ML19242D898
| ML19242D898 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 05/17/1979 |
| From: | Van Witbeck T ENERGY, INC. |
| To: | |
| References | |
| NUDOCS 7908280787 | |
| Download: ML19242D898 (16) | |
Text
l
'7 ~~~ j UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION l
1!
In the Matter of:
2 IE TMI INVESTIGATION INTERVIEW i
31 of Thomas Van' Witbeck l
Manager of Operating Plant Services Energy, Inc.
Si 6l l
7l t
8!
l Trailer #203 91 NRC Investigation Site TMI Nuclear Power Plant 10l Middletown, Pennsylvania 11l i
Mav 17,1979 12!
(
(Date of Interview) 131 July 9. 1979 (Date Transcript Typed) 14I 207 15l (Tape Numcer(s))
16!
17 18j 19i 20!
21l l
NRC PERSONNEL:
22' Bob Marsh l
Dorwin R. Hunter 23l William H. Behrle (Met Ed)
Tim Martin 24[
25:
790722 0777 f
I f!
b L) 2Y U; }
.~'t
,l j
f a
l
[
I MARSH:
Today's date is May 17, 1979.
The time is 1:18 p.m.,
and we art 11 2
located in Trailer 203 at the TMI site.
My name is Bob Marsh.
I am an 3{
investigator with the U.S. Nuclear Regulatory Commission, assigned to the 4
Region III office in Chicago, Illinois.
We are at the TMI site today to 5
conduct an interview on Mr. Thomas Van Witbeck, and at this time I would like the other people in the room to introduce themselves, spell their last Gj 7l name, and indicate their employment position.
You can start now.
8 HUNTER:
My name is Dorwin R. Hunter.
I am an inspector specialist, Perfor-g marice Appraisal Branch.
10 11l l
VANWITBECK:
My name is Tom Van Witbeck.
I am a manager of operating plant 12!
services with Energy, Inc.
13l t
14i BEHRLE:
My name is William H. Behrle and I am a project engineer with 15' Metropolitan Edison Company.
17!
l MARTIN:
My name is Tim Martin.
I am a reactor inspection specialist with 18!
the Performance Appraisal Branch.
19!
20!
l HUNTER:
Mr. Van Witbeck before we began here we had a few words before I 21:
began the tape addressing Mr. Page's memo which you had an opportunity to 22
[
read and I will provide you with a copy of this memo before you depart 23 today. As we indicated, the mamo covers the purpose of this investigation 24 and some of the goals we hope to achieve as well as the authority under 25!
i 4
I i
i 2
I which it is being conducted.
In addition there are several questions asked 2{
and at this time I would like to put those questions on tape and make them t
3l a matter of record on tape as well as on the printed form.
The first 4
question reads, do you understand the above making reference to the two-5l page memo?
61 VANWITBECK:
Yes.
7 l
8l HUNTER:
The Second question reads, do we have your permission to tape this i
g 10 11; VANWITBECK:
Yes.
121 13!
i HUNTER:
And third, do you want a copy of the tape?
14:
15:
VANWITBECK; Yes.
16:
17l HUNTER:
Okay, what I will do is at the conclusion of this interview, I 18{
will duplicat.e it and provide you a copy of the tape at that time and as a 19!
copy of the transcript becomes available, you will also be provided a copy 20i of the written transcript.
There is a fourth question not specifically 21l called out at the end of the letter, but covered in the body of it which 22l l
addresses your right, if you so choose to exercise it, to have a represen-23i l
tative from Metropolitan Edison present or a union member if you are an 24l l
employee of that class.
Can I ask would you like a company representative 25\\
present?
l
/04 093 l
l
(
3 i
i If VANWITBECK:
Yes, I would.
2' l
31 HUNTER:
Okay, I will indicate at this time that William Burley, Project i
4{
Engir.eer with Med Ed is present in the room and is in that capacity.
5l Mr. Van Witbeck to begin with if you would could you give us a brief resume 6
of your education, your work experience, your position with Energy, Inc.,
7 and how that addresses the TMI incident?
8 VANWITBECK:
g My background starts, in the nuclear business, in seven years 10j in the Navy, nuclear Navy.
I was an engineering and lab tech and an instruc-I tor at prototype, four years of college, and bachelor's degree in nuclear i
engineering, and nine years industrial experience, in startup and operating plant services, since.
Currently, I manage a start up and operating plant 14;,
services group for Energy, Inc.
As to the TMI involvement, we were brought l
in by Decamp to assist the client in identification of the sequence of 15-
,l events and evaluation of data and the assessment of the accident.
lb' 17 i
HUNTER:
Okay about Energy, Inc., could you provide me the address of the 18!
home offices?
19 20l VANWITBECK:
21!
Yes, It is P.O. Box 736, Idaho Falls, Idaho, and zip is 83401.
22I HUNTER:
Okay, and one other item.
I'd like to cove.- is the nature of your 23 relationship with Med Ed in the TMI incident.
Was tais brought about by a 24 recent contract or business agreement or has this been in existence prior 25j j
to the date of the incident?
I
- p. < -
k
() )
i i
4 1[
VANWITBECK:
This particular involvement came about after the incident.
2 Prior to the incident we have been involved with GPU through the Epri 3
Retran users group.
t 4j 5l HUNTER:
All right, one other point I'd like to make is that we address i
6' acronyms or jargon could we define as we go along to assist the girls who have to transcribe this?
7 8
VANWITBECK:
Sure.
The GPU being General Public Utilities, RETRAN being a g
c mputer code, thermal hydrulics computer code.
10 11!
HUNTER:
Okay, at that point then I'll turn it over to Tom Martin to begin the questioning.
14l l
MARTIN:
It's Tim Martin, Tom.
My interest at this point is in any informa-15; 16,'
tion you might have relative to loss of data in the computer system.
As a j
result of possible operator action or otherwise.
Can you provide us with 1/;
I any information in this area?
18j i
19!
VANWITBECK:
There was approximately an hour and half worth of data lost 20!
l from the alarm printer.
This information was lost because the alarm printer 21(
went out, was out of service and the printer was transferred over to the 22 print, the alarm print function was transferred over to the utility type-23 writer and in transferring back to the alarm printer after a new typewriter 24!
was installed, the data was lost during that transition.
As to whether or 25 l
684
[, J I
i i
{
5 i
i 1!
not it was operator action or a component misfailure, you know a failure of 2
a component, we don't know at this time.
We do know that it is missing and i'
3 it is well documented in sequence as you go down through the alarm type-4j writer that it skips.about an hour and a half worth of data.
I have got 5:
the exact times here if you want that.
6j MARTIN:
7 Tom, In earlier discussions you had informal discussions with, I g
think, Mr. Jim Creswell and you indicated that some data may have been lost g{
in the computer when NRC inspectors were interfacing with the computer.
Do y u have any knowledge of this?
If so, can you provide us with the specifics?
10 11l f
VANWITBECK:
There was a lot of data being taken on the day of the incident 12 5 and this is all third party inforamtion by the way.
I wasn't in the control i
13l
(
room at that time.
NRC investigators were calling things up on the type-
[
writer trending data, and in the process of this trending of the data, and 15i accumulating this data, other data, the data they were interested in was 16!
I removed, and the other data was dumped.
You know the paper was thrown away.
Some of it was data that some other people had been trending.
The data sheets, we have very few data sheets from these trends.
They seemed 19!
to have been removed or after they were used and the information taken from 20!
l them, they were discarded.
Something was cone with them.
They did not get 21 into the filing system at Med Ed.
l 2 31 MARTIN:
Tom, can you give an approximate time that this occurred, and.
24 l
25!
I n
?On o\\J > U j
OU1 i
[
6 If VANWITBECK:
I think what I am referring to is conversations I have had 2
with the people from the plant and from those conversations it would be in I
3 the first day and a half to two days where this was occurring.
i 4i 5l MARTIN:
Would this occur on the first day of the event or...
l 6l VANWITBECK:
Yes.
First and second days.
7 8
MARTIN:
g Do you know what people were involved in calling this data and p ssibly misplacing it?
10 11 VANWITBECK:
No, I don't know the names of the individuals.
I do know the, that it was the, there was data being trended by both NRC and by the plant 13l people and in the taking of this data frcm the computer tape, you know from the computer printout and removing it to other work locations in the plant (the control room) that this information never got back in the filer.
I 17l MARTIN:
Tom, So what we are saying is that if the data within the computer 1Sl was not disturbed, but in its printed form the portions of this printed 19l form were removed to other work locations e.d have not been 1;cated since.
20' 21; VANWITBECK:
True.
And this isn't with reference to the hour and a half 22 that was missing.
That data was just lost in the computer, the alarm data, 23{
and we are talking about other trending data information people had called up on the computer.
To my knowledge none of the alarm printer data was 25i missing because of people removing it.
t
., n 7 O( O h 0o I
U
[
t
I i
l I
11 MARTIN:
All right Tom, can you identify the information that was specifi-2f cally lost or must we refer to it in general terms?
I 31 i
VANWITBECK:
I, the only, I do know there were some thermocouples in it and 4
5 thermocouples being called up, but that is the extent of the knowledge right now.
I do know they were calling up more than that and you know g
using it, but as to what that data was I don't know at this time.
7 8
MARTIN;, Tom, to your knowledge is this data critical for the analysis of g
the event?
10!
(
11!
I VANWITBECK:
It would be hard to say whether or not it is critical becsuse 12I I
I don't know what it is.
It may or may not be.
The thermocouple informa-
,3l 14I tion would be significant as the plant here does not have a routine thermo-couple map as you would have on a Westinghouse plant, and these are called 15:
up on an individual basis.
17l l
MARTIN:
Tom, the information that we are talking about was it isolated to 18!
a single printer or typewriter system?
If so, what types of things would 19!
be printed on that typer that would be of interest to the investigation of 20i sequence?
21, 22l l
VANWITBECK:
The information that would be printed on that typer would be, 23l it would be on a utility typer.
It was the one that they were using at t
24i that time.
As to its value in the sequence, it probably wouldn't specify 25j 6BA CN
{
8 i
1 information as to when something occurred, that is a pump start and that gf type of thing.
What it would provide is amplifying information as to the 3
condition of the plant.
Whatever parameter was called up would be another i
4j data point on that parameter.
It may or may not be recorded elsewhere at c!
that time.
- t I
61 MARTIN:
7 So basically this probably involved d:;a that was called up because of a specific interest of the individual would not have been available g
normally unless someone else had called it up, and, therefore, it did not g
10j stud ne computer system specifically just because Gess, ac M ies were f
going on.
12' VANWITBECK:
It wouldn't have disturbed the computer system, however, that 13; f
you know you can say the time being used to call up this was interrupting trends thau other people were doing on the machine.
Each individual calling 16l up certain information would be wanting that nformation trended or printed l
out and that would interrupt and reduce the total availability of the 17 computer to other users or at least that typer.
I forgot the first part of 18!
your question, excuse me.
20j i
MARTIN:
Tom, let me go ahead.
I think you've provided the information 21!
needed there.
You indicated that this information you had was third party.
I 23!
VANWITBECK:
That is correct.
24 25; l
l l
pnl
,g.
g g, :,;
Ue i
i
I l
i l
9 MARTIN:
Which means that for me trace back and find out exactly what 1
2 information we are talking about, I am going to need to talk to those i
3j pe ple.
Can you give me the names of those people that might have first iy hand knowledge of these activities?
St I
VANWITBECK:
Some would be several of the operators in the control room at 6
the time.
7 I could have I would have to go back and look at my records.
I don't have them with me.
I do know that Ivan Porter who is an INC engineer, g
was one of the people that was using the typewriter at the time treading 9l I
some informatica.
10l l
lli MARSH:
Okay. I would like to request that data that you do have available 12j in your ocme records that I would have access to it.
14!
VANWITBECK:
Sure, right.
15i 16; MARSH:
Any names that you can recall or trace back that would have specific 17i first hand information I'd like to have access to it.
18l 19l VANWITBECK:
Fine.
20r i
21)
MARTIN:
Okay.
I just have one question the data that was missing would 22l include thermocouple data, has Med Ed or you done, have you done a trace in 23 l
the document trailer for that data?
24i 25!
r l
. rT i lO
\\ u eo*
s h
10 i
1{
VANWITBECK:
Oh, yes.
2l MARTIN:
It wasn't able, you weren't able to locate it?
3 l
4{
VANWITBECK:
No.
The information that was being trenced and as to what was 5
6 being called up would probably be best be gotten from the people who were doing it at the time that were calling the trend.
The value of that infor-7 mation would De you know best gotten from them as to what they were trending g
and what they were looking at.
It was my impression this information was g
being relayed to people offsite that were attempting to get a handle on the 10 situation.
Therefore, I would assume that it was of value.
lli 12 i
MARTIN:
Has any attempt been made to trace this data to these people since 13t obviously you had first hand or you've been informed that this was where 14!
the data is?
15!
16:
VANWIT8ECK:
Several requests have been made tc NRC to provide all informa-17!
tion they had.
The trending data that the other people were trending in 181 the control roca, we have gone back through all the waste material we could 19!
find in the control room.
This was three days later though, 3 to 5 days 20j later, and that in that type of time frame.
We could not find it and we j
went through all the boxes, the desks, the drawers, cabinets.
One of the 22' things the operators at this plant do is wnen they are taking data like 23 that they tear off the sheets and pardon me, they get the information that 241 they want, and taen those sheets that are torn out are put into the back of 25l l
. N. It
\\ &
- G 4
l hd h
l
- i I.
11 i
l{
the computer.
There is a cabinet behind it.
We pulled out the information 2
that was in there.
There were some 25 or 30 sheets though.
It wasn' t i
3l anything like we have had coming out of there at that time.
4{
[j MARTIN:
Tom, let me understand you then.
Are you indicating that you 6
can't determine how many sheets are missing or which sheets are missing?
i 7
ere is not at tMs dme any way to determine that because R 8
is off of a typewriter or a printer that doesn't run the, you know is not g
running in a routine sequence.
It has been interrupted to provide this information.
And as to the extent and the amount of information, no one, I can't.
All we know is, it was taken and it is not currently in our files, 12!
i and we have made a search of the control room for it.
Ana have asked the 13l operators and people to, you know anybody that was in there, to bring out 14!
of their files or in their files in the plant any information they hac.
16; MARTIN:
You indicate that there is no way to determine how many sheets are 17!
gone.
You know that some tear sheets were inserted in the back and you 18f indicated there were approximately 50 you found there.
19!
20l r
VANWITBECK:
Twenty-five or thirty.
21l 22!
I MARTIN:
All right 25 or 30.
How do you know that does not cover all the 23!
sheets that were actually typed?
24l 25I
,_ c
, v'-
t 1
h' I
i
(
12 i
If VANWITBECK:
In the volume of paper that goes th.udgh that machine was in 2,
continuous use and that would be more than that for that period of time i
3{
ver those several days.
Plus, there were no theraccouple data in there.
4!
gf MARSH:
For the sake of an understanding of the system here, could you i
6 briefly describe the system as we are addressing it what the input devices are, or what the output is?
Are you working off a buffer cr strictly addressing the computer straight? What's the nature of the printing devices?
g I
Are they IBM Selectrics or are they other devices? Are you using single g
paper or manifold paper?
Could you cover that briefly for us?
10 11; VANWITBECK:
Okay.
The computer system, I am not totally familiar with the whole system because you know this is not a plant that I have worked on in the past.
The paper we are using is a standard single sheet computer paper.
The printers are IBM Selectric typewriters.
The input devices are most of the plant parameters that are brought into the computer, and there 16!
j are some single point, there are quite a few single point parameters that 1/;
i can be called up on a code number point number and put on trend or called 18i up for single readings.
There are also some trend blocks that can be 19l called up, ana as to the content of all the trend blocks available, I don' t 20t i
know how all of those trend blocks that are available.
21; 22l l
MARSH:
Okay.
If I am understanding what you are saying earlier correctly, 231 I am looking at the possibility of laymen or other investigative teams 24; downstream listening to this tape that may not understand the intricacies 25!
/or 4 n, bl O.
r
- yj i
I
{
13 i
l{
f it, but are you saying that as this computer is calling these single i
2!
point pickups or taking information from those sources and is then asked to l
3j pr vide a printout that that interrupts that input of information?
4{
VANWITBECK:
It could interrupt a trend that was in process at the time, 5
i g
yes.
Or it will queue these and print them out from the queue.
71 MARSH:
All right, so as it is going through the process of queuing infor-g cation and preparing to feed it into the printer it no longer accepts g
f inputs or is that it still gets inputs but we just delay the actual print-10, I
t out of that information.
11!
l I
12' VANWITBECK:
The printout, I don't believe you will lose the printout information.
It is just a matter of an interrupt and then coming back.
Now you can and I don't know the hierarchy of commands, but there are certain commands that will come on and you could by setting up the correct i
trend rate, block out any printout, but I don' t know that that was going 17!
I on.
As far as somebody calling up a given point and putting it on trend to 18ll the extent that it would block out other inputs or if it the trend points 19!
they were looking at were of, let's say the trend commands were in the 20l l
hierarchy such that they would blank out anything else.
21{
i 22!
j MARSH:
The process of calling out a dump of particular information, would 23l l
this automatically wipe out the memory of information that was requested or 24I is that it just that it dumps and still maintains a fixed memory of that 25!
info?
r f D bb.
u
l t
{
14 1l VANWITBECK:
I do not know.
Whether as to the memory of this machine for 1
2 all single points, but based an the machine to my limited knowledge of it, 3l I 'Jn't believe that the information would be stored in the memory.
There 4
is a 12-hour dump, but I don't know what goes into it totally.
5 MARSH:
6{
Okay, but there is a 12-hour?
I 7l VANWITBECK:
Yes, but I don't knew what it, I just learned this a couple of g
g days ago here and I don't know the extent of the content of that storage.
10fj MARSH:
Okay.
What type of a dump is that?
Is it just a general erase or is it dumped out in a Mag tape or Mag card or is--?
12',
13!
VANWITBECK:
No, I don' t think it is dumped on Mag tape and I really don' t 145 know, but I immagine it is dumped out to this, through the Selectrics.
16i MARSH:
So there might still be a possibility that the total memory was dumped at a 12-hour period later and could possibly still be located? What 18[
I'm after is information that was lost, last forever or might we regain it 19l from a later dump.
20l 21!
VANWITBECK:
No, I don't think there was a later dump of that type of 22l information made for that machine and I am not sure what the capacity is of t
23 the machine and as to whether or not it just fills up the buffer and then 24 l
starts pushing the information out the back sida of the buffer.
25!
(> o q
}Jj e-a
l e
{
15 l
1{
MARSH:
That is all I have, thank you.
l 2:
VANWITBECK:
Okay.
3 4I 5l MARSH:
It appears that we are at the end.
Okay, r, one else indicates 6
they have any questions.
Tom, at this point I'd like to open it up to you.
l 7
We have asked you a few questions.
If there is any statement you would g
like to make or make a matter of record, I'd like to turn the microphone gj over to you at this time for your comments.
l 10!
i VANWITBECK:
Well, I guess I would ask the question that several requests 11l have been made to the NRC and are you aware of the status of any of these 12!
I requests that is the information that it was believed to be, have been in 13l NRC investigators' hands and the reps onsite at the time.
14t 15j MARSH:
Okay.
To the best of my knowledge, we had no investigators onsite I
during the event itself.
You had more operational people who were addressing 17l the operational conditions that were going on.
Any requests put to them, 18J no, I am not privy of any at this time.
19j 20; i
VANWITBECK:
Umhum.
2 22 MARSH:
And I speak that in a personal sense.
A 23 24j 25l n}
/ O$
v' g
b ()
't a
f
l
{
16 i
!lj VANWITBECK:
Yeah.
I understand.
El t
3l MARSH:
Our investigation as far as I know has not turned up any specific 4
requests.
I 5
VANWITBECK:
Okay.
That is the only questien I had.
6 l
7l MARSH:
Anything else? Okay, then the time being 1:40 p.m. and I am reading 8
410 on the meter.
I would like to say thank you for your time recognizing g
y u have a busy schedule and have an aircraft to catch.
Just say thank you 0
1 j
for your your input for showing up here.
I 12 VANWITBECK:
Well, thank you.
14 15j 16i 17l i
18[
19f 201 I
21l 22 23l i
24l 25l
,n7 l (' ?t e!
e 1
16 0 i
t i