ML20215A636
| ML20215A636 | |
| Person / Time | |
|---|---|
| Issue date: | 09/25/1986 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1551, NUDOCS 8610060163 | |
| Download: ML20215A636 (126) | |
Text
~!
03G NAL
~
O UNI 1ED STATES NUCLEAR REGULATORY COMMISSION IN THE MATTER OF:
DOCKET NO:
ADVISOR'l COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON WESTINGHOUSE REACTOR PLANTS
(
TPO4 DELEW B VaIITE
,rrUPS ORIGIN;L 'IO B.h1IITE,
- cas, 101 -H Thc.nks,
Barbara Jo Wnite, 4-1407
(
LOCATION:
WASHINGTON, D.
C.
PAGES:
1-105 DATE*
THURSDAY, SEPTEMBER 25, 1986 e. ;,
L..y f ^.s. 2 8!js,V-ipF
.a
- ' t.
u Jo Not Remove fmm ACRS OFice i
Y l
A E-FEDERAL REPORTERS, INC.
j O l
OfficialReporters 3
do 444 North CapitolStreet I
Washington, D.C. 20001 01 (202)347-3700 NADONWIDE COVERAGE t
PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSIONERS' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS THURSDAY, SEPTEMBER 25, 1986 The contents of this stenographic transcript of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards (ACRS), as reported herein, is an uncorrected record of the discussions recorded at the meeting held on the above date.
No member of the ACRS Staff and no participant at
)
this meeting accepts any responsibility for errors or inaccuracies of statement or data contained in this transcript.
l l
)
i
1 i l 1
UNITED STATES OF AMERICA O
2 NUCLEAR REGULATOR COMMISSION 3
4 5
6 Advisory Committee on Reactor Safeguards.
Subcommittee Meeting on Westinghouse Reartor Plants 7
8 i
9 1
10 11 September 25, 1986 i
Washington, D.C.
12 13 PRESENT:
Dr. David A.
Ward, Member.
Dr. Jesse C. Ebersole, Member.
14 Dr. C.
P.
Siess, Member.
Dr. Carlyle Michelson, Member.
15 DR. Charles J. hylie, Member.
4 16 l
17 l
l 18 i
19 20 l
21 22 i
23 l ()
24 l
25 ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
2 1
P R O C E E D I N G S:
2 (11:00 a.m.)
3 DR. WARD:
Gentlemen, our next topic is a 4
presentation by Dr. Speis and I guess others from NRR who 5
will discuss their analysis of the French PWR, specifically 6
the Paluel design.
I think it will be of considerable 7
interest to us.
8 I would like to recognize the presence of Dr.
9 Cheverdez, the attache f or nuclear energy at the French 10 Embassy and has an interest in this topic.
I would like to 4
11 invite him to comment during the presentation if he would 12 like to.
[s_-
13 Go ahead, please.
14 MR. SPEIS:
Thank you, Mr. Chairman.
I have i
15 with mc Mr. Rubin, project manager for this project and 16 also Bill and Ed from the staff who participated 17 extensively in this review.
18 As the chairman said, we will talk about our 19 review of the design of Paluel and more specifically, the 1
20 design diff erences between Paluel--
21 DR. MICHELSON:
I am not sure, if your speaker 22 is on, I have a speaker problem behind me that is really 23 buzzing in my ear.
()
24 Maybe we can turn it down.
25 (Off-the-record discussion.)
)
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
3 1
MR. SPEIS:
Can you hear me now better?
g-V 2
DR. MICHELSON:
Mike fine.
Much better.
3 MR. SPEIS:
We are here to discuss our effort 4
on Paluel, comparing their design to the U.S.
PWR design, 5
but more specifically comparing Paluel to a typical 6
possibly Westinghouse four loop design.
7 As you people may know, South Texas was the 8
reference plant for Paluel, and I understand from the 9
latest schedule that South Texas will go up to power 10 probably early
'87.
There are four Paluel reactors 11 operating also.
I 12 MR. HERNAN:
Ron Hernan.
NRR staff.
We also 13 have the South Texas Project manager here.
14 MR. SPEIS:
Yes.
I would I would like to 15 qualify the eff ort.
One 'of my objectives is to avoid some 16 of the detailed questions I heard earlier.
17 DR. WARD:
Good luck.
18 MR. SPEIS:
The effort was not really a 19 detailed design review.
I would like to make that clear 20 from the beginning.
21 The man-years expended on this effort was 22 about 1.5.
That includes the project manager, all the 23 other reviewers who participated.
That includes also my
()
24 involvement on the project.
25 If you recall we had done something similar ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
4 i
l 1
earlier back in 19883 involving the sizewell B design.
We 2
compared that to a SNUPPS plant.
The present effort is 3
really a continuation of that.
4 You people have expressed a keen interest in 5
our assessment, the objectives are to find out what they 6
are doing differently and more importantly, why they are 7
doing things diff erently than us.
8 How they are faulting experience, especially 9
f~ rom from their previous operating plants and how the 10 designs are evolving.
11 MR. EB ERSOLE :
May I ask a question?
12 MR. SPEIS:
Yes.
13 MR. EB ERSOLE :
You have also looked at the 14 British proposition, right?
i 15 MR. SPEIS:
Yes, Sizewell.
16 MR. EBERSOLE:
Well, let's go on.
Both of 17 these plants are, in the long run we are going to look at 18 them, too.
19 DR. SIESS:
How did you go about trying to 20 figure out why they did something?
21 MR. SPEIS:
Well, that is part of the l
22 discussion i
f 23 DR. SIESS:
Did you talk to people?
()
24 MR. SPEIS:
Yes.
25 DR. SIESS:
Okay, go ahead.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347--3700
5 1
MR. SPEIS:
I will describe it.
2 DR. WARD:
Go ahead.
l 3
MR. SPEIS:
I will answer that question 4
successfully, I hope.
5 I would like to say something in the 6
beginning.
Our effort on size well, even though it was 7
comparable, Sizewell is really a paper design.
8 Also, Sizewell had a PRA, and the sizewell 9
design was based as I said on SNPPS, Callaway, what is the 10 other one, Wolf Creek.
11 We had a kind of a more precise or coherent 12 evolution.
They started with SNPPS and then added or 13 subtracted exactly, you know.
14 Still, it was a paper design.
It hasn't been 15 built yet.
Paluel is a different case, four are running 16 now.
2 17 We were able to visit the plan, itself.
They 18 have procedures in place.
So our effort was more real than 19 theoretical, the thing that we did with Sizewell.
20 The scope of the review was, our major efforts i
21 is on design features, hardware.
22 Also, we did review very carefully the French 23 approach to safety.
Even though it is not listed here, an
()
24 integral part of our review was the emergency operating 25 procedures which are really closely coupled with design.
I I
__ ACE fEDERAI REPORTERS, INC.
Washington, D.C.
(202) 347-3700
6 1
will say something later on.
O 2
We talk about the French safety goals, when 3
they make an attempt to evaluate whether the design meets 4
or does not meet the saf ety goals, then they make decision 5
where to go beyond that point.
6 What they do is come up with procedures, very 7
precise procedures to take care of the issues or areas 8
mostly involving loss of redundant safety systems.
The 9
, procedures, again, are not safety procedures, but sometimes 10 they are both equipment changes and procedures, themselves.
11 But I will put that under procedures right 12 now.
13 The project was initiated in May of 1985.
The 14 way we started the effort, we got all the available 15 literature, and we came up with a draft report identifying 16 the differences, basically.
17 Again, the base was the SNPPS reactor to some 18 extent, even though we took into consideration some other 1
19 things.
20 We did five things.
We tried, we attempted to 21 understand the design as much as possible.
Then we 22 attempted to classify or delineate the differences between l
23 the French design and our design.
24 The next thing we did was to make an 25 assessment how important those design diff erences were.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
7 1
That assessment was kind of relative, among the changes, 2
themselves.
3 If, for example, we identified as you see 4
later on, 25 or so differences, then we make an attempt to 5
see how important those differences were among themselves.
6 The approach we used to arrive at that 7
evaluation was kind of a combination based on our 8
experience with PRA's, with resolving generic issues or 9
resolved safety issues, LER experience, operateing 10 experience.
It was kind of a synthetic PRA engineering 11 type of an approach.
12 So because you know, the Paluel doesn't have a 13 complete or detailed PRA, I don't even think they have a 14 PRA yet.
15 MR. EB ERSOL E:
Tom, I am reminded of research 16 projects that are stopped only because the money runs out.
17 MR. SPEIS:
Yes.
18 MR. EB ERSOLE:
I expect your job stopped 19 because the time ran out.
20 In approaching this job the general advice 4
l 21 from everybody is to take the cosmic view and get the big 22 picture and try to desues what there is here.
23 In view of the fact that these machines are i
] ()
24 built so complicated with so many elements of the safety 25 system in dark corners, how do you establish a point of cut ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
b 8
4 1
4 I
L 1
off at what you are going to look?
2 I might find a cotter key someplace and the i
3 whole thing will come across.
[
4 MR. SPEIS:
That might be clear when Al t
j 5
discusses six or seven of the key differences and puts l
6 pictures down.
He will tell you how far down we went.
4 j
7 MR. EB ERSOLE:
Okay.
I 8
MR. SPEIS:
How far down the support systems.
I j
9 MR. EB ERSOLE:
I am trying to find out what is I
10 your rationale about how far down you are going to go.
How 11 do you make that decision?
a
{
12 MR. SPEIS:
Resources were a fundamental 13 constraint to the process.
I 14 MR. EBERSOLE:
You run out of manpower.
I 15 MR. SPEIS:
We will.try to explain to you, for i
16 example, when we say we think the French have taken care of i
17 one of the important issues relating to the loss of l
18 off-site power or blackout issue, we will tell you, you i
19 know, how did we arrive at that conclusion, basically.
l I
j 20 MR. EBERSOLE:
All right.
l 21 MR. SPEIS:
I was discussing our process.
I j
22 Based on the report and documents, we put this draft report.
23 together.
()
24 The next thing we did, we send it to the j
25 French and a few months later we went to France ourselves, I
4 1
1 ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
9 1
about five or six of us.
The names of the people are fS V
2 listed in this report here.
3 We had detailed discussions on every issue.
4 You know, our obj ective was to make sure that we indeed 3
5 understood the design and we indeed understood the 6
differences and, more important, the rationale for doing 7
what they did, because I should have said earlier, the 8
French have three, maybe it is on the next vugraph, they 9
have three standard designs so far, you know, the 900 10 megawatts.
11 Now we are talking about the next generation, 12 which are the P 4 design.
13 So we went to France.
We met for a week or so 14 from people with CEA, PEF, their consultants and 15 contractors and took every issue, every difference in great 16 detail.
17 Also we spent a day at the plant, itself.
i 18 Some of the things we thought very important we went and 19 saw them ourselves, we touched them, we saw the drawings.
20 They were explained to us by the people running the plant.
21 So we got a very good feeling.
a 22 Still, it was a week's effort, so we weren't 23 able to go through all the wires and all the valves and I ()
24 pipes.
25 MR. EBERSOLE:
I don't know tde differences in i
1 ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
10 1
these plants, but let me presume there are some differences 2
in them.
That one plant can make itself safer at much less 3
cost than another, but the guidelines is drawn on a flat 4
basis and the inevitable argument comes up, why do I have 5
to, even if it's easy?
6 How do they handle that problem?
How do I 7
make it better when I don't have to?
Why should I?
8 Why should I make it better if I am already at 9
my minimum need?
10 MR. SPEIS:
I guess each country has a 11 different philosophy?
12 MR. EBERSOLE:
What do they do, make it better 13 and, ther ef ore, --
14 DR. WARD:
Why don't we hear what is going to 15 be different?
16 MR. EBERSOLE:
Okay.
17 MR. SPEIS:
I think the French, the 4
18 evolutionary designs, the three standard designs, each is 19 better than the other f rom a safety standpoint.
I think I
20 from an operational standpoint.
They have followed their 21 experience, not only their own experience but they have 22 their eyes and cars open to the world and read what happens 1
23 everywhere.
()
24 MR. EBERSOLE:
Was that done under regulatory 25 pressure?
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
11 1
MR. SPEIS:
The French have I think a very --
2 they have a system that is not as conf rontational as ours, 3
I think they work closely with the regulatory authorities.
4 MR. EBERSOLE:
Do they work concurrently as it 5
evolves in design?
6 MR. SPEIS:
I think so.
Maybe Mr. Cheverdez 7
can amplify but we asked that question.
They do work 8
concurrently.
9 I don't think the regulatory authorities can 10 tell the EDF, do something regardless of this or that.
11 MR. EBERSOLE:
Yes.
12 MR. SPEIS:
I think both sides use rational 13 approach to whether things should be done or not done.
And 14 they arrive at some mutually satisfactory conclusion.
15 MR. EBERSOLE:
Yes.
16 MR. SPEIS:
In the last analysis the 17 regulatory authorities have the final say whether a plant 18 should have this or that because of a number of 19 considerations.
20 MR. EBERSOLE:
Sure but --
21 MR. SPEIS:
You will see later the filter 22 vented containment approach they are using, even though it 23 was mutually arrived, you will see some of the
()
24 considerations that some of them are political or the 25 boundaries of those countries in Europe are such, the land ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
=-
12 4
l i
4 1
is not as plentiful as it is in the United States.
O 2
So there are many considerations that I am 3
sure are taken into account.
4 MR. EBERSOLE:
So they march along in 5
parallel.
6 MR. SPEIS:
Yes.
)
7 MR. EB ERSOLE :
As the design evolves.
There j
8 is no disturbance about the f act that they are a party to i
9 the evolution.
10 MR. SPEIS:
No.
1 11 DR. SIESS:
It's by it's been my experience i
12 that the French always use a rational approach.
Whether 13 you agree with them or not, it's a rational approach.
14 (Laughter.)
l 15 MR. SPEIS:
Yes.
l 16 (Slide.)
i i
17 MR. SPEIS:
I think I have mentioned this i
1 18 somehow.
The standard designs, they have 30 units in 19 operation of the 400 megawatt electrical type.
There are 20 some minor differences between those.
I don't think they 21 are all precisely the same.
l l
22 The effort that we focused on was the next i
23 generation, the 1300 megawatt electric.
You have four
[
i 24 units at Paluel and I think one at Plamanv111e and another l
25 somewhere else.
There are six units in operation, 14 under 1
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
13 a
i 1
construction.
2 They are also working on the third generation, 3
the 1400 megawatt electric.
I think two units are under 1
4 construction.
5 They have received, well, they have received 6
the construction permit, Schuise 1 and 2, I am not saying 4
7 that word wrong I hope.
8 As you also know about two-thirds of the 9
electricity in France comes f rom the use of nuclear power.
10 The first unit, first Paluel unit on line was 11 sometime in 1984, and the last one, was it this year?
It 12 was this year.
l 13 Let's go to the next one.
t i
14 (Slide.)
15 MR. SPEIS:
Based on the work we did we I
16 identified approximately 25 design differences.
As I said 17 we went about assessing the relative incremental saf e 18 significance.
19 We put two in the high category, two in the l
20 moderate to high category, seven in the moderate category, 21 low to moderate, threet low, seven and no difference, four.
I 22 Again I vant to caution on the use of the term 23 high, moderate, low.
They are not really absolute values.
l
()
24 They are relative among themselves.
i 25 As I said the evaluations were based on j
l ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
14 1
limited and generic PRA, based on operating experience, 2
familiarity with technical issues and our attempts to come 3
to grips with some of them f or the last year's or so.
4 DR. MICHELSON:
I thought I heard you say 5
there was no PRA?
6 MR. SPEIS:
Yes.
7 DR. MICHELSON:
Did I misunderstand when you 8
said you used some PRA in this other?
9 MR. SPEIS:
Our assessment was based on 10 limited generic PRA's, our PRA's.
11 DR. MICHELSON:
Oh, our PRA's.
12 MR. SPEIS:
Or whatever PRA's we are familiar O
13 with.
14 DR. MICHELSON:
It was only subjective.
15 MR. SPEIS:
Unless they are doing one now.
My 16 understanding in that there is not a PRA on Paluel.
17 DR. MICHELSON:
Just U.S.
PRA's.
18 MR. SPEIS:
Yes, PRA on Sizewell and some 19 others we are familiar with.
20 DR. MICHELSON:
Okay.
4 I
i 21 MR. SPEIS:
This aren is very key, areas of i
22 improvement.
Your question, Mr. Ebersole, on what basis i
l 23 they went and made the changes to this second, to the P 4
()
24 design, based on their operational experience with the the i
25 900 megawatts and world-wide experience they came to the AgEfyD3RALfpPORTpRS, INC.
Washington, D.C.
(202) 347-3700
- - - - - - =
15 1
conclusion that these were key areas that affect safety, 2
system dependencies, the reliability of electrical systems i
3 and, of course, human errors.
4 MR. EBERSOLE:
Did they work on 5
simplification?
Simplification of the fundamental 6
processes of shutdown and heat removal.
Simplification to 7
reduce the numbers of involved elements, the reliability 8
levels under hostile conditions, et cetera.
9 They work on that so a' person who'needs to 10 appreciate it, the public at large, can understand how this 11 trips out and how everybody goes home.
I keep seeking for simplification.
12 13 MR. S PEIS :
Our feeling was that that was an 14 important part of the evolution.
It wasn't strictly just j
15 add more systems f or the sake of adding systems.
16 There was an integral evaluation and in areas
{
17 where they thought more trains, either diverse or 18 redundant, will accomplish the goals, fine.
But simply it 19 was a part of their goal.
20 MR. EBERSOLE:
My impression has been that 21 some of the foreign reactors in order to get additional 22 shutdown they create another arm just like the one they 23 have got.
24 MR. SPEIS:
We don't have that feeling from 25 our review.
Maybe we can discuss --
i ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
16 1
MR. EB ERSOLE :
It's a little bit like, I am O
2 interested in if somebody tampers with my simple life boat, 3
because I think I understand that.
Maybe it will work even 4
in rough water.
5 MR. SPEIS:
Al Rubin will discuss what 6
changes, the changes that they have made that satisfy, 7
continue improving this area.
8 MR. EBERSOLE:
As a case in point I have a 9
standing difference with Westinghouse, even, you know, the 10 ACRS, I guess, about how you kill the power to make the 11 rods drop, invoking my own notion and that of others that 12 you should go to a pallet function to backup the straight 13 power function.
Kill the X station voltage.
14 DR. WARD:
Apparently, we are going to hear --
15 MR. EBERSOLE:
Will we get into that later?
16 MR. SPEIS:
Yes.
17 DR. WARD:
I don't know.
W'e are going to get 18 into 25 things that they identify as different.
I guess I 19 would like to hear those if we could get to them.
20 MR. SPEIS:
There are some people here who 21 participated in the review.
Some of them are not here.
22 Some detailed electrical types.
23 DR. WARD:
Let's hear about these.
24 MR. EBERSOLE:
Let's go ahead.
We will do it l
25 later.
i
(
ACE FEDERAL REPORTERS, INC.
Wauhington, D.C.
(202) 347-3700
=
_ = _
i 17 4
1 DR. WARD:
If there are other questions 2
later --
3 MR. SPEIS:
Again, we have concluded based on
- 5 4
this limited effort at Paluel, that Paluel offers 5
substantial improvements in safety compared to SNPPS for a 6
number of potentially dominant accident. sequences; station 7
blackout, failure to switch over following LOCA and 8
long-term cooling.
We will say more about those.
L l
l 9
Standardization, the French, it's part of 10 their way, part of their -- standardization is an important i
j 11 part of the process, itself.
1 i
12 Advantages in regulatory simplification, more l
13 efficient licensing, cost reduction, engineering design 14 equipment, shorter construction schedule, training of*
15 operators.
i 16 I also think their focus on safety is coupled l
17 also with standardization.
They want to make sure they l
18 eliminate as much as possible common mode failures and i
19 things of that sort.
I l
20 So since they have plants that are the same i
i 21 they don't want all of them to be shut down because a 22 common mode failure takes place, j
i 23 That is why they put forth such effort in j
()
24 eliminating as much as possible common mode failures to the 25 extent that experience helps, because common mode failures i
I s
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
18 1
by their definition, sometimes they appear out of nowhere.
2 Again, you do the best you can, a priori, I
l 3
either by design or procedures or diversity or whatever to 4
eliminate as many as possible.
i 5
I think it is important to say something about 6
their safety goal.
Just for completeness I will put down, 7
put up our safety goals.
8 You are familiar.
I don't want to go through 9
them in detail.
10 DR. WARD:
We can probably stipulate to that 11 chart, unless there is a particular point that you want to 12 make.
13 MR. SPEIS:
Maybe down here.
14 DR. WARD:
All right.
15 MR. SPEIS:
Regulatory obj ective.
Reasonable 16 assurance that the severe core damage accident will not 17 occur at a U.S. nuclear power plant.
18 The last one added recently, mean frequency of 19 large release of radioactivity should be less than so to i
20 the minus 6 per reactor year.
21 DR. WARD:
Okay, we have got it.
22 MR. SPEIS:
Let me say something about.
23 (Slide.)
24 MR. SPEIS:
I think their approach is more 25 practical than ours.
I don' t think they have spent
__ ACE PEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
19 1
hundreds of manhours or tens of man-years arguing and 2
writing reports and holding seminars and can I bitsing with 3
the ACRS.
4 DR. SEISS:
There is an even bigger 5
difference.
They haven't done a PRA.
Stating 6
probabilistic safety goals without doing a PRA is quite 7
different than what we are doing.
8 We are spending all our time arguing about the 9
PRA, not about the safety goals.
10 MR. SPEIS:
They are doing PRA's now.
- Also, 11 as part of their engineering analysis they also do PRA's, I 12 shouldn't use that word, some safety analysis, numerical, 13 fault tree events of important systems.
14 MR. RUBIN:
Al Rubin.
15 The French have done a PRA for the 900 16 electric plant so some of their results of that analysis 17 was done and used in their redesign, design, some of the i
I 18 differences --
19 DR. SEISS:
How was the PRA related to their l
20 safety goals?
21 MR. SPEIS:
Let me describe their saf ety goal.
22 First there is no document published with a NUREG number or l
23 some French equivalent number and say this is our safety 24 goal.
l 25 There are some documents, CEA and EDP I think f
i I
ACE PEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
l 20 t
I would have provided some summary reports to you in the 2
past.
i l
3 Plant perf ormance obj ectives are the 4
following.
Their goals are probability of acceptable 5
consequences should be less than so to the mineness 6 per i
6 reactor year.
7 I want to define what unacceptable consequence j
8 means.
That is a key part to their approach.
i 9
Also probability of unacceptable consequences li 10 due to f amily of events should be should *be less than 10209
{
11 mine must 7 per reactor year.
i 12 DR. SEISS:
How do they define a f amily of i
13 events?
l 14 MR. SPEIS:
For example, loss of heat sink is t
l 15 a f amily of events.
i 16 DR. SEISS:
If I make the family small enough 17 I can get that 10 to the minus 7 or 10 to the minus 8.
If i
18 I hit 10 to the minus 7 I can't have more than ten l
19 families.
So how you define the f amily is sort of i
j 20 important.
f 21 MR. SPEISt They are not very arbitrary.
22 Sometimes you have four or five or three or ten.
For i
f 23 example 787 24 A.
Loss of heat sink, and Al will tell you what l
l 25 is the f amily that constitutes the -- what are members of l
(
l ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
21 1
the f amily that constitute the f amily of heat sink events.
2 Station blackout, for example.
3 DR. SIESS:
Station blackout, is that one 4
family?
5 MR. SPEIS:
Yes.
6 MR. EBERSOLE:
One member of the family.
7 MR. SPEIS:
Containment perf ormance, 8
containment provides ultimate defense to reduce radioactive 9
releases to a level compatible with feasible off-site 10 emergency plans.
I will say more about that.
11 Let's go into the definition of unacceptable 12 consequences.
13 Now, for internal events this so 10 minus 6 14 per reactor year, they don't want to get into -- it means 15 core melt, not containment f ailure.
So this is really 16 their core melt goal.
17 MR. EBERSOLE:
10 to the minus 6.
18 MR. SPEIS:
Yes, per reactor year.
- Again, 19 it's a goal.
It will tell you very fast, when they attempt 20 to see whether the systems that they'have in place meets 21 that goal, they came to the conclusion very fast that it 22 couldn't meet that goal.
23 That is when they go to the procedures.
And
()
24 augment the design, enhance.
Either add some additional 25 equipment.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
22 1
But the procedure is an integral part of the O
2 process where in trying to satisfy themselves that they are 3
getting close to this goal.
4 But they know they are still, you know, it's a 5
goal.
They don't have to prove to anyone that they are at 6
10 to the minus 6 or a little below or a little above.
7 MR. EBERSOLE:
Is that a product of what I 8
would call numeralized competent peer group judgment 9
without all the paraphernalia of the little parts that went 10 into it?
11 MR. SPEIS:
It's the E P P, it's the C R A, 12 they have an ACRS there.
O 13 MR. EBERSOLE:
That number doesn't come out of 14 PRA's, does it?
15 DR. SIESS:
Not so far.
16 MR. SPEIS:
They have done one for the 9 900 17 megawatt.
18 MR. EBERSOLE:
You are talking in general.
19 MR. SPEIS:
Yes, in general.
20 DR. WARD:
Are you telling us about these 21 safety objectives because you perceive that they use these 22 objectives in making the decisions?
23 MR. SPEIS:
Yes.
()
24 DR. WARD:
About the design they went to?
25 MR. SPEIS:
Yes.
ACE PEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
23 1
DR. WARD:
Okay.
O 2
MR. SPEIS:
For example when they went and 3
evaluated whether the f amily of events constituted the 4
station blackout issue --
5 DR. WARD:
I mean, this is how they decided to 6
concentrate on the three issues?
7 MR. SPEIS:
That is part of it, yes.
Again, I 8
think the procedures are very crucial.
Al will describe 9
them in detail.
10 When I say procedures; they are not procedures 11 per se but are coupled with important additions or 12 subtractions of equipment, themselves.
Hardware.
13 DR. WARD:
Yes.
Go ahead.
14 MR. SPEIS:
For external events associated 15 with human activity, for example, airplane crashes, it 16 means that 10 to the minus 6, it means the radioactive 17 release at site boundary shouldn't exceed release from a B i
18 DBA.
19 It doesn't mean core melt.
It's somehow 20 possibly equivalent to our 10 to the minus 6 the Commission 21 wants us to evaluate.
22 MR. EBERSOLE:
The example you used bothers 23 me.
Iluman activity occurs inside the plant, too.
For 24 instance, cheer noble.
25 Did you make airplane crashes to characterize ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
24 1-human events as they think of it?
O 2
MR. SPEIS:
Maybe it's a poor choice of words.
3 Al?
4 MR. RUBIN:
External events.
5 MR. EB ERSOLE :, of course.
That was a 6
qualifier.
7 DR. MICHELSON:
Are they including in-plant 8
flooding as an external event?
l 9
MR. SPEIS:
That is internal.
i 10 DR. MICHELSON:
We define that as external.
I l
11 DR. SIESS:
We are not rational, though.
i 12 DR. MICHELSON:
Yes.
I just want i
13 clarification.
Is in-plant flooding an internal event for i
i 14 them?
15 MR. RUBIN:
We didn't go through each event j
16 specifically and how it was characterized.
i 17 DR. MICHELSON:
As a class is it inexternal or 18 an external event?
l 19 MR. SPEIS:
My understanding, A1, it's l
l 20 internal but we want to clarify that.
21 DR. MICHELSON:
This this country it
{
22 classified as external even if it is internal to the plant.
23 Cease ease you characterize that last one as parallel or 24 comparable to the general performance criteria.
25 It would help me, what is there permissible l
(
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
25 3
7-1 dose for design basis accident, is it the same as ours, 25 L/
]
2 rem whole body, or is it like the Germans?
3 MR. SPEIS:
I knew those numbers.
68 I don't 7
4 have them at my fingers right now.
We have them.
A1, I i
5 know, we have them back at Bethesda.
6 I don' t think they are -- Jack, do you recall 7
what the numbers are for design basis events.
i 8
MR. CHEVERDEZ:
I see it's the same.
We can 9
have it checked out.
10 DR. SIESS:
I know the German isn' t.
They use.
11 about a tenth of a source term and tenth of those.
l 12 Mr. McKue:
I am about the 9 pets percent 13 sure it the same.
14 DR. MICHELSON:
Another clarification.
Are 15 fires considered external?
16 MR. RUBIN:
The purpose of looking at the 17 safety objectives f or the French were to determine why they 18 were making certain design changes.
We didn't go through 19 event by event.
i 20 DR. MICHELSON:
Gee, though, this is a whole 21 class of events of --
22 DR. WARD:
Gentlemen, let's one speaker at a 23 time, please.
()
24 DR. MICHELSON:
I ask the question, is fire 25 internal?
ACE PEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
26 1
MR. RUBIN :
I think it would be considered an O
2 internal event.
3 MR. EBERSOLE:
You are not going to leave that 4-slide quite yet, are you?
5 DR. WARD:
Oh, I hope so.
Let's get going.
6 MR. EB ERSOLE:
Containment, ultimate release 7
but I don' t know what the state of the core is for that 8
MR. SPEIS:
Containment provides ultimate 9
defense to reduce radioactive releases to a level 10 coropatible with f easible off-site emergency plans.
1 i
11 (Slide.)
i j
12 MR. SPEIS:
You may have seen this before.
I 13 think I can explain it.
If not I have lots of help back 14 there.
15 I think they are being very practical about 16 this.
They sit down and say, my God, what can challenge 17 containment, steam explosions, hydrogen, core melt through, 18 slow over pressurization, you know, all the containment 19 failure modes that we have been talking about and 20 evaluating in the United States.
21 MR. EBERSOLE:
So they acknowledge.
22 MR. SPEIS:
There they kind of eliminate some 23 of them because of probability arguments.
They say, my
()
24 God, you know, based on our understanding, which is shared 25 to a large extent by others we think steam explosions are 1
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
27 1
of lower probability than some other f ailure modes that 2
would challenge containment.
Therefore there is not much I 3
can do about steam explosions based on the probability of 4
containment failure by steam explosion versus something 5
else.
6 They came to the conclusion that the most J
7 probable mode of failure is slow over pressurization and 8
included in that as a subclass, penetration of the 9
basement.
10 They say, okay, these are the doses, this is 11 the release fractions that get out if you have an early 12 failure, somewhere between 1 and 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />.
13 You see 80 percent noble gasses, iodines, 14 nonorganic, sixty percent, organic,
.6.
Cesium 40 percent, 15 strontium, 5 percent.
16 Doing their best analysis, deterministic 1
17 analysis assuming that, trying to understand where the i
18 containment will fail via this slow overpressurization l
19 process, they come to the conclusion that it will probably l
20 fail in one day and the activity or release fraction will 21 be 70 percent, noble gasses, without any filtering or i
22 anything.
23 Nonorganic is down to 2.7 percent f rom of 0.
()
24 This is the process of fallout and all the other things 25 that you guys know about it.
Cesium down to 5 percent, ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
28 1
strontium down to.6 percent.
73 V
2 They say, my God, I don't think they can live 3
with such numbers in making emergency planning meaningf ul.
4 All the other considerations are going on, about the land, 5
getting crap all over their neighbors and things of that 6
sort.
7 Then they say I want to do something better.
8 Theref ore I want to add a filter to my containment.
9 With filter, the U 5 procedure which Al will 10 describe in some detail, you see what they get.
They get 11 noble gasses, is basically still the same.
12 Nonorganic goes down by almost a f actor of 10, 13 cesium goes down more than a factor of 10 and likewise for 14 strontium.
15 So that is the rational that they have used 16 incoming up with simple filter vented containment system.
17 And then they think they can live with these numbers, they 18 can make the emergency evacuation and all the other things 19 more meaningful.
20 MR. EBERSOLE:
Let me ask you this.
I want to 21 go to the primary motivational f orces that make this 8
22 possible and impediments to them if any in that country as 23 contrasted 20 ours.
()
24 Motivational forces in this country says if 25 you do these things they must be put in the rate base.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
29 1
That is in a competitive context.
Everybody is looking for 2
the cheapest power.
3 The PUC's get involved but worse than that, 4
every_ Tom, Dick and Harry in the country gets involved.
5 Those purists who believe that you can't do 6
anything at all in the context of release or whatever, they 7
become a very high political, I don' t know, institutional 8
impediment to doing this kind of thing, and they are aided 9
and abetted by lawyers who make an awful lot of money out 10 of this process.
11 Are they plagued by that bunch of locusts?
12 MR. SPEIS:
I don't think so.
It's a 13 nationalized utility.
I don' t think they have that 14 problem.
15 MR. EBERSOLE:
Do the French public trust what 16 they do or does the media sell trash like it does here?
17 MR. SPEIS:
I think there is consensus from 18 all political parties that nuclear power is an integral 19 part of the energy.
20 MR. EBERSOL E:
So you don' t see the 21 spectacular articles in the TV and paper?
22 MR. SPEIS:
I think all parties f rom the lef t 23 to right are in probably some good agreement that nuclear
()
24 energy is an important source in meeting their energy 25 needs.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
30 1
MR. EB ERSOLE:
It's in agreement.
O-s 2
MR. SPEIS:
I think there is general 3
consensus, yes.
~
4 MR. EBERSOLE:
Ours is conflict.
5 MR. CHEVERDEZ:
Perhaps if you want I can make 6
general comment at this point on this rational approach of 7
the French or conflict aspects.
8 If you give me perhaps five minutes.
9 DR. WARD:
Why don't you go ahead.
10 MR. CHEVERDEZ :
The rational approach of the 11 French.
Yes, we are rational since Chernobyl.
We learned 12 that in school and so on.
13 But also the reason is ~because we have limited 14 family, you see, this is the reason we have decided to go 15 to one model of pressurized reactor.
We cannot af ford all 16 the options.
So we are really obliged to be rational.
17 About the safety organizations, let me say EDF 18 as parent of the plan, with the vendor, CEA as an 19 organization, R & D and improvement and safety authorities.
20 All are, their role, responsibility is very well-defined.
21 Let me say that the saf ety authorities are very, well, 22 competent, independent and very severe.
23 I am sure if there is some things they don't
()
24 want to accept, they can impose their will on EDF.
It is a 25 nationalized utility.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
31 1
But generally when they make a design and 7-)
LJ 2
discuss preliminary design, things like that, they try to 3
avoid to solve some conflicting problems, I would say of 4
minor interest, before they become big problems and then 5
result in losses of money or losses of time.
So generally, 6
we try to work in concert, yes.
7 But for some people, the last word is to the 8
saf ety authorities.
9 one thing very important about 10 standardization.
You say that we make each stand out --
11 Probably you are doing the same in this 12 country.
I am sure the plants built by the vendors, each O
k/
13 one is better than the other.
14 But in France it is probably more rational 15 because of the standardization and because we have one 16 utility with two divisions.
17 The design division and parent division are I 18 should say very well integrated.
They are working with 19 teams with periodical meetings so there is a system of 20 feedback.
21 If there is some incident, if there is any 22 problem, it is immediately reported to the design division.
23 I remember one important statement made by one
()
24 from the EDF when I ask him, he say, "Mr. Cheverdez, what 25 is important in standardization is to resist a long time to ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
32 1
the improvement the specialists want to make."
2 In fact, standardization, we do 3
standardization, but when there are enough reasons to 4
change the model and when we have been assured that we need 5
the improvement and it is worthwhile and so on.
Okay.
6 The last point to make is about the -- yes of 7
course, EDF is national, but it is managed as a private 8
company working for profit, produce kilowatt hours at the 9
best price f or the French user.
10 So generally the problems of money are not 13 involved in these problems of safety.
But supposing that 12 their exist some big problem that is raised.
9O U
13 I remember at the beginning of the program a 14 problem with cracks, detected in the first the 900 15 megawatts.
Cracks.
We were just at the beginning of the 16 starting of some plants.
l 17 Well, there was big discussions between safety 18 EDF and -- and they have the special groups,.netallurgists 19 and so on to study the problem.
20 Finally it was the decision, ministry itself l
i 21 as the sponsor, chief of saf ety, came to the parliament, 22 the Congress, and with samples of the steel explaining the 23 cracks were not dangerous as far as safety.
24 This of course was involving money because if 25 we stop, delay, construction delays of plants for five or ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
33 gy six years for a problem finally the specialists consider a 1
V 2
minor problem we probably would have lost millions.
So we 3
are not disconnected f rom problems of money.
4 These were some general statements.
5 DR. WARD:
Thank you, very much.
6 Chet?
7 DR. SIESS:
You said something in describing 8
this containment performance objective that, as I heard it, 9
seems to be diff erent than what was in NUREG as to have.
10 The footnote there says the emergency plan is 11 considered feasible if a gaseous release occurs more than 12 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> hours after the beginning of an accident and 13 maximum radioactive release includes a hundred percent of 14 the noble gasses inventory but only a few percent of the 15 iodine, cesium and other volatiles.
i 16 I look at your S 2 table up there and it seems 17 to me that that yould meet that criterion without the need 18 f or the U 5 leaded filter.
19 MR. SPEIS:
What page is this?
20 DR. SIESS:
Page 3-2.
You have gdt 75 percent 21 of the nobles, 2 percent of the iodine.
22 MR. SPEIS:
Right.
23 DR. SIESS:
How much of the cesium?
[(
24 MR. SPEIS:
5.5.
25 DR. SIESS:
See, that is a few percent.
i ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
34 1
MR. SPEIS:
You say 3.2?
2 DR. SIESS:
3.22.
It's it's a footnote there.
3 The footnote is on the feasible off-site 4
emergency plan state, statement, under 2 above there.
Do 5
you find it?
6 MR. SPEIS:
Yes, I find it.
I don't think 7
that is correct, the footnote.
We will have to check that.
8 I think the efficiency of emergency planning 9
and --
10 DR. SIESS:
I will take your word for it.
l 11 MR. SPEIS:
Yes.
12 DR. SIESS:
The one up there that says without 13 U S would not meet their requirement.
They need U 5 to 14 meet their containment guideline.
i 15 MR. SPEIS:
Of course, you know, they have 16 been operating these plants without filter systems up to 17 now.
18 DR. SIESS:
The difference is about an order 19 of magnitude on the iodine, cesium and strontium.
l 20 MR. RUBIN :
Was this slide for the Paluel or 21 900?
I 22 MR. SPEIS:
Both.
This is without filter for 23 both and with for both, approximately.
()
24 DR. SIESS:
Is that the U S another on the --
25 on the thousand down there, bottom line, it says with and ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
35 1
without.
V 2
Are those, is that U 5 or, I can't read it.
3 Looks like U 4, isn't it?
And U 4 is --
4 MR. SPEIS:
Is eliminate pathway for fission 5
product release.
6 DR. SIESS:
Yes.
I don't know what that 7
means.
Okay, I have got it.
U 4 is long-term total -- no, 8
that is H 4.
9 U 4 is a procedure to eliminate a pathway 10 through the drain system in the basement of the 11 containment.
12 It's U 5 up there on the the 900 but it's U 4 13 on the 1300, isn't it?
14 MR. SPEIS:
We better check this to make sure.
15 DR. SIESS:
This is talking about Paluel.
16 MR. SPEIS:
Yes.
17 DR. SIESS:
That is not inconsistent with the 18 NUREG; but it's of interest.
I think if U 5 is not needed 19 to meet their guideline on the Paluel, the next step would 20 be why?
21 You need it on the 900 --
22 DR. MICHELSON:
Different containment?
23 DR. SIESS:
I don't know.
They aren't that
()
24 much different.
25 MR. EB ERSOL E:
Filters?
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
36 1
DR. SIESS:
That is all right, go ahead.
2 MR. SPEIS:
I will check the paper.
3 DR. SIESS:
We are in the middle of this 4
containment business.
5 MR. SPEIS:
I will clarify it.
We just got it 6
this morning to explain this practicality of going from 7
here to here, but I see they distinguish between the the 8'
and 1300.
9 DR. SIESS:
Yes.
What is up there is not 10 inconsistent with the NUREG.
11 (Slide.)
12 MR. SPEIS:
I think the other vugraph that I 13 have here describes the procedures, themselves, total loss 14 of heat sink.
For example, all component cooling water, 10 15 RHE or emergency service water, total loss of feedwater, i
16 feed and bleed comes into being.
That is what the 17 procedure is all about.
18 H 3 isstation blackout, H 4 is long-term 19 total failure of saf ety inj ection or containment spray.
20 Then the ultimate procedures, U 1 to prevent 21 or delay core degradation, Al will talk in detail.
22 The second is containment leakage, so-called 23 beta failure we talk about in this country.
24 U 3, mobile units for core injection and 25 containment spray.
This is coupled with the H 4.
U 4, he l
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
37 4
1 be eliminate a pathway f or fission product release, and U 5 O
2 is the ultimate to prevent grass failure of the 3
containment.
4 DR. MICHELSON:
Is U 2 the containment vent i
5 valve failing to close?
,6 MR. SPEIS:
Yes..
7 DR. MICHELSON:
Is that what you said?
8 MR. SPEIS:
Yes.
9 DR. MICH ELSON:
Thank you.
10 MR. SPEIS:
Okay.
11 MR. EBERSOLE:
Dr. Speis, the last one, vented 12 containment, you didn' t say filtered.
13 MR. SPEIS :
Filtered vent containment.
14 MR. EB ERSOLE:
The other thing is, is that 15 really in the f ull spectrum of circumstances when its 16 f ailure because of failure to eject containment heat causes 17 it to over pressure can be carried away --
18 MR. SPEIS:
All containment cooling.
19 MR. EB ERSOLE:
Is it also in a selective way i
20 with a matrix of considerations vented when there is 1
21 activity in it but it's not yet really bad?
22 MR. SPEIS:
Yes.
23 MR. EB ERSOLE :
So they have a curve or
(
24 multi-element matrix or something that tells them how to 25 vent it?
Do they have those operational procedures ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
38 1
available?
~
b-2 MR. SPEIS:
Yes, they are available.
3 MR. EB ERSOLE :
Could we see translated what 4
they are?
5 MR. SPEIS:
Have they finalized the U S 6
procedures?
7 MR. RUBIN:
I think those procedures when we
~
8 were meeting with them were not yet completed.
9 DR. WARD:
They don' t have the systems 10 installed.
11 MR. SPEIS:
Right.
When they were there they 12 showed us pressures, what they were thinking of venting, 13 design pressure.
They were still evaluating these details.
14 MR. EB ERSOLE :
How close are they to 15 procedures to give to the operators, how long, how far away 16 time wise, do you know?
17 MR. RUBIN :
I am not sure.
When we were there 18 we were told they were doing research on the efficiency of 19 filters and -- probably complete by now.
20 MR. SPEIS:
I think now they have a schedule 21 more or less.
22 Mr. Cheverdez can tell us the schedule.
s 23 7
MR. CHEVERDEZ:
Yes, a few weeks ago I heard t-( j) 24 this has been officially agreed.
It was proposed by EDF 25 and agreed by the safety authorities and decided.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
39 1
D R. WARD:
To do what, begin construction?
,-O) 2 MR. CHEVERDEZ:
To have this filter vented, 3
filter in the containment.
4 MR. SPEIS:
Including procedures how to 5
operate and when to start and when not to start, things of 6
that sort.
7 MR. CHEVERDEZ:
Yes, of course, to be 8
consistent.
9 DR. WARD:
Wait a minute.
These are going to 10 have to be backfit to 36 or 40 plants.
11 MR. CHEVERDEZ:
Yes.
12 DR. WARD:
Has that started yet?
Os 13 MR. CHEVERDEZ:
This is another thing.
I 14 don't think we have the same approach on backfiting.
15 At this point these aspects of economy, 16 optimum of the electricity production will be probably 17 studies by EDF.
i 18 But if we make plans, if for the saf ety 19 aspects, backfiting cost more for EDF, this will be done.
20 But with some planning probably, not jeopardizing the 21 availability factor of the plant.
22 DR. WARD:
When you say decision has been i
23 made, you mean a decision has been made to put it in the
()
24 NH's or future plants or what?
I think it was agreed to put ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
40 1
on the NH's.
O-2 DR. WARD:
Okay.
3 MR. CHEVERDEZ:
I have to check that but I 4
think it was considered as a valid improvement.
5 MR. SPEIS:
My understanding is that, of 6
course, Mr. Cheverdez knows better, that they have also 7
decided to backfit them.
But they are trying to figure out 8
when.
Backfit to the 900 megawatts.
Right?
9 MR. CHEVERDEZ:
I cannot -- prefer not to --
~
10 MR. SPEIS:
Okay.
11 MR. RUBIN:
My presentation is primarily going 12 to focus on the design differences and some discussion of 13 them.
14 First let me say a little bit, briefly again 15 about Paluel.
16 (Slide.)
17 MR. RUBIN:
It's located in northern France on 18 the English channel.
There are four units on-site, all 19 operational f or a total capacity of 5200 megawatts electric 20 at that site.
21 In going through the project it was 22 interesting.
We received a number of French documents.
i ll 23 Sometimes the translations lef t a little to be desired.
()
24 I will give you a little example of literal 25 translations that I thought were from a list of pages.
i ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
~
41 1
Af ter power was really meaning decay heat.
2 Regular water was lightwater.
I can imagine 3
the translation in the reverse if someone was trying to 4
think what is lightwater.
5 Buildings in depression were not buildings 6
with tears coming out of their eyes, they were 7
depressurizeed buildings, vacuum.
8 Pulverized water wasn't water that was beaten 9
up, it was atomized water.
10 Just a picture of the site right here.
l 11 (Slide.)
12 MR. EBERSOLE:
The architectural features make
/*
13 it look good.
14 Do they use architects?
15 MR. RUBIN:
For the design, of the exterior?
16 MR. EB ERSOL E:
The overall visual aspects of l
17 the plant.
- I 18 MR. RUBIN
They did at the site excavated a 19 tremendous amount of material in construction so this would 20 be sunk even in along the coast and not visible for the l
21 approach to the plant.
22 MR. EB ERSOL E:
Is that a fact?
23 MR. RUBIN:
Yes.
It's built on cliffs along
(
24 the English Channel.
This entire area, you can see some of 25 the bedroom, tremendous excavation.
I ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
42 1
MR. EB ERS OLE:
That is a powerf ul regional 3
2 force to make the scene stay the way it ought to to look 3
good.
4 DR. MICHELSON:
How do they standardize the 5
plant and still fit it into local topography?
This must 6
have been somehow fitted in to the local situation.
They 7
can still do that with a standard plant?
8 MR. RUBIN :
Obviously there are some 9
differences in the construction or movement of material at 10 the site.
But the plant designs are standardized.
11 MR. EB ERSOLE :
It sounds like they dig out the 12 site to fit the plant.
l l
\\
13 DR. MICH ELSON:
Yes.
I l
l 14 MR. RUBIN :
For this site, particularly for 15 Paluel.
I am sure it had to do with the, some of the 16 esthetics of the site.
17 MR. SPEIS:
Imagine a great cliff.
18 MR. RUBIN:
The cliff was actually dug out.
19 MR. SPEIS:
You remember during the invasion 20 of Normandy.
Went into one of them and built the four 21 units.
22 MR. EBERSOLE:
We have one dam in the country 23 in which the mountains are carved to fit it.
()
24 MR. SPEIS:
Same thing here.
25 MR. RUBIN:
We determined there were ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
43 4
j 1
approximately 25 design differences.
All discuss today J
2 only 11, those we determined term were of moderate 3
significance or higher relative to all the features that we 4
looked at.
5 I won't discuss those of low, or low moderate 6
significance, but they are included in the report.
7 MR. EB ERSOL E:
Where does it get its steam?
8 MR. RUBIN:
I will get into each one of these.
9 These are the features I will discuss today one by one.
10 There are 11 on the list and 14 more discussed in NUREG as 11 to have.
12 Since we have talked about features that are i
s 13 either high significance, moderate to high or moderate, 14 there is certainly some subj ective judgment involved in 15 that.
We try to compare one to the other.
K 16 Let me read from a footnote to reiterate the i
17 use of the terms high, moderate, low.
They are not 18 intended to be absolute values.
Qualitative terms are used 19 to give insights on the relative importance of various 20 design differences.
i 21 If those features with a high safety i
l 22 significance were applied to U.S. designs, they could have f
23 a relatively significant saf ety benefit when compared to
()
24 the other design features that were identified.
l 25 The features designated as moderate or low if l
l l
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
~
44 1
applied to U.S. plants would have correspondingly lower 2
benefit.
3 For example, first the steam driven generator 4
would have higher safety significance than these two here 5
and in turn, these would have higher saf ety benefit than 6
the four aux feedwater pumps, for example.
7 DR. WARD:
You aren't making any judgment --
8 MR. RUBIN :
Not as to the absolute value, no.
9 DR. WARD:
Whether the top one is, would be 10 described as a major safety --
11 MR. RUBIN:
That's right.
These are relative 12 terms.
O
(/
13 MR. EBERSOLE:
Do they have a philosophy that 14 says they do or do not prefer in say some sort of safety 15 system that part of.is running all the time and thus will 16 surely f ail and you must operate on the residual fraction 17 of the system, or do they have them on total standby?
18 There is a difference in how you treat them, i
l 19 because the one that is running is going to be gone --
20 MR. RUBIN :
We will get into some of these 21 features.
You will see which are standby systems and how 22 there are interconnections, multiple trains, additional 23 pumps.
()
24 DR. SIESS:
Are all these systems what we 25 would classify as safety grade?
For example the first one l
l ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
45 1
up there.
2 MR. RUBIN :
No.
They are not all considered 3
to be safety grade.
4 DR. SIESS:
Do the French make the same 5
distinction between safety grade and nonsafety grade as we 6
do?
7 MR. RUBIN :
In terms of seismic criteria and 8
redundancy, let me give you an example.
9 DR. SIESS:
There is a lot more than seismic 10 criteria.
11 MR. RUBIN :
Redundancy, single failure.
12 DR. SIESS:
Quality assurance.
/^N 2
13 MR. RUBIN :
The steam driven generator for 14 example which is a design feature, the first I will 15 discuss, is used for an event, in event of loss of all AC 16 power.
17 If that were a saf ety grade system by our 18 definitions it would require to be redundant to accommodate 19 a single failure.
20 There is one steam driven generator.
21 MR. EB ERSOLE :
When they --
22 DR. SIESS:
Seismic isn't that much of an 23 issue.
l
[/
24 MR. RUBIN :
That's cor rect.
These are low
\\_
25 probabilities events relative to-he.
j ACE PEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
46 1
DR. SIESS:
Not protected against aircraft 2
crash?
3 MR. RUBIN:
They are inside, sometimes not.
4 DR. SIESS:
The first item is not, is it.
5 MR. RUBIN:
No.
6 MR. EB ERSOLE :
We have a lot of-trouble with 7
what I call philosophy of absolutism.
When it's safety 8
grade it's good, when it's not, it's no good.
9 There is always an intermediate stage in which 10 it's not safety grade but still mighty good.
What do you 11 do short of all the saf ety grade things like put pedigrees 12 on it, inspect it at intervals, but it still doesn't have 13 that last little touch which magically turns it into 14 something out of gold for us.
15 MR. RUBIN:
These equipment for example, steam 16 driven generator may not be safety grade but it's tested.
17 There are other --
18 MR. EB ERSOLE :
Pedigreed, QA'd?
19 MR. RUBIN :
Yes.
There are certain 20 requirements beyond just a little house generator.
You do 21 things to it you wouldn't do to it to make it run the 22 lights.
23 MR. RUBIN:
These are used for improvements 24 until safety.
25 MR. EBERSOLE:
Like flying an airplane with a l
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
47 1
gasoline engine.
It's a good deal beater.
2 DR. MICHELSON:
What about the things, it's 3
separated f rom events likely to get rid of normal service.
4 MR. RUBIN :
Right.
5 DR. MICHELSON:
Physical separation is an 6
important element.
7 MR. SPEIS:
Yes.
8 DR. MICHELSON:
Even though they may not go 9
through QA on the quality of the generator, itself.
10 MR. EBERSOLE:
Yes, they do.
11 DR. MICHELSON:
I don't know they do.
12 MR. EB ERSOLE:
He says there is an 13 intermediate level of qualifying.
14 MR. RUBIN :
We didn't look for example at 15 sys,tems interaction aspect in the review.
We assumed these 16 improvements wouldn' t have negative side ef fects.
17 MR. EB ERSOLE :
But they always do.
l l
18 MR. RUBIN:
For our scope of review, it was 19 not a design review.
Thatwqsthebasisofthe I
l 20 evaluations --
l 21 DR. MICHELSON:
Let me ask again.
Did you 22 check to see about physical separation?
23 MR. SPEIS:
Yes.
24 MR. RUBIN:
Yes.
25 MR. SPEIS:
We discussed that.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
I 48 T
1 MR. RUBIN:
In fact we were impressed with the 2
amount of room and availability for maintenance in lay down 3
areas and separation between --
i 4
DR. MICHELSON:
It's understood I assume, when i
4 5
I say ~ physical separation I mean including the services j
6 needed for this generator.
i
-7 MR. SPEIS:
Yes.
i 8
DR. MICHELSON:
As opposed to the --
9 MR. SPEIS:
Normal, to the normal, yes.
i l-10 DR. MICHELSON:
Okay, thank you.
I 11 MR. RUBIN :
Let me get back to this slide.
i 12 MR. SPEIS:
I want to comment, we haven't made gO) i v
13 the comparison with all the things that go to define our
{
1 14 saf ety grade versus theirs.
Let's be fair about that.
15 (Slide.)
i l
16 MR. RUBIN:
I will get into the meat of the I
17 discussion.
The first discussion is a steam driven 18 generator which is primarily used in event of a station 19 blackout.
20 In the SNPPS design, typical U.S. plants, i
i 21 reactor coolant pumps seal cooling is provided by two j
22 means, either injection using charging pumps or thermal l
23 barrier which requires availability of the component 24 cooling water system.
i 25 Both systems cooling modes are dependent on AC i
I i
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
49 1
power.
L 2
In addition at U.S. plants the instrumentation 3
and control systems necessary to achieve hot shutdown 4
condition are also dependent on AC or DC power.
5 MR. EB ERSOLE:
In a way this is a patch, you 6
go back to the seal itself.
Is there an effort to make it 7
heat resistant?
8 MR. RUBIN :
Not that I am aware of.
The 9
French looked at this approach, it was not very costly for 10 them to do this.
11 MR. EBERSOLE:
I know, it's practical.
13 12 MR. RUBIN:
Instead of spending a tremendous O-13 amount of money to, say, qualify the seals, they took this 14 practical approach and backfitted this on the the 800 15 plants as well.
16 DR. MICHELSON:
Single steam generator 17 overfill will get both these turbine driven devices.
18 MR. RUBIN:
This is again for the case of a 19 station blackout.
20 DR. MICHELSON:
Okay.
21 MR. RUBIN:
They have procedures --
i 22 DR. MICHELSON:
Just for that.
i 23 MR. RUBIN:
For station blackout.
I will
()
24 discuss features that apply to the station blackout 25 procedures as well.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
)
- =
50
~
1 The Paluel design is the same as for the SNPPS 3
2 plant, thermal barriers and injection for the component 3
cooling water system.
4 But they have added the additional AC 5
independent reactor coolant pump system which has the 6
turbine generator right here, draws steam from the main 7
steam line.
8 This is a generator of about 70 kilowatts of l
9 output.
It provides power to an existing hydraulic test 10 pump which is used normally to pressurize and pressure test 7
I 11
' the primary system.
12 The generator also provides power for use of 13 instrumentation and controls and emergency lighting that i
14 would be necessary to handle a station blackout event.
I 15 We on reviewing the safety significance felt 16 that this was of a high safety significance in terms of i
17 providing assurance of ability to cope with the station l
18 blackout through the use of providing seal coolings so the 19 seals would not degrade as well as providing DC power j
20 necessary for control and monitoring the situation.
I i
21 I will discuss later that this is one of t
22 several features that give the French a capability to cope l
23 with a long-term station blackout.
1 ()
24 DR. WARD:
So far as you are rating this l
25 important, I think during the period you made this review, l
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700 l
51 1
as I understand it, there have been some tests, perhaps f-w) 2 analysis of seal leakage which indicates that an uncooled 3
seal doesn't leak as much as it was thought perhaps to a 4
couple years ago.
5 I don' t know if, if that is true, has that 6
changed your ranking of this as one of the most important 7
features?
8 MR. RUBIN:
It would change it somewhat.
9 Eventually the seals will f ail.
It depends how long.
We 10 don' t know how long it is.
11 The French determined they have an objective 12 of reducing f amily of events, station blackout is one, to 13 low probability of event,10 to the minus 7 per reactor 14 year.
15 They have features that give them the 16 capability of coping with a blackout f or three days.
17 DR. MICHELSON:
Three days?
18 MR. SPEIS:
In conj unction with other 19 features.
20 MR. RUBIN:
Yes, I will summarize those at the 21 end.
This is one primary one.
22 DR. MICHELSON:
Part of the seal leakage 23 question I think was how much will they leak if they do
()
24 overheat.
25 DR. WARD:
Yes.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
52 1
MR. RUBIN:
Right.
2 DR. MICHELSON:
I thought there were numbers 4
3 like maybe 10, 15 gallons a minute.
The Catawba leak 4
recently of 43 gallons a minute, how does it fit into our l
5 thinking?
6 MR. RUBIN :
If the seals operate without i
7 failure, without the secondary seals failing or 0-rings, f
8 without the seals popping open, seal faces separating, if 9
they operate normally and you lose the charging system, I
10 thermal barrier, evaluations. by Westinghouse and also 1
11 confirmed by NRC independently would be ' leak rates of I
12 around 20 gallons a minute.
13 MR. SPEIS:
If the seals don' t f all, that is 14 important.
If you lose the 0-ring --
l 15 MR. RUBIN:
If you lose the 0-rings, if the 16 seal faces separate you have leakages of the order of a 17 hundred gallons a minute.
{
18 MR. EBERSOLE:
In the standardization context 19 seals leak at different rates, don't they?
i 20 MR. RUBIN:
They have diff erent designs and 21 could leak at different rates.
s 22 MR. EBERSOLE:
Why shouldn't we or they have 23 the best seals there are?
24 What prevents that, especially in the French 25 system?
l ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
53 1
MR. RUBIN:
The French had determined several O-2 years ago that based on some --
3 MR. SPEIS:
Testing.
4 MR. RUBIN:
Testing that they had done that it 5
looked like the seal temperatures were heating up.
6 MR. EBERSOLE:
Those seals.
7 MR. RUBIN:
Their seals.
By the way you 8
mentioned earlier where some tests that were conducted 9
recently in France of a French reactor coolant pump seal 10 which is similar but not exactly the same as U.S. design 11 pumps, different sizes.
12 MR. EB ERSOL E:
My impression --
k-13 DR. WARD:
Wait.
He is trying to say 14 something, Jesse.
15 MR. RUBIN:
They didn't run a seal test to 16 failure.
17 MR. EB ERSOLE:
They didn't?
18 MR. RUBIN:
No.
It looked like they were 14 19 starting to get increase in flow.
They said if we 20 extrapolate this over time we see a problem.
They didn't 21 really determine that in a station blackout we have 22 hundreds of gallons a minute.
23 MR. EBERSOLE:
My point is this, I think.
()
24 MR. RUBIN :
But they decided this was a fix 25 that would preclude that from happening.
They didn't need ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
54 4
1 to do all kinds of analyses and evaluations.
That is as
-sC-2 much as I can say on it.
3 MR. EB ERSOLE :
What you recognize is a 4
problem, a fault, def iciency.
Fine, you have to make a 5
fix.
6 In the background you say you should have done 7
it in the first place.
It should have been instrinsically 8
not needing a fix.
I think there are seals who don't.
9 DR. MICHELSON :
Whose seals don't?
10 MR. SPEIS:
I don't know what that means.
11 Have you gentleman been briefed on the issue of -- for 12 example Westinghouse in this country has identified all 13 rings are, could fail.
14 MR. EBERSOL E:
0-rings.
15 MR. SPEIS:
0-rings.
It's an integral part of 16 the seal.
17 MR. EB ERSOLE:
Yes.
18 MR. SPEIS:
I think one of the solutions is to 19 make sure they do change.
20 I think they on their own, you know, are 21 making the appropriate modifications.
22 MR. EB ERSOLE :
Let's say, I don't know this 23 but let say B & W has seals that don't need such cooling.
f()
24 How do we get our hands on them in a Westinghouse machine 25 or a French machine, if that is true?
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
55 1
MR. SPEIS:
We are looking at all pumps, both E
2 B & W and CE and Westinghouse.
3 MR. EB ERSOLE :
Is it true they don't leak as 4
badly?
5 MR. SPEIS:
I don' t have the latest 6
evaluations.
7 MR. RUBIN :
We have more evaluations on the 8
Westinghouse seals and they were felt more vulnerable.
9 That issue has not been resolved yet.
10 MR. SPEIS:
One of the things that made us 11 focus on Westinghouse, as Al said they did the test, the 12 French, they saw the temperature rising and, they didn' t iG' k-13 take it to ultimate failure.
14 MR. EB ERSOLE :
They found a vulnerability.
15 MR. RUBIN:
In f act the system is designed, 16 this steam driven generator is designed to start i
17 automatically and provide cooling within two minutes to the l
l 18 pump seals.
l l
19 MR. EBERSOLE:
I understand.
1 20 MR. RUBIN :
We don't think the seals will leak 21 in that short a time.
22 MR. EBERSOLE:
There is more than one --
23 DR. MICHELSON:
There is more than one aux
()
24 feedwater pump I assume, but is there just one of these 25 turbine generator arrangements?
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
56 1
MR. RUBIN:
Yes.
U 2
DR. MICHELSON:
Thank you.
3 DR. WARD:
How much more time do you think you 4
are going to need?
5 MR. RUBIN:
Depends (n1 your questions.
6 DR. WARD:
Okay.
7 MR. RUBIN:
Longer than I think, probably.
8 DR. WARD:
How much more presentation.
9 MR. RUBIN:
I can get through, at least half 10 an hour.
1 11 DR. WARD:
Let's hear what Allen's got to say.
12 MR. EB ERSOLE :
Okay, no questions.
13 DR. WARD:
And hold the speeches for awhile if 14 we could, okay?
15 MR. RUBIN:
I will try to cut down on some of 16 these and just hit the highlights the next feature I am 17 going to discuss, emergency core cooling system.
18 (Slide.)
19 MR. RUBIN:
I am going to discuss this before 20 I discuss a feature which we determined of higher safety 21 significance which was the interconnection between the 22 containment spray and low pressure safety injection.
23 This is to get a better understanding of the
()
24 designs of the emergency core cooling system.
l 25 This design shows the SNPPS ECCS with two l
ACE FEDERAL REPORTERS, IljC.
Washington, D.C.
(202) 347--3700
57 1
safety inj ection pumps, two centrifugal charging pumps and
.yJS 2
RHE pumps which are used for low pressure saf ety inj ection 3
following a LOCA.
4 The pumps, saf ety injection and containment 5-can draw suction f rom the ref ueling water storage tank 6
long-term following a LOCA near the RHE pumps which take 7
suction from the containment sump.
And then, if needed, 8
can piggyback or feed into the suction side of the charging 9
f or the safety injection pumps.
10 You will see the French have quite a simpler 11 design because they have added two pumps to their system.
12 DR. MICHELSON:
Those are full head also on 13 the SNPPS.
i 14 MR. SPEIS:
15 MR. RUBIN:
Yes.
16 DR. MICHELSON:
Up to the relief point of the 17 safety valve; is that right?
18 MR. RUBIN:
These are safety grade, 19 redundant --
20 DR. MICHELSON:
No, I asked a different 21 question.
Can they go all the way up to the relief point 22 of the saf ety valve?
23 That is what I meant by full head.
()
24 MR. SPEIS:
Do you have to depressurize, for 25 exaiaple.
That is the question.
l...._-_ _______ ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347 -3700
58 i
1 MR. LE FAVE:
I am Bill LeFave.
73
'd 2
They are full head.
3 DR. MICHELSON:
What does that mean?
4 MR. LE FAVE:
They can, normal operating 5
pressure, not like the lower medium.
15 6
DR. MICHELSON:
Do you have to depressurize to 7
bring them in.
8 MR. LE FAVE:
If you are up to the saf ety 9
valve setting, I don't know.
I don't know they can go up 10 that high.
11 MR. SPEIS:
I think they are very close.
4 12 DR. MICHELSON:
I think they are.
13 DR. WARD:
Perhaps Westinghouse could give us 14 a crisp answer to the question.
15 MR. SCHULZ:
You do have to depressurize 16 for SNPPS.
But the pumps also can open safety valves.
17 DR. MICHELSON:
Can they inject at normal 18 system pressure?
They can, thank you.
19 DR. WARD:
Questions, but no speeches.
Do you 20 have a question?
21 MR. EB ERSOLE:
Sure.
I am looking at a 22 singular header and valve.
Is that in fact true.
On all i
23 the pumps I take a singular suction header and valve
()
24 between it.
j l
25 MR. RUBIN :
Right here?
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
59 1
MR. EB ERSOLE :
Yes, sir.
2 MR. RUBIN:
This is again represented 3
schematically, representation.
4 MR. SPEIS:
Bill?
5 MR. LE FAVE:
I don' t remember.
6 MR. EB ERSOLE:
I think it's very important.
7 DR. WARD:
They don't know the answer.
8 MR. RUBIN:
It would be, and it's something 9
that was, we don't have the details right now.
10 MR. EB ERSOLE:
There is a rationale that would 11 defend it, we will make it super conservative, especially 12 for seismic.
13 MR. SPEIS:
I think that is a separate issue.
14 MR. RUBIN :
These are not engineering 15 drawings, these are schematic.
They were intended to 16 represent differences, to point out differences so they may 17 not have all the details that will be able to answer your 18 question.
I just apologize for that.
19 One other point to be made is that when the 20 tank is empty, the valves are realigned by the operators i
21 manually on the SNPPS plant which is somewhat different 22 than the Paluel plant.
23 MR. LE FAVE:
Excuse me.
On page 2-5 it shows j ()
24 them having two, train A and B coming f rom the ref ueling 25 water storage tank.
A('E FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
60 1
MR. SPEIS:
That is Paluel.
The previous one 2
is SNPPS.
3 DR._MICHELSON:
Yes, this is.SNPPS.
Paluel 4
does show it.
5 Next slide.
6 MR. SPEIS:
Yes.
7 (Slide.)
8 MR. EBERSOLE :
There we go.
9 MR. RUBIN : ~ Here is the -- let's forget about 10 the containment spray pump for a second on this slide of 11 Paluel.
12 I am showing you there are basically -- events 13 following LOCA requirements for safety injection.
What are 14 shown on this figure are the safety injection pumps and low 15 pressure safety injection pumps.
16 MR. EB ERSOL E:
May I ask a question?
17 MR. RUBIN:
Certainly.
18 MR. EBERSOLE:
What is the basis or rationale 19 f or separation under LOCA conditions when you don't 20 presumably have a coincident need for separation out in the 21 ordinary regions of the plant?
22 MR. RUBIN:
Part -- Well there are several 23 features.
One, they have the RHE pumps inside containment.
()
24 MR. EBERSOLE:
You are talking about the RHE 25 in context of heat removal.
I ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(2021 347 3700
61 1
MR. RUBIN:
This is some of the rationale the
. f3 V
2 French have used.
RHE are inside containment, they don't 3
serve the safety injection pumps.
Safety injection pumps i
4 are located outside containment.
5 We were told at additional expense, but partly I
6 for simplification of operation, it allowed them to use an 7
automatic switch over from injection to recirculation as 8
part of the operational mode and part of the safety 1
9 rationale that they took this approach.
10 As in the SNPPS design on loss of water, 11 refueling water storage tank, there is a changing mode 12 which is automatic change to take suctions from the i O'
\\-
13 containment sump either to the low or intermediate pressure 14 saf ety inj ection pumps.
15 Some of the rationale from a safety standpoint 16 was potential, the French considered was potential operator 17 failure in dealing with these manual operations of 18 switching over f rom injection to recirculation.
I 19 There are a number of valves that have to be 20 changed, not automatically.
Obviously f ewer valves with 21 the French design than on the U.S. design.
l i
22 Let me move on to another design feature.
t 23 (Slide.)
()
24 DR. MICHELSON:
Are you going to tell us l
25 eventually about their depressurization capability which ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
62 1
has to go with those intermediate pumps?
gd 2
MR. RUBIN:
They still have the charging 3
pumps.
I was not planning on discussing that.
4 DR. MICHELSON:
That is just low capacity 5
stuff.
6 MR.. RUB IN :
Yes.
7 DR. MICHELSON:
How do they do their, have 8
they got two PORV's or three or one?
9 MR. EB ERSOLE :
Three.
10 MR. RUBIN:
They have PORV's.
The number is, 11 maybe while I am talking, Bill, could you look and see if 12 we have got that in the report?
I will try and move along 13 with the discussion.
16 14 Certainly with the operation of the PORV's.
I 15 don't know if we identified the number.
I think we did.
I 16 don't recall offhand.
17 DR. MICHELSON:
I don't recall reading it l
18 either.
That is why I asked.
19 Before I talk on the Paluel design which is 20 shown on the screen, the SNPPS design, the low pressure 21 saf ety injection system and containment spray system are 22 independent of each other.
m 23 There are no interconnections which would 24 provide long-term cooling following a LOCA if the RHE 25 system were unavailable in the recirculation mode.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
63 1
These are two items that are related to these 2
design features and procedures, as well, that the French j
3 have included.
4 Bill?
l 5
MR. LE FAVE:
On the PORV's they have three 6
valves that can be operated from the control room.
I think l
7 two also serve as a saf ety relief valve but.they can be 8
pilot operated so they can be pressurized f rom the three 9
valves f rom the control room.
10 MR. RUBIN:
Thank you.
I thought we had that 11 in there.
12 The Paluel design, there is interconnection 13 between the low pressure safety injection pumps and the 14 containment spray pumps as shown here.
l 15 This low pressure, I am corry, this 16 interconnection is designed in the event of total failure I
17 of either both low pressure safety injection pumps or both 18 containment spray system pumps.
This is again the French l
i 19 philosophy looking at multiple failures of redundant safety 20 systems.
21 Should a, both spray pumps fail, the flow path 22 would be say from the containment sump to the containment 23 spray pump and then opening these valves to the injection 24 line, both this low pressure spray pump and this one, if l
25 they were unavailable.
That is for short-term cooling ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700 i
64 1
following a LOCA.
O 2
Longer term following a LOCA the French have 3
included an alternate means of cooling.
In a time period 4
of two weeks af ter a LOCA, decay heat is reduced to a lower 5
lev el, the French have mobile pump and heat exchanger that 6
can be brought to a site.
7 These pumps and heat exchangers can generally 8
within a two-day travel time to any of the French nuclear 9
plants, P 4 plants and will take about a day to connect.
10 There are flanges at the plant that are accessible where 11 these pumps would be hooked up.
12 The French also have, this is part of their U kJ 13 3 procedures.
14 Should all four these pumps fail, spray pumps 15 and low pressure safety injection pumps, af ter two weeks 16 this alternate means of cooling, use of mobile equipment 17 would take suction f rom the sump again and have a flow path 18 either through the regular heat exchanger if that were 19 available and through the injection lines.
If the heat 20 exchanger were also unavailable for one reason or another 21 it could flow through the heat exchanger as well, long-term 22 cooling.
23 MR. EB ERSOL E:
They look at long-term 24 degradation of the process, right?
25 MR. RUBIN:
Right he be.
l ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
65 1
DR. MICHELSON:
They use a potentially
- O 2
radioactive source, namely the sum.
{
3 MR. RUBIN:
The location of the mobile pumps 4
would be somewhere in the auxiliary building.
It's not 5
just outside, f
6 DR. MICHELSON :
They bring them in through the 7
doors and inside?
8 MR. RUBIN:
Yes.
9 DR. MICHELSON:
They are partly contained and 10 shielded.
11 MR. RUBIN:
They are in a shielded area where 12 they can work around it.
\\
13 DR. MICHELSON:
It's just another pump already 14 to be plugged in.
15 MR. RUBIN:
Right.
We thought that was an 16 interesting feature.
17 MR. EB ERSOL E:
Flanges are there.
18 MR. RUBIN :
They are there, sticking out about 19 this high off the ground, i
20 DR. MICH ELSON :
I guess it's a unique pump 21 then because they aren't worried about seal leakace and and 22 so forth.
23 MR. RUBIN:
They just want to keep the water
()
24 going.
25 DR. MICHELSON:
Even though it may be very hot ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
66 4
1 radioactively.
2 MR. EBERSOL E:
Does it have pump -- that is a 3
detail.
4 MR. RUBIN:
Let me get to one of the first 5
questions you asked which is how in-depth did we get into 6
the review.
This is about the level of depth.
We didn' t 7
get into the design of the pumps or seals or anything else.
8 MR. EB ERSOLE:
Yes, okay.
9 (Slide.)
10 MR. RUBIN :
I will go back to one of the 11 features that we don't have a slide for.
Talk for a minute 1
j 12 about the use of micro processors and computerization that 13 the French make wide use of in their plants.
4 14 In the SNPPS design for the reactor protection 15 systems sensors that are monitor system parameters are hard l'
16 wired through a logical processing system to provide l 17 17 reactor protection system.
Two out of four of the sensors, I
18 if parameters exceed the set point the signal is sent to 19 trip the reactor.
20 DR. MICHELSON:
They don't use multiplex, 21 then?
1 I
22 MR. RUBIN :
Yes.
23 MR. EB ERSOLE :
Do they buy quality grade stuff
()
24 such as Army, Navy quality solid state equipment?
l 25 MR. RUBIN:
The French?
-3700 ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347
.~
67 1
MR. EB ERSOLE:
I am talking about the earlier 2
questions we had about spikes, ambience.
3 MR. RUBIN:
I wasn't here for that part of the 4
discussion.
5 MR. EB ERSOL E:
I am sorry.
When you use this 6
stuff you bring up instantly the problem of the delicacy of 7
the system to the ambient conditions including cooling 8
within their own boxes not to mention atmospheric cooling 9
that supports that.
10 You build in the course of using these things 11 a new dependency of the cooling f unction f or the apparatus 12 which the old electro magnet particulars only crudely
(
13 needed.
14 DR. WARD:
Is there a question coming, Jesse?
15 MR. EB ERSOLE :
Yes.
The question is, do they 16 encompass in the use of these attention to guarantee that 17 if they get too warm and go crazy before they do so, they i
18 will take the --
19 MR. RUBIN:
They have ventilation and 20 circulation --
21 MR. EBERSOLE:
Oh, everybody does.
22 MR. RUBIN:
For the systems.
Obviously they 23 have some dependencies there.
That was beyond the level of
()
24 detail --
25 MR. EBERSOLE:
They don' t upset the f unction ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
68 1
of the systems if they get too hot?
2 DR. WARD:
An rope appropriate answer could be 3
you don't know.
4 DR. MICHELSON:
Did they take into account 5
loss of ventilation with loss of off-site power?
6 MR. RUBIN:
Which equipment that they would 7
need to monitor the reactor, yes.
That is obviously not 8
all the equipment --
9 DR. MICHELSON:
They don' t do a systems 10 interaction study on the effect of their elevated 11 temperature or do you know?
12 MR. RUBIN :
I don't think they have done it in 13 detail.
They may be starting to do some of that.
14 MR. LE FAVE:
As I recall they were doing 15 validation or verification test of all these systems.
They 16 hadn't finished them as yet.
17 So the electrical people actually looked, who 18 looked at some of that stuff are not here.
The program, 19 how they were approaching it, the electrical people were 20 happy with how they were approaching it as f ar as 21 validation and verification.
22 DR. MICHELSON:
I don't know if that helps me 23 any.
()
24 MR. RUBIN :
Unfortunately the fellow who did 25 the review of this particular item is on travel today.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
. ~ _. _.
69 1
MR. LE FAVE:
-- safety grade, busses and all
~
2 that.
3 MR. SPEIS:
We did discuss-this with them.
4 They are concerned.
They were aware that, you know, you 4
5 does want to just. add a system and then ventilation or some j
6 other problem will knock out everything.
I don't know to 7
what detail they went but it's something they did take into j
8 account.
1
{
9 MR. LE FAVE:
As I understand you have hard' 10 wired backup on the control --
11 DR. MICHELSON:
That is not the problem, of i
12 course.
1 1
i 13 Did you look into their, their cooling methods 14 at all?
i l
15 MR. SPEIS:
No, we just discussed it in i
]
16 general there.
17 DR. MICHELSON:
Do they still have to have i
18 chilled water systems for compartment cooling or do you i
19 know?
20 MR. SPEIS:
We didn't go into details, as I 21 said.
We were sensitive at that time because of us having l
l-22 gone through the blackout issue.
i l
23 They told us that they were sensitive
. h 24 themselves and were making sure to some degree those things i
25 were considered.
i l
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
70 1
We didn't go into great dets.il,
-b) 2 MR. RUBIN:
We had about 30 items to discuss 3
in a week time period and were able to spend a couple hours 4
on each.
5 DR. WARD:
Okay, let's go.
6 MR. SPEIS:
Five more minutes.
7 MR. RUBIN:
I have five more minutes?
8 DR. WARD:
Well, let's keep going.
9 MR. RUBIN:
Thanks a lot.
10 The design for the micro busses, I won't go 11 into details of it, but it allows quite a difference in the 12 control room, you will see on the next slide.
\\-
13 Use of computerization, a totally diff erent 14 capability in presentation of information to the operator.
15 When we evaluate saf ety significance of these 16 features, this particular feature is related to the control 17 room, which I will show on the next couple of slides.
10 (Slide.)
19 DR. MICHELSON:
I assume this is still done in 20 redundancy or triplicate, so if you get one spurious 21 channel of information you, what do you do then?
22 Do you have two more to look at, or just one 23 more?
()
24 MR. RUBIN:
They have redundant systems in the 25 control room.
l l
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
71 l
1 DR. MICH ELSON :
Then you have to guess as to
-~
J 2
which is spurious and which is right if you have only got 3
two to look at.
- 18 4
I don't know what their philosophy is on 5
f ailure of micro processors but they do f ail.
6 If th ey do, it~isn't always obvious which 7
channel is the failed one and which one is the one to 8
believe.
9 MR. LE FAVE:
They have a self-checking 10 feature, too.
11 DR. MICHELSON:
But when the processor fails, 12 the self-checking features can also fail.
13 MR. EBERSOLE:
They don't use --
14 MR. LE FAVE:
They have a two out of four 15 logic.
16 DR. MICHELSON:
Okay, now we are getting 17 closer.
So they have four channels to look at.
18 MR. RUBIN :
Yes.
19 DR. MICHELSON:
A lot of it I didn't 20 understand.
21 (Slide.)
22 MR. RUBIN:
I think you will find pictures of 23 the control room rather interesting.
The first slide I
()
24 will show is of the SNPPS control room.
25 DR. MICHELSON:
Is the plant running?
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
72 1
MR. RUBIN :
This was taken at Wolf Creek s
U 2
bef ore the plant was operational which is the reason you 3
will see a lot of alarms and indicator on the panels.
4 Since Three Mile Island U.S. control rooms have gone 5
through a design review to identify and correct certain 6
areas, human factors areas of concern.
7 Some modifications have been made, although 8
the control room design has not been changed.
9 Modifications such as including some mimic diagrams on the 10 panels, color coding for functional -- color coding 11 equipment and instrumentation equipment on the panel 12 according to functions, regrouping of some indicators and 13 relabeling some of the instrumentations and control.
14 I will show you in contrast the P 4, Paluel 15 control room.
16 (Slide.)
17 MR. RUBIN:
It may not show up bright enough 18 on the picture.
I 19 DR. WARD:
Do you want to kill the lights.
20 MR. RUBIN:
The design at Paluel relies 21 heavily on the use of computer technology.
22 SNPPS plants do use some cathode ray tubes for 23 information.
Paluel relies on them a lot more.
They have
()
24 11 color CRT's in the Paluel control room that have input 25 from a central data processing system.
,_ -3700 INC.
(202) 3 47
.-ACE FEDERAL REPORTERS,_
- Washington, D.C._ _ - _ _ _ _ _ -
73 1
Seven of these 11 display alarms, there. are 2
alarms displayed on the CRT.
3 DR. MICH ELSON:
Do they have two central 4
processing units?
5 MR. RUBIN:
I believe there are two, 6
redundancy, yes.
7 DR. MICHELSON:
Yes thank you.
8 MR. RUBIN :
Alarms on the screen and also the 9
windows that are shown here for some significant safety 10 systems.
11 MR. EB ERSOL E:
When things go haywire in the 12 room is there a nice place to go to in the closet where f) k/'
13 they can take it home?
14 MR. RUBIN:
With the computerization they have 15 prioritization of alarms so there will be a maximum of five l
16 alarms.
17 MR. EB ERSOL E:
I don't mean that.
I mean I 18 have, oh, an intrusion fire that takes this room out.
19 MR. RUBIN:
They have a backup control panel.
20 MR. EB ERSOLE :
Is it independent.
21 MR. RUBIN :
Yes.
22 MR. EB ERSOLE :
Of the wiring and everything 23 associated under and a letter above this control room.
(
24 MR. RUBIN:
Yes.
25 MR. EBERSOLE:
Good.
[
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
74 1
MR. RUBIN:
Safety significance is that the 7--
2 use of computerization, the alarms that are supressed 3
during certain situations and presentation of material, of 4
information to the operators.
5
( Slide.)
l 6
MR. RUBIN:
Some examples, a close up on the I
7 panel.
Slide shows the mimic diagrams and color coding of 8
the different trains for -- this is an active mimic diagram l
9 showing the pressurizer control.
10 DR. SIESS:
What does active mean?
11 MR. RUBIN:
That there is a flow path shown 12 here and there are operator controls that the operator 13 has -- these are switches called turn / push light switches.
14 the operator pushes it in, turns it and it will open or 15 close valves depending on the --
16 MR. EBERSOLE:
Is there a physical name on 17 each of the elements?
18 MR. RUBIN:
Each of these?
19 MR. EB ERSOLE :
Each thing you twist, does it 20 have a name on it?
21 DR. SIESS:
It's in French, Jesse.
22 MR. EBERSOLE:
You know, the Germans don't put 23 any, or they didn't.
()
24 MR. RUBIN:
No, it's the.e.
25 So this is what you call an active mimic ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
l 75 1
diagram.
2 (Slide.)
3 MR. RUBIN:
Here is an example of-generally 4
steam flow through the generators which is passive.
There 5
are no controls on here, but there are light indicators 6
showing valve positions and steam flow.
7 DR. SIESS:
What is that on top?
8 MR. RUBIN:
I am not sure what that is.
It's 9
very prominent in the control room.
10 DR. SIESS:
It's changeable.
I assume that is 11 a panel that lights up.
12 MR. RUBIN:
It does, and I am not sure what 13 that is.
14 DR. SIESS:
You can read it better than some 15 of those --
16 DR. MICHELSON:
It stands out.
17 (Slide.)
38 MR. RUBIN:
Back to the first slide of the
'19 19 control room.
It's a prominent feature, right up here 20 DR. SIESS:
Where are all the enunciators?
21 MR. RUBIN:
Where?
22 DR. SIESS:
Yes.
23 MR. RUBIN :
They are shown through the CRT.
()
24 DR. SIESS:
Oh, okay.
25 MR. RUBIN:
They are not on here.
There are ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
i 76 i-1 1
some here, two trains, about a dozen or so alarm windows.
I.
}
2 The rest of the alarm systems show up on the --
i 3
MR. EBERSOLE :
They are color coded aren't 4
they?.
i 5
MR. RUBIN:
Yes.
6 MR. EBERSOLE:
They require operators that 7
have full color capability, is that it?
q 8
MR. RUBIN:
We didn't check the color l
9 blindness of the operators.
Again, that. is a level of 10 detail we didn't get into.
11 MR. EB ERSOLE:
Sometimes the x-ray inspectors i
{
12 can't see, so it looks good.
i 13 DR. MICHELSON:
Those are the best kind.
14
'MR.
EBERSOLE:
You get a good product..
l 15 MR. RUBIN:
It's hard to quantify what kind of l
16 safety significance, the physical layout in the control i
17 room.
We felt this was really at least of moderate i
j 18 significance.
i l
19 DR. MICHELSON:
Are those TV monitors?
l 20 MR. RUBIN:
There are TV monitors that aren't i
i 21 shown.
I l
22 DR. MICHELSON:
What is that big thing in the i
23 middle way up above?
l
()
24 DR. SIESS:
The orange balls up there.
25 MR. RUBIN:
Lighting.
1 I
i ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
T 77 i
1 DR. MICHELSON:
All that big box?
That big 2
box --
t 3
MR. RUBIN :
I didn' t see anything hidden in f
4 there.
They were lighting the room.
5 DR. MICHELSON:
Are there TV. cameras in there?.
6 DR. WARD:
He said no.
Let's go.
7 (Slide.)
8 MR. RUBIN:
We can get out of the control 9
room.
10 DR. MICHELSON:
Do they have something proven 11 of S.P D S, is that a part --
12 MR. RUBIN:
Yes, I can talk a little more l
13 about the control room.
14 DR. MICHELSON:
I just asked the question.
15 MR. RUBIN:
They have a safety engineer who 1
16 would come into the room in event of an accident.
He has 1
l 17 his own separate panel.
f 18 He has the capability of evaluating the l
19 situation independently of the operators.
20 DR. MICHELSON:
Thank you.
21 MR. RUBIN :
He comes in.
Some of the features j
22 on that panel are very interesting, by the way.
With the i
23 computerization they can show trends for the past 30
()
24 minutes of plant parameters, even project trends ten 25 minutes into the f uture.
i l
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
78 1
MR. SPEIS:
Those are in the report?
U 2
MR. RUBIN:
Yes, they are in the report.
3 Thank you.
4 Let me move along.
5
( Slide.)
6 MR. RUBIN:
I think we can go through some of 7
the others fairly quickly.
8 This shows the aux feedwater system.
9 Basically, the SNPPS design has three auxiliary feedwater 10 pumps, two are motor driven, one steam driven pump which 11 can feed any combination of four steam generators.
12 (Slide.)
13 MR. RUBIN:
Paluel design has an additional 14 two trains.
The trains are independent of each other.
15 There are two motor driven pumps and two 16 turbine driven pumps that each train feeds a pair of steam 17 generators.
18 There is an additional turbine driven aux 19 feedwater pump on the Paluel plant.
Capacity of these two 20 steam driven pumps is the total capacity of two of them is 21 the same as the single pump at the SNPPS plant.
22 Capacity-wise, they have the same capability 23 but there is additional redundancy for the second pump.
h 24 This redundancy we feel is of benefit primarily if there is 25 a loss of all AC power and, theref ore, then they have two ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
I 79 1
steam driven pumps to provide core cooling.
gS U
2 DR. MICHELSON:
You said they can go for three 3
days on a blackout?
4 MR. SPEIS:
Yes.
5 MR. RUBIN:
Yes.
6 DR. MICHELSON:
They have taken care of the 7
ventilation questions and all that for three days?
8 MR. SPEIS:
Yes.
In fact that was the reason 9
we asked them, have you seen the other things that happen 10 as a result of blackout.
They said yes.
11 MR. RUBIN:
Yes.
12 DR. MICHELSON:
They are satisfied they can 0'
\\-
13 keep rooms cool if they have to be kept cool and whatever?
14 MR. SPEIS:
Yes.
i 15 DR. MICHELSON:
Thank you.
l 16 MR. RUBIN:
The pump rooms, there is a, there l
17 are large volumes of air in the pump rooms.
18 (Slide.)
19 DR. MICHELSON:
I was thinking more of solid 20 state control equipment and you know, control room, that 21 sort of thing for three days.
22 MR. RUBIN:
I think that is something they are 23 looking into still.
()
24 MR. SPEIS:
I think the CA required, or the 25 agreement was that the plant should be able to withstand a ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
80 1
station blackout f or much less but when they went back and 7g V
2 did the analysis to see how long it can cope given 3
everything that they installed, then they found out that 4
they thought it can withstand for three days.
5 MR. EBERSOLE:
I presume as do weather 6
critically dependent on direct current.
7 MR. RUBIN:
Yes.
8 MR. EBERSOLE:
What keeps the batteries 9
charged?
10 MR. RUBIN:
The first feature of the steam 11 driven generators provides power.
12 MR. EBERSOLE:
How big is that generator?
13 MR. RUBIN:
70.
14 MR. EBERSOLE:
Glory, that is house size.
20 15 MR. RUBIN:
Pretty small.
They don' t need a 16 lot of power during blackout.
17 DR. MICHELSON:
They won't do any ventilating 18 with 70 kilowatts.
19 MR. RUBIN:
They have manual remote control of 20 the aux feedwater system.
It's not done at the control 21 room.
22 MR. EB ERSOLE:
Full manual control.
23 MR. RUBIN:
Yes, f or the aux f eedwater flow.
()
24 That is part of their H 3 procedures for station blackout.
25 MR. EBERSOLE:
What do they do about gauge ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700 1
81 I
glasses and other things that tell them what they are 2
doing?
3 MR. RUBIN:
They are at the remote station 4
mechanical?
5 MR. RUBIN:
I don't know.
They are AC 6
independent.
7 DR. MICHELSON:
I think they are not 8
independent of this little generator though.
9 MR. LE FAVE:
Turbine generator supplies 10 minimum to you controls and instrumentation.
11 MR. EBERSOLE:
I have got it.
12 MR. RUBIN:
We are down here on the features.
13 There are a couple to go.
Reply of condensate storage 14 tank.
15 The SNPPS plant, condensate storage tank is 16 resupplied with water from the demineral licensed water 17 storage tank using AC powered pumps.
18 In the Paluel design the tank can be 19 resupplied independent of AC power.
20 MR. EB ERSOLE:
Queation.
21 MR. RUBIN:
Yes?
22 MR. EBERSOLE:
Also such a thing as you have 23 to have an open Michael or everything fills up solid and h
24 you are in big trouble.
f 25 How do you get rid of the water --
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
82 1
MR. RUBIN:
They don't open the valves unless 2
they need it.
It's part of their procedure.
Resupply of 3
the tank?
4 MR. EB ERSOLE:
No, I am talking about you pump 5
water in forever and ever and don't get it out.
6 MR. RUBIN:
No, this is condensate storage 7
tank.
8 MR. EBERSOLE:
I am sorry.
To that is vented 4
9 to atmosphere.
The steam that is made f rom that.
10 MR. RUBIN:
Oh.
Yes.
11 MR. SPEIS:
Yes.
12 MR. RUBIN:
It's vented through the --
13 MR. EB ERSOLE :
Do they have systems that pump 14 water toward containment but don't extract it from it at 15 the same rate?
16 We were just looking at one the other day that 17 did that.
18 MR. RUBIN :
Not that I know of.
19 MR. EB ERSOLE :
It fills up, of course.
20 MR. RUBIN:
This is for secondary side 21 cooling.
22 MR. EB ERSOL E:
I know that.
I am making it in 23 the general context.
()
24 MR. SPEIS:
I don't know.
25 MR. RUBIN:
The RHE system and their low ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
83 1
pressure safety injection pumps are separate as we talked
,--)
v 2
about earlier but the actual flow rates --
3 MR. EB ERSOLE:
They are closed cycle.
4 MR. RUBIN:
5 MR. RUBIN:
At Paluel it's gravity fed, so 6
basically to get flow the operator opens valves.
That 7
provides about 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> of water supply.
8 Additional water supply is provided at see 9
shore sites by additional tanks on the site.
10 Also gravity fed drain tanks could provide 11 water for three days and at sites that are inland or on 12 rivers they have diesel driven pumps that can, that are (rN.
'd 13 connected through the fire water system or similar to the 14 fire water system that can resupply the condensate storage 15 tank.
16 Saf ety significance of the feature obviously 17 is to provide long-term water supply for decay heat 18 removal.
19 The RHE system 1s inside containment.
I am 20 trying to get the --
21 (Slide.)
22 MR. RUBIN :
This slide shows the SNPPS design.
23 There is also a -- one of the features of the -- the t
(
)
24 diff erence of the automatic or manual isolation of these l
25 automatic control versus manual control of the isolation ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
84 1
valves to the RHE system.
2 In U.S. designs, SNPPS plant, these are 3
automatically closed when the reactor system pressure 4
exceeds a certain set point.
That is to prevent an 5
interfacing LOCA over pressurization of the RHE system.
6 French design, the RHE is inside containment.
7 A LOCA would not have the same consequences.
8 The French felt they wanted to have manual 9
control of the operation so in long-term cooling they would 10 have greater assurance that an automatic closure wouldn't 11 f ail the system closed and preclude operation of the RHE 12 system.
C'
(
13 MR. EB ERSOLE:
What is the pressure rating of i
14 the pipes relatively speaking?
15 MR. RUBIN:
I think similar to U.S. designs, 16 probably somewhere around 600 psi.
17 MR. EBERSOL E:
Okay.
l 18 (Slide.)
19 MR. RUBIN:
Getting down to the bottom of the l
20 list, there are a couple of pumps in the French Paluel 21 plant that are self-cooled.
This is to minimize system l
22 dependencies, minimize dependent on the component cooling 23 water system or --
24 MR. EBERSOL E:
They ride on the atmospheric l
25 system?
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
85 1
MR. RUBIN:
Either cooled by fluid themselves 2
or --
1 3
MR. EBERSOLE:
I see.
4 MR. RUBIN :
Or they have a radiator actually 5
on the motor that is cooled by a fan that is attached to 6
the pump shaft.
7 Let me tell you which pumps these are.
For 8
the charging pumps in the SNPPS plants they depend on 9
service water component cooling water for the bearing oil 10 coolers.
11 In th'e French plants the ch'arging pumps are 12 self-cooled, they have an air radiator attached to the pump 13 and the fan circulates air, the fan is attached to the 14 shaft.
15 MR. EBERSOLE:
That is fine as long as there 16 is open atmosphere but what if it is in a room?
17 MR. RUBIN:
It's in a large room.
i 18 DR. MICHELSON:
How do they get physical 19 protection of equipment if they put it in large rooms which 1
20 infers lots of other pipes?
'21 MR. RUBIN :
No, they are in a separate room.
22 Separate operation.
They are not in the same room.
23 Physically separate.
I ()
24 DR. MICHELSON:
Nothing else in the room, just i
25 a big room with those pumps in it, nothing else?
i ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
86 1
MR. RUBIN:
There is other piping but not the es 2
redundant pump in that room.
3 DR. MICHELSON:
Understood.
But there are 4
other -- okay.
5 MR. LE FAVE:
Mainly those things associated 6
with that pump and train.
7 DR. MICHELSON:
So they are claiming there is 8
enough steady state heat sink there to operate f or several 9
days?
10 MR. RUBIN:
Yes.
In the SNPPS plant, Paluel 11 plant as well, both turbine and motor driven aux feedwater 12 pumps are self-cooled, cooled by the pump flew witness.
13 In the SNPPS design the turbine driven aux 14 feedwater pump is self-cooled, but motor driven pumps 15 require essential service water cooling.
16 I will talk briefly about DC electric power 17 supplies.
Primary differences are in the number of 18 batteries that each design has.
j 19 SNPPS has four DC power supplies, five of l
20 which are nonsafety and four which are safety-related power 21 supplies.
Telephone they are divided into two trains that 22 provide power to vital busses.
23 In the Paluel design they have 17 DC batteries 24 power supplies and they provide dedicated purposes for 25 dedicated functions.
12 of the 17 batteries are l
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
87 1
safety-related and five are not.
gL) 2 Of these 12 safety-related they are divided 3
into two trains and there is diversity of the train.
One 4
train of batteries, are lead acid batteries and the second 5
train of batteries are nickel cadmium.
So there is some 6
reduction in potential common mode failures of batteries, 7
different trains.
8 MR. EBERSOLE:
Is this in the ultimate reduced 9
to two safety busses?
No matter how many batteries?
10 MR. RUBIN:
These batteries, there are two 11 redundant trains.
12 MR. EBERSOLE:
I don't know what a train is 13 ever until you tell me what the elements of it are.
14 MR. RUBIN:
The battery and charger, 15 converter.
There are chargers for each battery.
16 MR. EBERSOL E:
The hydraulic systems, you 17 could say it's two trains, but it's got four pumps if it's 18 got two pipes.
19 MR. RUBIN:
Two trains, each has six 20 batteries.
The reactor protection system for example is 21 powered by four batteries.
22 DR. MICHELSON:
I thought you said there were 23 several batteries dedicated just to particular functions, 24 not enter tied to common bus.
25 MR. RUBIN:
Right.
That's right.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
88 1
DR. MICHELSON:
So there is a number of
-~
_)
2 dedicated batteries which is another way of going?
3 MR. RUBIN:
Right, they are divided into two 4
divisions.
5 DR. MICHELSON:
But they are dedicated within 6
themselves to particular f unctions.
7 MR. RUBIN :
Right.
Four batteries provide 8
power f or the reactor protection system f or example.
9 DR. MICHELSON:
Nothing else.
10 MR. RUBIN:
Nothing else.
To provide power 11 for switch gear eg'uipment, two.
Nothing else.
One of the 12 nonsafety batteries, a battery particularly there for 13 station blackout events.
14 It provides power for a certain period of time 15 following a blackout.
Then that battery, if it should 16 deplete, the steam driven generator provides DC power and 17 charges the battery.
18 DR. MICHELSON:
In our plants there is less of l
j 19 the dedication concept.
They are using --
more of our DC 20 comes off of busses.
21 MR. RUBIN:
Right.
We didn't feel that was 22 the particular safety difference, the number of batteries.
23 DR. MICHELSON:
Just a difference.
()
24 MR. RUBIN :
Difference was in terms of safety i
25 there were different types of batteries.
I ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
89 1
DR. MICHELSON:
Yes.
j-g U
2 MR. RUBIN:
One train or another and the 3
station blackout battery.
4 MR. EBERSOLE:
What I think I heard you say is 5
the DC power system is about the same as ours.
i 6
MR. RUBIN:
Yes.
7 MR. EB ERSOLE :
Two trains.
1 8
DR. MICHELSON:
I didn' t hear him say that at 9
all.
10 MR. RUBIN:
The number, there is safety 11 significance.
~2 12 MR. EB ERSOLE:
If I fail switch gear I am i
13 finished.
14 MR. RUBIN:
The numbers of batteries was 15 not -- there certainly was a difference in the design.
16 MR. EBERSOLE:
If you lock up switch gear 17 because you fail two batteries on the switch gear nothing 18 is going to work.
19 MR. RUBIN:
Other things will work but you are 20 right.
21 MR. EB ERSOLE:
Not that driven by switch gear.
i 22 MR. SPEIS:
If you make that assumption, yes.
23 MR. RUBIN:
Right.
l [)
24 MR. EBERSOLE:
That is like we are.
l 25 MR. RUBIN :
Right.
It's not an additional ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
90 1
redundancy.
2 MR. EBERSOLE:
Yes.
3 (Slide.)
4 MR. RUBIN:
Paluel plant has a -- two more 5
slides here I think.
Gas turbine generator.
This is the 6
last design feature that I will discuss.
7 This generator is here on wheels.
It's 8
mobile, portable.
It happens to be located at the Paluel 9
site but it can be transported to other sites, any site 10 within a two-day period.
It can be connected within three 11 days.
I 12 It takes several hours to connect this gas 13 turbine generator.
It's a backup in the event of hospital w
14 of the off-site power and diesel generators.
15 MR. EB ERSOLE :
What is the rating?
16 MR. RUBIN:
It's a large gas turbine.
At 17 least the capability of one of the diesel generators.
18 MR. EBERSOLE:
Oh, okay.
19 MR. RUBIN :
There are two diesels for each 20 unit at Paluel.
Some U.S.
plants have gas turbines on-site 21 or nearby such as Indian Point.
22 MR. EBERSOLE:
They use two diesel generators.
23 MR. RUBIN:
Yes.
Let me show you
()
24 schematically.
25 (Slide.)
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
91 1
1 MR. RUBIN:
Two diesel generators for each 2
unit providing power to the saf ety busses.
If off-site 3
power is lost.
And if diesel generators fail the backup is 4
connection through gas, this cabinet right here.
5 That is part of the French H 3 procedures for 6
station blackout.
7 The reason that they don't require a longer, 8
shorter time for hooking up this gas turbine is that they 9
feel they have three-day capability for coping with the 10 blackout.
11 MR. EBERSOLE:
Can the units help each other 12 at these stations?
13 MR. RUBIN:
Yes.
14 MR. SPEIS:
Yes.
15 MR. RUBIN:
Let me go into the priority, 16
. priority of recovering AC power.
(
17 First off-site power fails.
The plants have a 18 design capability.
This gets into the next slide.
Let me i
l 19 do this.
There is enough right here on the diesel gas I
20 turbine.
21 (Slide.)
22 MR. RUBIN :
By the way, safety significance we 23 considered of the gas turbine to be moderate.
There are
()
24 other features of coping with the blackout.
If a plant 25 couldn't cope with a blackout for a long time and they had ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
~
92 1
two diesel generators -f or example and they connect the gas 2
turbine in a short period of time that could be that much
?
3 higher safety benefit.
4 MR. EBERSOLE:
I don't know what you are going 5
to tell us about the interchange between units, but do they 6
adopt a design philosophy --
t 7
MR. RUBIN:
Yes, you can intertie from diesel j
8 generator on one unit through this cabinet.
i 9
MR. EB ERSOLE :
Okay.
)
10 MR. RUBIN:
To another bus.
That is their 11-last priority.
12 First they go for the gas turbine, but they do 13 have this interconnection capability.
14 MR. EBERSOLE:
4160 or thereabouts?
15 MR. RUBIN :
I think so.
16 DR. SIESS:
How long does it take.
17 MR. RUBIN:
About three hours to go out in the 18 yard and make these connections.
They also can i
19 interconnect power through the switchyard f rom one unit to l
20 another with the plant operating at house load.
They have 21 the capability --
l 22 MR. EBERSOLE:
Can the plant operate at house 23 load.
()
24 MR. RUBIN :
Yes, their design capability 25 for --
t 1
?
4 ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
93 1
MR. EB ERSOLE :
That is interesting.
That is 2
rather important.-
3 DR. MICHELSON:
Yes.
i 4
MR. EB ERSOLE:
I guess it takes all the house 5
load f or stable, stability.
6 (Slide.)
7 MR. RUBIN:
If you lose off-site power the 8
plant is designed to drop a hundred percent of its load
)
9 without tripping.
10 MR. EB ERSOLE:
Is that a fact?
11 MR. RUBIN:
They have large bypass capability 12 and large condensor size.
13 DR. SIESS:
I thought I read 85 somewhere.
14 MR. RUBIN :
Certain percentage of the load is 15 providing the house-loads.
I 16 MR. EBERSOL E:
You mean it runs down to house 17 load.
18 MR. RUBIN:
Yes.
19 MR. SPEIS:
We are talking about D 4 by the
{
20 way.
21 MR. RUBIN :
Yes.
22 MR. EB ERSOLE:
That is unusual.
23 MR. SPEIS:
D 4 design.
I don't know about
()
24 the the 900.
25 MR. EBERSOLE:
It's a major difference.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
94 l
1 DR. SIESS:
That is where I read it.
l 2
MR. SPEIS:
Yes.
e 3
MR. RUBIN:
It's a difference from a safety 4
standpoint.
There have been some experience where the 5
Swedish plants also have this capability.
6 They had a large blackout in Sweden a couple 7
years ago that knocked out power to part of the country.
8 Although each of the nuclear plants had this
- - 3 9
capability to shed load and not trip, it was not 10 successful.
11 MR. EBERSOL E:
Is that so?
12 MR. RUBIN:
They' tested and during stable l
13 conditions it seemed to work okay, but I think the French 14 assume-that they would be 50 percent successful in shedding 15 load if off-site power were lost.
16 MR. EB ERSOLE :
It's still kind of nervous.
17 MR. RUBIN:
Yes, it's an additional i
18 capability.
What they put it in f or primarily is load 19 following.
l l
20 MR. EBERSOLE:
Don't they test it to see what t-
~
21 it will do?
22 MR. RUBIN:
They will it.
23 MR. EB ERSOLE :
Eventually it will lose itself, 24 wouldn't it, has it yet?
25 MR. RUBIN:
They have had experience with the ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
95 1
900 megawatt units which have been about 50 percent 3
J 2
successful.
3 MR. EB ERSOLE:
I understand it's instability 4
of steam flow due to the valving design.
5 MR. RUBIN:
Some inexternal problem.
I am not 6
sure what exactly was the reason, why they can't get it to 7
be more reliable.
8 Just in background we have talked a lot about 9
station blackout and what the French design capability is.
10 The French are particularly interested in that 11 area.
They have experience of losses of off-site power to 12 their reactors about once in every two reactor years as G
13 compared to the U.S.
experience, that is about five times 14 the frequency for U.S.
experience which is about once every 15 11.
16 DR. SIESS:
Per reactor?
17 MR. RUBIN:
Site years, excuse me.
18 MR. EBERSOLE:
Once every two years.
19 MR. RUBIN:
Right.
Back in the late '70s they 20 had a large blackout that affected a large part of the 21 country.
I think --
22 MR. EB ERSOLE:
I can understand they are 23 nervous about it.
()
24 DR. SIESS:
Is the grid nationwide?
25 MR. RUBIN:
I don't know how the grid is ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
96 1
divided up.
m 2
In case of off-site power lost there is load 3
rejection capability.
There is an interconnection to 4
provide power f rom one unit to another through the 5
switchyard with one unit operating at house load make Mike 6
is that automatic?
7 MR. RUBIN:
No.
8 MR. EBERSOL E:
You don't have to go around on 9
busses, do you.
10 MR. RUBIN:
No.
11 MR. EB ERSOLE :
Manual remote?
Must be.
12 MR. RUBIN:
Or they have to do something in 13 the switchyard.
14 I know for connecting up the diesel generators 15 or gas turbines they have to actually go out there in the 16 yard in these cabinets.
17 MR. EB ERSOLE:
They would rather see an 18 airplane engine concept.
They don't like the interties.
19 MR. RUBIN:
French do not like the interties 20 for the diesel generators between units.
They are 21 concerned about common mode fail common mode failures.
22 That is the has resort.
23 To reiterate, they have the steam driven
()
24 turbine generator for reactor coolant pump seal cooling, 25 instrumentation and control power as well.
Two steam ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
97 4
1 driven aux feedwater pumps, station blackout battery and g-2 battery systems and resupply of condensate storage tank 3
independent of AC power.
They have thought about that 4
fa.~ily of events quite extensively.
5 That will conclude my presentation.
6 DR. WARD:
Okay.
7 DR. MICHELSON:
General question.
8 DR. WARD:
Yes.
9 DR. MICHELSON:
You look at all the ways in 10 which the Paluel might have things that would be 11 improvements, prioritize the improvement.
12 MR. RUBIN :
Right.
'3 DR. MICHELSON:
Did you look to what extent 1
14 the U.S.
reactors might have things the French didn't have 15 and at least wonder why?
16 MR. RUBIN:
We did not look at that.
17 DR. MICHELSON:
Did you pick up any of that in 18 the process to share with us?
19 MR. RUBIN:
Not to my knowledge, no.
We didn' t specifically look for it.
Some features, there were 20 21 advantages and disadvantages.
l 22 DR. MICHELSON:
You didn't actually --
23 MR. RUBIN:
For example, automatic versus
()
24 manual switching has certain plusses and minuses.
I 25 DR. MICHELSON:
Nothing jumped out at you that ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
98 1
we have and wonder why they don't?
-s l
2 MR. RUBIN:
No.
3 DR. WARD:
Charlie.
4 DR. SIESS:
How about water level in the 5
reactor vessel, do they have that.
6 MR. RUBIN :
Measuring water level in the 7
reactor vessel?
8 DR. SIESS:
Yes, post-accident.
9 MR. RUBIN:
I am not aware that they do or 10 don't.
11 DR. SIESS:
Two -- monitors?
12 MR. LE FAVE:
If it was the same, it may have 13 been there, if it was the same --
14 MR. RUBIN:
We basically reviewed the French 15 documents and where we saw a difference between U.S. design 16 we focussed in on it.
If it wasn't, we didn't do the 17 reverse.
I 18 DR. WARD:
Charlie, you had a question.
19 MR. WYLIE:
I just wanted to ask, I noticed in i
20 the report that on the hundred percent load factor 21 capability you classified that as moderate to low as far as 22 importance is concerned.
4 23 Was that based on the assumption that it was
(
)
24 only going to be 50 percent effective in completing that?
25 MR. RUBIN :
Partially, yes, and the fact that ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
99 1
they had other - capabilities as well, even if that f ailed.
7-2 The French consider that not so much for 3
safety.
They even said themselves for safety that it's not i
4 that significant.
They consider it primarily for load 5
following, operational benefits.
6 MR. WYLIE:
It is quite a sizeable --
7 MR. RUBIN:
Oh, yes, it is.
There is a 8
difference in design.
An economic trade-off where you are 9
going to design --
10 MR. WYLIE:
75 percent bypass to condensors, 11 that is lot more than they have got in this country.
12 MR. RUBIN:
I think there may be a couple 13 plants here that can --
14 DR. SIESS:
Some old ones can.
15 MR. RUBIN:
Trip and shut load, too, from a 16 hundred percent.
17 MR. EB ERSOL E:
How do they regard turbine 18 explosion?
19 MR. RUBIN:
20 MR. EB ERSOLE:
Yes.
21 MR. RUBIN:
I think pretty similarly.
They 22 have very similar -- we reviewed as well some of their 23 criteria for design and it was very similar to U.S.
24 criteria.
25 MR. EB ERSOL E:
We have two kinds.
We have got ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
100 1
the parallel kind where it throws the missiles into the 2
critical apparatus and the other where it throws it at 90 3
degrees.
4 I think they had the first kind.
It looked 5
like they were all lined up end to end.
6 Was I wrong?
7 MR. RUBIN:
I think they are --
8 MR. LE FAVE:
Lined up perpendicular like we 9
would want them in ours.
10 DR. WARD:
Okay, good.
Anything else?
11 MR. SPEIS:
Yes.
12 DR. WARD:
Yes?
O 13 MR. SPEIS:
I would like to say a few things I 14 should have said at the beginning.
The French were 15 extremely cooperative in this eff ort.
They give us 16 everything we could think of.
17 In many instances they had in-house reports 18 that were in French and they made them available to us, 19 made drawings available to us.
There were really very 20 cooperative.
21 One of the primary obj ectives undertaking this 22 type of work is to see what type of technical solutions 23 these people pursue when they are faced with problems 24 because we have the same, you know, some of the issues in 25 fact we pursued, there were differences.
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
101 j
k_-
1 But our motivation was because we had some 2
concerns in these areas.
We want to know how-they go about 3
resolving the problems.
From that standpoint it's an 4
important consideration.
5 The other thing we have done, we have taken 6
this report and sent it to the owners group, sent it to 7
EPRI, for example.
8 You know, EPRI is working on future generation 9
of LWR's, brought it to their attention and told them to 10 look at it.
11 Maybe there are some things here that possibly 12 could be incorporated in f uture LWR's to reduce risk.
13 Scrutinize those areas much more, than we have been we have 14 been able to do.
15 We know ourselves we cannot arbitrarily cause 16 anything these days, we have backfit rules and all kinds of 17 other constraints.
18 But we did make this available to all the 19 organizations, all the people that possibly could be of use 20 to them.
21 We think this has been a very successful 22 effort.
We would like to continue this type of activity.
23 We have the comparison between us and the
()
24 Germans, since Chernobyl maybe we should make sure we 25 understand the Russian designs better, not what we learn or ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
102 1
don't learn.
2 But I think we should have all the information 3
we can, he was specially their WR's, LWR's, since one of 4
them or two of them will be built in Cuba.
5 I don't know how you people feel about it.
6 You did send a letter awhile back saying to pursue.
If you 7
still f eel the same way, maybe you could tell us more 8
formally because it is the age of diminishing resources.
9 Even putting one man-year in this type of eff ort, we would 10 have to think very carefully whether we should do that or 11 someplace else.
12 Again, it wasn't a detailed design review.
13 You had, if you have general questions, you had general 14 questions at some other level we didn't pursue.
15 MR. EB ERSOLE :
You made a statement, you said 16 we can't do anything without appropriate backfit rules.
17 There are procedures --
18 MR. SPEIS:
Well we have safety goals and 19 backfit--
20 MR. EB ERSOLE:
I don't mean that.
We still 21 have orders.
I mean that -- pardon me.
/
22 When you say that, we have a problem, you are 23 really violating our interpretation of the regulations when 5
' ()
24 you say ambiguity.
i 25 MR. SPEIS:
For example, you are familiar with ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
103 1
our proposed resolution to the station blackout.
We 2
thought that, well, using all the constraints and rules and 3
regulations which we have to follow, we came up with a 4
proposal the plant should be able to cope with station 5
blackout four to eight hours.
Well, these people have done 6
more.
I don't think it would be justified in asking for 12 7
hours or 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
8 MR. EBERSOLE:
Implicit in your statement was 9
if we ' find some drastic terrible thing wrong with the 10 reactor --
11 MR. SPEIS:
I didn't mean it that way.
12 DR. WARD:
I don't think he meant that.
Is 13 that all?
14 DR. MICHELSON:
Did you look at all the fire 15 protection and compare it with our practices?
16 MR. SPEIS:
Yes.
17 MR. LE FAVE:
Right.
They pretty much use the 18 same rules we do as far as separation barriers.
19 A significant diff erence was in the control 20 room, whereas one channel of all the cables, one channel 21 control were all in conduit and the instruments all have 22 metal covers over them.
It gives them a lot more time to 23 switch over.
()
24 They don't need to have redundant fusing or 25 worry about replacing fusing.
That one channel will be ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
104 1
protected in a control fire for at least an hour before 2
they have to transfer it.
3 DR. MICH ELSON:
All the way up to the 4
instrument?
5 MR. LE FAVE:
Right.
6 DR. MICHELSON:
How do they fight fire in the 7
control room?
8 MR. LE FAVE:
No difference.
Manual.
9 DR. WARD:
You didn't regard that as 10 significant?
11 MR. LE FAVE:
Apparently it was not regarded 12 as significant based on the fact of what is the probability 13 of a control room fire of that magnitude.
14 You have to look at what the risk of a control 1
15 room fire of that magnitude.
Pretty low probability.
I i
l 16 wasn't involved in making that determination.
l l
17 MR. EB ERSOL E:
You also said they had a way to l
18 shut down --
19 MR. LE FAVE:
Right.
They have redundant 20 shutdown panels.
They go all the way to cold shutdown 21 right at the panel, one operator can do it right there.
t 22 MR. EB ERSOLE :
He has everything to go with it i
23 so you can burn out the control room and live.
r( )
24 MR. LE FAVE:
Right.
25 DR. MICHELSON:
Maybe.
l l
{
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
105 1
MR. SPEIS:
Mr. Chairman, Mr. Denton wanted to 7~
V 2
be here himself today.
He was called to the White House, 3
the science advisor is going to explain some of the events 4
of the future as a result of Chernobyl.
5 He feels very strong about this program but 6
again the competition for resources.
He would like, 7
himself, to come and make a presentation to the full 8
committee.
9 DR. WARD:
Thank you.
Appreciate it.
10 MR. SPEIS:
He has a copy, here is a copy of 11 the letter.
12 DR. WARD:
We thank you, very much.
- Alan, 13 thank you.
14 Let's break for lunch and come back at 2:15.
15 We will be in closed session then.
l 16 (Whereupon, the open session was adjourned at 17 1:15 p.m.)
18 19 20 21 22 23
(
25 1
ACE FEDERAL REPORTERS, INC.
Washington, D.C.
(202) 347-3700
'i e
AY.
l
'i\\
!O j
NRR STAFF PRESENTATION TO THE ACRS 1,
l i
i
SUBJECT:
COMPARI' SON OF PALUEL TO U.S. PWR DESIGNS I
DATE:
SEPTEMBER 25, 1985 0
t PRESENTER:
- 1. THEMIS P. SPEIS
- 2. ALAN M. RusIN PRESENTER'S TITLE / BRANCH /DIV:
- 1. DIRECTOR, DIVISION OF SAFETY REVIEW & OVERSIGHT
- 2. PROJECT MANAGER, REACTOR SAFETY ISSUES BRANCH, DSR0 PRESENTER'S NRC TEL. NO.: 1. 49-27517 2, 49-28303 SUBCOMMITTEE: DECAY HEAT REMOVAL SYSTEMS O
',p,.,..
l ()
l l
COMPARIS0N OF PAUEL TO U.S. PWR DESIGNS PRESENTATION TO THE i
ACRS SUBCOMMITTEE ON DECAY HEAT REMOVAL SYSTEMS C:)
BY THEMIS P. SPEIS AND ALAN M. RUBIN DIVISION OF SAFETY REVIEW AND OVERSIGHT 0FFICE OF NUCLEAR REACTOR REGULATION i
SEPTEMBER 25, 1986 i
O 1
l l
l
t O
INTRODUCTION PURPOSE REVIEW DESIGN DIFFERENCES BETWEEN PALUEL AND TYPfCAL U.S. FOUR-LOOP PWR (SNUPPS)
ASSESS SAFETY SIGNIFICANCE BACKGROUND l
SIZEWELL B DESIGN COMPARIS0N (NUREG-0999, MAY 1983)
ACRS AND STAFF INTEREST IN APPROACHES TO REACTOR
()
SAFETY IN FOREIGN COUNTRIES (LETTER FROM ACRS TO EDO, JAN. 1984)
SCOPE OF REVIEW DESIGN FEATURES (HARDWARE)
APPROACH TO SAFETY (SAFETY G0ALS) 1 4
CE) e
s O
FRENCH NUCLEAR POWER PLANTS l
e STANDARD DESIGNS 900 MWE PWR - 30 UNITS IN OPERATION 1300 MWE PWR (P4 DESIGN)
- 6 UNITS IN OPERATION
- 14 UNDER CONSTRUCTION 1400 MWE PWR (N4 DESIGN)
CONSTRUCTION PERMITS FOR 2 UNITS ABOUT 2/3 0F ELECTRICITY IN FRANCE FROM NUCLEAR POWER I ()
I l
.. - - -.. - ~ - -,,,,..
O CONCLUSIONS o
25 DESIGN DIFFERENCES IDENTIFIED RELATIVE INCREMENTAL SAFETY SIGNIFICANCE - HIGH 2
MODERATE TO HIGH 2
MODERATE 7
LOW TO MODERATE 3
LOW 7
NO DIFFERENCE 4
o AREAS OF IMPROVEMENT O
SYSTEM DEPENDENCIES ELECTRICAL SYSTEMS HUMAN ERRORS o
PALUEL OFFERS SUBSTANTIAL IMPROVEMENTS IN SAFETY COMPARED TO SNUPPS FOR A NUMBER OF POTENTIALLY DOMINANT ACCIDENT SEQUENCES l
0 STANDARDIZATION l
i 1
O 1
O U.S. SAFETY G0ALS e
QUALITATIVE SAFETY G0ALS
- INDIVIDUAL BEARS NO SIGNIFICANT ADDITIONAL RISK TO LIFE AND HEALTH
- SOCIETAL RISKS TO LIFE AND HEALTH FROM NUCLEAR POWER LESS THAN COMPETING TECHNOLOGIES FROM ELECTRICITY GENERATION AND NOT A SIGNIFICANT ADDITIONAL RISK e
QUANTITATIVE OBJECTIVES
- RISK TO AVERAGE PERSON IN VICINITY OF PLANT LESS THAN 0.1% OF ALL OTHER PROMPT FATALITY RISKS FROM
([)
ACCIDENTS
- RISK TO POPULATION IN AREA 0F PLANT OF CANCER FATALITIES LESS THAN 0.1% OF ALL OTHER CANCER FATALITY RISKS e
REGULATORY OBJECTIVE
- REASONABLE ASSURANCE THAT A SEVERE CORE DAMAGE ACCIDENT WILL NOT OCCUR AT A U.S. NUCLEAR POWER PLANT e
GENERAL PERFORMANCE GUIDELINE (PROSED FOR FURTHER EXAMINATION)
- MEAN FREQUENCY OF LARGE RELEASE OF RADI0 ACTIVITY LESS THAN 10-6 PER REACTOR-YEAR O
O FRENCH SAFETY OBJECTIVES (NC FRENCH " SAFETY G0AL" DOCUMENT PUBLISHED) e PLANT PERFORMANCE OBJECTIVES
- PROBABILITY OF " UNACCEPTABLE CONSEQUENCES" LESS THAN 10-6 PER REACTOR-YEAR
- PROBABILITY OF " UNACCEPTABLE CONSEQUENCES" DUE TO FAMILY OF EVENTS LESS THAN 10-7 PER REACTOR-YEAR e
CONTAINMENT PERFORMANCE OBJECTIVE
- CONTAINMENT PROVIDES ULTIMATE DEFENSE TO REDUCE
([)
RADI0 ACTIVE RELEASES TO A LEVEL COMPATIBLE WITH FEASIBLE OFFSITE EMERGENCY PLANS e
" UNACCEPTABLE CONSEQUENCES"
- FOR LOSS OF REDUNDANT SAFETY SYSTEMS (E.G., INTERNAL EVENTS) MEANS CORE MELT
- FOR EXTERNAL EVENTS ASSOCIATED WITH HUMAN ACTIVITY (E.G., AIRPLANE CRASHES) MEANS RADI0 ACTIVE RELEASE AT SITE B0UNDARY THAT WOULD EXCEED THE RELEASE FROM A DESIGN BASIS ACCIDENT
()
I
1 FRENCH EMERGENCY OPERATING PROCEDURES e
H PROCEDURES Hy - TOTAL LOSS OF HEAT SINK H2 - TOTAL LOSS OF FEEDWATER sH3 - TOTAL LOSS OF AC POWER (STATION BLACK 0UT)
Hg - LONG TERM TOTAL FAILURE OF SAFETY INJECTION OR CONTAINMENT SPRAY e
U PROCEDURES O
U1 - PREVENT OR DELAY CORE DEGRADATION l
U2 - DETECT CONTAINMENT LEAKAGE 4
V3 - MOBILE UNITS FOR CORE INJECTION AND CONTAINMENT SPRAY (Uq - ELIMINATE A PATHWAY FOR FISSION PRODUCT RELEASE)
U5 - FILTERED VENTED CONTAINMENT O
4 s,-_. - -- - -, _
-,,.,,---,-_-___.__-._.-.-e.-,
-c...
J PALUEL 4
e LOCATED IN NORTHERN FRANCE ON ENGLISH CHANNEL e
4 IDENTICAL UNITS (P4 DESIGN) 1300 MWE (3817 MWT) EACH e
CONSTRUCTION
- BEGAN - 1976 1984
- FIRST UNIT CONNECTED TO GRID
- FOURTH UNIT CONNECTED TO GRID -
1986 j
(REFERENCE PLANT - SOUTH TEXAS) l l
- o l
l O
l
i DESIGN DIFFERENCES 4
e HIGH SAFETY SIGNIFICANCE
- SMALL STEAM-DRIVEN ELECTRICAL GENERATOR FOR RCP SEAL COOLING AND SELECTED INSTRUMENTATION AND CONTROLS
- INTERCONNECTION BETWEEN LOW PRESSURE INJECTION SYSTEM AND CONTAINMENT SPRAY SYSTEM, AND MOBILE EQUIPMENT FOR LONG-TERM COOLING FOLLOWING A LOCA e
MODERATE TO HIGH SAFETY SIGNIFICANCE
- TWO ADDITIONAL DEDICATED RHR PUMPS WITH HEAT EXCHANGER, AND AUTOMATIC SWITCHOVER FROM INJECTION TO RECIRCULATION FOLLOWING A LOCA O
- USE OF MICROPROCESSORS IN REACTOR PROTECTION SYSTEM e
MODERATE SAFETY SIGNIFICANCE
- CONTROL ROOM FEATURES
- FOUR AFW PUMPS t
- RESUPPLY OF CONDENSATE STORAGE TANK INDEPENDENT OF AC POWER
- RHR SYSTEM INSIDE CONTAINMENT, AND REMOTE MANUAL INSTEAD OF AUTOMATIC CLOSURE OF ISOLATION VALVES
- SELF-COOLED SAFETY-RELATED PUMPS
- DC ELECTRIC POWER SUPPLIES Q
- GAS TURBINE TO BACK UP DIESEL GENERATORS I
l
l O
n COMPRESSED AIR TANK MAIN STEAM LINE
[
~
m
, CST
\\
/
g
[
TURBINE-DRIVEN AFW PUMP O
ATMOS HERE TURBINE GENERATOR e
INSTRUMENTATION AND CONTROLS N
CONTROL ROOM LIGHTING if TO ATMOSPHERE RCSh RCS RWST j REACTOR I
MOTOR DRIVEN COOLANT PUMP TEST PUMP O
FRENCH BACKUP TURBINE GENERATOR FOR STATION BLACK 0UT
,A, REFUELING AT NTO RCS COLD LEGS TAriX A
J R
8o-f '
4 TO RCS HOT LEGS SAFETY INJECTION RHR h PUMPS u
9 M
R Oc Q 4
> TO RCS COLD LEGS AND CHARGIN PUMPS f
O RHR Pmo
='
e:
JL TO SAFETY INJECTION PUMP INLETS
'RCS HOT LEG F ' \\
TO RCS
.Ii9i, COLD LEGS CONTAINMENT I Ig
- HOT I
L RHR PUMPS
[
3 r i 1 i 1.
J W
[
LEGS
'RCS HOT LEG jg TO CENTRIFUGAL CHARGING PUMP INLETS O
SNUPPS EMERGENCY CORE COOLING SYSTEM i
REFUELING WATER STORAGE TANK f
INTERMEDIATE PRESSURE INJECTION PUMP 11619 poll
- TO RCS COLD LEGS
Z Q
- TO RCS COLD LEGS
- TO RCS HOT LEGS O
, r TO FITTINGS FOR MOBILE UN?T d
L L
'~
- TO CONTAINMENT f
CONTAINMENT SPRAY SPRAY PUMP CCW CONTAINMENT SUMP 1r TO TRAIN 8 (identical to A)
O entuet eneacency coas cootine systen
e
<a:S"
_ M W_
CONTAINMENT LPSI PUMP INJECTION o
4c
,;g;,,
0 g
~
"~'
j
///
/y/
l
/
~
SPR AY PUM P
. x =...
9/
LPSI PUMP
- INJECTION O
>Tc 0 3
"~'
/////
- /7 g[
exCa'o.
.P.,g_
/
MO;1te
/
~
.x = o.._3 f
1
/
f d
~~
U MOBILE PUMP 3 O O
PALUEL ECC SYSTEMS FOR POST-LOCA LONG-TERM COOLING
O O
O l
Q i
EU rGh w i
, gTOR. TK.
O O TANKS 4.J 124
",*L.
. k.1
., E.
'd l
v E"lD 'cv CO E i
1 AUX FEED PUMP MOTOR ORIVEN I
o z
t_.,
b 2
LCV T
S.
L PIECE NO. TOR. TK.
E RCW 'A'
/
e ERCW"S*
V E
LCi hUABINE DR N
^ 1
< Ac q_
i n
3 LCV
\\
ND. S OR. TANK 1
i ERCW'A'
/j I' AUX FEED PUMP V
LINE MOTOR DRIVEN i
N (h, Q
- i o
.a W-1 n..
4 LCV T
PRESSURE I
(INTERLOCK
'(
I m i,= reed S"Es"POilt E
4 v
)
1 l
\\
AUXILIARY FEEDWATER SYSTEM COwTAi=MEuT 1
i
O O
O VACUUM TANK LOCATED PRESSURE BREAKER INDOORS RELIEF CONTAINMENT
+
MINIMUM RECIRCULATION (TYP.I OUTSiDE INSIDE CST N -- 7 N.O. (TYP.i
/
STEAM f
i F.O.
PENERATORS I
/>
p W
-D<:
NgQ MOTOa.
MM4+
/
XX "u'E"~
I u
MAIN FEEDWATER)
+
7 V
TRAIN A
/
/
M h
h
~
TURBINE-
/
DRIVEN
/
2 PUMP,
p MAIN FEEDWATER)
(TYP )
+
L-k-< Sic 2
/
/
/
m ] M c?"ottni
/
^"
r I
ITYP.)
/
pg4 I
/
p N
O-W O
-D<:
N'gl d MmOa
/
- "due"
$Q---F~j-f e
3 TRAIN B Main FHOWATER) s u
ev,P,"s,
@a-A-3
!r i h
>o-ot:
9-s-F+
o TURBINE-
/
- pjy,
/
4
/
MAIN FEEDWATER)
+
Pil AFW SYSTEM
o bNIT1 l UNb2 I UNIT 3 I UNIT 4
)
1 I
I I
I l S) S l S S l S S l S S l 1
I I
I I
l Y
Y l
Y Y
l Y
Y l
Y Y
I l
l 2LHA 2LHB l
, gHA, gH B l 4LHBl ILHA 1LHB 4LHA i
i 0
0 0
0 0
0 O
I I
I 9
9 9
9 9
9 9
i i
"S"aN' I
I O
F--------
H I
I l
ggggig l
c, c,
l l
9 CABINET l
l GAS l
eE5" "3ya^'
l i
l C
I 2
I A
OO A
OO I
I I
t_______________________a O
BACKUP CONNECTION FOR MOBILE GAS TURBINE
O FRENCH DESIGN FEATURES FOR TOTAL LOSS OF AC POWER o
100% LOAD REJECTION CAPABILITY o
MOBILE GAS TURBINE o
INTERCONNECTION FOR EMERGENCY DIESEL GENERATORS o
STEAM-DRIVEN TURBINE GENERATOR FOR RCP SEAL COOLING AND INSTRUMENTATION AND CONTROL o
DC POWER SUPPLY 0
RESUPPLY CST INDEPENDENT OF AC POWER O
O V
m-_---,,-,,-,-_nrn-,.--
CERTIFICATE OF OFFICIAL REPORTER OG This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the matter of:
NAME OF PROCEEDING:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON WESTINGHOUSE REACTOR PLANTS DOCKET NO.:
PLACE:
WASIIINGTON, D.
C.
O DATE:
TilURSDAY, SEPTEMBER 25, 1986 were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission.
(sigt) 8/l(/'
(TYPED)
[
CRAIG L.
MOWLES Official Reporter ACE-FEDERAL REPORTERS, INC.
Reporter's Affiliation O