ML20149L749
| ML20149L749 | |
| Person / Time | |
|---|---|
| Issue date: | 11/08/1996 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-2082, NUDOCS 9611150136 | |
| Download: ML20149L749 (72) | |
Text
_
Officict Trcnscript ef Prseccdings 1
4
- g NUCLEAR REGULATORY COMMISSION 40?S/-208'2.
Title:
Advisory Committee on Reactor Safeguards 436th Meeting TRO4 (ACRS)
RETURN ORIGINAL TO BJWHITE Docket Number:
(not applicable) 7(s.;2s2s TaANxsi Location:
Rockville, Maryland i
i lO Date:
Friday, November 8,1996 i
9611150136 961108
- 08 PDR Work Order No.:
NRC-904 Pages 220-270 l
l l
n q'
1 1
i L V
i
/.h 150000 NEAL,R. GROSS AND CO., INC.
Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W.
i Washington, D.C. 20005 J202) 235g433 0(l ACRS ONe,C0py-Rei-
~0 "30
_ J'8 0" ".';8 00 miil0?
l i
f 4
O DI8 CLAIMER PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSION'S
)
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS i
NOVEMBER 8, 1996 I
1 The contents of this transcript of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards on NOVEMBER 8, 1996, as reported herein, is a record of the discussions recorded at the meeting held on the above date.
This transcript has not been reviewed, corrected and edited and it may contain inaccuracies.
I O
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1823 RHoDE ISLAND AVENUE, NW (202) 234-4433 WASHINGTON D.C. 20005 (202) 234-4433
.m._._
220 1
UNITED STATES OF AMERICA 2
NUCLEAR REGULATORY COMMISSION l
3
+++++
i 4
436TH MEETING 5
ADVISORY COMMITTEE'ON REACTOR SAFEGUARD (ACRS) l l
l 6
+ + +-+ +
l 7
FRIDAY 8
NOVEMBER 8, 1996 9
+++++
10 ROCKVILLE, MARYLAND 11
+++++
l 12 The Advisory Committee met at the Nuclear l
13 Regulatory Commission, Two White Flint North, Room T2B3, s
14 11545 Rockville Pike, at 8:30 a.m.,
Thomas S. Kress, i
15 Chairman, presiding.
i 16 COMMITTEE MEMBERS:
17 THOMAS S.
KRESS, Chairman 18
' ROBERT L. SEALE, Vice Chairman
'19 GEORGE E. APOSTOLAKIS 20 JOHN J. BARTON 21 IVAN CATTON 22 MARIO H. FONTANA l:
23 DON W. MILLER 24 DANA A.
POWERS
-25 WILLIAM J.
SHACK l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISt.AND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.~
-._..____-..__-...._....-_--m
_ _.. ~.. _
221 1
ACRS STAFF PRESENT:
2 JOHN T.
LARKINS, Executive Director 3
ROXANNE SUMMERS, Technical Secretary 4
SAM DURAISWAMY 4
5 RICHARD P.
SAVIO j
6 PAUL BOEHNERT 1
7 NOEL DUDLEY
-l 8
MADHAT.M. EL-ZEFTAWY 9
MICHAEL MARKLEY 10 AMARJIT SINGH 1
11 12 ALSO PRESENT:
O-AL CHAFFEE
]
'13
- 14-JIM TRAPP I
15 TOM CLEARY 16 PIERRE L'HEUREUX 17 JOHN MONROE 1
18 ERIC BENNER 19 20 21 22-1 23
?
24 25 i
i
~
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISt.AND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
222 1
A-G-E-N-D-A 2
Acenda Item Page
'g N_,,I 3
Opening Remarks by the ACRS Chairman 223 4
Nitrogen Bubble in the Reactor Coolant System 5
at the Haddam Neck Nuclear Power Plant 6
John Barton 224 7
Jim Trapp 226 8
9 ic i 11 12
-~
13 1
14 15 16 i
17 j
18 19 20 21 22 l
23 24
,,(,)
25 l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005 3701 (202) 234-4433
_ -.. -. ~.. _.
_. _ - -. ~. -
223 1
P-R-O-C-E-E-D-I-N-G-S 2
(10:08 a.m.)
O 1
3 CHAIRMAN KRESS:
The meeting will now.come to l
4 order.
l 5
This is the second day of the 436th meeting of 6
the Advisory Committee on Reactor Safeguards.
During 7
today's meeting, the Committee will consider the 8.
following:
One, future ACRS activities.
- Two, 9
reconciliation of ACRS comments and recommendations.
10 Three, report of the planning and procedures
.l 11 subcommittees.
Four, operating event involving nitrogen 12 bubble in-the reactor coolant system at the Haddam Neck 13 Nuclear Power Plant.
Five, annual ACRS report to l
14 Congress.
And six, proposed ACRS reports.
15 A portion.of today's meeting may be. closed to l
16 discuss organizational and personnel matters that relate 17 solely to the internal personnel rules and practices of L
18 this advisory committee, and matters the release of which l..
19' would constitute a clearly unwarranted invasion of
-20 personal privacy.
21 This meeting is being conducted in accordance l
l-l 22 with the provisions of the Federal Advisory Committee Act.
'23 Mr. Sam Duraiswamy is the designated Federal 24 official for the initial portion of the meeting.
l 25 We have received no written statements or NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 I
..--_....,.-.___...-__.._._.___._.....-_-_._m 224 l'
requests for time-to make oral statements'from members of l
2 the public regarding today's sessions.
A transcript of
[
O 3
portions of the meeting is being kept, and it is requested l
4 that the speakers use one of the microphones, identify l
r 5
themselves, and speak with sufficient' clarity and volume.'
l 6
so that they can be readily heard.
l 7
I think it's time to go ahead and start.
The l
8 next item on our agenda is the event at Haddam Neck, where
-l Li 9
a nitrogen bubble was introduced to the reactor coolant l
.10 system.
l 11 This is under the cognizance of John Barton,
'l I'
.12 our subcommittee chairman'.
So I'll turn it over to John.
13-MEMBER BARTON:
Thank you, Mr. Chairman.
The 14 purpose of the discussion for'the committee this morning 15 is to be brief on the findings and recommendations of the 16 NRC Augmented Inspection Team that investigated an event.
.17 at Haddam Neck on August 28, 1996, that resulted in the 18.
formation of a large. nitrogen bubble in the reactor 19 coolant system.
This briefing is for information only.
20 There is no ACRC' action anticipated at this time.
21-On August 28, with'the plan.to cold shutdown
{
'22 condition for refueling,'the plant operator inadvertently l
23 opened a wrong valve and the chemical volume control
{
24' system which resulted in nitrogen being admitted into the 25 reactor vessel.
This nitrogen gas continued to leak into NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
225 1
the reactor coolant system for a period of four days.
2
,7-The safety significance of this event is that i
\\
/
'~'
3 nitrogen intrusion into the reactor coolant system could 4
potentially result in gas binding and common mode failure 5
of DK heat remova) pumps and potentially interfere with 6
the ability of the steam generators to remove heat via 7
natural circulation cooling.
8 The NRC plans to issue a supplement to an 9
earlier information notice that discusses the issue of 10 undetected accumulation of gas in reactor coolant systems.
11 The key point noted in the IN is that the issue of gas 12 intrusion into reactor coolant systems and safety related 13 piping is a generic issue.
(, _\\
k s#
14 In the subject information notice, the NRC 15 notes that the issuance of eight previous notices on this 16 matter go back to 1988.
17 Representatives of the NRC will brief the 18 Committee on the particulars of this subject event.
The 19 AIT team leader will lead the briefing.
The staff has l
l 20 been asked to address the generic implications of this 21 event.
22 I believe you will also hear from the NRC 23 representatives about the lack of adequate performance on 24 the part of licensee in the areas of operations,
(,\\)
25 maintenance, engineering support, and plant management.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
'202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234 4433
. ~. _.
. - -. -. ~
. ~.. -.
.. ~. - -
.... -..- - - ~
-226
.1 Representatives of Haddam Neck plant are also 4
2 in attendance at this meeting.
Mr. Tom Cleary, licensing-V 3
representative,.and Mr. Pierre L'Jeureux, manager of i
4 nuclear. oversight, are in attendance, and are welcome to 5
make comments.after the briefing.
j 6
At this point, I would like to turn it over to i
7 the NRC, Mr. Al Chaffee, for opening comments.
.t i
8.
MR. CHAFFEE:
Based on that introduction that 9
you just provided, I don't really have to provide any i
10 opening comments.
I'll just turn it over.to Jim Trapp.
11
MR. TRAPP:
I was going to use the same-topic.
12.
I think you pretty much summarized everything I was going 13 to-say.
14 MR. TRAPP:
My name is Jim Trapp.
I work in 15 Region I.
I have been working in Region I now-for eight.
j 16 years.
I previously worked at Indian Point' Unit Two at
-17 Con Edison.
I was a senior reactor operator-there.
r 18 Previous to that, I was in nuclear engineering 19 school at Purdue for undergrad.
I went to Columbia as a l
L 20 graduate student in nuclear engineering.
21 The team consisted of four individuals from l
i 22 the region, four inspectors from the region and one from l
L 23 headquarters, an operator examiner from headquarters, John 24 Monroe'.
He is in attendance here too.
He conducted all 25 the interviews with the operators.
If you have any NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
I e...
i 227 1
questions on that, John can answer those for you.
2 If you recall, I was here about a year ago 3
this month, and discussed a BWR shutdown risk event at 4
Hope Creek, where a recirc pump discharge valve was left 5
open, which resulted in some boiling in the cooler.
6 l
So today I am going to give you the PWR 7
version.
Last year I gave you the BWR version.
So
'E there's some relevance between the two events.
9' Major events reviewed by this team, this AIT.
10 Our primary focus was on the undetected accumulation of i
11 the nitrogen gas in the reactor' vessel head.
There were 12 two related events which were inadvertent diversions of 13 reactor coolant system inventory from the RCS.
One being b
V 14 to a containment sump, and the other being back to the 15 IRWST.
I 16 One of these events occurred on August 22, 17 before the AIT arrived on site.
The other occurred on 18 September 4,'which was.while the AIT was in progress.
i 19 These events, a lot of the performance issues during these I
20 events were related to the primary focus of our
-21 inspection, so we kind of encompassed those events as 22 well.
I 23 There was another related event, decay heat 1
24 removal equipment failures.
There was a leak on an RHR
-25 heat exchanger inlet valve, which required that heat NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
228 1
exchanger to be isolated during this event.
There was 2
also a failure of a BRHR pump.
So both of these failures
(
i
~
3 would have abated the significance of the nitrogen gas in 4
the reactor vessel.
5 The discussion today, what I was going to do 6
was run through briefly, discuss the system configuration.
7 I have some slides for that, so we can all be on the same 8
page.
Then run through quickly a sequence of events.
9 I am going to discuss the safety significance 10 of these events, both from a reactor safety point of view 11 and from a human performance point of view.
There was 12 both those aspects to this inspection, and briefly discuss 13 the actions taken by the NRC.
,-g
}
r\\'
14 Now I am to jump over now to this diagram we s
15 have.
At the time of the event, the reactor was in cold 16 shutdown.
It had been shut down on July 22.
So it had 17 been shut down for a little over a month.
i 18 Licensee decided after the shutdown, the 19 shutdown was started to -- it was a tech spec required 20 shutdown for some design issues with some containment air 21 recirculation fans which on the 22nd caused them to shut 22 down.
Licensee subsequent to the shutdown decided to go 23 into refueling outage.
So during the event, which started 24 on August 28, the licensee was into refueling activities.
,-(,)
25 Major components.
Haddam Neck is a four loop NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
229 1
Westinghouse plant with loop stop valves.
What we are 2
depicting here is the reactor vessel.
I showed one of the 7-x
3 loops, one of the four loops, and showed the RHR system
/
4 and the vent header system.
5 RHR was in service right before this event.
6 The inlet to the RHR comes into the loop tube.
RHR flow j
7 then goes through the downcomer up through the core.
The 8
outlet is in loop one.
9 Loop stop valves at the time of the event were 10 closed, isolating this part of the loop, the steam 11 generators from the reactor vessel.
The steam generator 12 level secondary and primary were both full at the time of 13 the event.
(
)
V 14 The water level on the 28th was slightly below 15 the reactor vessel head when this thing got started.
As 16 you can see, I provided some elevations here.
You can see 17 that would put about four percent level in the 18 pressurizer.
19 I wanted to quickly run through the flow path 1
20 of the RHR system.
RHR system in this case, this is 21 pertinent to the event.
Suction from the hot leg through i
22 the RHR pumps at the time of the event, the A RHR pump was 23 operating on the 28th.
Both heat exchangers were in 24 service.
Then into the cold leg.
()
25 Status of the chemical volume control system.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 230 1
The nitrogen was being maintained in the volume control 2
tank here at 30 pounds psi.
The importance here is that
,3
\\
)
'~'
3 the nitrogen gas was being supplied by a regulator, so as 4
the nitrogen gas would be depleted, the regulator would 5
automatically make up nitrogen to the volume control tank.
6 You can see that these valves here would 7
become important to the event.
The 354 valve was open.
8 The 355 valve was closed.
Thus, isolating nitrogen gas 9
from entering this header here.
10 Charging pumps were not in service at the 11 time.
They were not operating at the time of the event.
12 However, there was one required boric acid flow path which 13 required this charging pump to be available.
You can see x-14 downstream here that there was a number of isolation 15 valves also closed.
16 other pertinent parts of this figure, level 17 indication that was in service, and this is critical.
The 18 level indication that the operators had would be a hot and 19 cold pressurizer level.
They had this cavity level 20 indicator.
21 The cavity level indicator is a temporarily 22 hooked up system.
It's only put in service during 23 refueling outages.
The intent here is to measure from the 24 bottom of the hotleg, which is at 170 inches elevation, A(_)
25 all the way up to the top of the refueling cavity full, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
i 231 1
which is 600 inches.
I 2
So you can sense this as sort of being just a O
3 pressure transmitter down here for the variable leg.
As 4'
the cavity level indicator, it indicates in the control s'
room.
Then there's a reference leg, a left reference 6
level transmitter coming off the vent header hose.
7 A key part of this cavity level indicator is 8
the way it was set up at the time of the event.
Is the 9
cavity level indicator indicated whatever the highest
)
i 10 elevation of water was in the reactor coolant system.
11 So ve'll see as water level'was decreased-in 12 the vessel, pushed up in the pressurizer, what cavity j
1 13 levels indicating would be the highest level, highest
~14 elevation in the pressurizer.
That's key to this event.
15 The' reactor vessel level indication system at 16 Haddam Neck is a CE design, with reference thermocouples.
17 So it's a discrete reactor vessel level indication.
It's l
4 18 not the Westinghouse reactor vessel level indication.
19 That was disconnected for refueling activities one day i
20' prior to the event.
It was disconnected on the 27th.
The 21:
event started on the 28th.
22 Core exit thermocouples were also 23 disconnected.
When you take the missile blocks off, the 24 missile shield, you need to disconnect, at Haddam Neck, l) 25 you disconnect the thermocouples and you disconnect the
- ?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
- ~.
232 i
1 reactor vessel level. indication system.
.i 2
Another key aspect to the system here we'll 3
see a somewhat unique design to Haddam Neck, is the i
4 pressurizer surge line which is right here.
It's on 5
different loops, but it's on loop four.
The pressurizer
[
6 surge line and the RHR suction line are at the same exact l
.7 elevation.
Both 10 inch lines, same elevation.
t i
8 So as you can see, from this drawing as level 9
would decrease, the pressurizer, if you have a vacuum in j
10 the pressurizer, the pressurizer level would maintain full i
11 until'you hit the point.
You know, you would lose RHR j
12 suction'before you are going to drop the pressurizer l
1 i
13 level.
That would be key to this event.
l 14 Vent header system, you'll see that this was l
15 also important during this event.
This is a temporarily
\\
l 16 hooked up system.
Its intent is to move radioactive gases 17 from the reactor coolant system.
The vacuum is about 10 18 inches of vacuum at the head.
It's also hooked up to the
'19 pressurizer.
20
.This system was in service.
There was a l
J 21 number of deficiencies with this system which contributed 22 to the event.
j r.
23 A brief description of what happened.
24 Operators went out to realign another boric acid flow 25 path.
The flow path that was currently in service using NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l
_(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4 33 I
_m
_ _ -. ~ _
233 1
this. charging pump, the diesel for that charging pump was 2
going to be tested.
The licensee declares the charging 3
pump inoperable when the diesel is being tested.
They 4
were lining up a new boric acid flow path, a new required 5-boric acid flow path.
In doing so, the operator 6
inadvertently opened the 355 valve.
7
-Inadvertently opened being that this valve 8
should not have been opened.
The 354 was opened.
What 9
this allowed nitrogen gas to do is leak through the 354 10 valve through the 355 valve into the charging system.
11 These valves closed, leaking by.
You'll see that theme 12 repeated.
RHR, through the regenerative heat' exchanges, 13 through the bypass.
There's a bypass valve open here.
' /~'s
%s/
14 Leaked by this valve or this valve, it's hard to tell
\\
15 which.
We assumed it was the cold leg valve, and into the i
R16 reactor vessel.
17 MEMBER CATTON:
How many valves in series 18 leaked?
19 MR. TRAPP:
Well, really you'll see later on L
20 in the event.
After they closed the 355 valve, the key 21 here is that the 355 valve continued to leak by.
So you 22 could count that.
So it was one -- at least three.
23 MEMBER CATTON:
Three in series.
L
- 24 MR. TRAPP:
Yes.
s g_,)
25 CRAIRMAN KRESS:
Thise are liquid valves?
i=
l-NEAL R. GROSS j.
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4413 WASHINGTON, D C. 20005-3701 (202) 234-4433 g
i
-. -. _. - _. - ~
. -. ~
J 234 1
MR. TRAPP:
That's correct.
Some of them, a 2
key here too is some are flow control valves.
So it
)
C 3'
wouldn't be the intent for them to be isolation valves.
4 They are not containment isolation valves.
So there was j
5 some reason for the -- you know, it wouldn't be expected l
6 that these valves would hold gas.
l 7
So now we have got the nitrogen going into the 8
-- the nitrogen gas.
Here's the charging line that I 9
showed there into the cold leg.
The nitrogen gas now 10 could enter the reactor vessel.
11 If you remember, downstream here, the loop 12 stop valves were closed, so it couldn't go the steam 13 generator way.
It had to go the reactor vessel way.
14 Plus, RHR flow would induce it to go into the reactor 15 vessel.
16 There are some flow. bypass holes in the lower 17 internals here, which would allow the nitrogen gas as it 18 accumulates here to accumulate in the reactor vessel head.
19 That was the start, the flow path for this event.
20 I kind of mentioned it before, level 21 indication response.
This is important to the event.
As l
l 22 the nitrogen gas came in, bubbles started to grow here.
23 Obviously water would decrease in the reactor vessel.
1 24-As the water decreased in the reactor vessel, i
t 25 water level decreased in the reactor vessel, it's pushed NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
235 1
up the surge line into the pressurizer.
t 2
So what the operator saw was an increase in O
3 pressurizer level, cold and hot calibrated.
.They also saw j
4 an increase in cavity level because what it was measuring 5
is the highest level in the reactor coolant system.
6-So'as level decreased in the reactor vessel, l
i 7
operators really didn't have any positive indication of l
8.
what vessel was it.
I don't want to use those words, but 9
it's kind of a TMI scenario, is reactor-vessel level isn't j
10 where you think it is based on pressurizer level.
11 The safety significance of this event.
As the l
12 level decreased, safety significance being that if level 13 had decreased to the top of the hot leg at the RHR flow ~
>O b
14 that was about 2,200 gpm at the time.
If, this is an if, l
15 if level had decreased to that level, RHR pumps would be 16-cavitated.
17 At the time of this event, of coursa the B 18 pump, unbeknownst to the operators at the time, the B pump 19 had previously seized.
If you had induced the nitrogen 20 into the RHR suction line, the A pump also could have 21 failed.
There's a number of issues involved there.
22 So RHR cavitation was the principle concern.
I j
23 The charging pumps, now you can see that the 355 valve --
J l
24 the operators closed the 355 valve after it was j
25 inadvertently opened, reclosed the 355 valve.
It leaked NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
236 1
by for four days, allowing the nitrogen in the CVCS system 2
to continue to fill with CVCS system so that the nitrogen p
\\
)
'~'
3 gas would have rendered the CVCS system inoperable as 4
well.
The charging pumps would not have operated with 5
nitrogen gas.
6 MEMBER SHACK:
When did they identify that 7
they should have closed the 355 valve?
8 MR. TR4PP-The 355 valve was identified very 9
shortly after by the operators.
As the nitrogen gas 10 initially blew the CVCS system water into the reactc r 11 coolant system, the level in the pressurizer went up about 12 two percent.
The operators picked right up on that, and 13 instructed the operator in the field to reclose that
, ~.,
e i
ks/
14 valve.
So that was a quick catch.
15 Unfortunately, when they reclosed the valve, 16 it didn't completely go closed, so you continued to leak 17 nitrogen into the same flow path.
l 18 MEMBER CATTON:
Is that the fault of the 19 operator or fault of the valve?
t 20 MR. TRAPP:
That's a hard question.
There was l
r 21 a number of valves.
You will see some of the other events 22' are also manual valves which the operators were unable to 23 get fully seated.
So I think it's a combination.
24 We saw that if you went out there with a valve
,,\\
(w.-)
25 wheel wrench and then you cranked down real real hard on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
1 237 1
1 the. valves, sometimes you could get them to close.
2' MEMBER CATTON:
You don't want to break them s.
'\\
3:
either.'
4 MR. TRAPP:
It's always a tradeoff, right?
5 How much to put on it.
If the valve isn't well.
6 maintained,.you.'can have the problem where the operators-1 7
can't close them.
So I think it's a combination of the
-8 two.
9 That's actually one of our contributing 10 causes, was the material condition of-manual isolation 11 valves.
12 Other key aspects here would be if the 13 nitrogen. gas had gotten entrained into the RHR system,.
-G 14 upon repair of the B RHR pump, the licensee attempted to 15 vent the B RHR pump and get it restarted.
The vents j
16 aren't at an optimal location.
It's very difficult to i
17 vent the RHR pumps at Haddam Neck.
The pump.actually j
.18 rotated for.about a minute and a half.-without catching I
.19 suction.
It_just sort of sat there and churned.
1 20 So the team was concerned that had the 21 nitrogen gas gotten entrained in the RHR system, the 22 operators.might have not been able to easily vent and
'23 return the RHR pump to service.
24 The other concern the team had would be one of
'25 the' mitigating strategies is to open the loop stop valves NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 238
(
1 and now use the steam generators.
There was a concern 2
that now the nitrogen gas could also have been induced
(
3 into the U tubes of the steam generators, and somehow 4'
adversely affected the ability to use natural circulation.
5 I would now like to sort of run through the 6
sequence of events quickly.
As stated before, the i
7 operators incorrectly opened the 355 valve, and it l
8 resulted in injection of water and nitrogen into the i
The operators caught this very 10 quickly, closed the valve, didn't see nitrogen continue to 11 leak from the volume controlltank into the reactor coolant-12 system.
13 The other key'here is, and I wanted to state, 14 is as the nitrogen gas would leak into the~ reactor coolant 15 system.
You'also had the -- it goes in is the CVCS 16 system, it goes out would be the spent header system.
17 We'll see from the event that a significant amount of the 18 gas that leaked in, the vent header system remov'ed.
19 However, there was a number of issues with the 20 vent header system, wrong diameter, hoses.
This is a 21 temporary system.
It's not the hydrogen, the required 22 hydrogen vent system.
It's a temporary system just hooked i
23 up for refueling.
24 A number of design deficiencies such as loop
()
25 seals full of water, there was dropout tanks which didn't NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W, (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
_. ~ _..
239 1
have adequate water level in them.
There was Tygon hoses 2.
that were held up with duct tape that had fallen off and.
Id 3
kinked.
So this vent header system was significantly 4
degraded at the time of the event, which didn't allow all 5
the nitrogen that was going in to go out, which allowed 6
the nitrogen to accumulate in the head.
7 As I stated before, the nitrogen gas 8
accumulated into the head, displacing water.
From the 9
traces, you could see that the cavity' level indication,
'10 cavity level continued to increase.
Pressurizer level 11 also continued to increase for approximately four days.
12 The magnitude of the increase and pressurizer' level was 13 about 10= percent pressurizer level.
14 On August 29th, in preparation to remove the 15 Cono seals from the reactor vessel, this is a normal
-16 refuelling evolution, the licensee drained approximately 17 5,000 gallons of water out of the reactor coolant system.
18 The intent was to drain the vessel level down to about the i
19 point of the reactor vessel flange.
20 Later that day, there were some issues of 21 working over the Labor Day weekend, so they decided to 22 fill back up and wait until Tuesday after Labor-Day to l
l 23-continue with refueling activities.
The operators then l'
24 threw about 1,000 gallons of water back into the reactor
()
25 coolant system.
This was key, because as they took out NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.~
240 1-5,000, they put in 1,000.
Cavity level and pressurizer O--
level indication were almost back to.where they had
.2 v
3 started.
4 So at that' point on the 29th, the size of the i
s' bubble was approximately 4,000 gallons.
So it was a big 6
bubble.
7 The bubble continued to grow on September 1 to 8
about 39 inches below the reactor vessel flange, which'is 9
about two and a half feet'above the point of. cavitation,
-10 above the hot leg piping.
It would be about seven feet 11 above the top of the fuel assembly.
12 On August 31, looking at the traces, it
-l 13 appeared that the size,-the magnitude of this bubble began 14 to decrease.
The reason for that was is now you were 15 building up a pressure in the reactor vessel..As the 16 bubble increased, obviously the pressure of that bubble 17 had to increase,.which slowed down the reactor coolant-18-system pressure increase, which' slowed down the 30 pounds
-19 driving force into the reactor coolant system.
It would 20 also help the vent header remove the gas.
So as pressure 21 increased in the reactor vessel, the bubble growth and the 22 cavity level decrease started to even off.
23 I guess the team sort of concluded it was 24 fortuitous that that was'the case.
With the condition of
()
25 the' vent header system, and with the valve leaking by, had NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W..
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
241 1
that valve leaked by slightly more or had the vent header 2
system been slightly worse, certainly that bubble growth 3
could have grown to a magnitude where you would have 4
certainly cavitated the pumps.
5 I think the total nitrogen flow through this 6
system was estimated to be 14,000. cubic feet of nitrogen, 7
which would almost fill a building 30 feet by 30 feet, 8
three stories high.
So a lot of nitrogen went through the 9
system.
10 MEMBER CATTON:
Don't they monitor their 11 nitrogen source?
12 MR. TRAPP:
Well, it's funny, and we'll get to 13 that.
But on the first -- the reason they actually found Ak-)-
14 this wasn't because of reactor coolant system problems.
15 It was because hey, we're running out of nitrogen problem.
16 That's really how this whole thing got discovered.
'17 on the 28th, they realized, which was early in 18 the event, they realized they had a nitrogen problem.
19 There was some documentation of some evidence that they 20 went out and looked at the I think it was the CST, because j
i 21 they keep a nitrogen blanket on that, thinking that's 22 where it was leaking.
23 Operators also felt that-the source of the 24 increase, there was some indications on the 29th that the
()
25 operators realized that they had an increase in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS j,
1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
242 1
pressurizer level without any additions and were confused 2
by that.
3 What the operators thought at the time was is 4
the. loop stop valves were what was leaking by, and water 5
from'the steam generators was somehow entering the reactor
]
6 coolant system.
They went down and tightened up on the 7
loop stop valves.
At the same time, they closed the i
8 relief valve on this side of the reactor coolant system to 4[
9 relieve inventory.
You know, this is a solid system now.
10 So you need a relief valve.
11 For some reason, they also isolated those j
1 12 relief valves.
That was another finding of the team, that 13 the relief valves were isolated for 13 hours1.50463e-4 days <br />0.00361 hours <br />2.149471e-5 weeks <br />4.9465e-6 months <br />.
We felt 3f
'1 14 that was inappropriate, j
. 15 So the 29th was key.
You kind of get a sense i
16 there for the size of the magnitude of the bubble.
The 17 31st, the bubble began to reach an equilibrium.
on
- 18 September 1, now I get to that point.
On September 1, the 4
19 reactor operators went out, trying to investigate the high 20 uses of nitrogen.
21 When they did that, the nitrogen input now was 1
22 cut off.
They isolated the regulator to the VCT, so now 23 nitrogen wasn't coming in here.
The vent header continued 24 to extract nitrogen.
25 The reactor vessel level started to increase.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234 4 33
243 1
Now the pressurizer level dropped out of the' bottom, as 1
i 2
you begin to refill the reactor vessel from the
[
3 pressurizer.
4 It actually dropped out of the bottom six L
)
5 times.
It tookLsix makeups to get the 4,000 gallons back I
6 into the reactor coolant system.
7 At the time, the first responsibility of the 8
operators of course was when they.saw pressurizer decrease 9
out of the bottom, is they thought they had a leak in the l
10 reactor coolant system and they took appropriate actions 11 to go'out and investigate those leaks.
12 Later on, actually then they opened up the l
13 nitrogen to the VCT because they wanted to put the system l (_/
14 back in configuration they had previous to the level 15 decreasing.
It was not until later on,-a matter of an l
16 hour1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> or so, that they actually discovered that this was 17 the issue.
18 The other two events that we looked at were an i
19 August 22nd event.
Like I said, this was before the AIT l
20 arrived on site.
The operators were attempting to place 1
21 the RHR purification pump back :bi service.
This is a 22 system used to filter the reactor coolant system when the j
23-plant is :bi shutdown condition.
It goes through the 24 chemical volume control system, de mins.
They were (k
25 placing that system back in service.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
=..
- -.. - ~.
244 i
1 The RHR purification pump has two of the many
[
l 2
suctions to that pump, one is from the IRWST, the other is
~
3 from the RHR system.
There's two suctions.
Operators l
4 inadvertently went to the wrong step in the procedure.
i 5
They opened up both of those suctions at the same time.
e 6
This allowed the RHR pump which was operating, just to i
7 divert flow right back to the IRWST.
i 8
MEMBER POWERS:
How did the operators go t
9 inadvertently to the wrong step in the procedure?
j 10 MR. TRAPP:
What happened, there were many t
11 reasons, at least.two reasons that we cited in our report.
I 12 One is is that the operators didn't'get an effective pre-f 13' job briefing.
But a key thing here is the procedure 14 quality.
I 15 Each step in the procedure had an initial l
l 16' sign-off as you go down the row.
iell, the taking the RHR i
17 purification out of service had all the steps with a sign-l
~
18 off.
They were all initialed down the row.
Then there 19 were two steps that didn't have any signoff.
The third l
i 20 step, you know, had a blank, but-the operator said okay, j
21 we'll go to the step with a blank.
That wasn't the right i
t 22 step.
They should have been a couple steps above.
So
-23 that was a-contributing cause.
24 On September 4, the team was on site.
At this
)
25' point, you have to put yourself in where we were on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
245 1
September 4.
We had an RHR heat exchanger'out of service 2
because of the inlet valve leak.
We have an RHR pump that O
3 we knew was seized.
The B pump was seized. JWe had one l'
4 RHR pump available which was operating.
5 It was also on the fourth that we're still 6
trying to really establish what reactor vessel level was.
-7 So at the time, the licensee couldn't really put in place L
8 a stop work order on anything that could affect RCS 9
inventory or the RHR, operating RHR system.
10 This activity which was to perform a PM on 11 some RHR pump motor operated spray valves, slipped through l
12 the net, the safety net that they had put up around the l
13 RHR system.
The plant maintenance personnel went.in to 14 conduct this PM on the motor operated valves,' which 15 required opening the motor operated valves, cracking them 16 open off their closed seats.
17 The valves are parallel valves going into a 18 manual isolation valve.
Again, the same theme.
Manual 19 isolation valve was closed.
However, it leaked by.
20 When the motor operated valves were taken off I
i 21 their closed seat, this allowed RHR now to flow through l
l 22 the two open motor operated valves, through the. leaking 23 manual isolation valve, and into the containment sump.
24 Again, in both these cases, in the first case,
()
25 they only lost about 500 gallons of water before the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.. _ _ _ _. _ -. - _.. _ =..
1 246 1
l 1
operators caught it.
In the second case, about 300 2
gallons of water.
So in both cases, operators 3
expeditiously terminated these events.
l 4
However, the team was concerned that a lot of l
5 the issues that were involved with these evente also were j
6 involved', especially in attitude and' decision making for 7
the nitrogen.
i i
0 In addition, had these events occurred at'the i
9 time when the minimum level was in the reactor vessel, it' l
10 could have been a more serious event.
l l
11 The number of gallons in the reactor coolant j
i 12' system on September 1, to get to cavitation, was about 13 roughly 2,000 gallons.
So had you had one of these l
i
~
14 diversion events, you're losing 500.
You had about 2.,000 l
15 left'.
You are getting close.
l 16 So that's why we review those events.
That's i
17 why those are a concern..
18 MEMBER POWERS:
Did your team look at things j
-19 like, liability, human reliability analysis to see what
)
20 probability it might be for the operators to fail to catch
}
21 these?
j
-22 MR. TRAPP:
No.
{
l i
l' 23 MEMBER POWERS:
It might be interesting to try 24 to do:that sometime,.given the time they have and the 25 amount of information they have, if it's highly probable i
NEAL R. GROSS i
COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
247 1-that they will catch these things or --
2 MR. TRAPP:
There are control room enunciators 3
on the cavity level which~ enunciate on high level.
So I 4
know that all kind of goes into the liability issue.
5 MEMBER POWERS:
Yes.
There.are a lot of 6:
things to take into account.
7 MR. TRAPP:
Decay heat removal, equipment 8-failures.
On August 31, a plant operator identified a 9
weeping small leak in an RHR heat exchanger valve.
This 10 valve then was closed to isolate the leak.
If you closed 11 the valve, this type of valve, it actually isolated the 12 leak.
13 On September 1, the licensee was attempting to 14 place the B.RHR pump in service.
Previously the'A RHR 15 pump had been in service.
With the A heat exchanger out 16 of service due to-the inlet valve leaking, they wanted to 17 run the B' heat exchanger with the B pump.
So they were 18 swapping over to the B pump and tried to start the pump
.19 two' times.
Both times the pump wouldn't start.
l 20 Later on, it was found out that that pump had L
i l
j 21 seized back on August 19th.
So the pump was not available j
l-1 l
22 after the last time it operated.
i 23 Safety significance of this event.
With the 24 loop stop valves, and I want to be real explicit here.
J
()
25 With the loop stop valves closed, timed to 200 degrees, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
\\
..~
.248 t
1 for that condition with the vessel level 39 inches below l
2 the flange, was approximately 52 minutes if RHR system was O
3 lost.
That's key.
The RHR system wasn't lost.
So we 4
weren't 52 minutes to a larger event.
It was if RHR S
system was lost.
6 If the loop stop valves are open, now you get 7
the addition of all the inventory.
That part of the RCS 8
and the time would increase by about three times.
9 Reactor vessel level decreased to about two 10 and a half feet above the point of RHR pump cavitation, 11 which is roughly 2,000 gallons of additional inventorys if 12 that-was removed, you would have been to the point of 13 cavitation.
It was about seven feet above~the top of the O
\\sI
-14 fuel, not the active fuel.
15 If the bubble had continued to grow,-and like 16 we said, it was fortuitous that it didn't, nitrogen gas 17 could have been entrained into the reactor coolant system.
18 You could have challenged the operating RHR pump.
19 The boric acid flow path required boric acid 20 flow path was obviously compromised from the gas.
You had 21' a common mode failure here, is that.both lost the chemical l
22 volume control system and would have lost the RHR system i
~
23 at the same time.
24.
We discussed a little bit that the gas, if you O
25 had opened the loop stop valves, there was a potential for
.q,j NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 2344433 l
249 1
the gas entering the steam generator U tubes.
2 Subsequently I guess, that has been looked at by NRR and
,s
/
i 3
the licensee and others.
They felt that if the RCS had 4
pressurized, that that nitrogen gas in the steam 5
generators could have compressed to the point where you 6
would have the ability to natural circulation.
You would 7
also have reflux cooling from the steam generators.
i 8
Had the RHR system been lost in not being able 9
to recover, the first mitigation strategy would be to use 10 the steam generators.
Place those in service.
You could 11 also, there's a lineup that at Haddam Neck, it's kind of a 12 unique lineup.
You can use the LPSI pumps in a closed 13 loop going back to the IRWST.
You can let down through
- k. '
14 the RHR system, which provides you a closed path for an 15 additional cooling path, which would have been available.
16 Now we said if the operators hadn't terminated 17 those two vents, we thought that was significant, could 18 have resulted in cavitation of the RHR pumps.
19 I discussed a little bit the venting, the B 20 pump, the venting wasn't really too successful after 21 maintenance.
The reactor coolant pumps really weren't 22 available for service.
So for circulation wasn't really 23 an option at the time of this event.
FRHR was lost.
24 Safety significance.
Go to slide 12 now.
The (y( ).
25 team identified a host of contributing causes for this l
NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l l
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
250 1
event.
Maybe the first and foremco-would be' sort of a
'l 2
lack of' questioning attitude by.the operators.
They had c
3 significant nitrogen usage.
There was. explanation for:
I 4
some of the nitrogen usage in the steam generators were i
5 being drained and nitrogen gas was being introduced.
So 6
you could in your head probably qualify'some of that,'but j
'7 you couldn't qualify the magnitudes that were occurring at 8
the time.
l i
9 Then we felt. that there was a lack of a 10 questioning attitude by the operators on the level j
i 11 increase.
Both in the fact that the imbalance that l
12 occurred on the 29th when they took out the_'five and put 13 in the pne, and a lack of questioning attitude on the fact j
! - /~T '
i 14 that pressurizer lwel continued to' increase for four days 15 without'what we thought were appropriate corrective
)
1 16 actions.
17 The poor decision making.
There.were a number l
18 of examples of that.
Specifically, the work on the RHR 19 pump spray valves.
That shouldn't have been performed.
20 Operators were well aware that RHR was in a challenging 21 position at the time.
That activity should not have r
l 22 occurred.
Closing the loop ut.op pressure isolation valves 23 was also a poor decision.
24 We felt there was a poor implementation of 25 generic information.
There was an information notice on a u
4 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2 % 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
251 1
Sequoyah event which is somewhat similar to what happened 2
at Haddam Neck.
The licensee had done a really good job 7-s' 3
of evaluating that information, provided a number of 4
recommendations.
In fact, there was supposed to be requal 5
training done on that information right before this 6
outage, but the outage as I said, started early with this 7
core shutdown from the tech specs.
So that requal 8
training was not performed.
I felt that contributed.
9 Also, a number of the corrective actions that 10 were recommended from that didn't seem to be really 11 effectively implemented.
12 There were a number of issues of poor 13 equipment condition.
The vent header, I stated before,
,a
\\
i_ /
14 the vent header system which is not safety related but 15 really was in a significantly degraded condition during 16 the event, and there was just several weakened valves, 17 manual isolation valves that we thought also indicated a 18 poor equipment condition.
19 There was, and this is kind of the same, back 20 in Hope Creek days, that event that I discussed with you, 21 in vessel inside the shroud, instrumentation was a real 22 issue.
The operators didn't really know what the 23 temperature was inside the shroud.
They knew what it was 24 RHR.
(
)
25 In this case, again, it's an instrumentation NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
I
252 1
issue.
They didn't have RVLIS available.
They didn't 2
have CETs available.
Their level indication looked fine.
7-(
)
3 They thought they had a full vessel.
So again, 4
instrumentation was a significant contributor to this 5
event.
6 Procedure quality, we thought for these 7
shutdown events anyway, was poor.
A numbered of required 8
procedures by tech specs and by the Reg Guide 1.3.3. such 9
as draining down the reactor coolant system weren't really 10 available.
What the operators did is they have an 11 administrative procedure where they can write step lists 12 for activities and evolutions that are not covered by 13 procedure.
They did this for the drain down and other
'a
)
K/
14 activities, such as venting the CVCS system.
The quality 15 of these step lists doesn't go through the same review and 16 approval process as a real procedure.
The team iden ified r
17 a number of what we felt were deficiencies for those 18 procedures which were also contributing causes.
19 We mentioned once before on the step list, 20 without the proper signoffs, that again contributed 21 significantly to that event.
l 22 There was also a failure to implement 23 procedures.
The boric acid alignment, which opened the 24 355 valve, operators weren't even aware that the guy was
/T l
y )
25 out there was supposed to be opening valves.
So the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS j
1323 RHODE ISLAND AVE., N.W.
j (202) 2344433 WASHNGTON, D.C. 20005-3701 (202) 234-4433 i
253 1
procedure indicates that you are supposed to call the 2
control room before you manipulate a valve.
That was not 73
('")
3 done.
4 There were other examples, such as realigning 5
the RHR purification pump where the procedures that they 6
had were out followed.
7 Inadequate training.
Discussed before.
But 8
the training that was going to be provided on what to 9
expect, what indications to expect with a nitrogen bubble 10 in the reactor vessel wasn't performed.
We thought that 11 contributed.
)
12 There's an administrative requirement at 13 Haddam Neck to perform pre-job briefings for evolutions,
,~ \\
kJ 14 such as aligning the boric acid flow system and placing 15 RHR purification system and service.
Neither of these 16 pre-job briefings were performed.
Both activities were 17 being performed by an operator for the first time, a non-18 licensed operator for the first time.
The team felt that 19 it had those expectations of the more senior operators 20 been transferred to the junior operators, that these 21 events could have been averted.
22 We thought there was a general weak use of 23 technical support and a weak technical support provided to 24 the operators.
A number of issues, such as the increasing g~3
(
)
25 level of nitrogen gas usage, that they thought the loop NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 254 1
stop valves were leaking.
They went in and closed down on fy 2
those.
The vent header condition, all these type of
(
)
3 issues we felt were prime issues to get technical support 4
involved.
Operators failed to do that.
5 There were a number of other issues where we 6
thought technical support should have been a little more 7
proactive and get in there and help the operators, and 8
that we felt that was also lacking.
9 There's some issues of inappropriate planning 10 and scheduling.
Delaying the outage over the weekend, 11 this was sort of a relatively higher risk period during 12 the outage.
Had that activity continued, when you broke 13 the cono seals, and in this case anyway, you would have
\\
1
'j 14 vented the vessel.
15 Then the planning and scheduling activities 16 that resulted in the RHR spray 7alves opening was also 17 inappropriate.
18 We looked at the event response slide.
This 19 was the activities taken by the licensee following this 20 event.
Licensee had available instrumentation and 21 procedures to go in and take manual rivless readings.
We 22 thought that that should have been done shortly after the i
23 gas bubble was discovered.tn the reactor vessel.
It took 24 two days to go in and take the manual readings.
AC 25 We felt that they could have done more with l
4 l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
255 1
the operating RHR pump.
We are down to one RHR pump now.
2 Really nothing significantly was done for five days
(
)
3 following the loss of the RHR pump and the heat exchanger 4
to provide additional surveillance of that operating RHR 5
pump.
6 The RCP and seal injection was not aligned for 7
several days following the event.
This would have been 8
advantageous had nitrogen gas accumulated for the steam 9
generators.
That could have vented the steam generator U 10 tubes.
11 Special tests to establish vessel level and 12 vent header operation were not performed for four days.
13 They had some special tests trying to establish what
\\_)
14 actual vessel level was.
These were delayed for a number 15 of reasons.
16 Their instrument response team was not really 17 placed in service and ongoing for four days.
Our team was 18 already there at the time doing our activities.
It tended 19 to be a delay.
20 Finally, it took 25 days to restore the B RHR 21 pump to operable status.
We thought there was a number of 22 avoidable delays in doing that maintenance activity.
The 23 RHR pump out of service would be under Mode 1 conditions 24 would be a 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> LCO.
(
)
25 The NRC actions.
We have completed our NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
-- - -.. - -.- -=._-..- -.. -. - ~
-_ -.+.
256
.i l'
augmented' inspection team.
We'have conducted the exit j
2 meeting.
That activity is complete.
There's still some 3
enforcement activities to proceed from this event.
The 4
information notice.
5 There's an information' notice.that's in draft 6
and being reviewed by the staff.
Additional generic 7
actions are being considered.
8 That kind of completes my presentation.
Are 9
there any questions?
f 10 MEMBER BARTON:
That was a good briefing, Jim.
11 Thank you.
1 J
12 Does the licensee have any comments?
i 13 MR. L'JEUREUX:
Yes.
I would like to say-l 14 something.
My name is Pierre L'Jeureux.
I am a manager 15 of the oversight group at Haddam Neck.
I was also the i
'16 team leader of the independent response team, the utility 17 team that looked into essentially the same events that.
18 Jim's group looked into.
19 Just a couple of things that occurred to me 20 that may be of interest to the committee.
There was a i
21 question about nitrogen monitoring.
The nitrogen level, 22 and we have a liquid nitrogen storage facility which is an 23 evaporative type.
It evaporates into the gas, which is 4
24 used in the plant.
That's logged on an eight-hour basis.
()
25 However, there was no trending of that information, which NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
257 1
probably would have' helped.
2.
Also, nitrogen is used in other activities at n
v 3
the plant above and beyond keeping a pressure on the 4
control tank at 30 pounds.
Several activities using 5
nitrogen for other things, such as draining and refilling 6
the' secondary side of steam generators, had occurred.
A 7
couple evolutions had occurred during this time period.
8 So even though there were noted changes in the l
1 9
liquid level in the facility, on the surface, that could j
10 be attributed to these other activities going on.
11 However, if they were trended, we believe that the l
12 operators would have been able to denote the negative 13 tren'd.
.l 14 Somebody asked a question about the 355 valve, i
]
15 whether or not the valve failed itself or was:it lack of l
16 proper closure by the personnel.
We don't know.
It was-l 17 our intention to'take the valve out and inspect it.
With i
18 the decision to decommission the plant, that may or may 19 not happen.
20 The operator involved, we believe, put enough 21 pressure on the handwheel in order to get reasonable 22 closure.
We believe this valve probably hasn't been I.
l' 23 operated in 28 years.
It's'the valve that connects strong 24 boric acid to the direct suction of the charging pumps, 1 -A
(,/
25 which there are lots of' ways of getting boric acid into 2
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
]
1323 RHODE ISLAND AVE., N.W.
i (202) 234-4433 WASHINGTON, D.C. 2000>3701 (202) 234-4433 i
258 1
the system.
That's probably the least desirable.
So x
2 chances are, it hasn't been operated in 28 years.
It i
3 could just be a very stiff operating valve that didn't 4
fully close or some debris that laid up against the side 5
of the valve seat, that when it was open and reshut, o
didn't fully allow closure.
7 Jim had mentioned that the 52 minutes existed 8
or would have existed to 200 degrees in the vessel if in 9
fact we had lost RHR.
I did want to point out that our 10 shutdown risk program does provide for a lot of equipment 11 being made available in the event of RHR, loss of RHR.
12 For example, we had reactor coolant system 13 integrity that could be easily established within several
/
u 14 minutes.
There was only a couple activities that 15 operators had to perform in terms of isolating certain 16 components that would have established pressure integrity 17 of the RCS, 18 Jim pointed out the fact that when the 19 decision was made on the weekend not to take the vessel l
20 apart, water level was raised in the pressurizer, and also i
21 they deferred the decision to disassemble the Cono seals, 22 which would have permanently prevented a closure in a 23 quick fashion.
We still had containment integrity 24 available.
/~N,
(, )
25 Jim had also mentioned the fact that if the
)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20MS-3701 (202) 234-4433
259
-j 1
vent header became less able to remove the gas or if this 2
valve would have been opened a little more, that the 3
bubble could have increased.
We agree with that.
j 4
There is a consideration that if in fact that i
5 would happen, the change in pressurizer level would have
\\
l 6
been substantial, j
i 7
We did~know, or the operators did recognize i
13 that after they had established that level or re-l 9
established the level after the drain down, that they had 10 noticed that the level started to creep up some.
They j
i 11 were looking at areas of leakage.
That's why they shut or 12 verified shut the loop isolation valves.
That is why they 13 shut the overpressure-relief isolation valves.
14 But if in fact the bubble'had gotten f
15 substantially bigger, the resulting increasing pressure-16 level would have been even more obvious.
We'are
'17 speculating, but we would hope that that would be 18 sufficient indication for a more intense look for what may 19 be causing the problem.
20 MEMBER CATTON:
Where does the surge line i
L 21 connect to the hot leg?
j l
22_
MR. TRAPP:
Right there.
The center line is 1
l L
23-185 inches.
1 24 MEMBER CATTON:
So you couldn't have emptied
'()
25 it below that.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
I
-._.__m
....m._
.?
~260
-l l
t 1
MR. TRAPP:
That's correct.
But given that' I
'. O 2.
you cavitate'it,-198.
The level got to 198.
l
.\\ r
+
3:
MEMBER CATTON: ' But you.have. pressure.
But 4
you have an over pressure on'the water.
'5 MR. L'JEUREUX:
I think what happens is you j
i 6
begin to get air entrainment when you'get down to 198 7
inches,_even though you haven't actually uncovered the
~!
.8 suction line to the RHR.
9 MEMBER CATTON:
That's a'10' inch. pressurizer i
10 surge line.
Oh, and you cavitate when you get to the top 11' of the hotleg?
12 MR. TRAPP:
Exactly, at that flow.
You know, 13 it's a flow dependant curve.
L 14 MEMBER CATTON:
No.
I understand.
Even
-15 though;you'have pressure in the head?
Nitrogen pressure l.
16 in'the head?
17 MR. TRAPP:,That probably wasn't factored into 18 it.
Well, it's not a temperature cavitation thing.
I i
19 think it's more vortex and more nitrogen entrainment.
- 20 MEMBER CATTON
It's a suction problem then.
21 MR. TRAPP:
Right.
l
{.-
22 MEMBER CATTON:
You will suck the gas into the
]
23 line.
24 MR. TRAPP:
Right.
So over pressure isn't
, g ;;
25' really going to help you much.
You know, it's not a 4 -
i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i.
1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
4 g
.r
261 1
temperature --
2 MEMBER CATTON:
No, no.
I understand.
7-
)
3 Although I don't think that would happen at the top.
You 4
have to get some level in the hot leg before that woulo 5
occur.
6 MR. TRAPP:
You mean it would have to drop 7
below --
8 MEMBER CATTON:
It would have to drop or else 9
it couldn't happen.
10 MR. TRAPP:
It's about the top of the hot leg.
11 I think it might be 197.
It's close.
12 MEMBER CATTON: I need a surface in order to --
13 MR. TRAPP:
To allow it to vortex.
,a t
i
~>
14 MEMBER CATTON:
For it to vortex.
15 MR. TRAPP:
Right.
It's right about the top 16 of the hot leg.
17 MEMBER CATTON:
Also, you need to have pretty 18 good suction before it will vortex.
Anyway --
19 MR. TRAPP:
Again, this is data from their 20 curve on their shutdown operation, midloop operation.
We 21 just tock the number off of their curve.
I don't know how 22 that was established.
23 MEMBER CATTON:
I don't know that there's much 24 difference between -- you know, one or two inches of level (p) 25 is not very much.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 G02) 234-4433
262 i
1 MR. TRAPP:
And given too, that this bubble 2
growth could have continued.
There's no reason it would p
l 3
stop at the top of the hot leg either.
4 MEMBER CATTON:
Yes.
l 5
Would a couple of. bubbles wipe out the RHR i
6 pump?
i 7
MR. TRAPP:
A couple bubbles?
.8 MEMBER CATTON:
Yes.
I mean it might wake 9
everybody up.
1 10 MR. TRAPP:
I don't know.
i 11 CHAIRMAN KRESS:
Has this incident made the --
12 MEMBER CATTON:
I don't know if this guy is 13 through, Tom.
O 14 CHAIRMAN KRESS:
I'm sorry.
15 MEMBER CATTON:
He's going to sit down, so I 16 guess he is.
I didn't mean to interrupt you.
17 MR. L'JEUREUX:
That's the extent of my 18 comments.
19 CHAIRMAN KRESS:
I was wondering if this 20 incident.has made the ASP program event system yet.
Does 21 anybody know?
22 MEMBER BARTON:
I haven't seen it.
23 MR. BENNER:
It is being entered.
It is being 24 evaluated for in the ASP program.
Whether or not it will OG 25 come up as a precursor or not is uncertain as of yet.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
263 1
MEMBER MILLER:
I had'a question on the core 1
2 exit thermocouples.
You indicated early on in your f
3 discussion they-had disconnected those?
I 4
MR. TRAPP:
That's correct.
{
5 MEMBER MILLER:
That's standard procedure 6
then?
7 MR. TRAPP:
That is.
8 MEMBER MILLER:
But then later on, you say 91 they weren't read.
l 10 MR. TRAPP:
Well what we felt is is that you 11-had the ability to go into containment with test' l
l 12 instrumentation and manually take readings.
You had the i
13 ability to do that with the rivless and the thermocouples.
O l
14 We felt that should have been done.
15-They also had farther on down in the event j
16 response, they installed a jumper for the thermocouples, I
17 and they installed a jumper for the rivless, so they had l
18 control and indication of both following the event.
i 19 Both of these activities were somewhat
[
]
20 indicated as part of the 94-36 information notice I
21 response.
That was one of our issues that we felt that 22 could have probably taken some of that information.
Had 23 they gotten the jumper installed previously, it would have 24 certainly provided the operators additional information.
(
25 MEMBER MILLER:
So it really wasn't a matter NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
WASHINGTON, O C. 20005 3701 (202) 234-4433 (202) 234-4433
~.,..
~ - - -. -. -
.. ~ -
l 264
'l of inadequate instrumentation.
It's just they didn't use 2
what they had.
f-q) 3 MR. TRAPP:
Being inadequate at the time of 4
the event.
5' MEMBER BARTON:
Well, I think you would, Don, 1
6 that there's a lot of temporary stuff hooked up during 7
refueling outage.
I think that leads to some of the i
8 problems you have had with' operators not familiar with j
9 normal indication being there.
They've got temporary, and 10 sometimes it doesn't work.
If the temporary header 11 doesn't work, that kind of tends to complicate these 12 events.
13 MR. TRAPP:
And I don't think Haddam Neck's 14 practices were too much different than the rest of the 15 industry.
16 MEMBER MILLER:
What's the reactor level 17 indicator system?
18 MR. TRAPP:
How does it work or --
19 MEMBER MILLER:
What kind of system is that?
20 MR. TRAPP:
It's a thermo couple system.
It 21 has I believe six discrete points.
One would point be 22 slightly below the top of the head, one at the flange, and 23 then there's four down in this region above the fuel.
I 24 believe it's one heated thermocouple and then another
()
25 thermocouple.
You look if there's water there, the heat NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 1
~
265
~1-t ransfer --
2 MEMBER MILLER:
That was disconnected also?
- (_)
3 MR. TRAPP: 'That's the rivless' system at
)
4 Haddam Neck.
What I just described is their rivless.
5 MEMBER MILLER:
Right.
So that had also been 6
disconnected?
l I
7 MR. TRAPP:
Right.
The thermocouples and the
?
8 reactor. vessel level.
It's an electrical connection.
The 9
missile blocks up above, it goes up through that area 10 where the rod, control rod drive connections are, so when 1
i 11 you lift the missile blocks off,.take all that stuff-12' apart.
l 13 MEMBER MILLER:
So it was standard practice to 14 disconnect both on what, all CE plants then?
15 MR. TRAPP:
It's Westinghouse.
And the
'16 Westinghouse rivless also has pipes that usually run up
- 17 along the missile blocks.
You have to take those apart 18 too.
There you don't even have the ability to do a 19 jumper.
2'O MEMBER MILLER:
'So you really have in a sense 21 lost instrumentation to maintain any knowledge of your 22 water level.
23 MR. TRAPP:
Inside the vessel.
24 MR. CHAFFEE:
But I think it's important to 25 point out that normally they do that for a short period of
+
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
=
266
[
-l' time, whereas in this case, they had a delay.
They were 4
f 2
four. days I guess or five days.
So again, the normal
- _<t t
3 window of not-having this type of level indication'is
.l
~4 shorter.
They normally transition into filling the cavity 5
when:they have a level indication in a more timely fashion 6
than happened in this event.
t 7
.MR. TRAPP:
I think we had something like
~
8 about five days to get the head off,-'get everything part, 1
9-detention head off.
Normally you would have maybe five 10 days without.
11~
MEMBER MILLER:
You indicated lack of -- you l
12 made a statement lack of technical -- asking for technical i
i 13 support I guess is the way to put it.
'l l
14 MR. TRAPP:
- Yes, i
15 MEMBER MILLER:
So in other words, the I&C I
16 engineers were not around.it?
17 MR. TRAPP:
We just. thought the operators 18 could have been more aggressive.
19 MEMBER MILLER:
Do they have an STA at Haddam 20 Neck?
21 MR. TRAPP:
I don't know the answer to that.
)
22 MEMBER BARTON:
Is the STA on duty during 1
23.
refueling outage, if you have one?
24 MEMBER MILLER:
I thought STA was supposed to 25 be.around all the time.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1
1323 RHODE ISLAND AVE., N.W..
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
267 1
MR. L'JEUREUX:
At Haddam Neck, we have the 2
dual function STA, where the shift supervisor also
,s
(
)
'~'
3 performs the STA function.
As such, she was in fact in 4
the control at the time.
5 MEMBER BARTON:
That's not necessarily 6
standard.
Some utilities will take STAS off STA duty 7
during refueling outages, except for critical evolutions 8
during a refueling period.
So it a not your normal shift 9
supervisor STA relationship unnecessarily on shift during 10 refueling outage.
11 MEMBER MILLER:
Is that practice of all 12 utilities or is that --
13 MEMBER BARTON:
I don't know all, but I know 14 that --
15 MEMBER MILLER:
I think everyone uses STA 16 differently, right?
17 MEMBER BARTON:
They use them differently, j
18 correct.
19 Any other committee member question?
- Jim, 20 thank you for the excellent briefing.
21 In summary, I think this points out again that 22 these events are never simple.
Many causes and many 23 lessons to be learned.
I don't think we have heard 24 anything here which is a surprise.
Planning and
(-()
25 scheduling, material condition, operator challenges, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4 33 WASHINGTON, D.C. 20005-3701 (202) 234 4433
l 268 1
procedures.
We have all heard them all before.
,-m 2
In addition, I think events like this point
\\
}
~
3 out the need for licensees to increase awareness of plant 4
conditions during periods of plant shutdown.
As was the 5
case during this event, plant configuration is somewhat 6
modified to accommodate outage activities.
The example 7
was the temporary means for measuring vessel water level, 8
also a temporary event header, which had a history of some 9
problems.
10 I think as utilities prepare for the 11 deregulating environment, operating cycles will lengthen, 12 refueling outages will happen less often.
Most utilities 13 are going to 40 month cycles, which I think adds to this
(
)
s' 14 problem in that operators will not be used to doing 15 refueling evolutions in dealing with temporary conditions 16 during those evolutions as often, since the operating 17 cycles will continue to increase.
18 I think with that focus changing in the 19 industry, there's an ever-increasing need for utilities to 20 heighten their awareness for shutdown risks.
21 VICE CHAIRMAN SEALE:
Don't turn your back to 22 the bull.
23 MEMBER POWERS:
I just can not resist saying 24 that --
(~N
!q,)
25 MEMBER MILLER:
I am surprised you have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
269 i
i resisted this long.
1 i
i i
2 MEMBER POWERS:
That you know, we're O
3-continuously told we don't need to look at these shutdown j
i 4
risks because we have scoped it out in our examinations, 5
crude examinations, of one mode of operation.
6 That it's obvious that we.have lots of 7
complicated engineering systems, multiple activities going 8
on with temporary equipment in these things.
That strikes I
i 1
9 me as exactly the kind of thing that PRAs were intended to 10 examine.
We are isolating ourselves from that.
1 11 We are also told that the industry-has now l
l-12 implemented lots and lots of new processes that are l
I 13 supposed to make much more disciplined activity during l-14 shutdown operations that reduces the risk even below the 15 leve.ls'that we've got in these scoping'PRA studies.
It's 16 not obvious to me that we've gone far enough.
17 I have just written a note to the planning 18 procedures that we're looking for areas to focus on.
I i
19 think we're just going to have to go look at the shutdown 20 risk business.
Id 21 MEMBER MILLER:
It seems to me that during l
l 22 operation, we have a requirement to measure certain l
23 parameters at all times, certain critical parameters.
It 24 seems like during shutdown, we ought to have the same 25 requirement.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
t i
270 i
V 1
.1 I know there is a requirement to make certain l
\\
i 2
you are measuring neutrons at all time, even when you are l
l 3
shut down.
I think that's still a requirement.
I'would i
L 4
think water level of.some type, they had a simple system l
5 to do that.
-6 I guess I'm in a sense supporting Dana on you l
7 need better procedures of that type.
Here are the things 8
that should be measured at all. times.
Also somebody l
i t
9' around who understands what the measurement means.
10 MEMBER POWERS:
We have got an engineering t
11 tool that allows us to identify what the critical 12 measurements are.
We have an engineering; tool for which L
13 under operations is now being used to guide day to day O
l 14 planning.
i-15 Why are we avoiding extending this tool to a t
l' 16 place where'it seems to be critically needed?
It's just' l
.17 surpathaffle human understanding here.
i l
18 CHAIRMAN KRESS:
Are there further comments?
19 MEMBER BARTON:
No further comments, I'll turn l
l 20 it back to you, Tom.
l 21 CHAIRMAN KRESS:
Okay.
At this time, I think i
I 22.
we can announce that we no longer need the transcription l
23 for the rest of the day.
l 24 (Whereupon, at 11:06 a.m.,
the proceedings b).
(_
25 were concluded.)
- NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
I O
CERTIFICATE This is to certify that the attached proceedings before the United States Nuclear Regulatory Commission in the matter of:
Name of Proceeding:
436TH ACRS FULL COMMITTEE Docket Number:
N/A Place of Proceeding:
ROCKVILLE, MARYIAND 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.
$1
}ht/A-
"C'ORBETT RINER' Official Reporter Neal R. Gross and Co., Inc.
AU NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVENUE, NW (202) 234-4433 WASHINGTON, D.C. 20005 (202) 234-4433
INTRODUCTORY STATEMENT BY THE ACRS CHAIRMAN 436TH ACRS MEETING, NOVEMBER 7-9, 1996 THE MEETING WILL NOW COME TO ORDER.
THIS IS THE SECOND DAY OF THE 436TH MEETING OF THE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS.
DURING TODAY'S MEETING, THE COMMITTEE WILL CONSIDER THE FOLI4 WING:
(1)
FUTURE ACRS ACTIVITIES (2)
RECONCILIATION OF ACRS COMMENTS AND RECOMMENDATIONS (3)
REPORT OF THE PLANNING AND PROCEDURES SUBCOMMITTEES' (4)
OPERATING EVENT INVOLVING NITROGEN BUBBLE IN THE.
REACTOR COOLANT SYSTEM AT THE HADDAM NECK NUCLEAR POWER PLANT (5)
ANNUAL ACRS REPORT TO CONGRESS (6)
PROPOSED ACRS REPORTS A PORTION OF TODAY'S MEETING MAY BE CLOSED TO DISCUSS ORGANIZATIONAL AND PERSONNEL MATTERS THAT RELATE SOLELY TO THE INTERNAL PERSONNEL RULES AND PRACTICES OF THIS ADVISORY COMMITTEE, AND MATTERS THE RELEASE OF WHICH WOULD CONSTITUTE A CLEARLY UNWARRANTED INVASION OF PERSONAL PRIVACY.
THIS MEETING IS BEING CONDUCTED IN ACCORDANCE WITH THE PROVISIONS OF THE FEDERAL ADVISORY COMMITTEE ACT.
MR. SAM DURAISWANY IS THE DESIGNATED FEDERAL OFFICIAL FOR THE INITIAL PORTION OF THE MEETING.
WE HAVE RECEIVED NO WRITTEN STATEMENTS OR REQUESTS FOR TIME TO MAKE ORAL STATENENTS FROM MEMBERS OF THE PUBLIC REGARDING TODAY'S SESSIONS.
A TRANSCRIPT OF PORTIONS OF THE MEETING IS BEING KEPT, AND IT IS REQUESTED THAT THE SPEAKERS USE ONE OF THE MICROPHONES, IDENTIFY THEMSELVES AND SPEAK WITH SUFFICIENT CLARITY AND VOLUME SO THAT THEY CAN BE READILY HEARD.
O I
i ww c
< - - --,,-~ -
~~
O O
O~T ACRS PRESENTATION HADDAM NECK DECAY HEAT REMOVAL SYSTEM CHALLENGES AUGUST 28,1996 PRESENTED BY:
JAMES TRAPP, (610) 337-5186 DIVISION OF REACTOR SAFETY, REGION I
...._--_-;7 O
O O-l NRC STAFF PRESENTATION TO THE ACRS
SUBJECT:
DECAY HEAT REMOVAL SYSTEM CHALLENGES AT HADDAM NECK DATE:
NOVEMBER 8,1996 f
PRESENTER:
JAMES TRAPP PRESENTER'S TITLE / DIVISION:
SENIOR REACTOR ENGINEER i
DIVISION OF REACTOR SAFETY PRESENTER'S NRC TELEPHONE NUMBER:
(610) 337-5186 l
l
--l
O O
O-MAJOR EVENTS REVIEWED i
i 1.
UNDETECTED ACCUMULATION OF NITROGEN GAS IN THE REACTOR VESSEL HEAD 1
2.
TWO INADVERTENT DIVERSIONS OF REACTOR COOLANT l
SYSTEM INVENTORY t
3.
DECAY HEAT REMOVAL EQUIPMENT FAILURES 4
i 3
i
O O
O '-
FORMAT i
1.
SEQUENCE OF EVENTS i
2.
SAFETY SIGNIFICANCE I
3.
NRC ACTIONS 4
o o
o
-l SEQUENCE OF EVENTS UNDETECTED INTRODUCTION OF NITROGEN GAS INTO THE REACTOR VESSEL t
i AUGUST 28: OPERATOR INCORRECTLY OPENED VALVE; e
RESULTED IN. INJECTING WATER AND NITROGEN INTO j
i e
VALVE CLOSED, BUT NOT SEATED PROPERLY; ALLOWED GAS
~
TO LEAK INTO RCS FROM VOLUME CONTROL TANK.
l e
GAS ACCUMULATED IN REACTOR VESSEL HEAD; DISPLACED WATER FROM. REACTOR VESSEL ENTERED PRESSURIZER.
t 5
i
SEQUENCE OF EVENTS (CONTINUED)
UNDETECTED INTRODUCTION OF NITROGEN GAS i
INTO THE REACTOR VESSEL (CONTINUED) e AUGUST 29: =5000 GALLONS OF WATER DRAINED FROM RCS l
IN PREPARATION FOR REFUELING ACTIVITIES. LATER, =1000 GALLONS WAS ADDED TO RCS (WHEN DECISION WAS MADE i
TO SUSPEND REFUELING ACTIVITIES).
l e
AUGUST 31: GAS BUBBLE GROWTH APPEARED TO SLOW.
i e
SEPTEMBER 1: GAS LEAKAGE STOPPED BY ISOLATING I
NITROGEN TO THE VOLUME CONTROL TANK.
OPERATORS ADDED =4000 GALLONS TO FILL RCS.
j 6
SEQUENCE OF EVENTS (CONTINUED) i INADVERTENT DIVERSIONS OF REACTOR COOLANT SYSTEM INVENTORY I
e AUGUST 22: OPERATORS IMPROFERLY ALIGNED VALVES PLACING RESIDUAL HEAT REMOVAL (RHR) PURIFICATION i
SYSTEM IN SERVICE; FAILURE INADVERTENTLY DIVERTED
=500 GALLONS TO REFUELING WATER STORAGE TANK.
l e
SEPTEMBER 4: OPERATORS INAPPROPRIATELY CONDUCTED WORK ACTIVITIES ON RHR CONTAINMENT SPRAY VALVES; DIVERTED ~300 GALLONS TO CONTAINMENT SUMP.
7 l
~
~'
O O
O SEQUENCE OF EVENTS (CONTINUED) i i
DEChY HEAT REMOVAL EQUIPMENT FAILURES i
e AUGUST 31: OPERATOR IDENTIFIED LEAK IN "A" RHR HEAT EXCHANGER INLET VALVE; VALVE CLOSED TO ISOLATE l
LEAK (REMOVING HEAT EXCHANGER FROM SERVICE).
I t
e SEPTEMBER 1: "B" RHR PUMP FAILED TO START, WAS FOUND SEIZED. PUMP PREVIOUSLY OPERATED ON AUGUST 19.
8 I
O O
O i
SAFETY SIGNIFICANCE i
i NITROGEN INTRUSION l
l e
TIME TO 200 F WAS =52 MINUTES IF RHR LOST.
e REACTOR VESSEL LEVEL DECREASED TO ~2.5 FEET ABOVE THE POINT OF RHR PUMP CAVITATION.
i e
IF GAS BUBBLE CONTINUED TO GROW, WOULD HAVE CAUSED CAVITATION IN OPERATING RHR PUMP.
e REQUIRED BORIC ACID FLOW PATH COMPROMISED BY GAS IN CHEMICAL AND VOLUME CONTROL SYSTEM.
r e
GAS COULD IMPACT ABILITY OF STEAM GENERATORS TO REMOVE DECAY HEAT BY NATURAL CIRCULATION.
i i
9
- t
O O
O SAFETY SIGNIFICANCE (CONTINUED)
DECAY HEAT REMOVAL CHALLENGES l
e TWO INADVERTENT RCS-DRAINDOWNS (HAD OPERATORS NOT TAKEN PROMPT ACTIONS TO TERMINATE) COULD HAVE l
RESULTED IN CAVITATING OPERATING RHR PUMP.
"B" RHR PUMP AND "A" RHR HEAT EXCHANGER NOT AVAILABLE FOR SERVICE.
i e
VENTING THE "B" RHR PUMP FOLLOWING MAINTENANCE WAS DIFFICULT.
j i
e IF RHR SYSTEM LOST, TIMELY RESTORATION OF FORCED CIRCULATION USING A REACTOR COOLANT PUMP WOULD HAVE BEEN DIFFICULT.
10
O O
O SAFETY SIGNIFICANCE (CONTINUED) i t
EVENT RESPONSE i
e REACTOR VESSEL LEVEL INDICATION (RVLIS), CORE EXIT THERMOCOUPLE (CET) READINGS NOT TAKEN FOR 2 DAYS.
i e
ADDITIONAL SURVEILLANCE OF OPERATING RHR PUMP NOT IMPLEMENTED FOR 5 DAYS.
I e
REACTOR COOLANT PUMP AND SEAL INJECTION NOT ALIGNED FOR SERVICE FOR SEVERAL DAYS.
1 e
SPECIAL TESTS TO ESTABLISH VESSEL LEVEL AND VENT HEADER OPERATION WERE NOT PERFORMED FOR 4 DAYS.
e INCIDENT RESPONSE TEAM NOT IN PLACE FOR 4 DAYS.
e 25 DAYS TO RESTORE "B" RHR PUMP TO OPERABLE.
i i
11 t
O O
O SAFETY SIGNIFICANCE (CONTINUED)
I OTHER CONTRIBUTING CAUSES i
e LACK OF QUESTIONING ATTITUDE / POOR DECISION MAKING i
e POOR IMPLEMENTATION OF GENERIC INFORMATION e
POOR EQUIPMENT CONDITION / LACK OF INSTRUMENTATION e
POOR PROCEDURES / FAILURE TO IMPLEMENT PROCEDURES j
e INADEQUATE TRAINING & PRE-JOB BRIEFINGS l
f e
WEAK USE OF TECHNICAL SUPPORT i
e INAPPROPRIATE PLANNING AND SCHEDULING 12 j
1 1
~~
~
O O
O NRC ACTIONS e
AUGMENTED INSPECTION TEAM SENT TO SITE ON SEPTEMBER 3-16,1996.
ISSUED INFORMATION NOTICE 94-36, SUPPLEMENT 1,
" UNDETECTED ACCUMULATION OF GAS IN REACTOR COOLANT SYSTEM," ON NOVEMBER X,1996.
ADDITIONAL GENERIC ACTION IS BEING CONSIDERED.
l 1
l i
f 13
9 O-THE UNDETECTED NITROGEN GAS INTRODUCTION INTO THE REACTOR VESSEL FROM THE CHARGING LINE L
g Vent Header generator F
[
V Pressurizer Pressurizer j
/
"0-
..n W
- 691" EL i
Head V
O
(*
Pressurizer Vent f
Li i
T)
Ai'
- Head Full 355" EL-12%
- 351" EL Top of Head
- 301" EL Wow 0..
Reactor i
Cavity
~
Nitrogen Vessel Level 1
r-233* EL Indicatar B
a Bubble a
,q Flange 270" EL l
0 Gas o
o o
qw o 4 --l
{^^^^^^^^-~
^- j 1@
'y f
q V
Nitrogen Hot Leg f
0*
Gas a
water I o+
a+O 3 c ajgng s
g l
Cold Leg f Velve LOOP h
RHR Stop Coo 8ng RHR inlet Velve Fuel piow
{
OW 10* Pressurizer 10" RHR R m tor Cold Leg y
Surge IJne Suction Line Cooient (LOOP 4)
(Loop 1)
EL 198" -
Loop #4
%p-1 J
r EL 185" ~
+
27v2"
+
EL 172" -
Rf, tor gagt,g pg,
p
'l FIGURE 1
O O
O-CHEMICAL & VOLUME CONTROL SYSTEM NITROGEN GAS LEAK PATH FROM VOLUME CONTROL TANK TO THE RCS i
i To RCS Loop #2 Volume Control Cold Leg Tank A
To RCS Q
f Loop #4 Gas 30s 6
Hot Leg
'^^^^=^-
eA-v-n4 Open Regenerative Hxs
% [
l sA-v-ass 7
A A
A y,iy,,
%s J L QosM l
Lg.j Leaking By g
'g: "
7, rA r
M
=
r, V
V V
~
Bypass I
n open i
i II Boric Acid Blender
)q Valves Closed Cliarging Leaking By Fumps rom Pumps L
FIGURE 2
e _._... _. _ _ _. _.... _ _ _. _ _ _. _ _ _ _ _. _ _ _ _ _.. _ _ _ _.. _.. _ _.. _ _ _. _. _.
- 7. -
O O
O t
RESIDUAL HEAT REMOVAL SYSTEM (RHR)
From RWST From RCS Loop #1 Hot Leg i
I o
u Failed Pump
.i n
X X
fB RH-V-794B RH-V-7918 i
^
To Loop #2 f
RCS J
Cold Leg i
RHR j
Pumps
[
>4
>4 RH-V-794A RH-V-791 A A
{
Heat Exchangers Pinhole Leak in Valve
- Recirc to RWST s
FIGURE 3 t
4 p 1,.%
ee:44 ve v W.
- 21 x* ao<<
raj 6.at ge m
- -N 7
YL f_s.3 -
m Y
o
%M DM re o 5 SE3O7/4 OM ise Dw O'. F M k.
%> P.*/f
)!
&I Tc.'l I
II h __,...,.h 3
( l M, 7'
1 1
s g
g
[
f
't- )
j~
~
, - ~
H Y.
l
'P,g.,y*
h M.m-M-
7 i
u fcs O
c' W1 e, d-.
V
,n 1
Got Uc3 W
Twers g.
I b
Y CAW.4 VCS j
pasr
(
1 J
g
+
,D
)
_'-X-v to2-1 Fi
/jj
?
w s__ g 4.
O ust To 9 hot 0C Tecm u ra r -