ML17058B516
| ML17058B516 | |
| Person / Time | |
|---|---|
| Site: | Nine Mile Point  |
| Issue date: | 08/23/1991 |
| From: | NRC - INCIDENT INVESTIGATION TEAM |
| To: | |
| Shared Package | |
| ML17056C371 | List:
|
| References | |
| CON-IIT07-119A-91, CON-IIT07-119B-91, CON-IIT7-119A-91, CON-IIT7-119B-91 NUREG-1455, NUDOCS 9305060241 | |
| Download: ML17058B516 (92) | |
Text
ORIGINAL cg-II'I0 ~~
OFFICIALTDNSCRIPT OF PROCEEDINGS Agency:
Title:
Nuclear Regulatory Commission Incident Investigation Team Nine Mile Point Nuclear Power Plant Interview of:
ANIL JULKA Docket No.
ICATION:
- Scriba, New York DATE
- Friday, August 23, 1991 PACES:
1 20 AI'&RILEYR ASSOCIATES, LTD.
1612 KR. N.W., Suib. 300 shingon, D.C 20006 (202) ZH-SNO.
ERRATA SHEET
~Pa e Line ADDENDUM Correction and Reason for Correction FSAP C,HA,PTER. 7 Iiucl vga ~i+ IMIOQ
'HPCS lesm~
oF Hpsc.
T po me 8h 0
0
}o Wo ~
Yrlo M ~ ~~
r ~ WTcJuz.
4e h
<yn ba m~wIL t
V 4 ~
c why.
al l,A-7 ov sr ~
b~ @cd l5 K~)
vugg~
~4 ~ 4 D~ co&~
Date I~
~/
Signature
4 ~
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION INCIDENT INVESTIGATION TEAM 6
Interview of 7
ANIL JULKA 8
(Closed) 10 12 13 14 15 16 17 18 19 Conference Room B
Administration Building Nine Mile Point Nuclear Power Plant, Unit Two Lake Road
- Scriba, New York 13093 Friday, August 23, 1991 The interview commenced, pursuant to notice, 20 at 1:10 p.m.
21 22 23 24 25 PRESENT FOR THE IIT:
Jose Ibarra, NRC Jim Stoner,
- Duke, NRC
E
P R 0 C
E E
D I N G
S
[1:10 p.m.]
MR.
IBARRA:
This is Jose Ibarra from the NRC.
4 I'm a team member from the IIT and with me I have Jim 5
Stoner.
Today we will be interviewing Anil Julka from 7
Niagara Mohawk.
Anil, would you please state your name, 8
your position, experience with the company'?
MR. JULKA:
Okay.
My name is Anil Julka.
I'm an 10 electrical design supervisor for Niagara Mohawk electrical 11 design group.
I have been with Niagara Mohawk for 12 approximately five years now.
13 My previous experience includes AEs and 14 Westinghouse.
15 MR.
IBARRA:
Can you tell me your involvement or 16 your responsibility in assessing what happened on August 17 13th?
18 MR. JULKA:
Okay.
After the event I'm also part 19 of the TSC task force for evaluation.
After the event I got 20 a call that we had declared a site area emergency so I came 21 up to the site and at first they were just evaluating what 22 had happened.
And I got here around 9 o'lock or so or a 23 little after that.
And our main concern, we were 24 hypothesizing at that point.
We knew there was a
25 transformer fault, but we were hypothesizing how the fault
1 had started.
So, I guess my primary responsibility that day was 3
the major support for, you know, reactor shutdown.
And 4
after that my primary responsibility was tasked with 5
reviewing the electrical distribution system to see what 6
type of transients had occurred, and reviewing if the system 7
had operated as designed and assist UPS people with the 8
trouble shooting efforts and also with the main transformer 9
wherever they needed help.
So those were my primary 10 responsibilities.
lh MR.
IBARRA:
One of the obvious faults was 12 lighting.
What assessments have gone on in lighting to 13 assure that what the operator saw was correct and also final 14 assessments as to what this is going to mean?
15 MR. JULKA:
Okay.
We have reviewed the entire 16 plant lighting system.
You know, we have five types of 17 lighting systems.
We have a normal lighting, emergency 18 lighting, essential, egress and eight-hour battery pack 19 lighting system.
20 During this event the essential portion of the 21 lighting system was lost because of the loss of UPS.
The 22 normal lighting was still available, the emergency lighting 23 was still available.
In some of the areas we do have 24 anomaly in the system that in some stairways there is a
25 concern which was expressed, I guess, back in '89 that if we
1 loose UPS, certain areas -- stairwell areas get -- the 2
eight-hour battery packs do not come on because they are fed 3
from the normal source and since the lighting there is fed 4
from UPS you could loose the UPS and the eight-hour battery 5
packs do not come on.
At that time we did make an evaluation using 7
Appendix R and it was documented that the Appendix R
8 lighting was required for loss of normal power also in this 9
case.
And at that time also we did not loose normal power.
10 So what -- last year, I guess, we started looking at the 11 entire UPS loading issue because the electrical group was 12 concerned about the loading on the UPS's.
And at that time 13 we decided that we should fix that anomaly also.
So, 14 rather than leave it as a modification of 89-042 which will 15 address that issue and there's more that's currently being 16 scheduled for refuel outage and thereafter.
17 So, our evaluation of the incident really says 18
- that, yeah, we did loose essential lighting.
They did -- my 19 understanding from talking -- looking at the operator's 20 report is that they had to use flashlight in certain areas, 21 but those are open stairways, so there is enough light 22 coming in from other sides, so it wasn'0 completely dark, 23 but there was some lighting coming in from the other sides.
24 Other than that anomaly we have not really found 25 anything else which is contradictory to our commitments.
I
1 MR.
IBARRA:
There was an issue then with lighting 2
in '89, an UPS redistribution of loads in '90, correct?
MR. JULKA:
State that again?
I didn't follow 4
that.
MR.
IBARRA:
In 1990, there was a study done on 6
the redistribution of the lighting according to the new UPS 7
that were going to be put in?
MR. JULKA: I believe that was
'91.
Don't quote 9
me on that.
I thought we did it earlier this year.
10 MR.
IBARRA:
But, is the lighting issue separate 11 from the distribution issue?
12 MR. JULKA:
That's correct.
That's correct.
13 MR.
IBARRA:
Two different studies?
14 MR. JULKA:
Different studies, right.
15 MR.
IBARRA:
Okay.
But the lighting issue was a
16 few years before the distribution on the UPS?
17 18 MR. JULKA:
The lighting issue came up in '89.
MR.
IBARRA:
Okay.
The instruments that failed in 19 the control room, can you tell us what happened there and 20
- how, and the assessment you'e done so far?
21 MR. JULKA:
First of all, in our commitment, we 22 did not loose any of the safety related instrumentation in 23 the plant, even control room.
Our commitment in USAR or 24 FSAR Chapter 74, is that we need four safety systems for 25 safe shutdown which is the reactor core isolation cooling
6 1
- system, which is called RCIC; you got standby liquid control 2
- system, SLS; RHR shutdown cooling mode and remote shutdown 3
system.
These systems and control and instrumentation for 4
these systems was available, at least electrically, 5
although I have heard that RCIC went in off for a while 6
because of a valve, but it was not due to the UPS loss or 7
any -- the event that happened that day.
And my 8
understanding is also that at the time RCIC was declared 9
inop, it was not really required and we still had HPSC 10 available.
We didn't loose plant annunciators, computers, 12 feedwater control.
Those are some of the things -- there is 13 a complete list of things we have which were lost and we 14 have evaluated those things and we have a separate group set 15 up who is going to be doing the safety assessment of all 16 these things which were lost to insure that they will not 17 require our preliminary indication as that,
- yes, they were 18 not required for safe shutdown of the plant.
They assist 19 operators, but in no way are necessary for a safe shutdown.
20 MR.
IBARRA:
Was Rixie or RCIC all of it 21 inoperable or only portions of it inoperable?
22 23 MR. JULKA:
Portions of it.
MR.
IBARRA:
So you did have one channel available 24 versus -- two channels?
25 MR. JULKA:
Well, in the initial portion of the
event it was available to support core cooling.
But there 2
was a lack of full close indication for the valves, AOV-156, that's the primary containment isolation valve.
But it was not really needed at the time it was declared inop.
MR.
IBARRA:
The computer systems were not 7
available, can you tell me what those were fed off of and 8
does that make sense that they would have lost those?
MR. JULKA:
Yeah.
Well, most of the plant 10 computer system is fed of 1G, UPS 1G which was lost during 11 this event and the LWS computer is fed off the UPS 1B.
So 12 we lost five of the non-safety UPS's which are all the same 13 14 type, Exide, 1A, 1B, 1C, 1D and 1G.
So loss of those will result in loss of plant computer system and that's -- since 15 they are non-safety and they do feed the computer systems.
16 17 MR.
IBARRA:
Your regulatory guide 197 instrumentation, was that all purple?
18 MR. JULKA:
I don't have full details on that, but 19 20 21 my -- I would suspect they were, but don' MR.
IBARRA:
How about the post-accident monitoring instrumentation?
22 MR. JULKA:
Yeah.
Post-accident should be 23 operable.
24 MR.
IBARRA:
Okay.
On the Appendix R assessment, 25 the assessment of whether the capability existed to put out
8 1
- fires, can you tell me a little bit about your involvement, 2
your group involvement in that assessment?
MR. JULKA:
We have an Appendix R engineer in our 4
group.
He looked at the Appendix R issue and our commitment 5
in FSAR for Appendix R is in accordance with the fire 6
protection guidelines, you know, reg guide Appendix R
7 scenario in our plant is considered with the loss of normal 8
off-site power identified in any given area you have a
9 capability to shut down the plant.
10 We didn't really get into that scenario in this 11 case because normal power was still available.
So our 12 evaluation really states, you know, that Appendix R
13 compliance was not really impacted.
14 MR.
IBARRA:,When was your group called to look at 15 that electrically?
16 MR. JULKA:
Our group was never really called to 17 look at anything electrically, except to look at the entire 18 electrical distribution system.
It was my decision, I 19
- guess, for now that within my group that we should look at 20 every system there is which is, in fact, to make sure that 21 loss of these UPS systems would not really affect any 22 electrical related systems in the plant.
And our compliance 23 to USAR was still valid.
24 MR.
IBARRA:
Since a few of the instruments were 25 inoperable previous to the event itself, are you going to
9 1
have any involvement with assessing the impact that some of 2
those instruments being inoperable would have -- did have on 3
the event -- being able to indicate in the control room?
MR. JULKA:
Yeah, that will be assess as part of 5
our safety assessment.
MR.
STONER:
Have you reviewed -- completed a
7 review of the electrical distribution system, and have you 8
made a determination whether the system -- the in-plant 9
systems as well as the switch yard systems operated as 10
- designed, including the associated protective systems?
MR. JULKA:
Yeah.
We had a pretty extensive 12 review of the protective relaying associated with the unit 13 protection system.
And we also had a fellow from GE, Mel 14 Crenshaw do an independent assessment for us.
From what we 15 have seen so far, our preliminary report has been put out.
16 We have not seen any anomalies in the protective relaying 17 area, I think everything operated as it. was supposed to.
18 Everything -- all the relays operated as designed 19 and isolated the fault.
So we don't really see any 20 anomalies there.
21 MR.
STONER:
Does any of the information available 22 indicate that there were any perturbations that were 23 superimposed upon the voltages that were supplied -- or, 24 excuse me, are the source for the UPS systems or the safety 25
- buses, and if not, would you expect -- what kind of
10 1
perturbations may you have expected to be seen there, if 2
any?
MR. JULKA: Initially, when the event happened and 4
we did not have the oscillographs from Scriba.
We did -- in 5
our initial information was that, UPS's tripped on lower 6
voltage.
We did suspect some lower voltage transients at 7
that time.
But from noon on the 8/13 -- the day of the 8
- event, we received the copies of the oscillographs from 9
Scriba and at that time it was very clear to us that from 10 what was shown on the 345 side that there were no audible 11 phase transients in the plant.
12 The voltage did that and we did experience 13 undervoltage, especially in the B phase.
You know, there 14 were the associated undervoltage in the other two phases and 15 that under voltage was carried through the plant and that 16 was evident from the trouble shooting that we have done on 17 UPS to date and also at the same time, since our safety 18 systems are normally fed from the 115 kV offsite source 19 which originates from the 345 kV at the Scriba station, we 20 did see a dip in that voltage.
21 The reason I say that is because we did see 22 undervoltage relay flags on the divisional buses come in.
23 However, at that time it was not sufficient to initiate any 24 actions required due to the undervoltage.
Mainly, the 25 diesel didn't start, you know, that undervoltage really
11 1
starts the diesel.
So, I guess overall we didn't really foresee any 3
extraordinary voltage transients.
We did find undervoltage 4
in the plant for a few cycles while the line 23 which is the 5
345 kV line was disconnected from the system.
We did 6
accomplish a fast transfer as designed and all the loads 7
were transferred over to the reserve station service which 8
is a normal feed for the safety related buses.
So, we don't really see any anomalies there in the 10 electrical distribution system.
MR.
STONER:
From the information available, could 12 you and have you approximated what the voltage is -- may 13 have been on the three phases at the input source to the six 14
-- to the UPS systems and to the 4160 volt safety systems?
15 MR. JULKA:
Yeah.
On the safety systems we 16 suspect the voltage got below 92.5 which is our commitment 17 for the graded voltage level relays.
It went down below 18 that and the 80 percent which is the backup protection for 19 loss of voltage, and that sat for three
- seconds, those 20 didn t come in.
So I think all we know is it got down to 21 below 92.5 and not 80 for three seconds.
22 I imagine looking then in a rough approximation 23 will be we were maybe around 78-79 percent for just cycles.
24 And on the
- UPS, as per our discussions before, initially we 25 calculated from the Scriba station, we found out what the
~
~
12 1
voltages there were and we transformed those voltages back 2
to the plant and UPS's are fed from 208/120*/Wye system.
We did find the voltages, especially in the B
4
- phase, dipped down to 65 some odd volts.
But that was 5
based on that fault that was on the 345 side.
Then we also 6
calculated a voltage based on if the low voltage winding 7
phase B was shorted out, what the voltages would be.
So 8
those two extreme voltage evaluations tell us that the 9
voltage was between 48 and 65.
10 In all reality, I guess we would suspect that 11 voltage was around 55, somewhere in the middle if those were 12 the two extreme cases and we know that neither one of them 13 really were true.
So, it's really a hypothesis and we have 14 bounded the voltage which is consistent with the evaluation 15 which is being done with the UPS right now.
16 MR.
STONER:
What protective relay actions 17 occurred during this event?
18 MR. JULKA:
Okay.
Since the phase B main 19 transformer had fault in it the differential relay for the 20 main transformer B operated and the unit differential which 21 also connects the generator, includes the generator and 22 transformer region; that also operated.
We had fault 23 pressure relays on the transformer which operated.
And 24 those were the primary relays which operated and they 25 operated the lock out -- different lock out relays and
13 1
different schemes which initiates a turbine trip.
At the 2
same time we also noticed start-up -- generator to start-up 3
overcurrent relays came in and operated a different set of 4
schemes.
Our conclusion is that those came in after the 345 6
kV line was disconnected from the system.
Since those 7
relays have an on-line contact in series with them, that 8
those relays should only come on when the unit is off-line.
9 So, we imagine those relays came in a little later after the 10 generator was disconnected from the system.
And they do 11 operate another lock, or relay which sends a slow transfer 12 signal and that was another evidence that fast transfer did 13 take place, because those relays came in later on and our 14 charts show that the 115 kV lines did pick up the load soon 15 after disconnecting from the 345.
16 MR.
IBARRA:
Anil, can you explain how come the 17 diesel generators did not come on?
18 MR. JULKA:
The diesel generators are started only 19 if the degraded voltage relays on the safety buses -- the 20 voltage stays degraded for 30 seconds.
And we did not -- we 21 did initiate the relays, but we did not initiate the timers.
22 Our timers did not have enough time because the voltage did 23 not dip for that long.
24 MR.
IBARRA:
What is the assessment so far, as far 25 as the -- where the fault occurred on the high side and the
14 1
low side, can you explain?
MR. JULKA: I think it's the hypothesis, right now 3
-- our initial indication we had was that the above ground 4
currents flowing in the 345 kV system.
So, I had initially 5
hypothesized the ground to be in the -- or whatever the word 6
is, to be on the high side.
With all the transformer exposed, I looked at the 8
transformer.
So far they are saying that the fault may have 9
started on the low side.
So, I don't think there is a
10 definite conclusion on that as yet.
I think we can only 11 determine that after the transformer is sent out for 12 evaluation and they open it up and see that it -- we know 13 definitely that the 345 kV system had a ground in it 14 ground current flowing through it.
MR.
IBARRA:
What would be the difference in 16 assessment if it occurred on one side and the other side as 17 far as the protection of the system?
18 MR. JULKA:
The protection schemes would operate 19 either way.
The only difference would be if there was a
20 ground fault on the low voltage side prior to short -- you 21 know, shorting out completely.
I think the ground voltage 22 relays on the generator side may have operated quicker.
23 They are disconnected after the lockout relay operates.
24 So it was a race in time, I guess with different 25 relays operating time.
But I would have expected that the
15 1
generator ground relays may have come in if the fault had 2
started on the ground side.
On the low voltage side.
MR.
IBARRA:
How long does it take to clear a
4 fault and get back on line?
MR. JULKA: If a fast transfer is going to take 6
place is going to take place we should clear the fault 7
within six cycles and transfer over to the other reserve 8
transformers.
So our commitment in the FSAR is the six 9
cycles from the initiation from the lockout.
After the 10 lockout relay operate the initiator timer, which is set for 11 six cycles, and if the transfer does not take place in six 12 cycles we disconnect the offsite sources and we go for a 13 slow transfer.
14 And there's another relay which monitors that and 15 only connects the buses after 30 seconds and it also 16 disconnects, feedwater -- all the large motors, feedwater 17 condensate
- booster, recirc pump is tripped to low speed, so 18 those connection did take place.
That's where there was 19 some confusion in it initially whether it was a fast 20 transfer or slow transfer because we did have a trip of the 21 feedwater recirc pump and condensate booster pumps.
But 22 looking at the charts we have established that fast transfer 23 did take place and we had to review other things on why 24 those pumps tripped.
25 MR.
IBARRA:
Can you explain the in-plant
16 1
monitoring system, what's available and what -- during this 2
incident?
MR. JULKA:
Not a whole lot.
I guess we have 4
we had to make all our reductions for the 345 kV side.
We 5
had an oscillograph in the plant which is an old style 6
old type which was not working at the time of the event.
So 7
we did not really have too much information on the plant 8
side.
We had to make our deductions from the high voltage 9
side.
10 MR.
STONER:
You indicated there was no operation 11 of the generator neutral ground relay.
Has that system been 12 checked to verify that it is in service?
13 MR. JULKA:
Right.
We did check the resistor.
We 14 checked the transformer and they have verified that the 15 system is in tact.
We had a concern after -- you know, the 16 different theories on other transformer came out, that since 17 a fault could happen on the low voltage side, so we wanted 18 to check out the generator grounding system, isophase bus 19 and the other transformers.
So they have checked that out 20 and they have verified that, yeah, it does work properly.
21 There is no damage to that piece of equipment.
22 MR.
STONER:
Have you been able to determine 23 whether the generator surge arresters may have operated on 24 the isolated phase bus system?
25 MR. JULKA:
We asked them to check that, but. I
17 1
don't have the final indication if there was a damage there 2
or if they operated at this time.
But we should have that 3
information by tomorrow.
MR.
STONER:
Do those arresters have counters or 5
would the determination of their operation simply be a
6 matter of inspection?
MR. JULKA: It will be just a matter of 8
inspection.
MR.
IBARRA:
Can you tell me what other 10 consultants have helped you in your assessment of this 11 event as far as distribution is concerned -- electrical 12 distribution?
13 MR. JULKA:
Well, the assessment was made by our 14 group, but we had some consultants verify what we had done 15 for an independent review.
And that was Mel Crenshaw from 16 GE.
He has prepared a report and I have given a copy to 17 Jim Stoner -- of his report.
Plus there were
- some, you 18 know, other people at Niagara Mohawk -- people.
There were 19 some consultants called in from Stone and Webster for 20 different evaluations to -- you know, make out some lists 21 and stuff; like load lists, and also helped with the UPS in 22 the plant.
Since I didn't think UPS was clearly system 23 engineering, design had to be involved in that so we had 24 people we called in from Stone and Webster to stay full time 25 at the UPS issue.
18 MR.
IBARRA:
Who are the people from Stone and 2
Webster looking at distribution?
MR. JULKA:
There was nobody from Stone and 4
Webster looking at distribution.
There was one guy from 5
Stone and Webster -- Steve Tsombaris who looked at the UPS 6
and helped with the UPS testing.
We had on our distribution 7
- side, we had Leon Blasiak, he's an ex-Niagara Mohawk 8
retiree who was initially involved.
So he was involved with 9
looking at some of the Scriba stuff -- voltages.
10 We had one guy, Ranjit Das who was here to help 11 with the fire protection and Appendix R from ASTA which is a
12 floor engineering -- he is ex-Stone and Webster, but now he 13 works for ASTA, so we called him for some assistance in the 14 lighting and Appendix R issues since he's a known Appendix R
15 engineer in the industry right now and he was involved with 16 the original design.
17 And we had three -- we had four more people who 18 were assisting us down in Salina Meadows with the 19 preparation of the load list, plant impact statements.
That 20
- included, you know, Steve Erikson, Pat O'rien, Roger Wyatt, 21 and there were two other people who were helping us put it 22 together over the week last weekend.
23 MR.
IBARRA: Initially when the UPS went down, 24 they were able to bring them up on the maintenance bus or 25 the maintenance side, yet I guess two hours into the event
19 1
they tried to line it up with the normal AC.
Do you have 2
any idea why they would try to do that?
And why not leave 3
it on the maintenance?
5 I'm MR. JULKA:
Repeat that question again, I think MR.
IBARRA:
Originally they brought the UPS back 7
up on the maintenance -- the alternate power supply.
And 8
yet later on into the event, a few hours later, they tried 9
to switch to the normal line up.
Do you happen to have any 10 reason why they would do that and not just leave it as it 11 was?
12 MR. JULKA:
No.
I guess I don't -- I'm not that 13 familiar with the operations procedures so I would rather 14 not answer that question.
15 MR.
IBARRA:
But as far as you'e -- the quality 16 of power from the alternate source in the normal AC source 17 going into the UPS, is it the same?
18 MR. JULKA:
Yeah.
The regulation is the same on 19 the voltages.
It's plus or minus two percent.
But I think 20 initially when the UPS went down the operations people 21 started to recover and after a couple of hours we had our 22 system engineer in, so at that time I think they may be 23 trying to connect that to the regular source.
24 MR.
IBARRA: If we looked past the UPS to let' 25 say the 120 site, would we notice any difference in the bus
20 1
-- in the AC whether it was off the inverter or whether it 2
was off the maintenance?
MR. JULKA:
No. It doesn'0 really matter, I 4
- guess, downstream if they get plus or minus two percent.
I 5
think that's covered by -- the UPS's regulate that.
And so, 6
you know, there was no concern on the -- the UPS systems do 7
have loads which are sensitive for the voltage regulation 8
and since the loads were disconnected so there was no 9
deviation farther down the line.
10 MR.
IBARRA:
Okay.
That's all that we have.
That 11 terminates the interview.
12
[Whereupon, at 1:45 p.m., the taking of the 13 interview was concluded.]
14 15 16 17 18 19 20 21 22 23 24 25
REPORTER'S CERTIFICATE This is to certify that the attached proceedings before the United States Nuclear Regulatory Commission In the Matter of:
NAME OF PROCEEDING:
Interview of Anil Julka DOCKET NUMBER:
(Not applicable)
PLACE OF PROCEEDING:
- Scriba, New York were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and thereafter reduced to typewriting by me or under the direction of the court reporting company, and that the transcript is a true and accurate record of the foregoing proceedings.
Mark Handy Official Reporter Ann Riley
& Associates, Ltd.
OFFICIALT1VHSCRIPT OF PROCEEDINGS Agency:
Title:
Nuclear Regulatory Commission Incident Investigation Team Nine Mile Point Nuclear Power Plant Interview of:
ANIL JULKA Docket No.
LOCATION:
- Scriba, New York DATE:
- Friday, August 23, 1991 PAGES:
] - 20 AI'&RILEYR ASSOCIATES, LTD.
1612 KR. N.'W., Suitt.'00 Washinyon, D.C 20006 (202) 293-3950.
'I P
ERRATA SHEET ADDENDUM
~Pa e Line Correction and Reason for Correction I-SQ p c.~A,pTEw 7 - +
/IvcLuoeD) /i~'& I~<loQ
'HPCS larnyx<> ot-HPSC.
7'fPo w< 4 h ~
u-sLncp Vo i~N IR~ ZK t4~ 5~
4Q k 4u ~
w<
.e s4o.V bg k
4 un&'c
<4I t 9 Le ~
c J4-.
LA 7
SG-4~
e
~cg<
ov s 4~
b4-Kc.lM~
<~~ 4
~e
( <
~s
-W~ )g~k 4 U~
cc ML Date
<~
~/
Signature
W
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION INCIDENT INVESTIGATION TEAM 6
Interview of 7
ANIL JULKA 8
(Closed) 10 12 13 14 15 16 17 18 Conference Room B
Administration Building Nine Mile Point Nuclear Power Plant, Unit Two Lake Road
- Scriba, New York 13093 Friday, August 23, 1991 19 The interview commenced, pursuant to notice, 20 at 1:10 p.m.
21 ej 22 23 24 25 PRESENT FOR THE IIT:
Jose Ibarra, NRC Jim Stoner,
- Duke, NRC
P R 0 C
E E
D I N G
S
[1:10 p.m.]
MR.
IBARRA:
This is Jose Ibarra from the NRC.
4 I'm a team member from the IIT and with me I have Jim 5
Stoner.
Today we will be interviewing Anil Julka from 7
Niagara Mohawk.
Anil, would you please state your name, 8
your position, experience with the company?
MR. JULKA:
Okay.
My name is Anil Julka.
I'm an 10 electrical design supervisor for Niagara Mohawk electrical 11 design group.
I have been with Niagara Mohawk for 12 approximately five years now.
13 My previous experience includes AEs and 14 Westinghouse.
15 MR.
IBARRA:
Can you tell me your involvement or 16 your responsibility in assessing what happened on August 17 13th?
18 MR. JULKA:
Okay.
After the event I'm also part 19 of the TSC task force for evaluation.
After the event I got 20 a call that we had declared a site area emergency so I came 21 up to the site and at first they were just evaluating what 22 had happened.
And I got here around 9 o'lock or so or a 23 little after that.
And our main concern, we were 24 hypothesizing at that point.
We knew there was a
25 transformer fault, but we were hypothesizing how the fault
1 had started.
So, I guess my primary responsibility that day was 3
the major support for, you know, reactor shutdown.
And 4
after that my primary responsibility was tasked with 5
reviewing the electrical distribution system to see what 6
type of transients had occurred, and reviewing if the system 7
had operated as designed and assist UPS people with the 8
trouble shooting efforts and also with the main transformer 9
wherever they needed help.
So those were my primary 10 responsibilities.
MR.
IBARRA:
One of the obvious faults was 12 lighting.
What assessments have gone on in lighting to 13 assure that what the operator saw was correct and also final 14 assessments as to what this is going to mean?
15 MR. JULKA:
Okay.
We have reviewed the entire 16 plant lighting system.
You know, we have five types of 17 lighting systems.
We have a normal lighting, emergency 18 lighting, essential, egress and eight-hour battery pack 19 lighting system.
20 During this event the essential portion of the 21 lighting system was lost because of the loss of UPS.
The 22 normal lighting was still available, the emergency lighting 23 was still available.
In some of the areas we do have 24 anomaly in the system that in some stairways there is a
25 concern which was expressed, I guess, back in '89 that if we
1 loose UPS, certain areas -- stairwell areas get -- the 2
eight-hour battery packs do not come on because they are fed 3
from the normal source and since the lighting there is fed 4
from UPS you could loose the UPS and the eight-hour battery 5
packs do not come on.
At that time we did make an evaluation using 7
Appendix R and it was documented that the Appendix R
8 lighting was required for loss of normal power also in this 9
case.
And at that time also we did not loose normal power.
10 So what -- last year, I guess, we started looking at the 11 entire UPS loading issue because the electrical group was 12 concerned about the loading on the UPS's.
And at that time 13 we decided that we should fix that anomaly also.
So, 14 rather than leave it as a modification of 89-042 which will 15 address that issue and there's more that's currently being 16 scheduled for refuel outage and thereafter.
17 So, our evaluation of the incident really says 18
- that,
-yeah, we did loose essential lighting.
They did -- my 19 understanding from talking -- looking at the operator's 20 report is that they had to use flashlight in certain areas, 21 but those are open stairways, so there is enough light 22 coming in from other sides, so it wasn't completely dark, 23 but there was some lighting coming in from the other sides.
24 Other than that anomaly we have not really found 25 anything else which is contradictory to our commitments.
MR.
IBARRA:
There was an issue then with lighting 2
in '89, an UPS redistribution of loads in '90, correct?
MR. JULKA:
State that again?
I didn't follow 4
that.
MR.
IBARRA:
In 1990, there was a study done on 6
the redistribution of the lighting according to the new UPS 7
that were going to be put in?
MR. JULKA: I believe that was '91.
Don't quote 9
me on that.
I thought we did it earlier this year.
10 MR.
IBARRA:
But, is the lighting issue separate 11 from the distribution issue?
12 13 14 15 MR. JULKA:
That's correct.
That's correct.
MR.
IBARRA:
Two different studies?
MR. JULKA:
Different studies, right.
MR.
IBARRA:
Okay.
But the lighting issue was a
16 few years before the distribution on the UPS?
17 MR. JULKA:
The lighting issue came up in '89.
18 MR.
IBARRA:
Okay.
The instruments that failed in ll 19 the control room, can you tell us what happened there and 20
- how, and the assessment you'e done so far?
21 MR. JULKA:
First of all, in our commitment, we 22 did not loose any of the safety related instrumentation in 23 the plant, even control room.
Our commitment in USAR or 24 FSAR Chapter 74, is that we need four safety systems for 25 safe shutdown which is the reactor core isolation cooling
6 1
- system, which is called RCIC; you got standby liquid control 2
- system, SLS; RHR shutdown cooling mode and remote shutdown 3
system.
These systems and control and instrumentation for 4
these systems was available, at least electrically, 5
although I have heard that RCIC went in off for a while 6
because of a valve, but it was not due to the UPS loss or 7
any -- the event that happened that day.
And my 8
understanding is also that at the time RCIC was declared 9
inop, it was not really required and we still had HPSC 10 available.
We didn't loose plant annunciators, computers, 12 feedwater control.
Those are some of the things -- there is 13 a complete list of things we have which were lost and we 14 have evaluated those things and we have a separate group set 15 up who is going to be doing the safety assessment of all 16 these things which were lost to insure that they will not 17 require our preliminary indication as that,
- yes, they were 18 not required for safe shutdown of the plant.
They assist 19 operators, but in no way are necessary for a safe shutdown.
20 MR.
IBARRA:
Was Rixie or RCIC all of it 21 inoperable or only portions of it inoperable?
22 MR. JULKA:
Portions of it.
23 MR.
IBARRA:
So you did have one channel available 24 versus -- two channels?
25 MR. JULKA:
Well, in the initial portion of the
7 1
event it was available to support core cooling.
But there 2
was a lack of full close indication for the valves, AOV-156, 3
that's the primary containment isolation valve.
But it was not really needed at the time it was 5
declared inop.
MR.
IBARRA:
The computer systems were not 7
available, can you tell me what those were fed off of and 8
does that make sense that they would have lost those?
MR. JULKA:
Yeah.
Well, most of the plant 10 computer system is fed of 1G, UPS 1G which was lost during 11 this event and the LWS computer is fed off the UPS 1B.
So 12 we lost five of the non-safety UPS's which are all the same 13 type, Exide, 1A, 1B, 1C, 1D and 1G.
So loss of those will 14 result in loss of plant computer system and that's -- since 15 they are non-safety and they do feed the computer systems.
16 MR.
IBARRA:
Your regulatory guide 197 17 instrumentation, was that all purple?
18 MR. JULKA: I don't have full details on that, but 19 my -- I would suspect they were, but don't--
20 MR.
IBARRA:
How about the post-accident 21 monitoring instrumentation?
22 MR. JULKA:
Yeah.
Post-accident should be 23 operable.
24 MR.
IBARRA:
Okay.
On the Appendix R assessment, 25 the assessment of whether the capability existed to put out
8 1
- fires, can you tell me a little bit about your involvement, 2
your group involvement in that assessment?
MR. JULKA:
We have an Appendix R engineer in our 4
group.
He looked at the Appendix R issue and our commitment 5
in FSAR for Appendix R is in accordance with the fire 6
protection guidelines, you know, reg guide Appendix R
7 scenario in our plant is considered with the loss of normal 8
off-site power identified in any given area you have a
9 capability to shut down the plant.
10 We didn't really get into that scenario in this 11 case because normal power was still available.
So our 12 evaluation really states, you know, that Appendix R
13 compliance was not really impacted.
14 MR.
IBARRA:
When was your group called to look at 15 that electrically?
16 MR. JULKA:
Our group was never really called to 17 look at anything electrically, except to look at the entire 18 electrical distribution system.
It was my decision, I 19
- guess, for now that within my group that we should look at 20 every system there is which is, in fact, to make sure that 21 loss of these UPS systems would not really affect any 22 electrical related systems in the plant.
And our compliance 23 to USAR was still valid.
24 MR.
IBARRA:
Since a few of the instruments were 25 inoperable previous to the event itself, are you going to
1 9
1 have any involvement with assessing the impact that some of 2
those instruments being inoperable would have -- did have on 3
the event -- being able to indicate in the control room?
MR. JULKA:
Yeah, that will be assess as part of 5
our safety assessment.
MR.
STONER:
Have you reviewed -- completed a
7 review of the electrical distribution system, and have you 8
made a determination whether the system -- the in-plant 9
systems as well as the switch yard systems operated as 10
- designed, including the associated protective systems?
MR. JULKA:
Yeah.
We had a pretty extensive 12 review of the protective relaying associated with the unit 13 protection system.
And we also had a fellow from GE, Mel 14 Crenshaw do an independent assessment for us.
From what we 15 have seen so far, our preliminary report has been put out.
16 We have not seen any anomalies in the protective relaying 17 area, I think everything operated as it was supposed to.
18 Everything -- all the relays operated as designed 19 and isolated the fault.
So we don't really see any 20 anomalies there.
21 MR.
STONER:
Does any of the information available 22 indicate that there were any perturbations that were 23 superimposed upon the voltages that were supplied -- or, 24 excuse me, are the source for the UPS systems or the safety 25
- buses, and if not, would you expect -- what kind of
10 1
perturbations may you have expected to be seen there, if 2
any?
MR. JULKA: Initially, when the event happened and 4
we did not have the oscillographs from Scriba.
We did -- in 5
our initial information was that UPS's tripped on lower 6
voltage.
We did suspect some lower voltage transients at, 7
that time.
But from noon on the 8/13 -- the day of the 8
- event, we received the copies of the oscillographs from 9
Scriba and at that time it was very clear to us that from 10 what was shown on the 345 side that there were no audible 11 phase transients in the plant.
12 The voltage did that and we did experience 13 undervoltage, especially in the B phase.
You know, there 14 were the associated undervoltage in the other two phases and 15 that under voltage was carried through the plant and that 16 was evident from the trouble shooting that we have done on 17 UPS to date and also at the same time, since our safety 18 systems are normally fed from the 115 kV offsite source 19 which originates from the 345 kV at the Scriba station, we 20 did see a dip in that voltage.
21 The reason I say that is because we did see 22 undervoltage relay flags on the divisional buses come in.
23 However, at that time it was not sufficient to initiate any 24 actions required due to the undervoltage.
Mainly, the t
25 diesel didn't start, you know, that undervoltage really
11 1
starts the diesel.
So, I guess overall we didn't really foresee any 3
extraordinary voltage transients.
We did find undervoltage 4
in the plant for a few cycles while the line 23 which is the 5
345 kV line was disconnected from the system.
We did 6
accomplish a fast transfer as designed and all the loads 7
were transferred over to the reserve station service which 8
is a normal feed for the safety related buses.
So, we don't really see any anomalies there in the 10 electrical distribution system.
MR.
STONER:
From the information available, could 12 you and have you approximated what the voltage is -- may 13 have been on the three phases at the input source to the six 14
-- to the UPS systems and to the 4160 volt safety systems?
MR. JULKA:
Yeah.
On the safety systems we 16 suspect the voltage got below 92.5 which is our commitment 17 for the graded voltage level relays.
It went down below 18 that and the 80 percent which is the backup protection for 19 loss of voltage, and that sat for three
- seconds, those 20 didn't come in.
So I think all we know is it got down to 21 below 92.5 and not 80 for three seconds.
22 I imagine looking then in a rough approximation 23 will be we were maybe around 78-79 percent for just cycles.
24 And on the
- UPS, as per our discussions before, initially we 25 calculated from the Scriba station, we found out what the
12 1
voltages there were and we transformed those voltages back 2
to the plant and UPS's are fed from 208/120*/Wye system.
We did find the voltages, especially in the B
4
- phase, dipped down to 65 some odd volts.
But that was 5
based on that fault that was on the 345 side.
Then we also 6
calculated a voltage based on if the low voltage winding 7
phase B was shorted out, what the voltages would be.
So 8
those two extreme voltage evaluations tell us that the 9
voltage was between 48 and 65.
10 In all reality, I guess we would suspect that 11 voltage was around 55, somewhere in the middle if those were 12 the two extreme cases and we know that neither one of them 13 really were true.
So, it's really a hypothesis and we have 14 bounded the voltage which is consistent with the evaluation 15 which is being done with the UPS right now.
16 MR.
STONER:
What protective relay actions 17 occurred during this event?
18 MR. JULKA:
Okay.
Since the phase B main 19 transformer had fault in it the differential relay for the 20 main transformer B operated and the unit differential which 21 also connects the generator, includes the generator and 22 transformer region; that also operated.
We had fault 23 pressure relays on the transformer which operated.
And 24 those were the primary relays which operated and they 25 operated the lock out -- different lock out relays and
13 1
different schemes which initiates a turbine trip.
At the 2
same time we also noticed start-up -- generator to start-up 3
overcurrent relays came in and operated a different set of 4
schemes.
Our conclusion is that those came in after the 345 6
kV line was disconnected from the system.
Since those 7
relays have an on-line contact in series with them, that 8
those relays should only come on when the unit is off-line.
9 So, we imagine those relays came in a little later after the 10 generator was disconnected from the system.
And they do 11 operate another lock, or relay which sends a slow transfer 12 signal and that was another evidence that fast transfer did 13 take place, because those relays came in later on and our 14 charts show that the 115 kV lines did pick up the load soon 15 after disconnecting from the 345.
16 MR.
IBARRA:
Anil, can you explain how come the 17 diesel generators did not come on?
18 MR. JULKA:
The diesel generators are started only 19 if the degraded voltage relays on the safety buses -- the 20 voltage stays degraded for 30 seconds.
And we did not -- we 21 did initiate the relays, but we did not initiate the timers.
22 Our timers did not have enough time because the voltage did 23 not dip for that long.
24 MR.
IBARRA:
What is the assessment so far, as far 25 as the -- where the fault occurred on the high side and the
14 1
low side, can you explain?
MR. JULKA: I think it's the hypothesis, right now 3
-- our initial indication we had was that the above ground 4
currents flowing in the 345 kV system.
So, I had initially 5
hypothesized the ground to be in the -- or whatever the word 6
is, to be on the high side.
With all the transformer
- exposed, I looked at the 8
transformer.
So far they are saying that the fault may have 9
started on the low side.
So, I don't think there is a
10 definite conclusion on that as yet.
I think we can only 11 determine that after the transformer is sent out for 12 evaluation and they open it up and see that it -- we know 13 definitely that the 345 kV system had a ground in it 14 ground current flowing through it.
15 MR.
IBARRA:
What would be the difference in 16 assessment if it occurred on one side and the other side as 17 far as the protection of the system?
18 MR. JULKA:
The protection schemes would operate 19 either way.
The only difference would be if there was a
20 ground fault on the low voltage side prior to short -- you 21 know, shorting out completely.
I think the ground voltage 22 relays on the generator side may have operated quicker.
23 They are disconnected after the lockout relay operates.
So it was a race in time, I guess with different 25 relays operating time.
But I would have expected that the
15 1
generator ground relays may have come in if the fault had 2
started on the ground side.
On the low voltage side.
MR.
IBARRA:
How long does it take to clear a
4 fault and get back on line?
MR. JULKA: If a fast transfer is going to take 6
place is going to take place we should clear the fault 7
within six cycles and transfer over to the other reserve 8
transformers.
So our commitment in the FSAR is the six 9
cycles from the initiation from the lockout.
After the 10 lockout relay operate the initiator timer, which is set for 11 six cycles, and if the transfer does not take place in six 12 cycles we disconnect the offsite sources and we go for a 13 slow transfer.
14 And there's another relay which monitors that and 15 only connects the buses after 30 seconds and it also 16 disconnects, feedwater -- all the large motors, feedwater 17 condensate
- booster, recirc pump is tripped to low speed, so 18 those connection did take place.
That's where there was 19 some confusion in it initially whether it was a fast 20 transfer or slow transfer because we did have a trip of the 21 feedwater recirc pump and condensate booster pumps.
But 22 looking at the charts we have established that fast transfer 23 did take place and we had to review other things on why 24 those pumps tripped.
25 MR.
IBARRA:
Can you explain the in-plant
16 1
monitoring system, what's available and what -- during this 2
incident?
MR. JULKA:
Not a whole lot.
I guess we have 4
we had to make all our reductions for the 345 kV side.
We 5
had an oscillograph in the plant which is an old style 6
old type which was not working at the time of the event.
So 7
we did not really have too much information on the plant 8
side.
We had to make our deductions from the high voltage 9
side.
10 MR.
STONER:
You indicated there was no operation 11 of the generator neutral ground relay.
Has that system been 12 checked to verify that it is in service?
13 MR. JULKA:
Right.
We did check the resistor.
We 14 checked the transformer and they have verified that the 15 system is in tact.
We had a concern after -- you know, the 16 different theories on other transformer came out, that since 17 a fault could happen on the low voltage side, so we wanted 18 to check out the generator grounding system, isophase bus 19 and the other transformers.
So they have checked that out 20 and they have verified that, yeah, it does work properly.
21 There is no damage to that piece of equipment.
22 MR.
STONER:
Have you been able to determine 23 whether the generator surge arresters may have operated on 24 the isolated phase bus system?
25 MR. JULKA:
We asked them to check that, but I
17 1
don't have the final indication if there was a damage there 2
or if they operated at this time.
But we should have that 3
information by tomorrow.
MR.
STONER:
Do those arresters have counters or 5
would the determination of their operation simply be a
6 matter of inspection?
MR. JULKA: It will be just a matter of 8
inspection.
MR.
IBARRA:
Can you tell me what other 10 consultants have helped you in your assessment of this 11 event as far as distribution is concerned -- electrical 12 distribution?
13 MR. JULKA:
Well, the assessment was made by our 14 group, but we had some consultants verify what we had done 15 for an independent review.
And that was Mel Crenshaw from 16 GE.
He has prepared a report and I have given a copy to 17 Jim Stoner -- of his report.
Plus there were
- some, you 18 know, other people at Niagara Mohawk -- people.
There were 19 some consultants called in from Stone and Webster for 20 different evaluations to -- you know, make out some lists 21 and stuff; like load lists, and also helped with the UPS in 22 the plant.
Since I didn't think UPS was clearly system 23 engineering, design had to be involved in that so we had 24 people we called in from Stone and Webster to stay full time 25 at the UPS issue.
18 MR.
IBARRA:
Who are the people from Stone and 2
Webster looking at distribution?
MR. JULKA:
There was nobody from Stone and 4
Webster looking at distribution.
There was one guy from 5
Stone and Webster -- Steve Tsombaris who looked at the UPS 6
and helped with the UPS testing.
We had on our distribution 7
- side, we had Leon Blasiak, he's an ex-Niagara Mohawk 8
retiree who was initially involved.
So he was involved with 9
looking at some of the Scriba stuff -- voltages.
10 We had one guy, Ranjit Das who was here to help 11 with the fire protection and Appendix R from ASTA which is a
12 floor engineering -- he is ex-Stone and Webster, but now he 13 works for ASTA, so we called him for some assistance in the 14 lighting and Appendix R issues since he's a known Appendix R
15 engineer in the industry right now and he was involved with 16 the original design.
17 And we had three -- we had four more people who 18 were assisting us down in Salina Meadows with the 19 preparation of the load list, plant impact statements.
That 20
- included, you know, Steve Erikson, Pat O'rien, Roger Wyatt, 21 and there were two other people who were helping us put it 22 together over the week last weekend.
23 MR.
IBARRA: Initially when the UPS went down, 24 they were able to bring them up on the maintenance bus or 25 the maintenance side, yet I guess two hours into the event
19 1
they tried to line it up with the normal AC.
Do you have 2
any idea why they would try to do that?
And why not leave 3
it on the maintenance?
5 I'm--
MR. JULKA:
Repeat that question again, I think MR.
IBARRA:
Originally they brought the UPS back 7
up on the maintenance -- the alternate power supply.
And 8
yet later on into the event, a few hours later, they tried 9
to switch to the normal line up.
Do you happen to have any 10 reason why they would do that and not just leave it as it 11 was?
12 MR. JULKA:
No.
I guess I don't -- I'm not that.
13 familiar with the operations procedures so I would rather 14 not answer that question.
15 MR.
IBARRA:
But as far as you'e -- the quality 16 of power from the alternate source in the normal AC source 17 going into the UPS, is it the same?
18 MR. JULKA:
Yeah.
The regulation is the same on 19 the voltages.
It's plus or minus two percent.
But I think 20 initially when the UPS went down the operations people 21 started to recover and after a couple of hours we had our 22 system engineer in, so at that time I think they may be 23 trying to connect that to the regular source.
24 MR.
IBARRA: If we looked past the UPS to let' 25 say the 120 site, would we notice any difference in the bus
20 1
-- in the AC whether it was off the inverter or whether it 2
was off the maintenance'?
MR. JULKA:
No. It doesn't really matter, I 4
- guess, downstream if they get plus or minus two percent.
I 5
think that's covered by -- the UPS's regulate that.
And so, 6
you know, there was no concern on the -- the UPS systems do 7
have loads which are sensitive for the voltage regulation 8
and since the loads were disconnected so there was no 9
deviation farther down the line.
10 MR.
IBARRA:
Okay.
That's all that we have.
That 11 terminates the interview.
12
[Whereupon, at 1:45 p.m., the taking of the 13 interview was concluded.]
14 15 16 17 18 19 20 21 22 23 24 25
REPORTER'S CERTIFICATE This is to certify that the attached proceedings before the United States Nuclear Regulatory Commission In the Matter of:
NAME OF PROCEEDING:
Interview of Anil Julka DOCKET NUMBER:
(Not applicable)
PLACE OF PROCEEDING:
- Scriba, New York were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and thereafter reduced to typewriting by me or under the direction of the court reporting company, and that the transcript is a true and accurate record of the foregoing proceedings.
Mark Handy Official Reporter Ann Riley
& Associates, Ltd.