are the approximate time periods.
23 .Taking 6:30 a.m. as the limiting time, 24 was there any attempt within GPU to determine whether 25 between 5:40 a.m. and say 6:30 a.m., HPI had been
590 1 Kooton IT reinitiated?
O 2 3 A I have heard some discussions of that 4 point but I c an ' t pinpoint them. It was not h 5 something that the task force explored specifically.
6 In fact, from the standpoint of the charter of 7 the task force, it war outside the scope of that:
/
8 charter.
9 Q Did the tas:t force ever determine when 10 core damage began at TMI-2 on March 28, 1979?
11 A The task force itself did not do such 12 studies.
b(~N 13 Q Did the task force have available to it [
14 any information concerning when core damage began?
15 A The answer is yes, or at least the members of the task force had information available.
16 I think 17 the information that was the most available to the 18 task force was probably the work that was done at the 19 Nuclear Safety Anal}
- tis Center.
20 Q What did that information show?
21 A That --
I don't remember the exact time l
22 interval but sometime shortly after the reactor coolant i
l l 23 pumps were turned off that the --
I believe the most 1
24 po.pular hypothesis is that core damage started at the '
(~'NI
\qj 25 time, j
l .
1 Konton 591 f]
\s 2 Q Have you,ever seen work done at GPU 3 that core damage occurred before the reactor 4 coolant pumps were turned off?
5 A I recall in gene ral terms some work that 6 was done within GPU as to the extent of core damage.
7 But I frankly don't recall any, at least significant 8 effort that was expended within GPU on the subject 9 of exactly when core damage occurred. I think that 10 one or more of the GPU engineers may have in fact 11 participated in some of the work that was done at 12 the Nuclear Safety Analysis Center on that subject
(~ 13 but I c an ' t recall any specific work that was done
%.)}
14 by GPU itself.
15 Q Which GPU engineer was that?
16 A I think there was may be more than one.
17 I am reasonably sure that Dr. John Luoma spent time 18 out there; in fact, I am sure. I am not sure what l 19 he was working on while he was there but it was i
! 20 something related to core damage.
t 21 Q Aside from whatever the Nuclear Safety 22 Analysis Center has done, are you aware of any other 23 work done by GPU or by consultants retained by GPU 24 to determine when core damage began on TMI-2 on CN
\'~/ 25 March 28, 19797
.. p
,y 1 KeatOn l'
sN hy s
()
u,
[' 2 A Not that I recollect right now.
3 Q Do you know whether GPU has ever used its i
4 RETRAN computer code to attempt to-determine when e
?,%s 5 core damage began?
iN 6 A I should explain that. The version of
s
,7 the RETRAN" code which was used to construct a model of s
1 8 TMI, as we have discussed previously, has limitations s
9 which make it unsuitable for that purpose. s
'T 10 -
Q Let me ask you to take a look at wh t' 1
11 has previously been marked as B&W Exhibit 341, which r
12 is a copy of Technical Data Report TMI-102 dated 13 May 18, 1979. '
14 If you would take a look at the section
- ' 15 t s labeled " Conclusions" which runs from page 21 through b 16' page 27 in the report, it is divided into a series s x. q
s 17 of sections, the first representing the time period 18 0 to 8 minutes into the transient, the second the
~
g 4
i 19 period of 8 to 30 minutes into the transient and the 20 final, 30 to 100 minutes into the transient.
ggg 21 Approximately 100 minutes into the 22 , transiynt is 5:40 a.m., has been termed in the
~. L' 23s.'
\ sequence of events which, according to the sequence 1
( '
24 _
of' events that have be'en filed by Met Ed to the NRC, l
(~'\ .C\i 8
\
I g s m x; 1 25 as s 3 the time that the last of the reactor coolant ; ,
'N -
g 3 _ . , . . . _ _ , _ -
. , y y . . _ , , , , _ _ , . _ _ . . _ _ _ , . .
_ =
593 1 Keaten i.
1
( 2 pumps were shut off.
3 w uld y u take a look at the last t
4 conclusion for the. period of 30 to 100 minutes.
5 It is labeled "g. Core Damage" and states, "Since 6 forced flow was available during this period, l
7 decay heat levels continued to decrease, and heat 8 transfer conditions were generally f avorable , it is 9 unlikely that any damage occurred to the core. The 10 period 90 to 100 minutes is of special interest 11 since the core voiding was large. Preliminary 12 COMETH analyses do confirm the conclusion that the
() 13 core was sufficiently cooled to present damage."
14 Do you recall seeing this conclusion 15 in B&W 341 in May of 19797 16 A Yes, I believe that I probably do and this 17 has jogged my memory.
18 Q How has it refreshed your recollection?
19 A There were, to the best of my memory, 20 some steady state calculations performed using the lll 21 COMETH code which were attempting to determine whether
{ 22 the core would be adequately cooled with forced 23 circulation of a mixture of water and steam. These l 24 were not transient analyses in the sense that the i
25 RETRAN code is but were rather the type of analysis l
l
_ _ _ . . _ - - _ , ___. _- _. . . _ , , _ _ . _ - . . _ _ _ _ _ _ . - _ - _ _ , _ . . _ ~
l l
1 K3aten 594 2 that can be described as looking at one moment in 3 time and taking the conditions that existed at that 4 moment in time and looking to see whether the core )
llk 5 cooling was adequate at that moment in time. I don't
. 6 remember over what range of void fractions that the l 7 calculation is extended. In fact, I don't remember l 8 that I ever personally saw the results of these 9 calculations, although I was made aware of them. I
, 10 remember, though, being told that the general results 11 of those calculations were that as long as the flow 12 circulation continued with a homogenous mixture of 13 steam and water, that the core could be adequately 14 cooled even with a relatively high void fraction of 15 steam.
16 Q Who in GPU was doing this work?
17 A My recollection is that they were being 18 performed under the supervision of Mr. Gordon Bond.
19 (Continued on the following page.)
20 21 22 23 24 i 25 4 _ _ - - _ - - -
hd1 1 Konton 595
() 2 Q Who is Mr. Bond?
3 A Mr. Bond was the section manager in charge 4 of the Nuclear Fuels Group.
lhk 5 ,,
Q To your knowledge, was any further analysis 6 done within GPU to determine whether any core damage 7 had occurred before the last of the reactor coolant 8 pumps was tripped, or was this the last thing that you 9 remember seeing on that topic?
10 A Frankly, it's the last thing I remember l
11 seeing. For example, I don't recall that we ever 12 published a report on the COMETH calculations , although 13 my memory might be faulty.
14 Q Aside from the calculations with respect to '
15 the makeup tank volume, did anyone ever contradict what 16 Mr. Zewe had to say about turning on HPI at the time the 17 last of the reactor coolant pumps was shut down?
18 MR. GLASSMAN: I thought that was asked 19 and answered before.
20 MR. WISE: I think the witness indicated 21 that there were others who had problems with what 22 Mr. Zewe said. We had some testimony involving l 23 the volume and the makeup tank. I want to know l 24 if Mr. Keaten heard any of the operators 25 interviewed say anything contradictory on that
l 2 1 Konton 596
/'T 2 issue.
O 3 A I have a recollection that I have been told 4 that others who were present in the control room recall lll 5 a different time at which HPI was reinstituted. I don't 6 recall any details.
7 Let me point out that this wasn't an 8 important element to the task force, so we made no 9 specific point to pursue this at length.
10 Q We will talk about that in a moment.
11 What did the others 9 call as to when HPI 12 was reinstituted?
13 A I'm sorry, I don't remember the details.
O 14 Q Was*it approximately the same time as Mr.
15 zewe remembered or something widely different?
16 A I recall it was sometime later, and I don't 17 remember how much later.
18 Q Do you know whether anyone within GPU has 19 made any analysis to determine whether HPI had been 20 reinitiated at the time the reactor coolant pumps were 21 shut down and left on, core damage would have occurred?
22 A I'm not aware of any such analysis, and 23 sitting here today I can visualize difficulties in 24 performing that analysis.
25 We did attempt to construct an analysis of
3 1 Konton 597
() 2 the system inventory of water as a function of time from 3 the reactor trip until the time that the pumps were 4 turned off, and an attempt to really find out what was I 5 the inventory at the moment in time that the pumps were 6 turned off.
7 My recollection of that work was that the 8 band of uncertainty with regard to inventory was so 9 large by the point in time that the pumps were turned 10 off, that it was very difficult to do anything with it 11 at all.
12 Q Isn't it a' fact that some of the operators 13 confirmed Mr. Zewe's recollection that HPI was
, 14 reinitiated at the same time that the reactor coolant 15 pumps were turned off?
i 16 A I simply don't remember.
17 Q Were you ever made aware that Mr. Foust 18 said that?
19 A Again, I don't remember.
20 Q Did the task force consider it worth [
21 investigating that two out of the four operators that 2 22 were there remembered that event? ./
i 23 MR. GLASSMAN: I should point out.that Mr.
24 Keaten has already testified that the area in O 25 which you are presently inquiring was not something l
l .
. . . - . _ _ _ . _ _ ~ . . . , .
4 1 Kcaton 598
() 2 within the task force charter, so I think the 3 implications contained in the wording of your 4 last question are totally inappropriate.
5 BY MR. WISE:
6 Q Mr. Keaten, wasn ' t the task force interested 7 in finding out what had caused the accident at Three 8 Mile Island 2 on March 28, 1979?
9 MR. GLASSMAN: May I have that read back 10 again?
11 (Question read) 12 MR. GLASSMAN: We are getting into an area i
() 13 of harassment. Everyone knows we are all 14 interested in finding out what caused the 15 accident. It in a silly question.
16 MR. WISE: I understood your question to be 17 that somehow tinis was not within the charter. I 18 am attempting to determine whether or not, in 19 fact, Mr. Keaten, the task force wasn't interested 20 in; pursuing any information that came to light 21 during its investigation which might shed some ggg 22 light on how core damage happened to come to Three 23 Mile Island Unit 2 on March 28, 1979.
24 A As I pointed out previously, the scope of O 25 the task force was to address the seven specific items
5 1 Koston 599 2 which were listed by Mr. Arnold in his establishment of 3 the task force. The issue of exactly when core damage 4 occurred and how the damage progressed with time was not 5 an issue which the task force purgued.
6 Q. Was the task force interested in operator 7 actions after the point where the reactor coolant pumps 8 were shut down?
9 A The task force did include some limited 10 consideration of the actions past that time, but by far, 11 the majority of the task force's efforts relative to 12 operator actions were directed towards those actions up
() 13 to and including the time that the last pump was turned 14 off.
15 Q Did anyone ask Mr. Zewe when he recollected 16 the HPI was reduced or terminated following its 17 reinitiation at.5:40 a.m. as he recalled it?
18 A I frankly don't recall whether there was 19 any discussion on that point, and if so what it was.
20 (Recess) r ggg 21 MR. WISE: Let me have marked as B&W 348 a 22 copy of a transcription of a speech given on 23 November 12, 1979. The report is titled " Analysis l
24 of TMI-2 Sequence of Events and Operator Response."
l 25 There are four names given: Mr. Keaten, Mr.
i vanWitbeck, Mr. Broughton, and Mr. Walsh.
l
6 1 Konton 600 l
l l
() 2 (Copy of a transcription of a speech given 3 on November 12, 1979 under a report titled 4 " Analysis of TMI-2 Sequence of Events and Operator lh 5 Response" was marked B&W Exhibit 348 for 6 identification as of this date.)
7 BY MR. WISE:
8 Q Do you recognize B&W Exhibit 348?
9 A I believe I do.
10 Q Is this a transcription of a speech that 11 i you gave?
12 A I believe that it is.
13 l Q Where did you give the specch?
{v~)
14 A I believe it was given at the 7.morican 15 Nuclear Society meeting in San Francisco, California.
16 g Who prepared the speech?
17 A There was no writte n speech prepared. I 18 gave this talk working from some handwritten notes 19 which I had prepared in the hotel ro'om earlier that day.
20 The view graphs, of course, were prepared in advance.
21 Q I was going to ask you, attached to this 22 transcription are copies of what appear to have been 23 some slides during the course of the speech.
24 Did you, in fact, have slides which you 25 used?
7 1 Konton 601 2 A They were either slides or view graphs, I'm i 3 not sure which.
4 Q Are the attachments which follow page 9 of I 5 the transcription of your speech copies, as best you 1
6 recollect, of the view graphs or slides?
7 A I think, to the best of my memory, they are, 8 although I notice in at least my copy there is none 9 labeled slide No. 7, so there may be a missing one.
10 Q That may simply be a problem with the copying ,
11 oscause I think the original of the microfilm does have 12 slide No. 7.
13 MR. GLASSMAN: Does your copy have Nos. 4
(
14 and 5? Mine doesn't.
15 THE WITNESS: I have 4, 6 and 8.-
16 Q Let's take a look at the one that has been 17 marked as the original. Maybe we can work from that 18 and have better copies prepared at a later time.
19 MR. WISE: It is also missing some, although 20 it has 1, 2, 5, and 7 and 8.
i 21 THE WITNESS: The combination of these two l 22 have all of them.
23 - _Q The one I am most interested in is slide No.
l 24 7 anyway, l
25 Who prepared these slides or view graphs?
l l
8 1 Konton 602 2 A Some of them, I personally -- you mean the 3 material? Some of them, I personally prepared; and some 4 of them were prepared for me.
)
lll 5 Q I mean who selected what would go into the 6 slides as opposed to what wouldn't. .
7 A Generally, I did.
8 Q Would you look at slide No. 77 9 MR. GLASSMAN: So it is clear, we obviously 10 would request a copy of the full --
11 MR. WISE: We will get ons put together.
12 During the copying, a few got left out. Before 4 13 we reshme, we will have one done completely.
14 PY MR. WISE:
15 Q You do have the original c: the exhibit 16 before you, which is one of the copies that happens to 17 have slide No. 7 in it.
l 18 Would you take a look at that TMI-2 19 accident subsequent events? Do you see that?
20 A Yes. I also see the parentheses "2," close 21 parentheses. My memory is that there were 2 slides 22 labeled " Subsequent Events."
23 Q Yes. On the original, there is also a 24 slide No. 6 which contains some additional material 25 concerning subsequent events in the sequence of events.
1 Konten 603 9
/)
V 2 A That's right. And these are chronological 3 in that slide 6 contains earlier events than slide 7.
4 Q Down the left-hand margin, there are some I 5 numbers. I take it those correspond to the elapsed 6 time of the accident sequence? .
7 A That is correct. Time zero is defined as 8 the point in time when the turbine tripped.
9 Q Would you look now at. slide No. 7, at the 10 time shown 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> 40 minutes and 37 seconds.
11 Do you see that?
12 ) A I do.
13 Q The entry there reads: " Operator stopped 14 RC pumps 2-A and 1-A and initiated EPI."
15 Where did you derive the information for that 16 entry?
17 A That entry was put there by me on the basis 18 of what Mr. Zewe had told me.
19 Q Had he told you that when the time nine pumps 20 stopped and HPI was reinitiated was 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> 40 minutes 21 and 37 seconds?
22 A No, I don't recall that he did say that.
23 What I recall him saying, is that he believed that he 24 had initiated or ordered to be initiated high pressure O 25 injection flow at about the time the pumps were turned
Konton 604 10 1 2 off.
3 Q Where did the time 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> 40 minutes and 37 4 seconds come from?
h A That came from the GPU sequence of events.
5 6 Q Do you know what the basis was for the GPU 7 sequence of events specifying that time down to the last 8 second?
9 A I don't specifically remember, but I do 10 recollect that the sequence of events itself does give 11 the basis for the information that is contained therein.
12 Q You mentioned that some work had been done 13 concerning the volume in the makeup tank, specifically 14 with respect to the issue whether HPI had been reinitiated 15 at the time the last of the reactor coolant pumps were 16 shut down.
17 Do you recall who performed that work?
18 A First, let me correct your impression. It 19 was not just the volume of the makeup tank, but also the 20 volume of the borated water storage tank, which is the 21 normal supply of water for the high pressure injection G The work was 22 system when it is operating in that mode.
l l 23 performed -- I believe that work was performed by Mr.
24 Lou L-a-n-e-s-e.
j L '\
l -
25 Q What was Mr. Lanese's position at that time, l l
11 1 Kenton 605 2 that is, when the work was done?
3 A He was a senior engineer in the Control and l
4 Safety Analysis Section.
I 5 Q So that he ultimately reported to you?
6 A Ultitately.
7 Q Who directed that Mr. Lanese undertake that 8 work?
9 A It is my recollection, and I should in all 10 honesty state that although I believe this is correct, 11 it is possible that I am in error that this work was 12 done as part of a more general task directed at
, 13 understanding the inventory in the reactor coolant 14 system, as I indicated earlier.
15 Q I am not sure that is an answer to my 16 question. I was looking for who it was that told Mr.
i 17 Lanese to do whatever he did in this regard.
18 Did you ask him to do it?
19 A I was certainly aware that he was doing it.
1 20 I am not specifically sure just who it was that asked 21 him to do this particular thing. It was part of the 22 general activities that we had under way in that section 23 associated with understanding and analyzing the transient.
24 (continued on next page) 25
1 1 Koaten 606 l
2 Q To whom else did he report besides you?
3 A At that time he also reported to Gary 4 Broughton.
lll 5 Q And Gary Broughton reported to you?
6 A At that time, that's correct.
7 Q was Mr. Lanese's work done, to your j 8 knowledge, before the time that Mr. Zewe indicated 9 that HPI had been reinitiated at the time the last of 10 the reactor coolant pumps had been shut down?
11 MR. GLASSMAN: Reported to whcm?
12 MR. WISE: To the task force.
13 f A The sequencee in trying to understand 14 the inventory was begun before our interview with 15 Mr. Zewe. I frankly don't remember exactly at what 16 point in time that this particular task was done or 17 completed.
18 Q Did you ever see a writeup of Mr. Lanese's 19 work?
20 A I remember seeing some written material 21 that was a result of his work. In fact, what I O 22 specifically remember seeing is a graph that showed his 23 evaluation of the upper bound and the lower bound 24 for the water inventory and reactor coolant system as 25 a function of time. To what extent I saw a more i- _____ _
~
-2 1 K30 ton 607
(^ 2 complete writeup, I just don't remember. ,
N_)} l 3 Q Are you aware of whether his work was ever j 4 embodied in any technical data reports prepared lll 5 within GPU?
6 A I frankly don't remember.
7 Q Do you know how far out in time Mr. Lanese 8 carried his analysis?
9 A I am not surb.
10 Q Did he carry it out past the time that the i
11 reactor coolant pumps were tripped?
12 A That_is the part I am not sure about. I 13 remember steing the analysis out to the poinu in time 11 the pumps were tripped and I don't remember whether 15 I have seen anything past that point.
16 Q I had understood your earlier testimony to 17 be that there were those within GPU who doubted Mr.
18 Zewe's recollection of when HPI was reinitiated, based 19 at least in part on the water levels in the borated
( 20 water storage tank and perhaps the makeup tank as well l-l 21 if that was a relevant source for HPI water.
22 Do I understand you to be saying that you 23 personally have no rec 111ection one way or the other 24 as to whether or not Mr. Lanese's work included the l
('
\_
25 time period following shutdown of the reactor coolant
i 1 Keaton 608 2 pumps?
3 A I may have unintentionally misled you.
l 4 It's my definite recollection that someone looked at ;
the inventory in the tanks which were supplies of lll 5 6 water to che high pressure injection and compared the 7 inventory as it actually existed with what would have 8 existed had high-pressure injection been turned on and g left on following the time that the pumps were turned 10 off. And the date that is availablo is inconsistent 11 with the assumption that the high pressure injection 12 was operating throughout that period at a normal flow rate.
1 O
s/
13 :
14 I have gone one step farther in sayirg .
15 that I think that was Mr. Lanese who did that work.
16 When I was referring to having seen results out to the 17 time of the reactor coolant pumo and uncertain about 18 the time thereafter, I was uncertain with the work 19 that Mr. Lanese did within the reactor coolant system.
20 Q I want to make sure I understand what you 21 are telling me. I want to focus on the work done by 9 22 whomever it was that led to the result inconsistent 23 with Mr. Zewe's recollection.
24 Was that work done by Mr. Lanese?
O
(_) 25 A I am not completely sure, t
4 1 Konten 609
- 2 Q Is there a universe of people within your 3 mind who would have been the ones who did it? I am 4 not asking you to speculate. I am asking whether you l I 5 have a recollection of one or more people who you 6 recall that it was one of that did the work.
7 A I understand your question. The problem 8 that I am having is that I know what.it's my impression 9 is correct, but I am not'sure my impression is 100 10 percent defensible. I don't mind telling you what it is, ,
l 11 My impression was that the work was done I 12 under the supervision of Mr. Broughton by someone in 13 ! his section, and I further am of the opinion that it O 14 was Mr. Lanese, but / also have to recognize the fact 15 as I stand here that I don't remember ever having 16 seen anything in writing, and there is some 17 possibility that the information that I got came from 18 somewhere else.
19 Q Do you recall whether whatever work was 20 d ne by whomever regarding the inventories of the tanks 21 supplying HPI specifically with respect to the time 9 22 period following shutdown of the reactor coolant pumps 23 was performed in response to the operator comments 24 c neerning HPI reinitiation, as opposed to a more I) - 25 9eneral project that was ongoing, separate and I n -
1 l 1 I
1 l
5 1 Konton .610 4 i l l
l r
(')
%.)
2 independent of their comments in that regard?
3 A I do not recall, and I don't know that I 4 ever knew.
lll 5 Q Did you ever consider the possibility 6 that if HPI had been reinitiated at the time the last 7 reactor coolant pump was tripped, core damage might 8 have been prevented?
9 MR. GLASSMAN: By "you," you are referring 10 to the utsk force?
11 MR. WISE: Task force or any time up until 12 today, has the witness ever contemplated or 13 considered whether if HPI had been reinitiated n/
- s. h i
14 at the time the last reactor coolant pump was 15 tripped, core damage might have been avoided?
16 MR. GLASSMAN: We are not here to 17 contemplate what the witness -- wc're here to 18 find out whether there was any analysis done, any 19 facts that he knows of.
20 MR. WISE: I am not asking him to 21 contemplate it today. I am asking whether as a 22 result of what he has learned on the task force, 23 he has given any consideration to that question.
24 A I don't specifically remember one way or the O
\l 25 other whether I might have thought about it at some
h I
5 1 Kanten .6611 2 time. I am not aware that I ever instituted any
! 3 serious effort related to that question.
l 4 Q Are you aware of whether there are any current studies in progress within GPU concerning the llh 5 6 HPI system on a B&W 177 reactor?
7 A I guess I don't understand the question.
1 3
- 8 For example, I have described the ongoing RETRAN 9 analysis program that we have been doing for certain i
l 10 types of transients would consider the HPI system.
t i 11 Q I am referring to a specific study that i
4 i 12 is directed to the HPI system, study of HPI, being i
l 13 performed by the safety analysis unit which reports to .
I i 14 you.
[
i 15 A I guess it is possible that they are doing 16 a study that I am not aware of.
i j 17 Let me tell you that I am aware of, very
! 18 specifically of a study that is going on by that 19 section with regard to the possible desirability of an 20 HPI system for a different type of reactor, but I 21 would have to say I am not familiar with any specific 22 study that they are doing on HPI related to B&W 177
! 23 reactor or other than the on' going types of activities 24 that they are doing in the R,ETRAN analysis and work we O. 25 are doing in conjunction with B&W on small break LOCA's.
l 7 Kooten 612 1
What type of reactor is the hpi study that (a) 2 Q 3 you are aware of about? l 4 A Performed for boiling water reactor.
lh , 5 Q Did Mr. Zewe ever tell the task force why 6 he reinitiated HPI at the time the reactor coolant 7 pumps were turned off?
8 MR. GLASSMAN: It has not been established 9 here that Mr. Eewe did so. All this witness 10 has testified to, as far as I recall, is that 11 Mr. Zewe has said that he thinks he did turn on 12 HPI ht that time. There is an assumption,
( 13 therefore, in your question and the lack of 14 foundation in the question is inappropriate.
15 MR. WISE: We are having trouble ~with 16 phraseology perhaps. The question is very clear.
17 The witness has testified that Mr. Zewe said 18 ,
that he turned it on. The witness has also 19 indicated that someone else has concluded that he 20 is inconsistent with certain other data.
21 MR. GLASSMAN: Off the record.
22 (Discussion of f the record.)
23 A I have to say I don't remember the exact 24 wording that well. He may have said that he 25 recollected that he did. I just don't remember the
1 Konton .613
() 2 exact words he used.
3 Q Mr. Keaten, it appears in your view graph 4 or slide for the November 12 speech you gave, it h 5 appears in the sequence of events. There isn't any 6 doubt in your mind at this point, is there, that Mr.
7 Zewe did in fact say that he reinitiated HPI at the 8 time the last reactor coolant pumps were turned- out?
9 A Let me say that as a result of discussions l 10 that I had in wnich Mr. Zews participated, it was my 11 opinion that Mr. Zewe believed that ha had 12 reinitiated HPI at about that time.
(} 13 Q Given Mr. Zswe's belief, did Mr. Zewe say.
14 why he"had done what he believed he did in 15 reinitiating HPI at the time that the last of the 16 reactor coolant pumps was shut down?
17 A I don't recollect whether he did or not.
18 Q Was the task force interested in his 19 rationale for doing that?
20 MR. GLASSMAN: Objection. We are not 21 looking for some vague interest or lack of I i 22 interest. You can find out if they pursued that
, 23 matter.
l 24 Q Did the task force pursue Mr. Zewe's l
25 rationale for doing what he said he did?
1; i
l l
l
1 K0atOn .614 2 A I do not remember whether there may have 3 been questions or comments at the time that t'ais was 4 discussed with the task force, although to my memory, 5 one occasion it was discussed. My recollection is 6 that outside of what muy have occurred in that 7 conversation, the task force did not pursue it any 8 further.
9 Q Did the task force ask Mr. Zewe what 10 conditions he believed existed at the time the lest 11 of the reactor coolant pumps was shut down and l 12 conditions within the primary system?
% 13 A I frankly don't recall whether we asked (A
\-
14 that specific question or not.
15 Q Did the task force have any information as 16 to what conditions Mr. Zewe, who was the shift 17 supervisor on duty that morning, thought existed at 18 the time those pumps were shut down?
19 MR. GLASSMAN: I would like to have that 20 read back.
21 (Question read) l 22 A The information which I can recall that was l
23 relevant to that question was one earlier statement by 24 Mr. Zewe that in general, that he had not recognized
(~ .
\ 25 that saturation conditions existed in the primary loops. !
l l
1 Konten .615 2 And second, that the vibration in the 3
reactor coolant pumps was outside of the established 4 limits and was increasing. !
h 5 Q Did anybody on the task force ask Mr. Zewe 6 whether at.the time he reinitiated HPI, if his 7 statement is to be believed, he still was unaware of 8 the existence of saturation conditions in the primary 9 loops?
10 A I just don't recall.
11 (Continued on next page) 12 1
13 14 15 16
/
17 l 18 19 20 21 22 23 24 h
- (kl 25 l
O
_ _y, _ _ . _ _ , _ _ , _ _ . _ ___ ._ _ _ _ _ _ , . , _ _ _ _ , _ , _
3/1 Konton 616 l 1
Q Didn't it occur to you, Mr. Keaten, during O 2 your conversation with Mr. Zewe that the reinitiation 3
of HPI might indicate an awareness on his part that 4
he had voiding in the primary system?
h 5 MR. GLASSMAN:
I don't know where we are 6
going with this line of questioning.
7 MR. WISE: The line of questioning is 8
to probe the thoroughness of the task force 9
investigation, what they were told by Mr. Zewe 10 and the other operators.and the degree to which 11 we can rely upon the task force conclusions.
12 MR. G LAS SMAN : I think it is very clear C:) that the length of the questioning and answering 14 today and throughout the whole week, you have 15 been probing that thoroughly. We have had a lot 16 of questions of what Mr. Zewe said to the task 17 force. I have let you probe that fully and we have also found out what the task force did and 19 didn't do.
20 You can keep on asking questions that way if you wish. To try to probe back whether at 22 some point when Mr. Zewe said something, something 23 l
else flew through the mind of Mr. Keaten I think 24
('-)' 25 is totally unproductive.
1 Konton 617 O
N' 2 MR. WISE: It is an important point to us 3 and I went to press it further.
4 MR. GLASSMAN: The objection stands, h 5 Irrelevant, inappropriate. Also object to the 6 form. '
7 You can answer it, Mr. Keaten, if you have 8 an answer. l 9 Q Didn't it occur to you, Mr. Keaten, at 10 the time that Mr. Zewe reinitiated HPI, that he might 11 have recognized that he had substantial voiding in 12 the primary system?
13 A Let me make two or three comments that I would perhaps help to clarify the situation.
14 Let me 15 first say that I frankly don't remember whether that 16 thought was in my mind at that point in time or not.
17 I do, however, recollect that from the very beginning, 18 the first time that Mr. Zewe made that statement that 19 I personally accepted it with reservations because I 20 was aware and in fact Bill Zewe and others had taken 21 special care to point out to the task force that their 22 memory regarding the exact times at which things happened 23 had proven untrustworthy in certain cases where it was 24 possible to cross-check and that in fact they seemed to 25 remember that things happened much more rapidly than in
1 Konton 618 2 fact they did.
3 My recollection is that that statement was 4 made prior to the time that Mr. Zewe made the statement h 5 about having started high pressure' injection at about 6 the time he turned off the reactor coolant pumps. .
7 The second thing that caused me to,let us 8 say, place some uncertainly in my mind on Mr. Zewe's l l
9 statement was that these interviews were held in I 10 believe August or September of '79 and by that time 11 Mr. Zewe and the other members of his crew had given 12 voluminous testimony on this event 'and had been asked 13 many, many questions with respect to high pressure 14 injection and saturation conditions and the primary 15 loPPs and so forth.
16 I recognized, and in fact I think it's 17 within a vague recollection, that they themselves i
i 18 poin'ted out to us that in some cases they might be 19 remembering what they had told others or.that their 20 recollection might have changed with time and so on_a 21 point like this in the vernacular, I personally took 22 it with a grain of salt.
I 23 I elected to put that item on this view-l 24 graph and I believe I am alos personally responsible A 25 for the fact that it's in the GPU sequence of events l
l l
1 Konton 619 2 because I did not want to directly contradict what Bill 3
zewe said that he believed he had done.
4 Frankly, at the time that I did both of lll 5 those actions, I did so with the opinion that it was 6 not a particularly important point. I recognize that 7 others may have a different view on that. I recognize 8 that now, at least.
9 Q Mr. Keaten, did you generally take 10 Mr. zewe's comments as to what he had done during the 11 morning of March 28, 1979 with a grain of salt?
12 A As I indicated earlier, his comments with gT 13 respect to timing, yes, I did, and in fact I believe V
14 he encouraged us to do that.
15 Q There has been a great deal of attention 16 in this case given to the operators' state of mind and 17 the reasons for what they did during the first few
'18 minutes of the accident.
19 Did you rely upon their statements as to 20 what they did during those first few minutes or did you 21 take those with a grain of salt as well?
22 A I want to emphasize that the grain-of-salt 23 comment on my part was directed towards statements with 24 respect to timing and based upon not only comments made 25 by the operators'themnelves that they were surprised l
\ ~
'i Konton 620
- s. ,
\
I ~\
s 2
when they,found out the actual times that things .
.
- occurred as indicated by the data as compared to their 3
memory of when or how fast things occArred', but also 4
statements that were made to me by others who appeared -
lh 5 6 to be knowledgeable in the field in the way_that 7
memory works with respect to timing of events._when a s'
8 person is under stress.
9 Q Were you as a member of the task force ,
10 interected during the course of the interview of
- . 11 Mr. Zewe in his understanding of thh primary system
~
12 conditions at the time the last of the re'.3ctor coolant s
13 pumps was tripped? '
1 >
14 MR. GLASSMAN: Objection to any implication
\
15 in the word " interested" as we hav'e used it.
~~'
16 , MR. WISE: We are curious to find out what
\,7 '
, 17 Mr. Zewe thought'when the last of the reactor 4 18 coolant, pumps was turned off.
19 MR.. GLASSMAN: I don't know where we are 20 leading. There was work to be done. It is 21 appropriate to find out what }Oc. Keaten did or
.g asked. I don't know where our curiosity gets us 22
, 1 i 23 here.
1 >
24 MR. WISE: I think it is-fairly obvious p if you read your complaint. The turning off of
.\ '25 ,
g
\ l 1
- l s - . 0.'
l l
1 Konton 621 2 the reactor coolant pumps was a direct cause of 3
the accident. It is extremely relevant and 4 obviously relevant to determine to what extent this witness, who headed up the task force, h 5 6 looked into the question of what the operators' l l
7 understanding of the conditions was at the time
, 8 they took that course of action.
9 MR. GLASSMAN: I can't imagine anyone to 10 allow you to go into more fully to find out the 11 answer to these questions. I just don't know that 12 whatever curiosity that he had makes any 13 difference.
14 The question is can we get to any facts as 15 to what happened?
16 Q Was that a matter of interest to you during 17 your investigation?
18 A The task force felt that it was part of 19 its' charter to understand what the operators were trying 20 to accomplish and how they were trying to accomplish 21 and what was the basis for this approach on their part.
22 My recollection is that Mr. Zewe told the 23 task force, and my recollection is also that he has 24 told others, that in turning off the reactor or reactor 4 .
25 coolant pumps that he was attempting to establish 4
1 1
-,,--ir- - - rc--- , -, ., y e-e---w-et,--,-+--r .e r - - - - - ,,r,r-e---,r-, - - -
- l 1 Kooton 622 2 natural circulation in the primary system.
3 Although I don't remember a specific questior, 4 on this matter, it was certainly my interpretation and lll 5 I believe the interpretation of the entire task force 6 that as part of that approach, that Mr. Zewe took those 7 steps under the assumption that the system was full of 8 water at least to the extent necessary to allow g natural circulation to be a heat removal mode.
10 Q Did it impress you as inconsistant with 11 that belief to reinitiate HPI at the same time?
12 A As I said earlier, I just don't recollect
~
13 the specific questions that we asked with respect to 14 the HPI initiation.
15 Q Did you as part of your task force work 16 review the relevant procedures with respect to 17 establishing natural circulation?
18 A We reviewed procedures with respect to 19 establishing natural circulation, t
20 Q Did you find that in any of those procedures 21 there was a requirement that 35 degrees subcooling be h
22 established before attempting to go to natural l
23 circulation?
l .
24 A My recollection is that in at least one i O' 25 procedure dealing with the transition to natural
1 Kooten 623 1
2 circulation that there was a requirement associated
) ,
with some degree of subcooling. I don't remember the 3
j 4 number.
lll 5 Q Did anyone ask Mr. Zews whether he had 6 confirmed at the time that he had attempted to go to 7
natural circulation that the primary system had 35 j
8 degrees subcooling?
3 9 A There was more than one procedure at l 10 TMI-2 at the time of the accident which contained i
11 instructions on the transition to natural circulation.
1 12 The requirements in the different procedures were not i -
) 13 all the same.
j 14 My recollection is that the task force was 15 informed that the procedure which the operators pulled 16 out and used in attempting to make the transition to 17 natural circulation did not include the requirement I 18 relative to subcooling.
i 19 Q Which of the procedures do you recall they 20 pulled out and used? Was it the one for a blackout?
21 A At this late date, my memory is fuzzy on 22 that. Except that I remember the specifics of this i 23 were discussed in the final task force report.
l 24 Q Did the task force ever determine why 25 Mr. Zewe chose the particular procedure that he did
1 Kocton 624 with regard to attempting to go to natural circulation?
O 2 A I d n't recall.
3 4 Q Regardless of which procedure happened to jll 5 be pulled out during the course of the transient on 6 March 28, 1979, did the task force inquire of Mr. Zewe 7 whether he was generally familiar with the procedures 8 for going to natural circulation?
9 A I don't recall that the task force ever asked 10 that particular question because I think the task force 11 assumed that Mr. Zewe would be generally familiar 12 with the procedures for natural circulation but that, 13 of course, is not the same as remembering every step O. 14 or precaution.
15 Q Are you familiar with something called 16 fuel pin compression ratio?
17 A I am generally familiar with that term, 18 yes.
19 Q Did anyone ever discuss with Mr. Zewe or 20 any of the other operators their kno1 wedge of the fuel 21 pin compression ratio?
1 1 16 I don't recall that I participated in such 22 A l
23 discussions.
24 Q Did the task ~ force ever attempt to determine 25 whether reference to the fuel pin compression ratio would
1 Kecton 625
' 2 have provided useful information to the operators
- .h during the course of the transient concerning the 3
existence of saturated conditions in the primary 4
system, gg, 3 A I don't 6(' call.
6 MR. WISE: It is 4:45.
7 g (Time noted: 4:45 p.m.)
9 10 11 Robert W. Keaten 12 Subscribed and sworn to before me 13 this day of 1982.
14 15 16 17 18 19 20
, 21 S 22 23 24 25
i i
11 1 626
,A Q CERTIFICATE 2 ,;
STATE OF NEW YORK )
3 .! : ss.:
!: COUNTY OF NEW YORK )
.; 4 I, CATHERINE COOK a Notary
!g ,
Public of the State of New York, do hereby l 6 ,
l certify that the continued deposition of a
ROBERT W. KEATEN was taken before me on Friday, January 8, 1982 consisting of pages 488 through 625 ;
I further certify that the witness had been previously sworn and that the within t
O transcript is a true record of said testimony; That I am not connected by blood or marriage with any of the said parties nor l 1 i interested directly or indirectly in the matter f in controversy, nor an I in the employ of any of the counsel.
18 i.
IN WITNESS WHEREOF, I have hereunto set my i 10 20 fg handthis) day of January , 1982. 4 21 i
k OILLVLD, Y i 23 4
- CATHERINE COOK 2.u :
i k I. !
.. - . . . . .