ML20136C313
| ML20136C313 | |
| Person / Time | |
|---|---|
| Issue date: | 11/14/1985 |
| From: | NRC COMMISSION (OCM) |
| To: | |
| References | |
| REF-10CFR9.7, REF-GTECI-A-44, REF-GTECI-EL, TASK-A-44, TASK-OR AB38-1-111, NUDOCS 8511210071 | |
| Download: ML20136C313 (113) | |
Text
_
ORIGINAL UNITED STATES OF AMERICA
(~ NUCLEAR REGULATORY COMMISSION In the matter of:
COMMISSION MEETING Continuation of 9/11 Discussion of Proposed Station Blackout Rule (Public Meeting)
Docket No.
(
Location: Washington, D. C.
Date: Thursday, November 14, 1985 Pages: 1 - 94 0511210071 851114 PDR 10CRR PT9.7 pyg ANN RILEY & ASSOCIATES I
Court Reporters 1625 I St., N.W.
\ Suite 921 Washington, D.C. 20006 (20;' 293-3950
- 1 D 1 SC LA I MER 2
3 1
4 5
6 This is an unofficial transcript of a meeting of the 7 United States Nuclear Regulatory Ccemission held on Tilursday, 3 lbvent)er 14, 1985-in the Commission's office at 1717 H Street, l 9 N.W., Washington, D.C. The meeting was open to public 10 attendance and observation. This transcript has not been 11 reviewed, corrected, or edited, and it may contain 1
g 12 inaccuracles.
13 The transcript is intended solely for general 14 informational purposes. As provided by 10 CFR 9.103, it is 15 not part of the formal or informal record of decision of the g
16 matters discussed. Expressions of epinion in this transcEipt 17 do not necessarily reflect final determination or beliefs. No 18 pleading or other paper may be filed with the Commission in 19 ar.y proceed i ng as the result of or addressed to any statement 20 or argument contained herein, except as the Concission may 21 authorize.
l 22 l l
23 24 25
1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 4
, 3 - - -
4 CONTINUATION OF 9/11 DISCUSSION OF
$ PROPOSED STATION BLACKOUT RULE 6 - - -
7 PUBLIC MEETING g - - .
9 Nuclear Regulatory Commission 10 Room 1130 J
11 1717 "H" Street, N.W.
12 Washington, D.C.
13 14 Thursday, November 14, 1985 15 10 The Commission met in open session, pursuant to 17 . notice, at 2:20 o ' c l o e.k p.m., NUNZIO J. PALLADINO, Chairman of 18 the Commission, presiding.
19 COMMISSIONERS PRESENT:
20 NUNZIO J. PALLADINO, Chairman of the Commission
, 21 THOMAS M. ROBERTS, Member of the Commission 22 JAMES X. ASSELSTINE, Member of the Commission 23 FREDERICK M. BFRNTHAL, Member of the Commission 24 LANDO W. ZECH, JR., Member of the Commission 25
1_
e .
1 2
_. 1 STAFF AND PRESENTERS SEATED AT COMMISSION TABLE:
. 2 S. CHILK 3 H. FLAINE t
i 4 H. DENTON i
! 5 T. SPEIS I 6 A. EUBIN
? P. BARANOWSKY !
4 8
i i 10 i
- 11 i
12 i
13 i 14 e
Y I 15 i i 16 4
L 17 I
i 19 l
-l 2o .
j 21 L
r 22
- a 4
i' 23 i
i 24 t
2$ +
i
-_a_ _ _ ___ _ . . _ _ _ _ _ . _ . _ _ . _ _ _ . _ _ _ . _ _ . . _ _ . _ _ _ _ _ _ . _ _ _ _ _ _ _ _ . _ _ . _ . _ . _ . _ _ - . . _.___._________._._._______...__._._m._ ___.___ _ _ _.___ _____ _ _ _ _ . _ _ __ _ _ _ __m._._ _ _. _ _ _ _ . _ . . _ _ _ _ . _
3 !
1 P RO C E E D I NG S 2 CHAIRMAN PALLADINO: Good attornoon ladies and 3 gentlemen. Commissioner Roberts has been delayed and his 4 ottice suggested we start without him. He will be joining us 5 shortly.
6 Today we meet with members of the statt to discuss 7 the proposed station blackout rule SECY-85-103. We last met 8 with the statt to discuss this subject on September 11, 1985.
9 At that meeting we decided to schedule an additional meeting 10 to hear from statt and from the Nuclear Utility Group on 11 Station Blackout known as NUGSBO and also to address additional 12 information from the statt.
13 (Commissioner Roberts enters the meeting.)
14 CHAIRMAN PALLADINO: After the September 11 meeting 15 NUGSBO advised they no longer wanted to meet with the 16 Commission. However, last week NUMARC expressed an interest 17 in this subject and requested the Commission hold in abeyance 18 the station blackout decision until NUMARC is able to provide 19 information on or about February of next year.
20 Regarding the Commission *s request for additional 21 information the EDO responded by memo dated November 1, 1985.
22 By way of background existing regulations established 23 requirements for the design and testing of onsite and olisite 24 electric power systems.
25 However, as operating experience has accumulated a
. _ ~ _. . .- --
4 1- concern has arisen that the reliability of both the offsite 2 and onsite emergency AC power systems may be less than 4
3 originally anticipated.
4 Therefore, several years ago a decision was made to 5 initiate a stati study as part of unresolved safety issue A-44 6 station blackout.
? P r e s e r. t NRC regulations do not explicitly require 8 that nuclear power plants be designed to withstand the loss of 9 all AC power at least for prescribed periods of time.
10 However, the statt studies indicate that station 11 blackout can be a significant contributor to overall plant 12 risk. Therefore, the statt is proposing that the NRC 13 regulations be amended to require that plants be capable of 14 withstanding a total loss of AC power for a specified duration 15 and to maintain adequate core cooling during that period.
16 At the end of today*s meeting I intend to poll the 17 other Commissioners to determine whether or not they are ready 18 to vote on SECY-85-163 in the near future and 11 Commissioners 19 teel they are not ready to vote what more we need to do.
20 Do any of my tellow Commissioners have any additional 21 remarks at this time?
22 COMMISSIONER ASSELSTINE: No.
23 COMMISSIONER ZECH: No.
24 CHAIRMAN PALLADINO: Then I propose to turn the 25 meeting over to Mr. Denton.
1 MR. DENTON: Thank you, Mr. Chairman. I think our 2 last meeting on this topic followed a Congressional hearing so 3 I hope we are not establishing a pattern here.
4 CHAIRMAN PALLADINO: I should point out to the 5 audience that the reason we had to delay the meeting was 6 because we had a prolonged hearing this morning.
7 MR. DENTON: At the last time we met on this, it was 8 our conclusion that station blackout could be in some plants 9 the dominant contributor to public risk and that sL bs t ant i al 10 reductions in core melt frequency could be achieved by this 11 proposed rule.
12 (SLIDE.)
13 MR. DENTON: Since that time a number of interesting 14 events have transpired. What I would like to do is to have 15 Themis Speis go over with you what has happened since our last 16 meeting and the work that has gone on since that time and then 17 have Alan Rubin and P. Baranowsky go into some more detail on 18 those. -
19 Our presentation is about 25 minutes long as planned 20 and without further ado, let me turn it over to Themis to 21 cover the topios of interest since our last meeting.
22 MR. SPEIS: Thank you. The next viewgraph, please.
23 (SLIDE.)
24 MR. SPEIS: We would like to present to you some 25 topios of interest since the September lith Commission meeting
, a 0
1 on station blackout. Since we last met there were two losses 2 of offsite power. One was caused by Hurricane Gloria and the 3 other one was due to a plant-centered loss.
4 COMMISSIONER ASSELSTINE: Which were the two plants?
5 MR. SPEIS: Millstone and Palo Verde and Rubin will
. 6 discuss more details about it.
7 COMMISSIONER ASSELSTINE: Great.
8 MR. SPEIS: Also, we will summarize for you some more 9 information regarding the French PWR designs for station 10 blackout, more specifically their ability to cope with station 11 blackout also their capability to reduce the probability of l
12 station blackout.
13 Back in June when we completed the statt report, 14 NUREG-1032 entitled, " Evaluation of Station Blackout, Accidents 15 in Nuclear Power Plants," we put out for public comments, the 16 period for comments closed October 15th and we have received 17 some substantial comments from the NUGSBO group, from 18 individual utilities and from vendors and we will summarise l
4 19 them for you today, 20 Also, as the Chairman has said we have provided our 21 responses to your questions on station blackout and we will be 22 happy to further amplify on them if you have any additional 23 questions on our responses to your questions.
24 May I have the next viewgraph, please?
25 (SLIDE.)
1.
7 1 MR. SPEIS: Again, the topics that we will be 2 covering today in some more detail is the recent events and we 3 will updste the information that we have regarding trends in 4 emergency diesel generator reliability. We will summarize for
- 5 you the station blackout experience compared to our analysis 6 because our analysis documented in NUREG-1032 takes into 7 consideration improvements.
8 So there are some differences of it one were to 9 estimate blackout events from NUREG-1032 versus the actual 10 number that have occurred up to now.
11 As I said, we will talk to you about the French 12 experience. We will summarise the industry and ACKS comments 13 and also we will give you some additional information and 14 perspectives in station blackout regarding the industry 15 comments.
16 Our conclusions ano recommendations from the last 17 Meeting basically as Harold said, they are unchanged.
l 18 So without any further delays, Al Rubin from our 19 statt will go into some more detail on these subjects.
20 MR. RUBIN: It we could have the next slide, please?
21 (SLIDE.)
22 MR. ROBIN: Since we met on September lith there were l 23 two events on losses of olisite power which are presented on 24 this slide.
25 These are examples of the types of events that have i
8 1 occurred in the past and we see them continue to occur. It is 2 not of particular interest to go into too much detail of these 3 because there have been numerous, some 50-odd losses of 4 offsite power, that we have looked at.
5 Of these two, one was weather related, the event at 6 Millstone in which salt spray from Hurricane Gloria caused 7 arcing in the switchyard. Power was lost for an extended 8 period of time. The plant had shutdown in advance of the 9 oncoming hurricane.
10 CHAIRMAN PALLADINO: Did the diesels operate?
11 MR. RUBIN: The diesels did operate, yes. In both 12 of these instances the diesels started and loaded successfully.
13 There was a similar incident at Millstone back in 14 1976 from a hurricane as well that also caused a long loss of 15 offsite power.
16 The Palo Verde event in early October, there was a 17 loss of offsite power from what we call plant-centered loss in 18 which a single failure here was a multiplexer that sent 19 signals to open two circuit breakers that caused the loss of 20 offsite power.
21 This was a new design. It has been corrected from 22- what I understand. During some testing when the plant was 23 shut down there was a loss about a week later within that same 24 week at Palo Verde but when we l o c 't at statistics that was not 25 something we count in our evaluation of the statistics on loss
9 1 of olisite power.
2 Similarly in October there was a loss of olisite 3 power at Lacross. That plant is unique in that it has only one .
l 4 olisite power line coming into it. It is a small plant and we 5 don't tend to count in our evaluation that plant because it is 6 so different from.the rest of the industry.
7 COMMISSIONER ASSELSTINE: On the Millstone 8 experience, do you intend to convey by the quoted language on 9 viewgraph three that they thought they had fixed this problem i
10 after the 1976 hurricane?
11 MR. RUBIN: This was some information in an EPHI 12 report. They knew they had problems because they did in 13 anticipation of some of the salt spray shut down the plant and 14 they took some very prudent procedures in restoring power.
15 Indicated here, actual loss or time to restore was shout 20 16 hours. They might have been able to restore power sooner but 17 they were cleaning the salt att the switchyard, ott the 18 insulators, checking the breakers before they actually went in 19 and restored power.
20 The diesels were operating successfully during that 21 event.
22 MR. DENTON: We put these up just to show that this 23 is not a rare event. We often talk about quite rare events 24 and these really aren't.
25 COMMISSIONER ASSELSTINE: You said there are 50-some
. __ _ . _ . . - - . _ _ __ _ m _ - - _ _ . _ - . ._ .-~ . _ . _ _ . __ .- _ _ _ _ - _ _ .
t 8' 10 1 of them over the time period?
2 MR. RUBIN: Yes, about 50 approximately.
3 COMMISSIONER ASSELSTINE: Over which time period?
4 MR. ROBIN: Twenty or 25 years.s a
$ COMMISSIONER ASSELSTINE: All right.
6 MR. ROBIN: About once every ten site years roughly.
7 CHAIRMAN PALLADINO: What?
8 MR. RUBIN: Once every ten site years approximately 9 there has been an experience of loss of olisite power. We 10 will get into that a little more in some later slide.
11 (SLIDE.)
! 12 MR. RUBIN: The next slide, the tourth o n e ,- again we i
13 are just talking about it because of this Hurricane Gloria did t
14 occur since our last meeting. There were a number of plan *s 15 that were in the path of the hurricane, some 11-plants up the 16 east coast.
. 17 This hurricane was a very tast moving one. It went l >
18 through very quickly. The only plant that lost all olisite '
19 power was at Millstone. Shoreham had a partial loss et 4
20 olisite power.
21 The winds at these two plants were on the order of 22 maximum of about 100 miles per hour, 23 CHAIRMAN PALLADINO: But in these cases, did the 24 diesels come on?
25 MR. RUBIN: In these cases, yes, the diesels both at
11 1 1 Millstone --
well, Millstone has a diesel and a gas turbine 2 and they both started.
3 CHAIRMAN PALLADINO: How about Shoreham?
4 MR. RUBIN: Shoreham didn*t need diesels. They still 5 had olisite power.
6 CHAIRMAN PALLADINO: I see.
? MR. ROBIN: They were running at low power.
8 CHAIRMAN PALLADINO: All right.
9 MR. DENTON: This was an interesting event for people 10 in the response center because the hurricane was moving so tast 11 that neither the plant people or the agency could respond to it 12 and it had the potential it it had veered inland perhaps 13 causing a wider spread loss of otisite power along the east 14 coast for a number of plants. That is why we put it up 15 because of that kind of an event.
16 It plants had not elected to do a shutdown, it could 17 have caused the loss of olisite power.
18 MR. ROBIN: It was a very strong event, the 19 hurricane, and the fact that it went through very quickly was 20 an advantage because it didn't stay in one area for a long i 21 period of time.
22 CHAIRMAN PALLADINO: Did Salem lose power?
23 MR. ROBIN: No.
24 MR. DENTON: The hurricane never came there.
25 MM. RUBIN: It didn't come nearby to those plants.
. e 52 1 Again, it was along the coast and didn't come directly over 2 those plants.
3 (SLIDE.)
4 MR. RUBIN: The next slide, number five, shows just 5 an update of recent experience. We have presented to the 6 Commission the history of losses of olisite power and we have 7 had some comments that there have been some improvements in 8 recent years in loss of offsite power events.
9 So this slide just shows representative events that 10 occurred during 1985. These were the events. There were tive 11 losses of offsite power, some were plant-centered losses, some 12 were weather related.
13 The overall frequency in 1985 was about equal to the 14- industry average for the past 25 years which is about one in 15 overy 11 site years. The events range trom short duration to 16 relatively long duration.
l 17 CHAIRMAN PALLADINO: Did any of these incidents have 18 any ditticulty with diesels starting'f 19 MR. ROBIN: One event at Quad Cities, one unit was 20 at 100-percent power. The second unit was down for refueling 21 and a diesel at unit-2 which was refueling was out for 22 maintenance. But that plant was shut down. So the only 23 diesel that was unavailable for maintenance because the plant 24 was in a refueling mode.
I 25 CHAIRMAN PALLADINO: Was there a diesel available to
_ _ ~. _
13 1 supply?
2 MR. ROBIN: I what they did is they cross-tied power 3 from one unit to the other to restore power.
4 CHAIRMAN PALLADINO: So they were able to have power?
5 MR. RUBIN: Yes, in about 45 minutes.
6 CHAIRMAN PALLADINO: All right.
7 COMMISSIONER ASSELSTINE: So you are saying that the 8 experience has been pretty consistent all the way along that 9 there is not dramatic improvement over the past couple of 10 years?
! 11 MR. RUBIN: We have seen some recent improvements in 12 reductions in grid-centered, plant-centered losses of oitsite i
13 power, not necessarily weather or grid related losses.
14 COMMISSIONER ASSELSTINE: All right.
15 MR. DENTON: I don *t think we and EPRI are too far 16 apart on the oitsite frequency. EPHI seems to do a very solid 17 technical job when they do an analysis. It is somewhat 18 problematical I think as to what the frequency will be in the 19 future.
20 11 we have excess reserve capacity, then you might 21 find grid reliability improvement. But if new plants are not 22 built and the grid gets less reserves, it might get a little 23 less reliable.
24 MR. RUBIN: The other point to make is it is 25 difficult to go by one year-and say, " Gee, this is
. s 14 1 significantly better or worse than what we have had in the past 2 or even a couple of years." You have to look at trends over 3 time.
4 One figure later, I will show some trends, compared 5 to what we are using in evaluations of estimated losses of 6 offsite power compared to the experience.
7 COMMISSIONER ZECH: In all cases the diesels came on, 8 is that right?
9 14R . ROBIN: In all cases when the diesels were called 10 for they came on.
11 COMMISSIONER ZECH: Yes.
I 12 MR '. RUBIN: One wasn't available and couldn't be 13 called for.
14 COMMISSIONER ZECH: Yes, I understand that.
15 MR. RUBIN: For these events.
16 COMMISSIONER ASSELSTINE: For these here.
17 MR.' RUBIN: Right.
18 COMMISSIONER ZECH: Thank you.
19 CHAIRMAN.FALLADINO: Do you have examples of where 20 they didn't come on?
21 MR. RUBIN: The nex' A ide goes into some recent 22 trends showing a comparisJn 43' e en diesel reliability during 23 testing and surveillance and diesel reliability during actual 24 demands.
25 MR. SPEIS: Next slide, please.
I
15 1 (SLIDE.)
2 MR. RUBIN: This is slide number six.
3 There has been some indication of slight improvement 4 in diesel reliability since we started looking at the data 5 which goes back to 1976. The average failure rate is about 6 two in 100 demands and when we look at these tailure rates the 7 statt has not in our analysis of station blackout assessments 8 counted auto-start failures or tailures that occur in ten 9 seconds or less as tailures because there would be sufficient 10 time to start a diesel manually even it it did not start in 11 ten seconds.
12 So 11 an auto-start failure was listed as the failure 13 cause but the diesel could start in a reasonably short period 14 of time afterwards, we would not count that as a failure.
15 The average failure rate during actual losses of 16 olisite power and this is for data through 1983 was about five 17 tailures per demand.
18 MR. DENTON: Not five.
19 MR. RUBIN: Five failures per 100 demands, I am 20 sorry, 0.05 per demand.
21 Part of the resolution which we feel is an important 22 aspect is that industry should maintain reliable diesels and 23 we have recommended some guidance which goes along with the 24 proposed rule of a minimum diesel reliability or a maximum 25 diesel rate of five tailures per 100 demands.
16 ,
1 1 COMMISSIONER BERNTHAL: I guess I don't understand
, 2 how you come up with the frequency of station blackout events 3 ranges between ten to the minus five and ten to the minus 4 three per year.
5 MR. RUBIN: You are on the slide coming up, right?
6 COMMISSIONER BERNTHAL: Oh!
7 MR. RUBIN: Number seven.
8 COMMISSIONER BERNTHAL: So I am, yes. Do you want 9 to say something first and then I will ask my question.
10 MR. RUBIN: All right.
11 CHAIRMAN PALLADINO: Before you do, I think Jim had 12 one on this one.
13 COMMISSIONER ASSELSTINE: I had a question on six.
14 You are saying that the average failure rate per diesel it 15 called upon to operate during a loss of offsite power event is 16 five per 100.
17 MR. RUBIN: That is the data that has been reported 18 recently by Oak Ridge.
19 COMMISSIONER ASSELSTINE: So out of every 100 demands 20 when you actually have a station blackout --
21 MR. RUBIN: When you have a loss of offsite power.
22 COMMISSIONER ASSELSTINE: That's right, loss of 23 olisite power, you would expect the diesels to fail tive of 24 those 100 times.
.25' .MR. RUBIN: That is what the experience has shown
1?
1 recently. There is much less data on actual losses in diesel 2 starts than there are over total surveillance tests.
3 COMMISSIONER ASSELSTINE: All right.
4 MR. DENTON: One problem you recognize is when you 5 have an actual loss of offsite power, you lose a lot more 6 . plant supporting equipment than you do during a test of the 7 diesel in which everything else around the plant is running 8 normally. So that is why we tried to separate them 9 MR. RUBIN: You might have a relay malfunction or l 10 something like that that may not be tested during a normal 11 surveillance test as an example.
12 COMMISSIONER ASSELSTINE: But one in 20 times the 13 diesels wouldn*t work either.
14 MR. RUBIN: When there was a loss of olisite power.
15 COMMISSIONER ASSELSTINE: That*s right.
16 MR. RUBIN: Right.
17 (SLIDE.)
18 MR. RUBIN: This next slide, number seven, is a 19 comparison of the actual experience on station blackouts where 20 all AC power has been lost for some period of time at plants.
21 'There have been four such events in about 850 reactor-years, 22 total reactor-years, of experience.which~is aaout five times 23 ten to the minus three per reactor year.
24 -CHAIRMAN PALLADINO: It says "per site year."
25 MR. ROBIN: I think we corrected that. We have some 1
I l
18 1 additional slides where that has been corrected.
2 MR. SPEIS: Let me pass these around.
3 (Previously referenced document was distributed to 4 the Commissioners and Chairman Palladino.)
$ CHAIRMAN PALLADINO: Some of us don *t get the word.
6 MR. RUBIN: I am sorry.
? COMMISSIONER ASSELSTINE: All of us didn*t get the 8 word.
9 (Laughter.)
10 MR. RUBIN: That was a recent correction.
11 CHAIRMAN PALLADINO: All right. Is everything else 12 the same? Should I go back to this slide?
13 MR. RUBIN: It is basically the same. That was the 14 only change.
15 MR. SPEIS: We have an additional slide in this new 16 package.
17 CHAIRMAN PALLADINO: All right.
18 MR. ROBIN: Use the new package. We will be on the 19 same slide, exactly the same slides.
20 'fou will recall the estimates of station blackout 21 trequency from NUREG-1032 was on the order of about ten to the 22 minus four per year with a range over a couple of orders of 23 magnitude.
24 I think the important points are the conclusions that 25 we show here.
19 1 One, NUREG-1032 estimates that the frequency of i
2 station blackouts in the future wall be better than the past 4
3 experience has shown.
4 CHAIRMAN PALLADINO: Why is that?
5 MR. RUBIN: For the following reasons, one, we 6 recognise that there have been some trends showing improvements 7 in loss of offsite power experience, that we don't expect as 8 many losses as have occurred in the past.
l 9 When problems do occur, we see that there are 10 corrections made and we acknowledge that there have been some 11 improvements made to reduce common cause failures which is of 12 concern.
'3 1 COMMISSIONER ASSELSTINE: Are those only ones where 14 you have identitied the specific common cause failure and that 15 one has been fixed because I take it there is not yet a 16 systematic review of the potential for common cause failures 17 either in design or in operations and maintenance, 18 MR. RUBIN: There is still some residual likelihood 19 of common cause failure or else we would be able to tell you 20 what is going to happen tomorrow and we can't do that.
21 COMMISSIONER ASSELSTINE: Right.
22 COMMISSIONER ZECH: You are working on it though, 23 aren't you?
24 MR. RUBIN: Sure thing.
25 (Laughter.)
20 1 COMMISSIONER ASSELSTINE: Except that the statt had 2 proposed a direction that in fact licensees do a systematic 4
3 review of common cause failures due to design detects and CRGR 4
4 told you that you couldn*t do that, right?
.5 MR. RUBIN: We did propose that and agreed with CRGE 6 in their comments that the licensees are told to have j 7 independent onsite AC power systems. There may be individual 8 plants that have not identified every single common mode 9 failure potential, 10 MR. BARANOWSKY: We also have to be honest and tell 11 you that the ability to identity potential common cause 12 tailures and determine which ones should be fixed and which
. 13 ones shouldn*t is not necessarily the type of thing we can do 14 that well today without maybe over killing in some instances, 15 .So that is the type of thing that will be developed I think in 16 the future as we look more at operating experience and develop 17 these. capabilities through research and so forth.
18 MR. ROBIN: .That is one of the reasons that we have 19 recommended this defense-in-depth approach of ability to cope 20 with the blackout. The French have gone through the same kind 21 of rationale and reasoning, r
, .22 MR. DENTON: I spoke with EPHI the other day and they 23 have proposed a detailed data collection scheme that would 24 really get at route cause failures of diesels and it seems to 1
25 me that kind of thing over time could eventually identity the
28 1 most probable causes and lead to correction, but it requires a 2 sort of rigorous route cause identification and not sort of the 3 essay type identification of cause that we are more used to in 4 the past.
5 I am very impressed by what EPRI is planning on doing 6 and I think that 11 they keep at it, they will improve diesel 7 reliability as they work at it.
8 COMMISSIONER ASSELSTINE: Have the licensees, some, 9 all, most, tow, done a systematic review to identity potential 10 common cause tailures either in design or in the manner in 11 which these systems are maintained and operated?
13 MR. RUBIN: For the diesels?
13 COMMISSIONER ASSELSTINE: Not only for diesels but 14 for control systems, the whole emergency power supply?
15 MR. BARANOWSKY: Some licensees have done a look for 16 common cause tailures as part of the FRA process. You look at 17 procedures and you look at interconnections and that kind of a 18 potential common cause failure but they don't necessarily 19 review similar components for design deficiencies or 20 manufacturing deficiencies that have not showed up as part of 21 say pre-operational testing or in-service testing.
23 COMMISSIONER BERNTHAL: I still have a question on 23 how you came up with your range, estimated range of frequency, 24 because 11 I just on the face of it you told me or told us 25 that the average, not the worst, but the average case is ten
22 1 to the minus one per reactor year so that is ten to the minus 2 one.
3 MR. HUBIN: Loss of offsite power.
4 COMMISSIONER BERNTHAL: Loss of olisite power, right.
5 MR. RUBIN: So you have this olisite power ano then 6 --
7 COMMISSIONER BERNTHAL: Right 8 MR. ROBIN: Okay.
9 COMMISSIONER BERNTHAL: Then you are telling us that 1
4 10 in the worst cases it is ten to the minus one, in fact, 90 11 percent reliability on the diesels in the worst case. The 12 average is 98 percent reliability. That gets me to ten to the 13 minus two and I guess it in every case you had a back-up 14 diesel, how do you get to ten to the minus three because if I 15 lose olisite power and my diesel doesn't work, the worst case
. 16 looks like ten to the minus two. How do you get to ten to the 17 minus three?
18 MR. BARANOWSKY: This is not a worst case analysis.
19 COMMISSIONER BERNTHAL: But you told us the range is 20 ten to the minus three to ten to the minus tive.
21 MR. BARANOWSKY: The range that we think is a 22 realistic range is on the order of ten to the minus five to ten 23 to the minus three.
24 COMMISSIONER BERNTHAL: An average range.
25 MR. BARANOWSKY: All right, average. There are going
23 1 to be times when the performance of equipment is going to be 2 somewhat outside that range, better in some cases and worse in 3 some cases.
4 We don *t normally expect to see two 90-percent 5 reliable diesels at a power plant and s high loss of oitsite 6 power troquency at the same power plant. In fact, we have not 7 observed that, have not.
8 MR. DENTON: There are redundant diesels so in your 9 math you have to take that into consideration.
10 CHAIRMAN PALLADINO: But these numbers are based on 11 these tour events, is that right?
12 MR. BARANOWSKY: No. The tour events, sir, are just 13 events that we are knowledgeable of The ranges are based on 14 reliability models in which we do the kind of thing that 15 Commissioner Bernthal was just talking about. We put in 16 reliability values and we select based'on our judgment what we 17 think is a realistic range and we can debate that, but that is 18 the range we chose,
! 19 COMMISSIONER ASSELSTINE: So your actual worst case 20 does not involve the worst case in terms of olisite system 21 reliability and the worst case --
22 MR. BARANOWSKY: Combinations of all kinds, here.
23 Yes. In fact, this range could go trem ten to the minus seven 24 to ten to the minus two 11 you want to put in the very, very 25 best and the very worst.
- . _ , - - - _ ~
24 1 MR. DENTON: Maybe you would like to show that backup 2 slide on improvements in olisite reliability just to show how 3 you picked some of the values that went into the calculations.
4 MR. RUBIN: I don *t think that slide makes that 5 point.
6 MR. DENTON: All right. Keep going then.
7 MR. RUBIN: Our conclusion then is that even with 8 these estimates, these realistic estimates, we feel are 9 valuations of estimated station blackout frequencies, the 10 station blackout still can be a dominant contributor to core a
11 melt frequency.
12 COMMISSIONER BERNTHAL: Let me make sure I understand 13 here. Now you are saying that the range of best and worst case a
14 in your judgment is four orders of magnitude, ten to the minus 15 two to ten to the minus seven. You are saying that the 16 uncertainty in effect causes you to make an average judgment, 17 ten to the minus three to ten to the minus 11ve. It sounds 18 like that is what you are saying.
4 19 COMMISSIONER ASSELSTINE: No.
20 MR. RUBIN: No.
21 MR. SPEIS: That assumes a combination of the worst 22 type of scenarios which we don't think they exist.
23 COMMISSIONER BERNTHAL: You do not think they do or 24 you know they don't'r 25 MR. RUBIN: In the reports we have seen, for example,
25 1 we have not seen a power plant that has two diesel generators, 2 each one of 90-percent reliability.
3 COMMISSIONER BEENTHAL: But somebody just made a 4 statement that said ten to the minus two to ten to the minus 5 seven.
6 COMMISSIONER ASSELSTINE: That is based on 7 unrealistic assumptions.
8 MR. BARANOWSKY: You will excuse us it those numbers 9 are ott a little bit. We are trying to remember them a little '
10 bit. They might be tive times ten to the minus seven or 1
11 something like that but the point is we didn't take the 12 absolute worst combination of every possibility or the absolute 13 best combination of every possibility to develop the range.
14 -
That would not necessarily be meaningful. It would 15 certainly bracket everything we have seen but it might mislead i
16 you in terms of what the real performance out there has been.
17 COMMISSIONER BERNTHAL: So what you are saying is 18 that any given plant in your judgment should at tall outside 19 that ten to the minus three number, is that what you are 20 saying?
21 MR. BARANOWSKY: As a best estimate, out best 22 estimate has uncertainty and that would catch the upper and 23 lower limits that we were talking about.
24 MR. DENTON: To some extent bad weather and bad sites 25 were reflected in the statt's original review. So sites where 1
a N
_ . . , . ,.-4 - - - - - _ -
26 1 they were thought to be vulnerable did tend to have more onsite 2 power and St. Lucie is a case where that issue was adjudicated 3 speoitically as to whether onsite power was adequate in view of 4 the offsite power losses.
5 So the system did tilt a bit to prevent the kind of 6 situation Pat described, namely, high loss of offsite power 7 and very poor diesels.
8 I think what a rule would require is a systematio i 9 look to find anybody who was up there in the high end and we 10 think some plants are much higher than others. I guess the 11 thrust of our proposal would be to bring them all to ten to the 12 minus tive so that you wouldn*t have any possible ones that 1
13 were high.
, 14 COMMISSIONER ASSELSTINE: Is it fair to say that the 15 uncertainties are such that you really don't know where the to plants tall right now with a high degree of precision until 17 you do the kinn of analysis and review that is called for 18 under the rule?
19 MR. DENTON: I think we otten talk about plants as 20 though they were static photographs. .s e y are actually 21 movies.
22 COMMISSIONER ASSELSTINE: That's true.
23 MR. DENTON: The diesel reliability changes with time 24 and some plants may run a long period of time with very 25 reliable diesels and then go to another mode. None of these I
27 1 think are real static issues. I think that is what lead us to 2 say that some ability to cope while they got diesels restored k
3 was really the way to get at it rather than trying -- I mean 4 everyone agrees you should reduce the likelihood of an AC loss 5 but it would be very hard to keep your diesels at 0.99 reliable 6 all the time.
7 Grand Gulf, for example, remember all the ottort we 8 put in those diesels and apparently just tailed due to an 9 overspeed event over the weekend. I don *t know what the 10 implications are. We will grind through it but diesels 11 occasionally tail.
12 CHAIRMAN PALLADINO: I don *t want to divert the 13 direction of your presentation but do I recall correctly the 14 Japanese had achieved or seemed to achieve a tailure of one in 15 7,000?
16 MR. DENTON: They have a very remarkable --
I -17 CHAIRMAN PALLADINO: Do we have any insight as to why j 18 they achieve such a high reliability?
19 MR. DENTON: Yes. I think it is the same reason that 20 they achieve high reliability in other engineering equipment.
21 They really do a detailed root cause evaluation and we have 22 recently had a presentation by Kensai Electrio on pump seals 23 and they identitled the 51 ways that pump seals could tall and 4
24 calculated through detailed tailure modes and etteot analysis 25 the probability of.each one of these tallures and took detailed
6 I 28 1 steps on roughness of the seals and triction and water 2 chemistry and really work at.
3 So apparently every tailure gets detailed attention 4 and you don't find the same tallures repeating. Somehow our 5 maintenance doesn*t -- and when they inspect a diesel they 6 don't really inspect it. They refurbish it so when it is put 7 back together there is no known problems with it.
8 So that combination of really identitying the root 9 cause plus high preventive maintenance during that three 10 months that they are down seems to assure a very reliable 11 diesel performance.
12 CHAIkMAN PALLADINO: Right now they don *t have many 13 root causes to explore.
14 COMMISSIONER ASSELSTINE: Yes.
15 CHAIRMAN PALLADINO: I do remember they said they had 16 dittiounty at the beginning.
17 COMMISSIONER ASSELSTINE: Do they use large diesels 18 comparable to those in the US plants, the two trains supported 19 by two large diesels?
20 MR. DENTON: Pretty similar design concepts.
21 COMMISSIONER ASSELSTINE: All right, but not like 22 many of the European plants that use the tour trains that can 23 be supported by smaller diesels?
24 MR. DENTON: They look pretty large but I do not know 25 for sure. They seem roughly comparable in size though to the l
I
29 1 type we use.
2 MR. RUBIN: The experience in most of the rest of the 3 foreign countries is kind of comparable to here, about one 4 or two failures per hundred demands on diesels. The Japanese 5 do seem to be doing something right.
6 COMMISSIONER ZECH: Do they have the same kind of a 7 testing program that we have or a similar program, do you 8 know?
9 MR. DENTON: We have looked at that.
10 MR. SPEIS: The Japanese testing program is not that 11 much dissimilar to curs. What is ditterent is the maintenance f
j 12 where at every refueling outage they take the diesel apart and
! 13 they examine it and refurbish it as Harold has said.
B 14 COMMISSIONER ASSELSTINE: Preventive maintenance.
15 MR. SPEIS: Preventive maintenance is the main toous to but they do the tast test that we do and most of the tests are 17 about the same.
18 MR. DENTON: Roughly comparable, but the difference 19 seems to be that they originate a preventive maintenance plan 20 at the time of licensing and that is sort of concurred in by 21 the regulator and they then carry it out wntoh involves l
22 detailed oil checking for debris in the oil and this sort of f
23 thing and they stick to that preventive maintenance program i
24 and they don't have very many unusual tailures.
25 COMMISSIONER ZECH: I know they have a relatively
30 1 good maintenance program. What I was concerned about and 1 2 think we have talked about it before, at least to some of the 3 statt I have mentioned before that my observation has been in 4 testing our diesels we test them pretty severely and just an 5 opinion of mine it seems to me that we load test them, we 6 really give them an awful good going over before the plant is 7 even authorised for start-up.
8 My concern has been and I think the stati has looked 9 into it to see whether that testing we are doing is realistic 10 and necessary and I think you discussed it with others, with 11 experts, who perhaps have been helpful but I hope you have 12 looked into that and perhaps you could comment on it.
13 MR. SPEIS: In fact, we have sent out a generie 14 letter last year eliminating tast starts. I think that is what 15 you referring to.
16 COMMISSIONER ZECH: That is exactly what I am
~
17 referring to.
16 MR. SPEIS: But the Japanese for some reason they are 19 still doing those tast starts because it is part of their 20 design basis LOCA.
21 COMMISSIONEN ASSELSTINE: That is interesting because
'1 22 my perception was the same I think as Commissioner Zech*s. In 23 tact, what I had heard at many of the plants was the testing 24 program is responsible for some of the tailures in the diesel l
25 program. We are testing them so much that they are wearing out I I
31 1 but it is interesting that the Japanese are doing essentially a l l
2 comparable test program even continuing the tast starts and yet i
3 have a much higher reliability.
4 MM. DENTON: Many plants now don *t have a requirement 5 for the tast starts. We came to the same view that the f
) 6 philosophy which seems simple on paper is every time it tails,
? you test it twtoe as atten was in etteot wearing it out and it 8 was far more profitable to go back to the root cause of the j
9 tailure and tix that rather than trying to test reliability and 10 it toally wasn't doing it.
4 11 COMMISSIONER ZECH: I hope we have a program to 12 continue looking not only of the Japanese maintenance but also 1
< 13 the testing program because we put a lot of stock in our 14 diesels. It has been my experience that they are a pret.ty good l
15 engine. They are reliable and all that.
4 16 On the other hand, we are really counting on them to 1
17 do the job for us and I think it behooves us to have an l I
18 especially vigorous and watchtul program for maintenance as 19 well as for testing.
I 20 I presume that is what you are doing. I hope that i
21 is what is going on. I think it is really necessary, 22 MM. SPEIS: In fact one of the bases of our proposal 23 is the reliability at diesels is an integral part of this i
24 proposal and what it involves is the development and j 25 maintaining some reliabilities.
2
32 4
I 1 MR. DENTON: I am encouraged by EPHI*s entrance into 2 this area and their willingness to undertake a real look j 3 at what is causing the tailures and how could they be fixed so ;
i >
4 it sounds like EPRI is going to come in in a scientific manner t
3 and start looking at this.
6 Up to now I think each utility relies to a varying l 7 extent on the vendor and they get varying qualities of servtce.
I I
l 8 COMMISSIONER ZECH: But the maintenance is an 9- important part, too, in my judgment. I have heard that the !
! I I
10 Japanese are doing a very good job in not only diesel engines j
11 but otaer types of maintenance but I think in this particular l 12 area, it is something that I would - egest we look at very 13 carefully.
- 14 COMMISSIONER SERNTHAL: It is a good point and well
! 15 taken, Lando, because the question of testing is something 16 that I think we do pretty well and fairly frequently to the 17 point in fact that some licensees have complained that we may 1
1 19 he testing so often that it may actually be detrimental to i
1 19 safety.
20 The question of maintenance, 11 I were to be so bold i
21 as to make a wild guess, it would be that maybe we are
- 22 overtesting and that maintenance is a bit underdone, j
! 23 CHAIRMAN PALLADINO: Why not go ahead? I think we f
. 24 have covered the points.
J 25 MR. MUNIN: All right.
}
4
d 33 ,
1 (SLIDE.)
2 MR. RUBIN: The next slide, number eight --
3 COMMISSIONER ASSELSTINE: At some point though it 4 would be helpful to understand a bit more what it is in this 5 package that will bring about the kinds of ettorts that would ;
6 result in that higher level of reliability by the diesels, i
1 j 7 perhaps not quite as high as the Japanese, but at least l
8 something approaching that level and is that built in, that in l j 9 order to achieve that kind oi reliability utilities would have !
j 10 to put in place a more ettective and more aggressive preventive 11 maintenance program.
12 I know it is in the reg guide. I guess what I am i
.)
j 13 wondering is, is it tied into this rule such that an assumption 14 about a high level of reliability of the diesel generators i
15 would in fact become reality at all of the plants it the rule 16 were to go torward.
17 MR. SPE!S: The duration of the co' ping capability
]
- 18 that we are talking about is coupled with the reliability of 19 the diesels so every plant would have to come forward and tell 20 us what is their reliability program and it has to be followed. ,
l 21 !! they go above or below those limits, something 22 would have to be done.
23 COMMISSIONER ASSELSTINE: Would that enable --
1 I
24 MR. SPE!S: We are asking for a voluntary program 25 trom them, define the program to make sure that the i
_.-___.,________.___.___________-.-.__u._-_.-.
1 34 1 re11 abilities that they can achieve are consistent with the 2 coping capabilities that we are prescribing in this issue here.
3 COMMISSIONEM ASSELSTINE: All right.
4 MR. DENTON: It doesn't really go directly to 5 improving maintenance which certainly is worthy of attention.
6 It sort of took what we had and more mcVed that you have to be 7 able to cope.
8 MR. SPEIS: We are askiaq tham to come forward with 9 their initiatives, you know, whst can they develop on their 10 own to oake sure that they achieve some levels that can be 11 maintained bastoally.
12 CO.tMI S S I ONEM ASSELSTINE: Would that approach still 13 tolerate a 90-percent reliability for some plants for their 14 diesels?
15 MR. WARANOWSXY: Not for long. There is going to be 16 11uctuatic s in any performance over time but the idea is to 17 have a program which doesn't allow continued unreliable 18 operation, one that calls for either making changes or 19 declaring a component so unreliable that it is not considered 20 a safety grade component any more until fixed properly, 21 something like what was done with the TDI diesels, for 22 instance.
23 Hopefully you don *t get into that situation +11 you 24 have a program that calls for reliability goals, it is a
{ 25 performance oriented type of.oriterion'and it-gives the burden i
35 1 of performance on the backs of the utilities who claim they 2 want the responsibility for making decisions themselves and it 3 is aimed in that direction.
4 COMMISSIONER ASSELSTINE: D you have a cutoit line?
5 You said the average is 98 percent and some plants do better 6 than that and some do as badly as 90 percent, Do you have a 7 cutoft where you say that this is so unreliable that you are 8 not maintaining equipment that --
9 MR. BARANOWSKY: We are aiming like for 95.
10 Ninety-five is lower level of acceptable continuous type at 11 performance but we would allow performance over short periods 12 of time at less than 95-percent reliability but we would expect 13 that the utility would be taking action and informing the NRC 14 of what they are doing and we would have to agree that that is 15 an appropriate action to fix the situation up.
16 COMMISSIONER ASSELSTINE: Sorry, Joe.
17 CHAIRMAN PALLADINO: When you talk about a diesel not 18 starting, just what do you mean? Do you mean it wouldn't 19 start no matter how hard you tried or wouldn*t start the first 20 time or wouldn't start the third time?
21 MR. BARANOWSKY: Eventually they all start, 22 CHAIRMAN PALLADINO: What time trame are you talking-23 about?
24 MR. BARANOWSKY: Really, we are talking about a 25 diesel that won't start either automatically or by manual
- . -- _ _ - . - _ _ . =- . , ..- . , . -
- t. 6 1 36 l
1 operations from the control room. If you have to dispatch an j 2 operator or maintenance person to make a correction, that is a 3 tallure and we take into consideration the types of repair
^
4 activities that go with ditterent failures and there is a 1
5 distribution on this.
1 6 In general, when we define a tailure it is something 7 that would not allow starting by operator action from the
]
4 8 control room.
,i 9 CHAIRMAN PALLADINO: I see. Now these tour instances 1
10 where they lost both the olisite and onsite power, you didn*t 11 say anything about the duration of the combined loss. Were 1
12 they minutes, hours?
13 MR. RUBIN: That i s the duration of the combined
{ 14 loss.
.j 15 CHAIRMAN PALLADINO: Were they without power five l
16 minutes, 20 minutes?
l 17 COMMISSIONER ASSELSTINE: Maybe you can give us just j 18 a quick run down of each of the tour events.
3 19 MR. ROBIN: There were tour events. One was in 1968, 20' two of them were about live minutes. The one in 1983 was 23 21 minutes, the one in 1984 was about ten minutes. That is the 22 total amount of time without AC power for those tour events, i
23' less than halt an hour, f
24 CHAIRMAN PALLADINO: .Did they get the diesels started 25 or did they get the AC power back?
_ . _ _ _ . _ _ _ . _ _ . _ _ _ . _ _ . _ _ _ -_ _m__ _ _ _ _ _ _ _ _ - _ _ _ _ . _. _ _ _ _ _ - _ - _ _ _ . . - _ ..-_-_.
i 37 1 MR. RUBIN: It depended. Generally, they got the 2 olisite power back.
3 COMMISSIONER ASSELSTINE: You mean Susquehanna?
4 MR. ROBIN: Yes.
5 COMMISSIONER ASSELSTINE: Where they ran power from 6 the other unit back over.
7 MR. RUBIN: Yes, so it was ofisite power. They 8 didn't get the diesels back. They got power from the other 9 unit.
10 COMMISSIONER ASSELSTINE: That*s right.
l 11 CHAIRMAN PALLADINO; All right. Thank you. Do you 12 want to go ahead.
13 MR. RUBIN: All right. In addition to the generic i
14 look at a probabilistic type analysis, we also looked at some 15 of the plant speoitic PRA's that were performed both by 16 industry and for NRC and although it is ditticult to compare i
i 17 PRAs because of the analyst and because of the number of i 18 sequences accounted for, basically the plant-specific PRAs 19 substantiate the findings that we developed from NUREG-1032 20 tht station blackout can be a dominant contributor to total 21 core me.t frequency.
22 Of the FRAs that v'e looked at and we are talking 23 about almost 19 of them, about half had core melt troquenotes 24 from station blackout of greater than ten to the minus liith l
25 per reactor year and about 25 percent were greater than five
38 1 times ten to the minus 111th.
2 The other halt had frequencies that were lower than 3 ten to the minus fifth. Some of these PHAs were done earlier 4 on and al the sequences were not included. For example, 5 battery depletion, reactor coolant pump seal leakage, 6 condensate storage tank availability was not considered 7 consistently throughout the PHAs especially the earlier ones.
8 There might be some modifications it they were looked at 9 again.
r 10 The station blackout contribution to total core melt 11 trequency ranged from very small, one percent, to almost 12 50-percent for some plants, greater than 20-percent 13 contribution from station blackout for about one-quarter at
! 14 the plants.
i 15 This is just to get some perspectives on what we 16 tound from industry PMAs. They don't ditter very significantly i
17 trom our generic analysis.
18 COMMISSIONEM ASSELSTINE: Which plant had an almost
' 19 50-percent contribution from station blackout?
20 MR. MUBIN: I don't recall. It might have had an 21 overall low core melt frequency.
22 COMMISSIONER ASSELSTINE: I understand.
23 MR. HUBIN: It was one plant and I don't recall !
24 could get an answer for you.
25 COMMISSIONER ASSELSTINE: That's okay. You can tell
l s
- 39 1 1 me later. That's fine.
2 MR. ROBIN: All right. I don't have that on my 3 fingertips.
4 CHAIRMAN PALLADINO: When you consider -- mine is 5 really not with oore frequency. I will wait until later.
6 MR. RUBIN: We just went through to corroborate that 7 this doesn't really change our conclusions when you look at 8 the plant-speoitic PRAs.
9 (SLIDE.)
I 10 MM. RUNIN: Onto a different subject trem PRAs, since 11 the last Commission meeting we have some additional information 12 on the French design for station blackout.
13 MR. DENTON: We talked about this during the safety 14 goal meeting, I think, and we have now had a chance to interact 15 with the French just on this topic and have what we think is a 4 16 tairly good understanding of their design for station blackout.
17 MR. RUBIN: The French have a target of achieving a 18 core melt frequency trom station blackout of ten to the minus 19 seventh per reactor year. In trying to achieve that target, 20 they have determined that common mode tallures are potentially 21 significant, that they are taking an approach where 22 defense-in-depth and an ability to cope with the blackout and 23 a design that they have included on their new 1,300 megawatt 24 plants and is also being backtitted on their existing 900 25 megawatt plants gives them a capability to cope with the
40 1 blackout for three days.
2 In that time period they feel that they car restore 3 power it not from an ottsite power source or some other source 4 from a mobile gas turbine that can be moved from one site to l
5 another.
6 CHAIRMAN PALLADINO: They can withstand it for three 7 days?
8 MR. RUBIN: They can withstand a total loss of AC 9 power for three days.
, 10 CHAIRMAN PALLADINO: How do they handle the heating, I
11 ventilating, air conditioning load? I am thinking specifically l
12 ci the incident in which the loss of air conditioning caused i
l 13 the temperature in the control room to raise above 90 degrees 14 and lo and behold it turns out the instruments go awry at 90 1
15 degrees and they began to show anomalies.
16 It sounds like air conditioning is very important for l
17 the control room. How do they handle that?
5 18 MR. RUBIN: It is one of the things we have 19 considered in the resolution package here in looking at the 20 environmental conditions following a station blackout. At the 21 time we didn't have the detailed report on this event you are 22 referring to at Maguire so we didn't specifically address that 23 with the French.
24 CHAIRMAN PALLADINO: Do we address it in our own 25 package?
41 1 MR. RUBIN: We do address it in our own package. The 2 evaluation of station blackout coping capability would include 3 assessing the operability of equipment that would be needed to 4 tunction during a station blackout.
5 CHAIRMAN PALLADINO: Doesn't air conditioning take a 6 tair amount of power?
7 MR. RUBIN: With a station blackout you wouldn't have
- 8 the normal heating, ventilation and air conditioning?
I I
9 CHAIRMAN PALLADINO: What would you have to keep the 10 temperatures from rising to where you lose the instruments?
11 MR. RUBIN: You might be able to open some doors or 12 some cabinets and get some natural circulation going for a 13 period of time. That would be one of the things that would 14 need to be assessed. i
{
15 CHAIRMAN PALLADINO: But you might be bringing in as 16 much heat load unless you have some colder places.
17 MR. DENTON: The heat load, of course, drops with no 18 AC or heat load is the DC power but that is an issue that will a
19 have to be looked at.
i 20 MR. BARANOWSKY: Also, the French design includes a 21 steam-driven generator that provides AC at least for charging i
22 batteries and they may tap a little bit off Just on those 23 instrument cabinets that they need to have a tan running for.
24 CHAIRMAN PALLADINO: I am surprised to learn that 25 instruments go awry at 90. Can't you get instruments that go
42 1 better than 90 degrees?
2 MR. DENTON: I don't know how representative that 3 case is.
4 CHAIRMAN PALLADINO: I don't either.
5 MR. DENTON: It is something that we have to look 6 into.
7 MR. ROBIN: I think in that case the instruments were 8 probably subject to temperatures above 90 but the ambient 9 room temperature might have been 90?
10 COMMISSIONER BERNTHAL: There would be a certain 11 irony 11 after all of the EQ things we have done, we proceeded 12 on the reasonable assumption that as long as human beings could 13 be in the control room at least that they were the most 14 sensitive mechanism there and it turned out that that weren*t 15 true, that it was equipment. instead.
16 CHAIRMAN PALLADINO: All right.
17 MR. RUBIN: Just to go into a little bit what the 18 French have included, they have two diesels per plant which is 19 similar to what we have here and they can also have the 20 capability for 100-percent load rejection without tripping the 21 turbine. If the grid goes down, they can reduce load and 22 maintain the plant on house load.
23 They have this mobile gas turbine that I mentioned 24 earlier.
25 CHAIRMAN PALLADINO: Are plants don't_have this
, - . , - , . r,n. , , -,n. , , . - , - . .
1 43 1 capability?
l 2 MR. RUBIN: Our plants generally don't have 3 100-percent load rejection. It is kind of around the 4 50-percent range normally.
5 CHAIRMAN PALLADINO: Do any of them have 100-percent i
6 capability?
? MR. ROBIN: There might be a few.
8 MR. DENTON: I think Palo Verde is attempting to 9 demonstrate so we may have a few, yes.
10 CHAIRMAN PALLADINO: Thank you.
11 COMMISSIONER ASSELSTINE: I think Diablo claims that 12 they do as webl. They mentioned that when I was out there.
13 MR. RUBIN: That is a benefit but in actual practice 14 at least the French have had about a 50-percent success with 15 maintaining the plant operable following a 100-percent load 16 rejection.
17 They also are able at the sites to interconnect both 18 offsite and onsite power sources from one unit to another.
19 So even with this AC power capability, they still 20 have design features to cope with the blackout for this 21 extended period of time.
22 The new French design has two steam-driven auxillary 23 feedwater-pumps in addition-to two motor-driven pumps. So 24 they have tour trains of aux 111ary feedwater.
25 They have added a. steam-driven generater that
44 1 provides power for a pump to cool the reastor coolant pump 2 seals and also to provide power for instrumentation and control 3 needed during station blackout.
4 COMMISSIONER BERNTHAL: It sounds like they must have 5 really taken this whole business very, very seriously because 6 three days is a large margin for error. If you have it for a 7 day even, I guess you are to the point where you should be 8 natural circulation.
9 MR. RUBIN: What their original objective was kind 10 of on the order of a day, 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />.
11 COMMISSIONER BERNTHAL: Yes.
12 MR. RUBIN: What they found though when they put in 13 the design modifications to achieve that objective, it wasn't 14 very difficult to show that they have the capability for the 15 steam generators to go for longer periods of time.
16 COMMISSIONER ASSELSTINE: The tough part is getting 17 to the 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />.
18 MR. ROBIN: Once they put in the steam-driven 19 generator that provided cooling for the pump seals and provided 20 some DC power for battery charging, that got them a long way 21 and they had a gravity feed for the condensate storage tank to 22 provide water source for a long period of time.
23 COMMISSIONER BERNTHAL: It sounds like their 24 objective is almost to have emergency power until they achieve i l
25 cold shutdown essentially. Is that right?
l
45 1 MR. RUBIN: They are not trying to get to cold 2 shutdown. They are just trying to maintain the plant in a 3 stable condition until they can get AC power back and then get 4 to cold shutdown. They won *t have the RHR systems operating.
5 COMMISSIONER BERNTHAL: I see. All right 6 MR. ROBIN: I mentioned again that they are 7 backfitting this design on the existing 900 megawatt plants.
8 COMMISSIONER ASSELSTINE: Do you have an estimate on 9 what the cost is for backtitting the requirements on the 900 10 megawatt plants?
i 11 MR. RUBIN: From what they told us, it was less than 12 5500,000.00.
13 CHAIRMAN PALLADINO: For each plant?
14 COMMISSIONER ASSELSTINE: Less than 5500,000.00 per 15 unit?
16 MR. RUBIN: Per unit. Now they already have a test 17 pump that they use for charging so basically when they add 18 some piping for a steam turbine and generator that provides 19 power for an existing pump.
20 COMMISSIONER ASSELSTINE: Do you have an estimate for 21 what it would cost to backtit those same requirements on US 22 plants, the ones they are doing for the 900 megawatt plants, 23 not the ones they are doing for the 1,300?
1 1
24 MR. DENTON: Since they have a standard plant, their 25 costs are no doubt cheaper than it would be in the US l
46 1 non-standardized plants.
2 MR. HUBIN: We have estimated a higher cost for doing 3 that here.
4 CHAIRMAN PALLADINO: Yes. It seems to me I remember 5 s600,000.00 on the average with a range of s200,000.00 to tour 6 million.
7 MR. RUBIN: We didn't estimate that backtit would be 8 necessary for achieving full radar coping capability. It we 9 are talking about longer times which at one time we were until 10 we looked at the loss of offsite power data in more detail 11 for a day or so, then we looked at coping capabilities and the 12 need for seal cooling, a separate seal cooling system, and we 13 estimated maybe two to four million for that backtit.
14 COMMISSIONER ASSELSTINE: Which would be comparable 15 to this.
16 MR. RUBIN: Which would provide the same functions 17 for this. But that can vary. It depends 11 the plant had an 18 existing pump or had similar design to the French, it could be 19 a lot less.
20 CHAIRMAN PALLADINO: But you did give a range.
21 kR. RUBIN: We did give a range.
22 CHAIRMAN PALLADINO: It was quite a range. I thought 23 ~ I remembered'an average of 5600,000.00'!
24 MR. RUBIN: That is right. That is for the kinds of 25 things we might see at creating battery capacity it needed,
- I
. o i
47 l
1 for doing station blackout assessment.
2 CHAIRMAN PALLADINO: The actual cost will depend on 3 the specific plant.
4 MR. RUBIN: It will be very plant specific. That is 5 right.
6 COMMISSIONER ASSELSTINE: But the range at least even 7 11 in crude terms for achieving one-day capability, that is 8 cooling capability for the reactor coolant pump seals would 9 be for the US plants somewhere in the neighborhood of two to 10 four million.
11 MR. ROBIN: Probably. That is our estimate.
12 COMMISSIONER ASSELSTINE: All right.
13 COMMISSIONER BERNTHAL: How does the French 14 requirement compare with Germany, Belgium,.Sizewell or am I 15 jumping the gun here?
16 MR. RUBIN: No. I have some information on that.
17 COMMISSIONER BEMNTHAL: Sweden.
18 MR. RUBIN: Most countries do not have specific 19 designs. The British do in Sizewell and we talked about this
) 20 at the last meeting. They have actually two steam-driven 21 charging pumps for seal cooling in their design.
22 COMMISSIONER ASSELSTINE: Right.
23 MR. ROBIN: In Germany, for example, their 1,300 24 megawatt plants, they have four onsite emergency diesel 25 generators plus from two to four additional diesels, the
48 1 bunkered system or separate decay heat removal system.
2 MR. DENTON: It is more like the dual system that we 3 have talked about.
4 MR. ROBIN: So station blackout is not a significant 5 consideration when they have so many diesels and redundant and 6 diverse systems.
7 The Swedish, for example, have either four emergency 8 diesel generators per unit or two diesel generators plus two 9 gas turbines either nearby or onsite. They also have some 10 additional redundancy or diversity that is not typical for 11 most US plants.
12 COMMISSIONER BERNTHAL: What are the coping targets 13 for those various countries?
i 14 MR. RUBIN: I don't think they have considered coping 15 targets but they consider that station blackout is not a very 16 likely event because of their large numbers of AC power 17 sources. I don't whether they have done a PHA type analysis.
18 MR. BARANOWSKY: The Germans consider loss of their 19 emergency AC power system but not the bunkered system so the 20 bunkered. system does their coping for them and it is unlimited.
I 21 COMMISSIONER ASSELSTINE: Yes, that is right, Is it 22 tair to say that all of the instances that you have just 23 described to provide better capability for dealing with loss 24 of offsite power than the average US plant?
25 MR. RUBIN: Some plants have additional-capability.
49 1 It varies. That is why we had such an extent of time looking l
i 2 at the diverse designs.
3 Obviously the more diesels up to a certain point, the 4 more reliability. You have the onsite power systems and you 5 still have the potential for common cause tailures which is one 6 of the reasons again I reiterate that we are looking at a 7 defense-in-depth and an ability to cope as a resolution of 8 this issue.
9 COMMISSIONER ASSELSTINE: How many US plants have 10 less than two diesels per unit?
11 MR. ROBIN: Maybe a multi-unit site you mean with 12 three diesels?
13 COMMISSIONER ASSELSTINE: YEs.
14 MR. RUBIN: About five. If you count plants that 15 have two units with two diesels, they both share, you might be 16 at about eight or nine, something like that.
17 MR. DENTON: It tended to be the early plants that 16 had the swing diesels and as we began to worry more about 19 interconnections, they went independent.
20 MR. RUBIN: About 15-percent, something like that.
21 MR. SPEIS: Commissioner Bernthal, the greatest goal 22 for coping capabilities is about a day.
23 COMMISSIONER BERNTHAL: A day?
24 MR. SPSIS: Yes, That is the goal.
25 COMMISSIONER BERNTHAL: So not so much different.
l
50 1 MR. SPEIS: It is not so much different from the 2 French, yes.
3 COMMISSIONER BENNTHAL: Except it looks like France 4
4 has a larger margin.
5 COMMISSIONER ASSELSTINE: But substantially beyond 6 even what you are proposing in this rule for the US plants.
7 MR. RUBIN: Hight.
8 ( S L I D E .- )
9 MR. RUBIN: It we can continue on to slide ten, we 10 mentioned earlier that we have received recent comments from 11 industry on NUREG-1032. This is a summary of some of those 12 comments, ones that were highlighted.
13 One is that there was some concern expressed that in 14 the 1032 analysis the staff did not give significant 15 consideration for credit for improvements, recent improvements 16 in loss of otisite power experience.
17 I would like '. o show a slide, it is a back-up slide 18 in your package.
19 MR. SPEIS: It is in the new package we gave you.
20 MR. ROBIN: In the new package we recently put 21 together there is an additional slide.
22 (SLIDE.)
l 23 MR. RUBIN: This is taken from a --
24 CHAIRMAN PALLADINO: Which one is this?
25 MR. RUBIN: It is B-2, the last slide in your set.
l
51 l
1 CHAIRMAN PALLADINO: All right.
2 MR. RUBIN: The curve and the circles of data were 3 taken directly from a report that was submitted by the Nuclear 4 Utility Group on Station Blackout and they tried showing 5 trends and recent improvements in average loss of olisite 6 power frequency as the experience from industry increases over 7 time and showing trends of declining frequency of losses of 8 otisite power.
9 The straight lines on this curve show that the 10 industry average is about 0.09, 0.88 shown on this slide, 11 losses of offsite power per site year.
12 The analysis in NUREG-1032 gave credit for what 13 appeared to be trends in reducing frequency of plant centered 14 losses of offsite power. These were relatively the short ones.
15 CHAIRMAN PALLADINO: How do you get those two 16 horizontal lines when the experience showing --
17 MR. RUBIN: The horizontal lines are not meant to 18 indicate trends.
19 CHAIRMAN PALLADINO: I know.
20 MR. RUBIN: The slanted lines show the trends. The 21 horizontal line just shows what is being used. It shouldn't 22 really go across time. It is just there. These are the 23 numbers used in the 1032 analysis.
24 CHAIRMAN PALLADINO: You are saying they are finally 25 beginning to achieve 0.088?
52 1 MR. RUBIN: These are the estimates that we are using 2 to predict future performance of 0.088. In fact, the actual 3 number used for assessing losses of offsite power frequency and 4 station blackout frequency are 0.071 which gives a credit for 5 reduced frequency of actual plant-centered losses.
6 So we are taking into account improvements in trends 7 in estimating future performance for loss off offsite power.
8 COMMISSIONER BERNTHAL: How does it happen that the 9 diesel generator group, the industry group, comes in and makes 10 a statement like this after much work and sweat presumably and 11 I am reading from their summary, "The stati's best estimate 12 approach does not use actual emergency diesel generator data 13 for failure to start and load. The range of EDG reliability 14 levels postulated in NUMEG-1032 is not representative of 15 actual industry experience in either extent or distribution."
16 The only thing I can see in that statement that you
- 17 may not have addressed at the table here with the numbers
, 18 presuming ~your numbers are right is the question of range or 19 distribution if you will 20 How can there be such disagreement on what should be 21 bare numbers?
, 22 MR. ROBIN: Are you going back on the diesel failure 23 rate? I just wanted to clarity.
24 MR. BARANOWSKY: I think it is fairly clear that they 25 either misread what we did or they just didn't bother to
53 1 read it all to be honest with you.
2 We state right in our report and we have talked to 3 people both at EPRI and everywhere else, the IEEE, about our 4 approach to this. We said it before the ACRS and everywhere 5 else that the large LOCA tast start requirement is not the 6 kind of thing that we consider in the diesel generator 7 reliability requirements.
8 Not only that, I think there were some reports cited 9 that gave some false implications of very high diesel 10 reliability. When we have gone out and done the best we can 11 in getting actual demand and tailure data from diesel 12 generators, we have talked to people from other countries, we 13 talked to utilities about actual experience and until we put 14 the stuff down on paper and you see a result that suggests 15 that maybe we have a safety issue here, we had not had much 16 comment about it.
17 The situation is we used the demand tailure rates as 18 we described here and that is we pruned out those auto-start 19 failures that they claim we used.
20 CHAIRMAN PALLADINO: All right.
21 MR. RUBIN: Shall I go back to slice tenf~
22 MR, DENTON: I think that EPRI getting involved, you 23 see up to now we have dealt with owners groups who definitely 24 have a vested interested in how this comes out, EPRI seems to 25 me has a program that will tackle this issue of diesel start
. =
$4 1 reliability in much more detail than the industry or the statt 2 has tackled it up to now and maybe will shed some light on 3 some of the disputes that still exist.
4 COMMISSIONER BERNTHAL: Let me ask a question about 5 the second one. You are about to go on to it, I presume.
6 MR. ROBIN: That is what your request related to,
? yes.
8 COMMISSIONER BERNTHAL: Yes. Let me read again what 9 the industry says here. Their comment is that in fact relevant 10 now to the second point, i.e., the stati does not realistically 11 evaluate reliability and therefore overestimates risk from 12 station blackout but in more detail they say, "The problem 13 is. "
and for example here and I am quoting now, "In fact, the 14 stait*s analysis indicates that only 25-percent of the total 15 core damage potential is represented by station blackout events 16 lasting longer than two hours."
17 It doesn*t strike me that that is a terribly strong 18 argument. That means 2$ plants have core damage potential 19 represented by station blackout events lasting longer than two 20 hours.
21 From that piece of information or complaint they go 22 on then to say, "Indeed the staff *s analysis. . " and this is 23_ their conclusion now, "Indeed the stati*s analysis shows that 24 the capability of coping with station blackout events longer 25 than two hours duration is not important in managing overall w - ~ , - , ,--. - , ,
55 1 station blackout risk, i.e., because only 25-percent of them 2 represent possible core damage events."
3 I have to say that 1 am unconvinced. Would you care 4 to explain what they are saying there?
5 MR. RUBIN: I would rather that they explain it.
6 MR. BARANOWSKY: I don *t think I know what they are 7 saying there either.
8 MR. RUBIN: I am presenting also in this slide 9 comments from industry. We don't necessary see exactly the t
10 same picture.
11 COMMISSIONER BERNTHAL: Well, go ahead.
12 MR. ROBIN: I think you did cover the second point 13 on this slide talking about the diesel reliability and I think 14 the point I made a little earlier on comparing the diesel 15 tailures during actual demands and during surveillance tests 16 is a point that we haven't had comments on but I think it is 17 pertinent to the question of the stait's really looking at 4 18 what is a real number for diesel reliability.
19 The third point is that there were some comments
~
20 looking at the statistical validity of estimating 21 loss-ot-offsite-power frequency and how we grouped the plants 22 in the station blackout assessment. It is true that'we did use 23 statistics and the best methods available with the limited b
24 number of data and-I think any time you look at statistics, you 25 are not going to have 100-percent confidence and we have made
56 1 some engineering judgments and some approaches that we felt 2 were reasonable to try to look at differences between plants 3 and have had some discussions with IEEE in looking at design 4 features we felt were important and apply some statistics to
$ look at that.
6 We may never feel with 100-percent statistical
? confidence that this is the way to parcel small amounts of data 8 but we have taken what we feel is the best approach available 9 and we haven't seen another alternative to that.
10 There are also recommendations from industry that new 11 source terms be used to make a more realistic assessment 12 of the consequences. This is for the value impact analysis.
13 I think the Commission had a meeting on source terms 14 recently and the guidance we are currently using are to use 15 the types of source terms that we have been using in the past 16 until the source term issue gives us another direction.
17 We have accounted in somewhat in the value impact 18 analysis for some modification of the source term because of 19 some additional time to containment failure and that is 20 reflected in the regulatory analysis which is part of your 21 package.
22 The conclusion that industry reaches is that station 23 blackout does not contribute significantly to risk and 24 rulemaking is not justified.
25 CHAIRMAN PALLADINO: How do they justify that i
$7 1 statement?
2 MR. RUBIN: Not to my own way of thinking.
3 CHAIRMAN PALLADINO: What is that?
4 MR. RUBIN: They have not substantiated that 5 statement to me.
6 CHAIRMAN PALLADINO: Then what could we expect from 7 NUMARC it we wait to February?
8 COMMISSIONER ASSELSTINE: More of the samet 9 (Laughter.)
10 CHAIRMAN PALLADINO: Unless they can give us 11 something now that justifies their conclusion, it just makes it 12 hard to think in terms of waiting until February.
13 COMMISSIONER BERNTHAL: I have to say it also makes 14 it hard when you see the rest of the leaders in this business 15 running as a pack in the other direction and here we are alone 16 in the crowd arguing or apparently arguing that this is not 17 important. '
18 The French are going very strongly, three or four 19 times greater coping capability, in the other directions. The 20 Germans clearly have. The Belgians have. The British are.
21 CCMMISSIONER ASSELSTINE: Swedes.
-22 COMMISSIONER BEMNTHAL: The Swedes ~have. It rings a 23 little hollow it seems to me.
24 CHAIRMAN PALLADINO: This is one of my pet subjects 25 because when I first cama on the ACRS I was preaching that
58 1 look, as long as you have power there are a lot of things you 2 can do to salvage a plant that is giving you trouble but 11 3 you ever lose power for any extended period of time, you just 4 don't have any options.
5 I think this is a very important subject. I think 6 that the nations that have recognized this have taken very 7 important actions and I believe it is up to us to do all that 8 we can here.
9 COMMISSIONER ASSELSTINE: Yes.
10 MR. DENTON: We have some drills, you will remember 11 Mr. Chairman, where the loss of oitsite power is postulated 12 and that really does leave no ability to diagnose and interpret 13 and that is what is so serious about this type of event.
14 COMMISSIONER ASSELSTINE: Yes.
15 CHAIRMAN PALLADINO: I remember being quite concerned 16 at Three Mile Island even though I wasn't on the Commission at
, 17 that time when I saw and heard about all of' these helicopters 18 flying around because I was afraid that one of them was going 19 to go into a pylon and then sure enough, we would be in the 20 middle of an accident without any ability to do anything.
21 MR. DENTON: I think there are some other views.
'22 Perhaps you want to go on to the next slide.
23 (SLIDE.)
24 MR. RUBIN: We have had comments.
25 MR. DENTON: We talked to industry for about a year
. ~ . ._ -_. _ . . .~
59 1 and a half on this matter and have not seemed to make a lot of 2 -- we have straightened out some differences, and I guess my 3 own view that EPHI bringing in sort of a scientific approach 4 may in several years shed more light on loss of offsite power 5 and diesel reliability and reliability programs and so forth.
6 But I don't think there is going to be a sudden 7 revelation by February in view of the years that we have been 8 struggling with basically the same set of data, t
! 9 COMMISSIONER BERNTHAL: This is after all only a
, 10 proposed rule.
11 COMMISSIONER ASSELSTINE: That's right. ;
i 12 MR. RUBIN: This is a proposed rule for comment. We 13 have gotten some comments already. We hav had a number of 1
i 14 interactions, meetings, with ACHS subcommittees and meetings 15 over the years on this issue, two letters that we have received 16 from ACES indicating support for the proposed resolution and 17 the second letter, 1985, reaffirmed the ACRS recommendation 18 supporting the stait's approach.
19 COMMISSIONER ZECH: Is it a unanimous opinion from e
20 the ACES or are they divided?
21 MR. RUBIN: The only divided opinion was that 22 possibly we should be doing more.
23 COMMISSIONER ASSELSTINE: Yes.
24 COMMISSIONER ZECH: Other than that, it is pretty 25 straight forward support of your views?
l
00 1 MR. RUBIN: Yes. We have some conclusions and 2 recommendations from the presentation today. Themis, do you 3 want to cover this?
4 (SLIDE.)
5 MR. SPEIS: Basically Harold discussed this at the 6 beginning but what has happened in the last month or so, we 7 haven't seen anything different to change our minds, our 8 conclusions and our recommendations, 9 COMMISSIONER ZECH: Excuse me, may I interpret for 10 Just a second before you go on to your conclusions?
11 MR. SPEIS: Yes.
12 COMMISSIONER ZECH: How about the CRGR? Have you 13 worked with them on this? Do they have any other views? You 14 didn't mention them I don't think unless I missed it.
15 MR. ROBIN: We mention it in the Commission package.
16 COMMISSIONER ZECH: Yes, but I mean you didn't 17 mention here today.
18 MR. ROBIN: We didn't mention here today because it 19 wasn't recent since the last meeting. We can over it.
20 COMMISSIONER ZECH: If you could just do it briefly, 21 I would appreciate it.
22 MR. SPEIS: We went to the CRGE and they recommended 23 that the rule go out.
24 CHAIRMAN PALLADINO: That the rule go out?
25 MR. SPEIS: Yes. They had. difficulty with the
61 1 structure of a rule as they said at that time. They are 2 addressing this. Their views are expressed very clearly in one 3 of the questions posed by Commissioner Asselstine.
4 COMMISSIONER ASSELSTINE: That's right.
$ MR. SPEIS: They thought that we didn't give enough 6 openings to the industry to come forward with additional 7 diesels or additional systems that will further enhance 8 reliability and they thought that our proposal was kind of 9 mono or uni-dimensional because our proposal is to go into the 10 coping capability because of our conclusions that there are 11 limits to the reliability of offsite and onsite systems and 12 therefore to overcome that our proposal is going in the 13 direction of achieving coping capability.
14 The CHGR thought that maybe there is more that can 15 be done in the front end, put more diesels or more systems 16 to further enhance rel. ability and thus avoid requiring coping 17 capabilities.
18 That was their view. But they had no problems with 19 our technical assessment, with the way we used PRAs in 20 comparison with engineering judgments. They say those things.
21 COMMISSIONER ZECH: Would they support the proposed 22 rule putting it out for comment?
23 MR. DENTON: I think they did, Of course, they 24 advised Bill 25 COMM~.SSIONER ZECH: I know they had some reservations
+
I l
62 1 that you have expressed and I have read those, too, but I just 2 want to make sure that there hasn't been any other more recent 3 review from the CHGR.
4 MR. DENTON: No. They have not reconsidered the 5 matter but they did provide the answer to question number two.
6 COMMISSIONER ZECH: Right.
7 MR. DENTON: I don't think we have anyone down here 8 from the CHGR.
9 CHAIRMAN PALLADINO: Could you check it and then 10 maybe send a little note.
11 MR. DENTON: I know that there have been no further 12 meetings and they spoke one time and that is what they said 13 and they haven't met again.
14 MR. SPEIS: Again our approach was a little bit 15 pragmatic because we are very mindful of cost benefit 16 considerations and proposing a system that will comprise tive 17 diesels or ten diesels to further enhance the reliability is 18 not a cost effective approach.
19 So our approach was pragmatic from the viewpoint of 20 cost benefit and from the viewpoint of course of common mode 21 failures because we thought that there were limits to the 22 reliability and you didn't want to put all your money on that 23 side of the equation. You wanted to.do something more in the 24 coping area.
25 Now of course as I said, CMGR thinks that maybe there
-. s 03 1 is more that can be done on the iront end.
2 CHAIRMAN PALLADINO: Could you do the following, 3 check with CRGH and if they have technical reservations on this 4 rule, then could you let us know, the offices know?
5 MR. SPEIS: They are saying the same thing I said 6 more loosely and more erudite in answering question two from 1 ? Mr. Asselstine.
8 CHAIRMAN PALLADINO: I think so.
9 COMMISSIONES ASSELSTINE: Let me take it one step 1,0 farther, Themis. As I recall the CMGR structure, 11 they have 11 problems with the rule then they make a recommendation to 12 Bill, Am I right about that?
13 MR. SPEIS: Yes.
14 COMMISSIONER ASSELSTINE: What is the EDO's position 15 on this?
16 MR. SPEIS: To go out for rulemaking.
17 COMMISSIONER ASSELSTINE: As is?
18 MR. SPEIS: Yes, as is, yes.
19 COMMISSIONER ASSELSTINE: Good.
20 MR. SPEIS: It is in both Commission' papers that have 21 been sent to you.
22 COMMISSIONER ASSELSTINE: So after Bill considered 23 whatever comments CHGR wanted to make to him, his view was go 24 with the stait package.
25 MR. SPEIS: Yes, i
l I
~
G4 j i
1 COMMISSIONER ASSELSTINE: All right.
f 2 CHAIRMAN PALLADINO: But to responsive though to 3 Commissioner Zech, if you find that there are some technical 4 reservation'that is significant it.would be appropriate to let 5 each Commissioner office know.
6 COMMISSIONER ZECH: I am just really reaching for 7 kind of a conclusive summary of their views. I appreciate what 8 they have said but it would help me just to know that they 9 still support the stait position.
i 10 CHAIRMAN PALLADINO: All right.
11 MR. SPEIS: As we have been telling you now for a 12 while the station blackout can be a dominant contributor to 13 total core melt risk and we feel that substantial reductions in 14 risk can be achieved by this proposed rule which will further 15 enhance defense-in-depth and assure public health and safety 16 from this issue' being a small' contributor to r i s k'.
l
, 17 So we are finally recommending that we go out with 18 this proposed rule and if industry or EPHI or others have 19 additional-constructive comments then we should consider them j 20 and possibly make some changes.
f 21 I would like to remind you that-the rule.itself does 22 not call for any specified duration. It just calls for the 23 ability to cope with a station blackout. Only the reg guide 24 itself talks about specific times like four hours and eight 25 hours depending on certain 1 actors like the configuration of t
, -, ,-s - -
65 1 the diesel generators or the reliability or the olisite 2 reliability and the time that might be important in power 3 recovery.
4 So I think we will have some time to maybe refine the
$ reg guide itself it other thoughts or other positive ideas come 6 torward. We can further refine the guide itself. .
? COMMISSIONER AS9ELSTINE: The Chairman mentioned 8 earlier the NUMARC reque.t that we wait.until February. It the 9 Commission were to go ahead and issue the proposed rule is 10 there anything that would prevent NUMARC within that time 11 period providing whatever comments they want to make and then 12 the Commission considering those as part of action on the 13 final rule?
14 MR. SPEIS: I wouldn't think so and you gentlemen 13 have the privilege of inviting them here to present their own 16 views to you.
17 CHAIRMAN PALLADINO: They weren't ready to present 18 any view they said until about February.
19 COMMISSIONER ASSELSTINE: But the comment period 20 would give them a chance to do that.
21 MR. SPEIS: Also, it is kind of unfair because so far 22 we have gone out in a partial way with some technical reports 23 and some segments of the industry have commented but maybe 24 there are others that would be interested in providing their 25 comments.
66 1 So I think it is important to send the whole package 2 out.
3 COMMISSIONER ASSELSTINE: This would give everybody 4 all the information and everybody would have the opportunity 5 to comment.
6 MR. SPEIS: Again, you are not voting for a final 7 rule. You have the chance to near NUGSBO or NUMARC in more 8 detail once they have developed their positions better.
9 CHAIRMAN PALLADINO: Themis, were you through, Jim, 10 on that question?
11 COMMISSIONER ASSELSTINE: Yes.
12 CHAIRMAN PALLADINO: On another question since you 13 were talking about the NOREG, remember slide 32 of the 14 September lith meeting where you gave the tour hours, eight 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> and indicated there would be discrete solutions to lack 16 of AC reliability.
17 It a licensee were to propose to interpolate between 18 these two discrete solutions, would the statt accommodate such 19 a request or do you see a problem with that?
20 MR. SPEIS: The answer is yes and we say so in the 21 reg guide. Isn't that right?
22 MR. ROBIN: Yes, that is correct.
23 CHAIRMAN PALLADINO: It came out in the presentation 24 it seemed to be either/or.
25 MR. SPEIS: No. I think the words itself in the reg
l
. i 07 1 guido say so. l 2 CHAIRMAN PALLADINO: All right. I have to confess I 3 haven *t completely absorbed the reg guide.
4 MR. DENTON: 1 think the four and the eight were just 5 to put the issue in context. What you would really like to 6 have is enough time to get AC power back and there are a lot at ,
7 arguments that perhaps it takes a shift. That would be B a time to assemble enough people and you could cope with it.
, 9 The longer you can cope, the more likely it is you will get i
10 the AC restored.
11 We thought those kinds of numbers for those 12 classifications of plants did give you around the ten to the 13 minus five assurance that it would not really be a dominant 14 contributor for future operation.
15 COMMISSIONER ASSELSTINE: So if you have a hurricane 16 it might take a day to get-AC power back.
17 MR. DENTON: Many plants can cope. I don *t want to 18 picture this as a universal problem and some plants might be 19 able to show considerable coping capability;with their present 20 designs but others may not.
I 21 MR. SPEIS: I think the other thing that drove us to 22 differentiate between four and eight was we are mindful, we 23 -don *t want to arbitrary regulators, there are some plants that 24 are better than others and actually maybe this should be a 25 spectrum.
- m. . _. ._~ __- __ _ _
. i 68 1 CHAIRMAN PALLADINO: That is what i was getting at.
2 MR. SPEIS: There are varieties and variations and 3 we understand that.
4' CHAIRMAN PALLADINO: All right. I think that answers 5 that question. Do you have any more?
6 COMMISSIONER BERNTHAL: Yes, I had a couple.
7 CHAIRMAN PALLADINO: No, I meant on their 8 presentation. I have a series of questions as well.
9 MR. DENTON: This completes our planned presentation.
10 CHAIRMAN PALLADINO: I will take the privilege of the 11 Chair and go ahead. I have a couple of technical questions and 12 then I have some procedural questions with regard to the 13 backtit rule.
14 Do the modifications that result from USI A-44 have 15 to be safety grade and do you address that question?
16 MR. RUBIN: We addressed the question and there is 17 not a requirement that they be safety grade in terms of being 18 qualified for seismic conditions.
19 CHAIRMAN PALLADINO: What is the premise? I could i
20 see the seismic situation being the cause of both the loss of
( 21 offsite power and perhaps a diesel problem.
22 MR. BARANOWSKY: If the seismic event was big enough 23 to knock out a seismically. qualified diesel, then it is going 24 to knock out a seismically qualified backfit.
l 25 CHAIRMAN PALLADINO: Not necessarily. I am thinking
. i 69 1 of the electrical end where you get shorts because other 2 things happen not because the diesel fails.
3 First of all, the switchyard is not necessarily and 4 I don't think it is --
5 MR. BARANOWSKY: No. What I am talking about is the 6 diesel generator which is seismically qualified for its
? function which is provide electrical power and all aspects of 8 the diesel and its subsystems and auxillary systems should be 9 seismically qualified.
10 CHAIRMAN PALLADINO: So you are saying that there is 11 no interaction from a seismically unqualified component that i
12 could short out the effect of the diesel?
13 MR. BARANOWSKY: If any additional hardware were 14 added that were not seismically qualified, they would have to 15 meet the same requirements that any other non-qualified 16 equipment which is in the plant has to meet. In other words, 17 the interfaces would have to be identified and there would 18 have to be an evaluation that no undesirable interactions 19 would cccur from a seismic event.
20 CHAIRMAN PALLADINO: So you would analyze that?
21 MR. BARANOWSKY: Yes. That would be a requirement, 22 yes.
23 M R .' DENTON: I think the general idea was that we
, 24 would not compound this loss of offsite power with another 25 design basis event so that you don't have to assume that loss
. t 70 1 of coolant accident occurs simultaneously with loss of AC 2 power.
3 CHAIRMAN PALLADINO: Yes, I was inclined to say the 4 same thing except that seismic events can cause problems.
' 5 MR. DENTON: That is an interesting point.
6 CHAIRMAN PALLADINO: It at least you are looking at 7 it as far as interactions are concerned so that you face up to 8 the situation.
9 MR. DENTON: I don't think we had focused -- we are 10 still trying to get over the hurdle of getting the idea 11 accepted and haven't quite focused on the qualifications yet.
12 MR. SPEIS: We discussed this with the CHGR, also.
13 MR. ROBIN: One perspective of the events that have 14 occurred, some 50 of them, none have been caused by seismic 15 events but we have seen a number of events from numerous 16 different causes.
17 CHAIRMAN PALLADINO: _All right.
18 COMMISSIONER BERNTHAL: It I could piggy-back one 19 question I had on that. It is not clear to me how you have 20 taken account of non-safety grade diesel generators or 21 generators I should say that might be onsite. .For example, 22 Turkey Point has some gas turbines or other things. Shoreham 23 was another such case.
24 So that is part of the question. Then the second l -25 question which touches more closely on the point the Chairman i.
l
i 71 1 made is whether in all of the cases that you have studied it 2 really mattered that much whether these things were safety 3 grade and/or bunkered.
4 Were the events relatively simple such that a turbine 5 or a non-safety grade diesel would have done the job? It is 6 the same question in a sense but I am wondering how much all of 7 the bunkering and safety grade and what-not really does for you 8 for the maybe 99-percent of events.
9 MR. BAMANOWSKY: For the events that we looked at, 10 the seismic requirement or any bunkering or that kind of 11 protection didn't really have, wouldn't have any effect.
12 COMMISSIONER BERNTHAL: That is 50 events?
13 MR. BARANOWSKY: That's right.
14- COMMISSIONER BERNTHAL: So it is fair to say 15 conservatively that at least less than one in ten events would 16 ever require something like that. That seems reasonable.
17 MR. BARANOWSKY: A small fraction and we really 18 looked at that as a separate consideration because it can 19 atiect decay heat removal systems. A seismic event can a11ect 20 those as well as the electric power.
~ 21 - COMMISSIONER BERNTHAL: Yes.
22 MR. BARANOWSKY: It is an issue that is really one 23 by itself to be honest with you.
24 MR. DENTON: I think, too, the fact- that just because 25 it is safety grade doesn*t mean it is reliable. We have often
72 1 confused the safety grade with reliability.
2 COMMISSIONER ASSELSTINE: That's right.
3 MR. DENTON: We are finding more and more that that 4 doesn't insure reliability.
5 COMMISSIONER BERNTHAL: In fact, let me state it the 6 other way around. It sounds to me like diversity is probably 7 based on experience is a lot more important than bunkering and 8 safety grade.
9 CHAIRMAN PALLADINO: In other words, we do use these 10 terms and we should know whether they are going to apply to 11 that s i t tia t i on or not or to what extent they apply.
12 COMMISSIONER BERNTHAL: It is the same question you 13 asked on seismic. How important is seismic compared to 14 whether the diesels start or not. It doesn't do any good 11 15 it is in a concrete bunker or not 11 it won't start 11 the 16 principal question is whether it will start. That is just the 17 point I am trying to make.
18 COMMISSIONER ASSELSTINE: At the same time the sense 19 I have is that reliability is really what is important perhaps 20 in the backup systems.
21 I guess I have the impression that gas turbines can 22 be very reliable-11 they are properly maintained. It they 23 are not, they can be fairly unreliable. So it may not be so 24 important whether the system is safety grade or not but it may 25 be. terribly important how that system is maintained.
73 1 It they just have it sitting around there and they 2 are counting on it sort of as a backup and they don *t properly 3 maintain it, it may not buy you very much. On the other hand, 4 11 it is maintained as an essential part of the electrical 5 power supply backup system as in some of the European 6 countries, then you have a reliable backup that may be worth
? something.
8 MR. ROBIN: Some of the points that you are making 9 we have worked with ANS who has just developed a draft standard 10 on station blackout that addresses some of these points, if you 11 need to have equipment to cope with a station blackout, what 12 kinds of qualifications or maintenance should be applied to 13 those equipments.
14 ME. DENTON: I think ones like Turkey Point and i
i 15 Shoreham and maybe others are unique cases and we should give 16 credit if they have non-safety grade electrical generating 17 capacity on site.
18 Typically the design basis approach excludes all 19 those things'but 11 they are there, they ought to be given due 20 credit whatever due credit would be or however how the utility 21 --
what importance they attach to it, that could be reflected.
22 COMMISSIONER BERNTHAL: But up to this point you have 23 not-attempted to make any kind of evaluation of these, l 1
24 MR. BARANOWSKY: We have on some aspects. For 25 instance, we would give credit: like at Turkey Point in that
i 74 1 situation there were let's suppose they had continuous grid 2 problems as they had in the past, they have several onsite 3 sources of power there which we would say would make a major 4 difference in the expected frequency of losses of power to that 5 site and restoration of power.
6 CHAIRMAN PALLADINO: Let me ask a series of questions 7 relating to the application of the backfit rule to this rule.
i 8 Does the backiit rule apply to the station blackout rule at 9 least in the sta11's opinion?
10 (Laughter.)
11 CHAIRMAN PALLADINO: That is only the first of five.
12 COMMISSIONER BERNTHAL: You have the right to remain 13 silent.
14 COMMISSIONER ASSELSTINE: Then you can di s' ingui sh 15 it from Part 20.
16 MR. DENTON: I think that is a legal question.
17 (Laughter.)
18 MR. DENTON: We have done the type oi analyses that 19 I think the backfit rule requires.
20 CHAIRMAN PALLADINO: That is where I was going.
21 MR. SPEIS: Yes.
22 CHAIRMAN PALLADINO: Then you must imply that you i
1 23 ought to do it. Does the cost benefit analysis already done 1
24 for station blackout rule conform to the backfitting analysis '
25 required by the backfit rule in your opinion?
75 4.
1 MR. SPEIS: Yes. In fact, we discussed this issue 2 last time in our presentation in viewgraph five, I believe.
3 COMMISSIONER ASSELSTINE: It is also addressed in a 4 number of the responses to the questions.
5 MR. DENTON: If you look upon the backfit rule as a 6 process, we have gone through that process and have done all
? the steps and I didn*t see that the backfit rule, that seemed 8 looking at it from a staff point of view would codify the 9 practice that was in existence for requiring these types 10 of analysis and they were all done here and we have the type 11 of substantiation that we will publish when we publish the 12 proposed rule.
13 CHAIRMAN PALLADINO: Incidentally, I have five 14 questions but we could stop at the end of two. Let me tell you 15 why I raise these. It was my feeling that you had certainly 16 made a cost benefit analysis that appeared to satisfy the 17 backlit rule.
18 However, one of the problems I have is knowing one 19 when I see one so I did indicate to the Commission and I am
- 20 proposing to send a memo to the stati saying that the 21 Commission requested the NRC stati develop stati procedures 22 to govern the need for and content of backfitting analysis for 23 generic and. licensee specific backlit and I would propose that 24 the stait consult with OGC in developing these procedures and 25 submit it for Commission review.
7@
1 MR. DENTON: I think we need to do that.
2 CHAIRMAN PALLADINO: This is what I was getting at.
3 I think the whole package really was a good justification for 4 having this rule. It did deal with costs. It did deal with 5 benefits. The only question I had was perhaps a quantification 6 of the benefit of reducing the risk from this contribution.
7 MR. DENTON: I think it is another one that was in 8 transition at the time the Commission was considering the 4 9 backfit rule.
10 CHAIRMAN PALLADINO: Let me ask you one other 11 question. You have other generic safety issues and USI*s. I 12 gather some of these do interact with the back11t rule or with 13 the station blackout rule.
14 It seems to me that these other things sort of depend 15 on what we do in the station blackout rule. I am thinking 16 specifically generic issue B-23, reactor coolant pump seal 17 failure. We returned that to the staff for further action.
. 18 Loss of control room instrumentation from loss of AC power and 19 then USI A-45, shutdown decay heat removal requirements, I 20 think those are all interrelated.
21 I guess my point is we ought to move on with getting 22 the station blackout rule settled because I think that could i 23 help guide solutions in these other' areas as well.
24 MR. DENTON: We have considered that point and maybe 25 you would like to comment on that, Themis.
77 1 MR. SPEIS: Yes. One of the things that always 2 concerns us for duplication, overlap and we are very mindful of 3 that. For example, we want to make sure that our proposed 4 resolution to A-45 doesn*t duplicate or overlay with our 5 proposal for A-44 and likewise for the other issues.
6 So in fact in our regulatory analysis on A-44 we do 7 discuss the relationship of A-44 with some of the issues that 8 you mentioned including some additional ones that you did not 9 mention.
10 MR. DENTON: This one is really a key to resolving 11 the other one is the way I look at it. The other ones would 12 not push this but this one pushes those.
13 CHAIRMAN PALLADINO: The reason for bringing it out 14 is I think it is important to get this station blackout rule 15 out.
16 MR. SPEIS: I agree with you.
COMMISSIONER BERNTHAL:
17 Since you keep saying
- 18 backtit, could I piggy-back a question on this?
19 CHAIRMAN PALLADINO: Go ahead.
20 COMMISSIONER EERNTHAL: I am curious to know if I can 21 switch to the other point of view for a moment just to see 22 what the response is, first of all I am surprised Harold that 23 what I thought I heard you say which amounted to the backtit 24 rule being a procedure whereas I would have thought and I 25 don't think that is the word you used but it amounted to that
78 1 and presumably we would have done a cost benefit and the 2 statt would have done it whether or not the backfit rule h ?. d 3 ever been passed.
4 So what are we supposed to cet out of the backtit 5 r u l e 'd I would have thovcht that the conclusion of the cost 6 benefit analysis then would have been a central issue and the 7 application there would have been a central issue then as we 8 seek to conform now with this backilt rule business.
9 I have to say as I look at what you have come up with 10 here knowing as you do what the errors in the analysis are, 11- they surely are bigger than a factor of two or three or four.
12 How in the world can the Commission justify promulgating this 13 rule based on the cost benefit'when by your own numbers we are 14 talking ratios of two to one or something like that or at best 15 a ratio of four to one? It is a pretty slim margin.
16 Let me just restate it. Very. broadly speaking the 17 stati conclusion is that 11 you impose this rule you will get 18 back two dollars for every one you put in it and put another 19 way, I would like to know what the errors are in that number.
20 ME. DENTON: Of course, that goes back to the 21 discussion we have been having about is the function of the 22 agency to prevent core melts or is to prevent offsite risk and 23 the dollars per man rem are only that offsite risk. It doesn't 24 speak to the other point.
25 I would think that a melt down due to loss of AC
. 79 1 would be cor.sidered something in its own merits we would strive 2 to prevent even 11 the olisite cost benefit ratio were very 3 close.
4 COMMISSIONER BERNTHAL: So you are arguing we should 5 t t_ k e into account onsite costs and in that case your number 6 comes out to be four to one which is still pretty slim. I
? will grant you tour to one then. What are the errors in that 8 analysis?
9 CHAIRMAN PALLADINO: I will take that any day four 10 dollars for one dollar.
11 COMMISSIONER BERNTHAL: But what are the errors?
12 MR. DENTON: Starting on our tolerance for risk is 13 really the issue. I think the backfit rule is a process and 14 it says consider all these factors and do these studies and 15 analysis and they have been done.
16 As we have said, a lot of people who have reactors 17 in their country have looked at this issue and decided that 18 more should be done.
19 We really don't know the exact FRA number and it we 20 knew it, like we have just found out someone sticks an earplug 21 in'an oil gauge, that is the kind of thing that leads to an 22 event.
23 You can't get an accurate number. It is not like we 24 have the same component and we are testing it over and over 25 and over. So there are guides. I~think what you=are getting
- 80 1 from the stati in this view is that this area is something 2 that we should do a little more than is presently done. It is 3 not based just on the s2,000.00 per man rem.
4 CHAIRMAN PALLADINO: When we started this cost 5 benefit concept we didn't expect that it was going to be 6 precise. We just didn't want to impose on industry those 7 things that seemed to have no cost benefit.
8 MR. DENTON: No disproportionate cost for the 9 benefit.
10 CHAIRMAN PALLADINO: I don't think we will ever get il rid of all the uncertainties is this kind of analysis.
12 COMMISSIONER ASSELSTINE: Is it fair to say that 13 there are several bases for the staff's judgment. One of them 14 is cost benefit analysis and there may be uncertainties but you 15 think it passes on that ground.
16 MR. SPEIS: Yes.
17 COMMISSIONER ASSELSTINE: The second is this basic 18 concept that you want to drive loss of offsite power station 19 blackout down as a contributor to core melt accidents, down to 20 the neighborhood of ten to the minus tive, something that the 21 ACRS I gather is supportive of and the third is, when you look 22 at what everybody else is doing they are doing at least this 23 . much and in many cases'they are doing substantially more 24 because they perceive on the basis of risk judgments based 25 upon prudent engineering judgment that this is a contributor
- B1 1 that ought to be reduced substantially much along the lines, 2 Joe, that you described, the basic judgment that power is very 3 important and therefore you want to insure that it is going to 4 be available and on all of those bases you think the rule is 5 justi11ed.
6 MR. SPEIS: Yes. I think we told you last time that 7 the engineering analysis was a very important factor in this, 8 not only the happening of this event but its consequences 9 because as we told you, you lose cooling capability but you 10 lose containment cooling at the same time.
11 So that drives the risk and provides more benefit to 12 the cost benefit equation in terms of the person rems. So it
(
13 is all these things together but the engineering analysis was 14 an important consideration. The PHA provided the guidance and 15 the push to say, "My. God, there is something here."
16 So you put all of them together. So this was a 17 perfect issue that all these things blended in a way that gave 18 some rationality to the process.
19 CHAIRMAN PALLADINO: Fred, since you have undertaken .
20 your questions, why don't you proceed.
21- COMMISSIONER BERNTHAL: Do you have more questionst 22 CHAIRMAN PALLADINO: I think I am finished.
23 COMMISSIONER ASSELSTINE: I guess I don't have any 24 more questions. I was just going to say to the staff that you-25 put together a package of information based upon the questions
82 1 that I think basically all of us sent to you. I thought you 2 did a real good job in putting together that information.
3 It was very helpful to me and very useful in terms 4 of getting a sense of where we are on this. If I have any 5 concerns, it is that we may not be going quite far enough in a 6 few areas as I would like to see particularly in insuring high 7 levels of reliability particularly on the maintenance side.
8 One of the things that concerns me is are we going 9 to be tooled in this area the way we were tooled with auxillary 1G teedwater systems at Davis Bessie in thinking that there is a 11 high level of reliability before we can identify specific 12 instances in which reliability of some of these systems is 13 below that level that we find acceptable.
14 Are we going to find those instances before we have 15 something more serious? Basically my comment was I thought the 16 information that you all put together was very, very helpful.
17 COMMISSIONER ROBERTS: Did you read the answers to 18 everybody's questions or just your questions?
19 COMMISSIONER ASSELSTINE: Everybody's, your*s, 20 Lando's, the whole works.
21 COMMISSIONER ROBERTS: We read different documents 22 then.
23 COMMISSIONER ASSELSTINE: 'I t was quite a package of 24 stuit.
25 (Laughter.)
. - - ~. , _ , .- _ _
- a. .
. 83 1 CHAIRMAN PALLADINO: Why don't we go to your 2 questions or comments. Go ahead, Tom.
3 COMMISSIONER ROBERTS: I want to review the answers 4 to my questions which I have not had a chance to do in great 5 detail. At first glance, I am not persuaded they are 6 responsive.
7 COMMISSIONER BEMNTHAL: I actually did have one 8 remaining question that I guess relates somewhat to Jim's 9 comment or question, whichever it was.
10 When you look at the French, the Belgians and the il Germans and the British and Swedish and the Fins who follow the 12 Swedes and the Swiss who follow the Germans or somebody I guess 13 and Japan, how did you decide that doing one-tourth of what 14 they typically do is enough?
15 MR. DENTON: Let me start with, Themis, did you tell 16 your anecdote?
17 COMMISSIONER ASSELSTINE: That is a good question.
18 MR. DENTON: I don't know it we have had Themis 19 recount his discussion with the French on this matter, did you 20 do it last time, about the tour to eight?
21 MR. SPEIS: We met with the French about a month ago 22 on the Paluel design and the first presentation I had to make 23 was to describe our approach to station blackout.
24 I started talking and.I came to the coping capability 25 and I starting mentioning the word "four" and before I
. 84 1 finished, they got up and they apologized. They said, "We are 2 not as conservative as you are."
3 (Laughter.)
4 MR. SPEIS: Then I said "four hours" 5 (Laughter.)
6 CHAIRMAN PALLADINO: That is interesting.
7 COMMISSIONER ASSELSTINE: I think that is a very good 8 question, Fred.
9 MR. DENTON: To some extent many other countries are 10 taking our minimum standards and building on them. They are 11 saying it is a good starting point but they are not necessarily 12 the best thing for their country.
13 Now they have other power concerns and PUC 14 arrangements and a lot of other factors that drive them to do 15 different maintenance practices and different approaches to 16 these problems.
17 CHAIRMAN PALLADINO: I think they are taking some of 18 the improbable accidents about which there seems to be debate 19 and making them so improbable they are no longer an issue.
20 MR. DENTON: Often 11 there are plants on borders 21 between countries, they do to extraordinary measures to assure 22 the acceptability of the plant.
23 COMMISSIONER BEHNTHAL: Population density and the 24 smallness of the countries certainly does play a role.
25 CHAIRMAN PALLADINO: Do you have mo r e 't
85 1 COMMISSIONER BERNTHAL: No, I guess that is all 2 CHAIRMAN PALLADINO: All right Commissioner Zech.
3 COMMISSIONER ZECH: Jim go ahead.
4 CHAIRMAN PALLADINO: I am sorry. Did you have 5 another question?
6 COMMISSIONER ASSELSTINE: I was just going to follow 7 on to what Fred said. Isn*t the fundamental driving fpree 8 behind many of those ettorts simply the judgment that they 9 just don't want to have an accident caused by this itetar?
i 10 They really want to drive it down to the level where it is not 11 a real concern anylmore?
12 MR. DENTON: I think the cost benefit that the 13 utility does as the owner might turn out different than the ,
14 cost benefit here. Many of these plants are a substantial 15 investment for that small country. They can't stand the loss 16 in that country of their nuclear program.
17 So they go to great lengths to prevent any kind of 18 accident leaks occurring in the core. Often you find us 19 working in the more otisite area in terms of doing cost 20 benefit. I think their cost benefits are nearer a fact of 21 protecting the investment.
22 CHAIRMAN PALLADINO: You did bring out a point that 23 I wanted to bring out before and I torgot. These utilities 24 are going to have to make their own cost benefit analysis when 25 they re-examine this situation on a plant specific basis. Is
~
, s 80 1 that not true?
2 MR. SPEIS: We attempted to estimate the costs.
3 CHAIRMAN PALLADINO: I know you did it on a generic 4 basis but different plants are going to have to do different 5 things.
6 MR. DENTON: Yes.
7 MR. SPEIS: Yes.
8 CHAIRMAN PALLADINO: They are going to come back with 9 a proposal and then I presume that is when you will see their 10 cost benefit analysis and I presume they would make one.
11 Otherwise, they wouldn't have a basis for arguing one way or 12 the other.
13 MR. DENTON: Yes, that is right.
14 CHAIRMAN PALLADINO: Lando.
15 COMMISSIONER ZECH: I would just like to say I 16 appreciate the effort the statt has put in in responding to 17 your questions. I, too, want to review them a little more J
18 carefully.
19 On the backtit discussion that we have had here 20 briefly this afternoon, it seems to me that at least I want to 1
21 be satisfied that wc have taken a rather disciplined look at 22 the whole station blackout process.
23 I know it has been going on for a long time but it 24 does seem to me that we are dealing in uncertainties and 25 estimates and things that at least in my mind I want to be I
87 ,
1 a little more satisfied that we are going to do the right 2 thing.
! 3 So I need to continue my review of the station 4 blackout proposal you have but I do think that you have tried 5 to wrestle with a very important issue in a very responsible 6 manner and I appreciate it. I just have to satisfy myselt 7 before I am ready to move forward, Mr. Chairman, but it s
8 shouldn*t be too long as far as I am concerned.
9 ~ CHAIRMAN.PALLADINO: I did want to poll the 10 Commissioners to see 11 there is anything more that we should 11 be doing before you think you are ready to vote.
4 12 COMMISSIONER ZECH: I want to review the answers you
, 13 have given us just a little more carefully than I have had a 14 chance to do. I may have some more questions but I will do 15 that soon.
16 CHAIRMAN PALLADINO: Rather than as of this moment?
17 COMMISSIONER ZECH: I am not quite ready yet.
J 18 CHAIRMAN PALLADINO: I understand, but as of this 19 moment there isn*t any something more we should be doing to 4
20 help you in your process.
21 COMMISSIONER ZECH: No. All I want to do is review 22 what I have here and if I have any questions I will get them 1
23 out promptly.
24 CHAIRMAN PALLADINO: All right. Tom, do you have
- i. 25 anything?
. 88 1 COMMISSIONER ROBERTS: Similar response.
2 CHAIRMAN PALLADINO; Same thing.
3 COMMISSIONER ROBERTS: Yes.
4 CHAIRMAN PALLADINO: Anybody else?
5 COMMISSIONER ASSELSTINE: I don't need anything 6 more. I am ready to vote.
7 COMMISSIONER BERNTHAL: I am ready to move ahead with 8 it. It is a proposed rule and we haven't talked here about the 9 request from NUMARC and obviously the industry group that is 10 working on this particular issue for the Commission to toss 11 this issue into their lap instead and in a way I feel wo 12 ought to say something.
13 CHAIRMAN PALLADINO: We said a little bit but not 14 very much.
15 COMMISSIONER BERNTHAL: Obviously, the first decision 16 will be an answer it we decide to put out the proposed rule but l 17 it seems to me we owe them a bit more than that.
18 I made the comment in the last meeting we had with 19 them and I have to say I still feel that way that there 20 are certain issues that in my judgment are things that do not 21 lend themselves well to analytical proceeding, to analysis I 22 guess is what I want to say.
23 Clearly, titness for duty is one of those, training, 24 in many cases where human factors 4re involved and human 25 factors vary tremendously from plant-to-plant and in those i
~
89 1 cases, I think those are areas where the approach that we have 2 taken to provide regulatory guidance rather than issuing rules 3 is a good one.
4 But in cases like this where you are really talking 5 about hardware matters that lend themselves to the analytical e approach I just am not convinced that that is the right kind 7 of issue to be tossing into the uncertainty of non-rule 8 framework.
9 With all due respect to the many good things that 10 NUMARC has done and the oitorts that the industry has 11 undertaken here in my judgment this is the kind of issue, it is 12 an unresolved safety issue that is most appropriately carried 13 out within the context of a rule where everybody knows where 14 they stand.
15 This is a pretty clear cut one. Other people are 16 putting in generators and steam-driven pumps and various things 17 that specifically address in a way that you can sit down and 18 count them the unresolved safety issue.
19 I think it is pretty clear how you deal with issues 20 like this.
21 CHAIRMAN PALLADINO: This issue was dealt with by 22 NUGSHO, the Nuclear Utility Group on Station Blackout.
23 COMMISSIONER ASSELSTINE: That's right, 24 CHAIRMAN PALLADINO: There was much interaction as !
25 understand it between the staff and them. The staff did
90 l l
1 address their comments and I think made some changes in the 2 rule as a result of some of their input but when they seemed 3 not to be able to go any farther then they dropped it and now 4 NUMARC is picking it up.
r 5 I think that is much too late. I think we have 6 certainly dealt with industry concerns. There is more 7 cognisant industry group on this subject than the NUGSBO and I B think they have been given consideration. At least that is my i 9 point of view.
10 COMMISSIONER ASSELSTINE: That is my perception as 11 well.
12 CHAIRMAN PALLADINO: I don *t know 11 you are 13 suggesting we need to write them a letter and deny but I thirk 14 we might write them a letter and say that we are not really 15 ignoring their otter to participate in this activity but rather 16 we suggest that they do it during the comment period.
17 That would give them time to deliberate on the matter 1
. 18 and come up with their proposals.
Id COMMISSIONER ASSELSTINE: Yes. I assume that the 20 comment period the statt is proposing and I don't remember 21 specifically what it was, but is long enough to aooommodate 22 comments in the February time trame.
23 MR. RUBIN: Three months.
24 COMMISSIONER ASSELSTINE: Three months, great.
25 CHAIRMAN PALLADINO: Is that what you had, 90 days?
I
o
- 91
)
1 MR. RUBIN: Yes.
2 CHAIRMAN PALLADINO: All right.
3 COMMISSIONER ZECH: Somewhere or other I think we 4 should hear from the group that was working on the station 5 blackout proposal.
6 CHAIRMAN PALLADINO: They withdrew their request to 7 come in today.
8 COMMISSIONER ZECH: I know that but they have given l
9 it over to NUMARC as I understand it.
l 10 In other words, I know they put in ettort. I know 11 they have worked with the stati It seems to me that it is 12 only appropriate that we figure out a way that we get their 13 views before we move on.
l 14 CHAIRMAN PALLADINO: We were prepared to have them l
15 appear today. That doesn't mean that we should cut them ott 16 from appearing any other time.
17 COMMISSIONER ZECH: That is all I mean. I think some 18 way or another we ought to figure out how to hear trom them.
19 CHAIRMAN PALLADINO: When they are ready to be heard, 20 I think we should hear them, That is why I asked them during 21 the NUMARC meeting when they think they might have some 22 feedback and they said February.
23 COMMISSIONER ZECH: As far as I am concerned we don't 24 need to keep it wide open. We can tell them when we want to 25 hear from them and 11 they don't want to come, that is fine.
O O 92 1 11 we don *t want to wait until February, I would imagine that 2 they would accommodate us at an earlier time.
3 I just feel that it would be appropriate. We have 4 discussed this for so long now that 11 they have any input, 5 'they ought to have an opportunity to make it.
6 CHAIRMAN PALLADINO: They had it today, the NUGSBO 7 group.
8 COMMISSIONER ZECH: But I understand NUGSBO torfeited 9 to NUMARC.
10 CHAIRMAN PALLADINO: That doesn*t say the industry 11 didn*t have a chance to at least come in before we are thinking 12 of making the decision.
13 COMMISSIONER ZECH: I appreciate that, Mr. Chairman, 14 I still think we ought to at least make an effort to hear from 15 them.
16 CHAIRMAN PALLADINO: Yes, and I think we ought to 17 communicate with them on this subject.
18 COMMISSIONER BERNTHAL: I have no objection obviously 19 to hearing NUMARC when they feel they are prepared to speak to 20 the issue to come in. But, Joe, I have to agree with you. I 21 think it is rather late in the game now to lateral the ball 22 here from NUGSBO which has to11 owed this issue throughout and
, 23 has presented a document to the Commission and now lateraled 24 the ball to NUMARC and have them come in and present the issue 25 to the Commission.
93 1 They are tree to do that. This is a proposed rule.
2 They are tree to comment as part of the formal comment 3 procedure here or they can come in and talk with us. That is 4 fine with me.
5 Underlying that, first of all, I think it is late in 6 the game and secondly, it is going to take some convincing to 7 convince me that this is the kind of issue that lends itselt 8 well to self-regulation. We are talking about hardware 9 here and by in large I think the removal of regulatory 10 uncertainty is very important on these hardware issues and 11 that means that there be some pretty clear guidelines in my 12 judgment.
i 13 CHAIRMAN PALLADINO: I am going to propose that we l
14 correspond with NUMARC and indicate the opportunity to come in 15 whenever they are ready and the opportunity to comment while i
16 this rule is out for comment.
~
17 COMMISSIONER ASSELSTINE: That sounds reasonable to 18 me .-
t 19 CHAIRMAN PALLADINO: All right. Anything else that 20 should come before us?
21 (No response.)
22 CHAIRMAN PALLADINO: Let me indicate that I was going 23 to propose a ten-minute recess between this meeting and 24 atitrmation so all of us can catch up with any last minute a
25 comments on the items to be attirmed.
. - . , - - - - - - - - - . - .n- ,n - , ,s . -
-- -- '^ >I---- . ,__ w _,2_,;__ _ _
e e 94 i
1 COMMISSIONER ASSELSTINE: Good.
2 CHAIRMAN PALLADINO: Before we break'up, let me 3 concur in the appreciation expressed by several Commissioners 4 4 for the work that you have done on this issue and I hope that 5 we can get a vote out reasonably-soon. So thank you very much 4
+ 6 for a good. presentation.
7 MR. SPEIS: Thank you.
8 CHAIRMAN PALLADINO: We stand adjourned.
9 (The Commission meeting was adjourned at 4: 15 o' clock 10 p.m., to reconvene at the Call of the Chair.)
11 4
12 13 ,
14 4
15 16 i
17 18 1
19 20 21 I
22 v 23 i
24 25 i
. b 1
CERTIFICATE OF OFFICIAL REPORTER i 2
(
3 4
, 5 This is to certify that the attached proceedings l'
6 before the United States Nuclear Regulatory Commission in the 7 matter of. COMMISSION MEETING 1
8 9 Name of proceeding: Continuation of 9/11 Discussion of ,
Proposed Station Blackout Rule (Public
- 10. Meeting) 4 11 Cocket No.
12 placet Washington, D. C.
j
! 13 Date: Thursday, November 14, 1985 14 15 were held as herein appears and that this is the original l
16 transcript thereof for the file of the United States Nuclear
- 17 Regulatory Commission.
13 19 (Signature)
V)t4Qw .pl %TL wu (Typed Name of Report 4r) Marilynn Nations 20 21 22 23 Ann Riley & Associates. Ltd.
24 25 ,.
i
s- .
e i
i .
i i
i PRESENTATION TO THE COMMISSION ON STATION BLACK 0UT (UNRESOLVED SAFETY ISSUE A-44)
BY H. DENTON, T. SPEIS AND A. RUBIN l OFFICE OF NUCLEAR REACTOR REGULATION i U.S. NUCLEAR REGULATORY COMMISSION l . NOVEMBER 14, 1985 3
i
+
i i
4 1
.).
SLIDE TITLE .
1 TOPICS OF INTEREST SINCE SEPTEMBER 11TH COMMISSION MEETING ON USI A-44 2 TOPICS TO BE DISCUSSED 3 LOSP EVENTS SINCE FIRST COMMISSION MEETING ON USI A-44 (SEPTEMBER 11, 1985) 4 HURRICANE GLORIA RESPONSE 5 TOTAL LOSP EVENTS TO DATE IN 1985 6 EMERGENCY DIESEL GENERATOR (EDG) RELIABILITY 4
7 ACTUAL STATION BLACK 0UT EXPERIENCE COMPARED TO NUREG-1032 ESTIMAIES i 8 INSIGHTS FROM PLANT-SPECIFIC PRAs 9 FRENCH PWR DESIGN FOR STATION BLACK 0UT 10 RECENT INDUSTRY COMMENTS ON NUREG-1032 11 ACRS COMMENTS ON PROPOSED RESOLUTION 12 STAFF's CONCLUSIONS AND RECOMMENDATIONS 1
1 4
TOPICS OF INTEREST SINCE SEPTEMBER 11TH COMMISSION MEETING ON USI A-44 TWO LOSSES OF 0FFSITE POWER (ONE CAUSED BY HURRICANE GLORIA AND ONE PLANT-CENTERED LOSS),
ADDITIONAL INFORMATION ON FRENCH PWR DESIGN FOR STATION BLACK 0UT.
l INDUSTRY COMMENTS ON DRAFT NUREG-1032.
ANSWERS TO COMMISSIONERS' QUESTIONS ON USI A-44.
1 I
l
TOPICS TO BE DISCUSSED RECENT LOSP EVENTS AND RESPONSE TO HURRICANE GLORIA.
IRENDS IN EMERGENCY DIESEL GENERATOR FAILURE RATES.
STATION BLACK 0UT EXPERIENCE COMPARED TO NUREG-1032 ESTIMATES.
FRENCH PWR STATION BLACK 0UT COPING CAPABILITY.
INDUSTRY AND ACRS COMMENTS.
AL:'TIONAL INFORMATION AND PERSPECTIVES IN STATION BLACK 0UT REGARDING INDUSTRY COMMENTS.
CONCLUSIONS AND RECOMMENDATIONS FROM LAST COMMISSION MEETINGS ARE UNCHANGED.
2
LOSP EVENTS SINCE FIRST COMMISSION MEETING ON USI A-44 (SEPTEMBER 11, 1985)
DATE DURATION CAUSE OF LOSP 09/27/85 20 HRS. TOTAL LOSP AT MILLSTONE DUE (RESTORATION TO SALT WATER SPRAY IN MIGHT HAVE SWITCHYARD FROM HURRICANE BEEN POSSIBLE GLORIA. BOTH UNITS SHUTDOWN IN LESS THAN BEFORE OFFSITE POWER WAS LOST.
20 HRS.) 0FFSITE POWER RESTORED IN APPROXIMATELY 20-24 HOURS. IN 1976 HURRICANE BELLE CAUSED 5 HOUR ,
LOSP DUE TO SALT SPRAY, "THE SWITCHYARD HAS SINCE BEEN RE-INSULATED TO PROTECT AGAINST THIS PROBLEM." (0UOTE FROM NSAC 85, " LOSSES OF 0FFSITE POWER AT U.S. NUCLEAR POWER PLANTS, ALL YEARS THOUGH 1984, JUNE 1985."
10/03/85 24 MIN. SIMULTANEOUS LOSS OF BOTH OFFSITE POWER CIRCUITS AT PALO VERDE DUE TO SINGLE COMMON CAUSE FAILURE IN SWITCHYARD. PLANT AT APPROXIMATELY 80% POWER. 0FFSITE POWER RESTORED IN 24 MINUTES.
3
HURRICANE GLORIA RESPONSE SITUATION: HURRICANE GLORI A THREATENED NUCLEAR FACIL-ITIES ALONG THE EAST C0AST BEGINNING THURSDAY, 09/26 THROUGH FRIDAY, 09/27.
FACILITIES NEAR PATH OF HURRICANE:
BRUNSWICK INDIAN POINT SURRY MILLSTONE CALVERT CLIFFS HADDAM NECK SALEM PILGRIM OYSTER CREEK MAINE YANKEE SHOREHAM ALL FACILITIES WERE OPERATING AT OR NEAR 100% POWER WITH THE EXCEPTION OF:
SHOREHAM - 2%
INDIAN POINT 2 - 75%
INDIAN POINT 3 - 0%
MAINE YANKEE - 0% (REFUELING)
ALL FACILITIES REDUCED POWER SIGNIFICANTLY (EXCEPT SALEM 2 AND CALVERT CLIFFS); MOST SHUT DOWN.
DAMAGES SUSTAINED:
MILLSTONE - LOSS OF 0FF-SITE POWER DUE TO SALT SPRAY.
SHOREHAM - LOSS OF ONE OF THE TWO SOURCES OF 0FFSITE POWER FROM THE 69 KV GRID.
SALEM 1 - POWER REDUCED TO 35% DUE TO CLOGGED INIAKE.
CONCLUSION HURRICANE GLORIA WAS MOVING S0 FAST, UTILITIES HAD TO DECIDE QUICKLY WHAT ACTIONS TO TAKE. POTENTIALLY MANY PLANTS COULD BE j AFFECTED BY ONE SEVERE HURRICANE.
l i
i TOTAL LOSP EVENTS TO DATE IN 1985 DATE PLANT RESTORATION TIME CAUSE OF LOSP 05/07/85 0UAD CITIES 43 MIN.* PLANT-CENTERED 05/17/85 TURKEY POINT 2 HRS. 5 MIN.* GRID / WEATHER (FIRE DURING DROUGHT DISABLED 500Kv TRANSMISSION LINE) 08/16/85 DRESDEN 5 MIN.* PLANT-CENTERED 09/27/85 MILLSTONE a' 20 HOURS SALT SPRAY FROM HURRICANE GLORIA
, 10/03/85 PALO VERDE 24 MIN. PLANT-CENTERED CONCLUSION LOSP FREQUENCY IN 1985 IS APPR0XIMATELY EQUAL TO THE AVERAGE INDUSTRY EXPERIENCE FOR THE PAST 25 YEARS.
i
- RESTORATION TIMES REPORTED BY EPRI.
5 l
l
EMERGENCY DIESEL GENERATOR (EDG) RELIABILITY TRENDS SHOW SLIGHT IMPROVEMENT IN EDG RELIABILITY SINCE 1976.
AVERAGE EDG FAILURE RATE 0.02 PER DEMAND * (FOR ALL DEMANDS).
AVERAGE EDG FAILURE RATE 0.05 PER DEMAND DURING LOSP.
MAINTAINING RELIABLE EMERGENCY AC POWER IS IMPORTANT PART OF USI A-44 RESOLUTION.
I l
l l
' FAILURE TO START'IN LESS THAN 10 SEC. NOT COUNTED AS A FAILURE, l
l 6
ACTUAL STATION BLACK 0UT EXPERIENCE COMPARED TO NUREG-1032 ESTIMATES ACTUAL TOTAL LOSSES OF ALL AC POWER 4 EVENTS TOTAL NUMBER OF REACTOR-YEARS 850 REACTOR-YEARS CALCULATED FREQUENCY OF STATION BLACK 0UT 4,7 x 10-3 EVENTS PER REACTOR-YEAR TYPICAL ESTIMATED STATION BLACK 0UT FREQUENCY 10-4 PER YEAR FROM NUREG-1032 ESTIMATED RNAGE OF FREQUENCY OF STATION BLACK 0UT FROM NUREG-1032 10-5 TO 10-3 PER YEAR CONCLUSIONS i NUREG-1032 ESTIMATES FREQUENCY OF STATION BLACK 0UTS IN THE FUTURE WILL BE LOWER THAN PAST EXPERIENCE.
RECOGNIZES TRENDS SHOWING IMPROVEMENTS IN LOSP EXPERIENCE.
ACKNOWLEDGES IMPROVEMENTS MADE TO REDUCE COMMON CAUSE FAILURES, l -
ASSUMES EDG FAILURE RATE DURING ACTUAL DEMANDS IS THE SAME AS DURING SURVEILLANCE TESTS, SB0 CAN STILL BE A DOMINANT CONTRIBUTOR TO TOTAL CORE MELT l
FREQUENCY, 7
INSIGHTS FROM PLANT-SPECIFIC PRAs PLANT-SPECIFIC PRAs CORROB0 RATE THE TECHNICAL FINDINGS OF NUREG-1032 THAT STATION BLACK 0UT CAN BE A DOMINANT CONTRIBUTOR TO TOTAL CORE MELT FREQUENCY.
APPR0XIMATELY 50% OF PRAs ESTIMATED CORE MELT FREQUENCY FROM SB0 GREATER THAN 10-5 PER REACTOR-YEAR (25% GREATER THAN 5 X 10-5),
ABOUT HALF 0F THE PRAs ESTIMATED CORE MELT FREQUENCIES FROM SB0 TO BE LESS THAN 10-5 REACTOR-YEAR, BUT SOME OF THESE OMITTED POTENTIALLY DOMINANT SB0 SEQUENCES, STATION BLACK 0UT CONTRIBUTION TO TOTAL CORE MELT FREQUENCY RANGED FROM 1% TO ALMOST 50% (GREATER THAN 20% FOR APPR0XIMATELY 1/4 0F THE PLANTS).
8
8* .
1 l
FRENCH PWR DESIGN FOR STATION BLACK 0UT FRENCH PLANTS ARE CAPABLE OF COPING WITH STATION BLACK 0UT FOR 3 DAYS. THIS " DEFENSE-IN-DEPTH" APPROACH WAS TAKEN BECAUSE OF CONCERN ABOUT COMMON MODE FAILURES OF REDUNDANT SYSTEMS.
POWER SUPPLIES AVAILABLE TO POWER ESSENTIAL AC BUSES.
TWO ONSITE EMERGENCY DIESEL GENERATORS PER UNIT.
CAPABILITY TO REDUCE POWER TO HOUSE LOAD WITHOUT REACTOR TRIP.
MOBILE GAS TURBINE GENERATOR.
CAPABILITY TO PROVIDE POWER FROM ONE UNIT TO ANOTHER THROUGH SWITCHYARD.
CAPABILITY TO INTERCONNECT EMERGENCY DIESEL GENERATOR FROM ONE UNIT TO ANOTHER. . ,
DESIGN FEATURES TO ACHIEVE 3-DAY STATION BLACK 0UT COPING CAPABILITY.
TWO STEAM-DRIVEN AFW PUMPS (IN ADDITION TO 2 MOTOR-DRIVEN PUMPS).
STEAM TURBINE-DRIVEN GENERATOR FOR REACTOR COOLANT PUMP SEAL COOLING AND INSTRUMENTATION AND CONTROL POWER.
GRAVITY FEED BACK-UP WATER SUPPLY TO CONDENSATE STORAGE TANK.
EMERGENCY PROCEDURES FOR LOSS OF ALL AC POWER.
ALL OPERATING 900 MWE FRENCH PLANTS ARE BEING BACKFIT WITH A STEAM-TURBINE DRIVEN GENERATOR TO ACHIEVE LONG-TERM STATION BLACK 0UT COPING CAPABILITY.
9 f
- i. . i l
~
RECENT INDUSTRY COMMENTS ON NUREG-1032 CONCERN THAT THE STAFF'S TREATMENT OF HISTORICAL LOSS-0F-0FFSITE POWER EVENTS DOES NOT GIVE SUFFICIENT CREDIT FOR TRENDS SHOWING RECENT IMPROVEMENTS, CONCERN THAT THE STAFF DOES NOT~ REALISTICALLY EVALUATE EMERGENCY DIESEL GENERATOR RELIABILITY (E.G., UNSUCCESSFUL FAST STARTS ARE COUNTED AS FAILURES), AND THEREFORE, THE ANALYSIS OVERESTIMATES THE RISK FROM STATION BLACK 0UT.
QUESTIONS RELATING TO THE STATISTICAL VALIDITY OF ESTIMATING LOSS-OF-0FFSITE-POWER FREQUENCY FOR GROUPS OF PLANTS, RECOMMENDATION THAT NEW SOURCE TERMS BE USED FOR MORE REALISTIC CONSEQUENCE ANALYSIS, CONCLUSION THAT STATION BLACK 0UT DOES NOT CONTRIBUTE SIGNIFICANTLY TO RISK, AND RULEMAKING IS NOT JUSTIFIED, 10
- i. .
't 4
ACRS COMMENTS ON PROPOSED RESOLUTION ACRS SUPPORTS THE STAFF'S PROPOSED APPROACH FOR THE RESOLUTION OF USI A-44 (LETTER FROM ACRS TO W. DIRCKS, JULY 13, 1983).
- ACRS REAFFIRMS SUPPORT OF STAFF'S PROPOSED RECOMMENDATIONS TO RESOLVE USI A-44 (LETTER FROM ACRS TO W. DIRCKS, MARCH 12, 1985).
t i
11
- s. .
STAFF'S CONCLUSIONS AND RECOMMENDATIONS CONCLUSIONS UNCHANGED FROM SECY 85-163/163A.
SB0 CAN BE DOMINANT CONTRIBUTOR TO TOTAL CORE MELT.
SUBSTANTIAL REDUCTIONS IN CORE MELT FREQUENCY CAN BE ACHIEVED BY PROPOSED RULE.
ADDITIONAL DEFENSE-IN-DEPTH (I.E., ABILITY TO COPE WITH SB0)-WOULD ASSURE PUBLIC HEALTH AND SAFETY FROM SB0 OVER EXPECTED LIFETIME OF PLANTS.
PUBLISH PROPOSED RULE FOR COMMENT.
l 12
BACKUP SLIDES 1
l I
r
FIGURE B.3 LOSS OF OFFSITE POWER LEARNING CURVE: 1959-1983 A 1 V -
E .
R -
A .
G -
E C
U ~
M U-
~
IT 0.1 -- O IrlDUSTRY AVG. 088 g OO l .- NN V .-
+ .071 USED Ill flUREG-1032 ANALYSIS E _ GIVING CREDIT FOR REDUCED PLANT - CENTERED LOSP F -
R ~
E O
U -
E ,
N C
Y 0.01 - -
i 7- i i
' i7 iiT 10 100 1000 CUMULATIVE SITE-YEARS
~
hYkkkkYYYQ0hQf0000000000kt0QhkkG0000000h)kkg0gQkgQgQgQqQgQgQ&
1 5 _
h
- g:; TRANSMITTAL TO: W=mt Control Desk, 016 Phillips 9/35 ADVANCED COPY TO: / / The Public Document Rxan
, cc: C&R
- == -- Ta'";-
papers)
Attached are copies of a C%mmission meeting transcript (s) and related meeting h=mt(s) . They are being forwarded for entry on the Daily Accession List j and plamt in the Public W=mt Boan. No other distribution is requested or required. Existing DCS identification ntunbers are listed on the individual docunents wherever knqwn.
S Meeting
Title:
b.malb o $ k l/ 'D r s s 5 h 0I Er e90' sed.
SO 3kkM Me. ,
a Meeting Date: Open /
] ll lIdl (5 Closed h
3 DCS Copies (1 of each checked)
] Itan
Description:
Copies 3 Advanced Original May Duplicate ib PDR , Wwnt be Dup
- Copy *
- l. TRANs.2urr 1 1
.. Men checked, DCS should send a ,
og of this transcript to the .
LPDR for: ,
L6fD16Aarads I *
- 2. ,
3
- h
- I
- 3. [
- 4. )
(PDR is advanced one copy of each doctznent,
- Verify if in DCS, and two of each S M paper.)
- 01ange to "PDR Available."
l
$ h $$ $ b bbh h