ML19199A507
| ML19199A507 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 04/23/1979 |
| From: | Ahearne J, Bradford P, Gilinsky V, Hendrie J, Kennedy R NRC COMMISSION (OCM) |
| To: | |
| References | |
| REF-10CFR9.7 NUDOCS 7905030055 | |
| Download: ML19199A507 (85) | |
Text
a Q ^ b DS
^
f NUCLE AR REGULATORY COMMISSION IN THE MATTER OF:
OPE 1 MEE"2ING BRIEFING ON PRINCIPAL FACTORS RELATED TO CURRENT STATUS OF OPERATING PLANTS Place. Washington, D.
C.
Date -
Monday, 23 April 1979 Pages 1 - 84 7eiechone:
(202) 347-37CO ACE. FEDERAL REPORTERS, INC.
OfficialReporters 444 Ner+h Cepitel Street 1i 1
p Washingten, D.C. 2CC01 II 8
U 7905030666 NATIONWID E COVERAGE. DAILY
~
CR4?65 1
(
g 1
\\
r DISCLAIMER This is an unofficial transcript of a meeting of the United States Nuclear Regulatory Cccmission held on f t:nd.ay, 23 April 1979 in the Commission's offices at 1717 H Street, N. W., Washington, D. C.
The This transcript meeting was c;en to pu lic attendance and observation.
has not been reviewed, corrected, or edited, and it may contain inaccuracies.
. The transcript is intended solely for general infor a'ticnal curposes.
As provided by 10 CFR 9.103, it is not part of the formal or informal record of decision of the matters discussed.
Excressions of opinion in this transcript do not necessarily reflect final detenninations or No pleading or other paper may be filed with the Carmission in beliefs.
any proceeding as the result of or addressed to any statement or argument contained herein, except as the Ccamission may authorize.
e e
4 e
i." e".
q ll
l!
i, a
=
ll 2
l l
1.
R4S IE UNITED STATES OF A' ERICA EL'14ER/rcca 2
NUCLFAR REGULATORY CO'4'1ISSION 4
i
.I 3
i Ocen ^1eeting 6
i 1
7i BRIEFING ON PRINCIPAL FACIORS RELATED TO i
i e
3 Il CURRENT STATUS OP OPERATING PLA'CS i
i i
9i o
41 10 '
11;!
Room 1130 12 ;
1717 H Street, ?I. N.
Washington, D.C.
13 ;'
Mondav, 23 Acril 1979 1.
15 Sieeting in the above-entitled matter was convened it at 1:30 c.m.,
JOSEPH kl. HENDRIE, Chairman, cresiding.
'7 PRESENT:
IE JOSEPH 'd.
HENDRIE, C HAI R'!AN l9 VICTOR GILINSKY, COTIISSIONER c
RICHARD KENNEDY, COAL'4IS S IO'E R 21 PETER 3RADFORD, C O'1'4IS S IO'IE R 22 JOHN AHEAR'E, C091ISS IONER 22 Also Present:
'!essrs. Denton, Gossick, Davis, Casa, 'Iattspn s' I
, un.,-..
Ii 6
2,,
Eisenhut, Ross, 'lo se le, Weiner and Bickwit.
3 I
I CR 55
_P_ _R _O _C _E _E_D
_I_ _N _G.S MELTZER/mm i
2 p, o
CHAIRMAN HENDRIE:
If we could come to order.
O 3 I l
The first item of business this afternoon will be i
4l l
a vote by the Commission to hold this meeting on short notice.
i i
5' i
The Sunshine Act requires us to do that,,since we did not j
i schedule it more than a week ago.
7 I would, therefore, ask -- and I might say that 8
the subject this afternoon is a briefing on the principal i
9 i
factors related to current status of operating plants -- an 10 '
update on status of review of Babcock and~Wilcox plants, and 11 a discussion of Saturday's Inspection and Enforcement bulletin.
12 1 l
Those in favor of the short-notice meeting please i
13 l indicate.
i 14 COMMISSIONER GILINSKY:
Aye.
15,
COMMISSIONER KENNEDY:
Aye.
I 16 !
1 COMMISSIONER AHEAR'iE:
Aye.
tlo 17 :
CHAIRMAN HENDRIE:
So ordered.
l 18 i p
Lee, we welcome you, and Harold Denton, and Ed i
19 Case, and Roger Mattson.
Please go ahead.
20 MR. GOSSICK: As you indicated, also we have got a 1
21 4 briefine bv Mr. Davis and his oeoole on the efforts that have 1
,, o
^^ i! been going on, apart from the bullatin itself.
23 L CHAIRMA'? HENDRIE: All right.
24
.y o r. ' t yN go ahead.
ke al Reporters, Inc,,
,c t'e,first slide, MR. DENTON:
Let's see if we can get g
I/
1.
4 1
.=;
clease.
il 2'
(Slide) l j
I will have copies of these slides available by
~
i a
the close of the meeting.
I didn't get a chance to make them l
l 5
f before the start.'
j 6 ll i
There are six areas I want to cover. First, we are i
7 continuing to learn from what happened at Three-Mile 2.
As 8
l we said, and a number of people have said in the past, tne 9l accident was caused and aggravated by human error.
Since t
I i
10 !
I've gotten back, I've gotten together with the staff and we've 11 talked about why are B&W plants more burdensome on operators 12 i
{
than others?
13 What is it about these clants that recuire more I
i 14 I i
operator actions?
I 15 l l
i So we have gone --
16 f
COMMISSIO5ER KENNEDY: That's on the assumption that, i
17 '
in fact, they do.
16 l MR. DENTON: We have asked the question if they do, h:l 19 9 and we have identified, I think, five design features of B&W 4
20 "
plants which, in off normal conditons place greater burdens 21 on ocerators than other PWRs.
i a,
v" :
Let me just identify these five.
Ones being talked h
2 ~' Yl about before is the inventory in the secondary side of the f
24 steam generators; the amount of water differs by a factor of s ee d at Reporters. Inc. i "E
three or four, up to a factor of eight with the other tyces of f
j[
r I
t
'I
5 l
1 l
1m 3 PWRs always having more water.
This means that the system 6
I 2 f:
1s more tolerant of loss of feedwater flow and has more time l
l 3'
l to restore it.
I 4
COMMISSIONER AHEARNE:
The other systems are more I
I 5'-
I tolerant?
,I l
l 6
MR. DENTON:
Yes.
i i
7 Another factor is the smaller purge capacity in 8
the BWR systems.
' lou look at the response of the pressurizer, l
l 9
pressure and level in a BWR compared to other types of designs,'l 10 '
it is less tolerant to withstand surges that come about because; 11 l
of secondary system changes.
I i
l 12 {
Another major difference is B&ie pl'nts rely, and i
13il - mre designed to rely on opening of the power-ocerated relief i
I i
14 !
j valve on the pressurizer to avoid scramming in a number of i
15 '
l transients.
I 16 lq None of the other PWR plants, those designed by 17 i j
Westinghouse or CE, have ever, to our knowledge, opened this 18 :j pressurizer relief valve'as a result of off-normal conditions, l
l1 19 ;i aad in the B&W plants there have been 150 incidents where this l
a d!
20 i valve did onen as a result of off-normal conditions.
- l 21 !
Another factor is the relative generation of the 22. steam cenerator --
1:
1
^3' CHAIR"? HENDRIE: Harold, do you know, aside from 24 the Three-Mile 2 case at hand, in any of the other identified A:
eral Reporters. Inc.
^R cases, where the relief valve has opened on one of the B&N
./.
4 - r l
i
6
=
=
i m
plants, has it hung open, do you know.
I 2
MR. DENTON:
I think there have been a total of i
t 3l i
l three cases, counting Three Mile.
The other two cases where it was open, either they l
r 5
i closed -- either reseated, or they closed the block valve.
j l
6' And one was related, I think, to mechanical failure.
l t
7 I
But let me ask someone else who maybe knows those i
l l
8 two instances better than I do.
j 9l i
DR. MATTSON: There were two instances. One was 10 '
at Oconee, one was at Davis-Besse.
l 11 Oconee, the explanation was some kind of depth I
t I
f 12,
l deposit on the valve that built up and allowed it not to close.'
I.
i 13 1 l
That was something they retrofit then to all of their plans j
14 l' l
to fix that problem.
CHAIRMAN HENDRIE:
The correction.
16 ll DR.MATTSON: The Davis-Besse problem was a unique 17 '
i l
valve.
It was the only plant that used the particular valve 18 !
that failed, and it occurred during startup testing.
I 19 h q
Apparently the difficulty was a
piece of electronic a
eo.,
' gear of one sort or another was left out.
I believe a relay.
1 21 "h And when that was fixed, the valve functioned properly.
MR. DENTON: Two other factors that make E&W nore
,3 h sensitive are the relative elevation of the steam generator 24 and reactor vessel heat sinks that E&W seems to have a less A CT ff 81 beDCrferl, if'C.
,C favorable configuraticn fcr natural circulation than the others 1
'?
4 M '
s
/
4 I
7 mm1 do.
I I,
2 I!
And finally, the fact that B&W nlants don't have h
i 3 i trins on secondarv system failures. This is in keeping witn i
4 their reliance on opening of the pressurizer relief valve.
i 5
Westinghouse and the CE plants trip on a variety I
6' of upsets in the secondary system, which lead to scrams.
l i
I 7
So, the B&W plant was designed to ride through the i
8 control system transients, which the other PWRs scrammed in l
l 9
the course of.
l 10 '
And I'll deal with these five later on, because in l
I i
11 looking at these five, some of these factors were the basis l
1 12 for the bulletin we issued over the weekend.
I i
13 COMMISSIONER AHEARNE: Harold, so what you are i
14 h saying is the plant is really designed for the relief valve to li 15q be open at a fair frequent number of times?
i Il i
16 ll MR. DENTON: That's right.
I ile 17 y COMMISSIONER AHEARNE: But it is also true, isn't it, a
l 18yl that there is no actual measurement device on it to indicate h
b 19 o that it has closed?
20 MR. DENTON:
I think there is an indicator on the
.i
,i 21 solenoid. But let me ask someone who is more familiar with p
the design of that valve.
23 i; DR. MATTSON:
The question is whether you get an 24 indication in the control room if the valve is open or closed?
a-at ReDorters, Inc.
25 COMMISSIONER AHEARNE: As I enderst and it, I thought P
4
g it was that there was a signal to indicate that you have given i 1
1 l
f I
2p the signal to the valve to either open or to close. But that 1
I 3i the actual accomplishment of the closing is not --
i i
I 4
MR. DENTON:
I think that's correct. There's not a --
l 5
DR.MATTSON:
There is more information than that I
6 available. There is the thermocouple in the tail pipe of the 7
valve line as it goes to the quench tank, which is a fast 8_
responding thermocouple, and it gives indication as to whether i
I l
9 1 or not there is steam coming down the tail pipe.
If there is, l
10 '
the valve is open.
i 11 The instrumentation in the quench tank varies from l
l l
12 i plant to plant, but I bdieve in the case of Three Mile there l
i, 13 is terperature, pressure and level indication in the quench i
14 i tank which would also be a further indication diat the valve l
l 15 ;
remained open.
16 '
COMMISSIONER AHEARNE: So you think, Roger, that U
17 ] there are other means of determining that it is staying open?
IS DR. MATTSON:
As I understood Mr. McMillan's summary n
1 19 ? to the ACRS last week, nearly, I believe every plant has some J
4 20 indication, secondary indication in the control room as to 21 j the status of that valve.
Scme directly off the valve, others 22 J of the type that I just described.
L 23 MR. DENTON: One of the crime acconolishments of 24 the bulletin we issued on Saturday was to -- and I'll get into re e 31 Recone s. Inc.
25 some detail -- was to eliminate reliance on the opening of j
,n l/
i v
g m
1 this valve for the types of transients where it opened in i
2 i the past. Sort of to avoid the sequential failure of feedwater !
i l'
l i
3l transients resulting in loss of coolant accidents, such as 1
4 Three Mile.
l 5
So what I will do ne>.t is, turn through the status i
l 6
of the operating plants in some detail. And I will include j
i 7
the actions in the bulletin that was taken.
1 l
8l As shown on this slide, we have issued bulletins to 9
all operating reactors that the B&W bulletins include a require-t i
10 '
ment for a design improvement.
The design improvement is, the:
i I
I 11 l reactor trip setting has been reduced such that it now trips l
12 !
at a pressure such that in all previous B&W transients you l
13 !
would not expect this valve to have openec.
s l
14 !
In other words, if the calculations are correct, i
l 15 :
all the 152-odd instances where this valve has opened in the l
I 16 ;
past, would have been precluded if the scram had been set i
f, 17 l where recommended by the bulletin, and the set point of the 18 valve had been raised as recommended by the bulletin.
i l
19 j COMMISSIONER AHEARNE: Will you talk about whether li'l 20 il there is any downside of that, your adjusting the basic 21 design operation of the plant?
i!
22 !j MR. DENTON:
I don'r think there's a real downside L
23 !!
from safety.
n 24 The downside is that the plant will trip more ACW-f 8f AfDOrters, IFC.
25 often than in the past on some transients that it might Pave n
4 --
\\/
s
- i
.i
i l
10,
l l
I ll ridden through. So it is a downside in terms of operational e
2 flexibility.
But I don't see a safety downside.
i I
3i Natural circulation, as mentioned in the bulletin, i
e
'j it is an area we are studying turther.
I 5'
And we are also looking further at the implications 6'
of having a stuck-open pressurized relief valve, since it would, I
i open for some remote transients. And even in the Westinghouse.
l l
l 0
1 and CE plants there are some transients in which this valve l
9i i
would be exc.ected to ocen.
i i
10 l So, let me go next to the second slide, just to get li l i
i an overview of wnere we are.
l 12 (Slide) 13 l I think our own staff is being stretched prettv i
i I ~d it; thin in the reactor systems area to accommodate the meetings up i
i 15 hi! at TMI and to do the reviews of the B&W plants.
B&W is I
16 lj probably also being stretched in this regard.
i 17' Other PWRs may have some of the same types of problens If but they are certainly less sensitive to off-normal plant 1
19 !:1 conditions in the secondarv svstem, and their staffs are not a
'O affected as the B&W one is.
a 21 '
COMMISSIONER AHEARNE : Have vou been able to draw oecole out of the other areas of NRC?
For examole, are there o
^*
people in Research that can be pulled off to assist you?
24 MR. DENTON: There has been a number of special W.
.ai Reporters, Inc.
ec
'~
areas where we have requested and gotten assistance fror Research i
4 I
(
l 8
11 m
1 and Standards.
i 2h But I think in the issues that are most critical l
t 3i here, reactor systems and systems analysis, there aren't very i
i l
4j many people in the agency in these areas, outside of NRR.
j I
5 COMMISSIONER AHEARNE :
I knew that there weren't l
6 on a sort of full-time basis.
i i
l 7
My question was more, were there people that it i
8 would be worthwhile pulling in to assist you, given that it 9l still would take a while for enem to become familiar, but, I
I 10 !
perhaps in the last year or two years they have been working i
11 those areas.
l I
i i
12 '
MR. DENTON:
Let me ask Roger, to see how successful, 13 ] we have been in the last week or two.
i i
l 14 l
DR. MATTSON: We have been using a number of those i
i 15 !
people both at the site and here in Bethesda. It is possible that if we were to bring them in and keep them in for a period 16 i h
F 17 1 of time that we could make more use of them. But it won't be i
is a lot.
1
}
19 MR. DENTON: So present policy is that we and B&W o
H 1
20 n are continuinc to meet fairly often to try to define and 1
s 1
21 L resolve the generic concerns in their plants.
22 $
We are meeting with specific utilities over specific r
23 f responses that they made to bulletins, and issues t h, a..t h a v e iq l/
iG-24 ' been. raised in those bulletins.
A:v-rai Reporters. Inc.
25 And we have a fairly extensive effort mounted on h
(
12 r-looking at natural circulation for B&W plants.
j I
And we are starting similar efforts with CE and 2
i
,j Westinghouse as time permits this week.
I t
Let me -- and then, I guess, the interim status, 4
the last item on this slide, is the issue that we are presently 5;
l rocusing on; what is the orobability of feedwater transients 6l i
I 7l in the various plants.
I Assuming for the moment that the bulletin has taken g
i I
care of the opening of the pressurizer relief valve for most 9
i transients so that we don't compound feedwater transients with 10,
I i
I 11 l LOCA, what is the probability of a LOCA.
l t
i i
i We have been looking at that over the weekend and 12 i
13 today, and I guess we are coming to the conclusion that the ll ja !! probability of a feedwater transient that requires operation o
'I of the ECCS in order to m'_tigate the transient is on the
{
15 i.
order of 1 in 10 to 1 in 100 for B&W plants.
16 'It it And it is probably a factor of ten higher in other j7 1
i jg designs.
i t
19 j CHAIP24AN HENDRIE:
Lower?
il
'I 20 ;
MR. DENTON:
Lower.
I
- l 21 p Yes.
And what we are starting -- with the starting l
assumption that the probability of a feedwate_ transient in
,2,
4 23 '
ny PWR is on the order of one or two per reactor year.
Then 24 you multiply tnis by the probability of the auxiliary feedwater i:e-si Reoorters, Ire.
73 pump system not operating when called upon to operate.
4 eo
- f ie i i e
i 13 l
i 1
Now the B&W plants are at a disadvantage because 2
of the factors I have described earlier, in tha tthey have i
i 3!
less volume. Therefore, there is less time available. And also 4
the integrated protection system such as the function at Rancho i
5 Seco transient yesterday, also seems to complicate the i
i 6i availability of the auxiliary -- of feedwater, under certain l
1 r
7 circumstances.
{
i 8
So, I guess if I had to put them in a probability i
i 9l order, I would say the probability of a transient in a B&W i
10 !
plant from feedwater caused conditions that recuires coeration :
i i
11 1 of the ECCS system is in the range of 1 to 10 to 1 in 100; I
I 12 l whereas the other PWRs seem to be about a decade lower.
I i
13 Now that's the kind of numbers we generated in i
14 looking this morning and yesterday, and they are not firm and i
i 15 they may change.
But, recognize the uncertainty that we 16 have not had an opportunity yet to see if there are compensating g
n 17 j factors that would make it different ia particular plants.
a
+
18 '
For example, at Oconee they only have one auxiliary feedpump i!
19 ' per plant, and that is a steam-driven pump, whereas at Rancho i
20 ; Seco they have both the steam-driven auxiliary feedpump and
!l 21 an electrically-driven feedpump.
And the probability of 22 a those systems failing to function is lower.
23 1 COMMISSIONER GILINSKY:
Can you explain the point 24 about the reactor protection ystem affecting the availability b
rat Reporters, tre.
25 of auxiliary feedwater?
Cr, did I understand you correctly?
! /
i 4
14 mm 1
CHAIRMAN HENDRIE: The integrated control system I
2[h rather than the protection system.
r1 i
P 3
MR. DENTON: Integrated control system.
I may have 4
said protection system.
I 5l COMMISSIONER GILINSKY:
No, I think you said control i
i I
6h system.
i I
i 7
Could you just explain that for me?
[
f i
a MR. DENTON: The integrated control system in the l
l 9
B&W olants adjust the water level in the steam generator j
i 1
10 depending on signals that it is getting.
i i
11 For example, in the Rancho Seco trio of yesterday,
{
12 it throttled down the feedwater flow to a low value and t
I I
13 maintained that value, but it did not trip on the auxiliary 4
14]
feedpumps to come on.
i 15 l:
COMMISSIONER AHEARNE:
What instrumentation is it i
I 16 I using to make these?
17 q MR. DENTON:
I think what we are learning is this Fl t
18 0 1s a very complicated control system.
We had not looked into a
17 n it in great cetail in the past becauce of the assur.ction that i
20 the plant itself was immune, largely, to these types of upsets.
I 21 But, let me ask Ed or Roger if they would like to 22 h explain these integrated control systems in a bit more de'; ail.
L 23 [
CHAIRMAN HE"DRIE: I would like to have the e:jolana-24 tion, but I think we should note the integrated control syster mF 1 Reporters. Inc.
25 is a stear side system as I unde rs tand it, which then adjusts
? l 1/
<io,
15 1
level of the steam generators and feedwater flow on the i
l 2l basis of the load that is being called for in the generator, l
3!
and hence in the turbine.
t 4
MR. DENTON:
It is something that all plants have 1
5l some type of control system, but the B&W one seems more 1
I 6
complicated and more difficult to analyze in terms of failure 7
modes and what it might do.
8 DR. MATTSON:
It is not entirely secondary.
When 4
9f it senses things in the secondary, it can also adjust the I
10 reactor.
11 CHAIRMAN HENDRIE: Reactor power.
12 COMMISSIONER AHEARNE:
It is an automatic system?
13 DR.
MATTSON: Yes.
14 MR. DENTON:
So it is control rods to compensate 15 for changes in the secondary system.
16 I
COMMISSIONER AHEARNE:
I would like to know more i
l 17 about it. In particular, why does this -- or does it -- is i
18 g it able to discriminate against some of the problems that we do 19i; tend to have been seeing in these B&W plants.
h 4
20 U DR. MATTSON:
Well, it is designed to sense less o
h 21 than full-power conditions, if you will, in the secondary l
hl 22 9, system, and to take actions with the crimarv svstem with the i
4 h
23 1 core to compensate for those things.
24 "ncipally, is an availability fearure of the B&W !
b I'
co eng co ccs j' i
25 design.
For e:.: ample, if the turbine trips it isn't necessarily
, :-i a..
p 0
i/
16 mm 1
true that the reactor has to trip.
It can be ramped down 2;
and power held at the low power using steam dumped from l.
3' secondary while the turbine is being put back on line.
4 From an availability standpoint, tha*'s good, 5
because you don't have to scran the reactor and then restart 6
from scrammed condition.
It increases the availability of the 7
machine.
8 The pilot operated relief valves in the ICS are l
9 a unique aspect of the B&W design to give that augmented i
10l availability feature.
I 11l MR. CASE: But it is not clear, even from a normal 12; operational standpoint, working as designed,that the ICS is 13l clearly advantageous from a reactor safety standpoint with l
14' regard to the reactor and how much heat is being dragged out 15 of the steam generators pending reactor scram.
j 16 l Nor is it clear at all from a single failure stand I
17 l
point, is it advantageous from a safety standpoint.
i 18i DR. MATTSON: That's probably worth elaborating on ll h
19: a bit.
l i
i t
20 h The staff has not reviewed the integrated control b
21h system on the B&W design. You have heard us talk in recent l
o 22 years, in recent months, about reviewing integrated protection!
23 -
t i
systems on things like the RESAR 414 cesign and the improved i
i d
i 24 ll integrated control system or the in tegrated protection system eno ccm, q 25 i! on the Bellefonte design by B&W.
<r!
- /
i '
I
17 l
Generally the philosophy has been, it is not 2 I safety equipment. That is, it controls normal operations, I
i l
3 makes secondary system adjustments and what havq you.
- And, 4
so long as it was kept isolated or separate from protection 5
systems or safety-grade systems in such a way that it couldn't I
6 interact with them and defeat them, it was not a matter of 7
intense staff study in the licensing review.
8 COMMISSIONER AHEARNE: But, does it interact with 9
them?
100 DR. MATTSON:
Pardon me?
I l
4 Il I
COMMISSIONER AHEARNE:
For example, in the Rancho 12 Seco system, does it interact with what you had classed as i
13 ll a safety system?
I I
I 14j DR. MATTSON:
So far not in a way that has shown I
t 15' us that there is a concern.
i 16i Now the Rancho Seco trip of last night begins to l
17 [
raise that question.
i 18!
Furthermore, in our discussions with B&W over the i
19!
last week or so, there appear to be questions that can be 20lj asked about interference with safety for which there weren't J
H 21l good ansvers.
22 Now that doesn't mean there aren't good answers, i
l 23j it just me, ins the last week or so we haven't been able to i
l n
24 J cenerate them.
j arm ce,em g.
j 25 COMMISSIONER AHEARNE: For example, does ir require --
I 4
l l7 1e-10/
18 7
1 does the system automatically take a pressurizer reading, 2l and then take action based upon wh1 the system believes 1
31 the pressurizer level --
4 DR. MATTSON:
No.
It does not conduct that kind 5l of control over the reactor system.
l 6
MR. DENTON: In the Rancho Seco case yesterday --
7 COMMISSIONER KENNEDY: Why don't you describe that, 8
since you have referred to it about 14 times.
9 MR. DENTON: All right.
Let's get Darrel Eisenhut, 10 who has got a story on what happened there yesterday, to walk 11 through it.
12 COMMISSIONER AHEARNE: I trust you mean an accurate 13 l description?
14 MR. EISENHUT: Let me just give you a very rough l
I 15 thumbnail sketch of what happened, because all we have at this i
16 ll point is a pretty rough thumbnail sketch.
J d
17 J There was an electrical comoonent failure yesterday 18 at about -- 'let's see, April 22nd at 1:08 p.m.
ll, 19 It led to a trip on an A inverter. Now the A l
U 20 g inverter supplies power to de A channel of the reactor i
f 21,a protection system.
e 22!
The reactor orotection svstem on this clant has l
t i
23 !
A, 3,
C, and D channels.
The inverter failure caused a loss l
24 o o f power to this reactor coolant -- I mean to the reactor w vco =-: n y l
25 :j protection systen channel A,
.thich is also being looked at by 4 -
i/
i ov 1,
o
19
- he integrated control system.
It doesn't interfere with what l
2 is going on in the reactor protection system, but it uses that 3
one signal.
4 It turns out the channel that did fail was the A 5
channel, and that happened to have been the channel that the 6,
integrated control system was looking at at that timo.
The 7
operator can use either A or B.
8i Looking at the channel, he got an indication at --
l'i 9 ' because of the loss of power to the reactor coolant flow 10 instrument, it caused a signal to be generated tothe integrated 11 control system which said that there was no reactor coolant i
12i flow.
I 13!
The integrated control system then shut off the 14 feedwater pumps, in the ense that it started ramping back the I
15 valves that are controlling flow to both steam generators. What' 16 ' it did was, it was starting to ramp down the valves to the I
i 1
17 point so you would get a low level in the steam generators, i
I l
18! somewhere between 15 and 30 inches, dependin g on the design.
4 4
19 Because the steam generator flow was being 20 4 the auxiliary feedwater flow was being ramped down, the 21 ] primarv system cressure becan to increase and caused the reactor o
t 22 h' trip on high pressure at about 2300 psig.
That was about 16 I
h i
P l
23 i! seconds into the transient.
l H
n 24 ll Now the 2300 osic is the new value we just -- there !
wvn; cr=,
i 25 p was a new value generated by the Licensee in response to our a
h 4,n l/
i G, h
20 mm I
bulletin that we issued on Saturday.
l The new set point caused the pressure to turn around; 2
31 and the actuaa,wassure peaked out at around 2330.
4 The aa.xiliary feedwater in this transient was 5
not called upon to start because the flow was still coming from 6
the main feedwater pumps.
It was not required to start, and 7
they did not start.
8!
The lowest point in pressure reached during the l
I 9j transient was about 1855 psig at about four minutes inte it, l
10 and it turned around and the system recovered.
Il TheiPORV set point, that is the new power operated l
12 relief valve set point on this plant was 2450, again in 13 response to the bulletin we issued on Saturday. And because 14j of that they didn' t get to the point.-This transient was some 15 l 120 psi below the point where you would pop a relief valve.
16 The pressure in the primary system did fall, as I l
17l said, to about 1855 psi. The operator manually initiated one 18 !
of the high-pressure pumps to bring back the system pressure, 44 19h even though he had not gotten down to the point where an I
20 automatic trip would have told him that he had to, or he 21 was just -- he took operator action.
22 He brought the plant back up and maintained the j
23 pressure --
!l 24 0 CO.'01IS SIO:iER KE: :IEDY :
How far would it have gone
\\
we,~; ce,ev l 25l had he not?
1j i^P I
i,u i
l
21 MR. EISENHUT: We are just not sure at this time.
)
i 9!
It would have automatically tripped HPI at 1600.
So if it
'l 3l would have gone down another 200 -- and he automatically 4l initiated at 1855.
The plant rode through the transient to the point 5
that, as I said, the valve didn't pop. Thepiant, as 6,
we understand it, came back on line this morning at about 7
^*"*
8 MR. CASE: But it is not clear to me that Se ramping 9
i down of the feedwater flow was the thing to do from a safety 10 jjl standpoint, which the feedwater controlling required.
0 1
MR. DENTON: With that big overview, let me go into 12 n w, some of the specific actions that were taken in the first 13 bulletin and that were taken in the second bulletin.
ja, If I can have the next slide.
15 16l (Slide)
I I
This is bulletin 05A, and I have just listed the j7,
dates:
It was issued on April 5th. The dates responses were 18 !:
a l>
I j9[ due back to us.
j f
201 Last week werwere evaluating these responses and this week we will be sending the results of our evaluation 21 I
I back to Licensees.
In some cases for action ad in some l
22!
I cases for more information.
23j l
So this week's acticity is the evaluation of j
24 n
r x teg s.rc ams.
25 responses to 05A.
{
i
+
4 - e t /
't i /
6 ',
i I
l 22 im I
Let me go to the next slide.
2 (Slide) i i
3 This will describe the impact of the bulletin issued 4
over the weekend.
l 5!
The bulletin had seven parts to it. The first 6.
part dealt with natural circulation, he need to train personnel l
7' on methods of establishing and monitoring natural circulation.
8' It picked up some of the ideas that were in the l,i il 9 l' ACRS letter and our own ideas about the need to maintain core i
10 subcooling before going to the natural circulation mode.
II It modified some previous instruction regarding 12 HPCI actuation to assure the vessel integrity, that temperature 13 limits were maintained.
14 The next slide shows the major change in my view 15 / that the bulletin required.
16 (Slide) 17 hl This was to modify the settings where the high-d a
181 pressure scram system and the opening of the pressuriner 19 [0 relief valve, such that for the type of transients experienced, h
20 in the past the valve would not open.
21 So I have listed here the old settings and the new i
22!
settings; and you can see the safety valve settings remain l
23 l
the same and the operating pressure of the plant remain the i
u i
24 j same. But, reactor trio coint has been firmed ccwn av accut w t -a camam ;
l 25 d 3
55 psi, and the pressuriner relief valve setting raised by f
7 i
i
1 23 b;
200 poi.
I l
s And calculations done by B&W indicate that this 3
would have been adequate to have prevented opening of that 4'
valve in all past B&W transients. And it was in the Rancho 5
Seco case.
6 CHAIRMAN HENDRIE:
Yes,you have already run a 7
test on it.
8 MR. DENTON: The fourth part of the bulletin required 9
that the operators be instructed to manually trip the l
10l plant on transients of this class.
I II As I mentioned earlier, B&W plants don't trip on 12 secondary system upsets, and this is sort of a softwire attempt 13 to provide the level protection that is available in Westinghouse 14 and CE planus to hardwire systems.
15, (Slide)
I i
16!
This next slide shows that we have asked these clants I
17 l to design and provide schedules for implementation of i
18 anticipatory trip systems, and this would replace the need for il 19 f -- (inaudible) -- scram.
l 20 (Slide)
I 21 The next slide notes the fact that it requires 22 liprompt notification from the utilities and the installation of,
23 ll an open communication channel, and requires them to submit 0
24 j c h a.. r e s to the tech scecs to confcrr with these chances to the
=mm ccmen j; l
25 [ license condition.
1 1
i I
ie a
24 mm j
COMMISSIONER KENNEDY: Where does this open 2
communications channel go?
I 3
MR. DENTON:
Let me ask -- in item 6, isn't this 4
to a regional office?
5 Where does the open communications channei go?
I 6'
I think it is to the local I&E headquarters.
7 MP. MOSELEY:
The regional office.
8 COMMISSIONER KENNEDY:
The regional office?
9 That means there is a 24-hour watcl. --
10[
MR. GCSSICK: Indeed.
ll 1
11 ;l COMMISSIONER KENNEDY:
-- in the regional of fices?
12l MR. GOSSICK That's been put into effect.
I (Slide) 13 MR. ENTON: This slide just putt into capsule ja form why we advocated those seven steps.
I have alretly 15 covered some of them, so I won't dwell on them again.
16 l
But the natural circulation issue is one that il dl we talk about -- has been talked about with the Commission 17l 18!
before.
i l
j9, I have covered the trip point in some detail, 1,
'i 20 j which I think that one ch mge will eliminate the principal u
lol 21 l' contributcr to the Three Mile 2 accident.
22!
The manual trip shown ca the next Vugraph, I have l
l 23 d covered.
24 (Slide' c
ocom c -en q 25 y So,
' won't go into those seven sceps, unless i
'l i
<^t I
l/
i'/'.
P
25 mm 1
there are questions about them.
2 So, finally I would like to then go to the six 3l factors the Chairman identified earlier in treating of this 4
accident, just list the total steps that have been taken in 5
response to each one of the six items.
6 1 (Slide) 7 So if you will go to the next slide, please?
8i (Slide) l 9l!
COMMISSIONER GILINSKY: By the way, you ought to l
10!
clarify on this keeping lines open. That's after they notify 11 us to keeo the line oc.en.
We don't have o.cen lines to all I
12; of them at the present time.
13 MR. DENTON: That's correct.
14 So, if I take the first contributor to the Three t
i 15 lI Mile 2 accident, the fact that both feedwater trains were valved Io 164 out, I have listed here the actions that we have taken through l'l 17 gl bulletins to preclude that condition from existing again.
18 4 I won't read them to you, but I think that should be fairly
'l 19 U effective action.
4 a
20 l COMMISSIONER AHEARNE: Is there any way that you 21 can devise a system by which there is a warning of some kind i
22 if both are locked ocen?
l l
~
l 1
23 MR. DENTON: Well, there is a wav that you could outi i
i i
24 h valve indicators on a computer, nor example, and have it sort wes c rea,1 I
I 25 of do a valve check for you.
With 1000 valves in a plant it
,6 i
1 /
l l /
6
,v I
r i
26 I
would mean a lot more electrical hardware.
mm 2l But, that certainly could be done not only for these 3
important valves, but other valves also.
I 4
COMMISSIONER KENNEDY: There already is an indication l
5 on the centrol panel as to the status of these valves?
6 MR. DENTON: Yes.
7 COMMISSIONER AHEARNE : Well, there may be an indica-8 tion on the control panel.
9 COMMISSIONER KENNEDY:
There is an indicator.
It 10 may not work.
That's a different question. But there is an i
11 indicator.
I i
12 i
MR. DENTON:
Yes.
1 13 COMMISSIONER AHEARNE : Do you know of any similar Id closing of both val es, locking them closed in other plants i
15 that have occurred?
16j MR. DENTON: Let me ask if anvone else --
i!
17 MR. CAEE:
I don't know of any.
i 18.,
DR. MATTF".4 :
I believe there have been valves of a
19I; that. type that have been found closed during inspections, h
h 20 4 h
- v. e s. Surveillance bv the c.lant ooerators.
hli 21 p Whether it was all be valves in Se aux feedwater dl i
i 2,4 system, I can't tell vcu the details off the too of T.v.
i l
23 j head.
I do believe they have been found closed.
I i
11 l
2.', ll COMMISSIONER AnEARNE:
Is the indicator that you w -c = =,,
9 25 were counting on in the -- on the control panel,that zellov i
^
1 ~/
i,-
\\d t
v
27 1$ tag that was put on the control panel indicating that mm Ih
? l'i! maintenance is being performed?
l' 3;
- 1R. CASE:
I think there is a valve indicator.
4l COMMISSICNER KENNEDY: The red and green light.
5 MR. DENTON: Yes, there is a light.
I i
6;l COMMISSIONER AHEAR*;E: If the tags weren't covering 7
it.
8i DR. MATTSON: Ther0 was some indication that may l
9ll ha:e been the case.
l l
10 ll CO?f1ISSIONER AHEARNE: Yes, i
11 MR. DENTON:
(Slide) i i
12; The second slide addresses the fact that the i
i 13!
relief valve failed to c. lose.
If the relief valve failed 14, to close,the actions we have discussed will ninimize the 15,l opportunity for that valve to open during off nornal and, I 16 !
think we have added instructions on how to detect instances
- I 17 where this valve nay be open.
Sort of secontari waya we 18 discussed earlier, such as quench tank temperatures and thermo-9 o
P I
19 ? couole indicators.
t l
20 9 COMMISSIONEP AHEARNE: Is that something that the o
21 )I operator would normally have been trained on?
22 ]
Is that a procedure that is embedded in the i
i 9
La 23 [l training of operators?
l 24 0 MR..DENTON. This crocedures was -- what I hae u
i
-engCver,q g
25 ij listed here are the actions thm the bulletin called out I
l!
H i:
4 l /
>i
I 28 iIh specifically.
mm I
I 2
l CO?CIISSIONER AHEARIE: I understand that.
t b
3 What I was speaking of, for example, as Roger g
I I
l-pointed out, you could use the thermocouple to indicate by 5
the fact that it hasn't come down, tat there is still steam i
I I
6 coming out.
7 Is that something that normally would be incorporatc)d 8
in the operational training?
9 l
MR. DENTON: I don't know.
I 4
10h DR. MATTSON:
I believe the answer is yes.
n ll 11 The operator knows when instrumentation is available 12 to him in the control room to indicate various things; some 13 of it more important information for maintaining the safety 14 of the plant than others, obviously.
15 MR. CASE:
It has not been completely clear to me, 16 hlat least, that this information is available for the operator i
17 h in the control room.
I 18 l, For instance, that thermocouple, I don't know if L
19[
it reads out in the control room or not.
Il 20!
DR. MATTSON:
It is my understanding it does.
l 21 COMMISSIONER KENNEDY:
If it didn't, where would it?
I J
I 22!
MR. CASE:
Oh, on the computer, or other places.
i 1
t' 23; CO:OiISSIONER KENNEDY:
Oh.
o'l 24 j CHAIRMAN HENDRIE: Or, it mignt be around on a orm
-=e, 9
25 backoaarc, or sometning 11.<e that.
l, kf I '
l l
l 11
29 I.
DR. BL;,TT SON :
It may not be on a front panel.
m j
COMMISSIONER AHEARNE: It is not absolutely sure 2j ih that we know that the infornation is directly, readilv 3
1 available?
4 MR. DENTON: My own view is that it is very important 5
i l
6l to keep this valve closed for these types of transients, and 1,
7 that was why I, myself, put more weight to the change than to the indicaturs.
g; I
1 9
COMMISSIONER AHEARNE:
I understand that.
h,i 10 ll (Commissioner Kennedy lef t at 2:05 p.m.)
11:
CHAIRMAN HENDRIE: I was coina to sav for whatever 12 m y be the case at Three Mile 2, this is the sort offdetail which could very well -
probably does vary from plant to 13 ja pl&nt.
DR. MATTSON: Yes, we know that.
15 i
I h
- 161, CHAIRMAN HENDRIE: So vou micht find it handy in i
I h
17, one olant and not necessarily handv in all clants.
18 h Let me ask, do we think we understand the i:
'l 19, secuence well enough to answer the following:
I 20 i The operators apparently recognize they have blocked feedwater, aux feedwater trains after a few minutes, 21,
22 something like 8 minutes he cleared those valves.
I i
23 l If, at about the same time he closed the block I
h 2,. ll l valve on the POR", would the plant have then settled down l
P wong cc cam,,
25 h, s o f ar as we
- ur.e.cithout --
- i l{,
l0 i
I e
I
30 mm MR. CASE:
I don't think it is clear.
2 I think the steam generator was dry at that point.
l 3l DR. MATTSON:
We would ha.<e had to kept HPCI l
ai on, keep the high pressure coolant injection system on.
I i
1 1
5 CHAIRMAN HENDRIE:
Had it fired at that -- something i
6l like 8 minutes?
7 DR. MATTSON:
E a r ls,' on he had taken it out.
g, CHAIRMAN HENDRIE: Been in and taken it out?
I 9
DR. MATTSON: Yes.
10l MR. CASE:
But the historv of on and offs of the l
11 HPI in that early pcrt of the transient, the first two hours, 12 is not completely clear.
I 13 I DR. MATTSON: That's a good point.
I ja COMMISSIONER GILINSKY:
Is there a record of that?
l 15; MR. CASE:
Well, Roger can speak to it. We've i
i i
16 h had a little discussion of it this morning.
17 DR.MATTSON: We are trying to put such a record i
18!
together.
'i I
19/
COMMISSIONER GILINSKY: But there is not this L
L 20.L speci:1c recorcer --
h 21 l DR. MATTSON: There is no scecific recorder of the 1
22 0 flow rate which is the hard cart.
l o
a l
23 j MR. CASE:
Or the interesting part.
o ti 0
24 DR. MATTSCS: We can
- ell uhen it was put on or a
wt-g Cc~ c e, p 25 g when it 'eas put off for all three pumps.
Two that are in the a
l f
/
- c. I s v
i 31 m:n 1
ECCS --
2 CHAIR'1AN HE"DRIE:
You have to know the system 3l pressure, and what the pump curves --
i 1
4 DR.MATTSON:
Well, no, it is harder than that.
5, The operator has manual c>ntrol of the throttle valve. And the statuc of the throttle valve is not recorded, 6;
7 printed or otherwise put down anywhere.
8; CHAIRMAN HENDRIE: Yes.
i 9 l' DR. MATTSON:
It is known that there was some I
10 throttling of high-pressure flow. The operators, evidently 11lhauestatedthat.
How much, when, seems to be not a very 12 clear picture at this point.
i 13 l What is clear is that th e A p ump, the B pump and I
14' the C pump, all the same kind of pumps -- B pump being in 15 the normal makeup mode, the A and C pumps being in the ECCS 16 mode -- were on and off for a period of hours.
15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> or i
17l so.
18i CRAIRMAN HENDRIE: Do you have any feeling, having n
o 19[
gotten started on this transient with the steam generators t
4 h
4 20 e going dry at what, 3 or 4 minutes, something like that, if 1
i i
21 1
you get aux reedback at something like S,
10 minutes, and i
i
?
22 h close that relief valve, could you then ridathat out for l
0 i
about a the rest of these circumstances?
l 23l l
24 Jo you have any feeling for that?
wtmg crea, 3 25 p MR. RCSS: We recentiv asked for some calculations f
I n.
o i
l l /
- t.. iJ s
32 I,
because there is a clear inference with natural circulation mm 2
path Tay well be lost at that point in time.
14 3l We are currently looking at some calculations and 4
the outcome is uncertain.
It looked like there is one of 5l two things that could happen:
6 The system -- where ycu have lost your heat sink --
t 7!
so vour system would presumably repressurize, and either f
8l collapsa enough voics tc tastere the natural circulation path, I
9l; or alternatively your operator may start hi~h-pressure i
h 10 p injection and leave it on.
d I I h'l In some cases there is three pumps, it may start l
I2l all three of them if he has enough power.
The.could also I
13l restore natural recirculation.
I
~
14l To answer vour cuestion, I don't have a feeling today 15 as to whether either these things would or would not work. We 16 daat have all the information that we need.
!l I7 N We have some calculation,but it doesn't show n
1 I8 0 saturation conditions aro""a "ma locp, it doesn't show what i
n I9 voids might form. So we just can't give an answer tcday.
h 20 h Several people are back at the shop today reviewing 1,
il 21; this recent analysis and hoping their car arrive at an answer 22'I to tnat question.
23 p
COMMISSIONER AHEARNE: So you are really saying it I
Il fl 24 ":
really is too early to reach any firr conclusicns eitner c:-
cm,no cc-o. g 25 !l what happened, or in the process of what happened if something!
a ij
< ~;
h i/
- c. o u
.1 l
l
I 33 mm 1
different had occurred, what the result of that would have 2
been?
i I
MR. ROSS: Right.
3l 4
Specifically, once you lose natural circulation 5
due to voids, what are the ways or way if any to restore it?
6 And that's what we are looking at today.
7 DR. MATTSON:
Part of the problem is, you don't l
81 know what the void is for certain, at that point in the i
i 9l transient.
If part of the void is the helium-fill gas I
10' from the perforated fuel.
Because if you went through an 11 initial' ramp and perforated fuel, probably not perking (?) _ fue:
12 in any subscantive way, but losing some plenum gas, then it 13 would be a problem with the non-condensable gases in the 14 system.
I 15 '
If there weren't non-condensable gases in the 16 system --
l 17!
C;GI?liAN HENDRIE : Probably not a great volume, 1
1 18 !
though at pressure.
191!
DR. MATTSON: Not a great volume, especially at l
20 b pressure.
21 If chey are not a problem then in the limit you i
22 can see if you drive the pressure up and reestablish the 231 secondary cooling, clearly the machine will go into I
i 24 q natural convection cooling.
r -
m s _mn ll 1/
4
's "
25 It is autonatic and a natural tning that happens.
34 So the question becomes the time that it takes to mm j
reestablish a natural convection situation within the primary 2l I
i system have you damaced fuel?
3 MR. DENTON: Let me run through the next three or 4
I four items.
5l 1
6l I.think we have hit -- maybe we haven't hit the l
third one which was the pressurizer level indicator.
7 (Slide) 8, I
9l Bulletins have dealt with this one in terms of 1
1 perator training and the need to recognize other information 10 I
jjj sources.
(Slide) 12 The next slide we got in the --
13 COMMISSIONER AHEARNE : Could you go back just a ja minute?
15 (Slide) 16 MR. DENTON:
Back to caracranh 3.
17 jgld COMMISSIONER AHEARNE :
In the evaluation of the j9 j information from other sources, are you now satisfied that i
i 20 !l there are enough other sources that would lead the operator l
21 to be able tp reach a non-ambiguous interpretation, or is it 22l that there are other sources that would lead him to raise --
i i
t 23!
question --
!i MR.DENTON:
I think vou have to look uOon aE these 2 n.
n cortmg Ccerans ;
25 bulletins as short-term instructions.
You %now, they don't h
4
-r n R 1i s.. { s I
i e
s l
35 I
preclude a long-term improvement inthe pressurizer level.
mm 2j
'I think the bulletin provides as nuch advice on 3
how to get information from the existing system as you can 4
get from it.
But I don't think we have completed our review 5j of pressurizer le rel instrumentation and whether or not in i
6 the long term it couldn' t be a much better system or not, il 7
COMMISSIONER AHEARNE:
Is it the 3taff sense that 8,
uhere are enough ways for the operator to c ' the indication i
I 9 !
that given the bulletin now alerting them that here's a real l
10l problem, make sure that you don'-t just look at the pressurizer l
Il level, is he able to interpret with the information he has i
12 available?
13 MR. DENTON:
We have a report being written by 14 Bob Tedesco that looked into this --
15 MR. CASE:
It really depends upon how many times 16 he gets -- what's the probability of him getting into a 17 situation where he has to infer things about level and i
18 pressure or what to do about it, getting back to natural il 19f circulation from other instruments.
l 20 h MR. n?" TON: Do you want to give us a forecast of 21 wnat our group is going to say in that area?
22l DR. MATTSON:
Well, I think they are going to say l
23 essentiall" the same kind of thinc vou can imo.lv from I
24 ll bulletins.
cc-., jj m ns 25 That if you look at the operating history of B&W D
[
'I
- j
..~a h
a
36 mm 1
design there are a lot of feedwater transients and a lot of 2
pilot operated relief valve openings, and too many of them 3
that stuck.
4 If you look at the design to try to explain that 5
operating history you sce a sensitivity of the design to 6
feedwater : transients not only in their occurring, but in 7
what happens once they occur, how hard is it to control a 8
machine, what kind of burden does it place upon the operator?
9 Then if you look at operator training for the off-10 normal, abnormal situation if you will,that is for the 11 anticipated transient, what kind of preparation does the 12 operator have, what kinds of "what if" questions has he 13 been prepared for, what kind of failure modes in his integrated 14 control system which under normal circumstances is like an 15 automatic pilot, would take him to a natural circulation 16 while he just stood and watched it.
I 17 What kind of preparations has he received for the 18 "what if" events, the equipe.ent failures or the controller 19 failures or the slightly different transient than the one 20i that was considered in the design.
21 Then finally Tedesco will ce looking at the 22 licensing area, how did we miss TMI 2 in our licensing i
23 review?
What are the holes, like not reviewing the integrated 4
h 24 H control systen, what are the "what ifs" that if we had asked 4
eng ce-ca v 25, them, even though the designer apparently wasn't asking enem, i
l
\\
"E t/
t.ou l
I
I 37 mm 1
we might have caught this transient in the licensing process.
2 Clearly we didn't.
3 And, how ought we to change to see that we ask 4
the right "what ifs" in the future.
S' (Commissioner Eradford arrived at 2:15 p.m.)
6 COMMISSIONER GILINSKY:
Is it fair to say that we 7
are developing a different view of what is safety-related?
8 DR. MATTSON: Oh, yes, sir.
9 Maybe I should say more than just "yes, sir."
I 10 Fundamental philosphy has been that the large 11 loss of coolant accident was the contributor to the pu _ic 12 l risk from these machines, and that anticipated transients I'
13 did not have pathways to significant consequences.
i4 There has been cautions to that approach that have 15 been offered before by the Reactor Safety Study and the Lewis lo Committee.
TMI 2 seems a stronger caution that there are 17 pathways between anticipated transients and off-site 18 consequences.
I, 19 And if you consider those pathways and look at i
l 20 what constitutes a safety system for interdicting those l
p 21 q pathways, then you will be looking at different ecuipment than il 22q you look at for loss of coolant accident. You won't just be e
b 23 q looking at emergency core cooling systems, you will be looking l
24 ;t at controllers, you will be looking at feedwater systems in It
- oc< ten g Comca v l; jf 25 q ways that they have not been viewed in the past.
e,
- /
6.
l/
- t.. V +
e I
38 m
1 COMMISSIONER AHEARNE: I guess, Harold, my concern 2
is that we have your staff, B&W people, all of the cluster of 3
people who are probably the most competent technically to 4
address these problems, wrestling with how co you control I
S the sensitive plant.
6' Do you have the sense that when you get through 7
with these bulletins which we are now giving to all the 8l operators, that they have a clear enough picture of what they 9;
ought to be doing ut. der these kinds of circumstances?
i 10l MR. DENTON: We asked I&E to ask their resident 1
11' inspector to look into that area.
Maybe this would be a good 12 time to get a report from I&E on how it is viewed by the 13 recipients of our bulletins.
14 COMMISSIONER GILINSKY:
How does it look from the 15 other end of the telescope?
16 MR. DAVIS: One of the items that we intended to 1
17 cover in our --
i 18 COMMISSIONER AHEARNE:
If you want to hold off unti.
19 l
you get to that, then I can just reraise the question.
20 I MR. DENTON:
I think my point in walking through i
21 these five areas is just to illustrate the bulletins have i
22!
issued instructions of required actions to deal with the I
i 23i six areas that we identified.
24 L And I think if these were the only six we woulc n
oting Cc-=ary 25 lt be more sanguine today about our ultimate views en B&W plants.
i/
l/
s U, v l
39 mp 1
But I point out that because of our view now that 2
auxiliary -- well, feedwater transients themselves and the 3
design of the plant being more sensitive to that kind of 4
transient, that maybe those six items don't do enough and i
I 5'
maybe there should be a seventh area as a contributor to 6
this cause of the accident.
7 That is the sensitivity of B&W designs to off-8i normal conditions.
Wh,reas the bulletins have addressed 9
each of these previous six areas that we have identified.
10 MR. CASE:
Could I add a few words to this?
I 11 I don't think I have any serious or any technical 12 difference at all with Harold as he has laid out the problems 13 or what we are doing about it.
14 My concern is thi the NRR staff is just about 15 saturated with current TMI-2 related activities. Wa have other 16 tnings coming up in the near future, seismic actions coming l
17l due, the five-plant shutdown and new seismic bulletins to 18 L respond to on which responses are due soon.
i l
19 We have laid out a fairly significant area of d
20 L action with B&W that is going on now and we all agree must 21 dentinue.
Those are:
Meeting with B&W to define and resolve i
22 the possible generic concerns. That incluces thermal hydraulic i
23 t analyses of stuck open pilot operated relief valves, and I
i 24l indeed stuck ocen safetv valves which, if you remove or increase ony cc-cev l l
25j the set point of the POR",then you have to concern yourself O
I I )t l
l
40 1
with safety valve opening.
2 Another point we are going to have to work with 3
them is on the hardwiring of the anticipatory scram to make 4
their designs comparable to other PWR vendors.
5 We have a.large job ahead of us meeting with the 6
utilities of the B&W operating plants.
We have to get into 7
details of the design of aux feed systems on these B&W 8
operating plants.
9l As Harold indicated,there are indications that the 10 reliability of the aux feed system is not what we would like 11 I
it to be in light of our post-TMI-2 knowledge.
12 We have to meet with them to discuss the design 13 in single-failure modes of the feedwater controller system.
14 We have concerns about it and I think aptly illustrated by 15 what happened at Rancho Seco.
16 Based on all these, we will have to come up with 17 some fairly definitive, I believe, operating instructions for 18 the plant operators because I believe that there currently 19' l
exists a considerable mismatch between operator training, 201 operator experience and the machine,the B&W machine.
A lot 21 is expected of these operators and clearly more San our i
22!
present training program and education and experience program 23 l1 contemplates.
1 24 i
Because we are so saturated at this point,and we
-sco-mvl!
25 have so many things to do, I believe I would feel more I/
^ '. p c.
v l
41 mm l'
comfortable as a precautionary measure, that the operating 2
plants, B&W plants be shut down pending completion of these 3
high-priority staff efforts.
4 CHAIRMAN HENDRIE: Let's see, Ed. We have begun to 5
cycle -- or the operating utilities have begun to cycle the 6!
operators back through the simulator running against i
l 7
transients of the Three-Mile kind, and associated secondary 8
side perturbations and so on.
10 CHAIRMAN HENDRIE: There are also, I suspect, a l
11 i
number of burdens thi the staff has at the moment that press 12 them hard. At least some of them, I know, are Commission-13 generated, and I.think we are goi.T to have to let Harold I4 sort out the priority and the order in which he gets to these 15 and expect that indeed, some of them will take a time to 16, answer.
17 There have been a number of requests for information i
t 18t of various kinds of activities and so on that have come down I
190 from us, and I think we will have to simply recognize that i
20 b the staff cannot do everything simultaneously and so on.
21 Harold, with regard to Ed's cc= ment,what do you 22 feel te balance -- what is vour coinionund what are the other 231 peoples' opinions on where does the balance lie.
24l COMMISSIONER AHEAR' E : I will have to say I a-
.oteg Cc-cany {
25!
leaning in Ed'c direction.
^1 3 -l,
\\
s-l
42 1.
MR. DENTON:
I guess I'm not quite as pessimistic l
mm 2l j
as Ed, but it may be that I haven't been back in town as long 3l as Ed has been and facing all the other issues.
4 I think the bulletin impact on the sequential 5
opening or this valve -- (Inaudible) -- transients is a 6
big step forward.
My main concern at the moment is with the 7
quality of the auxiliary feed systems in the individual 8
plants themselves.
And we haven't quite -- at least I haven't 9
quite reached the same conclusion Ed has about the proper 10l l
action on those.
il But, it is conceivable that in the next day or two 12 as we look into them and if they are as unreliable as some 13l people think, it is more down like 1 chance in 10 of having 14 to rely on emergency cooling systems to ride through these 15 types of transients per year, I think that's an unacceptable 16!
i value.
i 17l COMMISSIONER AHEARNE: That is what?
18I COMMISSIONER BRADFORD : Acceptable or unacceptable?
19:
MR. DENTON: Unacceptable frequency for having 20l to rely on ECCS perfornance for off-normal conditions.
21 If we, in fact, find that number may be too 22 pessimistic and the number is more up like 1 in 100 or t
23l higher, then I think I would have a different view.
24 !!
.,,y cm, d But the staff -- we just haven't had a chance ye f
25 a
H f
l/
t.
43 I
to focus in on that number on individual plants. Because in mm 2
order to get a number on individual plants you have got to I
3 look at fault trees and look at the type systems it has.
4 But it may be that low.
5 COMMISSIONER GILINSKY: When you compared it with 6
the other types of plants, you said their number was a factor 7
of 10 lower.
8 Does their range start at 1 in 100, or 1 in 1000?
9 MR. DENTON:
I guess you can start with the 10 I knowledge that in no CE or Westinghouse plant, has this Il pressurizer relief valve ever been called upon to open in 12 off-normal transients.
13 So whatever their~ failure rates are in the 14 auxiliary feed systems, they are such that in the experience 15 accumulated to date in PWRs of that type, it has never been 16 called uoon.
I i
17l We know he f act that it nas less water than the 18 B&W plants and therefore you have less time by a factor of 19h 3 to 5 in order to restart pumps or take other corrective 20 actions.
So there is a time problem in the B&W plants even 21 if the auxiliary feed systems are the same as CE or 22l Westinghouse plants.
23:
Then this is compounded by the uncertainty as to 24 how this integrated control system will really respond to eng c v,;
25f certain transients, and that may go the wrong direction.
Il i
~
- I u.-
44 I
rc So, if they are B&W plants who also have 2
auxiliary feed systems that are very unreliable, I guess my l
3 concern of a continued operation of those would be considerable 4
if they turn out to be as frequent as 1 in 10.
5 If we find they go the other way, then I view 6'
them, with the bulletins that are out there, I view these 7
plants as being more comparable to the CE and Wcstinghouse 8
plants.
9 COMMISSIONER BRADFORD: And are you saying that it 10 will really only be a day or two before 'you would have a 11 firm grip on those probabilities?
12 MR. DENTON:
I'm not sure that it would only be i
13 j
a day. This problem has been looked into, I think, with 14 Increasing interest.
15 My first concern was the fact that the valve 16 could open as a sequential failure.
And having dealt with 17 that, I have now started to focus on the reliability of 18 the auxiliarv. feed svstems themselves.
I 19 1 I would think this week we would have a much I
20 better view.
21 j Now we are meeting later -- we are meeting, I 22 think tomorrow with Westinghouse, the next day with CE --
23 DR. MATTSON:
Westinghouse today, Ccmbustion 24 4
tomorrow.
meo cream 25!
h DR. DENTON: So in a few days I snould have a much 9
"1:
l/
u..
I I
40 I
l' l
with safety valve opening.
2 Another point we are going to have to work with 3'
them is on the hardwiring of the anticipatory scram to make l
d their designs comparable to other PWR vendors.
i I
5 We have a.large job ahead of us meeting with the 6 !
utiliti es of the B&W operating plants.
We have to get into 7
details of the design of aux feed systems on these B&W 8,
operating plants.
9 As Harold indicated,there are indications that the 10 reliability of the aux feed system is not what we would like 11 it to be in light of our post-TMI-2 knowledge.
12 We have to meet with them to discuss the design 13 in single-failure modes of the feedwater controller system.
14 We have concerns about it and I think aptly illustrated by 15 what happened at Rancho Seco.
16 i
Based on all these, we will have to come up with 17 some fairly definitive, I believe, cperating instructions for 18I l
the plant operators because I believe that there currently 194 exists a considerable mismatch between operator training, 20' operator experience and the machine,the B&W machine.
A lot 21 is exc.ected of these coerators and clearly mora San our 3
221 i
present training program and education and experience program 23 l
contemplates, i
24 :
_,,,,,c,m[
Because we are so saturated at this point,and we 25!
have so many things to do, I believe I would feel more 1
l4 b
I 41 mm 1
comfortable as a precautionary measure, that the operating 2
plants, B&W plants be shut down pending completion of these 3
high-priority staf f e f forts.
4 CHAIRMAN HENDRIE: Let's see, Ed. We have begun to 5;
cycle -- or the operating utilities have begun to cycle the l
6 i operators back through the simulator running against 1
I 7!
transients of the Three-Mile kind, and associated secondary l
8i side perturbations and so on.
I 10 CHAIRMAN HENDRIE: There are also, I suspect, a li 11f number of burdens the the staff has at the moment that press i
I2l them hard. At least some of them, I know, are Commission-l 13l generated, and I :think we are going to have to let Harold I
14 sort out the priority and the order in which he gets to these 15 and expect that indeed, some of them will take a time to i
16 li answer.
I 17 l, There have been a number of recuests for informatio:.
I 18 of various kinds of activities and so on that have come down 1
19h from us, and I think we will have to simply recognize that i
i 20!
the staff cannot do everything simultaneously and so on.
21 Harold, with regard to Ed's comment,what do you I
22!
feel be balance -- what is _vour ooinion ard what are the other j
i 23l peoples' opinions on where does the balance lie.
l 24 ',
CO.'Oi!SSIC':ER AHEAR. E : I will ha'ze to say I am i
mes cc-en j
25 jl leaning in Ed's direction.
I I
i/
l l4 c
.c i
42 mm MR. DENTON:
I guess I'm not quite as pessimistic 2
as Ed, but it may be that I haven't been back in town as long 3!
as Ed has been and facing all tne other issues.
4 I think the bulletin impact on the sequential 5'
opening of this valve -- (Inaudible) -- transients is a 6
big step forward.
My main concern at the moment is with the 7
quality of the auxiliary feed systems in the individual 8
l plants themselves.
And we haven't quite -- at least-I haven't 9l quite reached the same conclusion Ed has about the proper 10 action on those.
11 But, it is conceivable that in the next day or two 12 I as we look into them and if they are as unreliable as some 13 people think, it is more down like 1 chance in 10 of having 14 to rely on emergency cooling systems to ride through these 15 types of transients per year, I think that's an unacceptable 16!
j value.
17!
l COMMISSIONER AHEARNE: That is what?
18 COMMISSIONER BRADFORD : Acceptable or unacceptable?
19 MR. DENTON: Unacceptable frequency for having 20!
to rely on ECCS performance for off-normal conditions.
21 If we, in fact, find that number may be too 22 pessimistic and the number is more up like 1 in 100 or 23 higher, then I think I would have a different view.
24' But the staff -- we just haven't had a chance yet m,,,, c,c,, q 25 l
1/
i lI I
43 I
to focus in on that number on individual plants. Because in mm l
2l order to get a number on individual plants you have got to 3
look at fault trees and look at the type systems it has.
4 But it may be that low.
5 COMMISSIONER GILINSEY: When you compared it with 6
the other types of plants, you said their number was a facter 7
of 10 lower.
8 Does their range start at 1 in 100, or 1 in 1000?
9 MR. DENTON:
I guess you can start with the i
10l knowledge that in no CE or Westinghouse plant, has this I
11' pressurizer relief "alve ever been called upon to open in 12 off-normal transients.
13 So whatever their failure rates are in the 14 auxiliary feed systems, they are such that in the experience 15 accumulated to date in PWRs of that type, it has never been 16!
called uoon.
I l
17l We know te fact that it nas less water than the i
18 ;l B&W plants and therefore you have less time by a factor of I
19 3 to 5 in order to restart pumps or take other corrective 20 actions.
So there is a time problem in the B&W plants even 21 if the auxiliary feed systems are the same as CE or t
i 22I Westinghouse plants.
i 23l Then this is compounded by the uncertainty as to i
24l how this integrated control system will reallf espond to mna co-c.,,,
25 certain transients, and that may go the wrong direction.
,s l#
V
Y l
44 I
rz So, if they are B&W plants who also have 2
auxiliary feed systems that are very unreliable, I guess my 3l concern of a continued operation of those would be considerablu 4
if they turn out to be as frequent as 1:in 10.
5 If we find they go the other way, then I view 6
them, with the bulletins that are out there, I view these 7
plants as being more comparable to the CE and Wcstinghouse 8
plants.
9 COMMISSIONER BRADFORD: And are you saying that it 10l will real]y only be a day or two before ' you would have a 11 firm grip on those probabilitics?
12 I MR. DENTON:
I'm not sure that it would only be i
13 a day. This problem has been looked into, I think, with 14 Increasing Interest.
15l My first concern was the fact that the valve 16!
could open as a secuential failure.
And having dealt with 17 that, I have now started to focus on the reliability of 18 I
the auxiliary feed systems themselves.
I t
19i I would think this week we would have a much I
20 i
i better view.
21 l
. sow we are meeting later -- we are meeting, I i
22l t
think tomorrow with Weitinghouse, the next day with CE --
23 li DR. MATTSON:
Westinghouse today, Combustion 24 il ton rrow.
m,a ceco ~ j 25 i,
DR. DENTON: So in a few cays I snould have a much
'iD 1/
45 mm 1
better view of it.
2 But it is quite correct that the knowledgeable 3
staff members in these areas are fully occupied and the 4
quality of our review and our ability to answer authoritativel:-
5 the types of questions that we would like to have answered, I
6l.
is down from where we would have been in normal circumstances.
7 COMMISSIONER BRADFORD: Is the calculation of the 8
probabilities there throwing you back on the Rasmussen l
9l fault trees and event--trees?
10 MR. DENTON: No.
It is more using some of that 11 methodology, but actually basing it on the plants as 12 designed.
13 COMMISSIONER GILINSKY:
How man;' plants are 14 operating new? Are there four?
15 MR.
DENTON: There are only four B&W plants now i
161 operating.
I 17 COMMISSIONER GILINSKY:
Crystai River is shut
- I 18 ll down?
dl 1
19 ]
I have a table if I can find it.
I 20 ll MR. CASE:
Oconee 1, 2,
3 and Rancho Seco.
21 MR. DENTON: Crystal River shut down over the 22 l weekend or early this morning, I have forgotten which.
23j COMMISSIONER AHEARNE: Do you have an estimate of l
24 L what the imoact 'eculf be if we did shut down those olants?
w uco cw=a, 25 What kind of impact onthe grid system?
j '/
.. ' [v
' ' ' 'i i
I g
ll
]l 46 l
1 MR. DENTON:
I asked someone to look into that, m
j h
,[
Let's see if he managed to get down here today.
n il 3
MR. WEINER:
Sir, I looked into the situation on I
a1 rather short-range this morning from a three-part standpoint:
F 5
The reliability of the grids, the voltage level of the grids, 6l and the stability of the grid'.
i The Oconee situation for all th ree units removed 7
g would leave them with a 15 percent reserve margin over the I
9, summer's peak if the Crystal River unit -- and this is 10 considering the Reliability Council that they all operate in.
i 11 1 They are in the same Reliability Council, they would be 12,
at 12 percent reserve.
I 13 [
Wi' h the Crystal River unit out, the tie support to e
H 14 !I the Reliability Council with these four units in it because U
4 il 15 [ of their ties to Three Mile Island Reliability Council and i
i 16 L Davis-Besse, which would be considered in this ultimately too 4
4 1
- 7 j for the same type of unit, would eliminate half of their tie h
4 18 capability.
U 19 So I would say from a reserve standpoint, this 20 system would be marginal to poor over the summer peak.
21 The situation now -- this is a time of the year 22, they are depending on doinc
,r system overhauls in their
-'e 23 other units, plus they art ing to swing into the peak 24 period, although they don't predict peak until July and a+
. n eoews. me.
25 August. The situation now is one that would still be marginal e o l $
l.9_
47 I
I I
I m
as far as bringing these units off the line.
That can be done, IL 2
but it would be marginal.
3; The voltage levels around Oconee can be cupported j
4 somewhat from the high-voltage system that they are tied to, i
C as best I can determine from what I looked at so far. And i
6' because of the larger units that are in the same area.
I 7 j But stability, system stability situation is again I
i 3l a marginal situation. They do depend on their neighbors in i
9' the north, ultimately tie through the Carolinas into the 1
10 ) Vepco system where they have the Surry units shut down for i
11 l the carthcuake consideration.
t I2 l So in summation, the situation from a voltage l
13 H stability and a reliability situation, I would say were all l
t
~
Id d marginal, and over.the summer peak would degenerate to poor.
P 15 f COMMISSIONER AHEARNE: And that's in a recion where i
16 ( Oconee and Crystal River --
f'l II d MF. WEINER: This is *.he Oconee/ Crystal River n'l I3 situation.
u CO:iMISSIO.4ER AHEARNE : How about Davis-Besse or 20 Rancho Seco?
21 MR. NEINER: Davis-Besse, the reserve situation there o,
is fair.
Again Three Mile Island 2 unit and Fitzpatrick
,a e a coorms, inc.
to a coor situation. Thev are in the same decenerate that
,C Reliability Council.
^~
.O
\\l 1
l 48 i
i i
I i
i h
1]
The Rancho Seco situation in my opinion,being im 2% quite familiar with California, I think is no problem. The 1,
3 ;i West Coast seems co be in a surplus situation over this period il il 4;
of time from good water conditions this past year.
1 5
COMMISSIONER AHEAR'JE: Thank yau.
61 MR. DENTON:
I guess where I come out is that my i
7l principal concern is on the reliability of the individual i8' auxiliary feed systems in these individual plants.
And if i
i 9'
it turns out that once we've had a chance to look in a little 10 more detail, and I've had a chance to get'a better feel for i
11 !
it, if it is low like 1 in 10, the probability of events 12 leading to operation of ECCS, I wouldn't hesitate to come back
!l 13 L1 and make the same recommendation Ed did,that these plants 14 also be shut down.
4 h
15 There probably are some things that dan be done to t
16 l ameliorate the situation, if it turned out to be that bad.
.i 17 o We've not had an opportur yet to explore ways to improve I:
P 18 1 the reliability of the exio ung aux feed systems, and the h
19 focus of at least;ny efforts have been to try to reduce the 20 e reliability of the pressurized relief valve opening.
And 21 ' we are now turning to this next problem, at least in my a
22 L mind down one tier, but not that far down that they can be 2
ignored.
2' COMMISSIONER BRADFORD: Do you have enough operating
..,. a
. Re x ners sec 25 data cn 3&W plants for probability judgments to have much meaning l
L.
49 i
l l
t-IU NIR. DENTON: The auxiliarv feed svstems are u
9 crobably about the same in mann clants, and there is not
,1 that much unique about the B&N systens.
1 4
'l Nhat is unicue is that the B&N system is a lot nore 5j vulnerable to uosets in the svsten so I want to look to see
,1 46j if_these coor systems are cresent in B&M svstems, and 7
whether there is anv way to comoensate for then or not.
i l
8; So we are down to iust four clants in the short-9' term that we have tc look at, Rancho Seco and Oconee.
Oconee 10 )
on first blush acoears to have a very -- has only one svsten dn II per olant which would out it in a lower reliability cateaorv, 12 '
whereas Rancho Seco has at least two systems which would 4
13 4h move it un a bit.
l Il L
But, I've only done iust a first-blush look and L
15l4 i
I've nct attemoted to make a detailed look in either olant 16 i
vet.
1 U]
Nhat vou are hearing ton ie sort of interruotion 1
1 ig a
. a meeting at noontime, and it is sti going on right now ot back there.
20 CO'NISSIONER GILINSKY:
Is the Oconee reliability 2I increased.2 there is less than one reactor coerating'
>ENTON:
That Nould be a nossibility that at 22 least that feed system would be available t) feed the other 2
two units.
They are alreadv cross-tied.
&r 1 ReDCffers IN.
25 CO'NISS IONER GILINS'"i-
"4"
' ass than three reactors a
h L '"
l o
1 0
50!
oI' I
d Io mm coerating.
I'm not sure whether I said that.
COMMISSIONER AHEARNE: Thev are interconnected?
i 3
MR. DENTON:
Thev are interconnected.
4 But there is some cuestion about how lonq it would i
5 take to realign t.t. valves to nake the interconnections act 6'
oromotiv.
i 7j COMMIPSIONER GILINSEN:
I see.
It can't be just 8
done instantiv?
I 9
MR. CASE:
No. It is 8-to 10-minute tyoe oueration.
q 4
10 j COMMISSIONER GILI'ISKv:
I see.
.i l.
Il So we are back to that oroblem.
12 COM'iISSIONER AHEARNE: That's how long it took at t
'1'
- m. it T.
m U
IJ DR. MA'2TSON : Mith one unit down though, we can nake 15..
the interconnection.
16 '
COMMISSIONER GILI'ISKY:
I see.
You can make the 17 interconnection beforehand.
I8 MR. DENTON:
Yes.
u H
I9
'Iow CE olants for examole -- I quess most of their 20 clants are not automated with regard to starting of auxiliary 2I feed systens.
This has to be manual.
So that's another om
-ractor to take into account in a comparison.
23 P.CE: 3ut, one of the allle & q rrr u re 4 that an oneren:
24 be assigned with that as his onlv function, nanels 2tuating
.,.c
, nexem w 25 aux feed.
So Ne have at least bridced that a bit b" One of l /
i i s
~
51 l
o
!l 8
m 1h the actions of the bulletin.
d a
2[
.4R. DENTON:
So the tvpes of actions that we are 31 talking about:
'l 4j One, to try to reduce the amount o# stored enerav d
5I following vour transient. You do that bv having anticioatorv a
d 6'
scrans such as Nestinghouse or CE cresentiv have, and we 7 ;.
are recuiring B&W olants to design and crocose.
d 8 )I, And in the iterin, we are using the lower nressure l
l 9
setting and the manual scrim to orovide anticioatorv signal.
10 Then vc" i
'Oe a reliable auxiliarv feed f
Il systen that ha:
9
.iity of not coning on in this tvoe 12 '
o f transient.
13 CO'4'1ISS & '... ME h 'E :
Do vou have data from the Ho 14j ooerating exoerience that will enable vou to net to the d
15 '
question vou were asking about the reliability o# aux feed 4
16 "
systems?
17 '
' iR. DENTO' :
I think it is iust finding the time 1
15 for experienced engineers in this area to look at the system 19 ticone detail and give us an estimate on how reliable thev 2C think it is.
21 CO'NISSIONER MEAR'IE : As ocoosed to using oneratinc 22 data?
22 DR. '!ATTS ON :
Be a cure ecuienent availabilit-- kind o 24 analvsis.
Ne would look at things like single failure, examine
- rs nex en sne 25 then in current licht.
4
^'/
{/
..v
t 52 I
mm 1 ll MR. DENTON:
Certainlv a mixture of stean driven u, -
L 2
and electrical driven provides that sort of redundancv we are 3I looking for.
I a.
4 '
CO'PIISSIONER GILI'ISKY:
Nhv are there more feedwater i
5' transients in the first olace in these clants?
6i MR. DENTON:
I think there are one to two feedwater '
l V
7l transients per year in each PNR.
)
i 8
COMMISSIONER GILINSKY: I thought vou said the
[
9h frequencv in these olants was greater.
10 DR. MATTSON:
Slichtly over two for B&N, slightlv h
11 d under two on the average.
i 12 CO'1'iI95IONER GILINSKY:
I see. So the di#ference i
13 there-is not that great.
I!
I 14 n,'
CO'DiISSIONER AHEARNE: But the volume of water --
o 15 DR. MATTSON: The sensitivity of the #ee^nater 16 "
transient is different.
I 17 MR. DENTON:
Let me ask, anvone else, Roger or J
1E Dennv, would you like to get vour views hear'd on the overall 19 situation, where we stand' 20 You've heard Ed and I --
21 DR. "ATTSON: Nell, I think I come out closer to Ed.
22 The design is nore sensitive.
3&M acoarentlv hasn't designed 22 or trained ooerators for the o f f-normal failure modes %inds 22 Of things that can hancen in anticinated transients.
n x,w m c.e e
23 Ao7arentl" there are scne older clants that dor't n
l/
m._,
I I
53 I
o i
Ih have the same qualitv of aux feedwater as is currentiv mm required for new olants.
I know Ehat in the licensing crocess we have not looked at the intecrated control systen and the 4,
failure nodes that it nortends for anticinated transients.
d 0 '
We never looked at the oilot-coerated relief valve.
It was
=
6l1 not a safety feature, it was an availability feature and 7
thought not to have safety significance.
8 Ne have not looked in detail at the effects on I
9 the reactor of loss of all feedwater. Thr comes close to one i
10,i of your unresolved safetv issues, that is station blackout, 1
II but not cuite the same.
It is a cernutation and combination i
I2 of that.
I3 f Ne have not looked closelv at a critical rance of o
j,!.
small loss of coolant accidents when exacerbated bv initiating H
15 d transients of the sort that occurred at Three $ tile Island 7.
1o i
16 '
The ability to look at all of those things, given i
'7 the need for close succort of the current situation of T'!I 1, a
I8 olus the need to look at Nestinghouse and Combustion Engineerina 19 designs and to nake sure we understand the unicueness and the 20 sensitivity of B&N designs, leads ne to cuestion whether we 2I can accomolish these verv incortant things in a reasonable time trane.
U And given two feedwater transients a vear, even with changing the oilot onerated relie" valve, there s'd
's y-r R e X r o's l a.C 25 a vulnerability of the machine.
-e
!/
..._v
i i
54 I
l I[
I'm not certain whe ther shutting then down makes l
T.m 2'
it harder or easier to get the work done.
Probably harder.
But, there is a risk thcre.
COANISSIO'IER AHE'sP2IE : Nhv does it nake it harder?
I 3
p DR. 51ATTSON: Well, sir, I was thinking of the last d
1 i
6 time that five olants were shut down.
l 7
(Laughter.)
i H
i 8
Ne scent two weeks at the Congress of the United e'i States.
10 u.
- at the mancower situation FIR. DENTON: I'm looking q
11!l with less of a aersoective than Ed.
Ed has been trying to 12 handle all the oroblens for the cast three weeks.
13 h I cuess my own view would be that we could continue 1*'
tLshovel un nanoower anywhere in NRR to Se higher orioritv U'
tasks until we flativ. run out of CP and OL licensing reviews.
n 16; The staff savs we are almost there, and I cuess at that ooint there would be just activities that we would have IE to droo off tha t we normallv do that we wouldn' t be doinc, 1 ;~
connectea. with issuance o# turther licenses.
20 CO'i'1ISSIONER AHEE.RNE : That was also one of nv
^1 earlier questions.
There are other branches that night have 2,
suf ficient recent technical exoerience that you could oull in
~2 to nelc you out.
C...,- -,b,_,
, -,., D R a,. _n:
,.._.. s a t air., v soeciall ea area e-
.~.s i n
, nn.
w aex.es m 2
and I susoect You hava cot oracticallv all of the sta#f that 4 -
~P l/
i 55 I
l l
L i
\\<
mm 1;
can make any useful contribution in the next month alreadv g
i 2
gathered, would be mv guess, c
3 MR. DENTON:
I'd like to also hear from Denny Ross, I
a 4 Yi who has scent a lot of time orecaring his bulletins and looking.
H i
5J at resconses.
I 6'
MR. ROSS: Nhile Ed was talking,I was writing.
7l Really, we are on item 3 of what was a six-cart 1
89 oresentation to the Congress a couple of weeks aco. And on d
90 the way down here we thought mavbe a seventh one would have 10 i been -- had to do with the rathe r extreme sensitivity that
!i 11 j B&N olants have to certain transients in a relative sense.
12 '
So I wrote down a few cuestions and answered then 13 1 and got to a bottom line.
!o if 1.1 With resoect to transients cosing a carticular 15 j challenge, the first cuestion is: Does this challenge cone I
16 along often enough?
e 17 '
And the answer, at least av answer is-Y e.
Once t
I I
18 '
or twice a year, and that is often enouch.
19 The next cuestion is:
Do we really unders tand what 20 we need to know abc" the transient behavior of B&N clants?
21 And I think, no, not well enough. At least it is 2
clear to me that there are some things that no on that we don't analv e as well as we ought to. In carticular, the cropensity of any void formation followinc these transients
- . s ae x
..e m:.
- S that heat the system un and then do cressurire it.
il
.4 V
g i 1
l 56 I
I H
n I
l lo And then there's a situation that we have not I
a A
L analyzed--
"Ne," both the NRC and the regulated industrv i
2 has'not vekk enough.
t 4il The third coint has to do with the control system.
1 i
ti 5
Is the feedwater control svstem a good, adequate and comolete d
4 -
system?
l n
/j I think the answer is no. And we saw what B&N said i
8 from the vice cresident last week. Thev said thev have a very d,
9j definite interest in fixing un the control svstem, redesigning
'l 10 1 and improving it.
II So the control sys tem, I wouldn't sav, is to the 1- :
ootinum coint,and I think all sides acree to that.
!l-I3 The fourth has to do, is there reliable ecuionent I4
.in the clants to cone with feedwater transients?
15 My feeling is bv and large, 'res. Turbines,whather p1 they are electric motor or stean-driven, and ECCS oumos and
'7 their cower suoplies, I would sav they bv and large are M
reliable except that we noted on Oconee thev are oeculiarlv U
susuectible to single failures since there is onlv one of them.
20 And any single-failure costulant would wice that out.
z1 So, getting to the bottom line,'. hat can be done
~
22 in the immediate future?
Like this week, what are some
'2 remedial c.ea sures that one could look at?
And, I'm talking about short-tern in carnicular.
a x.....,
25 Aux feed. Nhat better could be done, escecially ~or l l L.
I 57 t l
i 1
I i
I
'ra I'
a single-failure prone olant?
2' What better could be done on ECCS?
In carticular, n- ;
high-pressure injection?
q 40 Mhat's the ef fect of cart cower?
"ould this make 1!
5 the olant look better?
1 6
Could the once-through steam generator inventorv l
7 lI be raised to make the olant less susceotible to these volatile I
1 8h transients?
9 q
Could we get a better analvsis of the things we a
Il 10 ]
haven't done a good enough analvsis on, and in a hurrv?
Is n
11 U there some quick redesign features we haven't thought about?
12 These items that I wrote down, 4 throuch r, are I 2' things that are croper exoloration matters with the utilities l ~' F that are running.
And mv recommendation would be to have 15 h such a meeting with these ceople this week before anv I
14 taken.
crecioltant action is
'7' That's where I come out.
15 CO'NISSIOER AEAE: But Dennv, vou know all the M
questions you are asking, which are very good questions,thev 20 could lead to another whole set of bulletins and that would 2I be the coint that I would be verv interested in hearina I&r's v
answer.
23 I get the sense we have a larne bodv of extremel" comoe ten t technical oeoole workinc *ith a vert-comclex g
s m w.s,:
2 phenomenon. Complex in the sense that 4, unanalyzed; and, 3,
0 i e i
58 i
e i
I mm a reconstruction from, at the moment cartiallv-available 2
- data, funneling in a constant series of advice and direction
~
to the ouerators.
r 41 And I am just wonderina, at what point do the a
operators become saturated?
3 4o 6
MR. ROSS:
I anderstand the cuestion.
I believe 4
a il 7 ji I&E was going to speak to that.
Yes, I&E was going to soeak ti t
i 8j to that.
I d
a 9 il It is a valid concern.
4h 10]
CHAIRMAN HENDRIE-I must sav, I think it would be o
l' II useful to talk to the operating organizations.
I think Crvstal 12 '
River is down for refueling now, I believe. And accarentiv l
I2,
the single unit out West could be taken out without affectino, L
l ~' '
seriousiv affecting the cower supolv situa tion.
e
'l M,
Here at the Oconee station there aro dree units 1
16 running hard keeping Ehat system and its neighbors jn good shape.
And I think there is going to be trouble down in that 2
part of the country without those units.
I There is also, it seems to me,91e fact that Oconee 12 the first olace that B&N machines were braucht in.
Oconee wat 2-1 was the first in ooeration.
It has alwavs struck me as a 22 fairly comoetent arcanization, and I think s e discussion 22 with them about such things that Dennv raised nicht be useful here.
~
am,r i,:
2 I cuess I would advise that course of action in l
lI s.
a -
I i,
59 I
i I'
mm creference to anything else this afternoon.
2 CO'D11SSIONER GILINS KY:
Let's see. What are vou proposing?
o 4 '
MR. DENTON:
I am crocosing to follow un along 1
..i 3
the lines that Denny suggested to me, then go into these things 1,
6 and see how reliable the aux feed systems are. And if I don't 7
like what we see, come back to the Commission at that time.
i' t
B COM31ISSIONER GILINSKY :
It seens'to me the question 9
,i you have to ask is, 'iven the uncertainties, are we erring on i
I0 the side of caution here' II d And I guess I'm not clear what the answer is.
I2 MR. GENTON:
I don't think the answer is clear, I2 I was going to nake the coint that I think vou have Id to take the facts where vou find them. And so far we have 15 issued bulletins which deal with the most -- at that time, at c
16 '
that date--oerceived largest contributors to risk.
And I think we have dealt with several of the U
largest con trib uto rs, and this one is one that we have not 1;
stocoed to deal with.
It is a verv imoortant one and will
- C have to be looked at.
21 I guess I've tried to give vou the rance o# values; 22 that it might turn out to be between 1 and 10, 100 or higher.
22 And I really iuse can't, todav, give vou the kind of U
data that we would all like to have, b ecause we haven ' *_ generated R e x,*ee:
IN it vet.
2
-l li s-
I l
l l
60 i
11 I
!i I
mm CO'NISSIONER GILINSKY :
Nhat does gettina tocether 2
with the coerators entail?
How soon could one do that?
It is being done,but with other 4 i olants also.
1 1
5 In other words, we need to look not onlv at B&N 6
olants, but look at the other clants also to see if their 7
situations are really this much different or not, with it lI 1
8L the kinds of nunbers we have been using -- our 1
91 sort of numbers, reoresentative numbers that necessarilv 10 don ' t apoly to any carticular olant.
II COMMISSIONER AHEAT1E: As you found out when vou l
32 1
wen t to pressurizer level readings that you came un with, i
13 some of the other clants also had it.
l ~'
MR. DENTON: That's right.
15 !
CO'NISSIONER GILINST(Y:
I wasn't clear, Joe, were I
16 you suagesting thd=the single unit ao down, the three Oconee plants not?
M CHAIR'1AN HENDRIE: No.
9 I was just noting that the recort on t'n cower 20 cupplv situation indicated the single unit out Nest would be 2I a good deal less missed than three on this end, carticularl" 22 since thev are concentrated in one svstem and since there are 22 other units down in that system and surrounding systems.
'IR. DENTON:
I think first imoressions are that 3
a r e.,
i:
25 Rancho Seco has a better auxiliarv feed s'zsten than Oconee.
,_I.
s - -
l b
l/
n.._-
61 l
I l
mm i
DR. MATTSON:
Ves.
2 COMMISSIONER AHEARNE : I had heard somewhere that 3
Oconee fidn't have recombiner.
l 4
Do you know if that is true or not?
5 MR. DENTON: No.
6 DR. MATTSON:
I think that may be the case.
i 7J COMMISSIONER AHEARNE: That they don't.
1, f
8.l DR. BiATTSON:
I believe that's the case, i
.I 9j MR. DENTON:
Ne haven't let I&E give their reoort 10 '
vet.
II "
MR. GOSSICK:
I was going t3suggest cerhaos vou I2 would like to listen to the I&E reoort.
I3 COMMISSIONER AEEARNE : If we can ask some more la questions later.
IS DR. MATTSON:
No recombiners.
16 COM'iISSIONER AHEARNE: No recombiners.
17 Harold, what would vour comment be on that?
13 Nould that concern vou?
I?
'IR. DENTON:
No, not -- I guess av decision criteria 22 at the moment is more reliability of conservative feed systers.
21 I guess recombiners are something we didn't recuire for some 22 time. They were desirable. Ne will want to rethink whether or not in the lonc term we want to do something different.
2 But I don't see -- I'm more concerned about the
,am.... im
~
25 short-tern transients to have a fairlv high orobability of
.i
~,
l/
62 I
4ld f!
1 i
~
l
I occurring.
tm 2
MR. DAVIS: Nhat we would like to do this afternoon 2
froc I&E is to give just a few ninutes on our imoressions of 4
what has transoired so far in reaction to"the bulletins.
i 5
Norm 'ioselev will talk for a few minutes on the 6
augmented inspection orogram, the training -- what we have 7l.
done -- the training discussions we've had with Licensee i
3, ooerators, and our insoectors' imoressions and reactions to d
1 9'
the operators' readiness to handle the transients described a
i 10 in the bulletin.
II MR. MOSE LEY :
In terns of Licensee coerator indoctri-12 ]
nation, soon after issuance of Bulletin 05A, we had the I3 insoectors for the B&W olants to hold discussion sessions with Id all licensed coerators.
I5l These began Acril 6th and were comoleted Acril 12.
I6 The discussion centered on dae chronolony of events at Three Mile Island and the actions requested by Bulletin 05A.
I3 Ne, I should tell you, plan to do something sinilar for the 053, which was issued on Saturdav.
20 As far as Licensee training -- that is traininc be 2I the Licensees for their own ocerators, Bulletin 05; recuired 22 Licensee to review ooeraf.ing crocedures and traininc related 22 to not relving on a sincle instrument and the overriding or I'
shutdown emergencv safecuard features.
ua....,,-
Site inspections that have been conducced since
~ ~
\\l s.
63 e
h
.r I
then are verifving that procedural changes have been made or mm 2
are being made and that the training ascects are being 2
conolied with.
4 B&W simulator has been crocramned to simulate the i
3 Three Mile Island event, and coerators from all clants are a
6 being cveled through this training session.
7h As f ar as othe:- PNRs are concerned,there was the l
i 8
joint NRR/I&E team, or actually six different teams were d
91 formed to visit all non-P&W PWRs to discuss the Three ' tile 1
1 10 i event and the soecific NRC requests that were made in II Bulletin 06A and 06B.
5 I2 These teans Pegan visitina the sites last Mednesdav 12 and presently we expect that all sit'es will have been covered 14 by this coming Nednesday.
15 As far as augmented inspections are concerned M'
for all ooerating R9R reactors, an?aunnented insnection
'7 frequency was enacted.
It's an equivalent c# a resident 12 inspector with some additional insoection recuirements.
I?
And these additional recuirements are that there will be some 22 inspection done on each shift.
21 There has been increased surveillance of II control room activities and the insoection crocedure also 22 includes specific inscection requirements related to review I'
of the bulletin resconse, revice7 of ooerator traininc, w..,, : i:
~,
25 independent inspection or verification of encineered safetv n
4#
- i. "/
64 i
+1 I
H i
,l i
b I
features, readiness and assessment of changed operatina mm proceaures.
2 Ne instituted a soecial daily inspection report i
I 4j for the B&W facility. That was done in early Acril.
1 5'
A summary of these reports has been compiled daily J
6:
with copies sucolied to the Commission.
7 F Our clan now is to change to a weeklv recorting Jl 8l system on these.
d
'i 9l To get to a question that has sort of come uo a coucle 1
10 '
of times, as far as ocerator orecaration, we have discussed II d just this morning with the neoole who are doing the site I2 inspections at B&N facilities and the real feeling by the I3 '
insoectors is that the coerators are orecared, based on those Il areas highlighted in Bulletins 05 and 05A, the ocerating clant 15 procedures for operating clants, the crocedures have all been 16 modified and the operators have all been trained.
7 I must sav that for myself, cersonally, there is IS a point when continued reliance on adding more and more things M
for the operator to do begin to be a concern.
Like, I don't 2C presentiv believe we are there, but I don't know how far awav 21 we are fron being there.
22 C05NISSIONER AHEARNE : Uhat kinds of crocedures 2'
are they now following in order to get a sense of how nuch Il water is in the core?
A., g r + + r e la; 4 r 2
Thev are not coing to follow the cressurizer level r
ll 2,
u
65 l
i l
I e
i i
I i' as a unique indication.
What kind of crocedures have thev mm 2-put in?
MR. MOSELEY: Mell, thev use cressurizer level 4
together with oressurizer cressure is the orincioal means of o
5 making this assessment.
i 6l COMMISSIONER AHEARNE:
One of the recommendations M
7Y the ACRS has made is that they have steam tables, for 1
4 8 ll example, readily svailable so they will always be able to l
'i i
9 trs
' pressure and sa turation temperature.
10,
- n. 'tOSELEY:
. understand some of the Licensees i.
11 C have, indeed, done this. He have not, in our bulletins, so far I2 -
made that a recommednation.
- l I3 i DR. MATTSON
- Excuse me a moment.
I think we had la it in one last weekend -- that went out Saturdav did have that 15 in.
16 '
MR. ROSS: we were going to soecifically cover 7
this in our response to the bulletins,'ehich is under develoo-18 ment.
But we have seen charts that amount to a saturation I
meter. Call it a softwire, if you would, because it is a chart 20 right now, but some of the utilities are develocing a hardwire 21 system where the comouter would interrogate different things 22 and ther orint cut messages about our accroach to saturation.
B&W is heloinc on this.
2 So, there is really two things coing on in that y.h R..w - I:
resoect.
,,, n e
ll u
.v
i 66 i
1 i
i m
I!
MR. DENTON:
I guess I should coint out, too, in i
i 2
natural circulation it is hard to distinguish anonc the PNRs.
2 In other words, if veu assume we have taken care of the
]
l relief valve ocening, then the main difference is your time 4
5 available to get aux feed cump water going.
t 6,
But if that fails in PWRs, then all PNRs resconses I
1 7e for natural circulation or ECCS ooeration would be about the l
- i. I l
8" same.
So it is not a unicue oroblem for B&W if you reach a d
^
h 9j stage of an accident.
1, 10 l COMMISSIONER AHEARNE: Do the ooerators train to go i
II O into natural circulation?
l I2 MR. MOSELEY:
I don't Know if that was a cart of 1,
13 a simulated training prior to this e'ent or not.
I iust la don't know the answer.
15 MR. C ASE :
The new bulletin would require develoo-I6 ment and training, develoonent of crocedures and training in that area.
I8 DR. MATTSON: Nell, they do a loss cf offsite cower I
test and startuo testing.
20 MP. MOSELEY: That's not necesarily traininc for 2I all ooerators.
22 MR. DENTON:
I think the cuestion we are all
'2 gracoling wath is that we have done a lot throuch these bulletins.
It certainli' has incroved the sitatuion.
3,-
Ray-e.-
- m; 2-The real question is how much further do we need I
/
l $
l-
l t
67 l
I!
I i
I 1 q' to go?
7m 2
As we find difficulties t'.iat we didn't insoect out there, should we have done something different a week or i
two ago?
4 4
5j I guess I was crepared to wait another day or two n
6 before making a judgment in that regard.
7 CO'?IISSIONER AHEARUE : Norm, vour sense from vour l
e inspectors at the moment is that the ooerating personnel haven't o
i 9 0 reached that point where they are overloaded with requirements?
10 j MR. MOSELEY: That's the belief.
IIbt COMMISSIONER AHEARNE :
Is this orior to or after 1
I20 receipt of Saturday's bulletin?
o I3 MR. MOSELEY: This is based on Bulletins 95 and 05A.
I I#
I haven't asked them soecifically, because all this stuff that 15 is required by 05B has not yet been out into effect, as you l'~
know, it is not required to be, vet.
U COMMISSIONER AHEARNE: Yes.
M Except that it is still iust a whole host of new items which I imagine thev will be exoected to at least have 20 read and attemoted to absorb some of the information out of.
2I MR. MOSELEY: There are two additional ocerator action 22 type requirements that are established. There are certain U
things that we now ask them to initiate a manual scram for that do not now have scrams.
+
u,e.,
2-Also, the things that we discussed in the bulletin
,, m
, _/
Ii 4.
l 68 i
i I
related to anticioatory action orior to going to the natural na circulation route.
2 2
These are new requirements on the ooerator.
d al COMMISSIONER AHEARNE : In addition,didn't vou, on i
.1 the weekend bulletin, modify some o# the instructions in 3
li a
6 the previoua bulletin, namelv to essentially give them
.i 7d additional cautionarv for the integrity of the pressure vessel'c I
i.
8i MR. MOSELEY: Thev were new in the thev had not been 9
specifically transmitted to B&N.
g 10l However, those same instructions have been given Il to all the PWR plants. And, in fact, the discussions have been 12 held with various of the B&W olants along the sane lines.
j 13 But it just wasn't documented crior to that.
la COMMISSIONER AHEARNE: Are vou still e.xtractina nore 15 information from the TMI in the sense that we are still in the 16 major cart of the learning what hacoened?
17 MR. MOSELEY: Ne are certainlystill trying to -- in 15 our investigative effort that is ongoing, we are trving to M
sharpen the sequence of events to make sure that we have 20 identified all those things that are of significance that need 21 to be looked at in more death as tine coes on.
22 COMMISSIONER AHEARNE :
I guess m.v question would be:
22 If you are still in the stage where vou think you are learninc a great deal more, chat would be another item to add in this g
a -c.....,,:
25 balance of, "7e still aren't reallv sure we know what haco,ene.
ii e
{ /
4.
69 l h,.
h H
l I'
mm and therefore not really sure we know what to do to orevent it.'
2 MR. MOSELEY: Me have not in this coint in time, and the investigation team which I contact frecuentiv, 4
identified anythina of major significance in their view, i
a 5
which is not already under consideration as cart of the lessons 9
6 learned that we are now working with.
7 hll COMMISSIONER AHEARNE: Are they in a situation where l
e o1 they have comolete access to the oeople and get all the ll 1
9h records they need?
u I
They are moving. Thev have no oroblems II in moving ahead.
1
,- i
j It is a painstaking task to talk to a lot of ceoole I2 and try to sift through mountains of records and sift out tha 14 salient facts.
15.i COMMISSIONER AHEARNE : But vou would let us know if a
I6 they were runninc into anv difficulties, because clearlv it is
'7 critical that they do get access to all those oeople.
M Ne are now -- what is it, Harold, you had mentioned 19 H
days when vou were there, so we nust now bein the three-week 20 stage.
2I MR. DENTON:
Three weeks clus.
22 MR. GOSSICK: Tuesday will be four weeks.
22 CO.MMISSIONER AMEARNE : And as I recall, one of the II coints Steve Hanauer nade in Browns Terry
- as that after s
w nex<e: n 25 about a r.onth, neooles' memories begin to fade, or it is
\\/
u.
70 l
l H
i I
mm difficult, carticularly if thev have been involved in many i
2 things, to really reconstruct.
So I would be verv concerned if all our investigative team is not able to get immediate access to the records and 5
to the pecole so they can reconstruct that infornation.
'I 6
Do they have enough suoport in Se way of stenograohers i
7j and secretarial helo to get the records tegether?
H 80, MR. MOSELEY: Nell, one never has as much as he l
9" would like, I guess.
But we are sensitive to those issues and i
10 ;
are being responsive 2 11 il COMMISSIONER AHEARNE: As I raise that ooint, there is another coint that Hanauer made; that we ought to be verv I3 careful that we don't receat the oroblem of not cetting as good' 1#
a record as we can while menories are fresh.
15
- MR. DAVIS: We are tuned to that, Commissioner.
14 However, I would like to remind vou that we did delav our investigation a few davs to make -- so that the neocle on M
site could be very sure of the safety of the clant.
Not to interfere with that.
'C COMMISSIONER AHEARNE:
Of course.
Of course.
21 But the visible cresence of the gentleman across 22 the table leads ne to believe that we have a higher degree of assurance of that at the moment, because I'm sure he 2#
wouldn't be here --
- m. x.....,., ;
'IR. DENTON: The investigation started before I left.
'2 a;
s a.
f i
I I
71 i
I i
mm I j' MR. LiVIS:
Ne have asked our investigators to make !
us -- to be very.a ' ;rt to any oroblens they mav have, to notifv' 2
2 us.
And I've heard nothing of problems.
o 4 :i COMMISSIONER AHEARNE:
I have.
i n
5a MR. DAVIS: You have?
d 6
COMMISSIONER AHEAR'E: Yes.
1 7
MR. DAVIS: Nell, maybe we should ask if you hear i
l 8
of some to let us know.
I l
5j COMMISSIONER AHEARNE:
I just have.
I have just il 10 '
mentioned a few that I have heard of.
II l MR. DAVIS: We will pursue them.
12 MR. DENTON:
Nell, I don't know what timeframe 12 you heard them, but I was an advocator in the first fiew davs la that we not kick off the investigation before we made 15 progress --
16 CO'NISSIONER AHEARNE: I understand that.
17 But I think now we are at the stage where we le ought to be concentrating, making sure we ge t --
19 MR. DENTO'i :
I think it was about a week or ten 20 days ago that the investigation began in earnest.
21
'iostiv there have been tencorarv little eroblems, but this is to be expected, and we worked then out. Ne have, so far, been able to work out all of these things that have occurred.
- .c a.. :c,.
m:
If COM"ISSIONER GILI'ISKY : Could I a r', Ed, are vour l $
N -
1 i
h 72 pl, mm 4
I l' concerns tied to this auxiliarv feedwater oroblem?
2 MR. CASE: That's certainly one of them.
2 COMMISSIONER GILINSKY: Or, are they more general?
4 MR. CASE:
They are quite soecific.
5 Finding about the designs, the deficiencies in a
6, individual plants of un to the aux feed system, doing in d
7.!
effect a de novo review of the feedwater controller, ad l
E' failure modes and effects analysis of that to see what effect 1
99 that may have on oilot operated relief valve ooening, because IO without those two systems being good,the action taken to prevent PORV may not succeed because vou do relv on those 11 I2 two combinations of thos e two things.
I C' COMMISSIONER GILINSKY :
So the conditicn von would l'
place on resuming ooeration, would be resolving uncertainties 15 in that area.
la MR. CASE:
And perhaus hardwiring of the anticinatorv scram, which is a month or so tyoe ooeration, I would quess.
IE DR. MATTSO' :
I think one other things we have 0
BSN doing is better than back-of-the-enveloce calculations of 20 the critical range of small break LOCAS, which is about what 21 we have at the moment.
22 COM'IISSIONER GILI'; SKY :
Is there any overall 22 staff recommendation here that involves several offices?
MR. GCSSICK: No, I think not.
- m. c..,..,., :
2 MR. DENTON: L'e have not atte-cred to e'ren cet a
-a -
\\l
73
\\
I l
ll i
I4 coordinated one, but rather tell vou where we are todav.
mm 2
And I guess most offices did carticicate in this morning's meeting that we had.
4 MR. GOSSICK: I am concerned about the point Ed 5
makes,the saturation of the staff down there. There are just a
6 so many things that have been 'arought out.
I don't know I
7 h *whether that would be made better or worse by the course that E
Ed suggeste.
9 But I guess one could ask, is it easier to go down 3l I0 and in a few days decide to come un, to be on the safe side, II or is it worse from the standooint of addressina the whole I 2,i host of matters to go the route 'Ir. Case has suggested.
13 COMMISSIONER GILINSKY: It sounds like Ed is talking Id about more than a few days.
15 :
MR. GOSSICK: That's another cuestion, I'm not o
16 sure how long that duration is we are talkinc about.
U COMMISSIONER AHEARNE: I would like to again reiterate U
the related concern that I have about the instrucrions we are I
sending out.
20 Frankly, when I saw that over the weekend we had y2' sent out all of the ACRS letters to us, which I know we found 22 some dif ficulty in intercreting what exactiv was there, when U
we relayed those out,that was where I becan to cet concerned that cerhaps we were overloading rhe system.
1 _
/
\\i
~v n ec e..1
\\:
COMMISSIONER GILINS E : That's whv we are nct
I 74 I
ii i
l
\\
n I
mm operating reactors.
2 COMMISSIONER AEEARNE: But, we are almost at that stage.'
t 4
COMMISSIONER GILINSKY : It's an incortant coint.
. :i.
CONNISSIONER AHEARNE : We are almost at the coint 3
4 i
o 6
where we are giving daily a new set of soecifics --
- l 3
i 7 ;l COMMISSIONER GILINSKY: That's right.
]
d a
i 80 Ne have got all this high-oriced talent. Ne are i
n i
i n
9 just trying to figure out the right instructions.
10 +
COMMISSIONER AHEARNE: Highly-comoetent talent.
II (Laughter.)
- i I2 CO'NISSIONER GILINSKY : And highlv concetent.
J I3,
And all of that needs to affect the awareness and i
la actions of operators at the other end.
15 '
COMMISSIONER AHEARNE: TUere do we stand on Davis-M Besse at the moment?
We were talking brieflv about other clants.
I3 MR. DENTON: Davis-Besse is shut down. Their i
tentative restart date is the 25th of this month.
The 20 understanding with them is that thev will have to rescond to 2I these bulletins.and we have a hold on resuminc coeration until 22 we have concleted lookinc at their resconse tothe other U
bulletins.
,, n
{/
t-CO'I.'1:S S IO:iER nns.s c -
nnd I wou'd
'i>a o reiterate 4*x**.**
l":
~..
a request that I made several times, that before you do, I
i 75 l
i i
'I hI!
would let them go back up, I would very much accreciate the mm 2
opoortunity to review with whomever it is accropriate, because 2
I think I am particularly concerned about that clant, since 4
I
~q it seems that as I go throuch some of the infornation that 5
seems to be available -- for examole a bulletin, Bio r m, that 6,
you sent out to all the other licensing boards, Davis-Besse
,1 7]
seems to have enough indications of some difficulties, that h
n 1
-1 e?a given all the sensitivity we talked about of the design and 1
a 9]
the stress, t he requirements of all the procedures being 10 aopropriate, I would oersonally have a lot of concern about 4
I I
+"l letting that plant go uo.
1 12 And if vou tell me on Thursdav it is scheduled to 12 I'm double concerned.
go uo, M
FLR. DENTON:
I'm implying no nersonal judgment 15 about the caper. It is just tha t ' s their forecast.
I don't 16 think we have~ completed our review of their bulletin.
COMMISSIONER AHEARNE: It stands, they don't go 12 up until vou tell them it's all right.
Is that correct?
If MR. DENTON:
Yes, that's correct.
20 MR. MOSELEY:
I should sav, I talked with Jim II Kepper todav and he talked with the President todav after I had talked with him, and the President reiterated that he is 2
not going to let the olant go up until he is satisfied.
I' And that Thursday is an earlies: cossible target wxv...
-n i/
date, not any intention that thev would startuo.
i 76
)
I, I
l' I
mm CO"MISSIONER AHEARNE :
I noticed -- I'm rereading o
with some interest, Norn, the bulletin that vou sent out en 2
the 28th.
i 4j Had you planned on sending out a following bulletin?
3 MR. MOSELEY:
Nhich one are vou sceaking of?
6 COMMISSIO?IER AHEAR:iE : This was the notification of I'
1 7
licensing boards.
8 MR. MOSELEY: I really haven't gotten back to that a
9; issue yet.
The boards have been notified.
i 10 COMMISSIONER AHEARNE:
Yes, I know.
g l l MV only concern was, that here the Davis-Besse a
- l 3*a insoector had raised a number of croblens with the B&N design, j,
12 or questioned whether thev were croblems.
l ~'
And then I guess I'm not sure, NRR, cerhaus their I5 conclusions were that there weren't really necessarilv I0 problems.
~
U And we had, apparently, reached that conclusion and
.,R the insoector had insisted that he still wanted them II notified. And I uess our procedure is that if the originator 2C wishes it to be done, we will go ahead and do it.
But our 21 response seens to ne, here are the problems and here is whv 22 they are not a oroblem.
23 MR.MOSELEY-Nell, I think the inscector, his 2'
recommendation cas that the board ce notified.
a.x....
r:
l 4 L. -s 25 The response that we sent back to hin was that
77 I
i i;
If mm we disagreed that the boards needed to be notified because 2
it didn' t meet the c;::iteria. That is, it was not a new issue, nor did it shed new liqPt on an old issue.
a t
We did not believe,nor did we state to then that 5
we didn't think that these were things that needed to be 1
A looked at.
a II
,j "OMMISSIONER AHEARNE:
No. My coint, Norm, was that p
i 8 !,!
the information we sent tothe board seems to inolv that these
'i 9
really -- this would not b2 a maior concern with resoect to I
t 10 '
these designs.
11 a 1
MR. MOSELEY: That was based on a review of one i
I2 of the events, that's correct.
13 o
COT 1ISSIONER AHEAR'IT': Right, i
l ~'
Av.d I was wondering, maybe, cerhaps vou mi:tht have se reconsidered that.
M
~
MR' MOSELEY:
I think'we ar re'considering all of these issues right now.
R CHAIR. Mali HENDRIE: Peter did vou have a comment?
I COMMISSIO:iER BRADFORD: No.
Not on this ooint.
20 Have you had any indications as to whether 2I Westinghouse or CE are having any difficultv dicesting the U
material we have sent out to them?
~2 MR. MOSELEY:
I've had no such feedback from the teams. But, of course I'm not through.
But, I've had no a
n, x..,.. m indication from either of the mlacas that they have been to,
., - n il L.
s i,
il 78 t
I so far.
mm 2
COMMISSIONER BR;DFORD: Nhat kind o# a reduction 2
has this had on I&E's normal coeration?
4 MR. MOSELEY:
You mean in terms of doing the nornal 3
things that we normally do to everybody else?
6 COMMISSIONER 3RADFORD:
Ed's problem with regard
/
to NRR, as apolied to I&E.
g J
i 2
MR. MOSELEY: Ne are more reactive. we are spenA.inq 9I more time -- let's sav, we are becoming more oriented towards IO a reactive mode of coeration.
II l The routine orogram is suffering. I can't cive vou 9M a number, you know, how many things we should be doing that t,
12 we haven't done.
But obviousiv, that's hacoened to us.
15 COMMISSIONER 3RADFORD: Nith this focus on BNRs --
15 COMMISSIONER AHEA dE :
B&N, PWRs.
l' COMMISSIONER BRADFORD:
Yes.
-- does that mean PWRs are getting less attention than they normallv would?
I6 MR. MOSELEY: Some anount of routine inscection.
I Those things that the regional oecole feel thev can reasonab1v 22 droo, have been droceed. But I don't believe that I would sav 2I that the 3NRs are cetting short shrift in the inscection.
Me 22 are trying to drop off tnings that are of lesser imoo,rtance, n.1 o
a t...
but '.his will catch uo with us if it eersists.
2' MR. DAVIS: It is basically an e::tensicn of interval
~.a
- w. r.....,--
~*E
,Detween inspection of items on the routine insoection orogr27.
4.
79 i
U i
Ii mm No will eventually catch un, or hone to.
COMMISSIONER BRADFORD:
Let's see. Nould vou talk a little more about what that means, John?
4 It would seen to me the insoectors are erobably
?
oretty busy going from plant to clant anyway.
Since it is 6;
not as though thev review everything at each clant each tine 7
they went there. It seems to me inevitablv that some of this 3
6 material just can't be caught up with.
n 9
MR. DAVIS:
Tht 's right.
It's not a matter of 10 cerhaps -- perhaps catchina up is an incorrect term.
II Ne, in our routine orogram insoect on defined 3,"
frequencies, certain elenents of the plant. Now, what our a
I 2' regions will do is adiust those frecuencies when thev are l ~'
called uoon to move heavily into a reactor program. So where 15 we, nerhaps looked at -- for examole looked at a ersten annuall",
16 we nay slide it to 18 months or something like that.
COMMISSIONER BRADFORD :
How many ceople are IE involved in these teams?
MR. MOSELEY:
Each team was comoosed of a licensing
'C examiner, a team leader from I&E and the crincipal insoector e'
insoector tron the site involvec,.
or res cent.
22 CON 1MISSIONER BRADrORD :
And these are coverina just 22 the B&N or thev are coverinc all PNRs?
t i I d A.
s
'IDSELEY :
All PWRs other than 9&"; Mestirghouse MR.
w, a.x,.,
r:
25 and CE, since we have alreadv covered the 3&h earlier 'eith
]
80 l
i p!
i 1
I these peoole who are doing the resident-like inscection mm coverage.
2
- COMMISSIONER BRADFORD:
So what oercentage, then, 4
of your normal inscection work course is tied uo in PNR 3
problems, soecial PWR oroblems at the moment?
6 MR. MOSELEY: You know we have a contingent of 7]
people at Three Mile Island also, d
,a c I I just don't have a number right handy as to how o
9 many people are, in effect, nulled out of what thev nornally i
I2 would be doing.
I think thad s the cuestion you asked?
II COMMISSIONER BRADFORD :
Yes, basically.
I I2,
MR. MOSELEY:
I can't give vou a soecific number 12 right now.
la COMMISSIO IER AHEAR'!E : Well, Joe, we have had sort 15 of a nixed set of recommendations.
He have got some actions, 16 it seems to me a range from going ahead and renuesting all of them be shut down or cart of them be shut down, to IS allowing an additional cvele nhrough the reactor coerators, M
or an additional cycle through the vendor or vendors.
20 Ne ought to end uo with some --
2I CHAIPf!A I HE:iDRIE : What I an going to recommend to 22 you is that we allow Harold to take another cuick look at 2
this matter,and probably meet either late tomorrow or I'
Wednesday, depending on how his schedule works out and come u. e,..,:
, _[,
'Jb I
If to some conclusion on the matter at that time.
l 81 h
9 I
Il COMMISSIONER AHEARNE: And in the interim, Harold mm 0
will be doing what?
CHAIR'1A'i HENDRIE : Such things as he's outlined, i
talk to peoole -- thev are, I gather, as was said at one I
point, the of fice is in mid-meeting. It broke to come here to 4
lav out -- the erincioal staff have come here -- assorted
't 7 iI views, personal views, office and otherwise.
i 24 COMMISSIO'iER AWEAR'TE : Nould that involve talking 9,
with the operators, the utilities?.
I think if we can work it in, we 11 certainly will.
I2 We have tended to focus in our looks to date, on i
r" auxiliarv feed systems at those four olants that are still la ooerating. We have not, to my knowledge, looked at all of I "c the olants such asDavis-Besse, which are down.
,d We can get back with B&N and the utilities --
CO.'1MISSIONER AHEAR'IE : The reason I ask, if we were 2
to reach, let's say, the conclusion vere to be to shut down
'I' some operating clants, we might at that time have wanted to 22 know what the operators at those plants had to sav. That 2I was the only reason for thm c.uestion.
At that same tima, would it be nossible, John or Norm, to have rechecked with vour inscectors to find cut their estimate of how much we have oiled now on the onerators?
u. c.,. w:
MR. MOSELEY-Yes, si-7
^,/
\\t s.
s a
82 I
I q
mm 1
MR. DENTON:
I guess I would like to make the 2
point, I have',been trying to seoarate in av nind the technical 3
decisions about the safety of the clants from the administrative 4
overload.
The administrative overload is certainlv here But 5'
I would like to think that that hasn't yet reached the coint 6
where it should dictate coliev decision.
70 COMMISSIONER AHEARNE:
But, I think, Harold as vou 1
L S!
yourself made in a coucle of coint.3, that operator orocedural 1
l 9l problems -- I want to secarate it in my mind fron so-called t
i 10 '
ooerator error, because at2 east I'm not vet at the coint where 11 I would conclude that the ooerators should have'known to do i
12 one thing and did something else.
1 13 The crocedural ditferences from what thev should la have done, the number of administrative overlavs that vou 15 place on those coerators on these sets of crocedures --
16 MR. DENTON:
That's right.
17 CO'1MISSIONER AHEARNE:
-- can exacerbate that 13 oroblem in dealing with a sensitive, technical desian.
So I?
that's why I'm stressing that si?.e of it.
20 And even if you were to conclude that you really 21 now have the technical system well in hand, vou mav conclude that the crocedural recuirements are such,that given the 22 current level of instrumentation, or the current number of I'
operators, the current traininc, that it can't be handled.
zm..~
25 MR. DENTON:
I think that's a cood wav to look at it.
I/
- t_J,
p i
l i
L 83 tl F,
Ih If, ultimately we conclude that the system is mm so sensitive the way it is designed that the burdens on the 2
coerators are so large that the existing framework for 4
training and licensing ocerators doesn't really suffice, then
- _ ', we should have done something different earlier in the svstem, 6
and it involves a littla bit of croiection as to where we i
7 are going to come out in that regard.
1 i
c' COMMISSIONER AHEAR'IE: Yes.
9 COMMISSIONER BRADFORD : By administrative overload, q
j e' Harold, had you meant on the coerators, or had vou meant II within NRR?
, jl MR. DENTON:
Nell, we talked about it mainly in 12 NRR, but it is certainlv true on the operators as well.
Id COMMISSIONER BRADFORD: Mhen vou said that I5 shouldn't dictate the outcone, had vou meant the situation in
'6 NRR, or had you meant John's point?
MR. DENTO'I :
I mean within NRR.
M I think that should be a secondarv factor.
Nhat it would mean in NRR, there would be a lot of things that we
- C would not be doing. And if that leads to a crisis state, I would say so.
22 I just didn't think we were quite there vet.
It 22 seemed to me in an agency of 2500 oecole,we ought to be able
^r to tan enough resources --
l/
,. l.'
,s we-25 COM':ISSIONER THEARNE : I would have thought so.
hl l
84 mm I
CHAIRMAN HENDRIE:
Okay.
2 COMMISSIONER GILINSKY: Thank vou.
2 CHAIRMAN HENDRIE: Thank vou very much.
4 COM:iISSIONER AHEARNE: Could I ask one other
'l 5
que7tions, since we have alreadv been told that we have 6
a difficulty and we may be constrained in ask.ing further a
7' questions.
- i
'l 81 At some stage would it be oossible to get someone u
s 9l who is familiar with he radiation monitoring that has been
-l 10 i going on?
I would like to -- and I would inagine at least; II some of my fellow Commissioners would similarly like to get I2 that information.
Certainl" I would.
I 12 MR. DAVIS:
Are you talking in-clant, or la environmental, both?
I5 COM'!ISSIONER AHEARNE : Both.
I6 For examole,I'm having difficultv -- I'm not a radiation exoert in any wav, so when I read the data that's coming out, it is hard for me to immediately get a victure,
'O I
where is it being monitored and what's the significance o^
that.
'l And I would just like to get a better, clearer e
22 cicture of that.
y, ot e.m. n v..
- 1.,,
4 w
- c...
.s.
s a.
CHAIRMAN HENDRIE: Okav. Than:-: 'ou verv
.'ch.
a x.,,
(Nhereucon, at 3:20 c.n.,
the hearing 'zas adiourned.)