ML20149L635
| ML20149L635 | |
| Person / Time | |
|---|---|
| Issue date: | 02/19/1988 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1646, NUDOCS 8802240334 | |
| Download: ML20149L635 (294) | |
Text
/ic e r- /6 w ; l
- R 3 NA_ i UNITED STATES l NUCLEAR REGULATORY COMMISSION j
\
l I
ADVISORY CO*iMITTEE ON REACTOR SAFETY SUBCO"!!ITTEE ON JOINT SCRAM S"STE!!S RELIABILITY AND CORE PERFOR!!ANCE In the !!atter of:
GENERAL !!EETING 1
l t ,
t LOCATIONt, 'dashington , D .C . PAGES: 1 through 221 DATE: February 19, 1988 s* % .*
-a--======== -==
=======--------------- .......------- ..--
, lthis0 SIM Nb HERITAGE REPORTING CORPORATION onuwavaun 1224 L Street, N.W., Sake 644 WanWagton, D.C. 20005 (282) 628 4888 8802240334 880219 '
, 1 PUBLIC NOTICE-BY THE 1
2 UNITED STATES NUCLEAR REGULATORY-COMMISSION'S 3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
5 6
7 The contents of this stenographic transcript of the 8 proceedings of the United _ States Nuclear Regulatory 9 Commission's Advisory Committee on Reactor Safeguards (ACRS),
10 as reported herein, is an uncorrected record of the discussions 11 recorded at the meeting held on the above date.
12 No member of the ACRS Staff and no participant at-13 this meeting accepts any responsibility for errors or
( 14 inaccuracles of statement or data contained in this transcript.
15 16 17 18 19 20 21 22
. 23 24 25 i Heritage Reporting Corporation (202) 628-4888 e
, 1 1
1 UNITED STATES NUCLEAR REGULATORY COMMISSION 2 ADVISORY COMMITTEE ON REACTOR SAFETY 3 SUBCOMMITTEE ON JOINT SCRAM SYSTEMS RELIABILITY AND CORE PERFORMANCE 4
)
5 In the Matter of: )
)
6 )
)
7 GENERAL MEETING )
)
8 Friday 9 February 19, 1988 10 Room 1046 1717 H Street, N.W.
11 Washington, D.C. 20555 12 The above-entitled matter came on for hearing,
, 13 pursuant to notice, at 8:30 a.m.
14 BEFORE: MR. WILLIAM KERR 15 16 ACRS MEMBERS PRESENT:
17 MR. DAVID WARD !
Research Manager on Speciel Assignment 18 E.I. du Pont de Nemours & Company Savannah River Laboratory ,
19 Aiken, South Carolina '
20 MR. CHARLES J. WYLIE .
Retired Chief Engineer l 21 Electrical Division l Duke Power Company l
., 22 Charlotte, North Carolina 23 l 24 25 l Heritage Reporting Corporation (202) 628-4888 l
1
2 i
1 ACRS MEMBERS PRESENT (CONTINUED):
i 2 DR. PAUL G. SHEWMON '
Professor, Metallurgical Engineering Department 3 Ohio State University Columbus, Ohio
- 4 MR. JESSE C. EBERSOLE 5
ACRS CONSULTANTSt -
6 P. DAVIS 7 J. LEE W. LIPINSKI 8
ACRS COGNIZANT STAFF MEMBER: ,
9 PAUL BOEHNERT 10 :
NRC STAFF PRESENTER: j 11 D. FIENO 12 l'
, 13 14 l
15 l
16 17 18 -
19 20 21 22 , ,
23 !
24 !
25 Heritage Reporting Corporation j (202) 628-4888
3 1 PROCEEDINGS 2 MR. KERR: The meeting will come to order. This is a 3 meeting of the Advisory Committee on Reactor Safeguards, 4 Combined subcommittees on Scram Systems Reliability and Core 5 performance.
6 My name is William Kerr, I'm Subcommittee Chairman.
7 The other ACRS Members here today are: Mr. Ebersole, 8 Mr. Shewmon, Mr. Ward and Mr. Wylie.
9 Our Consultants are: Mr. Davis, Mr. Lee and Mr. ,
10 Lipinski.
11 The meeting is being held to review the current 12 status of light-water reactor plant operations with some -
j 13 emphasis on core reload designs as they may influence core 14 reactivity control operations; and also, as they may influence 15 the analysis and subsequent performance of certain hypothesized 16 transients, particularly the transients that might accompany 17 ATWS.
18 Paul Boehnert is the ACRS staff member responsible ;
19 for the meeting. Rules for participation in the meeting were 20 announced as part of the notice of the meeting in the Federal l 21 Register of February 2, 1988. l
., 22 A transcript of the meeting is being kept. I ask l l
23 that each speaker identify himself or herself and use a j l
24 microphone. j 25 Before we go to the meeting itself are there any l
\
i Heritage Reporting Corporation )
(202) 628-4888 l l
l l
l 1
4 1 comments that any members of the Subcommittee or Consultants 2 would like to make at this point? Anything you would like to 3 see emphasized, for example?
4 (No response) 5 MR. KERR We will proceed with the meeting, and our 6 first speaker is Mr. Ted Enos from Arkansas Power. I'm told he 7 repreaents the B&W Owners Group.
8 Mr. Enos.
9 (Slides being shown.)
10 MR. ENOS: Good morning, my name is Ted Enos, I'm the 11 manager of nuclear support for Arkansas Power and Light. Today 12 I am representing the B&W Owners Group ATWS Subcommittee.
- 13 We would like to talk with you for a few minutes this 14 morning about where we are on the ATWS issue and some of the 15 changes that have taken place in recent years that may impact 16 ATWS either positively or negatively.
. 17 Through the course of the presentation this morning 18 we want to discuss the ATWS requirements in brevity as to what 19 they mean for the B&W machines.
20 Programs and physical changes to the plants that may 21 have taken place since the '79 time frame when we did analysis 22 work which formed the basis for the ATWS rule and how those may ,
23 affect ATWS as it applies to the B&W machines.
24 Look specifically at fuel and fuel cycle changes and 25 how those things may have impacted the ATWS analysis.
Heritage Reporting Corporation (202) 628-4888
5 i 1 And address the adequacy of the current plant 2 configurations relative to the ATWS basis. ,
3 I'm going to do a very brief review here of ATWs and ;
4 -how it applies to us. The. purpose of the ATWS rule as was ;
5- promulgated in 1984 was to reduce the likelihood of the failure-6 of the reactor protection system and also to-mitigate the 7 consequences of an ATWS event, should one occur.
8 The rule requires for B&W plants to install a diverse 9 scram system and an AMSAC system for mitigation. Not included 10 to be or intended to be a final list here, but just a summary .,
11 of what the final rule requirements are for the B&W plants: the :
12 diverse scram system and some of the elements of that. And 13 also, the AMSAC or the mitigating system which requires auto ;
14 initiation of aux feedwater and also in automatic turbine trip 15 off of the ATWS event.
[
16 Over the years there has been a rather extensive ;
17 analysis that has been conducted to look at ATWS and see how it l 18 applies to the plants; starting in 1970, a lot of work in '70, i 19 '72 and '74, looking at a very broad spectrum of anticipated ;
20 transient type events. l 21 In 1980 we performed an analysis documented as what
., 22 we refer to as BAW-1610, and this is the analysis performed in l
l 23 response to NUREG 460, volume 3. That analysis really forms a l 24 fundamental basis behind modifications we will be making to 25 comply with the ATWS rule, i
Heritage Reporting Corporation (202) 628-4888 l
1
6 1 As we went through this series of calculations and 2 looking at events, you can see that there was a narrowing down ,
3 and a focusing in on the transients that were of particular 4 interest to the B&W machines. Most notably the singly largest 5 transient for ATWS in B&W's loss of feedwater event, you can .
l ,
6 see that we have tracked that one all the way across.
l 7 Loss of offsite power event is also of interest in 8 that it does degenerate to a loss of main feedwater event. So, 9 as we go through the final resolution loss of feedwater with ;
10 secondary concern of loss of offsite power is primary events 11 that we are concerned with.
12 In looking back over the years and the calculations l 13 that have been done, what we see is that for these eight ,
14 anticipated transients we are looking at calculaced projected 15 peak pressures, less than 2750 pounds.
Only t).e loss of l 1
16 feedwater event and the loss of offsite power event predict 17 peak pressures in an area which causes concerns from the ATWS j 18 standpoint; therefore we concentrated our attention on those i i
19 events. i i
20 In 1979, the report published in early 1980 our 21 i BAW-1610 was the calculations that formed the basis for the {
22 current way that we are looking at ATWS. In that series of 23 calculations we assumed a typical B&W-177 plant, taking 24 parameters that were not necessarily bounding, but more 25 representative of the plants.
(
Heritage Reporting Corporation (202) 628-4888
r c
7 1 We looked at loss of feedwater, loss of offsite 2 power, and two reactor coolant coastdown. There are some major 3 assumptions here; assumption of 95 percentile MTC. In this 4 case I will bring this up because we will be talking about it 5 later.
6 In 1979 we only had one plant that had moved from a 7 12 month fuel cycle to an 18 month fuel cycle. The MTC that we 8 picked 95 percentile MTC was based on the 18 month fuel cycle.
9 So, to some degree this parameter was picked as limiting our 10 bounding parameter at that point in time.
11 No single failures, and everything functioned. And 12 that's the general basis of the calculations that we did.
13 Again, not intended to be bounding or limiting, but 14 representative of the various plants.
15 In September 1981 we decided to perform a series of 16 plant specific analysis. One, to look at the effect of some l 17 minor changes in the assumptions. Two, to also validate that 18 the generic analysis that we had done in 1610 was indeed 19 representative of the various plants.
20 There were a couple of changes. We used in this case 21 a 90 percentile MTC which in 1981 was the MTC value that we i, 22 were using, added a one PCM penalty for rod maneuvering, and 23 changed some assumptions with regard to when we would admit i i
24 auxiliary feedwater into the generators.
25 The result of this was, in order to compare the plant l
Heritage Reporting Corporation l (202) 628-4888 l 1
8 1 specific analysis to the generic analysis in BAW-1610 one had 2 to add 300 pounds to the BAW-1610 analysis so that there could 3 be a comparison apples to apples between the two sets of 4 calculations.
5 With that correction made to the 1610 analysis --
6 MR. WARD: Where does the 300 pounds come from?
7 MR. ENOS: In 1610, in order to correct the 1610 8 analysis for this change and assumptions -- these changes and 9 assumptions between the plant specific analysis, the 10 combination of this change right here resulted in pressures 11 being about 300 pounds higher on the plant specific analysis.
12 Therefore, to go back and compare apples to apples to the older 13 generic analysis we had to add 300 pounds to the previous 14 results to accommodate the changes and assumptions on the plant 15 specific analysis.
16 so, it's a correction factor to accommodate changes 17 and assumptions between the two sets of analysis.
18 MR. KERR What does "B" mean?
19 MR. ENOS: I'm sorry?
20 MR. KERR: What does "B" under major assumptions 21 mean, AFW added per SG le'rel demand?
22 MR. ENOS: In the original analysis that we did we .
23 essumed a specific fixed point in time of which auxiliary 24 feedwater was injected in the steam generators. In the plant 25 specific analysis we chose the ta.me at which auxiliary Heritage Reporting Corporation (202) 628-4888
9
(-
1 feedwater was available based on the tech specs, and then we 2 allowed plant parameters to demand when auxiliary feedwater 3 would be injected into the steam generators.
4 So, in the first analysis we forced it in every time 5 at a specific time. The second analysis we waited until we got 6 a steam generator low level and let the plant automatically 7 demand it at that point. So it resulted in a somewhat delayed 8 auxiliary feedwater injection as compared to the generic 9 analysis.
10 MR. EBERSOLE: May I ask a question?
11 MR. ENOS: Certainly.
12 MR. EBERSOLE: Loss of main feedwater that occurs 13 when you lose AC turbine power either from loss of offsite or 14 failure to transfer; correct? If you have motor driven main 15 feedwater pumps. <
16 MR. ENOS: In our case we have all turbine driven, 17 steam driven main feeds, a 18 MR. EBERSOLE: All of your B&W plants have turbine 19 driven mains?
20 MR. ENOS: Yes.
21 MR. EBERSOLE: But they, however, are dependent on 22 the condenser. They are lost anyway.
23 MR. ENOSt They're dependent on condenser pumps, yes,
, 24 MR. EBERSOLE: So I'm saying, there's a cascade 25 there, so they're lost if you lose power right away. Now, I'm Heritage Reporting Corporation (202) 628-4888
10 1 trying to get some notion of the frequency of this. So you 2 lose main feedwater when you lose main AC power.
3 MR. ENOS: Correct.
4 MR. EBERSOLE: The electric driven feedwater pumps, 5 you generally have two of them and one turbine pump; right?
6 Aux feeds.
7 MR. ENOS: It varies from one motor, one turbine.
8 Some plants have one motors, two turbines. Some plants have 9 two motors, one turbine.
10 MR. EBERSOLE: So anyway, I'm getting to the point 11 you lose the motor driven pump. So you ride on the hope that 12 the turbine driven pump _will pick up and go.
13 MR. ENOS: Yes.
{
14 MR. EBERSOLE: What's the reliability standing on aux 15 feed steam turbine pumps startups and runs?
16 MR. ENOS: I don't exactly have those numbers. I
. 17 haven't looked at them recently. I know that over the past 18 years we have had problems with overspeed problems with terri 19 turbines.
20 MR. EBERSOLE: My impression is, all turbine driven 21 aux feed pumps have a question mark on reliability. !
22 MR. ENOS: To some degree I agree with you there. ,
i 23 However, over the last two or three years I know that a lot of l 24 work has been done in the overspeed area. ,
1 25 MR. EBERSOLE: You could tell me how many dollars you 1
(_ :
Heritage Reporting Corporation (202) 628-4888 ;
,1 0
11 1 spent, but it wouldn't erase it.
2 MR. ENOS: Well, I can also discuss it in performance 3 as well. I know some plants that really have had problems with 4 that in the past are not having problems with it at all now.
5 That failure rate I know on certain plants has been taken to 6 essentially zero from an overspeed standpoint.
7 MR. EBERSOLE: Well, I only want to call to the 8 subcommittee's attention the color of things, that's all; I 1
9 think I have done that.
10 MR. ENOS: I understand. With regard to the main l 1 :
l 11 feedwater question that you bring up, a little bit later I have 1 l
l i 12 some numbers for the Owners Group as far as loss of feedwater
, 13 events per year and upsets, so we can look specifically at l l :
l 14 that. j 15 MR. EBERSOLE: That would include the AC power I
16 failure cases of loss of feedwater, wouldn't it? )
17 MR. ENOS: Yes, sir.
18 MR. EBERSOLE: Thank you, j 1 19 MR. ENOS: So again, before I go forward, the 300 l 20 pounds that we're adding to the original generic analysis is 21 just an adjustment to accommodate a difference in assumptions l
l . 22 between the two sets of analysis.
23 What we see when we plot those together is, 24 fortunately, what we hope to see initially was that the generic 25 analysis that we had performed was indeed representative of the Heritage Reporting Corporation (202) 628-4888
i 12 1 plants. You see the generic peak pressure plotted here with 2 three plants and a small scatter on each side of that.
3 What they told us was that the generic representative ;
4 analysis was good analysis for us to use in doing further work 5 in looking at other things without having to be terribly 6 concerned about going back and doing things on a plant specific 7 basis. So it was a validation effort for us that our generic 8 analysis was indeed in the ball park.
i 9 Yes, sir, 10 DR. SHEWMON: Was this trend at the time of the i 11 commissioning of the plants or what?
12 MR. ENOS: This was based on actual plant !
13 configuration in 1980. So MTC's that would have been used 9
14 would have been 90 percentile MTC's based on the operating '
15 cycle at that point in time; and adjustments in there for !
16 relief valve capacity size differences between the two plants, :
17 differenoes in auxiliary feedwater setups, et cetera. !
18 MR. TAYLOR: Jim Taylor. The answer to your question 19 is no. ,
20 MR. KERR Jim, the reporter is missing your words 21 and we don't want her to. !
I 22 MR. TAYLOR: Jim Taylor from B&W. The answer to Dr. . ;
23 Shewmon's question is, the trend there has nothing to do with )
24 the order of commissioning of the plants. The Oconee plant was )
25 commissioned first, TMI second and so on. I think it was just I
Heritage Reporting Corporation (202) 628-4888 m
d 13 f
1 the way in which they were put up there on the chart. f 2 DR. SHEWMON: I appreciate that. But then this has
- 3 to do more with how the operators currently had them 4 configured?
5 MR. TAYLOR: Fuel cycle-wise, yes.
6 MR. ENOS: Fuel cycle and physical plant differences l 7 such as relief capacity, differences in PROV'S, et cetera.
8 Again, those differences being in 1980. These calculations 9 were done in that time frame -- or 1981.
10 MR. WARD: I'm still confused about the 300 PSI. Did 11 you use that in some way on the figures you just showed us 12 here?
- j 13 MR. ENOS
- Yes, sir, i
14 MR. WARD: What did you do?
15 MR. ENOS: What we did there was, from the generic ;
16 analysis, the number that was calculated on the generic It should have been plotted i 17 analysis was about 3400 pounds.
18 here. We added 300 pounds to it to make it 3764. And this 19 value that's plotted for the 1610 analysis is the corrected 20 value or the adjusted value to make the assumptions consistent.
21 MR. WARD: And you did that for each of the plants?
I 22 MR. ENOS: No, sir. The difference in assumptions, 23 there was one set of assumptions used for this calculation.
24 There was a different set of assumptions used for this 25 calculations. And the 300 pound adjustment was simply to i,
Heritage Reporting Corporation (202) 628-4888 4
14 1 accommodati difference in assumptionc.between the two sets 2 of analysis and make them directly comparable.
3 MR. WARD: You just brought the generic one up to be 4 consistent with the other plants.
5 MR. ENOS: That's correct. If we had not had done 6 that and had plotted raw values, we would have saw a generic 7 number down here and you would really be comparing apples and 8 oranges and drawing a wrong conclusion.
9 MR. WARD: Well, you just would have put it over on 10 the other end and everything --
11 (Laughter) 12 MR. ENOS: We put it down here so we have a nice 4
13 slope there.
14 MR. EBERSOLE: Before you take that down, are you 15 going to review the mechanical effects of these pressures 16 during this discussion?
17 MR. ENOS: I had not specifically planned on doing 18 that, although I will address the mechanical effects. We 19 performed back in the '79 '80 time frame stress calculations on 20 major RCS components, valves, piping, et cetera, and showed 21 that we would not expect any failures of component structures j 22 up to around 4,000 pounds, somewhere in that neighborhood. .
, j 23 From that standpoint that's the work that we have done to look j 24 at things mechanically, pressure withstanding capability, 25 MR. EBERSOLE: Some plant lost both of its "O" rings i l
Heritage Reporting Corporation (202) 628-4888 '
l 15 (
1 recently; those "O" rings wouldn't leak.
2 MR. ENOS: Yes, we would certainly predict that ;
3 somewhere around the neighborhood of 33, 3400 pounds we-would 4 expect to see leakage through the "O" rings on the head.
l 5 MR. EBERSOLE: Is there any possibility of that 6 cutting the head bolts?
7 MR. ENOS: In the looks that we have done at that, I l 8 mean, we have done some looking at what we would expect to 9 happen there, we don't see that as a significant concern.
10 MR. EBERSOLE: Water cutting is a popular mechanical 11 process nowadays, what does it take about 10,000 pounds, I ,
12 think. ,
i j 13 MR. ENOS: Something like that, right. j 14 In the case of leakage through the head seals we :
i 15 would expect that to start, thinking realistic world now and 16 getting out theoretical calculations, looking at bolt j E
17 elongations and all that, we would expect that to start ;
I !
18 somewhere in the neighborhood of 3300, 3400 pounds; and would l i
19 expect that peak reactor coolant system pressures probably !
l 20 would not exceed something like 38, 3900 pounds, with that l 21 relieving mechanism available. ,
22 MR. EBERSOLE: Would some pumps be dead-ended at this !
23 pressure and be damaged trying to put water in but unable to do 24 so? !
I t 25 MR. ENOS: The main pump we would be concerned with j i
Heritage Reporting Corporation [
(202) 628-4888 !
_ _ - _ _ _ __ - , . . . -m_,,___.,~ s _, _
l l
i l 16 I at that point would be the HPI pumps. All the B&W plants say, ,
i 2 one have the high head pumps which as 1 recall is about 5300 i
! 3 feet a head, which comes back and do my ments1 arithmetic, what i I 4 is that.
1 5 MR. EBERSOLE: Oh, about 25. !
6 MR. ENOS: Somewhere around in there. So we would l
I 7 probably expect during that region some dead heading of pumps, 8 those pumps are equipped with many recirs, so we should not j 9 have a problem with that dead head situation.
10 MR. EBERSOLE: Thank you.
11 MR. ENOS: I'm going to step off now for just a l
I l '
12 moment into three drawings and we will try not to spend a lot g
13 of time on these. One thing that I want to refresh memories 14 on, I know we have all looked at this before but it has been 15 some amount of time. There is the rather unique feature of the 16 B&W reactor trip system and that is the use of the silicon 17 controlled rectifiers for controlling and tripping the 18 regulating rods.
19 Within the B&W reactor trip system there are seven 20 groups of rods that we deal with, groups one through four which 21 are considered to be our safety rods; and those are rods that 22 are necessary, gets sufficient aid of reactivity to meet our ,
l 23 limiting licensing safety analysis, et cetera, l
I 24 We have groups five, six and seven. Eight is not 25 normally used, which are regulating rods or rods we use if we
(
1eritage Reporting Corporation (202) 628-4888
i 17 1 run a rodded operation or for control during power.
- 2 From the ATWS standpoint these regulating rods are ,
t 3 more than sufficient to mitigate an ATWS event. And we have to l i
4 have the safety rods for licensing type of events, but from an
- 5 ATWS standpoint the regulating rods are more than sufficient. ;
6 So this is the rods that we credit in mitigating, determining l 7 ATWS events looking at the diverse scram system. j 8 In the overall trip system we have two divisions of ,
t l 9 main power coming in, cross connected at the bottom so that ,
10 either division of power is capable of holding the rods.
i i
11 Through the first division of power we have one AC 12 breaker, which then goes through a power supply to a pair of DC L
i 13 breakers. The same thing here, one AC breaker and two DC [
4 { 1 i
14 breakers. So as I said, if we open -- it takes both of these
, 15 breakers or some combination across, both paths have to be !
l 16 interrupted. So this breaker and these two will cause a full !
l
]' 17 trip. This breaker and this breaker will cause a full trip. l i
18 These two breakers will cause a full trip. These two breakers !
- 19 down here by themselves will only trip the safety rods.
) !
j 20 Now, I'll explain a little further. Coming off here 21 is the power to the regulating rods.
4 .
> . 22 MR. KERR Excuse me, before you leave that. What !
l 23 happens if you lose offsite power to this system on the left? !
24 MR. ENOS: If you loss offsite power it will trip. l l
25 MR. KERRt But in your analysis you show loss of :
i r l Heritage Reporting Corporation )
q (202) 628-4888 ;
- i 1 l i
=
,p 18 1 offsite power as a contributive to RPS.
2 MR. ENOS: That's correct. Loss of offsite power 3 would be the anticipated transient. And although, as the plant 4 ~ currently exists, we would get a reactor trip directly-off of 5 loss of offsite power. In ATWS we assume, for some reason or 6 another, that trip does not occur, that the RPS fails to 7 produce that reactor trip. Therefore, we would rely on the new 8 diverse scram system to insert the regulating rods for us to 9 mitigate that.
10 MR. KERR But independently what you're assuming, 11 the analysis, electro-mechanically, if you lost offsite power 12 you would get a trip?
13 MR. ENOS: You would get a trip if we lose -- real 14 world, the way the plant is actually configured we get a trip 15 when we lose offsite power. In ATWS we assume that there is 16 some common mode electrical failure that prevents the RPS, for 17 some reason undefined, prevents us from getting that direct 18 reactor trip; and we have to generate that by use of the DSS 19 when we're doing ATHS calculations.
20 MR. KERR: Thank you.
21 MR. ENOS: Thank you.
22 Coming o of the main powers, again, two' divisions .
23 of power which co.'ae over to the regulating rods, that power is 24 passed through silicon controlled rectifiers; again, we have a 25 two channel arrangement so that either channel is capable of Heritage Reporting Corporation (202) 628-4888
l 19 l
~
1 -holding the rods.
2 When we -- that power coming from here. When we get ,
3- a reactor trip, either from the RPS or from the manual trip.
~
4 button, when'these breakers open here, this power supply is 5 interrupted, and therefore we lose holding power also to the 6 reg rods.
7 As a backup to that, when the RPS or the manual. trip 8 generates a reactor-trip and these two breakers open, there are 9 auxiliary contacts which we call our E&F contacts off of these 10 breakers which interrupt power to the gate of the SCR's, 11 therefore effectively degating the SCR's and allowing the rods 12 to drop.
13 So from a normal configuration of what we have, if 14 the reactor trip works. properly, both from automatic or manual, 15 we get a trip of those breakers; a trip of those breakers. We :
16 interrupt this power supply through the SCR's as a result of ;
17 opening those breakers. When these breakers open the auxiliary
, 18 contacts degate the SCR's. So we go at the SCR's from two 19 different ways; we interrupt the power through and also-the 20 gates to the SCR's.
21 For the diverse scram system that we'll be installing
. 22 for ATWS, what we will be doing is generating yet'another trip
, 23 signal and inserting a pair of contacts in series with these 24 existing EP contacts. The new diverse scram system then will 25 drop groups five, six and seven by interrupting power to the Heritage Reporting Corporation (202) 628-4888
l 20 1 gates of SCR's. We will therefore have complete diversity in 2 the tripping mechanism and will be using completely SCR's to 3 get rod drop as opposed to reactor trip breakers.
4 MR. EBERSOLE: Will those be breakers you'll put --
5 MR. ENOS: Contact -- that little contact.
6 MR. EBERSOLE: How much power do they handle?
7 MR. ENOS: Let me ask Howard Stevens from B&W, do you 8 know the current through that gate circuit.
9 MR. STEVENS: The gating circuit is at very low 10 level. The actual cut off by the gate drive is done at the 11 ampere level. So the relays that they're talking about adding 12 as f ar as ATWS is concerned are relays the t are dealing in the 13 million ampere.
4
-14 MR. EBERSOLE: What's the ampere level?
15 MR. STEVENS: I would say less than 100 million 16 amperes.
17 MR. KERR: Did you get his name?
18 MR. STEVENS: My name is Howard Stevens and I'm with 19 Babcock and Wilcox.
20 MR. EBERSOLE: I can't remember how this system l l
21 handles switching transients on the AC input side, how does it 22 do that? . l 23 MR. ENOS: Switching transients?
24 MR. EBERSOLE: Yes.
25 MR. ENOS: I'm sorry, you're talking --
l Heritage Reporting Corporation (202) 628-4888 1
i
21 1 MR. EBERSOLE: If the_AC input up there where the 2 controller drop system -- you call them AC main BUSS.
3 MR. ENOS: Here, where --
4 MR. EBERSOLE: Isn't that system subject to voltage 5 fluctuations? Do you have time delays of any kind?
6 MR. STEVENS: I can answer that question this way.
7 Since it requires both -- since both -- either one of the two 8 power supplies will maintain the reactor in the untripped 9 stage, you can take switching transients on either one of 10 the --
11 MR. EBERSOLE: But not both.
12 MR. STEVENS: -- powers without a problem. If you
(
13 get simultaneous switching transients on both power supplies of L
14 sufficient duration, and I suspect it's in the order of a few !
15 milliseconds, yes, you will get a trip.
16 MR. EBERSOLE: You just accept that uncommon. You 17 know, lots of them have rotator machines --
18 MR. STEVENS: Yes. Most of our plants do not have 19 that.
20 MR. EBERSOLE: Thank you.
21 MR. ENOS: The purpose of going through this was, 22 again, to review our use of SCR's and how we intend to use 23 SCR's for the diverse scram system. Again, the relays will go 24 here which will interrupt power to the gate drives.
25 MR. WARD: Before you leave there, just to give us I.
Heritage Reporting Corporation 1 (202) 628-4888 j
l 1
22
, l 1 some perspective. What are the worth of these rod groups in 2 terms of -- do you have some number?
3 MR. ENOS: I don't have a number. Do you have one, 4 Randy?
5 MR. ELLISON: My name is Randy Ellison from B&W. The 6 worth is in the one and a half percent to two percent range, it 7 varies with life.
8 MR. WARD: That's for which?
9 MR. ELLISON: Five, six and seven -- our calculations 10 only assume it's worth of about .7 on two of the groups sal 11 to show adequate margins during the plant specific analysis 12 done in the '81 time frame. Maybe I didn't answer tne right 13 question.
14 MR. WARD: You may have, I just probably don't 15 understand it. What's the worth of -- you've got eight groups 16 up there and are they all equal as far as reactivity worth?
17 MR. ELLISON: No.
Group is our aux power shaping 18 rods, so those wouldn't --
19 MR. WARD: Oh, eight is aux, okay. And the others 20 are worth what?
21 MR. ELLISON: Total worth for all those?
22 MR. WARD: Yes. .
23 MR. LOJEK: I'd like to answer that. My name is Jan 24 Lojek of Babcock and Wilcox. The question is, the worth of 25 five, six and seven separately or together?
(- ,
Heritage Reporting Corporation (202) 628-4888 l
l 1
23 1 MR. WARD: All of those.
2 MR. LOJEK: A typical worth of a group is about one 3 percent reactivity. Five, six and seven typically are worth 4 around three percent, three and a half percent reactivity.
5 MR. EBERSOLE: Every time I see SCR's I think about 6 Hanford and spiking of the rectifiers that caused the balls to 7 have to fall in. Have you done analyses on spiking and voltage 8 on these SCR's and what happens when you get it or if you get 9 it, the whole package?
10 MR. ENOS: I can't say specifically that we have done 11 that. Of course, the SCR's is something that we have used in 12 the plants now since the early '70s. The failure rate of those 13 SCR's has been something on the order of one or two in 10 f
14 years. We have experienced almost no failures of them at all.
15 They're in there in banks.
16 MR. EBERSOLE: That was the case of the Hanford ;
I 17 incident.
18 MR. ENOS: I understand. A single SCR failure 19 doesn't necessarily take us out, because they're in banks and 20 the minimum amount -- ,
. 22 do you have some sort of catastrophe on the back side or what 23 happens -- do you burn things up?
24 MR. ENOS: Not that I'm aware of. The failures that :
25 we've had have been what appear to be shorts to open, fell
. l Heritage Reporting Corporation (202) 628-4888
l l
24 l 1 short and burn open, and we have not had any equipment problems 2 or damages as a result of that. i i
3 MR. EBERSOLE: Thank you.
~
4 MR. LIPINSKI Let me generalize the question. If 5 you're going to two out of two system to avoid various trips 6 which reduces scram reliability, have you done a failure modes 7 and effects analysis in a numerical _ reliability of the system 8 to see what the probability is of the figures, the scram to the l
l 9 component figures and spiking, numerical figures.
10 MR. ENOS: Are you talking now about the existing RPS 11 or the new diverse scram system?
12 MR. LIPINSKI: Both, because you're going to add
, 13 additional components on the same equipment here. So you've 1
14 got a reference case where you had a reliability and now you're l 15 saying you're adding diverse scrams, what is the increase on 16 reliability. You need some reference base for comparison.
l 17 MR. ENOS: The only thing we are adding to the 18 existing RPS, which is the four channel safety grade fell safe ,
l 19 system is relay contacts -- the relay contacts in the SCR gate i l \
20 circuits. The design intent of DSS is to increase the 21 reliability of this system by a factor of 10. And that's the 22 two channel system that we propose. The two channel energize .
23 the trip system which would drive those relays. What we're l 24 looking at there is a factor of 10 improvement in an overall 1
25 reliability of the RPS. l Heritage Reporting Corporation (202) 628-4888 1
l
25 1 MR. LIPINSKI: You have done a numerical analysis of 2 the system before that?
3 MR. ENOS: No.
4 MR. LIPINSKI: Well, how would you demonstrate you 5 got the factor 107 o
6 MR. ENOS: The basis of the ATWS rule itself was 7 that, if one installs a diverse scram system as defined by the 8 ATWS rule, then one has achieved the factor of 10 improvement.
9 It was not required as a component of the rule for us to 10 validate that, that was something that was determined in 11 advance.
12 MR. KERR: Mr. Enos, I must say that response puzzles 13 me. I would think that the owner / operator of these plants
(
14 would be interested in the reliability of the scram system.
15 Now, perhaps you have convinced yc.urself that it's reliable 16 enough and that this added system is only to satisfy regulatory 17 requirements.
18 But if that isn't the case, it seems to me you would 19 want to know what effect on reliability this would have. I 20 believe if I owned a powerplant I would. Since I never owned 21 one, I can't speak from personal experience. I don't
. 22 understand an answer that says, the NRC assumes it will add a l l
23 factor of 10 and so it does.
24 MR. ENOS: I'm sorry, I understand your point. In 25 the development of the basis for the ATWS rule, during the
! I i
Heritage Reporting Corporation i (202) 628-4888 l l
! 26 1 years '82, '83, there was at that point in time an industry 2
utility group which was working with the-NRC in developing the 3 basis for the rule, et cetera.
There was a substantial amount 4
of PRA work that was done by the industry and submitted to the
- 5 NRC looking at the entire aspect or the entire element of the 6 ATWS arena.
7 What we showed in those PRA studies, which albeit 8 they were generic, but the PRA type work is that a diverse 9
scram system with a minimal set of requirements such as was 10 established by the rule, non-safety grade, single channel, 11 those types of requirements, would more than satisfy the design 12 goal of reducing the reliability by a factor of 10. And that's 13 backed by PRA type analysis.
34 MR. KERR I don't think you want the racord to show 15 that you came to reduce your reliability by a factor of 10, 16 which I think is what you just said.
17 MR. KERR You're absolutely correct. Improve the 18 reliability by a factor of 10; reduce the unreliability. That 19 was shown by PRA analysis and on a generic basis that, indeed, 20 that type of system bid, in the design that we have chosen 21 which we will look at on the nexc slide, we have put design 22 elements in these specific B&W DSS that takes that system more l
23 reliable than the one that was included in the original PRA 24 work that was done.
25 Are we, at this point in time, going to go back and
(..
Heritage Reporting Corporation (202) 628-4888
I I
27
. i 1 do another analysis? Our intent is not to do that. We are 2 satisfied on the background work that we have done.that this 3 system has a reliability or an unreliability significantly 4 beyond 10 to the minus one.
5 MR. LIPINSKI: May I interject another comment.
6 There is something missing in that because the rule was not 7 prescriptive in terms of the components that you used.
8 Whereas, depending upon the quality and reliability of.the 9 components, that will determine the element reliability. You 10 add components with a tremendous figure rate, you will not 11 achieve an improvement of a factor of 10, 12 MR. TAYLOR: Dr. Kerr, could I make a suggestion. I 13 want to try and put what Mr. Enos is saying in perspective and 14 hopefully get us back on track. We came down here to talk .
15 about the effects of the fuel cycle changes on previous ATWS 16 calculations, and Mr. Enos' purpose in going through this 17 background was to try to put that in a little bit larger 18 context and show some of the other things that have happened.
19 And everything that has been said so far is just in the way of 20 background. It was not to go back and revisit the ATWS rules.
1 21 I'd like to suggest we try to get back and let us get j 22 to the point which was the purpose of the meeting and that was 23 to show the effects of fuel cycle changes on the ATWS l
24 calculations. l l
25 MR. KERR: Mr. Taylor, I expect that we are both j l
f Heritage Reporting Corporation (202) 628-4888 1
28 1 interested in the same thing which is the likelihood that we 2 will be able to prevent an ATWS from occurring. And in order 3 to put the effect of fuel cycle in perspective it seems to me 4 one needs to have, as Mr. Enos indicated, some background in 5 where one is without making these changes.
6 And hence, I think some exploration of reactor 7 protection system reliability is appropriate. I don't think'we 8 should spend all of our time on that, but I don't think it's 9 inappropriate to spend some time on it.
10 Did you complete your question?
11 MR. LIPINSKI: Yes.
12 MR. SBERSOLE: Can I ask a question. I see all of 13 the trip channel outputs, A-B, C-D, feeding into UV trips on a 14 variety of breakers; correct?
15 MR. ENOS: Yes, sir, they feed under-voltage trips 16 and also shunt trips.
. 17 MR. EBERSOLEt I don't see tire shunt trips.
18 MR. ENOS: This is a figure copied out of NUREG 1000, 19 which was after the Salem event, and at that point in time we 20 did not have the shunt trips in the system; those have 21 subsequently been added.
22 MR. EBERSOLE: So the shunt trips are not shown. .
23 MR. ENOS: That's correct. They are there.
24 MR. EBERSOLE: Could you tell me what the voltage 25 incurred is on the DC trip breakers?
(
Heritage Reporting Corporation (202) 628-4888
29 1 MR. ENOS: Is that 400 amps?
j 2 MR. STEVENS: Are you talking on the load side?
3 MR. EBERSOLE: Yes, sir, the last one before you get 4 to the rods.
5 MR. STEVENS: It's over 200, but less than 400.
6 MR. EBERSOLE: And what is the voltage?
8 MR. EBERSOLE: Thank you.
9 MR. KERR: Mr. Ward.
10 MR. WARD: Just a quick question, again, background 11 for me I guess. Are the seven rod groups enough -- have enough 12 reactivity worth to take the reactor to cold shutdown hold it 13 or do you need boration in these plants?
14 MR. ENOS: You need boration.
15 The next slide.
16 MR. LEE: I don't understand your re,sponse, you said 17 the seven -- the rod banks may be worth one percent in 18 reactivity, so you're talking about seven, eight percent 19 reactivity; that could be more than enough to bring down the 20 reactor to cold shutdown condition from any condition 21 whatsoever.
22 MR. ENOS: I'm sorry, you're correct; I misspoke 23 that. When I think of going to cold shutdown I think about it 24 in terms of maintaining five percent delta K over K margin.
25 And you're correct, we are required to borate to get our Heritage Reporting Corporation 1 (202) 628-4888 l
30 1
1 margins that we need cold shutdown.
2 MR. LEE: Thank you.
3 MR. ENOS: I'm sorry. Thank you.
4 MR. WARD: So, boration is just to get them to 5 margin.
6 MR. ENOS: To margin; yes, sir.
7 The next two slides I wanted to just look at quickly 8 as far as what conceptionally or functionally we have put 9 together for the diverse scram system and the AMSAC system.
10 You will see the systems look very similar. They're two 11 channel systems. Two out of two, energized to trip, fully 12 testable on line. When we test one channel we'll deactivate
( 13 the second channel. During that portion of the test it
\.
14 generates an output signal which will drive the relays in the 15 gate circuits of the SCR's and trip routes five, six and seven.
16 It's a relatively simple straightforward two channel system for 17 DSS.
18 The input signal to that will be either high reactor 19 cooling system pressure and/or loss of main feedwater.
20 MR. KERR: Mr. Enos, has anybody in the Owners Group
\
21 been curious about the possibility that this new system may I 22 increase a false scram grade? ,
23 MR. ENOS: Yes, sir.
24 MR. KERR Have you done any calculations that would l l
25 give you some idea of whether and how much? i
(
Heritage Reporting Corporation (202) 628-4888 l l
4
31 1 MR. ENOS: 'the main thing that we did -- of course, 2 any time you put a new system in that can trip the machine you 3 have a concern about spurious trips of that machine.
4 The main two things that we did to address that was 5 from the design standpoint, that being a two out of two system 6 that energized the trip system; and on testing capability we 7 don't take it to a one out of one system during test, we 8 deenergize the entire system while we are doing the test 9 itself.
10 So, from that etandpoint, those are the major things 11 that we have done to improve that reliability from the 12 standpoint of getting an actual trip out of it or a false trip.
13 The types of components and things that we'll be 14 using in there, for the most part, will be components that we 15 have familiarity with out of either the existing -- in a lot of 16 cases we'll be using components out of our new EFIC, emergency 17 feedwater instrumentation and control system. So they're 18 components that we have familiarity with from other designed 19 uses in the plant, have some idea of what their failure rates 20 are and hopefully they're quite small.
21 MR. KERR So in response to my direct question, no, q
)
22 you have not tried to calculate the increase, but you have I 23 tried to minimize or to make it small by design and operational 24 practice.
25 MR. ENOS: That is correct, sir; yes, sir.
t 1:eritage Reporting Corporation (202) 628-4888
32 l
1 MR. KERR: Thank you.
2 MR. EBERSOLE: Did you say the two new input signals 3 is high pressure and loss of feedwater?
4 MR. ENOS: It can be either high pressure; it can be 5 loss of main feedwater; or it can be both.
6 MR. EBERSOLE: Loss of feedwater is anticipatory.
7 MR. ENOS: That is correct. i 8 MR. EBERSOLE: So you really execute in that trip for f 9 loss of feedwater. I i
10 MR. ENOS: From the standpoint of the tripping 11 system, yes, that can be done. High pressure is probably the 12 better signal. !
13 MR. EBERSOLE: What's the present pressure trip?
14 MR. ENOS: High pressure? f 15 MR. EBERSOLE: No, the one you have now? i 16 MR. ENOS: On the existing RPS?
i 17 MR. EBERSOLE: Yes.
l 18 MR. ENOS: High RCS pressure? f 19 MR. EBERSOLE: Is that what it is? ;
20 MR. ENOS: 2400 -- 2355.
i 21 MR. EBERSOLE: So it's already slightly higher than t 22 normal? *
.I i
23 MR. ENOS: Right. ;
1 24 MR. EBERSOLE: Are you going to have another one l 1
25 higher than that? !
l I ,
Heritage Reporting Corporation ,
(202) 628-4888 j
l l
l I
33 1 MR. ENOS: We would have another one higher than l l
2 that. And if necessary, we.might put a small time delay on it.
l 3 The design intent is that RPS would always have its opportunity l l
4 to trip the reactor first. So the set points would be set such j 5 that we would give RPS that chance. !
I 6 MR. EBERSOLE: So you never have to say you had an ;
7 ATWS trip.
8 MR. ENOS: Right, unless we really did. That would i
9 hopefully be the result. I 10 Moving to the next slide which is similar to that one 11 which is the AMSAC system. The same type of design concept, 12 two channel, two out of two, energized to trip. On line )
1 13 testable. There are some enables in there for low power, here. l 14 In this case we're looking at a loss of feedwater as the input i 1
I 15 signal.
16 The output of this is an auto actuation of emergency 17 feedwater and a turbine trip. 1 1
18 That being where we are at this point in time in the 19 analysis that we have done, the design work that we have done, ,
l 20 and the systems that we intend to place in the plant, I wanted l 21 to look ahead at what has happened since the analysis work was
. 22 done in '79 and '80, that in 1988 may potentially impact where I
23 we are in relative to ATWS and the use of the ATWS regulation 24 itself.
25 The specific questions I think we have been asked 1
Heritage Reporting Corporation (202) 628-4888 J
I l
34 1 prior to this meeting dealt with changes in MTC and fuel design 2 and core loads. We want to address those question. I would 3 like to start first by backing up and taking a little bit 4 bigger picture look at the thing, and across the board,.what 5 are things that have happened in the last eight years that 6 impact where we are in the ATWS arena relative to risk.
7 In order to do that, looking at the basic elements of 8 the ATWS rule itself, the overall purpose of the ATWS rule was 9 to reduce the ATWS risk by a factor of 10. The rule had two 10 principal elements to its improve the reactor trip system 11 reliability by a factor of 10 which was to be accomplished by 12 the diverse scram system; unacceptable consequences or that is,
{ 13 pressures less than service level C for greater than 50 percent r
14 of the time, that to be accomplished by the AMSAC cystem.
15 There was another element, not to the rule itself, 16 but in the statements of consideration to the rule which was a 17 reliability program; it was recommended that the utilities 18 undertake, to look at a variety of items.
19 MR. LEE: Can you give us some idea what the pressure 20 would be corresponding to the service level C?
21 MR. ENOS: Roughly 3200 pounds.
22 MR. LEE: Thank you. .
23 MR. ENOS: Looking first -- I would like to go 24 through first some of the reliability items that have taken 25 place in the last few years and the impacts that those
(
Heritage Reporting Corporation (202) 628-4888
35 1 potentially have on ATWS.
2 This particular slide, we have divided this into 3 transient reduction and trip reliability. We can look at this 4 kind of from the standpoint of these are things that attack the 5 AT part of an ATWS or the anticipated transient. These are 6 things that attack the WS part vi the ATWS or the without 7 scram.
8 Going through and looking, we have now completed a 9 PRA -- complete PRA look at the existing reactor protection 10 system to determine optimum and correct surveillance intervals; 11 to give us better -- what is the best surveillance interval to 12 get the best reliability out of the RPS.
13 So safety and performance improvement program that 14 the Owners Group has been undergoing fer about two years has 15 got a number of items, recommendations oriented towards that.
16 We've implemented a new root cause determination 17 program at the utilities; a generic program. We have 18 determined for doing root causes after trips and transients and )
l f
19 disseminating that information to the other utilities through I 20 the use of our transient assessment program.
21 If we go back and look at, how can we determine --
. 22 this stuff is nice, but how can we determine whether --
23 MR. KERR: Excuse me. You accomplish transient 24 reduction through PRA based RPS surveillance.
25 MR. ENOS: Yes, sir.
Heritage Reporting Corporation (202) 628-4888 i
1 36 1 MR. KERR: How does that cut down on the number of 2 transients?
3 MR. ENOS: Every time that we do a surveillance on 4 the reactor protection system, which currently we do one 5 channel of that RPS a week, we introduce the probability of a 6 spurious trip by the fact that someone in actually in the 7 cabinet.
8 MR. KERR: No, I understand that. If that's what you 9 mean, I understand the statement.
10 MR. ENOS: That's what we mean.
11 MR. KERR: So PRA based surveillance, you mean you i
12 have tried to balance the possibility of maintenance or testing -
g 13 induced trips against the increased reliability and changed by 14 the surveillance. :
i 15 MR. ENOS: All things considered, both positive and 16 negative aspects of surveillance, consider all of those things, 17 and what is the best most optimum interval to do that 18 surveillance test.
19 MR. KERR: What is an SPIP program?
20 MR. ENOS: That is the E&W Owners Group safety and i 21 performance improvement program which was initiated in early ;
22 1986 following the Rancho Seco event. ,-
23 MR. KERR: How much has that reduced the transient?
24 MR. ENOS: How much?
25 MR. KERR: Yes, sir.
(
Heritage Reporting Corporation (202) 628-4888 l
l l
37 l t
1 MR. ENOS: If you could bear with me just a minute, 2 'I'll.get.that. ,
3 This program has looked at about 15 to 17 major 4 systems in the plant. It has gone back over the history of B&W 5 plants since 1980 looking at every trip and transient that has 6 occurred in the B&W plants, trying to address what our root 7- causes of those and making recommendations to the owners as to 8 what equipment, procedures, programs, et cetera, need to be 9 corrected to prevent those type of things from occurring again.
10 Some 275 recommendations have now been forwarded to 11 the utilities as a result of that program. Essentially, all of 12 those dealing with either something tripped or transient 13 related in the performance of the plant.
7 14 If we look at performance indicators, if you would, 15 in this particular area, what has happened in the scram -
16 frequency at plants since we initially promulgated the ATWS 17 rule and the basis behind it. In 1981 when our 1610 analysis 18 was developed at the B&W plant, this is the collective body of 19 B&W plants, the scram rate was about five trips in '80 or five i
20 per plant; 6.2 per plant in '81. You see in 1987 the scram 21 frequency for B&W plants was two per plant. And our goal is j l
. 22 about 1.7 per plant in 1990.
I
- 23. Therefore the scram frequency now, as compared to the l l
24 time the PRA and supporting work was done for the ATWS rule for 25 B&W plants is more than a factor of two less; two as compared Heritage Reporting Corporation (202) 628-4898 l l
38 1 to five.
2 From the standpoint of upsets are significant, 3 transients or ATs, looking at loss of feedwater events for the 4 B&W plants over the same period of time, you saw in-1981 six 5 loss of feidwater events among the B&W plants. And 1986 and 6 1987 two; ti substantial reduction in the number of significant 7 anticipated transients as far as ATWS was concerned.
8 MR. EBERSOLE: Is that total?
9 MR. ENOS: Yes, sir.
10 MR. FBERSOLE: Of all the plants?
11 MR. ENOS: Yes, sir.
12 Looking again at the safety and performance 13 improvement program that I just mentioned earlier, of the 275 14 recommendations that have come out of that program and have 15 been forwarded to the owners, 23 percent approximately of those 16 recommendations have dealt specifically with main feedwater 17 reliability.
18 So although we have improved the main feedwater 19 reliability over the years, as we can see from that chart, 20 because some of the things have already been done and 21 implemented, the new control systems have been put in place at 22 various plants, we still have a number of items coming out of 23 the safety and performance improvement program yet to be done 24 which we believe will result in driving those numbers lower 25 yet.
(.
Heritage Reporting Corporation (202) 628-4888
l l .
l 39
' . Mk EBERSOLE: What are all those other acronyms 2 you've got up thers?
3 MR. EdOS: This is emergency feedwater, integrated
'
- This is specific 4 control system /non-nuclear instrumentation.
- 5 to, SSPC, I don't know what that one is. Secondary pressure.
6 Instrument air; and electrical, and then that's a general 7 category which covers procedures, programs, training, et 8 cetera, throughout the plant.
9 MR. EBERSOLE: I assume these are recommendations?
i 10 MR. ENOS: Yes, sir.
11 MR. EBERSOLE: That doesn't mean anything is 12 realized.
i 13 MR. ENOS: In some cases they are. About a third of l
14 those recommendations have now been implemented and a schedule j
.i 15 is being developed for the implementation of the rest of them. ,
16 The owners Group has just completed an audit internal !
17 to the owners Group where we send an audit team to each one of 4
18 the utilities to look at the program that they had in place for 19 implementing these recommendations and the schedules. And the a
20 report of that has been provided to each vice president in each 21 of the companies as far as where we are with regard to
. 22 implementing these particular items.
23 We have a second round of audits that has just now 24 started among the utilities to look at specific implementation 25 of the packages on a random basis to determine the timeliness
( l
. Heritage Reporting Corporation l l (202) 628-4888
- i l
40 1 and quality of implementation.
2 MR. EBERSOLE: What was IA again?
3 MR. ENOS: Instrument air.
4 MR. WARD: I'm sorry, I guess I missed it, those are 5 percentages of what?
6 MR. ENOSt- Percentages of about 275 recommendations.
7 MR. WARD: Individual recommendations.
8 MR. ENOS: Individual recommendations.
9 MR. WARD: Not dollar.
10 MR. ENOS: Nc. The rough number is 275 pushing its 11 way towards 300 I think at this point, and that is how they 12 break down categorically.
13 So we expect the message from that being that these 14 types of events have gotten better over the past few years.
15 Reactor trips have gotten fewer over the past few years. A 16 number of things are still in the pipeline which specifically ,
. 17 address trips and transients which we believe will further 18 drive those numbers down, and therefore continue to reduce the 19 ATWS risk. '
20 MR. EBERSOLE: The instrument air interface just with !
21 main feedwater?
l 22 MR. ENOS: No, sir, it's an entire instrument air .
23 system. We did a complete review of the instrument air 24 system, top to bottom. The same with -- it's nothing 25 specifically in relation to main feedwater. ICS/NNI top to i
Heritage Reporting Corporation (202) 628-4888 L
I l
41 l
1 bottom. l l
2 MR. EBERSOLE: Is there anything about the intrusion 3 of water and other things from the common design?
4 MR. ENOS:- Those things have been specifically looked 5 at, dryers and oil, the whole nine yards of the instrument air 6 system have been evaluated, all the way back to the original 7 TMI experience.
8 On the other side of that, those are things that we 9 have done to reduce the problem. Now, I'm not going to stand 10 here and tell you we have rushed out and done all of these 11 things just to push the ATWS risk lower; there's lots of other 12 reasons why these things have been done. They certainly do
,, 13 have an impact on pushing the ATWS risk lower.
14 Those thir.gs impacted from the standpoint if we don't 15 generate the need for a reactor trip following the loss of main 16 feedwater event, if we don't have a loss of main feedwater 17 event.
18 And the number of transients that are happening that 19 cause RPS the need to generate a trip are being fewer each 20 year. They're for the possibility of us initiating an ATWS in 21 the first place, has grown lower as the years have gone 1
22 through. l
, . \
23 The second half of this slide is, if we do initiate a 24 demand for a reactor trip, how reliable is the system to l
25 generate that reactor trip? What things have happened to make Heritage Reporting Corporation (202) 628-4888 1
42 1
1 that system more reliable?
2 We developed -- working with General Electric and B&W 3 we have-developed a generic maintenance and surveillance 4 procedure for the reactor trip breakers which all of the 5 individual utilities' procedures are based off of that generic 6 procedure, make sure that the RTB's have and are continuing to 7 get the proper and feeding based on what we learned in the '83, 8 '84 time frame. That was mentioned earlier, installed the 9 shunt trip.
10 We have across the B&W and Combustion Owners what we 11 call RTB on-line monitoring, and there is about four or five 12 plants that have installed the capability to monitor reactor 13 trip breaker response time in situ in the cubicle. Every time 14 a trip breaker is open for any reason, test, surveillance or 15 actual demand at these plants we would get a read out of what 16 the response time is. Therefore for the industry they can keep ,
17 tabs on anything happening to those breakers that is causing us 18 to have a concern and not let a Salem type problem slip up on 19 us again.
20 Generic trip breaker maintenance; generic RPS 21 surveillance and maintenance program, we have developed working 22 with B&W a generic program for doing post-maintenance testing ,
23 and maintenance and surveillance of the entire reactor 24 protection system.
25 Again, the safety and performance improvement program
(
Heritage Reporting Corporation s (202) 628-4888 l
43'
(
1 speaks to these components, root cause determination, transient '
2 assessment.
3 MR. EBERSOLE: In the transient reduction program, 4 your plants have a notably short time constant to trouble along
-5 feedwater flow level or parameter -- mixed parameter you use to 6 control it. Where did you find the biggest source of 7 improvement there, was it in the control valves?
8 MR. ENOS: On the feedwater system itself?
9 MR. EBERSOLE: Yes.
10 MR. ENOS: It's a combination of things. It's the ,
, 11 control system for the feedwater pumps themselves and the 12 control valves.
,e 13 MR. EBERSOLE: Well, the feedwater pumps themselves, 14 aren't they constant speed motors?
l 15 MR. ENOSt No. Not in all cases they're not.
16 MR. EBERSOLE: They aren't. What are they?
17 Turbines? !
18 MR. ENOS: They're turbines. They're all turbines.
19 MR. EBERSOLE: Oh, sorry, you told me they were all
]
- 20 turbines. Well, the speed control on the turbine, do you use 21 that?
22 MR. ENOS: The two things that we saw probably as the 23 largest contributor was just the actual control system on the 24 turbine itself. And actually backing up into the ICS system 25 where input to ICS would fail, and ICS would try to tell the
(
Horitage Reporting Corporation (202) 628-4888
1 1
I l
44 1 feed pump to go where that-failed input told it to go. So the 2 feed pumps would go where they were told to go, they're just ,
i
-3 being told to go in the wrong place.
4 And that issue and the combined issue of control 5 system ~on turbine itself is probably two of the biggest areas.
6 MR. EBERSOLE: Do you use control valves in
- 7 conjunction with speed control? ,
8 MR. ENOS: Some plants do, yes. ,
9 MR. EBERSOLE: But it's a mixed bag.
10 MR. ENOS: It's a mixed bag, yes, sir. Usually, [
11 there's control valves --
12 MR. EBERSOLE: When you use them both, do you have 13 less trouble?
14 MR. ENOS: No.
15 MR. EBERSOLE: More trouble. Can't tell.
16 MR. ENOS: Can't tell.
17 One piece of performance Indication that I would like j s
18 to present you as a result of things have been done to make the f I
19 RPS more reliability, and also in the way of kind of following 20 up on some things that we have' talked about before is reactor 21 trip breaker failures; failures of the actual trip breaker 22 itself to open on demand.
23 Across the B&W Owners Group we saw these kinds of 24 failure rates over the previous years. In 1986, 1987 --
}
25 starting in 1984 and particularly in '85 is when we began to do
('
Heritage Reporting Corporation (202) 628-4888
45 1 our frame change outs and change the bearings out with Mobile 2 28 lubricant. All the new things that we learned to do with 3 these breakers.
4 '86 and '87 we've had no failures to open a reactor 5 trip breaker across the B&W Owners Group.
s 6 So there again, being something that has happened 7 that has driven the reliability of the RPS higher, a more 8 reliable system, a more reliability capability.
9 MR. EBERSOLE: Are breakers used because they're more 10 reliable than contactors or are they cheaper or what? ,
11 MR. ENOS: In our case it's basically current 12 limitations on contactors.
13 MR. EBERSOLE: It is. Didn't I hear somebody say two 14 amperes of 150 volts.
15 MR. ENOS: Well, these reactor trip breakers tre 16 interrupting something on the order of 200, 400 amps; the 17 actual trip breakers themselves.
18 MR. EBERSOLE: That's not a big amperage. ,
19 MR. ENOS: I have to go back and remember this 20 discussion from several years ago. I know that there is one ,
21 plant in the country that does use contactors, and as I recall, 22 it has to do with vintage of the design. At the point in time i 23 that these things were designed we did not have contactors or 1 l
24 we were right on the upper end of contactors that would handlo ;
1 25 the type of voltages -- currency we were looking at. That's I l
t l Heritage Reporting Corporation
- (202) 628-4888 1
4
eS:
"l r
46 1 going back several years in my memory, but that's what I seem '(
2- to recall.- i
- 3 MR. EBERSOLE: Well, it could have been contactors,- !
4 it's not'an uncommon matter.
5 MR. STEVENS: Let me help to answer that question.
4 5 Howard Stevens, B&W again.
7 At the time when the design of this system was being :
8 done we wanted the current interrupting device to do two 9 things: to provide the interruption for the reactor trip within t
10 the time-frame of'50 to 80 milliseconds; and two, to provide 11 short circuit' protection for the control -- rod control system ;
12 which required an interrupting capability of in the 13 neighborhood of 22,000 amps.
14 And it was that requirement which led us to breakers. ;
f 15 And the biggest problem we then had which was to find a breaker 16 which would interrupt in the 50 to 80 millisecond time regiont t
. 17 and the breakers we selected are those that would do both jobs. ,
{ 18 MR. EBERSOLE: I see, Thank you.
a t
19 MR. ENOS: I recall it was current, f 20 MR. EBERSOLE: Well, they have a dual function. >
21 MR. ENOS: That's correct.
22 MR. LEE: If you look at the transference again, f 23 somewhere in the year of 1992 those are in the currencies of l 24 hreaker failures. Was there anything special with those.six or 25 so failures that you remember?
l Heritage Reporting Corporation f (202) 628-4888' !
1 i
I
47 1 MR. ENOS: Most of those were at one plant. They 2 developed a problem with breakers. And in 1982 we had not 3 figured out this problem of lubrication that we later learned 4 as a result of Salem.
5 There was a problem at one plant which I think o
6 resulted in four failures just right in a row. It failed.
7 Work on it, gee, it works great, put it back in the cubicle and 8 a week or two later it failed again; and we hadn't figured out 9 this hardening of grease, et cetera, and the things were 10 setting up on us. So it's principally driven by that one 11 event.
12 MR. LIPINSKI: Whose breakers are these?
13 MR. ENOS: General Electric, AK 2-A breakers.
14 MR. EBERSOLE: When you made this decision did you 15 consider amp traps to pick up the short circuit function? I 16 think it's commonly true that contactors would be a lot more 17 reliable for this kind of use.
18 MR. STEVENS: I think at the time we looked at 19 contactors, and because of the problem with short circuit 20 currency --
t .
21 MR. EBERSOLE: But I was just saying you could put in 22 tuses for that.
23 MR. STEVENS: Yes. l I think there were a number of 24 alternatives that we could have chosen. But I think it's not 25 surprising to see that all reactor manufacturers elected mostly l i
t Heritage Reporting Corporation (202) 628-4888 l
l
48
(
1 to put in breakers.
2 MR. EBERSOLE: I think it's the advent of the masters 3 of business administration.
4 MR. ENOS: From a-reliability standpoint the PRA work 5 for the ATWS regulation we assumed that an unreliability rate 4
6 of 10 to the minus three on breakers. Right now actual '
7 demonstrated unreliability rate is something more down to 10 to 8 the minus four range.
9 MR. KERR You said 10 to the minus 4 per demand for 10 breaker --
11 MR. ENOS: Unreliable.
12 MR. KERR: -- failure to open.
13 MR. ENOS: Failure to open is what was assumed in the 14 PRA work in the original ATWS, things that were done. Right 15 now we're demonstrating something in the area of 10 to the 16 minus 4. There are, believe it or not, enough demands on the ,
i 17 breakers to demonstrate things in the area of 10 to the minus 18 3, 10 to the minus 4. l 19 MR. EBERSOLE: Do you regard breakers as being over 20 tested which wears that out?
21 MR. ENOS: Well, I think over tested wears them out.
22 With the General Electric breaker there is not anything that is ;
23 particular sensitive to wear. Components rubbing against one <
24 another and wearing and scoring in friction related type wear 25 components. But obviously, you know, to some point in time
(
Heritage Reporting Corporation (202) 628-4888 :
49 L
1 there is an end of life device. It just requires more 2 maintenance at the frequency that we test them.
3 MR. EBERSOLE: Do they make breakers better now, with 4 better bearings for the same multi-purpose use that they used 5 to. You know, it used to be a breaker was not for persistent 0
6 use.
7 MR. ENOS: Right. Well, these breakers were 8 originally designed for something like 25,000 cycles which is 9 beyond our 40 year life. And we looked -- in 1984 we looked 10 specifically and quite hard at changing to a different kind of 11 breaker or to a lighter vintage breaker, and the decisions were 12 made not to do that. The best way to fix the problem was to 13 fix tho existing breakers. It's not as simple as taking one 14 out and putting another in.
15 MR. EBERSOLE: We have been listening to maintenance 16 policies and other thinga, do you use predictive maintenance or 17 ready to maintenance by replacing the whole thing at certain 18 cyclic accumulations or whatever?
19 MR. ENOSt Both. Predictive maintenance in that we 20 have the on-line monitoring capability. If somewhere at 21 several of these plants we see a problem come up, that 22 information will be disseminated and we'll look at the breakers 23 from a predictive standpoint.
l 24 PM from the standpoint of our procedure, 25 periodically, and it's usually done on a six to 12 month
(
Heritage Reporting Corporation (202) 628-4888 l
1 l
50 1 frequency.
2 MR. EBERSOLE: Isn't it true, however, that the 3 breakers just work or don't work, there's no hangup.
4 MR. ENOS: For the most part that is correct. But we 5 have seen some breakers. Our acceptance criteria, they have to ,
6 open in 80 milliseconds or 100 milliseconds. It varies between 7 plants. We have seen breakers open at several hundred.
8 MR. EBERSOLE: After the Salem case this fascinating parameter 9 of margin deforce came into being, do you monitor that?
10 MR. ENOS: No, we monitor margin deforce indirectly 11 through response time and trip shaft torquet.
12 MR. EBERSOLE: But that's a very crude measurement of 13 actual monitor.
14 MR. ENOS: Our experience has been, that's not really ;
15 the case. That response time and trip shaft torquet is a very 16 good indicator to the health of that breaker. And that tends 17 to stay very constant until you develop a problem.
18 MR. WYLIE: What is your test frequency? ;
19 MR. ENOS: It varies between the plants. For the 20 most part I believe it's annually -- well, I'm sorry. The ,
i 21 breakers are actually tested once a month in the cubicle.
22 They're given an input signal and verified that they open by ,
23 both under-voltage device and shunt trip device, individually. l l
24 They are pulled out with preventive maintenance performed on 25 them, normally every 12 months.
I Heritage Reporting Corporation (202) 628-4888 l
l
51
- 1 MR. EBERSOLE
- What is the time interval for 2 actuation?
3 MR. ENOS: Time interval for actuation, I'm sorry.
4 MR. EBERSOLE: In gauging a trip -- from the trip 5 signal to actual clearing of the contact. You don't have much 6 time to work with it, do you?
7 MR. ENOS: It's roughly -- it's less than a second, 8 as I recall.
9 HR. EBERSOLE: So you look at a variation in that 10 time?
11 MR. ENOS: No. When you do the RPS channel 12 functional test, once for refueling outage you look at your
, 13 response delay times from input signal through the logics
(
14 through the outputs through the relays, et cetera, and you 15 understand what those delay times -
16 MR. EBERSOLE: That's the whole string.
17 MR. ENOS: That's the whole string.
j 18 MR. EBERSOLE: I'm talking about after you get to the I
19 breaker.
1 20 MR. ENOS: That's right. But after you get -- you 21 know what the breakers contribution is when you do that whole I 22 string. And what we do is, we set a criteria, either 80 23 milliseconds or 100 milliseconds that that breaker has to 24 respond in. If it responds outside that time, it doesn't mean 25 it's overall contribution to the channel functional test.
i Heritage Reporting Corporation !
(202) 628-4888
52 1 MR. EBERSOLE: Thank you.
2 MR. KERR: Mr. Enos, my schedule shows that your time 3 is up. I don't want to cut you off before you have said !
4 something very important, but keep that in mind.
5- MR. LIPINSKI: One more question. The breakers are 6 mechanical, so they have high unreliability. What about your 7 SCR's, is that public data?
8 MR. ENOS: Yes, we do. I think we see about one or f
9 two SCR failures for 10 years of operation in a bank of -- how I 10 many is in a bank?
11 MR. STEVENS: Roughly 360. f i
12 MR. ENOS: 360 in a bank. i r
13 MR. EBERSOLE: And that's an open circuit? i 14 MR. ENOS: Usually, f 15 MR. EBERSOLE: Do you ever have any shorts?
16 MR. ENOS: Well, which -- when you have an open f
. 17 circuit, what came first. Most likely a short came. If it 18 shorts it usually winds up open.
19 MR. EBERSOLE: Does it do anything to the rod coil, i
- i 20 it chokes up. j 21 MR. ENOS: Right. !
t 22 MR. EBERSOLE: Thank you. , j 23 MR. ENOS: Going back where we were a few minutes l 24 ago, we talked about the types of things that have been done in j 25 the reliability program to reduce the front end risk of ATWS. !
. i I. I Heritage Reporting Corporation i (202) 628-4888 i
53 1 It's not insignificant, the change and the anticipated 2 transient rate impacts the overall probability of an ATWS very 3 significantly. You move that by just one in the assumptions, 4 it makes a big difference in the analysis.
5 From this part of the equation of the rule what 6 things have happened to impact reliability to change what we 7 might want to do with the diverse scram system, et cetera. We 8 have looked at this, we really don't see anything out there, 9 the world has changed much one way or the other. Things are 10 about the same as they were except for things like surveillance 11 and maintenance and better care and feeding of things that we 12 used to.
13 The DSS system looks like still the opportunity thing 14 to do, and the design for it looks app *opriate. So we don't 15 see any major changes that are taking place there.
16 In the mitigation aspect of things, there are some 17 changes that are taking place that potentially impact the 18 assumptions behind the ATWS rule in different directions.
19 What I would like to talk about for just a minute is 20 some of those things that we see impacting it. In looking at 21 the world we see basically two things that impact the 22 mitigation side.
23 First of all, aux feedwater reliability. I don't 24 want to go into that in great detail, but over the past five or 25 six years in the B&W plants all of the auxiliary feedwater
(
Heritage Reporting Corporation (202) 628-4888
t 54 1 systems have been upgraded with safety related, initiation and 2 control systems. In some cases we've added turbine drivers 3 where they weren't there before. A number of changes have 4 happened to make that auxiliary feedwater system better and 5 more reliable. And I'm sure you all are very much familiar 6 with those type of activities.-
7 So from that standpoint, we have improved the 8 capability to mitigate the consequences of an ATWS event, just 9 in the overall reliability of auxiliary feedwater.
10 The other thing that has changed and can change, that 11 could impact mitigation capability is core design changes 12 themselves. In looking at that world --
13 MR. KERR Before we get to that this question will 14 elicit a short answer. What is your calculated reactor ;
15 protection system on reliability on demand or failure to scram 16 on demand? You indicated you had done a number of PRA studies? '
17 MR. ENOS: I'm not aware that we have calculated an 18 overall PRA number for reliability of the or unreliability of 19 the reactor protection system as a whole. FEMA studies have i 20 been done. I'm not aware of one. I don't know, there's been 21 three -- well, I'll take that back.
22 MR. KERR So you really are particularly concerned
- 23 about what it is, as long as you follow the ATWS rules? I'm t
24 putting words in your mouth.
25 MR. ENOS: Yes, you are.
, ( .
Heritage Reporting Corporation l
I (202) 628-4888 i
55 1 (Laughter) 2 MR. ENOS: Yes, you are. We are very concerned about 3 the reliability of reactor protection syster and the tripping 4 system. We want it to trip, and it will trip. And we don't 5 want it to trip when it's not suppose to. It's an area that 6 we're paying a great deal of attention to.
7 FEMA's have been performed. On a plant specific 8 basis there has been three level 1 PRA's performed on the plant 9 specific areas; Oconee Florida, and I forget the other one. i 10 And I don't know what specifically was looked at in those 11 PRA's.
12 MR. KERR: It would seem to me with that in effect 13 two separate system setup that you have, that it would be 14 interesting to know whether the calculated failure rate is 15 significantly lower than on the more conventional systems.
t 16 Also, since we ask a greater reliability of scram 17 systems than any other system that I know of in nuclear 18 powerplants, and presumably with reason, I would -- I'm a 19 little puzzled that an owner / operator doesn't want to know what ,
20 it is, but then so be it.
21 MR. TAYLOR: Dr. Kerr.
22 MR. KERR: Yes, sir.
! 23 MR. TAYLOR: Those numbers were calculated, and I'm i l
24 not absolutely sure and we're not sure what the numbers were, 25 but they were done in connection with the reactor trip breaker 1 1
(
Heritage Reporting Corporation (202) 628-4888 l i
I
56 I surveillance test interval analysis that Ted Enos mentioned 2 beforo; and we can get you that number.
3 MR. KERR: I'm just sort of surprised where that 4 number is --
5 MR. TAYLOR: It's a very, very, very, very low 6 number.
7 MR. KERR: You mean like 10 to the minus 3?
8 MR. TAYLOR: No.
9 MR. DORAN: Ron Doran with B&W. The number that I 10 recall looked at a couple weeks ago was five times 10 to the 11 minus 9, 12 MR. KERR: I wish you hadn't told me that.
13 (Laughter) 14 MR. ENOS: Thank you. I forgotten in the process of 15 looking at the surveillance interval, we calculated several 16 different values to do comparisons.
17 MR. EBERSOLE: I want to ask you a question. Am I 18 correct in saying, I think B&W and CE have a good many more, 19 what I call, benevolence scrams, in that when you scram from a 20 turbine trip, for instance, you don't lose critical service 21 functions in the normal context like main feedwater, you ramp 22 them down; am I correct?
23 MR. ENOS: That's particularly correct on the B&W.
I 24 MR. EBERSOLE: And you hold the BISAC's and you don't j 25 lock up the heat sink or the water supply, you try to ramp down
( '
Heritage Reporting Corporation (202) 628-4888 ;
i i
J
l i
1 57 I 1 and hold. That's what I call a benevolence scram, because you I
2 didn't ask critical equipment to rise up and do anything, you i
3 just scrammed the reactor. 1 4 How frequently do you fail to get your benevolence, !
5 and you don't hold main feedwater? And when you're saying 6 that, how do you control differential flow'to the several 7 boilers from one feed pump?
8 MR. ENOS: The original design of the B&W machine is 9 such that it could run back to about 47 percent power on loss 10 of one main feed pump.
11 MR. EBERSOLE: What bypass do you have?
12 MR. ENOS: It varies, 15 percent on some of the older 13 plants, something more like --
14 MR. EBERSOLE: But you still have signals.
15 MR. ENOS: Yes. But the original design was for the 16 capability to run the plant back in loss of feed pump and stay 17 on line. Some of the changes that were made after TMI, we lost l 18 that capability for the most part and resulted in reactor
! 19 trips, where previously we had gotten run backs. l 20 In the last year or two we have begun to regain some 21 of those features and the plants are now, as we're changing 22 some things again, the plants are now beginning to get the 23 capability to run back again.
24 MR. EBERSOLE: I'm not talking about running back on 25 turbine output, just the main feedwater.
I.
Heritage Reporting Corporation I (202) 628-4888 !
58 1 1 MR. ENOS: Just the main feedwater, right. In the 2 last three cases that I'm aware of in the Owners Group where we 3 have lost a main feedwater pump, a singular pump, the plant has 4 stayed on line with the other feed pump running, everything 5 successfully running.
6 MR. EBERSOLE: When you trip the turbine, for 7 whatever reason, you can go kick a level switch on a heater and 8 trip the turbine.
9 MR. ENOS: Right.
10 MR. EBERSOLE: And there's zillions of those trips 11 out there which are singled to trips.
12 MR. ENOS: Right.
13 MR. EBERSOLE: All I want to hear you say is, I keep 14 the main feedwater and I know I have a heat sink because I've ,
15 got safety.
16 MR. ENOS: We keep the main feedwater in any case
. 17 that we possibly can.
18 MR. EBERSOLE: And so when can't you, what percent?
19 MR. ENOS: What percent power that we trip?
20 MR. EBERSOLE: No, what percent do you lose main 21 feedwater on a turbine trip?
22 MR. ENOS: I don't know that number.
l 23 MR. EBERSOLEt I'm trying to get the challenge 24 frequency on the feed pumps.
25 MR. ENOS: I understand what you're after.
(.
Heritage Reporting Corporation j (202) 628-4888 l l
1 l
l
l 59 1 MR. KERR: It's like 50 percent or five percent.
2 MR. EBERSOLE: That's what I meant.
3 MR. ENOS: Loss of main feed pumps on turbine trip?
4 MR. KERR: If you don't know or don't have any 5 ides --
6 MR. ENOS: Honestly, I don't know.
l 7 MR. EBERSOLE: It's important to differentiate 8 between a benevolence scram and a non-benevolence scram.
1 9 MR. ENOS: I understand. And I know where you're 10 headed with the aux feedwater challenge rate, and I just don't 11 have that information. I'm sorry.
12 MR. KERR: Why don't you continue. ,
13 MR. ENOS: In the fuel cycle we looked at a number of 14 things that are going on and changes that we're making at 15 various plants. These are things that are going on, going to 16 ZR grids, and axial blankets and low leakage schemes, et 17 cetera. And from the standpoint of ATWS these really have a 18 very minimal effect on ATWS analysis one way or another, mostly 19 in the graphs.
20 The only thing that is really going on out there that 21 has some impact on ATWS analysis is a move from an 18 month 22 fuel cycle to a 24 month fuel cycle.
23 As I mentioned earlier, the original analysis we did ,
24 for ATWS in 1980 was based on an 18 month fuel cycle; not all 25 the plants were there at that point in time. We now have
[
Heritage Reporting Corporation I (202) 628-4888 l
l l
60 1 everybody at 18 month, and in some cases moving to 24.
2 Some characteristics of moving to the longer fuel 3 cycle is adding lump burnable poison or soluable boron or both.
4 If we look at current configurations of plants in the 5 MTC's, et cetera, in 1979, 1980 when we did our generic <
6 analysis for all the B&W plants we assumed a 95 percentile MTC 9 7 of a minus 1.05. If you take the operating plants out there 8 today in an 18 month cycle, calculate their 95 percontile MTC, 9 average that, current 18 month cycles are running at a minus 10 1.1. The spread on that number is about .15.2; so everybody is 11 in there fairly tight.
12 on the operating 24 month cycle that is out there, 13 we've gone from 18 to 24, the current 95 percentile MTC is 14 approximately minus .43.
15 MR. WARD: Can I ask you, how -- I'll ask three 16 questions. How are those numbers calculated? Where and how !
i 17 are the calculations benchmarked to experiments? And what do i l
18 you regard as the uncertainty in those values?
l 19 MR. ENOS: I'm going to toss all three of those j 20 questions over here to B&W, if I could. l 21 MR. KERR: Did you understand the question? Would 22 you like him to repeat it? ,
i 23 MR. LOJEK: After hearing all three of them I would I l
24 like to go back to number one, the first question. Jan Lojek 25 from Babcock and Wilcox.
I
(- l Heritage Reporting Corporation (202) 628-4888
l 61 >
l 1 MR. WARD: How are those numbers calculated? ;
2 MR. LOJEK: They're calculated with our fusion code.
3 We had used PDQ in the past, we are going and presently ~using 4 NUTEL code. It has got a long history. Does that answer your l
l 5 first question? .
6 MR. WARDS. And how do you benchmark the calculations
! s 7 by the NUTEL code or whatever it is against the experiments?
8 MR. LOJEK: Well, if we're going to the NUTEL code 9 there's very good agreement between NUTEL and PRQ, so we go ,
10 back a long point in time, all the way back to 21 days of PDQ.
1 i 11 During every fuel cycle startup, reload cycle, there's a 12 measurement performed at power at or near 100 percent power to 13 produce a measured moderate coefficient. And that is compared 14 to the predicted. I 15 There are differences, of course, between predicted !
16 and measurement, as we expected. We did values on both sides 17 of the number. In other words, there's not a definite bias 18 between calculated measures.
4 19 And then your other question was, what is the spread .
20 between that or what is the uncertainty.
21 MR. WARD: Yes, what uncertainty do you regard that 22 was appropriate?
23 MR. LOJEK Well, fectoring in measured uncertainty i
24 and calculated uncertainty it is difficult to come up with a 25 number that you should meet all the time. We use our
(
Heritage Reporting Corporation (202) 628-4888
62 1 acceptance criteria .4 which does seem rather high.
2 MR. KERR I'm sorry, you said you use something?
3 MR. LOJEK: Use an acceptance criteria .4. In other 4 plus or minuu .4 times 10 to the minus 4, delta row for degree 5 Fahrenheit.
6 MR KERR What does an acceptance criteria mean?.
7 Does it mean that's what you=think you achieve or likely 8 achieve?
9 UR. LOJEK: We see differences much-smaller than 10 that,-in the range of plus or minus .2 from our measured and 11 predicted values.
12 MR. KERR Over the years you must have data from e 13 which you can apply some soit of spread. What is your 1 r 14 acceptance criteria based upon? Is it based on what you think 15 you are getting or what you would like to get or none of the l 16 above? !
i 17 MR. LOJEK The criteria is based upon what you !
18 expect to be seeing when you start up the cycle. Based upon i 19 past knowledge of the way in which the coefficient was )
20 calculated as well as the measurement technique you are using 21 when it is being measured.
22 There is a recent reg guide I believe on --
23 MR. KERR That's about a one sigma difference?
24 MR. LOJEK One sigma.
25 MR. WARD: And if you make this prestartup
( -
Heritage Reporting Corporation ,
(202) 628-4888 1 l
F
^
63 1 measurement and you don't fall within this spread of plus or
- 2 minus .4, what happens? ;
3 MR. LOJEK: Then it needs to be reviewed before it's -
4 accepted or not accepted. Only what it has done is remeasure 5 and determine if the difference is real.
i 6 MR. WARD: But it's reviewed by whom? !
7 MR. LOJEK: Utility and Babcock and Wilcox. If they 8 don't meet the acceptance criteria which were spelled out, 9 defined, then Babcock and Wilcox is suppose to be informed and
, 10 discussed before they go further in the test program or before i
11 continue operation really. But they don't shut down at that 12 point, as far as I know. It's just --
13 MR. WARD: After the review I guess you look for some f
14 explanation; then, nas there ever been a chse where the 15 utilities had to go back and make a change in the loading?
16 MR. LOJEK: Acceptance criteria is usually not even j 17 approached. You also have a tech spec value, which is really
. 18 the ultimate criteria of being, in most cases, negative at full 19 power. ,
1 1
20 MR. KERR: I guess I don't understand why the i 1
! 21 acceptance criteria is almost never approached if it's based on !
- 22 experience. j
\ l j 23 MR. LOJEK Once again. I 24 MR. KERR I said, I guess I don't understand why the 25 acceptance criteria is usually never even approached, whatever
, t Heritage Reporting Corporation (202) 628-4888 1
-, - ,, ,, - . - - -. - . - - - - - - - - - , - , - - - - w
64 1 that means, if it based on past experience. Have things got a 2 lot better.
3 MR. LOJEX: Well, to answer your question, you have 4 come up with a band which you believe the value should be 5 within, and then set the band at a reasonable amount.
6 MR. KERRt But if the band -- maybe I misunderstood. f 7 If the band is based on experience of a spread, it seems to me 8 occasionally one would approach the limits of that band. I 9 mean, one certainly did at one point otherwise the band 10 wouldn't have existed. If one never sees even a close approach 11 to that band anymore, things must be improving or the earlier l
l 12 data must not be typical of what one is not seeing. Am I 13 missing something. I l 14 MR. LOJEK: Maybe I should qualify that. There 15 probably have been plants that have come close to the band and 16 perhaps have gone over the band and have retested, remeasured ;
. 17 and resolved it before they went on.
i 18 MR. KERR Thank you. -
19 MR. EBERSOLE: I believe you said you tested at full ;
t 20 power. l 21 MR. LOJEK: The condition for the test is at or near !
l
! 22 full power. l l .
23 MR. EBERSOLE Wouldn't it be more conservative to l.
24 test it prior to power. l 25 MR. LOJEK For initial startups of a new core, which
(
- i l
Heritage Reporting Corporation l (202) 628-4888 '
l
65 I we have not had for some time, they do it at power levels going 2 up. ;
3 MR. EBERSOLEs- At steps?
4 MR. LOJEKt At steps.
5 MR. EBERSOLE: At what variations?
o 6 -MR. LOJEK: They. test at full power. They also test 7 at lower power on reloads, also.
8 MR. EBERSOLE: So what sort of spread do you get from 9 full power to low power?
i' 10 MR. LOJEK Full power to low power?
I 11 MR. EBERSOLE: Yes. You do get a difference, don't 12 you?
13 MR. LOJEK: You get a difference, yes. In fact ;
j 14 there's an intermediate power level, but there is a test 15 performed also. ;
1 16 MR. EBERSOLE: So what is it there compared to full 4
17 power? Is it less or more or what?
4 i
i 18 MR. LOJEK: When you se,y less or more, you compare 1
4 19 your predicted value which you already have precalculated at i 20 the lower power to your measure at the lower power. And then t j 21 when you go to full power you do the same type of test.
22 MR. ENOS: Let me put this graph up here, it may help
, 23 on this less or more question.
24 One of the main things we're concerned about on the 25 MTC here in startup is the tech spec limit, we can't be at full i
1
(..
Heritage Reporting Corporation
] (202) 628-4888 1
1 5
l 1
66 1 power with a positive MTC. So therefore we are looking at it 2 as we come up and we're most interested in it just before we 3 get to full power to make sure there.
4 Now, if you'll look at it during the 45 or 50 percent 5 power hold and then you come back later and you look at it 90, l
6 95 percant, is it more or less between those two. Well, it's l 7 going to somewhat depend because the time you get to 95 percent
- 8 power you're going to be at a different place on this MTC l
l 9 versus burnup curmi. So you would expect, what you're not 10 really looking for and you correct me now if I misstate again, .
11 what you're not really looking for is it the same at 50 percent 12 power as it 1s at 100 percent power. You're looking at how 13 close, what yotr measuring is to your predicted value as the
(
14 function of cycle burn up.
15 And you would predict that they would be different 16 two days ago versus today regardless of what power level that 17 one is at. That's what we are really looking at is comparison I 18 in that point in time.
19 MR. KERR Mr. Ward, I think you might not be 20 finished with your line of questions.
21 MR. WARD: Yes, I guess I don't have a real good 22 feel for how good this test is that's run. John, Lee, do you ,
23 have --
24 MR. LEE: My feeling is that at zero power, high zero t
25 power perhaps you can talk about .4 times 10 to the minus 4, 1 t.
Heritage Reporting Corporation l (202) 628-4888 l c
67 1 delta K over K type of accuracy with reasonable confident. At 2 anywhere above very low power level my confidence goes down 3 very rapidly, especially at full power I'm not even sure I can 4 talk about .4 or .5; I just don't have a good feel for that.
5 MR. WARD: You mean as far as being able to measure a
6 something that's really indicative of the real coefficient.
7 MR. LEE: Right. To the calculation at full power 8 compare with measurements, I won't be able to, myself, quote 9 anything like .4 times 10 to the minus 4.
10 MR. WARD: But yet they seem to be repeatable. I 11 mean, their experience in that they --
12 MR. LEE: You go through the same process, so you get 13 the same number, I guess.
14 MR. WARD: Sort of normalized.
15 MR. EBERSOLE: What do you do, force the reactor to 16 operate at full power at a somewhat lower than normal 17 temperature?
18 MR. LOJEK That's correct. That should change the 19 temperature.
i 20 MR. EBER90LE: And lots of things get mixed up in l 21 that kettle and soup, doesn't it. '
22 MR. LOJEK Yes, there are a lot of things going on.
23 One thing, at the beginning of life of the moderate coefficient 24 is less negative than at end of life. End of life you would 25 have more rod motion to compensate; at the beginning of life Heritage Reporting Corporation (202) 628-4888 l
68 ;
1 you have less rod motion. l 2 MR. EBERSOLE: You calibrate against the rods. ,
3 MR. LOJEK: Yes. t 4 MR.-LEE: But again, you cannot even measure rod 5 meaning for that full power. You just nessure at zero power, j 6 that's all. i 7 MR. EBERSOLE: Everything merges.
8 MR. LEE: Right. j 9 MR. LOJEK: I admit that the uncertainty at full 10 power is or at the power period is higher in the measurement l i
11 than zero power. [
12 MR. EBERSOLE: Do you have a feel for how much?
l 13 MR. LOJEK: I wouldn't know. [
14 MR. LEE: Just one quick question to confirm. This i
15 is all the MTC curve and curves you have in your package here, 16 are all for hot full power with zonon. i 17 MR. ENOS: Yes, that's correct.
18 MR. LEE: So at reduced power level with zenon, even !
19 without any uncertainty considered, you could be talking about 20 moving this curve out considerably? ,
21 MR. LOJEK That's correct.
i 22 MR. LEE: Could you give us some idea what this curve ,j 23 would be at hot zero power without zenon. i 24 MR. LOJEK: At hot zero power without zenon.
l 25 MR. LEE: Right. !
i Heritage Reporting Corporation f (202) 628-4888 ;
i
]
I
}
__ - ._- _ - - _ ,_=
.I i
69 !
1 MR. LOJEK The curve would be shifted up, okay, 2 qualitatively. Quantitatively it wouldn't be the same as with ,
3- the function of cycle device either. It would be difficult to i* 4 give a quantitative number.. At BOL for the 24 moi.th cycle f i 5 which Ted will show next, we measured a monitor -- a positive t *- !
- 6 monitor coefficient ut zero power with no zenon, and then at j 7 full power at roughly five to 10 full power days on the cycle, a value was measured negative. I 8
- 9 MR. LEE
- Just margin of the negative, according to
- i 10 the next curve.
1 L i 11 MR. LOJEK: It was more than marginally negative. :
i 4
12 Minus .5, in that range was the value measured already.
l l ,
13 MR. LEE: But if you now assume with some f i
l j 14 uncertainty, then you could be easily talking about zero MTC 15 value or even positive.
(
16 MR. LOJEK: If you assumed the bias. If you wanted f 17 to use values that are not nominal. l i i j 18 MR. LEE: Your acceptance criteria, j 19 MR. LOJEK: Then that curve would shift up, that's f J
t 3 20 correct. l 21 MR. LEE: Thank you.
] 22 MR. ENOS: The slide that I have up iare is a typical ,
]* 23 slide for an 18 month fuel cycle, MTC is a fun: tion of burn up.
I l
1 i I
l 24 An 18 month cycle we consider to be something on the order of i 25 424-40 EFPD.
I i Heritage Reporting Corporation (202) 628-4888
- , _ - - _ _ - - . _ . _ . - _ - . _ - - - . ., , - . _ . . .__ ._. J
'l 70 1 For the 24 month fuel cycle that we used, typical ;
2 burn up on that 540 EFPD. You'll see a curve that looks very l 3 much the same shifted to the right and up. Combine those two 4 curves, take a look at what does this mean from the standpoint
6 work behind the ATWS rule was an assumption, non-acceptable l 7 consequences less than 50 percent of the time or that being a 8 prediction of pressures below service level C, more than 50 9 percent of the time.
10 If we calculate that MTC it's approximately a minus 11 1.3 at which we would predict service level C pressures from a '
12 loss of main feedwater ATWS event for the B&W plants.
13 MR. KERR Mr. Enos, I don't believe that you want to 14 give the . impression that people who operate these plants are I 15 only interested in satisfying regulation. I think you're also 16 interested in operating a plant safely. And yet, in explaining I
. 17 things you frequently refer to the fact that you are, for '
i 18 example, get an unacceptable consequences less than 50 percent i 19 of the time, presumably, because this is the ATWS rule and not i 20 because you think -- well, I don't know. You don't do what you 21 do in your plant just because of the NRC regulations, do you?
i 42 MR. ENOS: No, sir, that's not the case and that's t
23 certainly not the message that I'm trying to leave.
24 MR. KERR Maybe you have talked with us in the NRC 25 so much about regulations that you have this mind-set. I want C
Heritage Reporting Corporation :
(202) 628-4888
{
_. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ - _ - J
71 1 to try to have all that eliminated.
~
2 MR. ENOS: It's not that case so much as I'm r 4 ,
3 beginning to think perhaps I'm answering a different question- !
!* 4 than maybe I'm being asked. The understanding in putting the j i
S presentation together was a question of, have things changed in ,
6 the last eight years such that the basis for the ATWS rule has-t
+
7 been invalidated. [
a 8 Now, I acknowledge that there are certainly many ;
9 things that we have going on and many concerns that we are ;
10 looking at. This presentation is focused on answering i
11 specifically that question, has the basis for the ATWS rule i
12 invalidated. And I apologize that I have apparently left the f j 13 wrong impression with regard to that, but that's why I keep g f
- i 14 referring back to the basis from the rule is, specifically in !
15 answering that question.
16 MR. EBERSOLE: I don't think this meeting is just to 1
17 look at the small preservations of what's at change as a result a'
18 to reloads, as to have another look at the whole show.
19 You all must have a great PR operation in your plant.
20 The shape of that curve is wonderful, it shows it going down.
21 It gives you a warm feeling. I could have drawn it the other
! t
! 22 way and it could have gone up and that would have left me with (
i 23 another feeling. !
24 The shape of that is basically due, isn't it, to the 25 fact you are changing the poison concentration in the monitor.
(- i Heritage Reporting Corporation !
(202) 628-4888 l l
72 1 MR. ENOS: The soluable poison, yes.
2 MR. EBERSOLE: And I could go back to say, what the 3 hell do you have any poison in that, why don't you shut it down 4 with iron rods, which has been one of my wonder months over the 5 last 25 years, how you ever got like that. You know, the 6 boilers can't do that. And it shows very well of where you are 7 on the left hand side because you've got poison in the water.
8 MR. ENOS: That's correct.
9 MR. EBERSOLE: If you didn't have it that would be a 10 grossly different curve.
11 MR. ENOS: That's right.
12 MR. EBERSOLE: But you say, oh, well, this is good 13 enough, and I reckon it is because nothing has happened yet.
14 MR. ENOS: I don't think that we want to be satisfied 15 that something is good enough on the basis that something 16 hasn't happened yet.
17 MR. 3BERSOLE: It's just interesting to know that the 18 shape of the curve sliding downward in the right direction.
19 MR. ENOS: Right.
20 I appreciate what you're saying and perhaps we have 21 missed the mark somewhat in the material that we have put [
22 together here. We're looking specifically at this question of ,
23 where are the plants now from the standpoint of ATWS risk 24 relative to where did we think they were in 1980 when we formed 25 the basis for the regulation; are they better or are they worse !
Heritage Reporting Corporation i (202) 628-4888 j l
t
'F 73 ,
t 1 in that ATWS arena. ,
2 I understand now that your questions are broader >
3 ranging than that, and I'm somewhat missing the mark here, and i 4 I apologize for that.
5 MR. EBERSOLE: It was inevitable, further than the l 6 narrow scope.
7 MR. LIPINSKI: Where is the 12 month curve. You have 8 shown 18 and 24. >
l 9 MR. ENOS: We don't have that up there. We don't 10 have any plants on a 12 month curve at this point in time, but i
11 it's going to be something along in there.
12 MR. LIPINSKI: But that is the old curve?
13 MR. ENOS: I'm sorry, the old curve? ;
14 MR. LIPINSKI Yes, the 12 month is your old cycle, i i 15 now you're going to 18 and 24.
16 MR. ENOS: That's correct. In 1980 when we did our ,
1 17 analysis for ATWS we assumed an 18 month cycle. At that point
]
18 in time we had one plant at an 18 month cycle with plans for 19 everyone to move. But we ut 3d this in our calculations. Since I 20 that time everyone has moved to 16 ,nonths. One has moved out 21 to 24.
22 MR. LEE: May I ask maybe one question. If you were i
23 to single out maybe two items that resulted in this substantial 24 decrease in the negative value of MTC between 18 months to 24 25 month cycle, what would you call it?
\
l (
Heritage Reporting Corporation )
]
(202) 628-4888 i
74 1 MR. ENOS: I would call it the single largest 2 contributor soluable boron. And that's really the only one 3 that makes a --
4 MR. LEE: How much of an increase in the 5 concentration are we talking about, typically taking that 6 cycle?
7 MR. ENOS: A few hundred PPM. Jan. l 8 MR. LOJEK: Yes, you're talking 3 to 500 PPM. !
9 MR. LEE: And that is primarily because the average, 10 core average burn up at the beginning of cycle in reloads has 11 come down, primarily.
12 MR. ENOS: Certainly the core average burn up at the 13 beginning of cycle has increased.
14 MR. LEE: You have to have --
15 MR. ENOS: Well, it's not really for that reason, you 16 go to an 18 month cycle you have to have more reactive core.
17 And you have to have more soluable boron to offset the 18 reactivity for the more reactive core; and that's what is 19 driving the increase in boron concentration.
20 MR. LEE: So do you feel enrichment has gone up?
. 4 21 MR. ENOS: Yes.
22 MR. LOJEK: Let me try to add something to that. The 23 longer fuel cycle you would have to have more access reactivity 24 in order to burn a longer time. You can hold that down through 25 one burnable poison, you can hold it down through higher Heritage Reporting Corporation (202) 628-4888
75 1 soluable boron concentration. You could also have other 2 mechanism such as more core rods in the full power, which we 3 considered is not desirable.
4 In our case we have shared the access reactivity that 5 we needed between the 18 and the 24 month cycle between 6 additional lump burnable poison and the higb9r cycle boron 7 concentration.
8 MR. EBERSOLE: What sort of diverse protection have 9 you got against excess boration? How many ways do you know 10 there is too much boron in it? There's a chemical analysis, of 11 course, I know that.
12 MR. ENOS: Yes.
13 MR. EBERSOLE: There's rod position, that's another 14 one. How many others?
15 MR. ENOS: Reactor power will tell you.
16 MR. EBERSOLE: Can't you just pull the rods out and 17 make it go with more boron on it?
18 MR. LOJEK: The rods are normally full drawn on those I
19 at full power.
l 20 MR. EBERSOLE: Well, are these the cross-checks, rod i 21 position and chemical analysis?
22 MR. ENOS: Well, rod checks and chemical analysis and 23 power level. If you pull the rods all the way out --
24 MR. EBERSOLE: Yes, of course, that's common 25 denominator.
I Heritage Reporting Corporation (202) 628-4888
76 1 MR. ENOS: Sure.
2 MR. EBERSOLE: But really that's the two cross-3 checks, rod position and boron analysis.
5 MR. EBERSOLE: And somebody will always notice one'or 6 the other.
7 MR. ENOS: Well, yes, you have requirements to 8 calculate your shutdown margin.
9 MR. EBERSOLE: The operators are keyed to knowing 10 that's what keeps them where they are.
11 MR. ENOS: Right. Shutdown margin is calculated once 12 or twice per shift. I forget the various criteria for it.
13 MR. EBERSOLE: And the -- shutdown are part of that.
14 MR. ENOS: Right. Rod position, boron concentration, 15 all that in a function --
16 MR. EBERSOLE: It's piece together.
17 MR. ENOS: That's correct. They actually do the 18 calculation in the control room.
19 MR. EBERSOLE: Is that in their exams? l 20 I MR. ENOS: I don't know that for a fact, but I would I 21 feel fairly confident that it is, because that is a calculation !
22 that is typically done in the control room.
23 MR. EBERSOLE: So they've got a seat of the pants 24 feeling of where they are.
25 MR. ENOS: Well, that's a fairly good calculation.
(
Heritage Reporting Corporation (202) 628-4888
77 1 Seat of the pants --
2 MR. EBERSOLE- Well, I mean, -- you know, what I 3 mean. They've got a confidence that --
4 MR. ENOS: They're required to maintain a five 5 percent shutdown margin. In Arkansas' case the procedure 6 requires five and a half. They know they've got more than five 7 and a half which gives them a margin to the tech specs, which 8 therefore gives them a margin --
9 MR. EBERSOLE: I'm just trying to get a feeling of 10 where they know where they are. ;
11 MR. ENOS: Yes, sir.
12 MR. WARD: What is the change in the fuel loading l 13 between the 18 and 24 months?
14 MR. ENOS: The number of the assembles is constant, 15 177 assembles. We're changing fuel and reactivity.
16 MR. WARD: So there is a higher number at the highest 17 enrichment, is that right, the larger number at the highest 1 18 enrichment.
19 MR. ENOS: The core average enrichment is increased 20 in order to get the -- l
- l 21 MR. WARD: How much?
22 MR. ENOS: Half a percent, something like that.
23 MR. LOJEK: Somewhere in that vicinity. It depends !
l 24 on the plant, of course, previous cycle how much -- what's 25 coming from the previous cycle into the next cycle a.,0 how much
(
Heritage Reporting Corporation (202) 628-4888
78 1 you're going to have to load up.
2 MR. ENOS: Dr. Kerr, bearing in mind this difference 3 of, I suppose the question that I'm answering and being asked, 4 this slide is sort of a bottom line slide of the presentation, 5 and I will make the concluding remark that the basis in the 6 regulation was pressures less than service level C, greater 7 than 50 percent of the time as one of the acceptance criteria.
8 Minus 1.3 is the MTC at which we calculate service 9 level C for the plants. Looking at the 18 month cycle you see 10 that those plants operate approximately 69 percent of their 11 cycle at pressures that we would predict to be less than 12 service level C.
13 The 24 month cycle operates at approximately 60 14 percent of its cycle at pressures that we would calculate to be 15 less than service level C.
16 Now, backing this up into the bigger scheme of things 17 and overall risk of ATWS as to what has changed, let me try to 18 contrast, again, going back to the basis for the rule, the risk 19 calculations that were done, et cetera. Let me kind of 20 contrast the significance of that change to some of the other i 21 changes that has been made.
22 If one goes back into those calculations and reduces .
23 the number of expected anticipated transients in a year by one, 24 in other words, we originally assumed four and plants are out 25 there at two now, say, we reduce that from four to three,
(
Heritage Reporting Corporation (202) 628-4888
79 1 reducing that number by one or the number of reactor trips in a 2 year by one will more than compensate for the difference in 3 moving from here to here. As a matter of fact, it will 4 compensate for by about a factor of one and a half, as far as 5 overall societal risk from the ATWS event.
6 So we would acknowledge that there is a difference 7 between here and here. Certainly this one being less than the 8 assumptions in the basis for the rule; and more than offset by 9 the performance of the plants in the area of trips and 10 transients.
11 Yes, sir.
12 MR. EBERSOLE: When you measure ATWS pressure 13 calculation, do you take credit for energy rejection to the 14 secondaries or any other place other than just bleeding steam 15 out the safeties.
16 MR. ENOS: There's energy rejection to the 17 secondaries.
18 MR. EBERSOLE: So even though one of your transients 19 has loss of feedwater you still take heat out that way?
20 MR. ENOS: Yes.
21 MR. EBERSOLE: How long does that last?
22 MR. ENOS: About 30 seconds, as I recall.
23 MR. EBERSOLE: So you take credit for sending heat to 24 the secondaries.
25 MR. ENOS: Yes.
(~
t Heritage Reporting Corporation (202) 628-4888
80 1 MR. EBERSOLE: The care to that, there would be 2 another answer, wouldn't there.
3 MR. ENOS: No, I'm not -- well, yes, there would be a 4 slightly different answer.
5 MR. EBERSOLE: Slightly? Isn't that a lot of energy.
6 MR. ENOS: What's happening there is basically a 7 function of -- let me back up. I'm not sure I know the answer.
8 There would be a differenco. It takes --
9 MR. EBERSOLE: I would think it might be a 10 substantial difference.
11 MR. ENOS: I'm not sure I can address that because 12 there's -- can you, Randy.
13 MR. ELLISON: Randy Ellison from B&W again. The 14 steam generators have an initial inventory that has to boil 15 off, just the physics of it.
16 MR. EBERSOLE: But in a B&W plant that's not much.
17 MR. ELLISON: It's less than two generators, for 18 example. Additionally, beyond that you're going to get 19 auxiliary feedwater coming in to provide some additional --
20 MR. EBERSOLE: At these pressures, with the turbine 21 driven pumps?
22 MR. ELLISON: We're talking pressures still around ,
23 PSI. These are primary system pressures we're talking about.
24 MR. EBERSOLE: Oh, I'm sorry.
25 MR. ELLISON: We haven't made that modification yet.
( ,
Heritage Reporting Corporation (202) 628-4888
l 81 1 So we're putting in auxiliary feedwater based on low level 2 demand and will continue to do that, and then plenty of 3 inventory around the entire ATWS event. But you do get a 4 significant amount of your energy removable through the safety 5 valves and core at primary.
6 MR. ENOS: We don't take credit for inventory that 7 isn't there. It's the actual inventory of the FTSG and it 8 takes about 25 or 30 seconds to complete that, and then you see 9 the pressure spike that comes along with it.
10 MR. KERR: Mr. Enos, you pointed out that reducing 11 the expected transients would move one out here to make the 12 risk rougnly equivalent, and that is certainly true. But what 13 you have done is decrease frequency and increase hazard for 14 consequences in order to achieve constant risk.
15 At some point one becomes a bit skeptical of that 16 kind of tradeoff, because consequences are just too severe.
17 One doesn't want a design for that.
18 Is there some point at which you people in the owners 19 Group would reach that level, and even though you can talk l 20 about a tradeoff you might decide the pressure or something or ,
21 other has gotten big enough in this calculation and we just '
22 don't think we ought to do that.
23 (Continued on next page.)
24 l
25 l
(
Heritage Reporting Corporation l (202) 628-4888 l I
l
82 i l
I 1 MR. ENOS: I understand your statement with 2 regarding to changing probabilities and hazards as well. One 3 has to look at it from that standpoint and one has to look at 4 it also from the total societal risk of what is going on, in a 5 particular case. l 6 Moving the probability downward is certainly not a 7 bad thing to do, it is a good thing to do.
8 And is it worth in a PRA calculation, are they 1 9 interchangeable?
10 MR. KERR: I am not talking about a PRA calculation.
11 I am talking about the judgment that you have to make once you 12 have done the calculation and have seen what it predicts.
13 MR. ENOS: That is right.
14 I think that when one -- years ago, when we studied 15 this thing ad infinitum, decisions were made that this was an 16 acceptable place to be -- less than service level C 50 percent
. 17 of the time.
16 In changing from an 18-to-24-month cycle, we have ,
R 19 made a move closer to that, but we are not at the point at 20 which we previously determined was an acceptable place to be. l 21 Now, --
l 22 MR. KERR Well, suppose your calculations predicted ,
23 that 3 percent of the time you would be at -- let's pick a 1
24 number out of the air -- 6,000 PSA.
25 MR. ENOS: Right.
(..
Heritage Reporting Corporation (202) 628-4888 l
1 1
83 1 MR. KERR: Would you be willing to accept that even 2 though the PRA would say the risk has not been increased?
3 MR. ENOS: I would say that is a possibility that we 4 would be willing to accept at, yes.
5 Again, looking at things in the total concept, if I 6 was concerned that one of the reasons why that we feel as 7 comfortable as we do about step over totally to the side, and 8 say core damage, is because we feel the reliability of the 9 safety systems are high enough to preclude that.
10 MR. KERR: I would feel better if you had not given 11 me that something time 10 to the minus 9 as your failure 12 probability for your reactive protection system.
13 What that says to me, without examining it, and I f.
14 apologize, but somebody has not looked carefully at that 15 calculation. Because you cv . must have ignored the 16 possibility of some common mc o r'ailures, for example, that one 17 could introduce.
18 MR. ENOS: I certainly don't know whether that number 19 considers common mode failure, because that is the purpose of 1
20 the ATWS rule in the first place, is to address potential l
21 common mode failure in the electrical portion.
22 The number that is used in the --
23 MR. KERR: Are you willing to believe as a 24 practitioner in nuclear power, a number which says that the 25 failure probability on demand is 10 to the minus 9 for the Heritage Reporting Corporation (202) 628-4888
84 1 demand?
2 MR. ENOS: Ten to the minus 9 for demand is a 3 difficult number to accept.
4 MR. KERRs Okay, we agree on that.
5 MR. ENOS: ' Ten to the minus 5 is a number that I .
6 don't have a lot of difficulty accepting.
7 MR. KERR: Well, 10 to the minus 9 concerns me a bit, 8 because I don't think it can be demonstrated experimentally.
9 There are things that can't be experimented.
10 MR. ENOS: Things on the order of minus 4, minus 5, 11 minus 6, I tend to have some confidence level in. Beyond that, 12 it is very difficult to prove except as an exercise on paper.
13 And the practicality is different.
14 MR. KERR You are an influential individual in this 15 business, if you will sort of encourage people to relook at 16 that 10 to the minus 9, won't you, please?
17 MR. ENOS: Yes, sir, we will take another look at 18 tnat.
19 The number that was used in generating the basis for 20 the ATWS rule was something on the order of 10 to the minus 5, 21 it was the number that was plugged into there, 3.5 or 22 something, or something like that as I recall. ,
23 MR. LIPINSKI: Question, you apply the structure co-24 efficient here, but that is only used in the calculation from i 25 the deep pressure. What would this plant look like if we were
(
Heritage Reporting Corporation (202) 628-4888
l l
85 1 finding deep pressure as a function of these temperature going I 2 into the analysis? !
3 What would the deep pressure be on the left side of 4 your line of analysis?
5 MR. ENOS: This line right here is going to 6 correspond with 3,250 pounds roughly. This somewhere right in 7 here is going to correspond to 3,500-something roughly. One 8 does theoretical calculation, holds the system in tact, and all 9 of that sort of thing.
10 This is going to produce a number probably 2,000 11 pounds higher than that, along that line.
12 MR. LIPINSKI: The 24-month curve is going to be on 13 the order of 4,000-some pounds?
s 14 MR. ENOS: No, it would probably more on the order of 15 5,070 pounds, 95 percent at that point right there in line.
16 MR. LIPINSKI: What happens to a primary system that 17 is subjected to a pressure on that crder?
18 MR. ENOS: Well, we have looked at the primary system l l
19 stress analysis standpoint to 4,000 pounds. We have not looked 20 at it at 5,000 pounds, at least quantitatively. Qualitatively, 21 engineering judgment or real world, if you would, we don't 22 believe the system would ever see any kinds of pressures like 23 5,000 pounds.
24 I can go do the calculations. I can assume 25 everything is in tact, and nothing leaks and all of that, and I Heritage Reporting Corporation (202) 628-4888
86 1 can calculate a number on up there. In reality, we know that 2 the 0-rings around the vessel head are going to leak. There is 3 probably going to be some sensing, the lines are going to break 4 and we really, in the real world do not expect that we would 5 ever see pressures-4,000 pounds or significantly above that, 6 even with this peaking on that curve.
7 MR. KERR: What if the core doesn't work or has been 8 blocked by the block valve?
9 MR. ENOS: In our calculations, we do not create the 10 fuel. I believe we send the block valves close. It is good 11 for about a delta of 100 pounds on peak pressure. It is not a 12 very significant action.
13 MR. LIPINSKI: The main relief then is through the 14 spring safety and the secondary injection?
15 MR. ENOS: Right.
16 I am sorry?
{
17 MR. LIPINSKI: 7t leaks a little bit, you said, gives 18 a little, you know, like --
19 MR. ENOS: At somewhere around 3,400 pounds, we are 20 going to see leakage around the 0-rings in the head.
21 MR. LIPINSKI: The bolts have been examined for their 22 capacity to stretch a little?
23 MR. ENOS: Yes, sir, in the calculations there, that 24 is going to happen. Really, we have done some work in 25 calculating cross-sectional areas, relieving and you know, you
(
Heritage Reporting Corporation (202) 628-4888
87 1 can come up with all kinds of different numbers you want there.
2 MR. LIPINSKI: Are they notch sensitive?
3 MR. ENOS: I. don't know.
4 MR. KERR: Mr. Lee?
5 MR. LEE: These adverse analyses, let's say, for loss 6 of feed-water transient, are based on some computer code, I 7 presume.
8 Do you account for a single value of MTC given our 9 point in the fuel cycle, throughout the transient?
10 MR. ENOS: No, the way that we calculate it,-is we 11 calculate the MTC that the plant is less than 95 percent of the 12 cycle, okay?
13 And we assume that as the initial condition in the
('
14 calculation and then from that, we calculate our feedback into 15 the plant, as a result of temperature changes during the 16 excursion.
17 MR. LEE: Right, but the MTC, itself, is held 18 constant during the transient analysis, am I correct?
19 MR. ENOS: The code input takes a look at moderator 20 of density feedback, the MTC is moderator density so that it 21 varies a little bit. It is not a specifically constant value.
22 MR. LEE: But on some of the theories, because you 23 are talking about 2,000 psi increase from normal almost 2,000 24 pound pressure increase from normal operating system pressure, 25 and in that case, density could change substantially.
(
Heritage Reporting Corporation (202) 628-4888
80 1 And it is even the MTC curve would at higher 2 pressure would be considerably shifted from these curves and I 3 wonder if you really account for these shifts consistently?
4 MR. ELLISON: Yes, to a certain extent but I am not 5 sure of the detail you are talking about.
6 MR. EBERSOLE: Since the bolts stretch a little bit, 7 is the stretch taken at the third level or is there a neckdown 8 section where it is designed to stretch?
9 You know, a constant cross-section bolt is not as 10 good as one with a program of reduced cross-section.
11 MR. TAYLOR: Mr. Ebersole?
12 MR. EBERSOLE: Yes?
i 13 MR. TAYLOR: The reactor vessel studs take down in I 14 the neck-down section. Our people who have been involved in 15 the design of these vessels indicate that they are going to 16 start to relieve the pro-load in the studs at about 3,200 to I
. 17 3,300 psi and so leakage will start across the 0-rings.
18 MR. EBERSOLE: But that is taken in the middle-down 1
19 section, not the third section? '
20 MR. TAYLOR: Right.
21 MR. EBERSOLE: Yes, okay, thank you.
22 That is all I want.
23 MR. WARD: My question I was going to ask was pretty l 24 much answered by your answer to Walt's, but it we look at the 25 uncertainty in temperature co-efficient, which is, you know, is Heritage Reporting Corporation (202) 628-4888
89 o
1 characterized as maybe, you know, a couple of tenths or, you 2 know, I don't know how much it is, of course, it depends on the 3 confidence level you want.
4 But that could move that curve up, I mean the dotted 5 line up a few tenths, perhaps. So if it goes all the way up to 6 .1, the pressure you calculate there, I gather would be the 7 order of something less than 4,000 psi, is that right?
8 MR. ENOS: Right along, minus 1, I would guess 9 roughly 3,400 somewhere, 3,500 minus 1.05, gives us 3,600.
10 MR. WARD: Oh, okay.
11 MR. ENOS: So, it is right in there, close.
12 MR. WARD: Okay, thank you.
13 MR. ENOS: One last concluding slide.
14 From the standpoint of the effects on the basis of 15 the ATWS rule, we would make the following conclusions from our 16 relook at that. Is that there have been a number of changes i
17 made in the past eight years that have impacted the probability j i
18 of an ATWS evunt impacted them in at, then we believe the I 19 probability of an ATWS event is now lower than it was in the 20 1980 time frame.
21 Additional changes are planned to be made in the
~
22 future, other than the changes from the ATWS rule, which we
, l 23 telieve would drive that probability lower yet. The design l 24 that we have for both the DSS and the AMSAC systems exceed the 25 minimum requirements that are in the rule, because of other
(
Heritage Reporting Corporation (202) 628-4888
90 1 considerations that went into that design.
2 And we don't see any plant or fuel changes that have 3 been made, nor do we expect to be making any with the way that 4 we are headed in the future, that would invalidate the basis 5 for the ATWS rule.
6 MR. KERR: Any other questions?
7 Mr. Enos, has the group looked at other possible 8 analyzed accidents that might be affected by the change of 9 moderated temperature of coefficient as well as the ATWS?
10 MR. ENOS: Yes, sir, as a part of analyzing each new 11 reload cycle that we do, we look at the advance, such as boron 12 dilution, and rod ejection, inadvertent rod withdrawal, low 13 power, zero power, beginning of life-type transients. ,
14 Those are revaluated and/or recalculated using the 15 tech-spec limits or using whichever is the worst MTC out of the i
16 tech-spec high or low for that particular event.
17 That is done at the beginning of each cycle.
18 hR. KERR And did you find marked differences in any 19 of the sequences that you analyzed? Didn't you change the i
20 moderate temperature coefficient?
21 MR. ENOS: On the 24-month cycle, ' don't know the 22 details, but I would expect that we found differences in boron 23 dilution accident.
24 MR. ELLISON: The boron dilution accident would be 25 evaluated for the longest cycle. All the other FSAR events i
Heritage Reporting Corporation (202) 628-4888
91
(
1 remain bounded by the original FSAR analysis which used 2 beginning of life and end of life values, the change in cycle 3 length would not put you outside the most positive and most 4 negative, that'we assume on the other side. And it affects how 5 fast you get through that period.
6 MR. KERR: In an earlier discussion, there was 7 mention of tradeoff between fixed in-core poison and poison in 8 the moderator coolant.
9 Is it feasible or possible to accommodate this 10 increased fuel life by putting all the additional poison in the 11 core without some extremely disruptive behavior other than 12 increased cost?
13 MR. LOJEK: I will answer that question.
i.
14 It probably is feasible, probably. It is probably 15 yes, however, the mention of poison is in there basically to !
16 control the local power distribution. So, you have to qualify 17 that statement in as much as you would not want to increase 18 power peaking by loading it up excessively.
19 MR. KERR: Any further questions?
20 (No response.)
21 MR. KERR Thank you, Mr. Enos.
22 I think we will take a 15-minute break at this point.
23 (A brief recess was taken.)
24 25 Heritage Reporting Corporation (202) 628-4888
i I
92 1 MR. ENOS: I have bad news and good news. The bad 2 news is I'm back up here again; the good news is I'm not going 3 to stay long.
4 (Laughter) 5 MR. ENOS: I'm going to put on my Combustion hat now 6 and the CE Owners Group would like to introduce the CE Owners 7 Group presentation, Mr. John Kaprinos from CE will be making 8 that presentation as some other people here at the table and 9 several of the CE Owners in the audience.
10 The presentation that we would like to make today is, 11 we want to look at the current trends and moderator temperature 12 coefficient, what is happening in the operating cycles out 13 there now. The impacts, the changes that have been made over 14 the past six or seven years on the, again, the basis for the 15 ATWS rule that was formed in the '79, '80 time frame. How we 16 address positive reactivity events. And response to the June 17 12 letter from the NRC asking questions pertinent to the 18 discussions that we're having here today.
19 I'll tell you the bottom line up front and then John 20 will go through and discuss that with you. Apparently, we 21 don't really see any overall trends among the operating plants 22 to move the MTC in really one direction or another. It's 23 relatively constant from that period of time.
24 In '79 the basis of calculations that were done for 25 the ATWS rule were conservative at that point in time. We
(. _
Heritage Reporting Corporation (202) 628-4888
93 i
1 believe that the current operating configurations that are 2 within those bounds are safety analysis work we do in the 3 licensing world, and the FSAR's that considers positive MTC' 4 events, we continue to look at those.
5 We discussed this information with the NRC back in-6 January and the feedback that we got from them is they felt the 7 information provided was responsive to their June 12, 1987 8 letter. We will be presenting essentially the same information 9 that we provided to them at that time frame.
10 I would like to introduce at this point, John 11 Kaprinos from CE and let him go through the discussion.
12 (Slides being shown.)
13 MR. KAPRINOS: In reviewing the handouts this morning 14 I noticed that a few of the graphs washed out. We had some, we 15 brought down from Windsor this morning and give Paul, they're 16 primarily the plots, so he can get better copies to everyone.
17 Basically, this presentation is responding to the i 18 agendo items. The purpose is not to go into a lot of history 19 on the ATWS rule, but brief chronology. We received the NRC 20 letter June 12th on the information request for ATC's. We've 21 had numerous telephone conversations with the NRC staff, Dr.
22 Thodone and his staff.
23 Sometime mid '87 the Owners had combustion review MTC 24 data and evaluated the impact on the ATWS analysis. The plan i
25 that we took was, and this was in response to an earlier ACRS i 1
(~
Heritage Reporting Corporation (202) 628-4888 :
94 1 agenda also, what are the current trends in MTC? How does it 2 impact the analysis, results that are on the books right now?
3 And the middle one there, what would the MTC be if we were 4 going to perform an ATWS analysis today?
5 As Ted said, we discussed this information with'the a 6 NRC mid-January, and this is basically the same presentation we l 7 gave them.
l 8 MR. KERR: I would think it would be improved the 9 second time around.
10 MR. KAPRINOS: Well, we did change it to respond to l
l 11 the agenda items; and we did add a few more data points. So l l
12 it's more -- the first one we're responding to Thodone's j 1
13 letter, the second is to your agenda and we did get some data l 14 points from our owners.
I 15 With regard to current trends at MTC, what we were j i
16 doing in we went back with out past cycles, the cycles that are l
. 17 operating now and future cycles. We also looked at operational l
l 18 data. And this is sort of leading up to the ATWS rule. Back j l
19 in '79 Combustion did not have many plants operating, so we did i 20 take in the database from the last 10 years. ,
l 21 We looked at some bias analytical calculations. And 22 these are bias -- you have the analysis, then you've got your ,
23 database, and you adjust your calculations for MTC predictions.
24 We also looked at hot zero power and hot full power 25 data. And then we took -- put all this on a database and l
l Heritage Reporting Corporation (202) 628-4888 1
95 1 sorted it on various fields to respond to the information that 2 you were interested in.
3 MR. KERR Excuse me. Am I the only one who doesn't 4 know what a bias analytical calculation is?
5 MR. WARD: I'll explain it to you at. lunch.
6 (Laughter) 7 MR. KAPRINOS: In looking at the database what we did 8 is, we went back to 1973 and picked a representative sample of 9 the Combustion plants. As you can see there's quite a scatter.
10 The triangles there indicate the hot zero power MTC's and the
, 11 circles are hot full power.
12 What we did with this database is to sort it on 13 various categories. One of the questions that was asked in the 14 agenda is, what if the function of MTC -- how does it vary with 15 cycle length? Well, we looked at that and we said, well, we 16 really don't see any clear trend upwards as you go with cycle 17 longth. There is a acatter on both sides, and we did not see ;
18 any great trend. l 19 The next question ---
20 MR. KERR: Excuse me.
21 MR. KAPRINOS: Yes, sir. 1 22 MR. KERR What do you mean by change in cycle 23 length; how much of a change?
24 MR. KAPRINOS: What we did is, we used actual cycle 25 length data in terms of effective full power days. And so we ,
i i.
Heritage Reporting Corporation l (202) 628-4888 l
l 96 l t i 1 have 11 months to two year cycles on here. And we don't see, ;
2 if you went from, let's say, the 12 month cycle up to the two 3 year cycle, we don't see any great trend upwards in the 4 scatter. Now, we do have plant specific we will present.
5 MR. KERR: Oh, I see. You're saying the scatter is 6 the same, but the --
7 MR. KAPRINOS: Scatter appears to be the same whether 8 you're 12 months or 24 months.
9 MR. KERR: But the data are not in the octual 10 moderate temperature coefficient are not the same.
11 MR. KAPRINOS: No, they're not the same, but we're 12 looking at a trend. The emphasis of this slide or our 13 presentation is to say, we don't see any great positive trend 14 in MTC's. And you can ask my fuel's guy questions on that. ,
15 MR. KERR: For a minute you were talking about ,
16 scatter in the data. Now, you can have about the same scatter 17 in cata and have data points that are different.
18 MR. KAPRINOS If you went and looked at, let's say, 19 worse case which is what we will got to later, or not worse 20 case, you could say -- let's say, this is all the same plant, 21 you've got a very definite positive trend.
22 What we found out in looking at the individual plants .
23 on a cycle by cycle basis is that we don't see this upward 24 trend.
25 MR. KERR I'm going to wait. Go ahead.
i Heritage Reporting Corporation (202) 628-4888
97 1 MR. LIPINSKI: This is experimental data?
2 MR. KAPRINOS: This is a biased analytical plus some 3 measured; is that right, Jeff. Jeff Brown.
4 MR. BROWN: Jeff Brown, Combustion Engineering. Hot 5 zero power is mostly measured data, and the hot full power is 6 nostly calculated. We were attempting to show here the most 7 positive limits in most cases. The full power measurement is 8 done only after some amount of core burnup and in some cases 9 not at full power. So that's what you have there; it's kind of 10 a mixture, but there wouldn't be that much difference if you 11 had showed measured or calculated or both, we had them 12 available.
13 MR. LIPINSKI: But with your calculated data you get 14 scatter?
15 MR. DROWN: No, that is measured data there. Each 16 data point represents a different plant on different cycle.
17 MR. LIPINSKI: That was my first question. Am I 18 looking at experimental data or calculated data?
19 MR. BROWN You're looking at --
20 MR. KAPRINOS: One of the differences between the B&W 21 design and ours is that we have plants with essentially three 22 different fuel types on here; you have 12 month, 18 month and 23 24 month cycles as opposed to everyone being at 18 or everyone 24 being at 24.
25 So we're saying, this is a mix of all the Owners Heritage Reporting Corporation (202) 628-4888
98 1
1 Croup plants.
2 MR. LIPINSKI: But the point is, each point up there 3 is taken from experiment on a particular reactor?
4 MR. KAPRINOS: Yes. ~
5 MR. LIPINSKI: Because he came back and answered some 6 was analytical.
7 MR. KAPRINOS: On the full power.
8 MR. LIPINSKI: You've got full power points spotted 9 up there, are they experimental or are they analytical?
10 MR. KAPRINOS: They are, to the best of my knowledge, 11 biased analytical.
12 MR. LIPINSKI: So then what I'm looking at here is a 13 mix of experimental scattered and analytical scattered.
14 MR. KAPRINOS: For various plant types of fuel design 15 cycle lengths.
16 MR. KERR So you're telling me that for a 24 month 17 cycle plant there was no change in moderator temperature 18 coefficient from one 24 month cycle to the next 24 month cycle; 19 is that right?
20 MR. KAPRINOS: We currently do not have any 24 month 21 cycles operating. There are some projections that we have for 22 these 24 month cycles, and admittedly there are other fuel 23 vendors involved, so we don't have all the information for all 24 the 24 month cycles.
25 MR. KERR So in the 18 month cycles you have some i
(
l Heritage Reporting Corporation (202) 628-4888
)
I j
I 1
1
l l
99 l l
1 that are operating, from an 18 month cycle? l 2 MR. KAPRINOS: .The 3800 class plant system 80's are !
I 3- around 18 month cycles.
4 MR. KERR So what you're telling me is that, the 18 5 month cycles don't show any trend in increased moderator 6 temperature coefficient as long as I keep them at 18 months. !
7 MR. KAPRINOS: Well, they don't show any trend. Let 8 me show one more slide and then we'll get into the plants, and 't 9 I think that will help out.
i 10 MR. LIPINSKI Before you take it away.
11 MR. KAPRINOS: Yes.
i 12 MR. LIPINSKI: That upper right hand triangle, I
13 experimental or analytical at 24 months?
i ,
14 MR. KAPRINOS: That I believe is a 24 month cycle. I 15 don't think it's operating, so it's analytical.
16 MR. LIPINSKI: You said that the zero power were 17 measured, yet you have thrown that up there. ,
j !
18 MR. KAPRINOS: We can X out that point. The purpose 19 of that was to say, all right, this is what we expect it to be
]
20 for a 24 month cycle. Without operational data it's tough to 21 give it to you.
22 MR. LIPINSKI: Then there's a mix. Not everything is 23 black-and white', you have to know which are -- specific points 24 are analytical and which specific points are experimental.
25 MR. KAPRINOS: The purpose of the first three slides
( )
Heritage Reporting Corporation !
(202) 628-4888 l i 1
100 t
I was to show scatter and to prove a point in terms of trending.
2 Joe Williams from Baltimore.
3 MR. WILLIAMS: Yes, my name is Joe Williams,-I'm with 4 Baltimore Gas and Electric. We do have an operating 24 month .
5 fuel cycle for hot zero power. We would set a measurement
~
6 within, I believe, plus or minus .2 of the analytical point.
7 So I don't know if that point is analytical or 8 measured. But that gives you some idea of what the difference 9 would be if that is an analytical point versus what the measure 10 would be. That would be the maximum.
11 It has been my experience that the measured is 12 generally much closer than that.
13 MR. WARD: John, it would help, what are you trying 14 to show, again, with these first slides; you're trying to show, 15 you said, that there's a lot of scatter and you're trying to 16 show that there's no trend with cycle length?
17 MR. KAPRINOS: Right.
18 MR. WARD: Are those the two points we're suppose to i 19 get out of this?
20 MR. KAPRINOS: There's no trend with cycle length, 21 and the third piece that was requested on the agenda was, is
- l 22 there any trend -- any current trend with regard to tech specs?
23 So what we have here is the. tech spec values for the 24 various plants for the last 10 years. And what this shows is 25 that plants have been requeeting certain tech specs for a long
(
Heritage Reporting Corporation (202) 628-4888 ,
l
l 101 l 1 time, positive MTC's at hot and zero power -- hot full and zero 2 power.
3 So it's not like we've, you know, started to creep up 4 with extended fuel cycle designs or longer fuel cycle designs.
5 MR. WARD: You're going to have to explain to me what 6 you mean by this, plants that are requesting tech specs; I 7 don't understand what you're saying?
8 MR. KERR: Dave, we, Paul and I, in discussing this 9 got the impression that there had been some change in the 10 plants which have previoinsly operated under one tech spec.
11 We're asking for a decreased negative coefficient or perhaps an 12 increased positive coefficient at zero power. And we were 13 trying to find out if this were generally true or restricted to 14 one or two plants or whatever.
15 MR. KAPRINOS: That's what we're trying to show here.
16 MR. EBERSOLE: May I ask a general question.
17 MR. KAPRINOS: Yes.
18 MR. EBERSOLE: I see positive in the MTC's. If I do l l
19 a parametric on maximum mechanical loading versus power level, j i
20 where do I get the maximum mechanical pressure loadings as a l
> , \
21 function of reactor power?
22 MR. KAPRINOS: For?
23 MR. EBERSOLE: Whatever you want to call it, the 24 worse or the average.
25 MR. KAPRINOS: For which event? Which type of event? l 1
(
Heritage Reporting Corporation (202) 628-4888 ;
l i
l I
I
102 1 MR. EBERSOLE: Oh, I'll pick the field events, I'm 2 just varying parallel. Wouldn't it be at part power, like -
3 half, 75 percent, with some head room, you know, to go up.
4 MR. KERR: Do you understand Mr. Ebersole's question?
5 MR. KAPRINOS: I guess I would have to respond saying ,
6 that, if you were looking at a -- it has been determined, and 7 I'm not sure if there's a new movement on to do part power 8 events. But determine that if you were at the zero power you 9 would run the worse MTC and full power the worse MTC for that 10 type of event whether it's a --
11 MR. EBERSOLE: If I just -- well, just let me pick 12 something. At 50 percent power would the full field of ATWS 13 events, do I get higher mechanical pressures than I would at !
14 full power before I turn around?
15 MR. KAPRINOS: I would suspect not. We have not done 16 parametrics on that.
17 MR. EBERSOLE: Well, it would appear to me you would
~
18 g.t worse mechanical loadings at part power than full power.
19 Just from the display of your data Pere. That's where your MTC 20 show positive, most posi.tive.
21 MR. KAPRINOS: But you've got the tradeoff in terms ;
22 of power level where you're drifting down. So somewhere in ,
23 there, let's say, I think this drops fairly quickly.
24 Jeff, the time span between hot and full, these !
25 values are like -- are they eight days or eight hours?
(
Heritage Reporting Corporation (202) 628-4888 ;
103 s
1 MR. BROWN: Typically during startup, typical times 2 are from say, fivo to 10 effective full power days, burnup upon 3 -reaching initially full power --
4 MR. EBERSOLE: I understand. But if I take a power 5 transient from 50 percent power other conditions being equal, ,
6 do I not get a higher mechanical pressure than I would if I 7 took it at full power.
8 MR. KAPRINOS: But you were considering an ATWS 9 event, everything but the core is at best estimate conditions.
10 So you would have to adjust your system to 50 percent power. ,
q l' You would probably have more inventory, have more feedback 12 effects on the heatup. ,
13 MR. EBERSOLE: I would have more room in which to
(
14 rise to high power before things begin to relieve. Am I
! 15 wrong?
16 MR. FENNELL: Pat Fennell from Combustion 3
17 Engineering. You're going to relieve based on pressure down to 18 the power that you have.
I 19 MR. EBERSOLE: I'm thinking about the rate which I'm i 20 summing up. I don't want to run through.
l 21 MR. LEE: To me certainly it is not inconceivable l 22 that, starting at high power or low power could result in most t'
23 severe pressure.
24 MR. EBERSOLE: Yes. It's like an ancient thing ,
1 25 coming out of the black hole, things go right on through into t
Heritage Reporting Corporation (202) 628-4888 1
l
104 1 trouble without even pausing.- And I'm only pursuing that same 2 line of thought.
3 MR. FENNELL: I believe that for ATWS the thing that 4 is-driving the power -- driving the pressure up is the 5 imbalance between the power being created in the core and the 6 way in which power is being removed. And hence, I think at a 7 higher power is running at a higher pressure.
8 MR. KAPRINOS: Because your balance -- let's say, 9 your safety valves are worth 30 percent of the core power., your 10 feedbacks would drive you down to 30 percent.
15 MR. EBERSOLE: It would seem that you would do some 12 calculations as a function of parallel level to come to grips 13 with what were the maximum mechanical pressure" 14 MR. KAPRINOS: I guess I would have so say the ground 15 rules for ATWS were estaolished at full power.
16 MR. EBERSOLE: Well, whatever they were, I'm going 17 back --
18 MR. KAPRINOS: There were also some zero power --
19 MR. EBERSOLE: I didn't say zero power. I said 20 somewhere in between. Zero power is so far down. It might be 21 that's worse, I don't know.
22 MR. KAPRINOS: I don't think it is, but I cannot tell 23 you we have done the calculations.
24 MR. EBERSOLE: Thank you. >
25 MN. KAPRINOS: In the CE Owners Group for the
(.
Heritage Reporting Corporation (202) 628-4888 3
105 1 purposes of ATWS analysis we have categorized the plants into 2 three different classes. They're the 14 by 14 plants, these 3 are primarily the 2750 plants. The 3410 megawatt class which 4 has 16 by 16 fuel. And then the 3800 plants which is the 5 system 80 plants.
6 In looking at the data that we saw previously on the 7 first three slides, what we did is to connect the various 8 plants in terms of cycles. So this would be a plant in terms 9 of its MTC, another one, another one, and another one.
10 What we're seeing is that, there does not seem to be 11 any definite trend upwards. There is scatter up and down on a 12 plant by plant basis. We found this for the 14 by 14 plants.
13 Now, with regard to the 3410 class which are CEOGs in 14 Louisiana, and I believe Arkansas is in here, also, because of 15 their fuel design, for the CEOGs unit we have less operating 16 data so we don't see the scatter. We start at '78. We do have 17 some predictions out to 1988.
18 We frankly don't see any great trend in there in 19 terms of going up or down, it varies on a cycle by cycle basis, i
20 MR. KERR: Mr. Lee. !
21 MR. LEE: Relative to what we have seen early in the 22 morning in the B&W presentat).on, Owners Group presentation, how 23 did you accomplish the increase in cycling that does not give l l
24 you an increase in the HTC value? l 25 MR. KAPRINOS: I'll defer that to Jeff Brown.
(
Heritage Reporting Corporation l (202) 628-4888
106 1 MR. KERR: Do you understand the question?
2 MR. BROWN: Yes. Simply by adding burnable poison at ;
3 cycle increase. We tend to have a target to MTC, that is, 4 essentially we just design it so that we maintain a comfortable .
5 margin to make sure that -- by comfortable margin that the MTC 6 'at full power is negative. And we use that same design target 7 for all of our cycles no matter what the cycle is. Therefore 8 if you keep'the same target it's just s.Lmply a matter of adding 9 traditional burnable poisons, so you achieve that goal.
10 MR. KERR: This is fixed burnable poison in the core?
11 MR. BROWN: Yes.
12 MR. KERR: Thank you.
13 MR. LEE: Thank you.
14 MR. KAPRINOS: And with regard to the system 80 15 class, the plants are new, we have limited data. But I believe 16 that this is unit 1 and things are a minimum stabilizing. I'm 17 not suppose to identify plants.
18 For another system 80 unit we seem to be following 19 the same trend. So we would see no great difference for the 20 system 80 class plants in terms of concluding that there is a 21 positive trend in MTC.
22 Therefore, to summarize this part of the 1
23 presentation, we do not see any overall trend towards more 24 positive MTC at full power, looking at all the data that we 25 have. In terms of tech specs this does not appear to be a new j
(
Heritage Reporting Corporation (202) 628-4888 i
l
107 ,
1 request. It has been reviewed, it's in our reload reports.
2 There appears to be no strong correlation to cycle 3 length with the information that we have, and with some 4 projection.
5 We do have design. goals to control the MTC at full 6 power, and these are listed there. And that it's our Owners 7 Group opinion that the concern about a trend towards positive 8 MTC for the CE Owners plants does not appear to be valid, based 9 on the information we have looked at; and that's been data for 10 the last 10 years.
11 MR. LFE: Just a question. I guess I missed a point 12 you made on the system 80 plant trends. There the MTC values 13 have become more positive or less negative.
I.
14 MR. KAPRINOS: Well, I'll let Jeff bail me out of 15 this one. But system 80 was a new design and it was 18 month 16 cycle. There was a lot of testing that went on early on in the 17 cycle.
18 MR. BROWN: I think the reason is that, you see on 19 here essentially section number one, the initial core for the J
20 three units, there is -- I mentioned about the margin of 21 comfort for design. We tend to make that margin of comfort 22 somewhat larger for the initial core, simply to accommodate any 23 surprises in terms of uncertainties or biases or something like 24 that.
25 So what you see here, you see a trend between initial l
I ,
Heritage Reporting Corporation (202) 628-4888 I
i l
I
108 1 cycles which tend to be more negative because you wish to 2 accommodate a larger margin and a typical reload cycles.
3 I think here if you throw out the first cycle point 4 and look at cycles two and three here, you'll see that there is 5 no trend in fact; there's a downward trend.
6 MR. LEE: Was a generic ATWS analysis performed in 7 terms of MTC values?
8 MR. KAPRINOS: We'll be getting to the ATWS section-9 next.
10 The next part of the process was to determine the 11 impact of the 1988 or current MTC values on the analysis that 12 was performed back in '79 and then again in '82. What we did 13 is, admittedly back in '79 we did not have a lot of operational 14 data, so we went and looked at operating data for various 15 representative plants.
16 We went and reviewed the work that was done back in 17 '76, '78 for -- to determine this MTC value, what was the basis 18 for it. We then said, all right, what would the new pressure 19 be estimated. And then we had drawn some conclusions on what 20 we had found.
21 We won't belabor this plot, but this is very similar 22 to what Ted Enos put up in terms of variation of MTC during 23 cycle length. This is for a typical CE plant.
It's not meant 24 to be quoted or anything, it's just we're showing the i
1 25 variation. It's very comparable to the B&W design fuel. l
(
Heritage Reporting Corporation !
(202) 628-4888 l 1
109 1 In terms of the basis for the 1988 ATWS MTC, we used 2 full power initial conditions as was done back then. The event 3 was the loss of feedwater event which was determined to be in 4 most limiting in terms of pressure.
5 The 95 percent value was 95 percent of the time spent 6 critical. And then we did factor an actual plant data.
7 MR. LEE: Can I go back to the previous, the one 8 before that one where you had the part of MTC as a functional 9 cycle length. That one. When you say 95 percent you are 10 looking at something like minus .5, something like that MTC 11 value; is that what you are saying?
12 MR. KAPRINOS: Yes.
13 MR. LEE: And that was used in the 1979, '80 series 14 of ATWS analysis?
15 MR. KAPRINOS: I think you're jumping ahead to a i
16 slide. But in '79 minus .63 was used. And you're right, it's 17 about minus .5 now.
18 MR. LEE: Sorry, I was jumping ahead.
19 MR. KAPRINOS: Now, this is typical. Remember that 20 we still have three plant classes and each of them have 21 different design fuels and different cycle lengths.
22 As I said, we did review this data by plant class.
23 We looked at reload analysis reports; plants physics data; core 24 reports. And when necessary these were supplemented by best
~
25 estimate calculations, primarily for the newer plants not
(
Heritage Reporting Corporation (202) 628-4888
110 1 having a big database on them.
2 What we arrived at are the two columns here. The 3 1979 value was the last big report we submitted to the NRC on 4 the ATWS issue. You can see the values were -- well, they are 5 listed there.
6 Because in '79 the 3410 and 3800 classes were not 7 operational or fully operational, these were estimated with 8 previous operating data. What we did in looking at it recently 9 is that we did come up with, for the 2750 class that the MTC 10 that would be used for an ATWS analysis is slightly more 11 negative. 3410 and 3800 are slightly more positive or slightly 12 less negative rather.
1 13 MR. KERR When you use the term, the estimated with 14 limited operational data, does that imply that those numbers 15 came from a combination of calculation and the limited data ;
16 that you had?
17 MR. KAPRINOS: Yes. t 18 MR. KERR: And now with operation, and I presume the 19 same calculation, you get the numbers on the right.
20 MR. LIPINSKI Based on the previous scatters that 21 you have shown and the single point values that are shown here, 22 what's the relationship? In one case you're showing the data ,
23 is scattered all over, pick a number, and now here you have 24 these numbers.
L 25 MR. KERR It's labeled ATWS analysis MTC, I would
(
Heritage Reporting Corporation i (202) 628-4888 i 1
i I
111 1 have assumed that meant those are the numbers he used in his 2 ATWS analysis.
3 MR. KAPRINOS: That's correct.
4 MR. LIPINSKI: Now, he's got one for 1988 and he's 5 showing the scatters. So how is he picking an even .57 from 6 these scattered.
7 MR. KAPRINOss Jeff, do you care to --
8 MR. BROWN: Yes. I think if you go back to the slide 9 that had the scatter on that you're talking about, and since 10 the analysis was done by plant class you draw sort of a cited 11 average there about where you're at today, 1988 -- beginning 1-12 88. You'll see that takes you, for instance, say it would tell 13 you, say, around the point of discussion, plus .3 at zero l l
14 power.
15 Then if you go back to the figure that showed the 16 variation of MTC, percent of cycle length. And using the 95
. 17 percent of cycle length value there, you can cite there what j 18 the value will be, you would see that the value reported here 19 for 1988 is very consistent with that method of using the 20 average for the plant class and taking credit for the 95 21 percent burndown.
22 In actually the value determined here was based on ;
23 looking at actual power history from various plant cycles, and 24 trying to do a little bit better job in terms of making the 25 actual plant history and seeing what effect this had. It
(.
Heritage Reporting Corporation (202) 628-4888
112 1 turned out it has a very minimal effect.
2 I would point out that these are what we consider 3 best estimate today in plant class. Our best knowledge we did 4 on an average plant basis.
5 MR. KERR And with these best estimate numbers 6 remind me of the peak pressures you get in your worse ATWS 7 situation. Are you getting to that?
8 MR. KAPRINOS: I knew you were going to get to'that.
9 What we did to estimate the impact of these different 10 MTC's is we went back to the '79 analysis. Now, this is loss 11 of feedwater ATWS. No turbine trip, I believe there was 12 auxiliary feedwater that had a minimal effect on pressure.
13 The first column shows that the pressures that were 14 calculated back in '79 using NRC reviewed models with 15 admittedly some controversy as to assumptions, methods, et 16 cetera, et cetera.
17 In '82 when the rule was being published the owners 18 commented on the rule and we updated our analysis using some 19 better methods on the secondary side and crediting the turbine 20 trip that was proposed in the rule. What we found was a net 21 decrease in peak pressure using these new methods.
22 Recently what we did is, we went back and looked at 23 the '79 report which there was a parametric on MTC versus peak 24 pressure. And based on the MTC's from the previous slide we 25 came up with an estimate of what the delta would be for using
( I l
Heritage Reporting Corporation (202) G28-4888
1 113 1 the new MTC. And they are relatively small as we expected.
2 Our bottom line is that the net change, if we had to 3 do the analysis -- the net change from '79 to today would be a 4- net decrease in pressure using a less negative MTC with 5 improved methods, et cetera, et cetera.
6 In-fact, I would suspect that if we had to do a 7 formal analysis then the system 80 class would be under the 8 level C stress limit; they're fairly close right now.
9 MR. KERR Mr. Lee.
10 MR. KERRt For the plants 280 PSI pressure increase, 11 peak pressure increase is associated with and due to a change 12 in MTC, one times 10 to the minus 5, delta K, am I correct?
13 (Continued on next page.)
14 15 16 17 l l
18 19 20 21 22 23 ,
24 i 25
(-
Heritage Reporting Corporation ,
i '(202) 628-4888 l 1 :,
I
i 114 1 MR. KAPRINOS: Yes.
2 MR. LEE: Approximately.
3 MR. KAPRINOS: Approximately.
4 MR. LEE: That is a small change, and the 5 uncertainties associated with measurements and calculations are 6 considerably larger than I would like to think, that number, 7 that change, that one to the minus fifth.
8 So if you take the uncertainties and scaling the data 9 points, and indeed you plot it, based on your calculations and 10 so on, would you be comfortable with just this 200 psi number?
11 MR. KAPRINOS: I would say that there was uncertainty 12 in the 200 psi delta than there was in the initial pressure 13 back in 1979 because of operating experience. So I would feel 14 more comfortable saying that it will only go up 200 psi, i 15 because we have looked at operation data, and we used the same 16 ground rules as we did back in 1979. So we had to be 17 consistent. We can change a lot of other things. ,
I 18 MR. LEE: I am not talking about the rules as such, i 19 I am asking more a philosophical question, if you will, you as ;
20 an owner of a power plant. There is a good bit of uncertainty
- l 21 regarding the MTC measurements and calculations. So this 200 22 psi pressure increase could double or triple up, if you would, 23 to account for some of these uncertainties.
24 Do you feel comfortable about not worrying about 25 those things?
I, Heritage Reporting Corporation (202) 628-4888
115 1 MR. KAPRINOS: Well, I am not an owner, I am a 2 designer.
3 MR. LEE: Okay.
4 MR. KAPRINOS: And if the owners want to say 5 something on that, that is fine. My reaction is that there is 6 a strong feeling that ATWS should not be design basis. It is 7 sort of like what sort of might really happen. I am not 8 putting that very well. So you have got to get rid of the 9 uncertainty aspect somewhere. The uncertainties are all in the 10 Chapter 15 events.
11 MR. KERR: Look, we are trying to talk about our 12 ability to calculate a hypothesized accident, and not about 13 Chapter 15. I mean maybe you are forced into a Chapter 15 14 mode, and I know that you are in many cases. But we are also 15 interested in what might happen in an extreme situation. So 16 help us. You are an engineer, I guess, I mean you are. You 17 have some idea what we are trying to get at, I think.
18 MR. KAPRINOS: I guess that the answer to both 19 questions is that I have a good deal of confidence in these 20 deltas. These are better methods. There is an ATWS mitigation 21 system in there with the turbine trip, and these deltas are 22 based on better data.
23 MR. KERR: Is not a delta based on a point value of 24 the moderator temperature coefficient?
25 MR. KAPRINOS: No. Our reactivity feedbacks are
(
Heritage Reporting Corporation (202) 628-4888
116 1 based on density. .
2 MR. KERR: You see, it is based on a peint valuo.
3 You have not taken any spread into account in saying the plus 4 200 psia. You have taken a value for moderator temperature 5 coefficient, and that permits you to calculate a value for the 6 delta. Now there is uncertainty in that change in moderator 7 temperature coefficient between what was 82 and 88 has an 8 uncertainty that is two to three times the value of the change 9 that it could have from what we saw. And then it seems to me 10 that 200 psia might have a similar uncertainty.
11 MR. KAPRINOS: Jeff, can you address she 12 uncertainties in the ATWS.
13 MR. BROWN Our documented calculation uncertainty, i 14 calculated minus measurements differences that we have seen, 15 mostly falls within .1 of the measured value, and falls mostly 16 within .1. I think that the order of uncertainties is on the 17 order of the change that we are seeing. The measurement of 18 uncertainty -- well, since we are dealing with the calculated 19 minus measurement, I think that roughly might encompass the 20 measurement also. The measurement of uncertainty is also 21 within .l. Zero power conditions where you control the test 22 conditions.
23 So I would say that since the uncertainties are on 24 the same order of magnitude of the change, I would probably 25 feel quite confident.
(
Heritage Reporting Corporation (202) 628-4888
117 .
i 1 MR. KERR: Are you saying that you would be confident ;
f 2 with a delta of 200, plus or minus 2007 3 MR. BROWN: Yes, i
4 MR. KERR: Thank you.
5 MR. EBERSOLE: I realize that we are here to talk 6 about the deltas. But inevitably, the absolute comes back and 7 hits you in the face all of the time.
8 MR. KAPRINOS: That is why I presented it.
9 MR. EBERSOLE: So I see that you put the numbers up 10 there, and I wonder if you might just review the structure of 11 pressurized and relief.
12 Do you also stretch bolts? i 13 MR. KAPRINOS: i ts .
4 14 MR. EBERSOLE: And do you have neck-down bolts, or do i 15 you take the stretch in the threads?
16 MR. KAPRINOS: I am in the mechanical area. From a
- i
. 17 thermohydraulic standpoint, the modeling looks like a safety i
18 valve characteristic. !
i 19 MR. EBERSOLE: Well, that did not tell me what I l
\
20 asked. Because you know, a neck-down bolt looks like a safety 21 valve spring than a constant cross-section screw through. It 22 is more amenable to stretching. But anyway, you do a ,
23 considerable more stretching than B&W does evidently in the 24 pressure here. l l
25 In the course of the pressure transient, you also j
(-
Heritage Reporting Corporation (202) 628-4888 1
I
i 118 4
1 take credit, and you have got these big wet boilers to do it, 2 of heat rejection to the secondary, I am sure, do you not?
1 3 MR. KAPRINOS: The secondary starting to recover.
4 MR. EBERSOLE: But in the years since then, you got 5 rid of the PORBs, but you did not count on then anyway, did 6 you?
7 MR. KAPRINOS: I do not believe that the early plants 8 used PORBs. They may have, but I am not sure.
9 MR. EBERSOLE: But your new plants do not have any.
10 MR. KAPRINOS: Right.
11 MR. EBERSOLE: Which is a controversial matter in ita 12 own right.
13 In the course of the pressure rise, do you get actual I.
14 boiling around the pins, which if you have a slight positive 15 coefficie - ould make matters worse?
16 MR. LEE: They do not have positive MTC at all.
17 MR. EBERSOLE: They do at more power, 18 MR. LEE: I mean they do not calculate low power 19 transients.
] 20 MR. EBERSOLE: Well, I know, but I askud awhile ago 21 should they not. I do not think that I have got an answer to 22 that yet.
23 MR. KERR But Jesse, if you have to moderator hot l
24 enough to boil.
25 MR. EBERSOLE: Well, if it is positive, and I have
(
1 Heritage Reporting Corporation (202) 628-4888
i i
119 i
(
I got it hot enough to boil. l 2 MR. KERR Yes, but it is no longer positive if it is -
l 3 that hot.
4 MR. EBERSOLE: If it is a direct measure of moderator 5 temperature, that is right, I agree. !
6 Anyway in the course of pressurizing and fluid !
7 expansion, in the course of this pressurizing, do you have i
8 liquid flow out of the spring safeties, and is there a 9 mechanical shock when you enter the liquid flow regime if you 10 do, do you follow me?
11 MR. KAPRINOS: Yes.
12 MR. KERR We will accept the answer I do not know. ;
13 MR. KAPRINOS: Well, concerning the mechanical shock, l
( ;
14 I can tell you what the thermohydraulics look like.
15 MR. KERR There is some possible assistance over !
i 16 here, if you want to make use of it. !
17 MR. FENNELL: This is Pat Fennell from Combustion.
18 We do assume that the safety valves would discharge 11guld on i 19 the pressurized. ;
20 MR. EBERSOLE: Is there a potential of spiking i 21 pressurized when water encounters the valve?
22 MR. FENNELL: This spiking of pressurized, this spike !
23 in pressure is more because you are trying to compress liquid. j 24 MR. EBERSOLE: I know that, I know what it is for. I .
25 asked if it was a significant spike when you are starting to (L l Heritage Reporting Corporation
{
(202) 628-4888 :
l s
4
_ - _ _ - _ - - _ _ _ - _ - _ _ - _ _ _ . . _ _ _ .J
L 120 l 1 force water through th9re.
1 2 MR. FENNELL: Yes.
3 MR. EBERSOLE: And how many psi is that?
4 MR. FENNELL: I would have to go back and look.
, 5 MR. EBERSOLE: I am looking for the jump, when you l
6 suddenly block the free passage of vapor through.
l 7 Is 200 pounds or 3007 8 MR. FENNELL: In that order of magnitude, in the 9 order of magnitude of maybe 200 pounds.
10 MR. EBERSOLE: 200 or 300 pounds, all right, i
i 11 MR. FENNELL: That is based on, I have not reviewed 12 the reports recently, but it is that order of magnitude..
13 MR. KERR: You recognize that an order of magnitude 14 would be a factor of ten in either direction?
l 15 MR. EBERSOLE: Yes, I understand that. ,
16 MR. FENNELL: That was a bad term to use, 17 approximately that amount. i 18 MR. KERR That sort of bugged me when you taid that.
19 MR. EBERSOLE: Well, he is correct in what he meant, 20 I think.
I 21 I MR. KERR Mr. Lee;. I 22 MR. LEE: Under the ATWS rule or guideline, do you ,
l 23 ;
also perform peak pressures aFsociated with 50 percent cycle l l 24 value of MTCs instead of 95 percent values of MTCs?
25 MR. KAPRINOS: Because of the results that are shown ,
( l l l Heritage Reporting Corporation (202) 628-4888 i
121 1 here, we would not get into the PRA. We did not go back to the 2 PRA arguments in terms of that type of evaluation. We 3 basically said, all right, we have our deltas based on the MTCs 4 that we believe in today, and we stopped the effort there.
5 MR. LEE: How do you address the question of pressure 6 going above the Level C limit?
7 MR. KAPRINOS: That has been a longstanding 8 controversy between the NRC, the ACRS, and the CE Owners Group, 9 and the other vendors. I do not know if this is the forum to 10 get into talking about an excessive level C stress allowance.
l 11 MR. LEE: My memory is foggy, but I thought that at 1
12 some stage the argument to overcome that problem was to use 50 13 percent cycle value MTC rather than 95 percent, so that the 14 calculated pressure would be considerably below even the Level -
15 C limit. But apparently, that is not the case. l 16 MR. KAPRINOS: Okay. Ilrst of all, it is not clear 17 to use where the 50 percent came from, where the NRC got it 18 from. So our position has been, or let me back up one step.
t 19 It is our perception that the NRC does not want the safety 20 injection check valves to remain shut following the pressure 21 excursion.
22 We have maintained that the pressures that we are 23 calculating will not cause these valves to stay shut. So ;
24 therefore, once you depressurize, you will get water back into l
25 the RCS. Now this is going back on history, and there are l Heritage Reporting Corporation (202) 628-4888 1
}
122 1 probably a lot of arguments in this room.
2 MR. LEE: But I guess my question is really at the 3 end of the B&W presentation they used this argument that the 4 Level C limit is not exceeded by more than 50 percent, so it is 5 okay, if I understood correctly.
6 MR. KERR No more than 50 percent of the time.
7 MR. LEE: No more than 50 percent of the time.
8 MR. KERR: Right.
9 MR. LEE: Apparently, in this case, that argument 10 does not have to be made.
11 MR. KAPRINOS: It did not go that far in the work 12 that we did. Because we saw the deltas and said considering 13 everything that these are reasonable. So we did not go into i
14 the PRA arguments or the 50 percent arguments.
15 MR. WARD: I do not think that is a PRA argument, but 16 what is your Level C pressure?
17 MR. KAPRINOS: It is the same as everyone, about l 18 3200. They are all designed for around 2500 psi. l
)
19 MR. EBERSOLE: Is this lock open on the safety valves 20 a common requirement for all of the PWRs, that is when you blow 21 them open that you want them to stay open?
22 MR. KAPRINOS: The only reason that I brought that up 23 is that I think that the last time that we were here a few 24 years ago that the PRA people were here talking about what the 25 criteria was and the safety valves staying shut and things like
(
i 1
Heritage Reporting Corporation (202) 628-4888 l
l 1
123 1 that. And we issued a rebuttal to the SER on the 263 report 2 which talked about the safety valves and the head lift and 3 everything else, and that is sort of history now.
4 So it goes into the 50 percent argument in terms of 5 risk reduction. It you look at the fault tree, it is 50 r 6 percent of the time you are okay, and 50 percent you are not.
7 MR. EBERSOLE: I see.
8 MR. KAPRINOS: So you examine the basis of the rule 9 and you examine it today, and what your analysis would look 10 like. We have showed you the analysis.
11 MR. EBERSOLE: Well, you actually hope to destroy the 12 safety valve, do you not?
13 MR. KAPRINOS: Excuse me.
f 14 MR. EBERSOLE: You actually hope to destroy the 15 safety, just blow their grids out, and then stay open, am I ,
16 correct, so that you can get water back in?
. 17 MR. KAPRINOS: My perception of the NRC concern was 18 the safety injection check valves jamming shut.
19 MR. EBERSOLE: Oh, no. I am talking about the 20 pressure safety valves.
21 Will they reclose, or do they have to reclose?
22 MR. KAPRINOS: I do not think that they have to 23 reclose. ;
1 24 MR. EBERSOLE: On the other side of the coin, are l l
25 they required to stay open? .
I i
t Heritage Reporting Corporation (202) 628-4888
l l l
l 124 1 MR. KAPRINOS: Weil, our analysis assumes that they 2 shut.
l l
3 MR. EBERSOLE: Okay. t 4 MR. KAPRINOS: If they did blow off, then you are in I 5 the LOCA scenario. .
6 MR. EBERSOLE: So you have got to put water in ,
7 against the relief pressure if they shut. I guess that the lid 1 8 comes back down and recedes itself, or does it?
- 9 MR. KAPRINOS: Do you really want to talk about this?
10 MR. EBERSOLE: Well, do you have to reintroduce water 11 at half pressure?
12 MR. KAPRINOS: For the CE design, we would have to l 13 wait. We do not have high head HZPs. So we would have to rely 14 initially on charging, which the charging would happen at about l 15 2500.
16 MR. EBERSOLE: In the transient, do you lose so much 17 water that you cannot get enough back in fast enough?
18 MR. KAPRINOS: I believe that the analysis will run l 19 out to about ten minutes, and the plant was recoverable at that l
20 time, if you assume that your safety is receded and your head
- l 21 receded. l 1
l 22 MR. EBERSOLE: Okay. , j 23 MR. KAPRINOS: That is about as far as anyone went f
24 back then. I 25 MR. EBERSOLE: Yes, okay.
( ,
I Heritage Reporting Corporation (202) 628-4888 l
i e
125 8
1 MR. KAPRINOS: In conclusion, our conclusions on the 2 ATWS impact evaluation is that yes, the MTCs are different.
3 One of the classes has a slightly more negative, and the other 4 two have a slightly less negative. Based on our analysis, we 5 see an overall decrease in pressure from a time before the rule 6 or the basis for the rule was formulated. So we do not really 7 see a problem there.
8 And I guess in sammary on this issue, the owners 9 Group Utilities are responding to the ATWS rule on an 10 individual basis, and that is currently under discussion with 11 NRC.
12 MR. KERR: Are there any questions?
13 (No response.)
14 MR. KERR: Would it be feasible with reasonable 15 engineering design to decrease that peak pressure by 16 introducing additional fixed burnable poison, so that it gets 17 down to semewhere around Service Level C rather than being as 18 high as it is now calculated?
19 MR. KAPRINOS: What does that do for the design, 20 Jeff?
21 MR. BROWN: You are asking if it is feasible?
22 MR. KERR: I am sure that it could be done, but does 23 it cost so much money or make the power level so distorted or 24 whatever?
25 MR. BROWN: I guess that I would reiterate the
(
Heritage Reporting Corporation (202) 628-4888
126' 1 gentleman's from B&W comment that it is feasible, of course, 2 that the burnable poison would. But that would in general at 3 the expense of an additional margin and power peaking. It is 4 economics. ,
5 MR. EBERSOLE: You mean the LOCA?
6 MR. BROWN: By increased LOCA power.
7 THE WITNESS: That would materializo in the LOCA [
8 case, would it not, which is'less significant than it used to 9 be?
10 MR. BROWN: That is a limit, but I am saying that 11 this would be in the power distribution. l 12 MR. DAVIS: D&B problems?
13 MR. BROWN: Bringing it closer to D&B. Less margin, l
(
14 less margin for maneuvering, and things like that. So it is 15 feasible. But at this point, probably a significant decrease 16 would not be desirable. ;
17 MR. FENNELL: The NRC consultants in SECI 83 through l 18 93 looked at adding additional burnable poisons, and their i 19 value impact analysis showed that the value impact ratio is f
20 substantially less than one meaning that it was not cost f i
21 beneficial. You asked whether it was feasible, and I assumed f 22 you meant economic feasibility as well. ,j r
4 23 MR. KERR I did not mean NRC constraint cost benefit i
I meant that it would cost so much that one would
! 24 analysis.
25 not build any more nuclear power plants. And maybe that i i
, ( !
Heritage Reporting Corporation (202) 628-4888 I
l
i 127 1 question does not make much sense. But I was trying to get 2 some idea of whether it would put one outside of the realm of i 3 being reasonable from an engineering point of view, or whether I 4 it would have some economic penalties, but maybe not terribly 5 serious ones. ,
6 Because it turns out apparently that there are some r 7 companies that own both CE and B&W plants. They feel 8 comfortable with the Combustion Engineering design that keeps a 9 moderator temperature coefficient as one goes to a longer fuel 10 life. But with a B&W design, it makes the moderator 11 temperature coefficient change in a way which at least those of 12 us who do not have to worry about economics would prefer not to l 13 see it change, i
4 14 I was sort of curious about how owners rationalize .
15 these two points of view. I guess that it is a balancing of a 16 large number of considerations of which I am unaware. I mean 17 presumably unless there is something markedly different about !
18 the CE cores, that it would be possible for them to keep the 19 moderator temperature coefficient about constant as they I
20 increase core life.
21 And presumably, it would be possible for CE plants to 22 increase or change the moderator temperature coefficient so 23 that one would drop these pressures. So then one has to judge 24 as to whether one really believes that these things are going 25 to happen or not.
( l i
Heritage Reporting Corporation (202) 628-4888
~
E :
t 128 i:
l 1 I guess that I have some lingering suspicion that an 2 ATWS just might occur and not with high probability, But I l
3 -would therefore want to look-for possible ways of designing 4 them out of plants if I could do it in some sort of reasonable 5 way. That is what I am trying to get at.
6 MR. LEE: If I may add my personal comment. I would 7 like to think that there is still a good bit of room in the 8 design that can give us a larger negative value of MTC without 9 getting into a reduction in the thermal margins associated with 10 the LOCA and the manuevers, and things like that. So there is 11 an economic consideration. For the design effort, that has to 12 be done.
13 MR. KAPRINOS: Just to quickly wrap up, another topic i
14 on the agenda and in listening to the conversations this 15 morning, we may have misinterpreted this. But there was a 16 topic there which dealt with --
17 MR. KERRt Let me point out that you may be giving us 18 more credit than we deserve. We were not sure exactly what we 19 wanted to learn.
l 20 MR. KAPRINOS: Well, anyway, this is not anything 21 mind boggling. Positive reactivity events use the limiting 22 value as prescribed in the toch specs with all of the 23 uncertainties. There was a question on analysis uncertainties.
24 I guess that I will have to leave that. That is documented in 25 this report. The material is proprietary, and I.could see no
{
Heritage Reporting Corporation (202) 628-4888 l
h i
129 1 reason to present it having all of those problems.
2 The types of uncertainties .ece reactivity related 3 dealing with control rods, power distributions, et cetera. But 4 they are all documented in that report, if you are interested 5 in our uncertainty methods.
6 MR. EBERSOLE: If I could go to the given condition 7 that you have an ATWS, and then do a conditional PRA on what B happens to the lid and the bolts, how do you come out about the 9 reliability of their moving to relieve pressure and not blowing 10 the head off.
11 Do you look at the structure of that, the thread 12 designs, the metallurgy, the chewing away of the steel with 13 boron and leaking pistil rings.
14 MR. KERR: Let's find out first if they really have 15 done it.
16 MR. EBERSOLE: Yes, right. We are talking about a
- 17 mini PRA in detail given that I have an ATWS and I want to keep 18 a head on it.
19 MR. KAPRINOS: I know that the bolt stretching and 20 receding was looked at by materials people. I am not aware of 21 all of the details. And to put that on your fault tree.
22 MR. EBERSOLE: I am looking at like a PTS, alloys, 23 the works.
24 MR. KAPRINOS: It is probably safe to say that it has 25 not been done.
(
Heritago Reporting Corporation (202) 628-4888
130
(
1 MR. EBERSOLE: Should it be?
2 MR. KERR: I cannot believe that a plant owner who 3 really thought that an ATWS was likely to occur would want to 4 rely on that mechanism to protect his plant or the public.
5 MR. EBERSOLE: Without certainly examining the 6 structure, even after that.
7 MR. KERR: I just have to conclude that people who 8 look at that analysis conclude that an ATWS is so unlikely that 9 one does not have to worry about the results of those 10 calculations.
11 Do you have anything further in your presentation?
12 MR. KAPRINOS: Just briefly. We have responded to 13 the June 12th NRC letter. We had met with them in January, and 1
14 not have received comments back from'them as to their receipt 15 of it or how they received it.
16 MR. KERR: You will recall that in Spril of 3986 that 17 there was a reactivity associated accident in the Soviet Union.
18 As a result of that, did Combustion .gineering 19 reexamine any of its reactivity related accident analyses 20 associated with water reactors?
21 MR. KAPRINOS: I was not personally involved in the 22 follow-up to and the hearings, et cetera involved with ,
23 Chernobyl. I believe that the fuels organization has looked at 24 things, and have testified in front of Congress, et cetera.
25 And I just not familiar with the details of that effort.
(
Heritage Reporting Corporation (202) 628-4888
131 1 Jeff, were you involved in that at all?
2 MR. BROWN: We did do a data MTC survey of a much 3 smaller scope than the kind of thing that had been done here at 4 that time. And on the basis of that, we concluded that both of 5 our MTC and overall power prohibitions were sufficiently 6 negative, you know, so as not to be of concern. And we did 7 convince ourselves that in terms of the feasible positive 8 reactivity transients that the Chapter 15 analysis addressed 9 those sufficiently.
10 As John mentioned, typically, the Chapter 15 analysis l' does assume the range of operating conditions including range 12 of allowed tech specs, MTC. And of course, the positive MTC is 13 used in circle events.
14 MR. KERRt Well, that is an answer, and I do not mean 15 to be critical of the answer. You know, the design basis 16 accident analysis approach which was developed after much 17 thought and careful consideration and a great deal of 18 experience was one which in effect assumed that serious 19 accidents were so unlikely that they did not have to be 20 seriously considered.
21 And we used things like single failure criteria, 22 which you know about much more than I probably, having looked 23 at it countless times and with the publication of WASH-1400 and 24 some subsequent events, which reinforced some of the 25 conclusions.
(_
Heritage Reporting Corporation (202) 628-4888
I l
l 132 1 1 I think that it was reasonable to suppose that things !
2 other than design basis accidents perhaps should be examined, 3 and indeed other scenarios. Both TM-2 and Chernobyl were 1
4 serious enough that it seems to me that all of us should look l l
5 again.
6 I am simply asking did you people do it. I am not 7 trying to be critical. I would have thought that you would )
8 have wanted to ask yourself have we really neglected some i
9 things that might be looked at again.
10 MR. BROWN: I think that to the best of my knowledge 11 that the answer is yes, that there was some thought given about 12 it. And of course, this type of scenario with TMI. And the 13 answer was no 14 MR. KERR: It is trivi, .o assume that yes, that it 15 was a reactor that had a lot of inherent characteristics that 16 we would build, and all that is true. But there may be some 17 applications of the thing that we should be applying. And I am 18 not claiming that I know what they are or that anybody does.
19 But I do think that we need as a community to look for them.
20 MR. BROWN: That is why we are here, I guess, rigi.t 21 now.
22 MR. KERR: If your reactors depend on guidance from 23 ACRS to be safe, I do not feel so good about them. They 24 probably depend a lot on you people and the people who operato 25 them.
I Heritage Reporting Corporation (202) 628-4888
l l
133 )
1 Are there other questions or comments?
2 MR. KAPRINOS: Even though I wag not J.nvolved.in the 3 effort, I do not think that we treated the Chernobyl event in a
- 4 trivial fashion saying, well, it is different. I think that-5 things were looked at.
6 MR. KERR: That sort of thing probably is not 7 available to people who are curious. If, for example, I came 8 to you not as an ACRS representative but as a member of the 9 university community and wanted to find out what you were doing 10 informally. Well, I will not pursue.it. But just as a 11 learning process, I would be curious.
12 Is there anything further?
13 (No response.) $
14 MR. KERR: I think that in considering the time ana j 15 schedule, that this perhaps is a reasonable time to break for 16 lunch. And that gives the next group of presenters a much more 17 receptive and comfortable audience. So we will reconvene at.
18 1:00.
4 19 (Whereupon, at 11:55 p.m., ;he committee recessed, to 20 reconvene at 1:00 p.m., this same day.)
21 22 J
23 l 24 25
(.
Heritage Reporting Corporation (202) 628-4888 i
134 1 AFTERNOON SESSION 2 (1:00 p.m.)
3 MR. KERR: The Westinghouse Owners Group, you have 4- waited patiently through a lot of interesting presentations, 5 and we expect more of the same.
6 MR. NEWTON: Good afternoon. My name is Roger 7 Newton, and I am the current Chairman of the Westinghouse 8 Owners Group. And I am also from Wisconsin Electric Power, the 9 Point Beach Nuclear Plant. And I am the general superintendent 10 of nuclear systems engineering and analysis at Wisconsin 11 Electric Power.
12 (Slides being shown.)
13 MR. KERR: What we would like to cover today are the f
14 five main bullets there and four main items. I will be 15 addressing the introduction and background as far as how the 16 Owners Group is involved in the current ATWS issue, and where l
17 we think we are going from here at least to set the direction '
18 for that.
19 Vinnie Esposito of Westinghouse will cover the fuel 20 management aspects that impact ATWS and the moderator 21 temperature coefficient. And then the other three items, i i
22 application of ATWS rule and bases to Westinghouse PWRS, the i 23 WOG, Westinghouse ATWS rule administration program and the 24 conclusions from that will be covered by Jim Little of 25 Westinghouse.
(-
Heritage Reporting Corporation (202) 629-4888
l 135 1 It has been interesting listening to the previous two 2 Owners Groups and vendors talk about the ATWS issue. Because I 3 have been involved in that or was involved in it all the way 4 through the 1970s. I had to drop out during TMI because of 5 other activities, and I back involved in it again when I 6 thought that I was over with that when the rule was passed.
7 But we are really not here to talk too much about the 8 ATWS transient analysis results, the large number that were 9 generated in the 1970s, and the difference between the 10 different Owners Groups and the Westinghouse plants. And we 11 are really not here to talk about the AMSAC design that was 12 proposed and is in the process of being implemented at 13 Westinghouse. These discussions pretty much culminated with j 14 the rule, and we are trying to implement that right now.
15 What we are really here to talk about is the ATWS 16 rule, and our new look at it because of questions that the NRC
. 17 has asked on that, and our plans to what we call administer the 18 rule. And we kind of had to come up with a name for the new 19 look that we are having at that. And for lack of a better I l
20 name, we have called it our rule administration program. !
21 Just a little bit of a background of why we are' !
22 involved again on this. It is that the NRC in June sent out a 23 letter to all of the PWR Owners Groups asking for information 24 on moderator temperature coefficient. They were concerned in 25 their observations that there might be trends that could
(
Heritage Reporting Corporation (202) 628-4888
136 1 potentially lead to a moderator temperature coefficient that i 2 was inconsistent with the ATWS rule basis. And that they would 3- like to have design information on our operating plants as to 4 what was happening, so that they could evaluate that.
5 For the Westinghouse Owners Group and the ,,
G Westinghouse Utilities with the large number of units and the 7 large numbers of refuelings that have occurred over the years, 8 we were a little concerned that we were going to be getting 9 back inte the transient analysis business and turning out new 10 analyses based upon moderator temperature coefficient which was 11 somewhat different than what the basis of the rule was. So it 12 took us awhile to 'stermine if we were going-to respond to it 13 and in what way. [
t 14 MR. KERR: Should I get the impression that you had i 15 not really thought about this problem until the NRC raised the 16 issue?
i 17 MR. NEWTON: To some extent, that is a true 18 statement. The ATWS rule was passed, and we all sighed a sigh 19 of relief and said we do not have to think about it again and 20 let's put our AMSAC in. That we have got sufficient margin in 21 the safety of the plant that the rule tried to address. From a 22 Westinghouse standpoint, we felt that our plants were safe and ,;
23 that we did not really have to think about it again.
24 And I think that I will show you in this background 25 information that we have had to struggle with this and look at
(
Heritage Reporting Corporation (202) 628-4888
137 1 what the NRC really was asking when they sent that letter to 2 us. We think that we have evolved to the point where what we 3 are going to describe to you addresses some of the concerns 4 that I have heard you asking today about ATWS and design 5 plants, and where we are going. So we met with the NRC staff a
6 in October.
7 MR. KERR: You said, I thought, that you had 8 concluded that from a Westinghouse viewpoint that this was not 9 a problem?
10 MR. NEWTON: That is correct.
11 MR. KERR: But you must also have your own viewpoint 12 which does not altogether depend on Westinghouse's.
13 MR. NEWTON: From a utility's standpoint?
I, 14 MR. KERR: Yes.
15 MR. NEWTON: That is correct.
16 MR. KERR: So from a utility's standpoint, are you 17 going to tell us eventually what the utility's view on ATWS is 18 as well as what Westinghouse's is?
19 MR. NEWTON: I can do that from two perspectives.
20 One is from myself as an operator of a utility. And two, a 21 little bit from the owners Group based upon the program that 22 was recently passed.
23 MR. KERR: I will wait.
24 MR. NEWTON: Okay. So in October, we met with the 25 NRC staff to discuss the letter that they had sent out. We
(-
Heritage Reporting Corporation (202) 628-4888
l l
1 138 l I still had not really assessed what the meaning of that was.
2 And our original intent was really to identify that from a 3 safety standpoint and from a risk standpoint, that what was 4 happening at the Westinghouse plants were within the rule and 5 that they should not be bothering us anymore, and that we 6 should just install AMSAC and go on our way.
7 We proposed to send them qualitative information to 8 back that up. They were not fully satisfied with our response 9 at that meeting, and asked the utility for more information, 10 because they were going in for a license amendment.
11 At the same time, we reviewed it internally, and 12 reassessed in December exactly what the NRC's real concerns 13 were and what concerns we might have with respect to trends in f
14 the operation and design of Westinghouse plants anyway, and 15 thought about exactly what is the basis of what we should be 16 looking at.
17 We have a rule there which we felt was appropriate.
18 It tried to address the risks due to ATWS. We felt that when 19 the rule was passed that we did not have to think about it 20 again. But the NRC was raising those questions. We said,
~
21 well, we as utilities, we have to be able to administer that ,
22 rule and to demonstrate or to prove to ourselves that we do 23 meet the intent of that rule.
24 So from a utility's standpoint, I started asking 25 questions saying how can we do this, how can we when we make an Heritage Reporting Corporation (202) 628-4888 i
l l
139 l 1 evaluation with respect to a 5059 and a core design, and what 2 have you, have we been ignoring ATWS. And when I asked j 3 Westinghouse, they said no, that they do look at ATWS in their 4 safety evaluations. And then I started asking, well, how do you 5 do that, and what should we be doing as utilities.
6 So we really reassessed and determined that we should 7 be looking at how we administer the ATWS rule for where we are ;
8 now and out into the future. So that we make sure that we do 9 not wander off the design of our plants, and do a number of 10 things that add risk in the ATWS area that we had not thought 11 about. And that things change gradually over that process, and 12 that we should develop the means to look at that closely.
13 We, meaning the Westinghouse Owners Group, asked 14 Westinghouse to put together a program that would develop a 15 framework that utilities could do and could implement into the 16 future. And we will be talking about that in the-second or 17 third bullet.
18 The program was proposed in January. The analysis j l
19 subcommittee looked at this program, and they endorsed it. In l l
20 early February, it was taken to the full Owners Group meeting, j 21 and the program was approved by the full membership to go l 22 forward. And the details of this program will be described 23 later after you talk about the field management aspects, and 24 then get into what the rule says and what we should be looking 25 at.
l Heritage Reporting Corporation (202) 628-4888
V l 140 1 Last week c n February lith, we met with the NRC and 2 described what was basically the same thing that we will be l 3 talking to you about. And I think that we caught them a little '
4 bit by surprise. Because not only did we feel that we were 5 addressing the immediate question as to where the plants were, 6 but we also tried to address where Westinghouse plants are 7 going in the future and that we make sure that we have a tool 8 available to utilities and to Westinghouse to evaluate that.
9 So that is the background of where we are today.
10 So to summarize some of the objectives. We will be 11 talking about the current field management activities, let's 12 say, within the Westinghouse Utilities. And then the 13 relationship of these management activities to the ATWS rule 14 and the basis of the ATWS rule.
15 And we also want to present to you the overview of 16 the Owners Group and the Westinghouse ATWS rule administration 17 program. Because Westinghouse has been very much involved in 18 ATWS over the last fifteen years, we got them to share with the 19 Owners Group program the cost of this on a fifty-fifty basis 20 versus a slightly different sharing on that. Because we can 21 both benefit significantly from this program.
22 The conclusions we will repeat at the end of the 23 meeting, but just to set the stage a little bit, is that on the 24 average that the Westinghouse plants have not seen a 25 significant or that there has been an insignificant increase in t
Heritage Reporting Corporation (202) 628-4888 l
l J
1 l
141 1 moderator temperature coefficient, and we will show you some 2 information on that.
3 That the current field management continues to be.
4 consistent with the conclusions of the ATWS rule. We will want 5 to point that out. But more importantly, I think that we will 6 want to identify and describe the WOG Westinghouse program that 7 provides for a systematic administration of the rule and the 8 basis for that, and also how that fits into what will be coming 9 on the severe accident policy, and that it is also incompatible 10 with that. And then address the program that we will be 11 putting together to administratively address the basis of the 12 ATWS rule into the future.
13 So that is basically the key items that we will want 14 to give you material on, so that we can demonstrate those 15 conclusions to you.
16 MR. KERR: Are you going to tell us what is meant by
. 17 the average insignificant increase in MTC?
18 MR. NEWTON: Yes, we will. That completes my 19 introduction. If questions come up with respect to what the 20 utilities feel about things, I will answer from the floor.
21 Vinnie Esposito will be covering the field management aspects 22 of where we are today with respect to the ATWS rule.
23 MR. ESPOSITO: As Roger stated, I am going to discuss 24 the fuel management aspects as they relate to MTC. To put that i 25 in some perspective, I have prepared a slide to talk about the
(-
Heritage Reporting Corporation (202) 628-4888 l
l 142 ;
1 core design process a bit, just so that we can put this all in i
2 perspective as we see it.
3 The major reason for this part of the discussion is 4 to make the point that there are many parameters that are 5 considered to address the safety, the operations, and the 6 economics of a fuel design. And that the parameters are 7- interrelated and that it is difficult to view each one in 8 isolation without looking at the entire set. So I felt that it 9 was hopefully useful to go through the process that we do in 10 doing a core design. And I will be brief. It will not be very 11 long.
12 I will start with the design requirements or the 13 inputs that are used. The key input which is supplied to us by 14 the utilities is the cycle energy output. If you wish, the 15 number of effective full power days of operation. And that 16 determines, as you can appreciate, many other cycle 17 characteristics like cycle burnup and things of that type given 18 the power of the reactor.
19 The other piece of input which is important is, of 20 course, the technical specifications and the safety limits. Or 21 another way of stating that is what requirements must the core 22 design meet. And we are talking about FQ, F delta H, insertion 23 limits, all of the FSAR safety bases. And they are obviously 24 part of that overall design.
25 And the last one is the fuel management
(
Heritage Reporting Corporation (202) 628-4888
143 1 considerations, what kind of discharge burnups are we talking 2 about, what kind of burnable poison types are we talking about, 3 a number of fuel management considerations that in many cases 4 are also contractual considerations that are folded into the 5 design requirements. So in a very global sense, those are the 6 three basic inputs that design the core configuration that we 7 are going to look at.
8 The output of that process, of the design process, is 9 the definition of the initial enrichment, the number of feed I
10 fuel assemblies, the specification of the burnable absorbers.
11 Those are basic outputs that come out of that process given the 12 input that we have.
13 The next step in the process after the final loading 14 patterns have been determined, and we have information from the 15 previous cycle of operation, is the safety evaluation. Here we ,
16 go through a detailed safety assessment using the reload safety 17 analysis procedure that we have in place. The results of that 18 process is the reload safety evaluation report.
19 The next piece of that to finish what I would call l 20 the output of the design is the operational data that we 21 provide to our utilities. These are startup physics constants, 22 core performance characteristics, and things that we have 23 predicted and supply that to the utility. So that is the 24 framework of the design process that we have used.
25 Given that process in place, and we have used that Heritage Reporting Corporation (202) 628-4888
144 4
1 for very many reload designs, and I am sure that well over a 2 hundred reload designs have gone through using that process, 3 given that process in looking at what we have done in the last 4 five years or so, the core designs have evolved over that time 5 with the experiences that we have gained towards more efficient 6 fuel assembly and fuel management designs which we believe has 7 provided significant benefits to our' utilities.
8 And I have listed some of them here. For example, 9 longer operating cycles. And we have already had quite a 10 discussion by a few people already. Approximately 60 percent 11 of the reloads that Westinghouse has performed are presently on 12 eighteen month cycles. So this is one of the optimization or 13 fuel management things that we have done.
14 In the area of optimizing fuel cycle performance, we 15 have introduced new fuel concepts, and have introduced new 16 burnable absorber types, so that we could continue to minimize 17 the parasitic losses, neutron losses. Yes.
18 MR. LIE: Sixty percent reloads implies that 19 approximately two-thirds of the fuel elements would be 20 discharged at each reload?
21 MR. ESPOSITO: No. It is about 40 percent.
22 MR. LEE: It would be 40 percent.
23 MR. ESPOSITO: It is closed to that. )
24 MR. LEE: And the 40 percent figure is a little bit 25 of an increase from previous shorter cycles?
(
s Heritage Reporting Corporation (202) 628-4888
l l
l 145 1 MR. ESPOSITO: In some cases, that is true. Remember 2 that we have a lot of different plants, and it is very had 3 sometimes to answer the question very specifically, because 4 'there are variabilities between them.
5 We had introduced a number of years ago like others 6 low leakage loading patterns. These have been used to some 7 extent to help the vessel fluent situation by removing the 8 placement of fresh fuel assemblies on the outside putting more 9 burned fuel assemblies on the outside to reduce the flux that 10 the vessel sees. And that is a kind of fuel management that 11 has to be incorporated into designs when it is used.
12 We have been increasing the discharge burnup again 13 for reasons of fuel utilization. Also obviously, to decrease 14 the number of fuel assemblies that are put in the spent fuel 15 pit or in fuel storage.
16 MR. WARD: With that second one, you seem to be 17 indicating that there is some neutron economy reasons for the !
18 low leakage loading patterns.
19 Would you have gone to this loading pattern even 20 without the PTS issue?
21 MR. ESPOSITO: Yes, we would have.
22 MR. WARD: There were some incentives there?
23 MR. ESPOSITO: Yes. In rough numbers, if you go from 24 what you call a standard loading pattern, leakage loading 25 pattern, to these low leakage loading patterns, it is worth
(
Heritage Reporting Corporation (202) 628-4888
146 1 about .1 in enrichment. So there is a benefit there in terms 2 of moving in that direction. So I am sure that we would have 3 done that.
4 The last one that I have here is the use of an 5 introduction of positive moderator temperature coefficients at 6 low powers. This is not at 100 percent power, but at low 7 powers. And this has been implemented for some time, and 8 permits some operating flexibility. It minimizes the number of 9 burnable poisons, and also has indeed economic benefits because 10 you do minimize the number of burnable poisons.
- 11 So the bottom line of this slide and the previous one 12 is that regarding fuel management that we have introduced many 13 changes over the years in fuel management. The impact on MTCs 14 specifically, some of them are negative, and some of them are 15 positive. And when we look at designs, we have to look at 16 those in an aggregate. And some of the numbers that I am going 17 to show you in a moment on the next slides look at the results 18 of these kinds of things and how they have evolved over the 19 years.
20 I think that we have had actually quite a bit of 21 discussion about this slide. Let me go through some of the 22 nomenclature first. We have plotted here temperature .
23 coefficient. And there are two types that are on here from our 24 nomenclature. One is the isothermal temperature coefficient.
25 This is at the beginning of life measured at hot zero power.
l Heritage Reporting Corporation l (202) 628-4888 1 i
147 1 So this comes from a measurement.
2 The other curve that we have here which I will talk 3 about in some detail is the MTC curve. This is again at the 4 beginning of life, but hot full power with all of the rods out 5 and equilibrium zero. Those are the conditions under which the 6 information'is being provided here. This is from predictions.
7 Now let me go back and look at each one of these.
8 MR. WARD: Which one did you point to?
9 MR. ESPOSITO: This one is predictions for MTC. This 10 is at hot full power.
11 MR. WARD: Okay.
12 MR. ESPOSITO: The two curves here are showing you 13 ITCs, isothermal temperature coefficients that are measured and 14 the isothermal temperature coefficients that we have predicted.
15 So some of the questions that we heard a little bit earlier 16 about the goodness of the calculations, if you wish, are shown
. 17 here. And in general, the agreements are not only good, but 18 they have gotten better.
19 This gives us some level of confidence that the ,
20 methods that we are using are reasonably good. And those same )
21 methods are applied here to do these calculations as they are 22 there.
. s 23 MR. WARD: I take it that those methods though are 24 normalized through observations or experimental points earlier, i 25 is that right?
(.
Heritage Reporting Corporation !
(202) 628-4888
l 148 1 1 MR. ESPOSITO: Those measurements have incorporated 2 knowledge over the years-based upon measurements that were 3 taken, yes. So that they have helped us to improve our 4 analytical capability. Yes.
l l 5 MR. LEE: Also I did understand your~ statement that 6 you would like to think that your prediction capability has l l !
7 become better, because you still have to plot uncertainty bars 8 even for isothermal temperature coefficient measurement.
9 MR. ESPOSITO: Absolutely. Yes, you do.
10 MR. LEE: And that is still I guess a point of l l
l 11 concern to me.
l l
12 MR. ESPOSITO: Well, we looked at some of the scatter l l I
! 13 from here to here, and that has come down. It has literally l l l l l 14 decreased.
l 15 Do you have a number, Dr. Casadei?
16 MR. CASADEI The scatter deviation for measurement l i
! 17 is predicted for the entire data base that vinnie referred to l I
l 18 as .8 pcm per degree. And this is going down in the 1986 and i l
19 1987 time frame. This represents more than a hundred reload 20 cycles in terms of measurement.
21 MR. ESPOSITO: Just so that we get it all in the l
i l 22 right terms. This is in terms of pcm per degree, the .8. Or 23 if we want, that is .08 times 10 to the minus 5 in the i
24 delta k/k per degree F. Just so that we are on the same units, l 25 okay. So this has all been plotted in pcm per degree F.
l
[
Heritage Reporting Corporation (202) 628-4888 l
149 1 MR. LEE: Also this is somehow an average of the 2 aggregate of Westinghouse plants that you are plotting?
3 MR. ESPOSITO: That is correct.
4 MR. LEE: So if you look at individual p.' ants, you 5 can have a different trend?
6 MR. ESPOSITO: Well, what we did was look at the 7 standard deviation, so that we can look at the whole 8 population. And that is the number that Dr. Casadei gave you, 9 .8 standard deviation value. You can have individual 10 variations plant to plant, that is correct.
11 MR. LEE: But also in terms of MTC trend and ITC 12 trend for individual plants, maybe you have seen some other 13 kind of trend.
14 MR. ESPOSITO: I think that some of the other slides 15 that we are going to present shortly will show you some of 16 that, again on an aggregate basie. We have not looked at each 17 separate individual point by point on a point by point 18 specific.
19 The second curve here is the MTC at hot full power.
20 Again from 1983 to 1987, and it is the average for all of the 21 Westinghouse plants that we performed the reload design for.
22 We looked at that in a little more detail in terms of the 23 different plants. Yes, sir.
24 MR. KERR: That latter curve, you said, is a 25 prediction.
(
Heritage Reporting Corporation l
(202) 628-4888
150 1 MR. ESPOSITO: Yes, sir.
l 2 MR. KERR Yes, okay. l 3 MR. ESPOSITO: A prediction; yes, sir.
4 In general, not in general, but I think across the l 5 board, the Westinghouse plants do not measure or try to measure ,
6 the MTC at hot full power for the reasons that Dr. Lee provided 7 before quite well in terms of what they say, the meaningfulness 8 of the measurement.
9 This is again MTC predictions as a function of cycle 10 start-up for two, three, and four loop plants. So we have 11 taken and gone one degree further in separating the average !
12 data that you saw before. The average data can be plotted on !
13 there. l l
14 And as one would expect, the four loop itself is very
- 15 similar to that, because that is the majority of the plants 16 that we have. The three loops which are shown here are 17 reasonably close to the average. And the two loops that you !
18 can see here have shown an increase in MTC with time. !
t 19 And there are some reasons for that, as one could i 20 expect. One is the issue of the low leakage loading patterns 21 that have been used for those types of plants in addressing 22 vessel fluents in particular. Old plants, pretty small, highly ,
23 coupled. Also the introduction of different fuel types which ,
i 24 have higher HTU ratios. And finally, the higher burnups that 25 plants have been going to.
Heritage Reporting Corporation (202) 628-4888 )
I
~1 1
151 !
1 The bottom line is that we have started very, very 2 low here in time. And in effect,:the two loop plants have come j upfto-bas 1cally the average of all'of the other plants over
~
3 4 time. And these are predictions.
5 The~last slide that I have is to show you'some of the .
6 scatter, the distribution of MTCs an a function of time. And 7 here we picked three times, 1979, the year 1984, and 1987, just ;
8 picked three years. In 1979, we lumped all of the months y 9 together, the start-up dates. And this just represents the 10 cycles that started up at the different months, just so we 11 could spread out the data. That is all we did it for.
12 And what we see here is from 1979 through 1984 and I 13 1987 that the range does not change very much. There is a lot i 14 of scatter, a great deal of scatter in the data and in the f
15 results. But the overali rate really has not changed 16 significantly. And that is giving us a little, bit more detail ;
17 in looking at the information that we have available.
18 The next part of the agenda, if there are no l i
4 19 questions, is going to be given by Jim Little, manager of 20 licensing. And he is going to discuss the other three points 4 -
21 of the agenda regarding the ATWS rule and the Westinghouse
. 22 Owners Group program.
23 (Continued on next page.)
24 25
. (.
Heritage Reporting Corporation (202) 628-4888 1
152 1 (Slides being shown.)
2 MR. LITTLE: Good afternoon. My name is Jim Little, 3 I'm the manager of licensing at Westinghouse Nuclear Safety 4 Department and I'm going to cover, very briefly I promise, the 5 basic elements of the rule and how we've looked at the ATWS 6 rule and how we've addressed both' today and how we are going to 7 address changes in the future. I have probably presented this 8 slide in these ACRS meetings a dozen times, but I'm just going 9 to use it one more time to make just a couple of points rather 10 than review the whole history.
11 Despite of what was said this morning by some other 12 organizations, we find the basis of the rule really to reduce 13 the risk imposed by ATWS transients, and to attempt to do that 14 a target was selected in the rule as 10 percent of the core 15 melt goal or probability. And you end up with a result of 1 16 times 10 of the minus 5 per reactor year for a core melt ,
I 17 frequency.
18 That was used in the rule primarily to select among a 19 number cf alternatives and choices that were being proposed by 20 three different organizations at that times the NRC staff, the 21 staff rule, the Henry version of the rule and the utility 22 petition group on ATWS.
23 What these options were evaluated with respect to 24 somewhat of a conservative approach on a consistent basis among 25 the reactor vendors with the probability of ATWS resulting in
(
Heritage Reporting Corporation (202) 628-4888
153 1 core melt, and evaluating that against the cost of an ATWS 2 event and the cost of these proposed hardware fixes.
3 You can see from the chart the base case was where, 4 for the BWR's on here, where these probabilities were 5 determined to be prior to the required changes in the ATWS 6 rule. And implementing AMSAC or diverse auxiliary feedwater 7 initiation in turbine trip signal function diversed from the 8 reactor protection system would achieve a reduction in risk 9 from 3 times 10 to the minus 5 to about 5 or 6. times 10 to the 10 minus 6 for the Westinghouse plants. And the impact was 11 assessed to be $2.8 million and it had a very favorable and 12 value impact ratio.
13 Some other options here at the time, this was after 14 the Salem event was --
15 MR. WARD: That's $2.8 million per --
16 MR. LITTLE: Per plant.
. 17 MR. WARD: -- per plant of some size, I guess, 18 typical plant.
19 MR. LITTLE: Yes.
20 Also at that time after the Salem events there was 21 some proposals to implement a diverse scram system instead of 22 AMSAC on Westinghouse plants, and that came about equivalent to ;
23 AMSAC in this analysis and in subsequent evaluations with the !
24 NRC staff and the Commission. That proposed rule was killed )
25 because this paper here SECY-83-293 didn't address all of the I
Heritage Reporting Corporation (202) 628-4888
154 1 issues associated with ATWS.
2 If we went further and had a diverse scram in 3 addition to AMSAC that would lower the risks of ATWS, but with
'4 a very marginal'value impact. So for Westinghouse plants the 5 final assessment was the installation of AhSAC.
6 For GE and B&W because of'the characteristics of 7 those plants, the designs of the cores, the features of those ,
8 plants, the installation of AMSAC alone could not yield an 9 ceceptable reduction and risk with respect to ATWS transients 10 and those potentials.
11 So, it was decided for those plants to install both 12 AMSAC and diverse scram. You can see the reductions in risk 13 down to 2 times 10 to the minus 5.
14 Now, what I would like to point out here is that, the ,
15 risk was a target, 1 times 10 to the minus 5 per reactor year.
16 In the case of Westinghouse this evaluation indicated we were 17 below that target. For CE and B&W indicated it was slightly 18 above. Yet again, it was a balanced look to look at what 19 hardware modifications should be done to evaluate the ATWS 20 rule.
.i 21 There are two points I think worth making here in 22 that, this is somewhat of a conservative approach because in ,
23 these analyses the NRC equated exceeding 3200 PSIA, the stress 24 level C limit with core melt and containment failure.
25 In the PWR unlike a BWR we don't have that situation, t'
t.
Heritage Reporting Corporation (202) 628-4888 ,
155 1 and the probability of the failure of the. containment in ATWS 2 events is much, much lower than one, it's on the order of 10 to 3 the minus 4 for conditional probability. So this was a fairly 4 conservative assessment.
5 MR. KERR: Excuse me, what is it, 10 to the minus 4?
6 MR. LITTLE: 10 to the minus 4 is the conditional 7 probability of containment failure given an ATWS transient 8 without mitigating effects; and that's documented in a number 9 of PRA studies.
10 But the point here, it's not so much 10 to the minus 11 3 or 10 to the minus 4, the fact in this analysis it was soon 12 to be 1.
13 This was, we think, an adequate approach to deal with 14 the issue of the need to protect the public, the need to 15 protect these investments. However, we don't feel that 16 necessarily this is where you stop with respect to ATWS.
17 MR. EBERSOLE: Before you throw that slide down.
18 MR. LITTLE: Yes.
19 MR. EBERSOLE: Just exactly what do you mean by 20 diverse? )
21 MR. LITTLE With respect to what, diverse scram, use i 22 a mechanism --
23 MR. EBERSOLE: The words you have up there? Do you 24 not mean diverse in aspect to trip signals but not to the trip 25 breaker functions?
( i l
Heritage Reporting Corporation i (202) 628-4888 !
l l
156 1 MR. LITTLE: We mean diverse from --
2 MR. EBERSOLE: You have diverse scram system added, 3 you know, items 1-B and 2.
4 MR. LITTLE: Yes.
5 MR. EBERSOLE: What does diverse mean?
6 MR. LITTLE: Diverse would mean it could share 7 sensing signals with respect to process parameters, but --
8 MR. EBERSOLE: But it all focuses down to two 9 breakers eventually, doesn't it?
10 MR. LITTLE: No, not with respect to a diverse scram 11 system, it would not.
12 MR. EBERSOLE: Oh, it doesn't. What are the other 13 means of tripping the --
14 MR. LITTLE: The other means would -- for example, 15 under diverse scram system might be tripping the fields on the 16 motor generator sets.
17 MR. EBERSOLE: That's what I wanted to get to.
18 MR. WARD: A couple questions. \
19 MR. LITTLE: Yes.
l 20 MR. WARD: The $1 million cost for item 2 from the 21 Westinghouse options; is that right? i 22 MR. LITTLE: No. If you looked at -- it's a basis of 23 comparison, given our brothers in Connecticut are pretty much 24 like us, the cost of diverse scram and AMSAC would be about 5.5 25 million.
And for Westinghouse plant in this regulatory
(
Heritage Reporting Corporation (202) 628-4888
157
( 1 analysis it would be 3.8 million.
2 MR. WARD: So one should read 3.8?
3 MR. LITTLE: It should read -- the total of those two 4 should be in the order of five and e. half million. We did 5 comment on this -- that specific point in a number of maetings 6 with the Commission on ATWS, but that number is not correct.
7 MR. WARD: When you said the conditional probability 8 failure of containment that some PRA's are shown as 10 to the 9 minus 4, given an ATWS.
10 MR. LITTLE: Yes.
11 MR. WARD: I guess that means -- a lot of that is 12 that there is a low conditional probability of core damage 13 given an ATWS; is that right?
14 MR. LITTLE: I think Mike Hitchler who is the manager 15 of our PRA group can answer that better than I can.
16 MR. HITCHLER: Yes. The basic point is that where 17 the number of conditional probability is somewhere between 10 18 to the minus 2 and 10 to the minus 4, and the definition of 19 that just means that we don't see a link between the phenomenon 20 that causes an ATWS, causes a core melt and the functioning of 21 the containment safeguard system.
22 The reason for that is two-fold. One is that we're 23 dealing with a different set of sensors. In other words, for 24 Westinghouse plants after an ATWS would startup containment 25 spray or fan coolers based on high containment pressure.
I Heritage Reporting Corporation (202) 628-4888
158 1 The other aspect is that the containments are large, 2 so therefore we don't need to have these functions operete 3- until well into-the ATWS event, meaning on the order of 30 -
4 minates to a couple of hours to actually protect the I 5 containers.
6 So therefore, we don't see any kind of a link between t 7 what caused the ATWS and what could also impair our containment 8 safeguards. }
9 MR. WARD: The core melts has kind of a link.
10 MR. HITCHLER: But from a standpoint of any kind of ,
11 phenomenon, the other aspect la there ways in which we would i 12 reach the containment to the ATWS. Would we go to such high !
13 pressures or factors that would cause breaching of the i
14 containment turbine, i i
15 In our ATWS events we're talking about --
l 16 Westinghouse PWR's we talking about relatively benign transient I i
17 turbine peak pressures. !
18 MR. LITTLE: I think the analogy would be there, if 19 you had an ATWS transient which gave you 10,000 PSI pressure _:
l 20 response you vould be concerned about the integrity of the ,
21 steam generator tubes; and that would be a containment bypass 22 and you could have a potential damage in your ECCS system and a ,
l 23 bypass for at least your containment. !
l 24 We're saying in this case, given the characteristic 25 of a Westinghouse plant in that assessment, we don't have those [
(~ !
l Heritage Reporting Corporation I (202) 628-4889 l l
1 I
l 4
l 1
1 159 1 kinds of responses. l l
2 MR. EBERSOLE: Just to get a feel for these options, 3 could you tell me what do you contemplate with the APWR in this 4 context?
5 MR. LITTLE: The APWR has the inherent features of l 6 AMSAC within it. And it is functionally separate from the 7 reactor protection function. So we design the features of 8 AMSAC within the APWR. . !
l 9 MR. EBERSOLE: Does it have diversity in your former i 10 identification of that?
11 MR. LTTTLE: It has diversity with respect to the :
12 initiation of AMSAC. We don't have a diverse scram system in 13 that plant design.
(.
14 MR. DAVIS: Isn't it true, Jim, though that the r
15 transient that initiates the ATWS could also have some effect i 16 on the reliability of containment safeguards, like, loss of
. 17 offsite power, for example. !
18 MR. LITTLE: Loss of offsite power could.
19 MR. L/.VIS: So there could be a link between the 20 probability of ATWS. )
l 21 MR. LITTLE: Yes. And our PRA should onow those l l
22 linkages. 1
. l 23 -
MR.~ DAVIS: And they do. i l
24 MR. LITTLE: Yes, they do. l 25 MR. DAVIS: And you don't get 10 to the minus 4.
i
(._
Heritage Reporting Corporation (202) 628-4888
160
(
l MR. LITTLE: Well, we don't get 1. It might be 10 2 to the minus 2, as Mike'says, but it's in that range.
3 MR. HITCHLER: It's the same for a non-ATWS core melt 4 indicated event, as for an ATWS.
5 MR. LITTLE: But again, the point here is the 6 intention of putting this slide up is not to argue hoir the ATWS 7 rule was so conservative, was really unnecessary. We do think 8 it was a necessary thing with respect to ATWS. It is a 9 contributed core melt and has shown to be.
10 This is one of the event trees in SECY-83-293 which 11 contains the basis of the ATWS rule. And what it -- it's a 12 fairly simplified event tree that we have here, tha" considers 13 the number of initiating transients per reactor year. The i.
14 reliability of the reactor trip function, both electrical as we 15 discussed in part of the presentations today, but also it 16 addresses the mechanical one.
17 And from NUREG 460, this number in WASH 1270, this 18 number was decided by a number of discussions over the years, 19 the reliability being 3 times 10 to the minus 5 per reactor 20 year, per failures per demand was this number, two-thirds of 21 which was associated with the electrical. This is really what 22 the ATWS rule deals with, is the electrical portion of this 23 risk which represents really two-thirds of the issue.
I 24 The next point in this node here is the condition we ;
I 25 call MTC over-pressure. And this wasn't really adequately )
1
(
Heritage Reporting Corporation (202) 628-4888
i 161 .
I 1 described in the basis, but the definition of this is, are 2 those set of conditions which exist, which would-give you a 3 condition which would exceed the stress level C limits of 3200 4 PSIA. And in this node 90 percent of the time it would not 5 exceed for these non-turbine trip transients for those 6 conditions and 10 percent of the time. For the CE and B&W t
7 plant these numbers wera 50 percent.
8 Then relying on auxiliary feedwater the reliability >
9 of that, you can see the number from the NUREG 0737 activities, 10 post-TMI is given there; the availability of high pressure 11 injection. And then the, what we call unacceptable 12 consequences probably, which really equates wich core melt and l 13 gives a direct release in rem because we take no credit for 14 containment mitigating functions in this NRC traatment.
15 I guess the points we like to make here is that, the 16 rule really dealt with the basic issues, that factors that 17 affect ATWS on an overall basis, and it had to do it in a i 18 simplified form so that the various vendors could be treated on 19 a consistent basis without complicating the rule and the l 20 treatment of these alternatives, these hardware alternatives in 21 an overly complex sort of way. So it was simplified to the 22 level that PWR's could be compared to PWR's.
23 We do feel that it dealt with the basic issues on an 24 overall basis, but it also provides the framework that one can 25 use in assessing MI as a licensee with the particular
(
Heritage Reporting Corporation (202) 628-4888 l
l I
162 1 characteristics of my plant, my core design. Am I consistent 2 with what level of risk wo sought to reduce in the ATWS rule.
3 These issues are important in themselves. We're not 4 just worried about some of these core melt nurbers, but we 5- really are from an operator standpoint concerned about the ,
6 reliability of our protection systems. Concerned about the 7 transient characteristics and response of our plant.
l 8 A lot of these features here have been the subjects l
I 9 of numbers of studies, not only by the NRC but thrs utilities 10 themselves, and particularly Westinghouse Owners Group.
11 For oxample, we did a program in 198T and 1983 on 12 rapid protection system reliability to address surveillance 13 frequencies rapid protection reliability. We didn't treat only 14 the analogy portion of that system, we address it all the way 15 through the reactor trip breakers. We aiso did numbers of 16 studies, post-Salem on the reliability of reactor trip 17 breakers.
18 We have certainly done a lot of work in the aux 19 feedwater reliability area. I'm rjoing to talk a little bit 20 about the ease conditions are, e.nd what -- if you develop this 21 in more detail, really what met of parameters shvald you map 22 out as an envelope that you'+:e going to have to manage the ,
23 configuration of your plant, within to be con =istent with the 24 objective we sought to achieve.
25 MR. EBERSOLE: Does that mean you reflect the current
(,
I Heritage Reporting Corporation j (202) 628-4888 1
L 1
163 1 concern about breaker trip' malfunctions that we have been 2 seeing so much of recently? l 3 MR. LITTLE: Yes. Yes.
4 MR. EBERSOLE: You mean there has been no change?
5 MR. LITTLE: I think -- let me first say that, with 6 respect to reactor trip breakers it's not -- we are not e 7 unconcerned with that. We are working very hard on that, in 8 particular the recent situation we just sent out a technical 9 bullet on the DS-416 breaker situation at the McGuire plant.
10 We think we have that under control right now. We've 11 got a number of actions and recommendations working very close 12 with the staff on that.
13 But what we're saying here is, in 1984 we all raised li our hands and said, ATWS was resolved. Really what was left 15 is, well, how are you going to manage this after the fact?
r 16 We're going to put these hardware changes in. Well, you can't 17 make an argument about the reliability of reactor protection j 18 system and ignore the facts that the data might change in the 3
19 future.
- 20 MR. KERR Well, you also can't make an argument 21 about the reliability of the reactor protection system without 22 saying something about the reliability of the scram breaker.
23 MR. LITTLE: Absolutely.
24 MR. KERR And be concerned about it is desirable, 25 but it doesn't change -- it doesn't have a lot of influence on )
( j Heritage Reporting Corporation ,
(202) 628-4888 l l
l l
l
164 1 the reliability.
2 MR. LITTLE: Well, I think it would. We're not only 3 concerned with respect to --
4 MR. KERR It would if the concern leads to 5 improvement. But I think Mr. Ebersole was asking whether the '
6 recent data had led you think that perhaps those numbers are 7 too optimistic; and I didn't hear an answer to that.
8 MR. LITTLE: The answer to that is, no, it would not 9 lead us to that.
10 MR. EBERSOLE: I have always been bothered by the 11 advertised diversity in the scram system at large, but the fact f
12 that it focused down to two breakers.
13 MR. LITTLE: Yes. I think Mike has got an answer to ;
14 that. ;
15 MR. HITCHLER: In '83 when we did the details of the 16 reliability assessment of the RPS, we came up with an i
17 unavailability for the reactor trip function, between 2 times 18 10 to the minus 5 and 6 times 10 to the minus 5 for the correct ,
t 19 system. Of that number virtually 90 to 94 percent of the 20 unavailability came down to common mode failure of the 21 breakers.
22 Where that appears -- and this isn't part of the ,
23 Westinghouse numbers, these are the NRC numbers that were used i 24 in part of the SECY paper. But where that occurs is that, in !
25 terms of when you have an integrated model is, trying to take c 1 Heritage Roporting Corporation 1 (202) 628-4888 I'
165 1 credit for manual scram in the reactor is not going to be a 2 very reliable function because the manual scram -- l 3 MR. EBERSOLE: You can make manual scram in the
~'
4 context of the operator killing the power supply to --
5 MR. .HITCHLER: Longer term. If you have the' power 6 supplies from the control room, yes, you can do it. Most 7 plants you can't, it's more like a 10 minute action. !
8 But what you can do in models -- and Tim will talk 9 about that later is, the Westinghouse procedures have put in 10 tri-manual scram, but recognize it may not work. If that 11 doesn't work, drive the rods in manually, drawing that maximum i
12 will give you a significant benefit here; and the procedures 13 are geared to recognize failure modes that can exist.
14 MR. EBERSOLE: I understand. You're telling me
- t 15 you've got time. r 16 MR. LITTLE
- Yes, i 17 Now, this is just an outline of the points I just 18 went through on this event tree. And again, to repeat myself, j 19 a simplified approach with respect to the elements of that 20 tree, but it really didn't address moderate temperature 21 coefficients per se. It mixed them together and assessed a
! 22 certain time where unfavorable conditions would exist that 23 might give you a result above the 3200 PSI limit. And the i 24 number was put without AMSAC, 3 times 10 to the minus 10 per 25 reactor year with AMSAC; 5 times 10 to the minus 6 per reactor
(
, Heritage Reporting Corporation (202) 628-4888
166 1 year.
2 So we not only, I think, met with respect to this 3 assessment the impact of the AMSAC modification; we met the 4 goal, but we've always been in the past been able to 5 demonstrate very favorably analytical results with respect of ,
6 ATWS transients outside of PRA assessments alone.
7 The treatment of those coefficients as I have 8 mentioned label them MTC over pressure. Again, there were two 9 time frames associated with turbine trip events and non-turbine 10 trip events, and if you read SECY-83-293 the numbers are 11 somewhat arbitrary, but within the ball park of what one -- one 12 might assume; some of them are differences to different steam 13 bypass systems designs, but it's difficult to understand within >
14 the particular numbers there are. But the bottom line, it 15 acesn't really make a large difference. ;
16 I guess the point we would like to say here, one 17 thing that the tr'ee does do is, what we think overemphasize the 18 contribution of MTC as the source contributor to ATWS risk; we 19 don't think that to be the case. We think that an ATWS result 20 is a function of the configuration of the plant including the 21 moderator temperature coefficient, but many things --
22 MR. KERR Do you think it contributes anything to , ,
23 ATWS risk? l 24 MR. LITTLE: Yes.
25 MR. KERR: Do you think it contributes significantly
(
Heritage Reporting Corporation (202) 628-4888 ,
I 167 1 to ATWS risk?
2 MR. LITTLE: Less significantly than the rules said.
3 There are other things which the rule event tree did not point 4 out which are contributors to risk.
5 What we are saying here is that --
6 MR. KERR I think you're answering a question that I l 7 didn't ask, and that's perfectly okay. But what I was asking, !
8 if you thought it was a significant contributor, and by 9 significant I mean that one should be concerned about it?
10 MR. LITTLE: Yes. What we're saying in this !
11 evaluation is that, this one factor dominates-other factors !
12 that are also important, not only from a risk standpoint but 13 from a safety standpoint as well.
14 MR. KERR: We'll stipulate than that's i 15 representative, but neither we nor you accept that as 16 completely definitive.
17 MR. LITTLE: Correct.
18 MR. EBERSOLE: What bothered me was when Salem 19 occurred there had been a lot of chit-chat for many years that 20 it might and it did; and you uncovered the margin of force 4
21 problem. And you went about the process of patching up'the
. 22 breakers.
23 MR. LITTLE: Well, I guess to answer --
24 MR. EBERSOLE: It looked to me because of its 4
25 localized viewpoint on the fix, you know, the patchitis 1
('
Heritage Reporting Corporation (202) 628-4888
i 168 1 approach which is popular within NRC anyway, and I guess l
2 equally popular with Westinghouse.
3 MR. LITTLE: Well, I don't know if it's --
4 MR. EBERSOLE: Anyway, you know, it was just the 5 localized fixed than a diverse spreading fix to de something 6 about it.
7 MR. LITTLE: Well, I think the answer to that, number 8 one, the margin of force consideration with respect to reactor 9 trip breaker is more in issue with the GE design. The issur 10 with respect to the Westinghouse design was lack of proper 11 maintenance and testing, which were the dominates --
12 MR. EBERSOLE: Well, those are administrative 13 procedures that are subject to all sorts of populations, 14 they're common modes.
15 MR. LITTLE: Yes, I understand. But I won't speak 16 for the NRC on patchworkness. Now that you have fed me the
. 17 punch line for this next slide.
18 (Laughter) 19 MR. LITTLE: I guess the point here with respect to 20 ATWS rule is that it stopped a little short in that it proposed l
21 a hardware fix and said, hey, the party is over. And I think l
22 after the dust settled after about 18 years of this the ,
23 questions were there, well, what are Ne going to do about the 24 future? And I think Commissioner Asselstein asked the right 25 question. If you read his comments in the rule, what are you l
( i 1
Heritage Reporting Corporation (202) 628-4888 I
l
. = . .-
169 1 going to do for future reactors?- Interpret that a little 2 better at saying, hey, the arguments you made here in ATWS: rule 3 going to change.
- 4 And Roger Newton talked about, well, we-all felt 5 confident. But everybody did, not just the licensees, the 6 Commission, the staff, everybody said, finally, I can't 7 believe, we've had our last meeting on ATWS, and that certainly 8 hasn't been the case.
9 MR. EBERSOLE: Well, there's always a reluctance to 10 change because it cast a disparage on what you have already 11 done.
12 MR. LITTLE: True.
13 MR. EBERSOLE: Of course, GE is a good example of 14 that.
15 MR. LITTLE: I won't comment on that. But the point 16 here is that, the bottom line basically ended up in a hardware 17 fix and the regulatory attention focused on what does'AMSAC 18 look like? What does -- what are the requirements for-diverse l 19 scram? How many signals? Let's get the SCR's out. What is l
1 20 the schedule for implementation. And the other aspects of ATWS i 21 were kind of left there. And that's really what we are 22 attempting to do in our program, to make sure that we don't 23 back into a problem. .
24 I think it wasn't -- you asked Roger Newton, did you 25 really back into this problem and not realize it? It's not a
( l l
Heritage Reporting Corporation )
(202) 628-4888 l l l
l
n 170
(
1 problem yet. I think if we back into the realization, 2 something ought to be put in the place. And you will see as we ,
3 discuss that, and we will discuss that. ,
4 We don't have a problem right now. The present fuel 5 management is enveloped by the ATWS rule basis. We're well 6 within the arguments we made then. The rule does provide a 7 basis, a foundation on which to establish a program to evaluate 8 future changes. We don't want to have a one dimensional 9 program that says, well, moderate temperature coefficient is '
10 it. We've made some other changes which may invalidate those 11 arguments that we made with respect to ATWS rulemaking, and the -
12 benefits we sought to achieve. And we do think a process like !
13 this is worthwhile.
I 14 What our objectivos in this program are first of 15 all, to better document and develop the application of the rule 16 on a bases to the Westinghouse PWR features. Better than it is 17 done in SECY-83-293 paper in more detail. To say, these are -
18 those considerations, those sets of parameters which affect 19 ATWS transients, to develop a process of procedure for 20 licensees to use in evaluating changes under the guidelines of 21 10 CFR 5059 in answering those specific questions. Here lies 22 in this mechanism, probably consequence scenario.
23 Everybody still always has to comply with 10 CFR 24 5062. You don't necessarily comply with the hardware 25 implementations alone, you have to comply with what benefit you
(..
Heritage Reporting Corporation !
(202) 628-4888 ;
171 1 sought to achieve.
2 With respect to the approach we want to develop a 3 more detailed risk model. What we mean by risk, it's not just 4 a problemistic model, but a risk model that incorporates the 5 deterministic analyses, the success criteria that developed 6 from 15 years of ATWS analyses that says, what are these 7 sensitivities. Develop those guidelines, those steps for-8 utilities to use in evaluating these changes.
9 The schedule we're on for this program, we have 10- initialized it in February, we expect to have the program 11 completed in May and we're going to share that information with 12 the NRC right after the end of that program in May.
13 MR. KERR: You're talking about a program -- I'm 14 getting lost here. What is the program that is being 15 initialized in February of 19887 16 MR. LITTLE: The program is a mechanism by which a t
17 utility can evaluate a change to the blank configuration and 18 determine if there is an impact with respect to the ATWS rule.
19 MR. KERR: So this is in the process of complying 20 with the ATWS rule --
21 MR. LITTLE: Yes.
22 MR. KERR: -- by Westinghouse Owners Group associated 23 plants?
24 MR. LITTLE: Yes.
25 MR. KERR And program completion means that by May
(
Heritage Reporting Corporation (202) 628-4888
i l
i 172 1 you will be able to tell them what you think should be done?
2 MR. LITTLE: Yes.
3 MR. KERR And then at that point somebody 4 establishes a schedule for implementation.
5 MR. LITTLE: Well, at that point we intend to 6 implement that program.
7 MR. WARD: It's more of a kind of methodology or 8 administrative approach.
9 MR. LITTLE: Yes.
10 MR. WARD: And the NRC hasn't asked you to do this.
11 MR. LITTLE: No. No. We decided this was the best 12 way to avoid problems in the future. Not only to back into a ,
13 problem, also, we're down here and we have another meeting and 14 talk about reactivity coefficients and models and 15 uncertainties. We want to, number one, address the question 16 with respect to moderate coefficients that we've done, but also 17 put something in place that we never have the need for meetings 18 like this again. So this is part and parcel to that.
19 MR. EBERSOLE: Speaking of administrative changes, do ;
20 you know the value of splitting maintenance and testing so that 21 you don't touch all your resources at once, you know, like the 22 airplane out of --
23 MR. LITTLE: The Eastern Airlines flight to Nassau.
24 MR. EBERSOLE: Right. It's not smart to touch your 25 resources.
(
Heritage Reporting Corporation (202) 628-4888 l l
173 1 MR. LITTLE: Yes.
2 MR. EBERSOLE: Do you therefore spread your attention I l
3 to your breakers over some logical span of time?
4 MR. LITTLE: I think the elements of this are, number 5 one, to make sure other programs that have been done by 6 Westinghouse and the Owners Group and utilities and industry as 7 well, but let's not go and redo everything there. We want to 8 make sure we take maximum benefit from what exists and what's 9 ongoing. So we're not specifically going to address that 10 aspect, but l'i is an important one.
11 This is focusing primarily on documenting those sets 12 of conditions which will ensure as a set that you're in 13 compliance with the rule. ,
14 This approach is going to apply to all the <
15 Westinghouse PRW designs. We're not only going to develop a 16 more detailed risk model to determine impacts on the basis, but '
17 we're also going to make sure this is compatible with what 18 we're doing on severe accident policy. Because ATWS is one of 19 the contributors to core melt and risk. So this methodology 20 should be consistent with those methods being developed now, =
21 IPE, for example, on severe accident policy, so the utilities 22 could take the benefit of this study being done generic and not i i
23 have to repeat this work. And it will be consistent among the 24, licensees. l l 25 MR. WARD: Bless you.
- (
Heritage Reporting Corporation )
(202) 628-4888 1 1
l l
174 1 MR. EBERSOLE: What effects do you think successfully 2 mitigated -- what effect do you think a successfully mitigated 3 ATWS would do to your plant in a reputational context or 4 whatever, you know, like TMI-27 I'm talking about one that you 5 really mitigated, but it, you know, you had a neat challenge. ,
6 MR. LITTLE: You mean if we-were successful in 7 mitigating a consequence?
8 MR. EBERSOLE: It would say, it proved I was right 9 and it didn't hurt a sole. On the other hand you could take 10 the other view, the containment it survived and I wish they l 11 hadn't.
12 MR. LITTLE: I don't know what the question is.
13 MR. EBERSOLE: Well, I'm just saying, would you 14 consider that in the positive context that you would say, see, 15 it proved the reactor safe. l l
16 MR. LITTLE: Well, that question could be asked --
l . 17 you could ask that about TMI, was TMI a good thing; it taught t
l 18 us things.
19 MR. EBERSOLE: It's the same question. I'm thinking l 20 about the effects of TMI which hurt nobody except --
l 21 MR. LITTLE: Except GBU. If I were a utility I would 22 say, not in my plant. ,
23 MR. EBERSOLE: Not in your plant, right. Not in a 24 Westinghouse plant either.
25 MR. LITTLE: That's right. I would leave that to Heritage Reporting Corporation (202) 628-4888
?
175 -!
I someone else to do in a state far away from me. !
2 The approach again is to make it consistent with the f 3 severe accident policy statement. Make it consistent with the
'* 4 targets, the' goals that were in the SECY paper. fWe're not l 1
5 going to out and try and reargue that what's rule.
6 We're. going-to use this model to evaluate the effects [
7 of changes in the core and core melt frequency. The products i 8 of this program will be that model definition, risk model !
, 9 sensitivities, studies and the guidelines, so I can' understand ;
i
! 10 what the effects of these changes are and implement them in f
11 this model, i
)
12 And just give an example of, I showed you the four or -
t i 1 13 five things that were in the SECY paper, this is just a parcel !
t i 14 listing of the things that you put in a more developed risk 15 model. ;
16 The point is that you have to consider all these !
17 attributes and not just the good ones. You can make arguments l I
] 18 that the number of challenges it was reducing, and that offsets {
l
, 19 ATWS risk, but there are other things happening out there, like l
} ;
] 20 reactor trip breakers, for example, j
. i 21 You've got to look at the whole set, the good and the ;
i 22 bad together. I 4
23 This is a very good story with respect to initiating 24 frequencies of events. Okay. We've had studies on the [
l 25 availability of reactor trip, as I mentioned before. We've }
(. . l Heritage Reporting Corporation i (202) 628-4888 i f
i I
u . _ - - _ , _
176 1 done a lot of work in emergency response guidelines in 2 particular ATWS events. The availability of main feedwater 3 studies are being done now. The AMSAC design we have licensed 4 with the NRC. We did have a reliability goal for this system.
5 There was a failure modes effects analysis on this.
6 MR. EBERSOLE: Can you tell me why your design is 7 unique in aspect to tripping main feedwater than trying to 8 restart it? The other plants don't do that, they try to ramp 9 them down.
10 MR. LITTLE: That's right. It depends on the 11 characteristics of the plant. You can do rapid turbine run 12 backs on a super heat s*.eam generator. There are other things 13 you can't do with a super heat steam generator.
14 The problem we have --
15 MR. EBERSOLE: What do you think about CE and GE, 16 they all run by --
17 MR. LITTLE: Thinking about what?
18 MR. EBERSOLE: CE and GE all run --
19 MR. LITTLE: CE has a partial reactor scram. Okay.
20 And we find that's really nice, but if you look in reality, how .
21 many times that function is challenged and the money you have 22 spent for it, it's really not worth that feature.
, i 23 MR. EBERSOLE: I'm just looking, you know, try to 24 minimize the challenge frequency on aux feedwater. !
t 25 MR. LITTLE: I understand.
(
Heritage Reporting Corporation (202) 628-4888
l 1
1 177 J 1 On AMSAC we did do reliability studies with the top l 2 of the reports we submitted to the NRC, we did do failure motor j 3 analysis. We did have a risk goal.
4 Our logic on this system isn't a two-out of two.
5 It's a two out of three to energize and a three out of four to 6 energize depending on the logic design you pick. So we're 7 worried about spurious challenges to the system. But we're 8 also not going to make this thing prone to the failure of one 9 component not making function.
10 We made it testable with power without taking it in a 11 bypass, in a partial trip mode where it would light off if the 12 technician did the wrong thing.
13 MR. KERR: Do I conclude that you are not depending ,
14 on a single failure criteria, but you actually are doing a 15 reliability analysis in order to achieve appropriate 16 availability of AMSAC7 17 MR. LITTLE: We want -- we don't want the system .
18 itself to be prone to the failure of one component spuriously 19 initiating.
20 MR. KERR: But you're going beyond the single failure 21 criteria --
22 MR. LITTLE: Yes.
23 MR. KERR: -- and are specifying a reliability.
24 MR. LITTLE: Yes.
25 MR. KERR: Thank you. ,
\
Heritage Reporting Corporation i (202) 628-4888 )
J
178 1 MR. LITTLE: We also want to look at those other 2 functions. We talked about containment heat removal, are they 3 available. Long term shutdown. What about the availability of 4 ECCS, and the availability of the operator trip the MG sets in 5 this case. We're going to put all these things in context. .
6 Have we blocked PRV's on this plant? You know, in 7 those ATWS analyses we took credit for relief valvo capacities 8 on the pressurizers. Utilities block these valves.
9 MR. WARD: I thought that didn't amount to much, 10 though, as I recall.
11 MR. LITTLE: They amount to about 300 PSI. It's very 12 important for any PWR design to have relief capacity. Look, 13 for example, on the system 80 designs, why the pressure -- no 14 relief valves.
15 So the point here is that, having relief capacity 16 ATWS event is very, very important. Well, don't make the 17 argument in the top of report in 1974 and then block the valves 18 all the time. Well, if I block the values and I look at all 19 these other things, am I still exposed?
20 So what I'm saying is, look at the good, but also 21 look at the bad things you might be doing. And then consider 22 the reactor core, feedback characteristics as well. When you ,
23 do that, you start looking at event trees that are a little 24 more detailed.
25 And now you give the utility the tools to make the i
Heritage Reporting Corporation (202) 628-4888
179 1 decisions by saying, well, if I want to do this with respect to 2 the core and I want to do this with respect to the relief !
3 valves, what does this do? Where do the sensitivities go?
4 Does the dominance change? Are the things that were in order 5 of dominance switch, and those things that switch are very, 6 very sensitive, I want to pay a lot of attention to those.
7 We're going to set up the model consistent with the 8 reference plant, the bounding analysis that we did for 9 Westinghouse designs in the '70s, making sure we're still 10 consistent with those arguments and then defining those ,
11 sensitivities to those that address the other plants, the three '
12 loop plants, the two loop plants, the lower power level plants. !
t 13 So I want to make sure it's a conservative bounding i 4
14 case and give them the tools, the sensitivities to evaluate 15 their particular situation. j 16 MR. LEE: In the events tree it just showed, there is L 17 no SS high level graded in 40 percent decided for your 18 terminology, correct. I'm just curious, why the 40 percent is 19 chosen.
20 MR. LITTLE: Because we did sensitivity studies to 21 answer another question about part power ATWS, okay, to support ,
22 the AMSAC design. When should AMSAC be armed. So we did part 23 power ATWS analyses in the past, so we know that lower power 24 ATWS, the low 40 percent power not a concern. So we have a 25 power level condition that says, this is the AMSAC set point. l l
( l Heritage Reporting Corporation l (202) 620-4888 l
180 1 This considers the availability of AMSAC. And is AMSAC armed.
2 MR. LEE: Thank you.
3 MR. LITTLE: Here is one example of, I can look at 4 information that relates to other programs, we've had a large 5 programs for three years with Westinghouse owners on trip 6 reduction and assessment, and it has reduced significantly 7 incorporation with a lot of other programs on what the 8 utilities are doing with cells and INPO; a number of 9 challenges. And that's a good story. I want to be able to 10 factor that story in.
11 But if I'm a utility that has had a bad experience 12 with respect to these transients, I should use the numbers that 13 apply to me and not just plug in an entry avercge for that.
14 That is a good story. The trips have been reduced by 15 a factor of two.
16 So we have -- these are the plots. You can see the
. 17 Westinghouse line on here.
18 (Continued on next page.)
19 20
~
21 22 23 24 25 (1 !
Heritage Reporting Corporation (202) 628-4888 l l
t
181 1 MR. EBERSOLE: You have done that by looking at and 2 doing things to safety systems and trips, at large.
3 But you have not gone back, have you, to the, into 4 the balance of plan?
5 MR. LITTLE: We are now.
6 MR. EBERSOLE: Are you?
7 MR. LITTLE: Yes. We are looking at turbine trip l 8 initiators now.
l 9 MR. EBERSOLE: You are going to contemplate 10 coincidence in such things as heater level trips and all that?
I I
11 MR. LITTLE: We are looking at -- we started at this 12 and we looked at the biggest contributor which was steam 13 generator level, and we looked at that in 1985 and we focused I
14 on a number of things: procedures for controlling level. We had 15 seminars, we looked at the reactor protection system logic and 16 set points, we looked at the analysis in trying to develop 17 better set points, better instructions.
18 After that, we looked at safety injection initiators.
19 We looked at the order of their contribution.
I 20 MR. EBERSOLE: It is an insult to the system to find
~ ,
21 that the janitor hit his broom against the heater level switch !
22 and you went through all of this stuff, t
23 MR. LITTLE: That is correct.
24 We are looking at the single point failure analysis 25 in the BOP now.
(.
Heritage Reporting Corporation (202) 628-4888 l l
182 i'
1 MR. EBERSOLE: Okay.
2 MR. KERR If GE were plotting this set of curves, 3 would they look like that?
4 MR. LITTLE: Well, the numbers don't lie. You might 5 turn it this way, I don't know.
6 MR. KERR I was just wondering, there are a couple 7 of ways that people can interpret SCRAM, as I have discovered 8 in the past and I wonder if there are any different ways of 9 interpretation that could lead to a different set of curves.
10 MR. EBERSOLE: That approach does not yet reflect 11 going back into the balance of plant in the bunt structure, 12 does it?
13 MR. LITTLE: That is right. We insist that it is 14 going to get better.
15 MR. EBERSOLE: It is getting pretty good.
16 MR. LITTLE: Now, we have submitted a large amount of 17 topic reports on trip reduction to the NRC for their 18 information. You will find that these are very much consistent 19 with the AOD reports, too.
20 We have, every plant in the United States is linked 21 with electronic data system where all the trip information is 22 loaded and maintained in a large data base. So that we are ,
23 talking about 44 units, here, in the United States. This is a 24 large data base.
i 25 MR. EBERSOLE: What is the Japanese experience k l Heritage Reporting Corporation (202) 628-4888 i
183 1 compared to that?
2 MR. LITTLE: Well, it is probably below this line.
3 MR. HITCHLER: The Japanese are about 25 to one trip 4 per year.
5 MR. LITTLE: Yes, it turns out that the Japanese have 6 a totally different philosophy with respect to maintenance on 7 their plants, the way things are done. You have longer 8 outages.
9 MR. KERR Do you have any idea, qualitatively, what 10 fraction of the current trips are caused by errors made during 11 surveillance testing of their reactor protection system?
12 MR. LITTLE: I don't know.
13 MR. HITCHLER: It is about 26 trips per year, due to 14 that kind of testing.
15 MR. KERR: Thank you.
16 MR. LITTLE: We found before that the automatic 17 trips, two-thirds of them were steam generator level and that 18 was the first element of our program in 1985.
19 To conclude, really, now we have seen what the !
20 definition of average was in any significant ones. We are not 21 any where near violating the arguments we made with respect to :
22 ATWS.
23 The current fuel management continues to be 24 consistent with the conclusions of that rule, and we have ,
25 instituted a proactive program that systematically addresses in
( :
Heritage Reporting Corporation (202) 628-4888
l 184 l
I' 1 the future, that will also be consistent with a severe accident l 2 policy approach.
l 3 And we do think this is worthwhile. We want to do it )
! 4 ourselves. If we are responsible for the licensing and safety !
5 of these plants, we think that we have the wherewithal to 6 develop a program that will administrative 1y deal with this 7 issue very effectively and address all the factors which affect j 8 ATWS risk.
9 Are there any questions? ,
10 _MR. LIPINSKIt Back to your breakers, you made 11 reference to maintenance, but on July 2nd, there was an event 12 McGuire II, again with an after-trip breaker which is totally 13 unrelated to prior events. ;
14 MR. LITTLE: That's correct.
15 MR. LIPINSKI But in terms of diverse scram that is 16 not of interest to you at this point.
17 MR. LITTLE: No. i 18 MR. LIPINSKIt You're still going to rely on two 19 reactor trip breakers fixing whatever occurs to them which is 20 totally unrelated to any prior failures. i 21 MR. LITTLE: Yes, that's correct.
22 MR. LIPINSKI: That leaves to a conclusion that these ,
23 breakers are going to continue to be a source of problem, and 24 hopefully it will not occur simultaneously.
25 MR. LITTLE: That's correct. You wouldn't see j
- l L
Heritage Reporting Corporation !
(202) 628-4888
185 1 anything in that argument that would indicate, that was a 2 dominant contributor to common mode failure. But we do think 3 it's important that those breakers do function reliably and do 4 their function.
5 We have a lot of evidence that indicates, we know 6 what the problem is to be. We have a lot of steps underway to 7 address those problems, and we're going to address them one at 8 a time. There is nothing that would make me change my argument 9 with respect to ATWS risk.
10 MR. KERR: They have a backup, they know it's bad GE 11 breakers.
12 MR. LITTLE: Well, I don't think we would do that.
13 (Laughter) s 14 MR. EBERSOLE: It's fascinating how this argument --
15 MR. LITTLE: I think when we see diversity with 16 respect to unload failure and that's --
17 MR. LIPINSKI: This particular one simply indicates 18 that it's a patchwork, each time something happens it catches 19 your attention, you fix it, until the next thing happens.
20 MR. LITTLE: Well, I think it's consistent with 21 enything you discovery, you've got to make sure you address it, 22 you know that. I think if you had an unlimited budget you 23 could go out and design many, many things that would address 24 all future unknown problems.
25 The point is that, we did have redundancy in that
(
Heritage Reporting Corporation (202) 628-4888
186 1 plant. We do test those breakers. We can take credit for 2 . ..a t . Okay. We do inspect those breakers. Now we have 3 improved the. inspection teci.niques on those. We're going back 4 and looking at the manufacturing process, and isolated that one 5 particular model breaker. But it's not the only model breaker 6 in Westinghouse PWRs. So you have to learn from that 7 experience.
8 MR. EBERSOLE: I think that you and GE had some 9 ATWS's at Salem and Browns Ferry.
10 MR. LITTLE: I don't know if you can call them ATWS.
-11 MR. EBERSOLE: I mean, they were beginnings of that.
32 MR. LITTLE: Well, we had some reactor trip failures.
13 MR. EBERSOLE: I always look at the common dump 14 volume with GE and parallel with the two single breakers at 15 Westinghouse as being a localized fixed problem. Aitd we keep 16 trying to bench them both.
. 17 MR. LITTLE: Well, I don't think Salem was anywhere 36 nears Browns Ferry.
19 MR. EBERSOLE: Well, of course, it was -- well, the 20 handle came off, that made it nervous.
21 MR. LITTLE: It was designed to be removed.
22 MR. EBERSOLE: Under those circumstances. Anywcy, ,
23 it's an interesting parallel, the dump volume and the two 24 breakers.
25 MR. LITTLE: Nell, certainly you're not getting more C
Heritage Reporting Corporation (202) 628-4888
187 1 events like that. We're going to pay a lot of attention. We 2 have been very aggressive in pursuing VDS 416 problem and we're 3 going to continue to be aggressive in addressing that. Right 4
4 at this point in time I wouldn't say, I'm going to change my 5 mind with respect to ATWS risk.
6 MR. KERR- . Any questions?
7 (No response.)
8 MR. KERR: Earlier you said, I believe, that the 9 recent experience with trip breakers had not chariged your 10 feeling about the reactor protection system reliability, or 11 something to that effect.
12 MR. LITTLE: Yes.
13 MR. KERR: Now, it occurs to me that a number of data 14 points have been developed, and if I add those to a statement 15 which I heard from a very nice young engineer representing 16 Westinghouse a year or so ago in which he said that there had 17 never been a Westinghouse breaker failure to open when the 18 shunt trip is being used.
19 MR. LITTLE: That's correct.
20 MR. KERR: It would seem to me that one would take 21 the recent data and do a reanalysis to see if, by some use of 22 statistics, a prediction of reliability of that breaker might 23 be modified a bit. This has been done and there's no 24 modification.
25 MR. LITTLE: I guess the point is, when that young
(
Heritage Peporting Corporation (202) 628-4888
i 188 4
1 engineer told you that, that was a statement of fact and you 2 never heard predictions of what happens in the future. Never 3 said it wouldn't, okay. But the point then we were --
4 MR. KERR .But having had it happen, it seems to me l 5 some change in your reliability prediction might be in order.
6 MR. LITTLE: Some assessment of that should be 7 evaluated.
8 MR. KERR: I would think so.
9 MR. LITTLE: Yes, it should be.
10 MR. HITCHLER: One comment there is, we have done ;
11 that. The comment that was made was saying, the failure mode 12 that we're seeing seems to indicate that the breaker stayed 13 closed and therefore the use of a manual trip is not 14 necessarily going to give us the benefit, and we have to rely 15 more on the manual run assertion. So we are modifying the 16 model consistent with that.
17 The other aspect, as I said --
18 MR. KERRs I was referring specifically to your i
19 calculation of the failure probability of a trip breaker. I 20 don't know what it is, I have never seen it, I'm not even sure.
21 MR. HITCHLER: But the fundamental part of the model j 22 that would be impacted here would be 94 percent of the reactor l
. l 23 scram unavailability comes from common mode failures to l
24 breakers. You go back and assess whether the failure modes of l 25 the event identified recently indicate a common mode failure 1
( .
Heritage Reporting Corporation 1 (202) 628-4888 i
)
1 i
189 4
1 mechanism that's applicable to all of our plants.
2 MR. KERR I understand common mode failures. But .
3 since I don't know what they are, I also understood in failures 4 that are not common mode because you're likely to see more of 5 these and it seems to me to pay more attention to them or you 6 can perhaps de something about them.
7 The reason I keep on this is because of what appears 8 to me, there's a large allowance made for. reactor protection 9 system unavailability. I mean, it's considered to be very low.
10 Low enough so I don't have any idea how to demonstrate that one 11 actually is achieving thia. Maybe Westinghouse can demonstrate 12 it to its satisfaction; maybe utilities can. I would find 13 myself completely unable to demonstrate that number. And 14 therefore it seems to me one looks very carefully at anything 15 that can be done to add some understanding or additional good )
16 feeling or whatever one needs when one does not have l
17 experinnental verification.
18 Now, I must say that I like the fact that if I 19 understand the situation Westinghouse design la to say, it's 20 the pressure buildup given ATWS is less. I wish that other 21 people who are building reactors could somehow have had the, 22 whatever it is, the lid to that. But I think it's a good 23 thing. They are out of the compensating features. So maybe if 24 I were Westinghouse I would worry a little bit less about the 25 trip breaker reliability if I had the relieving capacity. But i
Heritage Reporting Corporation (202) 628-4888
1
'l l
i 190 1 I hope, as you said and I agree with you, one neet.9 to look at ,
2 tnis whole panoply of things of tripping; and it is encouraging i l
3 that you're doing this. You really are, aren't you. l 4 MR. LITTLE: Yes. i 5 MR. KERR: You just didn't make the slide.
6 (Laughter) 7 MR. LITTLE: I guess the point is that, we don't want 8 to give you the impression that with things like the DS 416 !
I 9 breaker that we're complaining about, oh, well, it was a well 10 break and we'll just reweld that and everything will be all 11 right. Every time we have something we do a full reassessment 12 of what that means with respect to the arguments.
13 You heard Roger Newton talk at the introduction is 14 that, when we have been designing cores we haven't put our 15 hands over our eyes with respect to ATWS in the past. We've 16 kept an eye on it and we know where we are. We have always 17 known where we were with respect to ATWS and the way the 18 coefficients were treated in the rules. So we're not going to i 19 be bt .d about the same things about reactor trip breakers per 20 se as well.
21 A lot of those programs that came out of Salem were 22 directly related also to ATSW. As a matter of fact, when we 23 were looking at the ATWS . rule, one of the programs was, we 24 studied feedwater transients in Westinghouse PWRs over a five 25 year period. And that program, that report that we developed I. l Heritage Reporting Corporation l (202) 628-4888 1
191 i
1 was the genesis of our. trip reduction program. We said, hey, 2 there are some nice fixes to these problems. We've had the 3 opportunity to study them. We have a lot of information now.
4 And we have been working on that program for three years and 5 seen a dramatic improvement. So it's not --
6 MR. KERR: To me this is also, perhaps, an 7 illustration of an economic draft because all I had to do to 8 recognize that needed to be done is to look at how much it cost 9 to have plants down because of feedwater trips. So for 10 whatever reason it's desirable to do that.
11 Are there further questions?
12 MR. EBERSOLE: Yes. I'd like to ask, you have so 13 little time to watch a breaker do what it's going to do, it's 14 in milliseconds. I have difficulty feeling myself that you can 15 do any kind of a predictive measurement utilizing just that 16 time, considering both the shunt as well as the UV trip. Maybe 17 the UV trip has got a little more time, I think it's slower.
18 And every time invoke preventive maintenance they always have 19 the chance somebody is going to put it back together wrong.
20 MR. LITTLE: That's correct. )
21 MR. EBERSOLE: And I don't know what those odds are, l l
22 but it's not zero by a long shot. l 23 MR. LITTLE: That's true.
24 MR. EBERSOLE: So what is your philosophy about the 25 balance you strike between these, do you do any predictive
(
Heritage Reporting Corporation (202) 628-4888
192
(- 1 ' maintenance as a result of watching --
2 MR. LITTLE: Well, I think the key thing --_one of
~
3 the key lessons that was learned on the Salem event, after you 4 fix something you really should make sure you test it after you 5 fix it. And I think TMI taught us that as well with respect to ,
6 alignment. So that's certainly one element of it, the need to 7 do a functional test after you have maintained it.
8 MR. EBERSOLE: That's'after you have fixed the part.
9 MR. LITTLE: To use'that word loosely -- if you've 10 done that. I think the other thing is that, hey, naturally you 11 want to look for components that aren't going to be so reliant 12 on very, very detailed maintenance inspection procedures. And
, 13 you have got to have that defense in-depth in that.
I 14 We have always made the arguments in ATWS events 15 that, there was a lot of defense, inherent defense in the -
16 natural characteristics of the design.
17 MR. EBERSOLE: Well, the UV trip was a case in point, 18 it was highly susceptible to maintenance.
19 MR. LITTLE: Yes. But it also was directly corrected 20 with a manual trip right at that point in time. The shunt trip ,
21 attachment in the breaker is not an item'that requires 22 maintenance; it's a very simply component. That was the case .
23 where deenergized to actuate created the complexity that really 24 was a contributor to that problem.
25 We really feel that, hey, inherent in any testing,
( ,
Heritage Reporting Corporation l (202) 628-4888 '
193 1 inherent in any maintenance program is adequate means to 2 determine is it operable after I've maintained it. So'we 3 really believe in that as well.
~
4 MR. KERR: Mr. Davis.
5 MR. DAVIS: I had a question on the reason that you 6 have been able to reduce your scram frequencies per year.
7 MR. LITTLE - Yes.
8 MR. DAVIS: Has any of that been related to 9 adjustmentlof set points or changing the kinds of signals.
10 MR. LITTLE: Yes. It has been related to a number of 11 things. One is, we have looked at the set points we're using 12 in the protective systems as saying, are we giving the 13 operators particularly low power adequate margins to be able to 14 control this plant without generating a lot of trips and a lot 15 of challenges. And we found out there were a number of margins 16 that we could add with respect to, particularly steam generator 17 level in the sensitivities in controlling level from a 18 transient nature, that was one.
19 Other ones were, we ran work shops with operators and 20 developed better control procedures, better startup procedures 21 that would avoid getting the scenarios where you're balancing 22 steam -- levels and forced steam generators in trying to line 23 up the turbine at the same time. And we have seen some direct 24 improvements in those as well.
25 MR. DAVIS: My concern is, and I know you're aware of
(
Heritage Reporting Corporation (202) 628-4888
194 1 this also, you can go too far in attempting to eliminate the 2 number of scrams.
3 MR. LITTLE: That's correct.
4 MR. DAVIS: And you can actually make the scram 5- function less' reliability because you have removed some things 6- that scramble.
7 MR. LITTLE: I think you can'think back to TMI and 8 come to that conclusion without even thinking about ,
9 Westinghouse. Westinghouse has always had, and in the past 10 been criticized for having too many trip signals in its 11 protection system. We don't intend to take a lot of that 12 diversity that we claimed back in the early 1970's out. We 13 want to eliminate unnecessary complenity, but we want to 14 maintain a level of redundancy in diversity as well, 15 We're not going to rely on the last trip, the last 16 course of resort in those things. So you'll see in 17 Westinghouse analyses and in Westinghouse designs numerous of 18 trips. The final safety analysis report you'll see the numbers 19 of trips for single events.
20 MR. KERR There is, it seems to me, on an associated 21 question and that is, whether this current approach toward trip 22 decrements, decreasing trip rate will convince operators that 23 they should never intervene.
It seems to me that would be 24 unfortunate because there may be times when intervention is ;
25 desirable. And if there seems to be developing, as they
( 1 Heritage Reporting Corporation (202) 628-4888
l 195 1 probably did before TMI-2, a man says, you know, you don't ever 2 do this toward tripping, I would think that would be-3 unfortunate.
~
4 MR. DAVIS:- We have a similar situation with the 5 BWR's on liquid poison injection. There was a great reluctance 6 to ever use that.
7 MR. KERR: Tough questions.
8 Thank you, Mr. Little.
9 I'm going to suggest a 10 minute break at this point.
10 (Recess.)
11 MR. KERR: Presentation by NRC staff.
12 MR. FIENO: I'm Daniel Fieno and I work in the l 13 reactor systems branch at NRC , and I will be making the staff l
14 presentation.
15 Basically, a lot of the points that I'm going to talk 16 about have already been covered by the Owners Group, so a lot i
17 of what I say will be in the nature of a summary. l l
18 (Slides being shown.)
19 MR. FIENO: Let me start with, over the years there-20 have been a lot of changes in PWR's. And basically, for 21 example, there have been changes to extend the cycles of 18 and 22 24 months, low leakage core designs, discharge burner has been 23 increased. There have been a lot of new fuel designs. And 24 moderator temperature coefficient technical specifications have 25 been changed in quite a few plants.
(
Heritage Reporting Corporation (202) 628-4888
1 1
196- j 1 These type of things over the years have led the-2 staff to'lxa concerned about transient PWR MTC's. And in 3 particular we have been worried a little bit about what has 4 happened to the ATWS analysis basis. So that has been-the type 5 of reason that led us to send the letter on June 12, 1987 to 6 the various ?WR Owners Group.
7 We certainly recognize that the MTC is just one 8 factor in the ATWS analysis; and also, that there are other 9 factors such as was talked about earlier in terms of systems, 10 et cetera. So we're certainly not naive in this area. . But our 11 concern was related to strictly the analysis aspects, and in 12 particular, the MTC.
13 The objective of the letter is very simple, there is 14 no hidden agenda in the letter. We have a very simple mission, 15 and that was to uncover any trends in the MTC for the various 16 classes of PWR's. And particularly, the letter' asked for 17 justification for continuing applicability for conservatives in 18 an ATWS basis MTC. And we wanted to see if'there were any 19 outlier or unique plants that we needed to worry about.
20 21 22 23 r 24 25
(.
Heritage Reporting Corporation (202) 628-4888
--.____---.--a_
197 1 MR. FIENO: And we wanted to see if there were any 2 differences in current MTC's with ATWS basis MTC's. And we 3 wanted to see what kind of data was used. And we wanted the 4 vendors, Owners Groups to discuss their methodology and the 5 assumptions that they were making in this analysis.
6 One of the questions that was on the ACRS agenda, I
? going to skip around a little bit here, but I'm basically going 8 to follow the ACRS agenda.
9 And the second item on the agenda basically asks, are 10 there potential concerns with current reactivity operational 11 limits and potential for analyzed reactivity accidents 12 including operational experience or incidents of concern. And 13 as far as the staff is aware there is no incident that has 14 occurred or any operational experience regarding unanalyzed 15 reactivity incidents that we are aware of.
16 However, we would like to point out that RES does 17 have a program at B&L where they intend to assess previous 18 judgments on the adequacy of reactivity accident analysis that 19 formed the basis for design approvals, basically the Chapter 15 20 type of analysis. So there is the.t program that is being 21 formulated and initiated.
22 MR. WARD: Let's see, I'm interested in that program.
23 I understand that it hasn't really -- has it started yet or is 24 the contract signed or what is the status?
25 MR. FIENO: Well, I'm aware that a request for a
(
Heritage Reporting Corporation (202) 628-4888
198 1 proposal went out for a 189, but I'm not aware if the program 2 has actually been funded and started at this point in time.
3 And I don't have anybody here from RES to respond to that 4 question.
5 MR. RICHINGS: I can respond to that. Howard .
6 Richings, NRC. I have been relatively close contact with that 7 particular program. It's about to start now and it is 8 essentially the paperwork on fund is essentially done. The 9 first get together meeting on it, formal get together meeting 10 on it will be within a week or two as I understand it. We 11 already had many preliminary _ discussions on the subject. But 12 the program has not formally started, but will very shortly.
13 MR. WARD: And it's with B&L7 5
14 MR. RICHINGS: B&L, yes.
15 MR. WARD: Thank you.
16 MR. DAVIS: Is that to cover both, BWR and PWR?
17 MR. RICHINGS: Yes, it's the cover the activity 18 events of great significant and their probability and likely j 19 course wherever we can think of them.
20 MR. DAVIS: Does that include examination of the 21 analytical methods that are used for ATWS analysis?
22 MR. RICHINGS: It will examine types of events and ,
23 will use the Brookhaven resources on thair analysis methods.
24 -MR. DAVIS: Thank you.
25 MR. FIENC: There were some questions raised earlier
(
Heritage Reporting Corporation l (202) 628-4888
199 1 about accuracy, for example, measurements and calculations, and 2 I would like to bring out some data here.
3 Incidentally, there's a word missing here, I'm'really 4 talking about the accuracy of this particular measurement.
5 Here the isotherma temperature coefficient measurement accuracy 6 at hot zero power. .And basically, I picked this number out of 7 the ANS standard. They assert that for across the board for 8 PWR's that we could expect about a plus or minus two PCM per 9 degree for the ITC measurement.
10 And from the data I've looked at from the various 11 people doing calculations on PWR's, this seems to be a fairly 12 reasonable number. A lot of the groups do better than this.
13 So this is the kind of accuracy one sees for the ITC type of 14 measurements. And the recent standards speaks of that number.
15 MR. WARD: Maybe I misunderstood, but there are some 16 attempts to measure -- m6ke a measurement at power, also.
17 MR. FIENO: That is correct.
18 MR. WARD: And dcas the ANS standard say that at all?
19 MR. FIENO: No, it doesn't. It does not speak about 20 measurements at power, i.t speaks strictly about the hot zero 21 power case, the ITC in particular.
22 Now, just for the sake of completeness what I did l I
23 was, I took a Westinghouse PWR, vogtle FSAR, and I took some 1
24 numbers from their analytical methods. And again, this is sort 25 of typical of what other groups do in this kind of cal C. And
(
Heritage Reporting Corporation (202) 628-4888 i
200 1 I'm not going to run through the whole list here, but the 2 important point here is that for the various reactivity 3 coefficients, power defects, et cetera,'the calculations are 4 getting to bc fairly decent. One can see good accuracies.
5 tor example, on the critical boron concentration.
6 MR. KERR: What is meant by the accuracy of the 7 analytical methods?
8 MR. FIENO: In comparison with, for example, measure 9 data.
10 MR. FIENO: I think here -- the point here is that 11 the kind of cals that one sees for PWR's are getting to be 12 fairly decent and in terms of calculation of things like, 13 moderator temperature coefficients, et cetera.
14 For example, in the vogtle FSAR they quote plus or 15 minus three percent for part distribution measurements. They 16 don't break this particular number down, I presume that means 17 given noble basis for part distribution.
18 And again for the MTC, this particular FSAR quoted 19 two PCM per degree F on an MTC calculation.
20 Well, this particular agenda item concerned 21 basically, I think, the reactivity operational limits, the MTC 22 technical specificetion. And the kind of technical .
23 specifications on te, for a lot of plants run something like 24 this, the spec will stay something like the MTC should be no 25 more positive than say five PCM degree F, say, zero to about 70 1
Heritage Reporting Corporation (202) 628-4888
l l
I 201 i f
1 percent power. And usually what happens then, they'll ramp 2 down to about zero PCM per degree F from 70 percent power to-3 100 percent power. Sometimes they will step down.
4 But. basically the point here is that many reactors 5 now have that type of technical specification. And usually l 6 these specs are usually for the purpose of operational 7 flexibility, and also for performing Chapter 15 type analyses, j 8 MR. LEE: If I understand your statement about 9 ramping down from five PCM degree Fahrenheit.
10 MR. FIENO: Variation from 70 percent power to 100 11 percent power. In other words, if it's five percent or seven 12 percent, it would be zero here and you would draw a straight 13 line.
14 MR. LEE: So it could be five PCM degree Fahrenheit 15 plus five PCM degree Fahrenheit at 70 percent power.
16 MR. FIENO: Well, usually at -- or less than at 70 17 percent power, so there would be a break point.
18 MR. KERR: I think what he's saying is, at 70 it 19 can't be bigger than five, and at 70 and 100 you draw a l 20 straight line from five to zero.
21 MR. FIENO: Right. j 22 MR. LEE: What I'm curious is, at 70 percent _ power 23 tech spec allows up to five PCM degree Fahrenheit of MTC.
24 MR. FIENO: For a number of reactors that is correct.
25 MR. LEE: It has been like that all along?
(t.
Heritage Reporting Corporation (202) 628-4888 l
l l
202
(
1 MR. FIENO: Well, there's a lot of reactors that I
2 have come in for tech spec changes over the last few years._
3 That's part of the reason why we're here today, I think, is 4 that some of these changes that we have seen in tech specs have 5 led us to question, where are we going in MTC?
6 MR. KERR: From what we saw today we would have to 7 assume that permits some margin because nobody is really 8 operating with that much of a positive, at least if they are 9 they didn't tell us.
10 MR. FIEFO: At zero power they will be operating, 11 say, they could operate year their limits. They could operate, 12 say, plus two, plus three PCM. And some plants have done this 13 and bumped and come close to those limits.
14 MR. EBERSOLE: On the other side of the coin when you 15 look at the boiler with a void collapse the negative moderate t
16 coefficient gets to be a big problem. And there, do you do 17 actual tests of what the collects coefficient is? You know, 18 you've got a substantial on every turbine trip.
19 MR. FIENO: That is correct. Well, first of all, 20 those are usually a factor only to pressurization type of 21 transients, turbine trips, low rejections and things like this.
22 MR. EBERSOLE: But you've got a substantial power 1 23 spike when you trip a turbine.
24 MR. FIENO: That is correct, you do get a power 25 burst, t
Heritage Reporting Corporation (202) 628-4888 1
t
203
(
1 MR. EBERSOLE: Are there any cases where you -- with 2 faster quarters you measure that actual flux peak.
3 MR. FIENO: This has happened. We have done three 4 tests -- well, not we, but the industry, General Electric and 5 Peach Bottom conducted three tests at Peach Bottom at three 6 different power levels and these were heavily instrumented.
7 And basically they have been used by the industry to benchmark 8 their methodology for those type of events.
9 MR. EBERSOLE: Were the recorders fast?
10 MR. FIENO: They were special recorders specifically 11 set up for that test program. These tests were conducted back 12 in the '78 cime frame, three tests here and there was one test 13 overseas ct KKM, I believe.
14 MR. EBERSOLE: How did they match the calculation?
15 MR. FIENO: Well, the methodology had to be -- I 16 think everybody that is doing PWR counts basically had to go 17 from say, point kinetics to one dimensional neutron calculation 18 as a minimum. So that is one of the things that has happened 19 out of this, they found that point kinetics wouldn't do the job 20 basically; that in some cases it would be ccnservative and 21 others not, depending on consideration of the analysis.
22 So I think those tests -- that testing did lead to 23 significant changes in how one does BWR over-pressurization 24 type calculations.
25 MR. EBERSOLE: You have done three tests then; three
(
Heritage Reporting Corporation (202) 628-4888
204 1 tests?
2 MR. FIENO: There was actually three tests that are 3 available to the industry, but I think there's a couple more
~
4 that General Electric has -- KKM tests, overseas tests.
5 MR. WARD: When you agree to a tech spec of the sort 6 you have shown there, which is -- what do you accept as proof 7 that they met the tech spec?
8 MR. FIENO: Well, in the past we have only been 9 concerned about Chapter 15 type of analyses. However, in the 10 back of our minds we've always had -- again, this is -- I would 11 like to make tha point that this letter we sent out in June is 12 not an all of a sudden type of thing. Over the past few years 13 we have been bothered a little bit about these type of tech 14 spec change in terms of how do you handle vis-a-vis ATES.
15 And so in a certain~ sense we -- I guess you might say 16 we ignored the ATWS question in the past to a certain degree as 17 to how it fit in. Although, we certainly discussed it amongst 18 the staff. But eventually it's culminated in this particular 19 letter that we sent out to the industry.
20 So in the past you could say that we addressed it 21 only in support of Chapter 15 type of analyses and reviewed it 22 on that basis.
23 MR. WARD: I didn't make my question clear. It'n 24 really for any tech spec, not number, what do you accept an a 25 demonstration by the licensee that this reactor loading meets 1
Heritage Reporting Corporation (202) 628-4888 l
205 1 that requirement?
2 MR. FIENO: You mean if a utility has this particular 3 spec?
4 MR. WARD: Yes. If he has got that spec --
5 MR. FIENO: Right at this moment, again, I'll just 6 repeat what_I just said. That the Chapter 25 analyses is the 7 type of a basis on which we accept this, that they meet all 8 the --
9 MR. WARD: No, no. I mean, you require that he makes 10 a zero power measurement?
11 MR. FIENO: Well, all the PWR's in general will make ,
12 the zero power measurement for the ITC, that is correct. I am 13 not aware of any that don't make this particular raeasurement.
14 MR. WARD: But you don't require that. I'm trying to 15 find out how you enforce this tech spec or how you inspect for 16 compliance with this tech spec? ;
17 MR. FIENO: Well, as usual the compliance with the i 18 tech spec, the utility have to comply with these specs in some 19 fashion. he don't have, for example, a_-- we don't require l 20 that this PWR or that PWR do a specific set of startup tests.
21 However, we do endorse the ANS standard 'in a sense as 22 being a minimum set of tests. And we have worked on that 23 stendard. And so that seems to be an industry consensus.
24 So at this stage we sort of say that's the kind of 25 testing that should be done as a minimum. !
(
Heritage Reporting Corporation (202) 628-4888 I
206 1 MR. RICHINGS: The tech spec itself requires -- has a 2 surveillance requirement which requires a measurement of 3 moderator coefficient at the beginning of life and end of life. ,
4 MR. WARD: By the licensee?
5 MR. RICHINGS: Yes, by the utility.
6 MR. WARD: But then that measurement is at zero .
7 power, I guess.
8 MR. RICHINGS: Yes. It's a one point measurement.
9 MR. WARD: So then he -- .
10 MR. RICHINGS: There's little uncertainty. If you ,
11 measure one state point of the reactor there's little --
12 there's very little uncertainty in knowing what the remaining 13 state points of the reactor are about. So that's all that's 14 really necessary, one confirmation that the reactor has been 15 loaded correctly and can meet one well measured state point. '
l l 16 The rest of the state points follow.
17 MR. WARD: So you're saying that the relationship 18 between a measurement, let's say a fresh charge, a measurement i r
! 19 at zero power and that the correction of that number to full 20 power is very straightforward, well known, everybody agrees, 21 there is no doubt about it.
22 MR. RICHINGS: Yes.
23 MR. FIENO: Then also, as Hnward has indicated, not l 24 only do you have those measurements, but you also have the core j 25 follow that you do. So if anything goes wrong in that cycle I
(
1 Heritage Reporting Corporation l (202) 628-4888 i 1
207 1- you would know about it in terms of whether 'the power 2 distribution.was getting wacky or something. So you do have 3 that e pect of the surveillance that goes on normally in a 4 core. And you know the cycle is going to start here,-it's 5 going to end here and you're going to followfa certain boron 6 letdown curve and things like this. So you have a good idea 7 what's happening core-wise. And if it doesn't happen then you 8 know you have accounts for bad or something, but as far as we 9 know this happens, if at all, very rarely. ;
t 10 I don't know if this answers your question or not.
11 MR. LEE: I'll try to see if I understood you j 12 correctly. So for certification that this particular tech spec ;
13 has been met, one needs to take single measurement near the t,
14 beginning of cycle at hot zero power and show that it is below 15 plus 7 PCM degree Fahrenheit, accounting for this ANS.
16 MR. FIENO: Wait-a minute. Go ahead.
17 MR. LEE: ANS standard allowance plus or minus-two 18 percent. Then you would say the license has met the 19 requirement.
20 MR. FIENO: Well, I would have to be a little careful 21 here because sometimes the -- yes, all right, .if you want to I
22 look at it that way.
23 But again, I would point out that the important thing 24 here is that, not only the startup measurements show a ball of 25 wax, but-you've got to look at how the core operates tnroughout
(
Heritage Reporting Corporation (202) 628-4888
208 1 that cycle. And presumably if it operates correctly, then the 2 coefficients that you calculate are going to be okay. That you 3 don't feel that they're going to stray outside of your 4 celculation bounds.
5 MR. LEE: This is again for all rod out on condition .
6 aa measured.
7 MR. FIENO: You mean hot zero power, yes.
j 8 MR. LEE: All rod out -- all control rod withdrawn.
9 MR. FIENO: Right, that's an all rods out, yes.
10 What I'd like to do is just briefly go over the 11 status as I see if. In response to our letters we've had two 12 meetings with Westinghouse and the Westinghouse Owners Group.
13 The first one occurred back in October and the second one just i
14 a week or so ago. And we met with the Combustion Group back in l
15 January. And the B&W Group yesterday morning.
16 The CE -- and the CE Owners Group meeting was the 17 first one where we actually got some MTC data. And the first 18 Westinghouse meeting basically was an exploratory type of
! 19 meeting as was indicated earlier by the Owners Group.
20 As you have seen from the Westinghouse presentation, 21 we did receive some MTC data in the second meeting.
22 The B&W presentation, I think you saw that earlier ,
23 this morning, that gave us some indication of the trends in 24 their coefficients.
( 25 And I would like to summarize the data in this l
I (
l Heritage Reporting Corporation (202) 628-4888 l
209 1 particular slide here as I see it. I apologize.
2 Here I've put everything in the same units, PCM per 3 degree F. In the Westinghouse case their previous number on a 4 95 percent basis was minus 8. And from their data it seems'to 5 be that their number is around 10 right now.
6 And for CE these numbers are numbers you saw earlier 7 this morning for their 2750 megawatt class, slight improvement ;
8 and for their other two classes a slight decrease in the 1
9 coefficient. '
10 And for the B&W class of plants I have put down two 1: numbers here, for 18 month cycle number the differences is very 12 small, minus 10 and a half to about 11. However, for their 24 13 month cycle there appears to be a significant change. So this 14 is the status as of now -- as we see this factor.
15 And basically the bottom line as far as the staff is i 16 concerned on MTC data is that, from the data we've seen, we 17 don't see that there has been any adverse trend in the actual 18 MTC's that the plant sees, in spite of the fact that we have 19 changed the tech specs in various degrees, et cetera.
20 So at least from this part here we feel that --
21 MR. WARD: Well, there's no adverse trend but that's 22 only because the B&W calculations, the current 24 month 23 coefficient, they come out with a number that -- fcr the level 24 C pressure, whatever you call it.
25 MR. FIENO: Service level C.
(.
Heritage Reporting Corporation (202) 628-4888
. - . . . ~.
210 I i
1 MR. WARD: But there certainly has been an adverse 2 trend.
3 MR.'FIENO: For that particular -- there is just one l 4 plant that has gone to a 24 month cycle for their class of 5 plants. And if all their plants went into that direction, I
- l 6 guess you would call that -- l 7 MR. WARD: One plant don't make a trend, I guess.
8 MR. FIENO: I-guess at this stage, perhaps not, but 9 maybe later it would.
10 Basically, I think this is the presentation as I see 11 it. Are there any questions.
12 MR. KERR: You do not propose to deny other plants 13 that come in for the request for 24 month cycles and these 14 kinds of MTC -- you would handle those the way you would handle 15 the first one, as far as you know at this point.
16 MR. FIENO: Well, I think -- well, in a certain sense 17 I think in Westinghouse's case that they're going to eventually 18 tackle the whole ball of wax, so to speak, and tackle both '
19 systems aspects and operational data and what have you. So in 20 that case their -- I don't anticipate any problems.
21 MR. KERR And your criteria will be what, that they 22 satisfy Chapter 157 23 MR. FIENO: No. We were looking to see if there were 24 any adverse trend to the analysis basis, the ATWS rule, at l 25 least as far as the MTC was concerned.
)
Heritage Reporting Corporation (202) 628-4888
=
211 1 MR. KERR: What would the adverse trend be?
2 MR. FIENO: Well, for example, if -- pick any number 3 there. If the number say went from minus eight to minus two in 4 the Westinghouse case, if I saw a number like that I think I 5 would be quite concerned, especially if their other plants now, 6 say, would exceed service level C; then I would say, wait a l 7 minute, what's happening, you know. I think that's the type of 8 thing we are looking for.
9 MR. KERR: Well, my impression was that some of the l
10 plants, not Westinghouse, already exceeded level C and they're l 11 willing to exceed it even more, although still for a fraction I l
12 of the time less than 50 percent. Did I misunderstand l 13 something? I 14 MR. FIENO: I don't think --
15 MR. KERR: It seems to me if the moderate temperature l
16 coefficient goes down you're going to get more pressure --
17 MR. FIENO: Absolutely.
18 MR. KERR: But that's not a trend that you worry 19 about. I'm not trying to put words in your mouth, I'm just 20 trying to understand what basis.
21 MR. FIENO: Let me back track a minute and let me try 22 to explain. The staff at this point here has not really fully 23 resolved how they're going to handle MTC type changes. We have 24 not fully resolved that as to how we expect licensees to deal 25 with this. That's still under discussion amongst various staff
{
Heritage Reporting Corporation (202) 628-4888 v
212 1 members. That is the real answer to your question.
2 MR. KERR: And it will be -- you expect to resolve 3 this question by next Tuesday, maybe.
4 MR. FIENO: I doubt that soon, but we will. Is there 5 any special -- what's special about Tuesday.
6 MR. KERR: Well, it's just better than Wednesday, 7 maybe.
8 MR. FIENO: We still have it under discussion amongst 9 ourselves and we're still not sure how we intend to treat these 10 items on a plant by plant basis.
11 MR. KERR: Any further questions?
12 (No response) 13 MR. KERR Thank you, sir.
14 This concludes the presentation part of the meeting, 15 but I do want to get some comments that any of you have and 16 suggestions for any further investigation you think we should 17 undertake.
18 I'll start with you, Mr. Ebersole.
19 MR. EBERSOLE: I certainly noticed the absence of GE, 20 and I never have felt that the GE issue has ever been put to 21 bed; it has been compromised by a variety of fixes.
22 I get a confused picture about the effectiveness and 23 degree of mitigation of the ATWS event and a BWR as it's 24 represented by operator participation. I don't understand how 25 much the operators got to do and what time.
(- 8
, Heritage Reporting Corporation (202) 628-4888
213 1 I hear a variety of_ stories about trying to get more 2 neutron loss by running the level down. That we have automated 3 boron injection. But we still have that blasted dump volume e
4 which is, as far as I'm concerned, is a black mark on the 5 industry until we do something about it. And as I look, one 6 possible -- I would just as soon put it on the record -- one 7 possible augmentation to mitigation would be to have operator 8 control or automatic control boil down of the common dump 9 volume te the torquet, which is a slight approach to what 10 they're going to do BWR. There it's automatic, every rod has l 11 its own discharge to the torquet. Here I would just invoke 12 that sort of approach to award, what I regard, as an ;
l 13 unimportant fear of a modest amount of radiation leaks to the ;
14 torquet which would occur if you had a release there.
15 And two, if you had obtained a signal that the rods i 16 were seeded.
17 As to the other things we've heard here, I'm much j 18 interested in the systematic changes that Westinghouse i 19 mentions, and I don't know what they mean but I would like to 20 follow that with much ir.cerest as to whether they're going to 21 stick with just two breakers as they have now or do something 22 else.
23 I get at least a slight flavor that they might 24 augment their, "diversity" in the killing of power to the rods.
25 That's all I've got to say.
Heritage Reporting Corporation (202) 628-4888
214 1 MR. KERR Thank you.
2 Mr. Ward.
3 MR. WARD: Well, just a couple comments. Let's see, 4 Mr. Fieno said the staff will be developing some sort of 5- procedure or something to-follow up the ATWS rule in this 6 rega rd, and perhaps there are other things that should be 7 controlled, too, in addition to this. I think that's needed.
8 I'd like to hear how that's progressing at some time in the 9 future, perhaps later than next Tuesday would be all right.
10 I would sure like to hoar what the boiler water 11 reactors have to say about this or what the situation is. -
i 12 We didn't hear from them today, but I guess could, ;
13 what is your position on it, you could have presented us in the 14 summary slide some information on the boilers. Is it not 15 available or you haven't been able to --
16 MR. FIENO: We did look at the BWR case quite 17 extensively, but it's a much more difficult reactor type in (
t 18 terms of the analysis basis. So we're atill thinking about 19 that in terms -- we haven't come to any conclusions on it.
20 It's more difficult to come to grips with that. So that's part l
21 of the reason why they're not here.
2? We haven't formulated how to approach it. And for e ,
23 PWR meeting 1? .uldn't h,sve been appropriate for them to come.
24 But eventu- w will er no grips with that.
25 l' ,.
et's see, a couple more comments.
. i Heria. Reporting Corporation (202) 628-4886 l
1 3
l 215 '
)
1 You know. this whole ATWS question, Westinghouse seemed to be i
2 in the best shape going into the issue and seems to be in the 3 best shape coming out. I was glad to see that they seemed to 4 acknowledge it. They just have better PR or.something, but at 5 least they seemed to acknowledge it as a safety issue of some 6 significance and not just a regulatory nuisanco.
7 I guess I didn't really appreciate what I -- rome of 1 8 the looseness of the ATWS rule for four or five years ago.
, 9 Maybe that's all right. I was kind of surprised at the CE 10 analysis, they seem to be in the weakest shape, you know, among i
11 these three PWR's. l l
12 Then there's -- I couldn't quite figure out whether 1
13 they really fully comply with the ATWS rule or whether that's 6
14 still being negotiated. ;
l 15 Do the CE plants fully comply with the ATWS rule with 16 their present status or is that still under discussion?
17 MR. FIENO: If you mean, are they complying with 5062 18 in terms of AMSAC and diversq scram, I think the answer is, i
19 yes. I'm not familiar with the status of their SCR in that i
]
I 20 area. .
- l 21 MR. KERR: When you say comply, do you mean is the 22 hardware in place and operational?
23 MR. WARD: Well, no, I guess it's not all in place. 1 24 MR. KERR No, I just didn't understand what you
(
25 meant by comply.
(.
Heritage Reporting Corporation (202) 628-4888
+
216 1 MR. WARD: No. What I meant was whether the results 2 of their analyses even with new hardware in place, bringing 3 them in conformance with the original intent of the ATWS rule.
4 MR. DAVIS: Exceeding level C.
5 MR.-WARD: Yes. I guess I just didn't have a clear ,
6 enough picture of that. But I still don't -- that's up to me 7 to straighten out for myself on this.
8 MR. KERR Kr. Wylie.
9 (Continued on next page.)
10 11 12 13 14 15 16 17 ,
18 i
19 :
20
~
21 22 !
\
23 ;
24 25 C
Heritage Reporting Corporation (202) 628-4888
l l
217
{
1 MR. WYLIE: The other point is that I think that i 2 some effort should be extended by B&W in coming up with the 3 reliability of the reactor protection system, as we discussed 4 earlier. And also in light of the problems with the 5 Westinghouse circuit breakers, I think that the diverse scram 6 system deserves more attention than is being giving. I would 7 like to see something more concrete done in that area.
8 MR. KERR: Mr. Davis.
9 MR. DAVIS: I guess that I am also concerned _about 10 the BWR question as raised by Mr. Ebersole and Mr. Ward. That 11 is a much more complex system, and I have always felt that the 12 problem there may be a little more serious. I am also
, 13 concerned I guess a little bit over the reluctance of B&W and ;
s 4 14 CE to perform realistic unavailability analyses of their scram 15 systems. Weetinghouse seems to have done a . lot more in that 16 area.
17 I have not heard anything at this meeting I think 18 that is cause far concern about many of the trends. I think 19 that I agree with the staff's evaluation. I am also confused 20 about CE's compliance with the ATWS rule. It seems like they 21 are marching to a different set of rules somehow. And I guess 22 that I will have to try to sort that out myself also. That is 23 all that I have.
24 MR. KERR: Mr. Lee.
25 MR. LEE: My impression of the overall trend on the i
(
Heritage Reporting Corporation (202) 628-4888 l
218 1 MTC is perhaps that with CE and Westinghouse that there has not 2 been an overall trend toward less negative MTC at full power.
3 But B&W from the cycle perhaps has seen that trend. And in the 4 case of the Westinghouse reactors perhaps for two plants, such 5 a trend has taken place. ,
6 And I would like to see how thoso trends translate 7 into MTC values at low power. Not at hot zero power exactly, 8 but at low power below 40 percent or 50 percent power, 9 somewhere there. And I suspect that even with Westinghouse's 10 two plants perhaps at substantially low power. And you may be 11 approeching even positive MTC at some part power.
12 I would like to see some actual data for the two 13 plants in particular, and in general for a few selective 14 individual pir.nts that have seen the trend toward less nagative 15 MTC even at hot full power I think. 4 16 And in general, what this trend for some selective 4 17 plants means in terms of the transient that has not been 18 traditionally considered as the limiting. I would like to see 19 if indeed those traditional limiting trends will remain as 20 limiting trends even with this trend in MTC.
21 And a very small technical point again. If you go 22 above 3000 or 4000 psi, I would like to see if indeed some of ,
23 these transient models that I usod to predict power plant 24 behavior can be trusted as much as at normal operating i 25 conditions. Those are my two comments, I think. !
(.
l Heritage Reporting Corporation I (202) 628-4888 l
219 1 KR. KERR: Mr. Lipinski.
2 MR. LIPINSKI: I would like to support this last 3 comment. If you are talking about a plant going to 5000 or 4 6000 psi, whether it is in the five percent window or not, I 5 think takes special consideration in terms of what would happen 6 to a plant under those conditions.
7 The other one, I will reiterate what has been said 8 earlier with respect to the Combustion Engineering solution to '
9 ATWS. Based on the discussion that took place during the eart.y ;
1 10 hearing, I am surprised to hear what I did today relative to I
11 Service Level C.
12 And the Westinghouse scram breakers. During the 13 original ATWS discussions, they had relief capacity, so that 14 the diverse scram system was not required by Westinghouse. But
~
15 the performance of these breakers leaves a lot to be desired to 16 see the frequency in which failures are occurring.
17 I would like to hear the NRC's staff comment whether !
18 they are considering anything on the liability of those ,
19 Westinghouse breakers.
20 MR. EBF.RSOLE: Bill, may i add another comrmat?
21 MR. KERR Yes, you may.
22 MR. EBERSOLE: All of these improvements to the PWR
, j 23 scram systems still result in high pressures, and Westinghouse i 24 is in the best shape as I see it. And then B&W is next, and CE l 25 is last. And in looking at these high pressures at least with i
(
Heritage Reporting Corporation l (202) 628-4888 t
I 220 1 B&W and Westinghouse, it seems to me that if we are going to 2 introduce those high pressures as a concept of survival, 3 mitigation, or whatever you want to call it, that it would be 4 mandatory that you look piece-wise in the fine structure of the
! 5 survivability of the vessel, in short not blowing the head off.
\ =
! 6 And that would be a much more detailed study i 7 analogous to the PTS type of study than we personally have on i 8 board now. Since it is here and we have to deal with it. I 9 realize that you said that if you have to invoke survivability 10 at that level, that you are not very happy, but what else are 11 you going to do, put some more hardware on the scram system, 12 add some more rods.
l 13 MR. KERR: There are a number of things that you can 14 do. I do not know how practical they are.
15 MR. EBERSOLE: I like Service Level C as a common 16 goal.
17 MR. KERR: One could put in more capacity in the 18 pressure vessels. Whether this is practical, I do not know.
19 It may be impossible with vessels that have been used. And one 20 could put in more burnable poison in the core.
21 MR. EBERSOLE: By the way, what is the boiler peak 22 pressure now, does anybody remember? I do not.
23 MR. KERR: Are there any further comments?
24 (No response.)
25 MR. KERR: Well, we thank all of those who had the N.
l Heritage Reporting Corporation (202) 628-4889 i
221' 1 persistence to stick around-this long for their interest and 2 participation. And I appreciate the information that has been 3 provided. I hope that we can do something sensible with it.
4 And I thJun that the presentations have been informative and 5 well-organized. Thank Eyou.
6 (Whereupon, at 3:22 p.m., the committee adjourned.)
7 8
9 10 1
11 12 13 i
14 15 t
16 17 18 19
- 20 21 22 j 23 24 1
25
(
Heritage Reporting Corporation (202) 628-4888 t r --
1 CERTIFICATE
[ 2 3 This is to certify that the attached proceedings before the 4 United States Nuclear Regulatory Commission in the matter oft
- 5 Name: GENERAL MEETING 6
7 Docket Number:
8 Place: Mashington, D.C.
9 Date: February 19, 1988 10 were held as herein appears, and that this is the original 11 transcript thereof for the file of the United States Nuclear 12 Regulatory Commission taken stenographically by me and, 13 thereafter reduced to typewriting by me or under the direction ,
14 of the court reporting company, and that the transcript is a i I 15 true and accurate record of the foregoing proceedings.
16 /S/ ch du 17 (Signature typed): Joan Rose 18 Official Reporter 19 Heritage Reporting Corporatien 20 21 1
. 4 22 ]
23 1 24 i
25 k
He:1.ag. naporting Corporation (202) 626-4888 l
t f
l l
BABCOCK & WILCOX ;
OWNERS GROUP ,
ATWS COMMITTEE EFFECTS OF PLANT AND FUEL
. CHANGES ON.ATWS BASIS
- (
PRESENTATION TO: L ADVISORY COMMITTEE ON REACTOR SAFEGUARDS I FEBRUARY 19, 1988 J
- l l
6 i
PURPOSE OF PRESENTATION o REVIEW ATWS REQUIREMENTS FoR B&W PLANTS i 1
- o EVALUATE PLANT AND PROGRAM CHANGES -
WHICH HAVE AFFECTED ATWS BASIS
( o EVALUATE FUEL AND FUEL CYCLE CHANGES WHICH HAVE AFFECTED ATWS BASIS o ADDRESS ADEQUACY OF CURRENT PLANT j CONFIGURATIONS RELATIVE To ATWS BASIS ,
1 t
I
{ i
- r
I l
1 PURPOSE OF ATWS RULE IMPROVE THE DESIGN AND OPERATION OF NUCLEAR POWER PLANTS TO:
- 1) REDUCE THE LIKELIHOOD OF A FAILURE OF THE REACTOR PROTECTION SYSTEM TO SCRAM FOLLOWING ANTICIPATED TRANSIENTS, AND
( 2) MITIGATE THE CONSEQUENCES OF ANTICIPATED TRANSIENT WITHOUT SCRAM EVENTS.
THE FINAL RULE REQUIRES THE INSTALLATION OF DSS AND AMSAC AT B&W DESIGNED
. PLANTS TO i MEET THESE OBJECTIVES.
l '
l
(
FINAL RULE e DSS (DIVERSE SCRAM SYSTEM) e NON-SAFETY RELATED e SINGLE CNANNEL e DIVERSE FROM RPS e MILD ENVIRONMENT EQ e NO SEISMIC e QA GENERIC LETTER FOR GUIDANCE
{ e POWER - DIESEL BUT NOT IE e TESTABLE AT POWER
- e INPUTS DEFINED BY UTILITY e
AMSAC (ATWS MITIGATION SYSTEMS ACTUATION CIRCUITRY)
I e AUX FW INITIATION ON ATWS SIGNAL o TURBINE TRIP ON ATWS SIGNAL I
i I
O h
EVENTS ANALYZED ;
t i
I r
e f
4
, . f
. I l
i l
l i
t i
I !
6 L
i b
I
^-
B W PLANT ATWS Al "lSES BAW-10099 PLANT DSS BAW-10019 BAW-10016 BAW-10099 REV 1 BAW-1610 SPECIFIC ANALYSIS EviENT/ DOCUMENT (1970) (1972) (1974) (1977) (1180) '(1981) (1985L_
stuck OPEn PSV X X X DRAID LINE BREAK X Loss or Loan X X Loao INCREASE X 501 Loss or FEEDWATER X Loss or FEEDWATER ,I. X X X X X X Less er OFFsITE POWER X X X X CONTRot Roo MALoPERATION X X HooERAToa DILUTION X X 1 PuMe CoAsTooWN X X 2 PvMr CoasToown X X X X 4 Pune CoAsTooww X X LOCKED rotor X Pune STARTur X Roo WITuonAWAL FRon STARTur X X Roo WITwonAwAL FROM Futt power X X X X 1 ::o EJECTION X
i F
PEAK RCS PRESSURE FOR ATWS EVENTS (NO ROD INSERTION)
NR-0460 EVENT CATEGORIES PRESSURE, PSIA 177-FA
- 1) LOSS OF MAIN FEEDWATER* 3464
- 2) LOSS OF 0FFSITE POWER
- 3207
- 3) ROD WITHDRAWAL <2750 .
- 4) BORON DILUTION <2600 '
- 5) LOSS OF PRIMARY FLOW <2500
(
- 6) LOSS OF ELECTRICAL LOAo <2500
, 7) LOAD INCREASE <2500
. 8) INACTIVE PRIMARY LOOP STARTUP <2500
- 9) EXCESSIVE COOLDOWN <2500
- 10) PRIMARY SYSTEM DEPRESSUR.T.ZATION <2500 L i
- ASSUMES AMSAC FOR MITIGATION '
i l
i h
. l 2 ( i l
i
f i
CONCLUSION i
LOSS OF MAIN FEEDWATER (LOOP) !
- .r
. IS THE SIGNIFICANT ATWS EVENT i
[
i f
i s
}
-i I
.i :
) !
l i
l i
1 i
i I
l 1
i i
I l 1
i
.BAW-1610 ANALYSIS OF B&W NSS RESPONSE TO ATWS EVENTS JANUARY, 1980 e PERFORMED IN RESPONSE TO NUREG-0460 AND THE REQUEST OF FEBRUARY 15, 1979, FOR "EARLY VERIFICATION" REVIEW
, EVENTS ANALYZED: 1) LOSS OF MAIN FEEDWATER
( 2) LOSS OF 0FFSITE POWER
- 3) TWO REACTOR C00LANT PUMP COASTDOWN MAJOR ASSUMPTIONS: 1) 95% MTC
- 2) NOMINAL INITIAL CONDITIONS
- 3) NO SINGLE ACTIVE FAILURES
- 4) NO CRA INSERTION, ALL OTHER SYSTEMS FUNCTION NORMALLY i
- 5) TURBINE TRIP AND AUTOMATIC AFW (AMSAC) FUNCTIONED AS REQUIRED
I i
i PLANT SPECIFIC ATWS ANALYSIS FOR A LOSS OF MAIN FEEDWATER SEPTEMBER 1981 o PERFORMED TO DETERMINE PLANT SPECIFIC DIFFERENCES ;
! ON PEAK RCS PRESSURE
- EVE,NT ANALYZED: LOSS OF MAIN FEEDWATER i
i (
l MAJOR ASSUMPTIONS: SAME AS BAW-1610 EXCEPT: ,
A) 90% HTC WITH A 1 PCM PENALTY FOR '
e FOR COMPARATIVE PURPOSES, BAW-1610 GENERIC RESULTS J
CAN BE ADJUSTED BY 300 PSI (3464 TO 3764) TO BE i
- CONSISTENT WITH PLANT SPECIFIC ANALYSIS '
),
ASSUMPTIONS.
! l
- 1 4
i i (
- l 4
)
l
, n PLANT SPECIFIC PEAK PRESSURE FOR LOSS' OF MAIN FEEDWA ER 4.6 - - - -
4.5 -
4.4 -
4.3 -
4.2 -
3 a
4.1 --
v 4.0 -
wa +
ru 3.9 -
3C ,
$0 u 3 3.8 -
+
EE l
3.7 - + +
y t 3.6 - .
g 3.s -
3.4 -
3.3 3.2 -
3.1 -
3.0 i i i i r i i ANO-1 DB-1 SMUD 161O OC- 1.2.3 TM1- 1 CR-3
_ , , , _ _ _, n- 'v " ' "' ' ^
e i
8i
- t. ica t
3 g$ 5: s! t w
, I It I 13 5.
b@ .
@ C I -*P-*
- a a nb* -
I r-C
~
]I t
'}~
s P. .
e T
~
I
-4 4: a 5
j i C iEI -
a 1~ -
3 s.. ,
..l.I;. - . ) II) - -
g -
o !
- ilii
!!n.
Ii! t
- - i! ri
- .l I
i a
u!i.:
i! ii 5 gi 2 I s
!i j' g 1 a '
-l-ri W!
qiw .~
1" 1 -
,j E.
a j g e
( /
_4 .e.
I 11 1
_3
.I
~
g
'tw g
m .
t, ,
g,g I
<st F9 I-l4L E '
!jh!
I 3h M \ ai 1
}If; i== N a-G : {
- M
- . u) b. :. -
. .. i lg.sr. !!
._ 'g.
.lil !
^ ** :- i n L- 1 !
. .i t .r .
. II.
./
J' -
2,
[ 3
- '{
_ g a g .. _ 3 , !
w .
i[
> 4 -
>t ei t 1i -
! --@ II A! f t gi A. v IC A' ' - l l i* s*! -
A L"j 4 !
A -
3r '
( ;i !!i l i -
v! _< ;
6 3 39 t
t
/ /
a
% % > > a
% M N M e g -* S5 -- E5 -. MWt' MW a
$$ -MEW
, <8m EW Ea <58 m5o$
-=
-= =82 d si d d n 6
- u
_ m e w
5 w 5 %
2 - 2 m m a & 5,.E 52 '
= a m
$ 5 %,
- 5
<% 5 5a
< = <
C -
' - O k
d a
t w
W X % wJ
~ Z Q
- e2 ~
2 C 2 -* s=
= m C e
cm w m
o b
W $ *--ti <+ W 5
< < = <
8- m 5 M 5 8-g~e _- O c W :
"$ t
< < 5 = 5 3~ l a a a a E G E W 5 W- 0 W 8 W-m 5~
_$ 8 g
8 "
$ 5" a g m ~$
ea
( w s s<,m
= - <
m -
w a ,ea 4 l
W W l 8h5 "
8 h "t 0
O
< e l
d a d \
2== e c b g y -* s ,, = $ 5 = - ; g' + Emg " y" OO & &
< v C. CO A v C.
n A 4 a !
~
d w d 0 5 $ ,
5 S '
d gh s a
- c. c.
E
! a: > h a j =
=:,= 5 m
h Wd e EE 5 g 4
.E Bv Ea.
n4 4h i
$ HW HW $
a s Wd N" 8.- 3 u G
E e$ e
^E" a$" E W
=
l n n n b <
w s 1 b E e o- 0 3 l
- 5 - 5 l
I
. 8- = G 5+ 8- 8- + G 5 ; 8-1- W t "O G- "O G- W t a e
< < R < < = m n 3 a a 3 i B $ $ $ B 5 e m a g a g m e m t
" ' a 5 W W W E W 5 W 2 : g : : ~ p E $ g a
- 0 g$ g a i
= g a g o -
e = = =
w- E_m W.= W= n
< m
_ = =
m.
i l
l I
~
\
I ELEMENTS OF ATWS RULE !
PURPOSE OF ATWS RULE: i REDUCE ATWS RISK BY t FACTOR OF 2
(
r- -
j r----- 1----- , -
1 RELIABILITY l IMPROVE RTS UNACCEPTABLE l PROGRAM l RELIABILITY BY CONSEQUENCES i j L_____________J FACTOR OF 10 < 50% or TIME :
i i 4 I i
i i
I
( j i
s 4
PLANT AND PROGRAM CHANGES I WHICH HAVE AFFECTED ATWS BASIS 4
i e
e I
I 1
l l l 4
i l l I
__ - . _ _ . . - , . _ _ _ _ . - - _ _ _ . ...-_.. ___ .1
i.
PLANT IMPROVEMENTS WHICH
, REDUCE ATWS RISK I
TRANSIENT REDUCTION PRA BASED RPS SURVEILLANCE l -
SAFETY AND PERFORMANCE IMPROVEMENT PROGAM (SPIP)
ROOT CAUSE DETERMINATION PROGRAM IRANSIENT ASSESSMENT PROGRAM ,
I(
TRIP RELIABILITY RTB MAINTENANCE AND SURVEILLANCE SHUNT TRIP RTB ONLINE MONITORING GENERIC RTB MAINTENANCE GENERIC RPS MAINTENANCE PROGRAM SPIP ROOT CAUSE DETERMINATION PROGRAM TRANSIENT ASSESSMENT PROGRAM I
(
i
n - .,
li Unplanned Aulomalic Scrams While Critical .
- Industry Average l i I
8- 7,4
~
6.1
.1980 1981 1982 1983 1984 1985 1985 1987 1990 goal l -
i
n ~ .,
1 l B&WOG LOFW EVENT FREQUENCY 7
8-2 E 5-b M
N b 4-a b \
u 3-u fo 2- - 3
.4 1-0 i i i i i i i i 1981 1982 1983 1984 1985 1988 1987 i
O , O 4 A .#3 . .-
Breakdown of SPIP Improvement Recommendations EFW (7.2%)
~
OTHERS (24.3%)
MFW (22.7%) t i
t IA (18.8%)
ICS/NNI (16.6%)
/
ELE (5.0'/c)
) SSPC (5.5%)
\
Reactor Trip Breakers Failures (Falls to Open er Opens > 100 m:)
e 7-i e-s.
G e
i
> 5- '
s.
e 4-e e
le 3 3-I be E-1-
0 , , , , , i i i i y ;
1s7s isso 1ssa iss4 issa 198a
(.
l l
1 l
4 FUEL AND FUEL CYCLE CHANGES f WHICH HAVE AFFECTED ATWS BASIS f
a l
l I
i I
i l
l i
l i CHANGES AFFECTING l l ATWS CONSEQUENCES r '
' s AFW RELIABILITY s CORE DESIGN CHANGES j O
i
! I i
1 n- .._.-n.,
- , , - _ _ , ._..,..._..,...__.._.-____.,._,..,,,-,___,,,,--._n.+.--n,, -_
l l
1 l ,
DESIGN CHANGES OVER THE LAST 5+ YEARS SOME IMPACT ON ATWS ANALYSES d
o 18 - 24 MONTH FUEL CYCLES MINIMAL IMPACT ON ATWS ANALYSES i
o ZR GRIDS 8
1
! \
o GRAY APSRS l o AXIAL BLANKET o EXTENDED BURNUP MECHANICAL CHANGES 1 1 1'
o HIGHER BilRNUP AT BOL i
O LOW LEAKAGE i
d o VARIATION IN FEED BATCH SIZE 4
d l
- . . _ . - .= .. . . . -- . . . .
l
. (
P i
t t
4 t
5 3
CHARACTERISTICS OF LONGER FUEL CYCLES i
i l (' e Higher LBP Loading !
i e Higher Soluble Boron Concentrations 1 i
l I
1 j
l 1
I i
!f i
I l
- - - - - - , . . - .-. ,--_ _ , . - . - . , , . , . , , - , , - - ..,._ -_,, - _,_--n _--,_.,--, .-- ,n,-, ., n-..r--
4
( l l
l l
1 i
MTC VALUES FOR ATWS MTC 95% or CYCLE DURATION (10-4 DELTA-K/K/ O p)
(
BAW-1610 -1.05 (GENERIC ANALYSIS)
Avo. or 18-MONTH
-1.10 i OPERATING CYCLES '
OPERATING 24-MONTH -0.43 CYCLE I
/
(
V TC Vs. Burnup for '8-N onth Cycle O
e l
a.
?~
n'
'S 5 .
1 g.
.L h
t c i 2
5O e*.
o u
i a 2
o m
o (9
?~
1
(
?- , , ,
o ico zoo soo 4;, 3f, ,,,
Cycl. Lif etime. EFFD
MTC Vs. Burnup for 24-Month Cycle
(.
O r
l
's 3
S i-9
( ~
5.0,-
o a
O 1 ,-
1 i
h
. e I
4
\
7 i i i
i i i
(. )00 200 300 400 SCO 600 1
Cycl
- t,,,,i... EFPD
l MTCs for 18- and 24-V ont, Cyc es l O.
?-
r I
- u. _ . -
9m 18 24 1 /
r <
\
g-
$ ----_-___._____.\_..________________________
ATWS PRESS < SERVICE LVL C
=*-
- i ,(
,Y 69%* 60%*
2 '
8 O
E ;-
l
- c. .
?- i l
I * % OF TIME IN CYCLE < SERVICE LEVEL C l
? , , , , ,
o 100 200 300 400 soo soo Cycle Lifetime. EFPD l
l 2
(
l 1
CONCLUSIONS l
l MANY CHANGES HAVE BEEN MADE TO REDUCE THE I PROBABILITY OF ATWS.
ADDITIONAL CHANGES WILL BE MADE TO REDUCE THE PROBABILITY FURTHER.
1 1
- CURRENT DESIGN OF DSS EXCEEDS THE 1
OBJECTIVES OF THE ATWS RULE.
', CURRENT DESIGN OF AMSAC EXCEEDS THE 1
l OBJECTIVES OF THE ATWS RULE. !
NO PLANT OR FUEL CHANGES HAVE BEEN MADE OR ARE EXPECTED WHICH WILL INVALIDATE THE i ATWS RULE BASIS.
I i
' I
! l i l i
l !
l i
l TESTINGHOUSE OWNERS GROUP / WESTINGHOUSE PROGRAM ATWS RULE ADMINISTRATION !
1 l
l
- . COMBINED CORE PERFORMANCE / SCRAM l SYSTEMS RELIABILITY l 4 :
ACRS SUBCOMMITTEE MEETING !
I FEBRUARY 19, 1988 l l
I (
'l i
o
)
i 4
OUTLINE
- INTRODUCTION / BACKGROUND 4
e FUEL MANAGEMENT ASPECTS } l
- APPLICATION OF ATWS RULE / BASES TO W PWRS
! e WOG/W ATWS RULE ADMINISTRATION PROGRAM l
e CONCLUSIONS i
j
'I e l i
l l
i
- e l
l
\
'j i
i BACKGROUND e NRC INFORMATION REQUEST TO ALL PYR OWNERS GROUP - RINE 12, 1987
[ PROVIDE MTC INFORMATION TO DEMONSTRATE ATTS RIILE/ BASES CQ i NRC OBSERVATION:
RE!AAD CORE DESIGN TRENDS COULD POTEhTIALLY <
. IJtAD TO MODERATOR TEMPERATURE COEFFICIEhTS j (MTC) INCONSISTENT TITH ATWS RULE / BASES l
- l. I l
3
\
.- a
.I '
i r
I BACKGROUND
- BSTINGHOUSE OWNERS GROUP RESPONSE:
OCTOBER,1987 NRC/WOG MEETING TO DISCUSS REQUEST NOVEMBER,1987 YOG IEITER TO NRC e WOG/YESTINGHOUSE PROGRAM EVOLUTION:
l NOVEMBER,1987 INITIAL REVIEW l
DECEMBER,1987 REASSESSMENT TO ADDRESS ADMINISTRATION JANUARY, 1988 ENDORSEMENT - WOG ANALYSIS SUBCOMMITTEE FEBRUARY,1988 PROGRAM AUTHORIZATION - WOG MEMBERSHIP
- NRC/WOG/YESTINGHOUSE BRIEFING MENTING - FEBRUARY 11, 1988
)
l l
l t
l l d
OBJECTIVES l
- SUMMARIZE CURREhT FUE MANAGEMEhT :
- REATIONSHIP 0F CURRET FUE MANAGEMDT TO ATWS RULE AND BASES e OVERVIEW YOG/TESTINGHOUSE ATWS RUI2 ADMINISTRATION PROGRAM ;
CONCLUSIONS f
( e ON THE AVERAGE, INSIGNIFICAhT INCREASE IN MTC
- CURREhT FUE MANAGEMEdT C0hTINUES TO BE CONSISTEhi TITH CONCLUSIONS OF THE ATNS RULE
- e YOG/ WESTINGHOUSE PROGRAM ESTABLISBS SYSTEMATIC ADMINISTRATION BASIS ALSO COMPATIBLE WITH SEVERE ACCIDEhT POLICY APPROACH l
I I- e 10G AND YESTINGHOUSE BEID'E ACTIVITY TO BE WORTHTHILE I
TO ALLOY UTIIJTIES TO ADMINISTRATITEY ADDRESS BASIS OF ATWS RULE
(
j l
_g i
j .
! l
- r
- . CORE DESIGN PROCESS ;
i :
l
- CORE DESIGN IMCORPORATES MULTI-PARAMETER CONSIDERATIONS l l TO ADDRESS SAF3TY, OPERATION, AND ECONOMICS l
) !
4 DESIGN REQUIREMEhTS: l 4 :
- l CYCLE ENERGY OUTPUT
.l
- TECHNICAL SPECIFICATION AND SAFETY IAUTS i i
- FUEL MANAGEMENT CONSIDERATIONS !
l !
1 I
DESIGN PROCESS: ;
! I j_ e FEED REGION / BURNABLE ABSORBER SPECIFICATION l i e SAFETY EVALUATION (10CFR50.59,10CFR50.92)
J
!.
- OPERATIONAL DATA i
i i
1 1
I i
l
'l l
I L
i
! I i
FUIL MANAGEMIhT !
i
! l i :
I IMPLEMEhTATION OF EFFIC1EhT FUE MANAGEMEhT ED OPRATIONAL l 4
i
! STRATEGIES PROVIDES SIGNIFICANT BENEFITS TO 10G UTILITIES: l l !
- i I !
I l
J e LONGE OPEATING CYCIES l
i i .
- t
- 60 PRCENT RE0 ADS ON 18 M0hTH CYC12S i
l i
)
- OPTIMIZE FUE CYCLE PERFORMANCE l i 1
) - IFFICIEhT FUE AND BURNABLE ABSORBER ASSEMBLY DESIGNS
! - !AY LEAEAGE IAADING PATTENS (PRESSURIED TERMAL SH0CE/
! NEUTRON ECONOMY)
{- - INCREASED DISCHARGE BURNUP (FUE UTHEATION)
! - PMTC TECHNICAL SPECIFICATIONS (OPERATIONAL FIEmm1TY/
ECONOMICS) 4
! (
7 I
Average Westinghouse U.S. Plants 1
^ 5- ITC: BOL, HZP u_
o
$ MTC: BOL, HFP, ARO, EqXe N O- _, #) '
E w m ***
_._m_ . . ; -:A, . . _~.-.......# w e _.,, .
%K ~~~.= mer..._, . o.
O_
v -
c
~
e 5 ~ y ,~. ,
, -I
.__~~~I
___ -I
, -g o
O
- 1 ,_. -
O t
3 4
l 0 u -z,o-
! e -X- ITC MEAS -O ITC FPED -I- MTC PRED o_
C
@ n H c5 i i ; i i 1983 354 1955 1956 1987 Cycle Startup
,~
O Average Westinghouse U.S. Plants
^
u_ o-MTC: BOL, HFP, ARO, EqXe a
F N
E o O_
v C 'B
.e
- ',s,- N~,N ,r'g s ,.~ ,,
.9 _10_ , -
____8 ~~..
W .-
o .. -
c) *
e ,.. ,.. -
N .. . . . . . . . . X' u
3 .-
o ,.
t G $,- .
~
O_
E -I- 2 LOOP C-}- 3 LOOP M 4 LOOP 9- AVERAGE i ,
53 SJ 35 S5 87 82 Cycle Startup i
i
- , - - _ _ . ~ . . _ _ _ - - . _ _ _ _ . _ - - _ _ _ _ - . _ _ _ _ _ _ - _ _ _ _ _ - - _ _ _ _ _ _ - _ _ - - _ . . _ - _ _.
i r
r J !
I
.........@.Q...................................- l
\
00 0 -
l O l O O -
g EO c .
O -D
.w ;
C O O -$ :
W W lw:
l t
L y E -
e LaJ t e6 . ...{........,....
. O CE o 3 4 . . . . . .
...........0- e- -
a G a I
~
d a m La.
Z i O -
~
3 0 -
O _f -
_'-2
.E: o O
_y D
v1 Q m g ..
c- O O ils W b O -
M E -
G O -
We ..........................g............................................... _
m I'
E Inst FTTifTn m H
I l l l 1 I I O O O O O C O C e l c e N N ,
l l l l )
(3 6e0 LUOd) {UO[O!jj000 OJn[DJOdt.UO.L
l ATWS RULE BASIS l
(SECY-83-293 BASES, JULY 1983 (REGULATORY ANALYSIS OF ALTERNATIVES) ,
d l
0 RULE BASIS: ACCEPTABLE LEVEL OF RISK AND RISK REDUCTION i
0 PWR RISK TARGET: 1.0x10-5fgy . l 0 COMPARISON OF ALTERNATIVES IMPACT
- i
/YR WESTINGHOUSE GENERIC OPTIONS P
ATWS (MILLIONS) V/l* {
O. BASE CASE 3.7x10-5 .. __ ;
1A. DIVERSE AUxlLIARY FEEDWATER 5.8x10-6 $2.8M 3.3 INITIATION AND TURBINE TRIP (UTILITY PROPOSAL) !
1B. DIVERSE SCRAM SYSTEM 5.3x10-6 $2.8M 3.4 1 2. DIVERSE SCRAM SYSTEM, ADDED 2x10-6 $1.0M 1.1 f
{ TO UTILITY PROPOSAL CE/BSW GENERIC OPTIONS f
- 0. BASE CASE 8x10-5 __ ..
l
- 1. UTILITY PROPOSAL 2.2x10-5 $5.5M 3.2
- 2. SAFETY VALVES OR MODIFYlNG 7.2x10-6 $10.0M 0.44 CORE, ADDED TO UTILITY PROPOSAL
{
r I
- DONE ON AN INCREMENTAL BASIS. ;
O NRc VALUE/ IMPACT STUDIES TO ASSESS PROPOSED EQUIPMENT J
ALTERNATIVES TO ACHIEVE APPROPRIATE LEVEL OF RISK REDUCTION
(
k t
J
. t WestinJhouse A,w.5A.C
- Diverse Initiation of AN an:! Turbint Trip Isolation Tranittats (Non-Turbina Trip Transients) i AW3 Unacceptable R73 RPS NTC Over-eressure Re11sht11ty El, Centeevences reb. l
_AT h Mech l
CK i
CX
\
.01 l 0.9 CD 1.1x10*7
- l
'U 4 l 1x10 5 CD 1.1x10 l
1 0.1 .
CD 1.2x10 4 ',
2x4 c1. 2/7,y i
CX .
l
~
.C1 0.9 CD 2.1x10*7 1x10 5 ,g3) q ,
CD 2.2x10 l
- 4 CD 2.4x10 4
Total = 4.cx10 I
3 t' SECY-83-293 - JUIX 1983 i
- l
- NRC EVENT TRH M AMSAU
- - EVENT TRH TOP EVEh7S i - - RPS ELECT - FAILURE OF SENSORS, BREAURS OR IDGIC i
- RPS MECH - PHYSICAL INTERFERENCE OR BINDING PREVENTING C0FTROL ROD INSERTION
- MTC OVERPRESSURE - FRACTION OF TIME FOR UNFAVORABLE l MODERATOR TEMPERATURE COEFFICIENT l
(TWO VALUES ASSUMED M AND MOUT j TURBINE TRIP) j
- 1 - EQUATES COM DAMAGE M OVERPRESSURE
- l 1 - AFWS RELIABILITY - AUXILLARY FEDWATB UNAVAIIABI1JTY, j GIVEN AMSAC ACTUATION i
- - HPI - HIGH PRESSURE INJECTION OF BORATED YATER (AUT0EATIC ACTUATION ASSUMED) l I~ - FMQUENCY OF COM MELT l' - MTHOUT AMSAC
- 3.75-5/RY l - MTH AMSAC: 5.8E-6/RY 4
l I
It i
J l
]
. f TREATMENT OF MODERATOR TEMPERATURE COEFFICIENTS IN ATWS RUIZ i
l
- NRC RISK ASSESSMEhT l "MTC OVERP h sdJRE" !
- EVENT TREE NODE:
)
- FRACTION OF TIME FOR UNFAVORABLE REACTBTIT FEEDBAC:'
(REACTBTIT FEEDBACK RESULTING IN > 3200 PSIG)
- UNFAVORABLE IIPOSURE TIME (UET) ASSUMPTION l l
1 PERCENT UET, TURBINE TRIP f 10 PERCENT UET, NON-TURBINE TRIP l l
f
- SIMPIRIED APPROACH HIGHLIGHTS DOMINANCE OF MTC l OVERPRESSURE NODE AS A CONTRIBUTOR TO ATWS RISE l
l 1
8 I
(
FINAL ATWS RULE (100FR50.62): PWR DESIGN
- AD0PTED U""dIIT PETITION GROUP PROPOSAL FOR ALL TYPES i
'l 1
m
l ;
.i 1 CONCLUSIONS e PRESENT FUEL MANAGEMENT IS ENVELOPED BY THE ATWS RULE BASES e SECY-83-293; ATWS RULE BASES PROVIDES THE BASIS TO ESTABLISH ADMINISTRATION PROCESS FOR FUTURE CHANGES l
e DEVELOPMENT OF A PROCESS TO EVALUATE CHANGES IS FEASIBLE AND WORTHunlLE l
11 l
WOG/YESTINGHOUSE PROGRAM OVERVIEW 1
I OBJECTIVES
- DOCUMENT APPIlCATION OF ATNS RULE / BASES TO YESTINGHOUSE PYR FEATURES ;
i e DEFINE GENERIC PROCESS FOR IJCENSEE USE
(
APPR0ACH
- DEVELOP MORE DETAILED ATWS RISK MODEL - BASE IJNE
- DEVELOP IMPLEMENTATION GUIDEllNES TO EVALUATE CHANGES SCHEDULE
- PROGRAM INITIAllZATION: FEBRUARY,1988
- PROGRAM COMPLETION: MAY,1988
- WOG/YESTINGHOUSE NRC INFORMATION MEETING: MAY,1988 F .
12
l l
l ATNS RISK CONTRIBUTION METHOD
)
APPLICABILITY: ;
ALL T PWR DESIGNS l l
APPROACH: l 1
DEVELOP MORE DETAILED RISK MODEL TO DETERMINE IMPACT ON ATNS BASIS l ASSESS RELEASE POTENTIAL TO ASSURE COMPATIBIllTY TITH ATNS RULE AND SEVERE ACCIDENT POLICY APPROACH FOR ADDRESSING BEYOND DESIGN BASIS EVENTS
.)
l 13
I i
l l
RISE MODEL GUIDEUNES: f r
SEVERE ACCIDENT POUCY; CORE MELT FREQUENCY GUIDEUNE; < 1.0 E-4 /RY i
SECY-83-293: ATES CONTRIBUTION < 10 PERCENT OF "0TAL CORE MELT l FREQUENCY; 1.0 E-5 / 3Y f l
i
.- ATWS RISE MODEL CAN BE USED TO EVALUATE THE EFFECTS OF CORE /NSSS CHANGES ON CORE MELT FREQUENCY PROGRAM PRODUCTS _
e BASEUNE RISK MODEL DEFINITION e RISE MODEL SENSITIVITY STUDIES e IMPLEMENTATION GUIDELIh%
(
14
(
l RISK MODE GUIDEMNES:
l l
SEVERE ACCIDENT POUCY; CORE MET FREQUENCY GUIDELINE; < 1.0 E-4 /3Y SECY-83-293: ATNS CONTRIBUTION < 10 PERCENT OF TOTAL CORE MELT FREQUENCY; 1.0 E-5 / RY
.- ATNS RISK MODE CAN BE USED TO EVALUATE THE EFFECTS OF CORE /NSSS CHANGES ON CORE MELT FREQUENCY PROGRAM PRODUCTS
- BASEINE RISK MODEL DEFINITION
- RISE MODEL SENsmvai STUDIES
~
- IMPLEMENTATION GUIDEIhTS
(
14
l !
ATUS RISK MODEL / BASE IJNE MODEL CONSIDERATIONS !
I I
l i
e CURRENT FREQUENCY / DISTRIBUTION OF l INITIATING EVENTS
- AVAILABILITY OF REACTOR TRIP l
- OPERATOR MANUAL SCRAM / MANUAL !
CONTROL R0D INSERTION
- l AVAILABILITY OF MAIN FEEDWATER i
- AVAILABILITY OF AMSAC i e AVAILABILITY OF AUXILIARY FEEDWATER
- AVAILABILITY OF LONG TERM SHUTDOWN
-
- AVAILABILITY 0F CONTAINMENT HEAT REMOVAL
- AVAILABILITY OF PRESSURE RELIEF TO MAINTAIN PEAK PRESSURE <3200 USING EXISTING ANALYSIS /
EVALUATION BASIS
- CORE REACTIVITY FEEDBACK CHARACTERISTICS .
l
- a
i LTS REC CHR
' FWA MRI AR hr PR BF ET RT PL 1 SUCCES 81188881188818 18818881188888811118888118tt111188118188888888818188888888
( !!!!!!!!!!!!!!!!!
1 tittttttttltittititttitilittllllligt!!!Illittillt 2 $UCCES I i i tittlltillfttttti 3 $UCCES i ! i tillittitttttttti I Ittillllittittititittttti t t tiltilitt 4 TLC
! I t I ttillttil I 118888182 5 TL 11881188888811888 I t
I 6 TEC t tiltilitt t stttillittllittitittititt1888888!!
t 7ft ;
t t ' tillttllttttttiltilltttti S $UCCES littillt I litittllttitilltittiliitt i tillitill 9 ATEC t i I :
t 88811:18888888t!I t 188881188 10 ATE t t
I tilttllllliititillitillilitttitti I i I titt!!!!! 11 ATEC ,
I t I >
I I tillttttltilitillitillittlltiltlittttttti!!!!!!!!! !! ATE I t tiltilititttttillittisti: 13 SUCCES '
I i I t liitltittttttttitillitttilllitttttiltillitttltilli litttttt: 14 ATEC 1 I t tilitittilitittti tiltillit I 181188888 15 ATE 8 8 t I ltttitillitillittiltlitt 16$U*CES I I ;
till!!!!!It!!!!t!
t t I I I !!!Illitt 17 A!!C i I
i t i liitttttttillttti Ist Ittit iga!l t I !!!!!!!!!
I I t i '
/
I i 181111111 19 ATEC 1
tillittti 111111!!! tiltttitttttttttttttttttill!!!st!tillstill 20AIE I I I ,
I I I 181118881 .*; CIC l I I I !
t littllitt littlltlltilltilettlltilllitittttttttttt!!!!!Ittt! :: Ai[ '
! I i I I I 318881881 llA*IC I I I I I Illtillitttttttillttltistittt!!!!!!!!!!!!!!!!!!!!3:1883111 ;4 *!
! I I t Illtillillitttttttttttitt 25(U;CES l t i I ittlittitttittitt f 1
!!!!!!!!! t tillittil 26AIEC l I I tiliittittttiti!!
8 !!!!!!!ll stillitti 27 A!E i i ! 8 t I I tittllill 29AIEC t liittttti tillttttttilltttttttttttiltitttt!stIIIIIII ;9Ali 8 8 i t
- I I 181881888 30A!!C I
Ittlitist siit t ttit t t tit t t t t t t t titillt t t t tit t t tis t!
! 118:11:18 3:A!!
I I .
Ittitt II 30A!!C t il t illit ti t i tilli t t llt i t t ll t tit t tt t t t t titilla t t IIIIIttti 7:A?!
CATE6CRYt!!CR!PilCN EVENT EVEri NAMI
- Ri AEACICRTRIP TL TRAM 1. Lait IE6RAC. Il0 CC PL POWER LEVEL ) 401 ftAEi. EH LT REERAD. glin CC FuA FEEIsATER AV!;LAILE TEC it itasi. eat f vl0 CC MRI Rat X 10 Ik!!AflCN ATEC Atul eat.LT tiltA; g!TH CC AR AMSAC ATE Aful EARLT MIRAD U10 CC AF AUI. FEED #ATER M PRESSURE RELIEF BF ICITT 1 FET)
Lil LGh4 TERM $CC;aN RfC RECIRCULAi!CN CNR CChiAihMENTHIAIREAC/AL
1 i
i BASEllNE/ SENSITIVITIES OBJECTIVE: DEFINE ACCEPTABLE ENVELOPES OF MODEL ASSUMPTIONS YHICH ARE COMPATIBLE WITH THE ATWS RULE BASES
- BASIS FOR GENERIC ASSESSMENT - REFERENCE PLANT ;
NSSS CONDITIONS l 4 LOOP PLANT CONFIGURATION MODEL 51 STEAM GENERATOR l l
CONSERVATIVE PLANT PERFORMANCE CHARACTERISTICS l l
CORE PHYSICS CHARACTERISTICS
- SENSITIVITIES TO RISK MODEL ASSUMPTIONS
)
a
(
CURRENT FREQUENCY / DISTRIBUTION OF INITIA?dG EVENTS TREND: TOTAL NUMBER OF INITIATING EVENTS SIGNIFICANTLY REDUCED OVERPRESSURIZATION INITIATING EVENTS SIGNIFICANTLY REDUCED t.
REASON: 10G TRIP REDUCTION AND ASSESSMEhT PROGRAM 20 ATTS IMPALT: REDUCTION IN ATWS RISK DIRECTLY PROPORTIONAL TO REDUCTION IN 1REQUENCY S
N
)
I i
i l
CONCLUSIONS l i
l i
i
- ON THE AVERAGE, INSIGNIFICANT INCREASE IN MTC f
e CURRENT FUEL MANAGEMENT CONTINUES TO BE CONSISTENT l WITH CONCLUSIONS OF THE ATWS RULE !
1 !
.
- Y0G/YESTINGHOUSE PROGRAM ESTABLISHES SYSTEMATIC ADMINISTRATION BASIS ALSO COMPATIBLE YITH THE l 1
, SEVERE ACCIDENT POLICY APPROACH i
)
- Y0G AND YESTINGHOUSE BEIJEVE At;ny1Ti TO BE !
YORTHTHII2 TO ALIhY UTILITIES TO ADMINISTRATIVELY ADDRESS BASIS OF ATWS RUIR 16
i.
COMBUSTION ENGINEERING OWNERS GROUP PRESENTATION TO THE ADVISORY COMMITTEE ON REACTOR SAFEGAURDS .
CORE PERFORMANCE / SCRAM SYSTEMS RELIABILITY SUBCOMMITTEES CONCERNING
( CURRENT TRENDS IN MODERATOR TEMPERATURE C0FFICIENTS l
1 J. TED ENOS (CE0G, AP&L) l JOHN F. KAPINOS (C-E)
FEBRUARY 19, 1988 i
1
(
l l
1
.=
PRESENTATION CONTENI O CURRENT TRENDS IN MTC 0 IMPACT ON 1979 ATWS ANALYSIS 0 POSITIVE REACTIVITY EVENTS 0 RESPONSE TO JUNE 12, 1987 NRR LETTER CONCLUSIONS 0 NO OVERALL TREND TOWARDS POSITIVE MTC 0 1979 ATWS ANALYSIS RESULTS ARE CONSERVATIVE NRC BASIS FOR ATWS RULE NOT CHANGED 0 SAFETY ANALYSIS CONSIDERS POSITIVE MTC
~
0 FAVORABLE NRC RESPONSE AT JANUARY 11, 1988 MEETING 1
INTRODUCTION 3
0 NRC JUNE 12, 1987 INFORMATION REQUEST ON MTCs 0 CE0G / NRC TELEPHONE CONVERSATIONS ;
i i
0 C-E REVIEWS MTC DATA AND ATWS IMPACT FOR CE0G l CURRENT TRENDS IN MTC EVALUATION OF ATWS MTC MTC IMPACT ON ATWS ANALYSIS !
1 0 CE0G / NRC MEETING - JANUARY, 1988 l
a.-. - - -
e- ,m- - , .. -. -- -, - , , -
- s. :
. CURRENT TRENDS IN MODERATOR TEMPERATURE COEFFICIENTS t
O REVIEW 0F PAST, CURRENT AND FUTURE CYCLES i
0 OPERATIONAL DATA P
h 0 "BIASED" ANALYTICAL EVALUATIONS
(. !
0 HOT ZER0 POWER AND HOT FULL POWER i
O DATA BASE SORTED ON VARIOUS FIELDS 9
Y i
l i l I
a -
l O :
H l < 0 g !
4 4 00 O
l
'Ra
<O O
~
< q O O W < 6 O O O p <
< <a
< < 0' 0
O 00 O O 6 C ,q O< g O E m ; < 0 m ' < 0 a- ! _
W << 0 O 4 i ,:-
~
N (f)
O:
OO gy <
O 0
O O O C
<3 0 m
=-
j < O ~
( < 0 bL n <
t O
N n l < O 0
k O ^- l11
< 0 ^
x N:
~
(w; '
n:
)
9 )
w 4
- i
- Z_'
M -
i f
I s
- ., e 1 i
$ k o 0 2 -e
= -
- < c i _
i
=
I
. t.
p I
=
_= --
=
' = = c o o a =
N
.y 2 . .
v .
t3 CO O i W i i i i i q_ .' .
l
, , ,MTC ,Cx10E,4 drho/dT) ,
O O O O O O O O l O co o s N O N s o
~
C O O O O O O O O l . . < i . . 4
}
l > O l
I N
- M L
T Z Z i,
2 -
c' n samme .
, c y dE 7
C C>
. C _. O O O
'=
>> t>
0 li C,
. o Q
O0 9, O C> C> C>
O O 0, >
,f-o C, C
O t>
c O v o >t.> > 1 n
'I O OO O
CCXO O C>
C> C> b [> t>
Iii T 0,
0 y > g e--
eN - O -
3
.-S O C>
w< m m -
3 -
H
~
- O l !
l l
l l
O W
L d
@ 4 0
o ou ss e
< <' b o zr .
4 d o o E
l
<i o 6 h& l O 0 <
4 g
o o
o 4
I I
- 1 l
W t 'i e e o 8 !
pC -
& o 6 i
0 v
g "a Q < o O_ L E
o 9 o
S b u!J < o i C.
J
< o i
O L J < o 3 sa
.O .__
< o em
.s ,
W# L < o - o k, E Q >O a
i b c U
O Z i !-
O <
?
Z " ,y, w
(
5 ! , , , , , , ,
>=== o o o o o o o o o o o 9 m 9 3 9 9 9 m 9 o 0 o o o o o o o a o _
C1P/04JP PSOlX) 1!w!$ oe'ds yoel ' '
I O .
H E .,
l
- T
- =
O ,, < r i
'\
,' \
/
. . E o f
~) ?
,' y
.4 / <% *
- f **
. . j,'
- e s I l
- t
/
< g ~~
M - 'A g -
! T
\ \ \
~
G,
\
w a %
y
} *
~
O
,/ y g w~- !
' #s be p*
$ q,. . #
s .
N. s , ~, *
\\ ! 's - . =
1 _ g i
s
\. \- s s, -
- ~
- r '
f,
. \ .
~
O , _ .
. m E
s, l
em M
4 Z
emmesme N
O t
m 6
N. _
WU g e ,D ,)
m ^"*
g -um Z .
e518 m.
4 H _
g 4
e.m g
i .w w .% -
l . . . . . . l l
m W W manuar W *nMF W _
6
( e m. e . == = * , 8
- --s-
, , , T.. <4 l
l
m , - - - - - , --- - - - - - - , - - - - - - - - - - - - - - - - - - - -
2 i
. I
~
-e. .
a e h 0
t / l
/ /
/ /
/ ,/ f
. + l M
.~
e, g
g
/s ~
I _-
/
i l
i.
~~
/ l
\
i
==== / 's 4 44 -
- l \; - ,
~
p \ - i O _w; ,.
s ;
! / \ g
, _=
, i f {\ -
, / \ \
\ \ ~
' =
( _-
^
\N j
\
-\
\
g
~ ~ ~
- l i l g
3
- \ 1
~
A \ '
\
1 O- - '
4 '
i
\
g g
\
g -.
j' \ '
t 4 ~
I e
Z~ e-om me>
alip t .
N
'- l-- _
.- =
=
=
~ . - , . -
e e e t 6 e p
= eum - N eu. . .
W e e w & W w w p = A *
- g
- w w we we <me e one W * ,,, ,p 4
l I
~
i -
I r '"
r$ - -
- i i
.g m
=
I.
s
/
i C '
s l
l' I \
\
i g ~ ,
g
/ -
J \
s O
\ .
(=~ j \ d \ C - C -
d \ s N 4 g \ O-g
\ \
~
\ \ l ::"
- i
. g j
\ _
\
\j \ i 1
b_
g ,.
\
g g
m 1 _
\ % ,
\
O _A _
g
\
\
\
\
l l _
g % --
~'
_ s 1 ~
. f \
a %
\
g n s -
sammum _
\ g L
~
4 Z_ -
W
\
E \ -:
( H -
~- O
~
~
\
,"% .: P
_ = - w -.s .w _ .
i e e e e a e e e .
w
,5 w
yw.w m
_ _ s V *""" *""
i i
- i i m s.nm . m
- s R
% m f _$
CONCLUSIONS FOR MTC TREND ANALYSIS i
0 NO OVERALL TREND TOWARD MORE POSITIVE CYCLE MAXIMUM MTC AT FULL POWER i
0 POSITIVE MTC TECH SPEC REQUEST NOT A NEW ,
ISSUE
! O NO STRONG CORRELATION TO CYCLE LENGTH 0 DESIGN ASPECTS USED TO CONTROL MTC AT FULL
. POWER ,
BURNABLE POIS0NS i ENRICHMENT l NUMBER OF BURNABLE POIS0N RODS CONTROL RODS ,
t
. O CONCERNS ABOUT POSITIVE MTC TREND ARE NOT VALID FOR CE0G UTILITIES l I
i i
F IMPACT OF MTC TREND ON 1979 ATWS ANALYSIS O REVIEW 0F OPERATIONAL DATA O BASIS FOR ATWS MTC VALUE 0 RESULTS OF C-E ANALYSIS O IMPACT OF CURRENT MTC ON ATWS ANALYSIS O CONCLUSIONS ,
)
l i
i
1 1
REVIEW 0F OPERATIONAL DATA !
0 DATA REVIEWED BY PLANT CLASS ;
RELOAD ANALYSIS REPORTS '
J PLANT PHYCS DATA .
i CORE FOLLOW REPORTS l t
i I
0 SUPPLEMENTED BY BEST ESTIMATE CALCULATIONS .
l j
l i
1 J a g i ;
-~~
e
> .- O "
CD G
l
_a ,
c0 l i
_) I O _C' C
Qo a O9 1
- g
' i w 'J ;
O-( -
,nj,O
]O U l
i
' Ca O
r C C
O .Em'l --
a E
o a_C o i g3
- ~3 l
__ - _i . L __ t _. ! _.1 ___._... _ __ i '
f.
g 4
N O N 7 O CO O N vO CO O N 7 O CO O O O Oi O O . -
Ni N N N: N i i i i
(.!.9 'L v.J P F 3 0 ! x ) 3J
.c 4
BASIS FOR 1988 ATWS MTC l
i 0 FULL POWER INITIAL CONDITIONS i
0 95% OF TIME SPENT CRITICAL
' 0 ACTUAL PLANT DATA 4
1 i
i i
l k
l i
i 4
(
ATWS ANALYSIS MTC
- PLANT CLASS 1979 VALUE 1988 VALUE 2750 MWT (14X14) -0,20 -0.26 3410 MWT -0.63* -0.50 i !
( !
3800 MWT -0.68* -0,57
- ESTIMATED WITH LIMITED OPERATIONAL DATA FROM PREVIOUS :
DESIGNS l i
. I l
(
l
ATWS ANALYSIS IMPACT ;
CHANGE IN PEAK PRESSURE ,
PLANT CENPD-263 ATWS RULE' MTC"" NET i CLASS (1979) (1982) (1988) CHANGE (1979-1988) 2750 MWT 4220 PSIA -258 PSIA -50 PSIA -308 PSIA l (14 X 14) l l
l l l
3410 MWT' 4290 PSIA -347 PSIA +210 PSIA l -137 PSIA l
3800 MWT 3800 PSIA -882 PSIA +200 PSIA l -682 PSIA i
INCLUDES TURBINE TRIP
- ESTIMATED
(
I
l CONCLUSIONS OF ATWS IMPACT EVALUATION 0 CURRENT ATWS MTC 2750 MWT CLASS SLIGHTLY MORE NEGATIVE 3410 MWT AND 3800 MWT CLASSES SLIGHTLY LESS NEGATIVE
, O CHANGE IN ATWS PEAK PRESSURE SINCE 1979 i- -
OVERALL DECREASE l'e ATWS PEAK PRESSURES 0 CE0G UTILITIES RESPONDING TO PREVENTION AND MITIGATION REQUIREMENTS OF ATWS RULE O
~ -
4 441 a - 4 e
POSITIVE REACTIVITY EVENTS O SAFETY ANALYSES CONSIDER POSITIVE MTC TECH SPEC VALUE 0 ANALYSES YIELD ACCEPTABLE CONSEQUENCES p
l'
( ANALYSIS UNCERTAINTIES 0 UNCERTAINTIES DOCUMENTED IN CENPD-266-P ROCS AND DIT COMPUTER CCOES t :
l
RESPONSE TO JUNE 12,1987 NRR LETTER 0 CE0G AND NRC STAFF MEETING, JANUARY, 1988 O FORMAL SUBMITTAL 0F TECHNICAL MATERIAL TO !
STAFF, FEBRUARY, 1988 .
I i
i i
i 4
h I
i f
i .-
l
- i. j i
t
( !
a j i.
I f'
b
, e . _ , _ _ _ _ _ _ - _ . - ._,_ .-. _ . . _ , , _ _ _ _ _., _ ., _- _ ,
e a
SUMMARY
/ CONCLUSIONS 0 N0 TREND TOWARDS MORE POSITIVE MTC 0 CURRENT ATWS EVALUATIONS INDICATE LOWER PRESSURES THAN 1979 ANALYSIS 0 LEVEL D STRESS LIMITS FOR ATWS NOT EXCEEDED FOR 1988 MTC VALUES
( 0 1979 ATWS EVALUATIONS STILL VALID 0 CE0G UTILITIES RESPONDING TO PREVENTION AND MITIGATION REQUIREMENTS OF ATWS RULE 0 SAFETY ANALYSES INDICATE ACCEPTABLE RESULTS FOR CURRENT MTCs
- 0 CE0G HAS RESPONDED TO NRR LETTER
(
e .
b J
l h i
i Re ATWS MODERATOR TDPERATl'PE COEFFICID,TS STAFF PRESOTTATION
, TO l _
ACRS 00PSINED COPE hiniMNCE AND SCRAM
(
SYSTEPS RELIABil.lTY SLEW.lTTFS l FEBPtJt.RY 19,1908 j
n
! I l
4
.1 .
i l
l- l 1
'l I
(.
1 .
1 I
i i
.j ACRS AGENDA ITEM 3A POTDTIAL CmCERNS WlE CUPPENT REACTIV11Y OPF.PATIONAL LIMITS ,
(CHANGES IN OPERATING Ort!DITIONS OR OTHER INFOPf% TION ON ,
' i VAllDITY (T EARLIER ATWS ANALYSES)
CHANrfS RELATED T0:
EXTDED CYCL.ES OF 18 AND 24 MMHS LOJ LEAKAGE COPE DESIO:S ;
INCPEASED DISCHARGE FAIPNUP
- NEW FLEL DEFIGJS
- P0DEPATOP TBPERATURE CnEFFICIEhT (MIC)
TEOfilCAL SPECIFICATION OWE REnlFSTS i
I~ LED TO STAFF CONCEPH WAT PWR MTCs MAY PE IKMSISTENT WITH ANALYSIS BASIS PRECEDING DFM10PPENT OF ATVS PRE ,
- i. :
l STAFF C m CERN LED TO JUNE 12, 1987 l
LETTEP TO PWR M GROUPS
( '
l l
l
t OBJECTIVES OF JLNE 12,1987 LETTFR:
(MERS CRUPS SHOULD ASSESS MTCs l' SING CURRENT DATA AND REPORT TO T E STAFF:
- JUSTIFICATION FDP THE CONTINUED APPL.irMILITY OR CONSERVATi m IN A M BASIS MTCs PPMIDE RESULTS FOR LMOVE OR OtrR.lER PLANTS
(
- DIFFEPENCES IN CUPRENT MTCs WilH ATWS BASIS MCs
- PLAM DATA USED
- KTFnD0 LOGY USED TO DEPlVE AM MCs
.-
- ASSitFTIONS PADE
Is 4
ACPS AGENDA ITB1 Ir.
POTENTIAL CWCERMS WITH 0.'RRENT REACTIVITY OPEPATIONAL LIMITS (POT &TIAL FOR LNANALY7ED PEACTIVITY ACCIDENTS /INCireffS ,
IKLUDING OPEPITIfWAL EXPERIENCE OR INCIDRTS OF CONCERN)
( STAFF IS NOT AWARE OF AhY OEPAT!ft'AL EXPEPIENCE OR INCIDRITS REGARDING LMFALYlED REACTIVITY ACCIDENTS THAT ARE OF CONTPN IN LWRS l HOWEVEP, RES HAS P90GP P AT BNL TO ASSFSS PREVIOUS JUDG9 9 TS ON THE ADE00ACY OF REACTIVITY ACCitBT ANALYSES THAT F0PP THE i
BASIS FDP DESIGN APPRNALS I
i
(
6
(
ACPS ArfNDA TTEM Ic 4
POTENTIAL. CfhCERNS WITH OJPRBG REACTIV!TY OPERATIONAL LIMITS (ACCURACY / UNCERTAINTY OF PEACTIVITY ANALYSES (FOR EXAPPLE, MC))
ISnTFER*AL TUPEPATUPE C0 EFFICIENT (!TC)
EASUP997 AT HOT, ZER0 POWER (PZP) 22x10-5 gyjgf.F (MS 19.6.1 STANDAPD) i FP m V0GTLE FSAR (PAGE 4.3-50) ACCURACY OF OPPSR ANAL'GICAL ETTODS IS:
1 02% 4R FOR DnPPLER DFFECT 22x10-53 gfgf.F FOR FTC 50 PPP FOR CRITICAL BOPnN CnNCOURATION r
_3% FOR POWER DfSTRIPUTIONS
- f.0.2% Ax/K FOR POD BANK WRTH l
.ia]O-54 K/K/ STEP FOR DIFFEFtiTIAL POD WRTH
~
f.0.5x10-54 K/K!PfH FOR BORON WPTF 0
2.1%ox/K FOR PU)ERATOP DEFECT I
( '
ACRS AGBIDA iT B la POTENTIAL CONCERNS WITH CUPPfhT EACTIVITY OPERATIONAL LIMITS (PEACTIV!TY OPERATIONAL LIMITS VEPSUS CORE POWER LEVEL.)
MTC TECHNICAL SPECIFICATIONS HAVE PECTE MPE R1SITIVE FOR SME PLANTS, FOR EXAFPLE, TE0iNICAL SPECIFICATIOPS MAY STATE:
+5x)n-5g gfgf.F FROM ZERO TO M PGEP.
PAMP 0F +5x]D-541K/K/*F FPOP 707, POWER TO ZER04K/K/*F AT 1007, PMER 7 TECHNICAL SPECIFICATION OMNGES ARE FOR:
OPERATIONAL FLEXIBILITY FSAR 0%PTER 15 ANALYSIS ASStPPTION ,
?
- In EVER, IF 10 m POWER ATWS MTC HAS NOT BEC0 E mRE POSITIVE, TIB' LOWER P09 ATYS MTC YlLL :
NOT BF. CUE MPE POSITIVE (AS FISTORTED BY ACTUAL MTC DATA) l
(
s ACRS AG90A ITEM 2
. g STATUS OF ESP 0tGES TO JttE 12,1987 STAFF LETTER 3
(ACTI0fG PUMED (IF At!Y))
STAFF EETitlGS WITH OWNERS GROUPS 7
- W/WOG OCTOBER 7, 1987 FEBRUARY 11,1988 .
- E/E0G JANUARY 11,1988
- B&W/BWDG FEBRUARY 18,1988
( !
P E/E0G KETING PROVIDED STAFF WITH FIRST OWERS t
GROUP ESP 0f6E CONCERNillG PRESENT MTC DATA AND RESULTS c
W/WDG FIRST EETING WAS EXPLORATORY ON EQUEST FOR l ifFORPATION RAISED BY STAFF LETTER 3
. W/WDG SEC0f0 EETING PROVIDED STAFF WITH MTC DATA AND ATWS RULE ADMIfilSTPATION PROGRAM ;
ECW/BWOG EETillG PPOVIDED STAFF WITH MTC DATA t
i s
0
E f
a 3
l3 ATWS MTCs (10-5 g gfgf.p) i i
A1KS ANALYSIS BASIS CURRBU VALUES W -
8 -10 (STAFF ESTIMATE) i CE (275 W ) -
2 -2,6 '
- r (341 % ) - 6,3 -5,0 l (. (380 W ) -
6.8 -5.7 l B&W (177FA) -
10.5 -11.0 (18 MONTH CYCLE) ;
i
-4,3 (24 K WT}i CYCLE)
! +
1 1*
.i i
J J
(.
t
i a L
t i
4 STtFF CONCLUSifW BASED ON INF0PMATIOP PROVIDcD BY OWtO S GROUPS, STAFF CONCLUDES THAT QJPPFNT FTC DATA IS CftSISTFF WITH PREV 10!S ATks MTC AFALYSIS
( BASIS FOR ALL PLANT TYPES, ;
STAFF 00NCLUS10N WILL BE STATED IN AN EVALUATION, j P
e C