ML20138R318
| ML20138R318 | |
| Person / Time | |
|---|---|
| Issue date: | 11/07/1985 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1468, NUDOCS 8511180540 | |
| Download: ML20138R318 (150) | |
Text
ORJGINAL o .UlN11ED STATES NUCLEAR REGULATORY COMMISSION IN THE MATTER OF: DOCKET NO:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 307TH GENERAL MEETING
- i
"- ;pg b Oih O
LOCATION: WASHINGTON, D. C. PAGES: 1- 115 DATE: THURSDAY, NOVEMBER 7, 1985 l
., e~. (T\",i
.-c?f? ,a,, -. n -- a h\@hk t" q u m i) i \i[$5'. s C
]h kc,m S~q QQB(}
DO OISBb0ACE-FEDERAL REPORTERS, INC.
OfficialReporters 444 North Ca itol Street Washington, C. 20001 (202)347-3700 6511180540 051107 NATIONWIDE COVERACE PDR ACRS T-1468 PDR
CR25017.0 1 DAV/dnw I UNITED STATES OF AMERICA
( )
'~'
2 NUCLEAR REGULATORY COMMISSION 3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4 307TH GENERAL MEETING 5
Nuclear Regulatory Commission 6 Room 1046 1717 H Street, N.W.
7 Washington, D. C.
8 Thursday, November 7,19 85 9
The committee met at 8:45 a.m., Mr. Jesse C. Eb ersole ,-
10 chairman, presiding.
11 ;
I 12 PRESENT. ,
[h L/ 13 MR. JESSE C. EBERSOLE l
l MR. DAVID A. WARD :
14 DR. ROBERT C. AXTMANN il DR. MAX W. CARBON 15 h DR. WILLIAM KERR
'i DR. HAROLD W. LEWIS 16 ! DR. CARSON MARK I
DR. FORREST J. REMICK '
19 DR. PAUL G. SHEWMON l DR. CHESTER P. SIESS l 20 MR. CHARLES J. WYLIE l
'21 22 23 24 has-F Reporters, Inc.
25
( )' PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSIONERS' ;>
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS THURSDAY, NOVEMBER 7, 1985 The contents of this stenographic transcript.of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards (ACRS), as reported herein,,is an uncorrected record of the discussions recorded at the meeting held on the above date.
No member of the ACRS Staff and no participant at
() this meeting accepts any responsibility for errors or inaccuracies of statement or data contained in this transcript.
I l
o ..'
0170 01 01 2 (m,)DAVbur 1 PROCEEDINGS 2 MR. EBERSOLE: I call on Mr. Jerry Haynes, of 3 Arizona Power, to do this from 9:10 to 9:40. This is the 4 condensed version of the entire day's meeting.
5 MR. HAYNES: Good morning. My name is Jerry 6 Haynes. I am Vice President of Nuclear Production for 7 Arizona Public Service Corporation.
8 The purpose of my presentation is to give you a 9 brief overview of the power ascension testing experience on 10 Palo Verde Unit 1.
11 I will then turn over the podium to Mr. Robert 12 Butler, Director of Electrical Services for Arizona Public
!'l s/ 13 Service, and he will discuss the auxiliary pressurizer spray -
14 enhancements and the modifications to enhance offsite power l 15 reliability.
16 The first part of my presentation will deal with 17 the power ascension test program, specifically the schedule l 18 ascension test programs.
l 19 (Slide.)
20 This slide shows various power levels, shown on 21 the left side of the slide, as a function of time.
22 Green represents the schedule we eventually laid 23 out for the power ascension test program. That was based on 24 experience at other CE plants, in consideration of the test
) 25 program at Palo Verde and the test programs at other OE ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 66M
0170 01 02 3 s
I
_j DAVbur 1 plants.
2 For Unit 2 we have taken into consideration the 3 actual schedule we achieved at that point and the reduction 4 in testing.
5 As you can see, early in the test program we had 6 20 percent power and performed the testing we did, quickly 7 moved ahead of our planned program and went to 50 percent 8 power testing.
9 At near the completion of 50 percent power 10 testing, we discovered a deficiency in the post-accident 11 sampling system. We committed to correct that deficiency 12 before we came back to power.
() 13 We then came back and had a short amount of test 2
14 at 50 percent power, which completed that phase of the 15 program, then went to 80 percent power.
16 Two other significant outages shown here were the 17 event that raised questions about their auxiliary 18 pressurizer system. We will talk about that.
I 19 The other is the event associated with loss of 20 offsite power. We are also going to discuss the 21 improvements that we have made there.
22 In summary, even despite six weeks of unplanned 23 outages, we have been able to maintain our schedule pretty 24 well. At the present time we finished, with the exception
() 25 of one test, the 80 percent power testing, and we will ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33&M46 l
E 0170 01 03 4 o
(,)DAVbur 1 expect to run the remaining test next week and then proceed 2 to 100 percent power.
3 As far as experience goes -- .
. 4 (Slide.)
5 -- I already described how we developed the test 6 program. The results of the testing have shown that 7 particularly in the physics area the predictions -- the 1 8 results agree very well with the predictions.
9 We have had nothing there out of the ordinary in 10 the plant's transient area, with the exception of two tests 11 which resulted in trips. The transients have not interfered 12 with activities, and the plant has performed very well.
<a k-) 13 One indication of that is that during the now 11 14 months of testing we have only experienced seven trips.
15 Only two of those were actually during the conduct of the 16 test. I mentioned those in the previous slide, and we will 17 talk about these later.
18 Two of the trips were associated with loss of the 19 main feedwater pump. One of those was the result of the 20 need for adjustment of recirculation flow, loss of suction 21 pressure of the feed pump, and loss of the pump.
22 The other was associated with, also, a low 23 positive suction pressure due to a startup strainer that was 24 left in the system.
((MJ 25 DR. MOELLER: Excuse me. I don't follow. You ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverase 800 336 4646
0170 01 04 5 DAVbur 1 are saying five trips occurred.
f~)/
q, -
2 Are you saying without the reactor being 3 critical?
4 MR. HAYNES: No, we had seven trips in total.
5 Two of those were during transient testing. The others were 6 not associated with transient testing.
7 So it was in effect an unsuccessful transient 8 tect.
9 DR. MOELLER: Thank you.
10 MR. HAYNES: One of the trips was associated with 11 a CEAC circuit board failure.
12 Three of the trips were associated with loss of
( )) 13 power. One of those was during testing. It was a load 14 rejection test. One was an actual loss of power event not 15 associated with testing. The other was an actual loss of 16 power event during troubleshooting. That trip actually 17 wasn't what I would normally classify as a trip. The 18 reactor was subcritical, and we were not even in the startup i 19 phase. The rods were withdrawn, but the circuitry was 20 actually working. The rods did drop, so we classified that l
21 as a trip, also.
22 The last trip which was associated with testing 23 was due to a false steam generator level signal subsequent 24 to a test.
() 25 The other thing I would like to mention is that ACE-FEDERAL REPORTERS, INC.
I 202 347 3700 Nationwide Coversee 800 3 % # 46
0170 01 05 6 f')DAVbur
\_/
1 in no case have we had an unplanned trip due to operator 2 error. There is no operator error and no maintenance 3 error.
4 So we -- also, as I am sure you have 5 recognized -- had seven trips during the initial power 6 ascension testing on the first unit. It is certainly not a 7 large number of trips. It is more than we would like to 8 see, but it is lower, substantially lower than most people 9 have achieved.
10 Any questions?
11 DR. REMICK: A question on your two main feed 12 pump trips:
(m,) 13 Were those cases where one pump tripped or both 14 pumps tripped?
15 -MR. HAYNES: In both cases, as I recall, we were 16 esely in power. It was early in the test program, and we 17 only had one pump in service. We lost that pump.
18 DR. REMICK: You used the word " trip" here.
19 Are all seven of those unanticipated events? Is 20 that how you are using it? They were not intentional trips?
21 MR. HAYNES: Those were unplanned.
22 Are there any questions?
'e 3 (No response.)
'N_
I MR. HAYNES: I will turn the program over to Bob n
(_) 25 Butler, Director of Electrical Services.
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336 4646
0170 01 06 7
( )DAVbur 1 MR. BUTLER: Good morning. My name is Robert 2 Butler, Director of Electric Services for the Arizona 3 Nuclear Power Project.
4 The purpose of my presentation this morning is, 5 at the request of the subcommittee, to present a brief 6 description of the documentation of the plan of the Palo 7 Verde auxiliary pressurizer spray system.
8 By way of background, the modifications -- the 9 experience we have had in the power ascension test program 10 on Unit 1, specifically on September the 12th, and we had a 11 load rejection test at 50 percent power.
12 The plant experienced a loss of charging flow.
[) 13 This was the result of depletion of the inventory of the 14 volume control tank. It was caused by inaccurate level 15 indications to the operators in the control room.
16 As a result, we took short-term measures, 17 including increasing the surveillance on the VCT level 18 instruments plus administrative and procedural controls.
19 In the long term, we planned these modifications 20 to address the test experience. These modifications were 21 developed by our engineers and, in addition, discussions 22 with the NRR staff.
23 (Slide.)
24 As the viewgraph indicates, this is a summary of (nj 25 the modifications to the plant.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
u>
l '
0 1 7 0 0 1 __ 0 7'.- 8 DAVbur. l' The first is to add a second diverse reference j_ 2 leg to the volume control level transmitters. This, we 3 . fee'1, addresses the root cause of the September 12th 4 problem during the test.
'5 The second is to provide power'to the VCT outlet 6 and refuel water tanks' gravity feed line from a lE motor 7 control center. -The purpose of'this is to eliminate the-8 'need for the operator to take action outside the control 9 room.
10 MR. EBERSOLE: May I comment at this point?
11 You are.providing power to those valves; however, 12 they remain not qualified. Am I correct?
l:
13 MR. BUTLER: That is correct.
14 MR.-EBERSOLE: So I want the committee to observe
'15 that.
t l 16 MR. BUTLER: The third modification is to add l 17 automatic actuation of the transfer from the VCT supply to t
l 18 the refuel water supply. This automatic action will i 19 eliminate operator action in the event of a loss of offsite 20 power.
.- 21 The fourth modification is to lock open two i
I 22 normally open valves in the aux spray flow path, one being l
j 23 on the suction line from the refueling water tank, the other l 24 being on the discharge of the charging pumps at the
-O 25 e neatnmene. ;
I ACE FEDERAL REPORTERS, INC.
202 347 3700 Nadonwide Coverage 800 336 6646
_ . _ , . _ _ . - _ ~ - - , _ . - , . , _ . _ _ _ . __ _ _ _ _, _ . . _ _ _ . _ _ - _ _ . _ _ _ - - . - . - -
'0170 01 08 9 3DAVbur 1 The purpose of this is to reduce the potential f/
x_
2 for isolation of the gravity feed or charging lines by 3 spurious actuation or operator error.
4 MR. EBERSOLE: Mr. Butler, do you intend to show 5 a diagram of this system to the full committee here?
6 MR. BUTLER: At your pleasure.
7 MR. EBERSOLE: Let me very briefly comment on 8 what this is. What they have done is use the volume control 9 tank as a source of water for auxiliary spray, that tank 10 being the one which is used primarily for deboration, to go 11 into power.
12 They thus take suction pumps for two purposes:
(~)
V 13 for the usually charging pump function of keeping inventory 14 up, and this time the same charging pumps are used for the 15 function of spraying water into the pressurizer when the 16 tank is of small volume.
17 And you have heard the discussion about improving 18 the level check on it.-
19 I want to just call out the fact that the 20 reciprocating charging pumps, of which there are three, now 21 have the dual function of maintaining inventory and the 22 auxiliary function, which is quite important for providing 23 spray to depressurize.
24 There is no special pump for depressurization.
() 25 They have taken the edge-up function out of the charging ACE-FEDERAL REPORTERS, INC.
20247-3700 Nationwide Coverage 800 336 6646
0170 01.09 10
,\
D DAVbur 1 pumps.
2 I suppose you could argue that you could have 3 . dedicated pumps to do this spray with considerably more 4 ability.
5 Am I correct? You have used the charging pumps 6 as simply a convenient source of spray?
7 MR. BUTLER: Right.
8 MR. EBERSOLE: With the attendant disadvantages 9 of the small tank, of not having an auxiliary suction line 10 in the event that line fails.
11 I think the committee would like to see the 12 diagram.
( 13 Yes, sir.
14 MR. REED: I would like to ask a question of 15 Mr. Butler.
16 Recipient charging pumps are good. They have a 17 long history of not being so good, but I still think they 18 have reached pretty good performance, and of course these 19 will be used all the time to maintain pressurizer-levels.
20 But you have some tech spec commitment one way or 21 the other to availability. I am sure you are not committing 22 all three to be operable all the time when you are in power 23 operation. So you are either down at one or two operable.
24 In your tech spec commitments, do you remember 25 what you are?
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee 800-33H646
0170~01-10 11
- )DAVbur 1 MR. BUTLER
- I believe it is never less than 2 two.
3 MR. REED: Never less than two. So you have 4 spray of never less than two. You could have one out for
-5 maintenance. They generally do have a high maintenance 6 requirement.
7 Are these Ferry drives or what kind of drives?
8 What is the maker?
9 MR. QUAN: This is Terry Quan, Arizona Public 10 Service.
11 The maker or the. manufacturer of the pump is 12 Galling.
13 MR. REED: A Ferry drive or what?
!( ) ,
14 MR. QUAN: I am not sure.
15 MR. REED: I guess you could visualize that in an 16' instance you might actually be down to one pump.
17 Does that bother you actually? One pump is 18 satisfactory for your auxiliary spray?
- 19 MR. QUAN
- Right. The analyses which we require 20 only use one pump.
i l 21 MR. MICHELSON: One other small matter that was ,
t l 22 discussed during the subcommittee meeting and should be 23 reemphasized, that although this auxiliary spray function is 24 an essential function, at least as I view it, only a portion
() 25 of the system being used for that purpose is seismically f
i l ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
0170 01 11 12 c,
.( jDAVbur 1 qualified.
2 MR. BUTLER: The system in its entirety is 3 designed for seismic load.
4 MR. MICHELSON: Then I got misinformatipn. I 5 asked about the transfer arrangement, for instance, between 6 the valves. You said that was not a seismically qualified 7 transfer.
8 MR. BUTLER: In the mechanical pump it is not.
9 MR. MICHELSON: The pressure boundary. Do I 10 understand that the pressure boundary is seismically 11 qualified, but not the control?
12 MR. BUTLER: That is correct.
(q_) 13 MR. MICHELSON: And of course the control is 14 perhaps more susceptible to seismic disturbance than is the 15 pressure boundary?
16 MR. BUTLER: I have the diagram up.
17 Is there anything you would like?
18 (Slide.)
19 MR. EBERSOLE: I would like the committee to ask 2.0 any questions about the diagram. I would just like you to 21 notice the dual suction, the primary suction, the chosen 22 suction, the amount of the value control tank rather than 23 the refueling water storage tank. However, the head suction 24 is from the refueling water storage tank.
!( ) 25 MR. REED: It is also important to point out that ACE-FEDERAL REPORTERS, INC.
N2 347 3700 Nationwide Coverage 800-3364646
0170 01 12 13 DAVbur 1 in this transfer what they have concluded is that the volume 2 control tank, which is hydrogenated water, is not really the 3 best source of supply in view of its boron concentration 4 changes and feed and so on, and it is hydrogen. It is not 5 the best source of supply for the auxiliary spray when you 6 have an incident.
7 So what they have concluded is that they want to 8 make a transfer, and I believe they have made it automatic 9 now where a 501 valve will close and a 536 valve will open, 10 putting the suction supply to the refueling water storage 11 tank, which is borated, nonhydrogenated water.
12 These valves are not parallel, and although the 13 electrical supplies are safety grade, they are not parallel 14 valves.
15 MR. EBERSOLE: What do you mean?
16 MR. REED: They are not series parallel 17 arrangements for single component failure.
18 MR. EBERSOLE: As a matter of fact, if the one 19 under the right tank stays open, that suction will override 20 and defeat the suction you are trying to encourage, is that 21 correct?
22 So the system is not safety grade.
23 You will hear later about another adjunct system, 24 which is a different set of valves, that presumably m
) 25 overcomes the nonsafety aspects of this design.
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 33(H5646
0170 01 13 14 f'jDAVbur 1 By the way, are the charging pumps always safety
.w 2 grade?
3 MR. BUTLER: Yes, they are.
4, MR. EBERSOLE: Are they used for purposes other 5 than just system makeup? There's other functions for it 6 which are semi-safety grade?
7 MR. REED: You said you assumed one pump operable 8 with this 44 gallons of fluid. But seals take 20 gallons of 9 fluid, right?
10 MR. QUAN: Correct.
11 This is Terry Quan again.
12 In our analyses we assumed that the seal water to
() 13 the reactor coolant pumps was being taken from the charging 14 pump discharge. So that is a total of 18 gallons per 15 minute.
16 When we looked at auxiliary spray, we assumed the 17 44 minus the 18.
18 MR. REED: You got down to the bare minimum for !
19 the spr y?
20 MR. EBERSOLE: So let me just say one more 21 thine,.
f 22 This aystem affords the most rapid form of 23 depressurization if it w3rks. If it doesn't work, then it l 24 isn't safety grade.
~
25 You will describe later the other system, the
(-) ,
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 4646
0170 01 14 15 F~TDAVbur 1 one-inch vents which you use, which is a longer time vent
.\ )
.2 process.
3 Thank you.
4 MR. WYLIE: I still have a question on 501 and 5 536.
6 Are those individual controls on each valve? Do 7 you control them individually?
8 MR. BUTLER: Yes.
9 MR. WYLIE: I am a little puzzled why you chose 10 to put them on the common power supply downstream.
11 Why did you do that?
12 Before you made that change, you had them on
() 13 individual circuits, completely independent except for the l 14 control center. They came from a common mode control l
15 center, and now you show them tied together.
. 16 I am a little puzzled as to why you did that.
! 17 MR. QUAN: As far as I recall, with the l
18 modification there is the same degree of independence as l 19 there was before -- as far as I recall.
20 MR. WYLIE Not necessarily, because if you go l
! 21 off the lE motor you have a common circuit and then you 22 branch out to two wells.
23 MR. EBERSOLE: When you use this system and you 24 have trouble with it, how long do you try to use it before
() 25 you invoke the other system which you are going to show us ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6M6
0170'01 15 16
.T'TDAVbur 1 later?' 'Do you struggle with this one a while before you go
\~)
2 and use the other, or do you just jump to this one right 3 away? ,
4 MR. BUTLER: I think that has to come from the 5 operator's sense of the situation at the time and what he 6 has got to work with. I believe the process that we did 7 assumed a couple of hours.
8 MR. ETHERINGTON: When we speak of 9 depressurization, do we mean only to system saturation 10 pressure?
11 MR. BUTLER: Yes, that is correct.
12 MR. WYLIE: I am not sure I got an answer. Maybe 13 you don't have one.
(])
14 MR. QUAN: I guess the diagram that we provided 15 is a simplified sketch and it is not a detailed wiring 16 diagram.
17 As far as I can recall, we aren't overly 18 compromising the independence of power to those two valves.
19 MR. WYLIE: Maybe it makes no difference, but the 20 way it shows, the diagram you have provided, is you have a 21 fault because of the tripping of the breakers. It will take l
22 out both valve supplies, and then you have no control over l
23 either of the valves.
24 MR. QUAN: That is correct. We would lose that O 2s 1E meeor cenero1 center.
ACE FEDERAL REPORTERS, INC.
202 347 3700 Naionwide Covernoe 800-))6 4646
~
0170 01 16 17 T' DAVbur 1 MR. WYLIE: No, just that one branch circuit.
'V 2 You have got two circuit breakers in series on a common 3 circuit.
4 MR. HAYNES: If you have a slide, I think it is 5 worth putting up because that is a simplified slide.
6 7
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 l
I ACE. FEDERAL REPORTERS, INC.
202 347 3700 Netionwide Coversee 800 336 6646
0170 02 01 18 1 (Slide.)
f')DAVbur
.G' 2 MR. WYLIE: On the left is what you have 3 previously from a non-lE motor control center. There's two 4 separate circuits, two separate controls. Over here you are 5 tied together, and both of the controls are on the common 6 circuit and then branch out.
7 It would seem to me you compromise the 8 independence of the two circuits.
9 MR. QUAN: As far as I can recall, a fault in 10 either one of these, the breakers are sized such that we 11 would open this breaker before we would lose power to this 12 branch right here.
13 Also, if we were to lose power to these valves,
(])
14 there would be ample time, adequate time, in which to take 15 manual action to properly align these valves.
16 MR. WYLIE: So you don't need the electrical 17 controls?
18 MR. QUAN: We would prefer to maintain control 19 with the rollers. If we were to lose power of these valves, 20 we believe there would be ample time to take operator action 21 to properly align those valves.
22 MR. EBERSOLE: Could you comment on the potential 23 for spurious depressurization and your realization that it 24 is happening? Do you have flow or other indication that
() 25 spray flow is occurring, or do you just have the ultimate ACE. FEDERAL REPORTERS, INC.
202 347.)?00 Nationwide Coverase 800.))6 4M6
=0170 02 02 19 DAVbur 1 ~ effects of spray?
2 In short, if I notice them getting depressurized, 3 do I know whether or not it is due to the spray being 4 inadvertently open, or do I simply have to take like a 5 blanket analysis that I am being depressurized for whatever 6 reason?
7 You know, when you have a nonsafety 8 depressurization system, I think you have to be a little l 9' concerned about inadvertent spray, the other side of the 10 spectrum.
l 11 MR. HAYNES: Perhaps I can respond to that. j i i 12 There is a flow indication at the charging pump discharge.
13 MR. EBERSOLE:
(]) But it is going to be doing other 14 things besides spraying, won't it?
15 MR. HAYNES: That is correct. During normal 16 operation, the discharge of course would not be to the spray 17 system.
18 MR. EBERSOLE: You wouldn't take the common flow 19 as being indicative of just spray, flow, would you? Do you 20 have spray flow?
21 MR. QUAN: We don't have specifically indication 22 of spray flow. We have now position indication within the 23 control room.
24 MR. EBERSOLE: Just that the valve is open?
() 25 MR. QUAN: Right.
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Cowrees 800 336 4M6
0170 02 03 20 1 MR. EBERSOLE: Is that direct that the valve is i()DAVbur 2 open or just the relay, a secondary indication?
3 MR. QUAN: I think it is more the latter 4- indication.
5 MR. EBERSOLE: Of the ste'a position?
6 MR. QUAN: Right.
7 MR. WARD: I didn't get that answer. It is an 8 . indication of the stem position on the valve or of the 9 demand signal on the valve?
10 MR. QUAN: I think it is the stem position of the 11 valve stem. We use the NAVCO limits.
12 MR. WARD: Does that satisfy you?
() 13 MR. EBERSOLE: I would have preferred indication 14 of spray flow.
15 MR. WARD: Charlie, did you get an answer to your 16 question or did you finish that?
17 MR. WYLIE: Well, as I understand it, they are 18 not really relying on the electrical operation of these 19 valves. They are saying they can do it manually.
20 MR. EBERSOLE: Really what is coming out of this, 21 I think, is the realization that although the spray system 22 has been improved here, in the long run it still has some 23 questionable aspects, and I am sure we will get to the other 24 system shortly that backs this up.
() 25 MR. WARD: Well, wait a minute. Is the answer l ACE FEDERAL REPORTERS, INC.
] 202 347 3700 Nationw6de Coversee 900 344M4
0170 02 04 21 f'lDAVbur v
1 here that mechanically the pressure boundary is fully safety 2 grader although the control system is not, you do have the 3 ability to operate the system manually from the field as a 4 backup?
5 MR. BUTLER: That is correct.
6 MR. MICHELSON: It is not clear the operator can 7 be fast enough in the unlikely event you get an inadvertant 8 retransfer back to the boron tank when it is already empty 9 and it starts sucking hydrogen immediately. It is not clear 10 then that the operator can run down and close the tank to 11 prevent gas binding, to give an example. That takes a 12 spurious operation.
() 13 MR. WARD: Is that a reversible situation?
14 'MR. MICHELSON: That is something you have to I 15 look at the circuitry to see what the credibility is of 16 getting inadvertent retransfer.
- 17 MR. EBERSOLE
- Well, if you gasify the pumps that i
j 18 takes a while to clear out, and that can happen really 19 quick.
20 MR. MICHELSON: It was claimed, I believe, that 21 they had perhaps time. They are providing some better 22 vents, as I understand it, and so forth. It took them quite 23 a few hours the last time. Now, I think they have better 24 vents, as I understand it.
() 25 MR. HAYNES: May I respond to this line of ACE. FEDERAL REPORTERS. INC.
202447 3700 Nasionwide Coversee 8004446d6
0170 02 05 22 1 questioning in general?
()DAVbur 2 We did experience a problem with the charging 3 pumps due to a faulty indication of the level control tank.
4 We corrected that level indication, so the operator has a 5 much better sense of the level in the volume control tank.
6 That is one point.
7 The second point is we have improved all of our 8 procedures such that if he does get a low level, as 9 indicated either by the level or by charging flow, there are 10 precautionary statements that give him some direction as to 11 what to do. .-
12 So in the event you have a seismic event and the n
(_) 13 system does transfer back to the volume control tank, it is 14 already empty, the operator has some direction.
15 In this particular case where we did experience 16 gas binding, the operator had no direction, had not 17 , experienced that previously, yet was able under the 18 circumstances that existed -- the loss of power incident --
19 with safety injection initiation to'not have a steam 20 generator tube failure. He was able to reestablish charging 21 flow within an hour and a half.
22 our analysis did not give the auxiliary 23 pressurizer spray until two hours into the incident, and 24 with that assumption in the analysis they are still well
() 25 within the radiological offsite limits.
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nauonwide Coversee 800336464
0170 02 06 23
.DAVbur 1 MR. EBERSOLE: It should be noted, I think, that 2 what is being done here is he is riding up on the limit for 3 direct malfunctions of the system. Should they have trouble 4 with this system, he rides out to the limit and then gets it 5 fixed within the limiting time.
6 MR. REED: I think it is important, Jesse, as you 7 said, to hear about the other system. The other system has 8 similarities in common mode failure potential. Keep in 9 mind, there is common mode failure potential here because 10 they are just single valves.
11 MR. EBERSOLE: This is a dual track at least, but 12 one that is orificed.
, 13 , MR. MICHELSON: And of course, if there is no 14 mechanical damage to the pumps in the process of 15 l gasifying. Although it didn't have mechanical damage last 16 time, it is not clear that that would be the same situation 17 next time.
18 MR. HAYNES: I agree with you, although in the 19 actual incident the operator, given no direction, tried to 20 start all three charging pumps at various times and more 21 than one time, yet before returning to service we 22 ASME-tested all three charging pumps. All three charging 23 pumps showed no evidence of gasifying. Flow was as good as 24 it had boon previously.
25 So they do have some resistance to damage from
(~}
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverase MG33HM4
0170 02 07 24 DAVbur 1 gasifying.
2 MR. WYLIE: Let me ask you if your indication in 3 your volume control tank is -- I don't know what point you 4 set it at. That is where you do you transfer.
5 Do you have some link you call for that to 6 transfer?
7 And those circuits do trip out, and you do not 8 close 501 and you do not open 536.
9 Do you have sufficient time then to go out and 10 realign those valves?
11 MR. HAYNES: Yes, we do. We have got two hours 12 from initiation of the incident.
- 13 MR. EBERSOLE
- We have mentioned that those pumps l
14 l have low suction trips. So that protects them from that i
15 l kind of thing.
16 If there is no further question on this diagram, 17 i I think we should proceed to the backup system, which is a l
18 ' dual track, small system that providos a very slow method of 19 depressurization, further challenging this limiting 20 dischargo from the steam generator tubo failuros.
21 (Slido.)
22 MR. BUTLER: I would point out that this is not a 23 modification. This is appropriated in the plant as it is 24 now, and it is a fully safety grado system.
25 MR. EDERSOLE: May I commont on that comment?
(~'
ACE FEDERAL REPORTERS, INC.
202 347.)?00 Nationwide Coverage MX)-))6 /M4
0170 02 08 25
~NDAVbur 1 What this is is a system that came about as a (O
2 result of TMI-2 to get rid of noncondensables. I think it 3 was found useful here, as was the charging system, to make 4 it perform an adjunct function of discharging vapor, but at 5 a much slower rate than the other system, en aspect of 6 depressurizing.
7 So it serves the original TMI-2, a noncondensable 8 venting function and now is being used for another 9 function by necessity of backing up the previous system that 10 we described.
11 Carry on.
12 If I am wrong, tell me about it, but that is the 13 way I see it. This was a noncondensable venting system. It (J~)
14 is now used, happily, as a backup for the vapor vent 15 process.
16 MR. REED: Jesse, you might be a little more cool 17 by saying that it is now a mini-PORV.
18 (Laughter.)
19 MR. EBERSOLE: I am not going to say that. I 20 don't regard it as such.
21 MR. BUTLER: What further discussion would you 22 like on the system? ,
23 MR. EBERSOLE: Is the system, as seen by the 24 committee here, do they understand that it is a one-inch
(~T 25 piping vent as though it were a PORV7 It is fully safety
\.)
ACE.FEDERA1. REPORTERS, INC.
202.)47 3700 Nadon*Ide Coverage En))MM4
0170 02 09 26 1 grade.
' ._ ' DAVbu r 2 It has no capacity for fluid flow in a bleed / feed 3 context. It is a slow way of discharging steam. It results 4 in a time to depressurize which is substantially extended 5 above and beyond the prior system.
6 I believe you have some curves that show those 7 relative depressurization rates.
8 MR. MICHELSON: Before you go to the curves, let 9 me ask a couple of questions which I didn't ask at the 10 subcommittee meeting.
11 Do you normally plan on venting to atmosphere or 12 ; to the reactor drain tank?
13 MR. BUTLER: The proferred route would be the 14 i reactor drain tank.
l 15 l MR. MICHELSON: How much of the heat load can the i
16 reactor drain tank handle?
17 r You talked about 10 to 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> deprosaurization 18 using this devico.
19 Can the reactor drain tank koop up with tho hoaL 20 I input from this vont for a 10-to-12-hour poriod?
21 MR. QUAN: I think you have to understand, also, 22 that the main prosaurizor safety valvos also dischargo into 23 this tank. Wo would bo dopressurizing over a longer period 24 of timo, doponding on our ability to quench and our ability
[l 25 to drain the tank. Wo may or may not rupture this.
Acu. FEDERAL REPORTERS, INC.
202 347.)m) Nadonokte Coverage kn)%#M4
0170 02 10 27 DAVbur 1 MR. MICHELSON: You think you are near 2 equilibrium with the heat input rate, then, but you are not 3 quite sure? ,
4 MR. QUAN: Right. We haven't taken that close of 5 a look.
6 MR. REED: This drain tank has a coolant in it, 7 or is it a fill and drain type of cooling tank?
8 MR. QUAN: More the latter.
9 MR. MICHELSON: Does it have a quench tower in 10 it?
11 MR. QUAN: No, it doesn't.
12 MR. MICHELSON: Just inventory. Then it has no
( 13 ; heat removal rig then?
14 MR. REED: Only by refilling with cold water and 15 i draining out.
I 16 j MR. MICilELSON: That is not very fast.
17 f But you say it is in equilibrium?
18 MR. QUAN: I think you havo to understand that wo 19 would bo draining to a tank through -- from our analysis wo 20 looked at using the orifico lino, which is fairly small.
21 This lino isn't as of much concern as far as tho tank being 22 ablo to handio that load.
23 MR. MICl!ELSON: It is a simplo quantion. I was 24 just wondoring, whon trying to doprosaurizo in 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />
^
( 'T 25 using this dovico you can go to the reactor drain tank and ace FEDERAL REPORTERS, INC,
- o:.m.nm m oon.a.co m .. mnwu
I 0170 02 11 28 l 1 you do not have to go to atmosphere.
{)DAVbur 2 Is that your answer?
3 MR. QUAN: Yes.
4 MR. MICHELSON: This is a safety grade system, 5 but it wasn't intended to be single failure; even though 6 they are parallel valves, there is of course only one valve '
7 to the reactor drain tank?
8 MR. QUAN: The intention is that it would be 9 single failure proof.
10 MR. MICHELSON: Not at least to the reactor drain l
11 tank?
12 MR. QUAN: That is correct, but we have two paths
() 13 or two places to which we could discharge. One would be the 14 reactor drain tank. ,
15 MR. MICHELSON: I am assuming one is on one train 16 of power. 106 is on one train and 105 is on another.
17 The problem I have is: how did you divide this 18 up relative to auxiliary spray? Because loss of power will 19 get auxiliary spray in the single failure points, like I 20 valves, for instance.
21 How did you divide the valve power up on that 22 system versus the valve power on this system?
23 I don't want to get into details, but I assume 24 you worked that out so that no single power failure would i
() 25 get both this system and the auxiliary spray.
ACE. FEDERAL REPORTERS, INC. l m.147 3700 Neuon.we covmee m))ume !
0170 02 12 29 T~~^DAVbur 1 MR. QUAN: That is correct. No single power V
2 failure will totally disable power to the valves and the 3 equipment required for aux spray and affeet both vent 4 paths.
5 MR. MICHELSON: My concern wasn't the vent 6 paths. You might have to go to atmosphere under this 7 circumstanco, is that right?
)
8 MR. QUAN: We may have to.
9 MR. EBERSOLE: I just wanted to note to the 10 committee this is a typical four-valve font, guarantooing 11 opening as well as closing.
12 liR. ETHERINGTON: That is a pretty small line.
( 13 j Why do you further restrict it with an MRS?
14 MR. REED: That is the TMI backfit.
i 15 MR. QUAN: Just if wo did have a failure of this 16 valvo, the inventory loss through this could be made up by 17 one charging pump.
18 MR. EDERSOLE: But you have a backup valve for 19 that?
20 MR. QUAN: With thoso throo closod, that is 21 correct.
22 MR. EDERSOLE: Couldn't you just got rid of the 23 orifico and invoko the two valvos in socios?
24 MR. QUAN I am not quito nuro. I don't think --
( ) 25 we really woron't looking at taking crodit for thono valvos ACE. FEDERAL REPORTERS. INC.
202 147 1700 Nothin*kle Coversee NG))M/4
1 0170 02 13 30 1 as being redundant isolation valves.
()DAVbur 2 MR. EBERSOLE: If you don't, you are subject to 3 an inadvertent blowdown by a single valve opening right 4 there, RC-103, unless you invoke those backup valves in the 5 vertical lines.
6 MR. HAYNES: Not shown on the drawing is a code 7 change in the piping.
8 MR. EBERSOLE: I see. There is a code change to 9 a lower grade pipe. So then you are subject to inadvertent 10 steam blowdown if you have a spurious opening of that, and 11 you run that commercial risk, whatever it is.
12 MR. QUAN: Also, another point that I would like
, () 13 to bring out as far as this system is concerned, the 14 analyses which we pdeformed assume depressurization that had 15 a discharge through the orifice. If we were to use the 1
16 three-quarter inch line, the nonorificed line, we would get 17 a depressurization capability that is comparable to the aux l
l 18 pressurizer spray system.
19 (4 R . EBERSOLE: Although there is a code change in 20 that one-inch line, that doesn't mean there is any reason to 21 believe it wouldn't hold 2250, or is there?
22 MR. HAYNES: I am not sure. But, no, there is no 23 reason.
24 MR. EBERSOLE: You have it closed presumably; it J
() 25 is at 2250 now?
ACE. FEDERAL REPORTERS, INC.
202 147 1 2 ) NationwWe Covetese M 1) H 444
0170 02 14 31
,7~]DAVbur 1 MR. HAYNES: Yes.
U 2 MR. EBERSOLE: Any comments on this systam?
3 So you note this is a physically simple miniatire 4 PORV design, with zero capacity for mass flow of any 5 substantial amount.
l 6 So the residual issue, the major issue about 7 bleed and feed is to whether all we have been talking about 8 is just the transitional operation of golt.g tron high te low 9 pressure, its rapidity, and efficiency, and se forth.
10 Carry on.
11 MR. BUTLER: Would you like to see the curves?
12 MR. EBERSOLE: I think maybe some words about the
(~)
V 13 ultimate problem, which is riding up on the dose as you 14 provide the discharge time and the statement that you can 15 afford the legal limit that we have on your radioactive 16 releases, that you can afford to ride up for quite long ;
17 periods of time.
18 (Slide.)
19 MR. BUTLER: This shows the two-hour and 20 eight-hour doses calculated for the aux spray.
21 The first analysis -- these are the three 22 analyses -- the first analysis is the FSAR, Appendix 15A, 23 with aux spray. That is at 17 minutes.
24 The next is with aux spray, but two hours.
(* 25 And then this is with the absence of aux spray l l l i ACE FEDERAL REPORTERS, INC. 202 347 1700 NadonwMe Coverage 800.))HM4
0170 02 15 32
. cs 1 and the pressurize event at two hours, showing the (J DAVbur 2 relatively insensitivity with regard to the dose curve 3 configuration.
4 MR. WARD: Let's see, Mr. Butler, those numbers 5 assume the pre-accident iodine spike. 6 Do you have any idea approximately what the 7 numbers would be without that assumption? 8 MR. BUTLER: Just a minute. We will try and find 9* it. 10 (Pause.) 11 12 0 12 14 15 16 17 18 19 20 1 ! 21 l 22 l 23 24 O 25 ace. FEDERAL REPORTERS, INC. ll02 347 3700 Neuon** Coversee 800136eM4
( -7 l~ l
- 0170.03 01 33 I f~SDAVbw 1 DR. MOELLER
- While he's looking for that, could G
2 we ask the Staff to refresh my memory. They're quoting here 3 the 10 CFR 100 dose limit to the thyroid to the 300 rom, and
> 4 for, I guess, plants at the CP stage. I realize that's not 5 the case here. You were using the 150, were you not? So 6 could you straighten me out? Does 300 apply in this case?
P' 7 MR. LICITRA: This is Manny Licitra of the NRC 8 _ Staff. You're correct that at the CP stage, we do consider 9 the 150 value as a limit just to account for any eventuality 10 of a step in the curve during the construction of the plant 11 at the OL stage. We go to the 300 rem. 12 DR. MOELLER: And to Chairman Ward's question, do () 13 you consider a pre-spike of iodine for the calculations? Is 14 that the standard approach? 15 MR. LICITRA: It's my understanding that that's 16 what the Staff requires and is considering; yes. 17 MR. EBERSOLE: May I ask kind of a clarification 18 question. 19 What is the limit allowed for a full LOCA? 20 MR. LICITRA: The limit for a LOCA? It's the 21 same thing. 22 MR. EBERSOLE: What I'm having trouble with. You 23 know two failures are going to occur. We know that LOCAs 24 are not going to occur, yet you have the same limit. I () 25 don't understand this. It seems to me, there would be some ACE. FEDERAL REPORTERS, INC. 202 147 3700 Nationwide Coverage 8(A )) H 646
r 0170 03 02 34 1 gradation in consideration of the relatively high
-(])DAVbw 2 probability of tube failure, almost a guaranteed case versus 3 a zero case of full LOCA. I can't see how you can use the 4 same number for systems so tremendously different than the 5 probability of occurrence.
6 MR. WARD: Wait now, Jesse. The LOCA is a 7 different situation. 8 MR. EBERSOLE: But they're using the same dose 9 limit. 10 MR. WARD: If you take the chapter -- well, I 11 don't know about that. You don't get numbers anything like 12 this for a large-brake LOCA within containment.
' ,)
q 13 MR. QUAN: This is Terry Quan from Arizona Public 14 Service again. These analyses that we did, these two 15 different rupture analyses assume one full break of a steam 16 generator tube with a loss of off-site power and a fully 17 stuck-open atmospheric dump valve throughout the event. 18 This first set of number is for the FSAR Appendix 15 A 19 analysis, which assumes aux pressurizer spray application of 20 approximately 17 minutes, even generated iodine spike, 21 preexisting iodine spike, for the two-hour and the 22 eight-hour dose. 23 DR. SHEWMON: Before you leave that, a 24 preexisting iodine spike is what was already in the (') 25 coolant. Is that one gap, or where was it preexisting? ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-336-6646
'0170 03 03 35 f'kDAVbw 1 MR. QUAN: That's right in the reactor coolant V
2 system. 3 DR. SHEWMON: In the water? 4 MR. QUAN: Yes. 5 MR. EBERSOLE: That's not a very high probability 6 statement up there, is it? 7 MR. QUAN: We don't believe so. 8 MR. WARD: But it's what the Chapter 15 analysis 9 requires; isn't that it? U 10 MR. QUAN: It's what the Staff requested, as far 11 as steam generator rupture analysis. 12 MR. EBERSOLE: A much more realistic dose is one ( )' 13 that would be generated by the event itself. Okay. 14 MR. QUAN: And as you can see in the lower 15 portion here, this is for aux' pressurizer spray, and the 16 numbers would be comparable for the pressurizer event, using 17 the orifice line. It's integrated iodine spike and 18 preexisting. And the numbers are fairly comparable. The 19 point to be made here is, there is a large delay in the 20 actuation of the aux pressurizer spray system. 1 21 MR. WARD: Thank you. 22 (Slide.) 23 MR. BUTLER: This is the curve we showed the 24 subcommittee to try and put in relative terms the operation () 25 of the aux-spray. This compares with the operation of the ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 3 %-6646
0170 03 04 - 36 DAVbw 1 pressurizer vent, and as you can see, starting at time zero, 2- the operator takes control of the plant, opens one ADV, 3 depressurizes below, the HPSI pumps will come in. The 4 accident analysis assumes the arbitrary two-hour delay. , In 5 the real situation, you could expect that that could be 6 faster.. The solid line is the case where we bring in the 7 auxiliary space system and the dotted line is bringing in 8 the lower or orifice leg of the pressurizer vent. 9 If we had used the larger of the two legs, I 10 believe the depressurization rate would be approximately 11 equivalent to that with aux spray.
'12 MR. EBERSOLE: Why do you assume aux spray's 1
(} 13 initiated two hours? There's no reason not to initiate it 14 right way; is there? You just arbitrarily assume? 15 MR. BUTLER: That's assumed to give you the time 16 for operator action. 17 MR. EBERSOLE: Two hours? 18 MR. BUTLER: You get your valves lined up. 19 MR. EBERSOLE: Is there any real physical reason 20 that he can't with it in two minutes? 21 MR. BUTLER: No. 22 MR. HAYNES: The reason we assume two hours is, 23 the delay of depressurization for two hours, maximizes the 24 two-hour off-site dose. (} 25 MR. EBERSOLE: You just did it to display the ACE-FEDERAL REPORTERS, INC. M2 347-3700 Nationwide Coverage 800-336-6646
)
0170 03 05 37 l (~3 DAVbw 1 the pessimistic case. You notice that the operation of the V 2 ADV doesn't depressurize the plant, except down to about 3 1500. The hot bubble hauls it up. 4 MR. WARD: Let's see, Jesse. 5 On our schedule now, do you still want to cover 6 the licensing presentation and the off-site power 7 reliability? We've only got five minutes for that. 8 MR. EBERSOLE: I think we can do that next. 9 MR. WARD: We've spent a good bit more time on 10 this. 11 MR. EBERSOLE: This is the guts of the meeting 12 here. (} 13 MR. REED: I'd just like to ask a question. 14 Maybe the reason that you hold up your' pressure for two 15 hours, high pressure injection functioning is to better 16 assure decay heat removal by natural circulation to the 17 steam generators; isn't that perhaps the case? 18 MR. BUTLER: Not in this analysis. It was for 19 the reason Mr. Haynes gave. 20 MR. REED: If you were to come down using 21 auxiliary spray, the auxiliary spray will bring you to the 22 saturation pressure. 23 Now the only way you get below saturation 24 pressure is to have natural circulation conduction for the 25 steam generators in to cool the primary mass itself. (]} ACE-FEDERAL REPORTERS, INC. 3)2 347 3700 Nationwide Coverage 800-336-6646
0170 03 06 38 1 I thought maybe you were holding up in order to {-)DAVbw 2 better assure natural circulation, because there are a lot 3 of questions, and a lot of people have these curves now that 4 the NRC has required. With respect to maintaining 5 overpressure from natural circulation, if you're going to go 6 the steam generator route, in th.is design now, you do not 7 have a bleed and feed route for decay heat removal. 8 MR. QUAN: This is Terry Quan again. It is my 9 understanding that the pressure hangs up, basically because 10 high pressure signal injection is making up the primary to 11 secondary leakage. What is happening here is, by use of 12 either the aux pressurizer spray or the pressurizer vent (^ ' . 13 system, that level in the pressurizer is being retained such t/ 14 that the operator has met his criterion in which to throttle 15 the HPSI flow, allowing depressurization of the system. 16 MR. EBERSOLE: Any further questions? 17 I would like to move to the next topic, which is 18 going to be very brief, I think. 19 MR. BUTLER: I think it can be very brief. 20 I would like to say in summary on this issue that 21 we believe that these modification fully address the 22 experience of the September 12th test results and that they 23 enhance the reliability of the system. 24 The next topic, I believe, covers the loss of
'~ ' ; 25 off-site power. I'll get to that very quickly. This had to ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336 6646
0170 03 07 39 DAVbw 1 do with our multiplexer systems, which had remote operation 2 and indication of the nonsafety equipment, including the 3 nonsafety 13.8 kV switchyard breakers. It is a 4 computer-based system utilizing fiber optics. We had loss 5 of power events on October 3rd and 7th. We believe this was 6 due to spurious signal from the plant multiplexer. 7 We corrected the problem by taking the 8 multiplexer out of that loop and hard wiring a modification 9 for the control of the 13th breaker. 10 MR. EBERSOLE: So the essence of this is, you 11 hard-wired it and got rid of the problem. Would you still 12 argue, if your multiplexer had worked, that you complied (' 13 with GDC 17, even though it's kind of a commonality in 14 electronics to have a very expensive power supply? 15 MR. QUAN: As far as meeting GDC, yes, we believe 16 that when the high multiplexer, that by design, the off-site 17 power sources did meet GDC 17. The high multiplexer, as far. 18 as I understand, is supposed to be fully independent, with 19 redundant computer controls, such that one failure or common 20 failure would not affect both sourcese of off-site power. 21 MR. EBERSOLE: Your experience has been
~
22 otherwise, though. 23 MR. QUAN: Our experience has been otherwise. 24 MR. BUTLER: What happened was not supposed to f"') v 25 happen. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
~
0170 03 08 40 T' DAVbw 1 MR. EBERSOLE: Any further questions on this V 2 matter? It's been fixed by hard-wiring. 3 (N,o response.) 4 Now we can proceed to the next topic. Thank
-5 you.
6 MR. CREWS: I'm Jess Crews from Region 5. 7 I want to offer comments regarding the next item 8 on the agenda, which is operating experience on the Unit 1 9 Power Ascension Test Program. The Power Ascension Test 10 Program on Palo Verde Unit I commenced in early June of this 11 year, and when the plant resumes operations, there remains 12 about three weeks or so of testing to complete the power () 13 ascension test program. 14 We've had an opportunity over the past 15 approximately seven months to observe the performance of_the 16 operating crews, the technical support organizations and the 17 plant systems themselves, as well as the overall performance 18 of the management systems, not only for the power test 19 program but for the overall contact operations as well. 20 From our observations and my evaluations to date, 21 I.have some comments addressing each of the areas that-I 22 spoke of. Tha operating crews have been chal.enged rather 23 substantially as you've heard dibcussed on thh events of 24 September 12, in particular, where there's a loss of () 25 .off-site power, a failure of the charging pumps and ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverase 800 336 M46
0170 03 09 41 1DAVbw 1 instrumentation failure which further complicated the 2 operator's response in that event. 3 Under the circumstances, the operators performed, 4 we think, remarkably, realigning power to valves and pumps, 5 restarting pumps, realigning valves, providing suction and 6 alternate suction to the refueling water storage tank and 7 restoring the charging pump's operability within a period of 8 approximately an-hour-and-a-half. 9 We think that that experience reflects well on 10 the training and qualifications of the operating crew. 11 We had an opportunity, which I would mention, 12 during the initial criticality low-power testing phase. We i 13 had a team of inspectors who spend some time around the a 14 clock for a period approximately three weeks at Palo Verde, 15 and at that time, they were favorably impressed with their 16 performance. During the power ascension test program, their 17 performance has been quite good. 18 So overall, we rate the performance of the 19 operating crew quite high. 20 The overall technical support organization 21 performance has been acceptable. They've demonstrated the 22 capability to understand the problems encountered during the 23 power ascension program, the procedural actions to correct 24 the problem. (} 25 MR. REED: I'd like to make a point. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 33(H5646
'0170 03 10 42 DAVbw Regrettably, Dr. Okrent and Dr. Kerr aren't here, ( 1 2 but I am trying to sell something, so I'll make the point, 3 that these remarkable operating crews, Mr. Crews is talking 4 about, were POS test selected people, and in some cases, 5 they were pirated people that had been selection-tested for 6 natural ability elsewhere. 7 DR. SHEWMON: From that we conclude that it was 8 all equipment failure?
- 9. MR. REED: I assume we have concluded that.
10 MR. CREWS: To repeat, we think the overall 11 performance of the engineering support organizations has 12 been acceptable, and in some instances, generally, as () 13 appropriate, the vender expertise that was called upon, 14 persists in better understanding some of the complicated 15 problems that-have been encountered. Licensee management 16 has taken steps to improve upon the utilization of the 17 technical staff, as the testing program has progressed and 18 experience has been gained. This staff, like that of the 19 operator's, has been challenged rather substantially. 20 Many have worked long hours. We think they'vemet 21 the challenges pretty woll, and their utilization of 22 performance has improved, as experience has been gained, 23 particularly in the area of post-trip and post-event review 24 and evaluation. With regard to plant systems, the reactor () 25 controls and instrumentation performed well, as have ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33H6d6
0170 03 11 43 l'3DAVbw 1 reactor protection and engineering safety features, when V 2 challenged. 3 The secondary plant systems have generally 4 performed well and reliably. This is especially true of the 5 main feedwater system and the main feedwater system 6 control. 7 Some of the systems, however, have not performed 8 reliably. As discussed earlier, the auxiliary pressurizer 9 spray system and that of the multiplexer control for the 10 distribution system have been disappointing. 11 In the case of the auxiliary spray system, the
. 12 licensee has implemented procedural changes to minimize the (yj e
13 repetitions and problems that have been encountered and 14 proposed design improvements which we feel will 15 substantially improve the overall reliability of this 16 system, which is not fully safety grade. 17 The multiplexer control system has been abandoned 18 in favor or hard-wired typos of controls to important 19 breakers, which we believe will improve the reliability of 20 this system, substantially, as well. 21 In terms of the overall effectiveness of 22 management systems for the planning and conduct of the power 23 ascension test program and the conduct of operational 24 activities, we have been generally satisfied with the (') 25 performance, and there has again been a steady improvement. ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336 6646
f 0170 03<12 44 l-DAVbw 1 . Management has demonstrated a health att.itude [ 2 toward critically examining the performance of the programs 3 and the. managing systems upon which they must rely for 4 effective control for operational activities. 5 Steps have been taken to substantially improve, 6 where experience has revealed the need in such areas as I've 7 mentioned, as well as the control of vendor information, 8 particularly as it relates to the secondary part of the 9 plant. 10 I would point out that the lessons learned during 11 the power ascension test phase in the early operational 12 period of Palo Verde Unit 1 are, in most cases, directly 13 applicable to Units 2 and 3. 14 This is true, both in terms of the physical 15 plant, which is essentially identical in all three units, 16 and the management systems, as well. 17 The improvements which have come about, as the 18 result of Unit 1 experience, we feel, should fully expect to 19 result in measurable improvements in the overall performance 20 of Unit 2, as it enters today's startup. 21 MR. EBERSOLE: Any questions? 22 (No response.) 23 MR. EBERSOLE: If not, thank you, Mr. Crews. 24 And I see no reason for us not to proceed into 25 the licensing status. ACE. FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 6646
0170 04 01 45 1 MR. LICITRA: My name is Manny Licitra. I am the {)DAVbur 2 NRC Project Manager for the Palo Verde plant. I am here to 3 discuss the licensing status for Palo Verde Unit 2. 4 I would like to point out that Palo Verde Unit 2 5 is really the same design as Palo Verde Unit 1. They are 6 all identical, including Unit 3. 7 We actually reviewed the package as a whole and 8 the previous reviews that we have done for Palo Verde 1. 9 The initial license on 1 also applies to Paio Verde Unit 2. 10 However, there were some individual issues that needed to be I 11 addressed for Palo Verde Unit 2 before we would proceed with 12 issuing a license. 13 I would like to put on the slide, (]) ' l 14 (Slide.) 15 What this slide represents is what needs to be 16 done before we proceed with Palo Verde Unit 2. 17 The three remaininc issues are the post-accident 18 sampling system, ECCS reanalysis, and pressurizer auxiliary 19 spray system. 20 With regard to the PASS, the issue'there was the 21 location where the previous PASS system for collecting 22 containment samples, air samples, was located, it turned out 23 to not be feasible for meeting GDC -- I believe it is 17 -- 24 as far as doses to individuals, particularly if you were to () 25 postulate a design basis accident and someone would have to ace FEDERAL REPORTERS, INC. 202 347 3700 Nanonwide Coverase 800 336 6646
0170 04 02 46 1 go to the area than previously chosen to take a containment ()DAVbur 2 air sample, and it did not appear feasible that that sample 3 could be taken and still meet the dose guidelines. 4 So what they have done is to propose a new 5 location. They have submitted information regarding where 6 that location will be. The staff is completing its 7 evaluation, and it appears that we should be able to resolve 8 that issue very shortly. 9 The next item concerns the ECCS reanalysis. 10 Basically, this concerns potential error that is found in 11 the CE model for evaluating the performance of the ECCS, 12 having to do with the point along the axis where the peak () 13 occurred. 14 They had done some studies -- I guess it was in 15 Westinghouse plants -- and determined that the peak had been 16 in a different location than they previously thought, and 17 when they checked it out with CE plants the same thing 18 occurred. 19 What this resulted in is that they had 20 reevaluated the peak clad temperature and the revised peak. 21 It would bring it slightly over the regulation limit of 2200 22 degrees Fahrenheit. I think it is to something like 2203. 23 It is small. It is not a technical issue, but because there 24 are specific numbers in the regulation, in order to say that () 25 this plant complies with the regulation we had to have an I ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-6646
0170 04 03 47 7' 'DAvbur 1 analysis of record with an acceptable mark. V 2 The third item concerns the pressurizer auxiliary 3 spray system which we have been discussing today. 4 Currently, the staff is still reviewing that issue. 5 On Monday we received additional information 6 regarding this issue. It was in that submittal that they 7 pointed out to us that they had this other means of 8 performing a depressurization in the event you have a steam 9 generator tube rupture. 10 At this stage, we don't anticipate there will be 11 any problems in completing the review, but it is still under 12 review. () 13 Before we issue a license, as we do with all 14 plants nowadays, we require that the utility provide us a 15 certification that design, construction, and testing of a
- 16 i unit has been completed in conformance with the FSAR and 17 other docketed commitments. This was done on Unit 1, also 1
18 on Unit 3, and has been done on other facilities. 19 Finally -- this is not directly related to i 20 licensing -- but prior to initial criticality of Unit 2, we 1 21 would prefer to see that the power ascension program in Unit 22 1 is complete. 23 So in summary, Palo Verde Unit 2 is close to 24 licensing. We don't see any problems at this point. As () 25 soon as we wrap up our pressurizer auxiliary spray system ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverase 800 336 6646
0170 04 04 48 1 reviaw, we should be ready to reach a decision to go to low ()DAVbur 2 poser licensing. 3 MR. REED: I would like to ask -- you were 4 pointing out you think you are going tc *: satisfied -- are . 5 you satisfied with the fact that instead of rapid 6 depressurization, as we have used the words over the years, 7 five years, you are satisfied with the slow 8 depressurization, with this design? 9 I recognize that the CE design has large water 10 inventory. It has large steam generators, with the issue of 11 steam generator overfill. All the same, there will be 12 radioactivity discharge until the pressures are either () 13 equalized or something. 14 Are you satisfied with this slower 15 depressurization versus what PORV systems do and a more 16 rapid depressurization to saturation? 17 MR. LICITRA: I can't address the PORV issue. 18 DR. NOVAK: This is Tom Novak, of the staff. 19 I think, Mr. Reed, the question you raise is the 20 question that was discussed at length prior to the licensing 21 of San Onofre 2, for example. 22 I think the staff was of the opinion that the 23 power-operated relief valves would be a beneficial safety 24 feature to a Combustion Engineering design. O 25 the Commission decided at ehat time that the ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336 M46
i 0170 04 05 49 ( )DAVbur 1 merits of it did not support the equivalent of a backfit to 2 the Combustion Engineering design. That decision, in 3 effect, is in limbo until the decay heat removal generic 4 issue is resolved. 5 And I think all the discussions we have heard 6 this morning ace second order effects with regard to the 7 basic question. The improvements in the auxiliary 8 pressurizer spray system that we feel should be applied now 9 give you the feeling that you have enhanced the reliability 10 over a shorter period of time. 11 I think we are expecting that by the first 12 refueling of Palo Verde Unit 1 the A-45 issue will be on the () 13 table and decisions will be made, and depending on the 14 outcome of that, I think there will be some looking back at 15 some of the specific units. 16 These are not items -- if you looked at San 17 Onofre 2, the Waterford plant, and you looked at Palo Verde, 18 you would not find identical systems in terms of the 19 charging systems, and so forth. They are different. They 20 do have some different features, and you want to go back and 21 look at all of them I think individually. 22 At this point in time, the emphasis is on getting 23 A-45 completed. 24 MR. REED: So really the dominant issue here is () 25 not rapid depressurization; it is decay heat removal in l ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coversee 800 336 4646 [
r , 0170 04 OC 50 1 A-45? If PORVs do become a bleed and feed capable
' .)iDAVbur
~
2 requirement, you get rapid depressurization q.e.d.? 3 DR. NOVAK: That is right, but as long as you
- 4. talk in terms of the classic spectrum of accident -- steam 5 generator tube rupture and other events -- you can show by 6 analysis that this system can meet the criteria.
7 MR. EBERSOLE: May I ask a question on this 8 general matter? 9 I think you did, however, ask for a qualified. 10 method of depressurization to low pressure, is this correct? 11 I am talking about a safety grade type system to 12 get depressurized. () 13 DR. NOVAK: Yes. The answer is: in the event of 14 an accident such as steam generator tube rupture, you must I 15 rely on safety grade systems. 3 16 MR. EBERSOLE: Would that have permitted, for i. 17 example, a bunch of hand valves to depressurize? 18 DR. NOVAK: The first answer: we are, I think, 19 relying on the analysis to demonstrate. l 20 MR. EBERSOLE: Was there a claim that this was a 21 safety grade system, the auxiliary spray system? 22 DR. NOVAK: Yes. 23 MR. EBERSOLE: I want to call that out in 24 particular as sort of an administrative problem. I think it (n j 25 has been visible to everybody. ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 233M
0170 04 07 51
- f]DAVbur v
1 This is clearly a system subject to half a dozen 2 or more single failure points, all of which were missed by l 3 both the Arizona Power Company and the staff. That is not 4 E 4 something to be overlooked. It was loaded. 5 What happened here is they reached out and 6 grabbed another system alraady there and used it for an 7 adjunct purpose not heretofore revealed, not to me anyway, 8 as being useful for that purpose. 9 Am I correct? 10 DR. NOVAK Let me say it this way, I think, l i 11 Mr. Ebersole l 12 The licensee, in his documentation, described the ; i i () 13 auxiliary pressurizer system as a safety grade system. His 14 definition of what constitutes a safety grade system hinged 15 on the staff's requirements for being able to take a plant i 16 to cold shutdown using only safety grade equipment. 17 Now, this position had various acceptance l l 18 criteria. Depending on the vintage of the plant, you were 19 permitted to take certain manual actions. The Palo Verde 20 application fell into that vintage where manual actions l 21 would be permitted to be taken outside the control room if 22 enough time was available. l i 23 So that was the window of the Palo Verde unit. l 24 When it met that position as far as the licensee was l () 25 concerned, he had an acceptable safety grade system. i ACE FEDERAL REPORTERS, INC. l 202 347 3700 Nationwide Comage 800 33646d6 ,
0170 04 08 52 LDAVbur 1 MR. EBERSOLE: So he could have had a faucet on 2 the pressurizer and he could have run up and turned it on 3 and that would have been a safety grade system, is that 4 right, in that context? 5 DR. NOVAK: You are not going to put words in my 6 mouth. 7 DR. REMICK: A question. It is my understanding 8 from the subcommittee report that if these three issues are 9 resolved and if you receive the certification from the 10 applicant that the staff would still not allow a Unit 2 11 startup until Unit 1 ascension tests are completed, is that 12 right?
, 13 MR. LICITRA: We would be able to visualize this, 14 but we would not look favorably upon having two units in a 15 i transient startup mode.
16 ! I think the utility also recognizes that, and 17 they do not plan to do it either. 18 DR. REMICK: So you are not necessarily 19 preventing them from doing it; you jointly come to that 20 position? ! 21 MR. LICITRA: They are going to provide us a 22 commitment to the offect that they will not do that. 23 DR. NOVAK: I think the position is that when we 24 go forward with the issuance of the low power license the 25 fuel loading can take place. That does take a substantial ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nation =ide Coverage 800 336-6646
L 0170 04 09 53
/~'\DAVbur 1 period of time.
t \-) l 2 If the Unit 1 startup program comes close to j, 3 being on schedule, it would have been completed within that 4 window of time necessary to load fuel. 5 So the licensee anticipates that he will complete 6 his power ascension programs within the time that Unit 2 7 completes fuel load. 8 Our experience has been that while the operators, < 9 and so forth, are certainly qualified and there is a 10 sufficient staff to perform these functions, the concern is 11 as you go up into the management of the plant that you are 12 now competing with Unit 1 operation versus Unit 2 operation, () 13 and it has just proven out that I think you want to keep 14 them as separate as you can and not overtax your management 15 with competing issues on two different units. 16 DR. REMICK: I don't disagree with the prudency. 17 I am interested from the standpoint of the authority for 18 this. It seems like this is kind of an ad hoc decision. 19 When they get more units in operation, they are 20 going to be faced with those same things. There are going i 21 to be units refueling, and so forth. 22 I realize these are the initial stages, and I 23 really don't, as I say, differ with the prudence of it. I 24 am interested in where that decision is made. () 25 Suppose they don't finish the ascension tests for ACE FEDERAL REPORTERS, INC. 202 347 3700 Nadonwide Coverage 800 336 4646
0170 04 10 54 ( ,)DAVbur 1 Unit 2 -- or Unit I when Unit 2 is ready to start its 2 ascension tests. 3 I am not clear whether the staff is preventing 4 them from starting up, and if so, where is this decision 5 made in the Commission? Is it at the region, at the Office 6 Director level? 7 DR. NOVAK In this case, there is agreement both 8 from the Office Director and Mr. Mothami that this would be 9 a prudent action. We discussed it with the management of 10 Arizona Power, and at this point in time it looks like this 11 approach can be accommodated. 12 . Granted, if there was a substantial problem that
/y (s_j 13 came up in terms of power ascension, if you had a major 14 outage in the unit with one operation, I think there you 15 could take the other side of the coin. If you had the 16 feedwater pump break down or something or the plant was done 17 and it really didn't represent a challenge to operation, you 18 could perhaps go up in low power testing of Unit 2.
19 So I think a major delay can be based on facts 20 that can be evaluated at that point in time, but right now 21 it looks like you can clear the Unit 1. 22 This was kind of the approach that was followed 23 on the Diablo Canyon facility, and we think it is just a 24 very good way to run things. (O_) 25 DR. REMICK: This sounds like the applicant l ACE. FEDERAL ) REPORTERS. INC. 202 347 3700 Nationwide Coverage 800 336(M6
0170 04 11 55 1_ DAvbur 1 apparently has committed to do it'. 2 If you had an applicant who disagreed, what is 3 their appeal process? Has this been faced before? 4 Maybe it is a legal question that you can't 5 answer. 6 DR. NOVAK: It has not been faced. . 7 DR. REMICK: Thank you. 8 MR. MICHELSON: In the ACRS letter of December 9 '81, the ACRS recommended that Arizona Public Service should 10 expand its studies of system interaction and system 11 reliability. 12 Wao there ever any follow-up on that 13 recommendation? If so, what was it? 14 MR. LICITRA: I am not aware of what Arizona 15 Public Service has done in that regard. . 16 MR. MICHELSON: The staff has not received or 17 seen any materials that might have constituted what would 18 appear to be a system interaction study or a system 19 reliability study? 20 MR. LICITRA: I don't believe so. 21 MR. MICHELSON: Did the staff ever pursue this 22 issue, since it was a recommendation in the ACRS letter on 23 Palo Verde 17 24 MR. LICITRA: We didn't pursue that. We wrote a 25 letter to the utilities. ACE FEDERAL REPORTERS. INC. 202 347 3700 Natonwide Coverase 800.))6 Md6
0170 04 12 56 1 DAVbur 1 MR. MICHELSON: Let me ask, can you answer that? I )
\/ 2 MR. EBERSOLE: No. I don't think we have ever 3 had any explicit good explanation as to what has been done 4 on this.
5 As a matter of fact, the indications with the aux 6 spray system indicate that this sort of process has been 7 working very well. 8 MR. MICHELSON: What bothers me a little bit is 9 do we just just idly make these recommendations? Do ve not 10 press one way or the other as to whether or not they are 11 carried out and if they are not, why not? I 12 I mean, why do we make the recommendations if we 13 don't believe it? If we believe it, why haven't we gotten (~)
\~' 14 an answer?
I - 15 MR. EBERSOLE: I wonder if we could get Arizona 16 to comment on the degree to which you do system interaction 17 studies. 18 MR. WARD: Let's clarify the answer from the 19 staff first. 20 Mr. Novak? i 21 MR. MICHELSON: I assume our letters are to the i 22 staff, in essence, and therefore the staff either goes back 23 and says, no, we are not, or, yes, we carried it through. l 24 DR. NOVAK: I don't think it is quite that way. 25 At least what we would do is ask the applicant to respond 'f ACE. FEDERAL REPORTERS, INC. i 202 347 3700 NationwideCo ersee 800134*M4
0170 04'13 57 l_ DAVbur 1 to those issues directly. When we. review their response and I)
in our minds -- depending upon the rigor with which you 2
3 express your desire to reach a conclusion, then I am sure 4 that in a supplement to the Palo Verde safety evaluation we 5 address that concern. 6' MR. MICHELSON: I have not been on Palo Verde's 7 subcommittee meetings before, so I didn't know the history. 8 That is why I was asking because I assumed it was handled 9 further back. 10 I wanted to go back and look at these studies 11 relative to the questions of auxiliary spray as well as a 12 couple of other questions. I was really looking at the 13 reference. 14 And apparently, to your knowledge at least -- 15 I MR. WARD: Wait a minute. He thinks there l 16 probably was something written in an SER somehow. 17 DR. NOVAK: Generally speaking, the first safety 18 evaluation supplement that we issue addresses those concerns 19 identified in the ACRS letter. So we do very shortly after l l 20 the issuance of the safety evaluation meet with the 21 committee. If we have written a favorable letter, then we l 22 pursue those comments with the applicant. He would address 23 them, we would review them, and we would follow-up the 24 safety evaluation. l 25 Now, it may not be very specific, but in some ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coversee 900136 MW4 i
0170 04 14 58 1 DAVbur 1 senso it should be addressed. 2 MR. WARD: Perhaps we could ask the staff to get 3 us a copy of the SER. 4 Dave, did you have something you wanted to say? 5 DR. OKRENT: I was just going to repeat what you 6 said. After the break, we ought to have before us just what 7 the staff said. 8 MR. EBERSOLE: Let mo make a comment. 9 I think it is rather clear evidenco in the case 10 of the design of the auxiliary spray system itself that 11 whatever may have been said in general terms the system 12 doesn't reflect a very high degree of system interaction 13 study capability. 14 15 16 17 18 19 20 21 22 l 23 24 25 ACE FEDERAL REPORTERS, INC. 202 147-)?00 Nationwkle Coverage M10136 6M6 i
0170 05 01 59 1 DAV/bc 1 MR. MICHELSON: Could we get any comments from 2 the licensee in this regard, as to what studies they might 3 have done on reliability as well as system interaction? 4 MR. WARD: First of all, I guess, are you 5 familiar with a' request from the staff with that 6 recommendation in the ACRS letter? 7 MR. HAYNES: Personanlly, I'm not. I haven't 8 been with the project that long. I do know that we have 9 accomplished a reliability group. I believe that was 10 established subsequent to the unit one. That group is 11 active in doing reliability analyses on the systems and, in 12 fact, has looked at the auxilliary pressurizer spray system 13 subsequent to September 12th of that. And I have thoroughly 14 evaluated the changes in reliability that result from the 15 modifications that we're doing. 16 But they're also activo in doing reliability 17 studies on other systems. As to responding to this specific 18 item, I'm not aware of the status of that. And, 19 unfortunately, wo don't have anyone here with us today who 20 can respond. 21 MR. MICHELSON: On system interaction, do you 22 have something ongoing? Have you dono a system interaction 23 study for Palo Vorde, or something of that sort? 24 MR. HAYNES: I don't think so. I'm sorry that I 25 can't respond moro. (
- )
ACE FEDERAL REPORTERS, INC. 200 347 3700 Nationwide Coverase 800 336 6M6
0170 05 02 60 1 DAV/bc 1 MR. EBERSOLE: You would have done of course a 2 study, a system interaction study, that always precedes the 3 occurrence of the event in order to prevent this sort of 4 thing from occurring. 5 MR. HAYNES: I'm sure if we were requested to do 6 such a study, we would have done it. 7 MR. EBERSOLE: Any further questions? 8 (No response.) 9 MR. EBERSOLE: I think we should have a break 10 before we have a discussion here. 11 MR. WARD: And I hope we can get the SER 12 supplement. Let's take a 10-minute break. 13 (Recess.)
~
14~ MR. WARD: We'll reconvene. 15 Mr. Ebersolo, I guess, first, we would like to 16 hear if there's any information on Mr. Micholson's question 17 in the SER supplement.
, 18 MR. EBERSOLE: Mr. Licitra.
19 MR. WARD: Does staff have anything? 20 MR. LICITRA: Yes. What I'll do is refer you to 21 supplomont number one on the Palo Vorde SER. In supplement 22 one to the SER, we did address that comment by the ACRS in 23 the December '81 letter. What I'll do is just read from the 24 documents. First, it says the committoo stated that the _ 25 Arizona Public Sorvice Company should expand its studios of N ACE FEDERAL REPORTERS, INC,
, 202 347 3700 Nationaide Coverage 800 336-6M6
0170 05 03 61 2 DAV/bc 1 systems interactions and systems reliability. The response 2 that was provided in the supplement was as follows: 3 Item A-17 in Appendix C to the SER discusses the 4 ongoing staff efforts to reach generic" resolution of the 5 systems interactions in nuclear power plants. It is
. 6 expected that the development of systematic ways to identify 7 rank and evaluate systems interactions will go further to 8 reduce the likelihood of intersystem failures resulting in 9 the loss of plant safety functions and hence improve systems 10 reliability.
11 After resolution of this generic issue, the staff 12 will determine whether additional studies by Arizona Public 13 Service Company are required. / 14 ' MR. EBERSOLE: I think the aftermath of that is 15 the finding that the aux spray system is not working would 16 require investigation of its fundamental weaknesses. At 17 least my response to this meeting would be take a letter of 18 l two paragraphs: 19 "We see no reason to impede further progress on 20 fuel loading or full power for unit two." 21 The basic issue still remains the same about the 22 PORV versus some other method of cooling in addition to the 23 high pressure cooling and the decay heat removal system, and 24 that we will probably write a paragraph pertinent to the 25 absence of any interactive studies that are going to
\
ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336 4 646
Y fg 0170 05 04 62 2, DAV/bc \1 preclude any findings on the aux spray system. t > 2 Glenn... A 3 MR. MICHELSON: That's only half the answer. 4 A-17 is the reliability study, what did the SER say? You 5 answered it for the interaction study, who asked for two p 6 things -- interaction studies and reliability studies. They 7 are quite different. 8 MR. EBERSOLE: I didn't take them to be 9 different, Cdrl. 10 MR. MICHELSON: They're quite different. The y 11 interaction study is quite different from the reliability s 12 study. ,
, 13 MR. EBERSOLE: Are you talking about direct
( 14 liability of the direct systems? 15 MR. MICHELSON: Yes. There's a great difference. 16 MR. EBER' SOLE: What do you have to say about the 17 reliability studies? 18 MR. MICHELSON: You can have a situation where 19 you have a very unreliable system that has no interaction 20 with other systems. 21 MR. EBERSOLE: I agree. 22 MR. NOVAK: This is Tom Novak. I think that the 23 response the staff gave is in the form of if you do a 24 systems interaction study, you learn from it. It follows
.,- 25 that you snould if you make changes, enhance the
%,)
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
0170 05 05 63 1 DAV/bc 1 reliability. And that was all that the staff was saying. 2 The staff did not specifically respond to the committee's 3 recommendation that the applicant perform, that he should 4 expand studies on system interaction and systems 5 reliability, and systems reliability. 6 MR. MICHELSON: The two studies that we asked 7 for, you didn't seem to address very much the latter. If 8 you read the one in RHR, decay heat removal, you will find 9 that you did say you were going to put that in a later SER. 10 So I assume, if you search the SER's, you'll find it. 11 MR. LICITRA: I think that was addressed in the 12 supplement, supplement one. i 13 MR. MICHELSON: Do you know which one? [D
14 MR. LICITRA: I'm guessing. I think it's number l 15 five.
16 MR. MICHELSON: Thank you. 17 MR. EBERSOLE: Perhaps that can be looked at. 18 Glenn, do you intend to make a statement? 19 MR. REED: Jessie, are we sort of wrapping up 20 what we might put in the letter? 21 MR. EBERSOLE: Yes. 22 MR. REED: I'll not talk about...I've been 23 involved quite a bit in Palo Verde in the last year. I'll 24 not talk about the good and bad things I've seen. I really 25 have noted a good operating organization. I do agree with ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
0170 05 06 64 l_ DAV/bc 1 Mr. Crews. I think there must be some frailties in design 2 thinking and technical support, because it appears to me 3 that this test for electrical continuity and its impact 4 unfolded deficiencies in the auxilliary spray system. And 5 that's not a real engineered way to get to a discovery of 6 the vulnerabilities. 7 I happen to think that the applicant, the 8 licensee, has made a good point with respect to the need for 9 rapid depressurization versus slower depressurization. As 10 an operator, I perhaps would like to be more able to rapidly 11 depressurize to saturation, to turnaround steam generator 12 tube rupture.
~
13 But with their large steam generators at their
~,
14 Site, it looks to me that slower depressurization would be 15 okay. I do feel very much that the final wrapup of Palo 16 ' Verde is tied to what Mr. Novak said. 17 The A-45 thing we must get on with and we must 18 find out whether the reverse bleed and feed cooling 19 technique is important. I think something to that effect 20 should go in the letter. 21 I also have been concerned that there are a 22 number of single component failure potentials in the systems 23 we saw today and in other systems, such as auxilliary 24 feedwater and in safety injection.
,, 25 This comes about from the fact that what I think
( u-ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 80 4 336-6646
0170 05 07 65 1 DAV/bc 1 I see is a policy on the part of either combustion or Palo 2 Verde to have lots of closed valves. 3 And I wonder if that shouldn't lead somebody to a 4 limited PRA review of safety systems. 5 With respect to these closed valves, many of them 6 I think we could leave open. I think we ought to consider 7 whether we should have a mini-system, a limited systems PRA 8 review some time along in the next year or so, to see if 9 there shouldn't be some changes in valve positions to 10 improve our reliability. 11 MR. EBERSOLE: Any further comments on this? 12 (No response.) 13 MR. EBERSOLE: I presume you all know that we are 14 obligated to provide a reply to Commissioner Asselstine 15 about these recent events we've heard about. My position at 16 this time is that we write about a two-paragraph letter that 17 takes up the several matters that we've heard about today, 18 and have it ready for discussion Saturday. 19 MR. WARD: Let me ask the staff beyond the 20 committee's request and the need to be responsive to the 21 Commissioners, Commissioner Asselstine in particular, does 22 the staff, in terms of licensing the second unit, does the 23 staff see any need for further ACRS comment? 24 MR. NOVAK: No, sir. We can go forward without a 25 specific letter from the committee on the acceptability of f ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336-6646
0170 05 08 66 1 DAV/bc 1 licensing unit two. k- 2 MR. EBERSOLE: This will of course come prior to 3 the next SER, which is like this week. 4 MR. MICHELSON: Would that statement be f 5 predicated on the assumption that you took care of our ( 6 comments on unit one? In other words, would you proceed i l 7 without a letter from ACRS on unit two if it did not cover 8 things that the ACRS had asked for on unit one? The system 9 interaction question? t 10 MR. NOVAK: I guess what I perceive from this 11 commit. tee, and I'm prejudging the letter, that the committee 12 would not have written or would not have advised the staff 13 that prior to issuance of unit two, certain things must be
\s 14 completed. I think what the committee is saying, there are 15 some recommendations you're going to make. They do fall in 16 the period of time that goes beyond issuance of the license 17 and commitment has to be obtained from the applicant 18 regarding certain studies.
19 MR. MICHELSON: Thank you. 20 MR. EBERSOLE: It's yours. 21 MR. WARD: If there are no other comments, we,'ll 22 close out this agenda item then and move on to the next 23 topic. l 24 (Whereupon, at 10:30, the meeting was adjourned 25 and the committee went into an unrecorded session.) O
'% )
ACE. FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 336 6646
0170 06 01 67 1 AFTERNOON SESSION (3:25 p.m.) gDAVbur 2 MR. WARD: Our subcommittee chairman isn't here, 3 but I think we can begin. We will give him the opportunity 4 to say what he wants to say. 5 But my understanding is that we were going to get 6 some presentations from the staff. 7 Here he is. 8 Do you just want to start hearing from the staff, 9 or have you got anything to say? 10 MR. EBERSOLE: No, I have nothing to say. 11 MR. WARD: Why don't you go ahead, Mr. Rossi? 12 MR. ROSSI: I am Ernie Rossi, from the Office of 13 Inspection and Enforcement. -( } 14 I have a number of other staff members with me 15 today. Rob Hernon, from the Office of Nuclear Reaction 16 Regulation is here, and we have a number of people who are 17 going to make presentations. 18 We are going to start with a problem that was 19 discovered October 10th in Maine Yankee. It involved 20 improper installation of Rosemont transmitters. 21 Pat Sears, from the Office of Nuclear Reactor 22 Regulations, will give this presentation. 23 MR. EBERSOLE: Dave, I want you to hear this. 24 This has to do with the exodus of what you call (} 25 primaticity. It also relates to what I said earlier that ACE. FEDERAL REPORTERS, INC. ll02 347 3700 Nationwide Coverage 2 336-6646
'0170 06 02 68 f~}DAVbur 1 you should use a steam genie on uvery installation to see %) 2 whether you do a good job for a steam jet. 3 MR. SEARS: My name is Pat Sears, the Maine 4 Yankee Project Manager. 5 This concerns improper installation of Rosemont 6 transmitters and certain other problems we found 7 subsequently. 8 With the plant down for refueling, Maine Yankee 9 discovered that pressurizer pressure, pressurizer level, and 10 steam generator level transmitters had not been installed in 11 complete agreement with EQ test configuration. 12 The installation was completed in 1982. The 13 transmitters were locked, and Maine Yankee believed that (]) 14 they could be environmentally qualified, and it turned out 15 that, yes, they passed the test. The problem was Maine 16 Yankee did not go back and check to see that the 17 installations were exactly as tested. 18 It was discovered by an operator who found that 19 he could move fairly easily the leads coming out of one of 20 the transmitters, one of the connectors to the t 21 transmitters. 22 They went around and tightened everything up to 23 the test configuration. Then they started looking for other 24 stuff. () 25 They found two other fairly small problems: ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nadonwide Coverase 800 336-6646
. . . . . _ . . . . . _ . . . . . . . . . . . - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - * - - - - ' ^"
l l 0170 06 03 , 69
- Tm DAVbur 1 one, dra n holes in the bottom of the junction boxes were
! V ! 2 not drilled. Those are the connectors that are in the 3 junction boxes. ( 4 Now, the question I was asked Monday that I 5 couldn't answer for sure ist what relationship do those 6 junction boxes have to the Rosemont transmitters? 7 The answer is none. They are not connected to l 8 them, nor are they part of that. 9 The junction boxes have power cabling coming into 10 them. Rosemont transmitters, of course, just have 11 instrument cabling coming in. They are physically remote. 12 They are not electrically connected. ! 13 The junction boxes carry power cabling, with j f 14 limit switches for those valves. The junction boxes : 15 themselves are not required to be environmentally I 16 qualified. 17 However, in my view, the junction blocks that 18 they are going around are environmentally qualified. The 19 environmental qualifications report requires that the 20 junction box be drilled for pressure equalization and to 21 allow moisture to escape. 22 MR. MICHELSON: That is the same as saying there 23 are not open terminals or terminal blocks within the 24 junction box, just splices. 25 Is that really true? ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336 4646 l- - -- - - _ _ . _ _ _ _ _ .
0170 06 04 70
~ 'DAVbur 1 MR. SEARS: There is a person here who can answer 2 that.
3 Stan, could you get up over there to the 4 microphone? 5 MR. HERBANOWSKI: Stan Herbanowski. And the 6 question was again? 7 MR. MICHELSON: What is insido the junction box? 8 Are these just pigtails spliced together with those 9 connectors, or are they terminal strips with screw-down 10 connectors or whatever? 11 MR. HERBANOWSKI What is in thoso boxes are 12 terminal blocks like the one being passed around. That is a 13 single junction point from a terminal block. Typically, 14 terminal blocks will have eight to twolve positions, and 15 similar to the viewgraph that is on the board there. 16 (Slide.) 17 MR. SEARS: It is not a very good viewgraph, but 18 it is the best I can do. 19 MR. EBERSOLE: That is a nonseal block, correct? 20 MR. HERBANOWSKI Yes, sir. 21 MR. EBERSOLE: Therefore, if it is initially 22 cold, it suffers from condensation upon steam reentry into 23 the box. That means it suffers from voltage leakage to 24 some greater or losser degree from polo to pole and pole to
) 25 ground?
ACE. FEDERAL REPORTERS, INC. l 202 347 3700 Nationwide Coverage HX)-336-6646 i
0170 06 05 71 1 MR. SEARS: Yes. {]}DAVbur 2 MR. EBERSOLE: Has it been tested for that 3 condensation leakage condition, and has it been shown that 4 in a_ dirty typical condition there will not be excessive 5 short-circuit currents? 6 MR. SEARS: Can you answer that? 7 MR. EBERSOLE: This is the classical question. 8 We had all thought hermetic sealing of electrical 9 apparatus was mandatory. We found that this situation has 10 been compromised in numerous places in which open terminals 11 have been exposed to this transient condensation and leakage 12 currents do flow as a result of the initial condensation () 13 phase. 14 MR. MACHINTONI I am Armand Machintoni, from the 15 Equipment Qualification Branch. 16 The concerns you are voicing are a concern in low 17 current type circuits and surfaces of that nature. 18 Now, I am assuming that these terminal blocks 19 were used in control and power circuits; whereas, the small 20 leakage currents that have been observed as a result of the 21 steam impinging on these terminal blocks would not be a 22 problem In this application. 23 MR. EBERSOLE: There is a notorious report from 24 Sandia that says in the testing of these in various () 25 laboratories they have gone so far as to preheat these ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33HM6
0170 06 06 72
'DAVbur 1 terminal blocks before spraying steam on them, thus 2 precluding any condensation phenomenon.
3 I ask you, have those terminal blocks been , 4 subjected to the initial normal condition when they are cold 5 and then submitted to a steam environment wheroin the 6 maximum condensation rate takes place? 7 MR. MACHINTONI I can't answer that. I haven't 8 soon the qualification reports. 9 MR. EBERSOLE: We need to get an answer on that. 10 i MR. MICHELSON: When the qualification testing 11 was dono, was a hole drilled in the junction box that was 12 tested? Did they just drill them in the field, or is this a 13 now feature added without going back on the test? ( '. 14 MR. SEARS: The qualification test report from 15 { Woidmuller, the manufacturer, required that they be vented. 16 MR. MICHELSON: They did include venting when 17 they did the testing. 18 Thank you. 19 MR. EBERSOLE: Could the required testing then be 20 done with an initially cold condition with a steam 21 onvironment imposed upon this vintage construction? 22 MR. MACHINTONI As far as I am aware, the 23 terminal blocks woro in the normal ambiont status before the 24 test. They woro not prohoated.
') 25 MR. EBERSOLE: They were in a normal ambient ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 804336-646
0170 06 07 73 1 status and then the leakage currents were measured; is this {}DAvbur 2 what you are saying? 3 MR. MACHINTONI From what I recall, yes. I do 4 not recall the terminal blocks being preheated. 5 MR. EBERSOLE: I think we would appreciate a 6 confirmatory piece of paper referring to this. I think it 7 is a universal problem and a rather critical one if we are 8 faced, you know, with excessive short-circuit 9 phase-to-phase / phase-to-ground current. 10 DR. SHEWMON: Jess, did you agree that if these 11 were power cables that the condensate leakage currents would 12 be a "no, never mind," whereas with instrument systems they () 13 would be a problem? 14 MR. EBERSOLE: I don't know. I don't know if 15 they would be a "no, never mind" or not. 16 MR. MICHELSON: They could be in both cases. 17 Clearly, in the switched gear cases excessive condensation 18 leads to a flow of current around. 19 MR. EBERSOLE: Some of these currents might be as 20 high as 4.0, mightn't they? 21 DR. SHEWMON: How could a cable with condensate 22 carry as much current at 440 volts? 23 MR. EBERSOLE: If it is dirty and the poles are 24 closed, it will do it. () 25 DR. KERR Mr. Chairman, if we are really ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coversee 800 336 6646
0170 06 08 74 gggDAVbur 1 designing a qualification tost, it sooms to me wo ought to 2 do this at a separato mooting, not at this one. If thoro is 3 some ser,ious concern about the qualification test that is 4 being used, then it sooms to me that ought to be taken up by 5 a subcommittoo. 6 MR. MICHELSON: It was, except they didn't havo 7 the information that they are now giving us. 8 DR. KERR: But I am suggesting that the kind of 9 information that we apparently need is something that ought 10 f to be explored in subcommittoo, it scoms to me. I MR. EDERSOLE: This is extrapolated from the 11 l 12 subcommitteo information hero. Wo didn't got this 13 inf o rma tion. 14 DR. KERR I moan, if we nood further information 15 i of this kind, it sooms to me -- 16 MR. EDERSOLE: I think they would say this could 17 be resolved by some sort of a lottor or some statomont from 18 the staff to us at a lator date. 19 DR. KERR: It would be nico, though, if tho staff 20 know exactly what it is that we wanted them to document. 21 MR. EDERSOLE: I think the staff knows precisely 22 want. 23 DR. KERR Maybe the staff doos. I don't. 24 MR. ROSSI: Lot mo ask you a question. I think
') 25 if wo are going to be asked to como back and answer, it ACE. FEDERAL REPORTERS, INC.
202 347.)?00 Nationwaje Coverage s m )36.u 4
0170 06 09 75 ; t Tm DAvbur 1 might be good if we had a letter from the committee with l-U ( 2 specific questions that were being asked because a lot of l 3 times when we come to these meetings we get a lot of 4 questions, and we aren't really sure how much staff effort l 5 to put in to coming back and trying to answer each one. f 6 So I would much prefer -- and I think Ron Hernon j i 7 would, too, if we could have the questions that you really a feel you need answerm to in writing. Then we could devote [ 9 the staff resources to it and come back. 10 MR. EBERSOLE: Only since it is a very reaching, l t 11 widespread problem, I think it would be easier to put a ( 12 letter out on this matter. , [ t 13 MR. REED: Are you going to tell us who was 14 responsible for the modification installation of these 15 Rosemonts? 16 MR. SEARS: There was no modification. These l 17 were installed. t 18 I don't understand the modification part. 19 .MR. REED: Is this an aspect from Three Mile i i 20 Island? Did they change? ! 21 MR. SEARS: I believe so. 22 MR. MACHINTONI It was kind of an equipment : i 23 qualification upgrade and Three Mile Island. i i 24 MR. REED: So some installer, either the j 25 contractor or the utility did the work. Who did the work? I I ACE FEDERAL REPORTERS, INC, l 202 M 3700 Nailonw w Covweee R S) M ee6
h ! 0170 06 10 76 I NDAVbur 1 MR. SEARS: The utility, I believe. It was done f~/ x_ l 2 in 1982, before the EQ test was performed. 3 MR. REED: As at Three Mile Island. I was 4 wondering why the circuits didn't get complete. l 5 Is that a contractor fault or a utility weakness? 6 MR. SEARS: It is a utility weakness. It is a QA l 7 fault. It is clear to me, anyway. l l 8 MR. REED: I guess I should put that ; 9 differently. The licensee is always responsible. I was l 10 wondering whether he had it done by contractor. Was there a 11 contractor gap there? i l 12 MR. SEARS: I don't see that there could have 1 13 been a contractor gap. The equipment was installed, but the (]) 14 tests were run, the test reports were given to Maine l 15 Yankee. Maine Yankee did not go back and see that this l 16 equiprent was in enhanced testing condition, a clear QA 17 fault. 18 on the follow-up Maine Yankee checked for other 19 like problems and found none, except as noted up there. 20 On September 10th, J. T. Beard briefed this 21 subcommittee on the steam generator, pressure 22 instrumentation that was compromised. Now, taken together, 23 these events are being considered an abnormal occurrence. 24 The safety significance, of course, ,is the steam line break ! 25 could have compromised the systems that were served by the (]) l ace. FEDERAL REPORTERS. INC. 202 347 3700 Natlon=hle Coversee 800 )J4 M 4
0170 06 11 77 DAVbur 1 Rosemonts. 2 MR. EBERSOLE: Would it be more accurate to say 3 that any kind of a steam or water vapor leak of substantial 4 amounts in the containment rather than the steam line break 5 itself, any kind of a leak, a high prossure lino leak would 6 have done that? Wouldn't it? 7 MR. SEARS: It could have, yes. My problem horo 8 is that I don't know how tight thoso connections woro. 9 What Maino Yankoo did, as soon as they found a 10 looso one, a OA man was calloo in. Ho couldn't answer how 11 tight they should be. 12 MR. EBERSOLE: Isn't the so-callod tight tent
~
13 establishod by how tight you do it in foot-pounds, 14 i inch-pounds, or howevor you do? 15 l MR. SEARS: In this caso it was clearly 16 specified. All of the connections had inch-pound. 17 MR. EBERSOLE: So this in an intrinsic weaknosa 18 we already know about in a tight test concept? 19 MR. ROSSI: I think thoro is a misunderstanding. 20 My undorstanding of what happonod was that they installod 21 tho oquipmont prior to comploting tho oquipmont 22 qualification test in anticipation that the equipment would 23 pana tho tost. Thon they ran the tonta under cortain 24 conditions, and they didn't food back the cortain conditiona
') 25 to bring the plant equipmont into conformanco with the ACE FEDERAL REPORTERS, INC.
am.nm moon.a. co,mu www
0170 06 12 78
^ " 'DAvbur 1 conditions under which the tests were run.
w/ 2 So I am not sure it means that there was a 3 problem with the tight test. There was a problem in making 4 sure that the installed equipment was inscalled the way it 5 ultimately was tested when it passed. 6 MR. EBERSOLE: Only what I am saying is that is 7 what is wrong with all type tests, whether the field 8 installation matches the test? 9 MR. ROSSI: That is true, but you wouldn't run 10 the equipment qualification tests necessarily on the 11 installed one. 12 MR. EBERSOLE: No, but you must follow rigidly ()) 13 procedural details. 14 MR. REED: This is not a unique thing that is 15 occurring out there with Three Mile Island backfits as far 16 as the sequence of events. A lot of Rosemonts are here that 17 haven't been equipment qualification tasted, and they aro 18 waiting for this significance and waiting in the great rush 19 to get these things turned out. 20 llave you gotten out a bullotin to other people 21 who are involved in this Rosemont backfit? 22 There might be the possibility that it exists for 23 others. 24 MR. ROSSI: I don't beliove we havo gone out with () 25 oither a bulletin or an information notico. I bolieve that 1 ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nation *lde Coverast $$))MM
I 0170 06 13 79 (' ',DAVbur 1 we are still considering an information on the problem. v 2 MR. SEARS: I know we are. 3 MR. ROSSI: So you know it is a breakdown in the 4 QA that ought to be out there, and we are looking at whether 5 a notice may help. I think it is unlikely that we would go 6 with a bulletin. 7 MR. SEARS: I talked to the region yesterday. 8 They are going to run a special OA audit in this area to see 9 that the as-billed condition matches testing conditions. 10 This is Maine Yankoe now. 11 DR. OKRENT So far we have been talking about 12 the Rosemont connections. What is not clear to me is 13 whether there may be other connectors made by a different ([ ) 14 manufacturer which have boon one way or another left in a 15 nonhermetic condition and how the staf f would know if this 16 woro the caso. 17 MR. SEARS: They did find one. 18 DR. OKRENT: You are talking about this 19 particular plant, and I am talking about 100 plants. 20 MR. SEARS: You mean the generic issues? 21 MR. ROSSI: I don't think there is any question 22 that overy so ofton we find problems where the aquipment is 23 insta11od in a way that is inconsistont with the way that it 24 was tosted or analyzod. () 25 That is true of soismic devicos, also. It is a Ace. FEDERAL REPORTERS, INC. 202 347 3700 NaHonwide Coverage Aa4))fkA444
0170 06 14 80
' DAVbur 1 QA problem, and I don't know that noticos and bulletins and 2 that kind of thing will totally solve the problem.
3 It doesn't soom feasible to go out with a 4 bulletin that says chock ovary picco of equipment in ovory 5 plant. It doosn't soom too useful to go out with a bulletin 6 that says chock particular piccos of equipment in ovary 7 plant. 8 What wo would normally do is look and soo if 9 thoro woro something unusual in the instructions or 10 somothing also about ono ploco of equipmont that other 11 liconsoon might not havo the right information to uso in 12 their QA program, and then wo would go out with a notico or
-x ,
13 l a bullotin. i 14 It is difticult to know oxactly how to handle it, 15 ! but wo run into thoso probloms, I guons I would say, now and 16 thon. 17 18 19 20 21 22 23 < 24 ( 25 Acu. FEDERAL REPORTERS, INC.
- 02 34mm Nanon.iae coversee mn))uM4
0170 07 01 81 f~ DAV/bc 1 MR. SEARS: Maine Yankee, by the way, does have a U
- 2 good QA program in place.
L 3 DR. OKRENT: I heard the answer but I'm not sure 4 why you think it's adequate. ,
- 5 MR. SEARS
- The QA program?
6 MR. ROSSI: No, why my answer is adequate. I'm 7 not sure how to reply to that question. Clearly, what we're 8 doing, well, not 100 percent ensure that every problem is 9 found on every plant. You know, the responsibility for the 10 QA lies upon the licensee. What the NRC does is try to 11 audit what they're doing, make sure they're doing it right 12 and make sure people have the right information to use in 13 their QA programs. (]} 14 And where we find they aren't doing these things 15 in our inspection programs, we take appropriate 16 enforcement. It's a matter of practicality and how far we 1 17 can go in doing the job for the utilities. 18 Clearly, if we think they're missing information 19 they need to do their job correctly, then we issue an 20 information notice or a bulletin or a generic letter, 21 depending on the severity of the problem and how important 22 we think it is to get feedback on what we did. 23 MR. REED: Let me try to help you a little bit 24 with this because I feel that you don't have to go out and 25 generate research on ground zero to this. There's a (]) ACE FEDERAL REPORTERS, INC. 202 347 3700 Natkmw6de Coverage 800.))&M4
0170 07 02 82 T 1DAV/bc 1 different pattern in this Rosemont installation -- that's 2 what I was trying to make a point -- from the normal pattern 3 of received equipment qualified goods, putting them in 4 place. 5 These Rosemonts, I think many licensees put them 6 in place, waiting for lator equipment qualification 7 inf orma tion. 8 MR. ROSSI: You're absolutely right because they 9 may have put it in. They didn't have all the original 10 information when they originally installed it and now they 11 havo to come back and perhaps modify the installation. 12 You're right, that is different and that is a smallor 13 problem. This may be the kind of thing that the notico is (]) 14 quito appropriate on. : 15 If wo do go with the notice, it would seem that 16 it ought to be worded more broadly than just "Rosemont 17 transmitters". It ought to cover everything that's been put 18 in prior to doing the E0 testing. You want to make suro 19 that the loop was closed, to make sure that it's installed 20 the way it was ultimately tested and that the testing was 21 dono after the installation. 22 I think that's your point. 23 HR. EDERSOLE: May I ask a question, Ernio? 24 In the broador context, can you provide us with
~T 25 any kind of statoment going back to day one as to what's (O
ACE. FEDERAL REPORTERS, INC, 202 347.)?00 Nat'9awide Coverage , 800-)36 W6
0170 07 03 83 DAV/bc 1 been required of the utilities, beginning at the highest 2 voltage we have inside these containments in these hostile 3 environments, and what they do with the variety of terminal 4 types that they have in regard to either hermetically 5 sealing them or permitting them to be open and not subject 6 to leakage currents. 7 As a case in point, I must think the 4160 8 leached, the main coolant pumps, are hermetically sealed; 9 correct? 10 MR. ROSSI: I'm reluctant to try to answer. 11 MR. EBERSOLE: I'm just starting with that, okay? 12 Now, I'm going to como down the line to 440, 250 DC, et (~ 13 cetera, et cetera. Are there any general requirements that
\_-)
14 say that you must hermetically seal these as a guaranteed 15 concept of adeqacy? Or if you're going to run the risk of 16 leakage currents so qualified by whatever method you 17 proscribe? 18 l MP. ROSSI: I suspect that I'm not absolutely 19 sure of this, that we do not tell them prescriptively when 20 and when you don't hermetically seal things, that our 21 requirements are more generally worded, that they have to be 22 qualified to do the job thet they're intended to do. And we 23 don't tell them how you go about doing that. i 24 That's the gonoral approach that we tako. 25 MR. EDERSOLE: So we have no knowedge at this (]) ACE FEDERAL REPORTERS, INC. 202 347 3700 Natbn*kle Coverage 8(01364M6
l I 0170 07 04 84 llgDAV/bc 1 time what they do. 2 MR. ROSSI: I do not have any. There are j
- t 3 probably general things which are done, but I suspect that ;
i 4 we don't prescribe it, that our regulations are more 5 general when you say it's got to do the job under the j 6 conditions that exist when it has to do the job.
. I 7 And we tell them that they've got to test or i e
8 whatever. Leave it at that, and then audit, testing, and so I 1 9 forth. 4 i 10 MR. EBERSOLE: Ever since I heard that Sandia 11 report in which the testing laboratories, it's so biased it ! 12 makes things worse, that's been bothersome. - MR. ROSSI: That seems to me a case in which the (]) _ , _ 13 14 way'they tested was probably the problem. I guess we have 15 somebody here from the Equipment Qualification Branch. ; 16 Do you have anything to add or to correct to what 17 I said? t 18 MR. MACHINTONI Yes, the requirement on j 19 utilities is that the equipment must be installed in a l. 20 manner which does not violate the test which was performed i 21 on it during the qualification test. If the terminal block j 22 or the equipment was tested and exposed to the environment I 23 that it sees during an accident, then it can be installed j i-l 24 the way it was tested. Our requirement on all qualified equipment in (]) 25 ACE. FEDERAL REPORTERS, INC. l' 202 347 3700 Nation =We Coversee 800 ))6 Me4
0170 07 05 85 1 that it has to be installed the way it was tested. (}DAV/bc 2 MR. ROSSI: And we don't prescribe whether it has 3 to be hermetically sealed or whatever, as long as it passes 4 the test. 5 MR. MACHINTONI: That's correct. , 6 MR. EBERSOLE: On the other hand, you didn't 7 define the testing conditions to preclude their preheating 8 the system and they escape through that narrow crack. 9 MR. MACllINTONI: I would say, if they prehoated , 10 the equipment, it's a violation of the testing requirement. I 11 They should not preheat it. They should test the 12 equipment. It should be when it sees the accident ; (f 13 environment. If preheating it is done, then it's an invalid 14 tost and wo wouin't accept that. 15 MR. EBERSOLE: Okay. Good. 16 DR. SIESS: Preheating to what temperature? 17 MR. EDERSOLE: Whatover temperature would prevent 18 condonsation when it's impacted by steam. 19 DR. SIESS: That would be higher than the normal 20 containment temperature, which is what? A hundred and 21 twenty? 22 MR. EDERSOLE: I think it would be, yes, probably 23 higher. 24 MR. ROSSI: I'm not familiar with the details of () 25 that report but what you're describing to mo sounds like a l ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage $ 6 336 6644
1 0170 07 06 86 f ';DAV/bc 1 clear-cut case of where they simply went out of their way to
\_/
2 do the test in a way that wasn't valid, so they'd pass it. 3 That's assuming that it was the way you described 4 it. 5 MR. WYLIE: This could occur. Maybe the 6 Equipment Qualification Branch might want to speak to it, 7 but it has occurred in the past whero equipment was not 8 deliberately done that way, but it was done because they had 9 limited test facilities. And in order to get the 10 temperatures up, they'd first run steam through the chambers 11 and then hit it with the pressure later, in which case they 12 dried it out when they wore running the steam through 1,t. () 13 That is an erroneous way to do the test, but that 14 was becauso they didn't have an adequate steam sourco to do 15 the tost. 16 That has occurred in some of the verification 17 tests. At the National Labs also, where they had an 18 inadoquato steam source. 19 MR. ROSSI: I think that's a general problem, 20 too, that when you run the test, you have to run it in some 21 way that it approachos as nearly as you can practically do 22 the conditions you're going to see during the accident. , 23 There may be cases whero peoplo have just technically not 24 dono it correctly because of a lack of knowledgo or () 25 whatevor, which is a differont situation than I understood Acn FEDERAL REPORTERS, INC. 202 347.)?le Nationwide Cowrage $10-))66646 n
0170 07 07 87 T'?DAV/bc 1 you to be describing. v 2 One is a deliberate way to circumvent the test 3 requirements. The other is something that they didn't know 4 at the time that made the test not valid. They're very 5 different situations and, you know, we obviously look for 6 both in what we do. 7 MR. WYLIE: But the events that I would cite here 8 is the breakdown on the QA program, where the OC inspector 9 didn't sign off on the way something was installed. The 10 recommendations by the manufacturer that it be installed in 11 a certain way, even though in this case the testing was 12 after the fact, wo should have come back and done something 13 about it. ( ') 14 MR. EBERSOLE: By the way, did the papers show i 15 that everything was in order? 16 MR. WYLIE: That's the question, whether the QC 17 inspector signed off on the installation or not. 18 MR. ROSSI: I don't know, but I thought we had 19 somebody here from the utility. 20 MR. SEARS: Stan, do you know if there was any 21 documentation on the original installation? 22 MR. IlERBANOWSKI: I've not seen the package. 23 MR. WARD: Would you identify yourself, please? 24 MR. IIERDANOWSKI Stan tierbanowski from Yankee i j ; 25 Atomic, representing Maino Yankee. I have not noen the l Acn. FEDERAL REPORTERS, INC. 202 347 3700 Natione6de Covernse 8%)S%4
0170 07 08 88 ... design package that the transmitters were installed under, gDAV/bc 1 2 so I can't verify what's in the package, the signof fs. 3 ,MR. WYLIE: I would assume that the quality 4 assurance program requires that they have a complete design 5 package or installation spec before they install it. 6 MR. HERBANOWSKI: That's true. j 7 MR. WYLIE: In general though, you have to have 8 qualification testing before you install them. 9 MR. HERBANOWSKI True. 10 MR. WYLIE: Isn't that correct? 'You quarantine 11 equipment until you get that kind of information. Then, 12 once it's installed, then the QC inspector signs off that he O 13 saw it installed and verifies that it was installed the way 14 it was supposed to. , 15 MR. HERBANOWSKI This is the case where the [ 16 transmitters were installed with conditional approval on 17 EO. 18 MR. WYLIE What do you do where you don't have , 1 19 the conditional approval? i 20 MR. HERBANOWSKI: The normal case is that the 21 equipment is qualified first and we're aware of the 22 qualification test report and have it in our possession. i 23 In this particular case, it was part of a atassive , 24 industry program to qualify the Rosemont series i O 25 transmitters. And all the transmitters that were installed i e ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nulon Wecoversee son n u d6 l
0170 07 09 89 ' ',DAV/bc 1 during this time woro conditionally approved for EQ based on 2 successful completion of the EQ test. 3 MR. WYLIE: What ensures then that you are going 4 to install the correct one? 5 MR. SEARS: The QA system broke down there. 6 MR. EBERSOLE: I think we can close out on this. 7 I think we can write the letter you want, Ernie. 8 MR. WARD: All right. We've used about half the 9 time. 10 MR. EBERSOLE: We'd better accolorato the rost. 11 MR. WARD: Ton minutos a picco for the next 12 items. [N 13 MR. ROSSI: The next prosentation is on an event 14 that involved reactor coolant system leakage to the reactor 15 building through a leaking scram solenoid, and then through 16 the scram dischargo volume vont and drain valves. 17 This presentation will be given by Eric Weiss of 18 the Offico of Inspection and Enforcement. 19 MR. EBERSOLE: While he's getting roady, lot mo 20 remind you of the flatch ovonts and tho Oyster Crook ovents. 21 Again, this is a recurring problem with the standard 22 solonoid valvo and the dump volumo of the boilor. 23 MR. WEISS: On September 19th, Drosdon had a leak 24 of roactor water in the reactor building caused by what wo (~') 2S now understand to be the configuration of the pilot scram ACE. FEDERAL, REPORTERS, INC. 202 347.)?00 Nationwide Coverage 800 136 6644
-0170 07 10 90 I'NDAV/bc 1 solenoid valve'and the characteristics of those valves.
V 2 The reactor was operating at 80 percent power 3 when they had a spurious turbine control valve closure. 4 They got a pressure spike on APRM high. The operators 5 attempted to reset the scram but due to a problem with the 6- mode switch, they could only reset channel A. 7 This turns out to be one of the necessary 8 conditions for the sequence to occur. You have to have the 9 A channel preset:and the B channel still tripped following a 10 reactor scram. And then we will get this leak. And we 11 'would expect it to occur in other boiling water reactors 12 except.those with fast scram drives. 1 () 13' g The operators noticed steam in the reactor 14 building. They got radfation alarms. There was 15 contamination on the first three levels of the reactor 16 ". building. They got some false indications which misled them 17 as to where the leak was coming from. And, ultimately, they 18 terminated the event by closing the vent and drain valves,
/ 19 which I'll show you later.
I 20 Three individuals were contaminated during the 21 event, and I understand a fourth was contaminated during the 22 cleanup. ~7 23 There are tests which were rather extensive both 24 on the reactor and in place. These indicated that 51 scram (} 25 solenoids were leaking and there was a reconstruction that 4 ACE FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6646
- / ,,, _ , .. __--
0170 07 11 91 ' T"NDAV/bc 1 indicated that the mode switch was left inbetween positiono \_) 2 because it was part of the normal scram recovery procedure 3_ that they move the mode switch to the refuel position to get 4 the white light on rod position. It verifies that all rods 5 are in and they inadvertently left it between positions. 6 (Slide.) 7 The leakage path, which I'll show you later, was 8 from the vessel through the scram discharge valves and into 9 the reactor building via the reactor building equipment 10 drain tank and the instrument volume. The leak lasted for 11 about 23 minutes. 12 (Slide.) 13 We did an investigation of this event. We went ( }) 14 out and examined the licensee's extensive testing efforts 15 with the help of the resident inspector. We concluded the 16 following things, that most BWR's will have this leak if 17 they have A channel reset, B channel tripped following a 18 full scram. 19 We think the operator should be able to recognize 20 this event because they have adequate instrumentation in the 21 control room to recognize. They have a blue light on the 22 full card display that indicates the scram outlet valve is 23 open. They also have red and green lights indicating the 24 position of the scram discharge volume vent drain valves. () 25 The research that was done by the licensee ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 33(H5646 l _a
0170.07 12 92 "~ 1 suggests that the exhaust diaphram on one of the two valves (iDAV/bc
/
2 in the culprit and that it tends to leak due to bad 3 pressure. There may be a contribution due to maintenance. 4 I bought the valve with me. It is a three-way 5 air solenoid valve with a pilot solenoid. It has two rubber 6~ diaphrams in it that have to pop as the result of solenoid 7 action. 8 I've also brought pictures of the work in place 9 and some of the bench testing that's done. 10 (Slide.) 11 I'd now like to briefly illustrate how this event 12 could occur. This shows the reactor protection system in () 13 the reset condition prior to a scram, with the air pressure 14 holding the scram inlet and outlet valves closed and holding 15 the vent drain valves open. 16 So these are air to close, these are air to 17 open. 18 (Slide.) 19 During a scram, the scram solenoids which I've 20 indicated here, the 117 and 118 valves, will change 21 positions and they'll cause the scram inlet and outlet 22 valves to cor.e open. And these valves over here, the 20 23 valves, will cause the vent and drain valves to go closed. 24 This gives it a solid reactor water underneath 25 the drive piston; the control rod drive removes it from the (]) ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nadonwide Coverage 800-33MM6
0170'07 13 93
;()DAV/bc 1 top of the piston and deposits it into the reactor building 2 equipment drain tank where it stays bottled up.
3 There is a condition called a half scram, which 4 occurs quite frequently. It doesn't look a lot different 5 than the reset condition. We have one channel tripped. We 6 have one of these solenoids tripped, but the net effect is 7 we're still sitting there very much like we were in the 8 reset condition, even though we've got one of these 9 solenoids tripped. And one of 20 valves tripped. That's 10 the one out of two taking place. 11 The logic is that we must have both solenoid 12 pairs go in order to get the valves to change position. O 13 (stiae > 14 The problem occurs if you have a ledk in the 15 exhaust diaphram of the 117 valve. We get a degraded air 16 pressure here; following a scram, we actually have zero 17 pressure hare due to the backup scram solenoid. But when 18 the backup scram solenoids close, we don't build up air 19 pressure to the full 75 psi if this is leaking out through 20 the vent. We'll have a degraded air pressure here, say, of 21 38 psi, which is sufficient to hold these valves open, and 22 ineufficient to close these over here. 23 So now we have a leakage path out the vent and 24 _out the drain into the reactor building. p O 2s (slide.) ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33M446
0170 07 14 94 1 This diagram shows the scram solenoid pairs, the lT]DAV/bc 2 117 and 118 valve, solenoids in both exhaust and pressure 3 diaphrams in both, and in a nutshell, what happens or what 4 we believe happens is that there's a pressure built up on 5 this side of this particular exhaust diaphram which keeps it 6 from completely seating due to the pressure drop across this 7 open diaphram. 8 This incidentally shows you in a half scram 9 condition, or half reset condition if you want to be more 10 precise -- by half reset, I mean a half scram following a 11 full scram, where we have the A channel reset.1 and the B 12 channel tripped.
-'O ,g 13 14-15 16 17 18 19 20 21 22 23 24 m
U 25 ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646 2
0170 08 01 95 T'~ ' DAVbu r 1 (Slide.) L.) 2 The solenoids I passed around the room appear
, 3 here under each hydraulic control unit. They are indicated 4 by the number 7. So they are addressed, and there are 177 5 pairs of these solenoids.
6 And in response to this event, we are issuing an 7 information notice. 8 I have also included in your package a more 9 detailed drawing showing some of the intricacies of the 10 system. 11 Are there any questions? 12 MR. EBERSOLE: Are there any questions on this? (] 13 DR. KERR: What do you think will be the effect
- 14 of your investigation on future incidents?
15 MR. WEISS: I expect that plants will modify 16 their procedures and training and that if it should occur 17 again -- which we think is a fairly low probability event -- 18 but if it should occur again, I would expect that operators 19 will recognize it more quickly and have a procedure in place 20 to deal with it. 21 DR. KERR: Thank you. 22 MR. EBERSOLE: Any other questions? 23 (No response.) 24 MR. EBERSOLE: Let's move along to the next one, (~T 25 then.
%)
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
0170 08 02 96 DAVbur 1 MR. ROSSI: Next is going to be a discussion on 2 problems with the auxiliary feedwater system at Turkey Point 3 Unit 3, which occurred on July 22nd. , 4 We are going to have two people giving us 5 presentations. First, Bob Baer, from the office of 6 Inspection and Enforcement, is going to review the specific 7 event and causes for it. 8 Then we have Dan McDnald, the Project Manager, 9 from the Office of Nuclear Reactor Regulation, who is going 10 to give a more general discussion of the auxiliary feedwater 11 system design at Turkey Point Unit 3 and some of the changes 12 that have been made to it over the years. 13 Bob Baer will start with the discussion of the (a) 14 specific event. 15 (Slide.) 16 MR. BAER: With the shortage of time, I will try 17 and run through this fairly quickly. This was an event the 18 committee heard about at its September meeting. This is 19 just to try to refresh your memory. 20 The event occurred just before midnight on July 21 21st, 1985 and carried on into the next morning. The Unit 3 22 tripped from full power, as I said, just before midnight. 23 The level in the steam generators were maintained 24 with the main feedwater pumps. They were run back to the () 25 control level, but the high level occurred in the steam ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646 i
0170 08 03 97 T^'iDAVbur 1 generators, and this caused the main feedwater pump to V 2 trip.
)
3 Turkey Point has three safety grade steam turbine !
.4 driven aux feedwater pumps. These started automatically, 5 and all of them started and ran. One pump, however, had 6 problems with an air-operated aux feedwater control valve on 7 the discharge side of the pump, not directly related to the 8 steam admission to the turbine.
9 Sometime later the main feedwater pump was 10 restarted, and the aux feedwater pumps were shut off. -The 11 aux feedwater pump turbine governors were reset locally. 12 This involves among other the steps draining oil () 13 from the governors. I will get back to that point in a f- 14 moment. 15 The pump then had problems with the air flow 16 water control valve. The steam supplied to the turbine that 17 runs that pump was put ir an off-normal configuration so 18 that they could do troubleshooting of the flow control 19 valve. And what this amounts to is there is an interlock on 20 one of the steam admission valves whose position is 21 interlocked with the air supply to the flow control valve so 22 that normally closed -- those two valves are in series. The 23 normally closed valve was open, and the normally open valve 24 was closed so they would have air in this feedwater control P () 25 valve. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6646
0170 08 04 98 T ;DAVbur 1 Sometime after that, while operating on the main 2 feedwater pump, again there was a high level of steam 3 generated. This time when the aux feedwater pumps were 4 restarted two of them tripped on aux feed. Those were the 5 two in the normal configuration. 6 The third pump, which was in this abnormal 7 configuration as far as the steam supply valves, operated, 8 but the fuel oscillated. The pump went from on to off. 9 The basic cause of the problem was the governor 10 oil was not drained properly when the aux feedwater turbines 11 were reset locally. 12 What this amounts to is there was oil in the
/
) 13 governor when the pumps were called upon to restart and the v
14 control system was biased to a higher speed. It starts up 15 faster, and it is more likely to have a mechanical overspeed 16 trip. 17 The procedure was not sufficiently explicit. The 18 generic actions that were taken with I&E is we have obtained 19 about a week and a half ago a great deal more information 20 from Terry Turbine on the various governors and auxiliaries 21 that are offered for Terry turbines, and we plan to prepare 22 an information notice that will describe the event and make 23 the information available to the industry. 24 MR. REED: I am a little surprised to see here () 25 again -- this, I believe, is a Westinghouse unit -- that ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
0170 08 05 99 TDAVbur 1 all the auxiliary feed pumps are steam driven, the same as 2 Davis-Besse, only there is a better steam supply. 3 MR. BAER: Dan Mcdonald is going to talk about 4 the actions taken at the plant. 5 MR. EBERSOLE: There is a part two to this, 6 Glenn, there under licensing. Part two is coming. Part two 7 is critical to this presentation. 8 MR. BAER: You are on, Dan. 9 MR. ROSSI: We have Dan Mcdonald now, and he will 10 progress the design of the auxiliary feedwater system. 11 MR. EBERSOLE: There is kind of a correlation to 12 the Davis-Besse problem here, and some actions are
/~ ) 13 contemplated to bring it up to a high level of reliability.
v 14 By the way, these three pumps are shared with two 15 units, right? So there is not really three turbine pumps 16 per unit but three turbine pumps per two units. 17 (Slide.) 18 MR. MC DONALD: Attached to the briefing package 19 is the memorandum to the Commission, in which they asked the 20 staff how were we responding to the letter you sent to the 21 Commission expressing your concern about this event, and 22 attached to that memorandum is a response to the ACRS 23 letter. 24 I believe your concern is broader than the event 25 itself, but more concerned about the regulatory process and
} })
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
0170 08 06 100 ( jDAVbur 1 how we are attempting to identify a combination of things 2 that individually may not lead to an event but collectively 3 contribute to it. 4 In order to do this, I want to very quickly go 5 over the original design of the Turkey Point auxiliary 6 feedwater. Then I will go over the current design that has 7 resulted in the changes in the post-TMI. I have highlighted 8 the key changes that have been' implemented subsequent to 9 that. 10 FP&L has installed two motor driven pumps, 11 nonsafety, 850 gallons, and an additional 500,000 gallon 12 demineralized water storage tank.
) 13 After that, I will go through the follow-up of a
14 the event by the staff and also actions taken by the 15 licensee, and very briefly I will discuss a special I&6 eam 16 inspection, which was not the result of the event itself but 17 concerns the staff has had that was highlighted by the 18 Davis-Besse event in that maybe we were not going into 19 enough depth and looking collectively over all the things 20 that could contribute to it. 21 (Slide.) 22 In order to get a feel for the original design, I 23 have a one-line. 24 As indicated, there are three turbine driven [v', 25 pumps. There is a three-loop Westinghouse plant, Unit 3 ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
0170 08 07 101 1 and Unit 4, with a steam supply from each of the steam lines {}DAVbur 2 just outboard of the concaimment on each of the units. 3 They go through a pressure reduction stage. 4 These were low pressure turbines, two condensate storage 5 tanks, 250,000 gallons, 185,000 gallons, dedicated to the 6 aux feedwater system. Then the outboard flow control valves 7 to the steam generators of each of the units. 8 I would like you to note the piping and the 9 number of valves, both at the inlet and the outlet. 10 The next one-line in the package -- 11 MR. REED: Why did you make that point of the 12 number of valves. I was sitting here pleased that they were 13 in the open position versus other manufacturers and reports ()~ 14 we have heard. 15 MR. MC DONALD: As a result of the TMI 16 modifications, this was just simply three pumps shared 17 between the two units as a result of the modifications. Let 18 me just highlight them again. 19 (Slide.) 20 The system was manually initiated. It became 21 automatically initiated. There is redundant flow 22 indication. There is redundant level alarms on the 23 condensate storage tanks. Two separate full flow capacity 24 AFW trains. () 25 When you look at the other one-line attached to ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
0170 08 08 102 TDAvbur 1 the package -- I will put it on the projector in a minute -- V 2 you will note there is more redundancy in the inlet valves, 3 plus some of the passive piping was even added to ensure 4 that even passive failure would not result in failure of the 5 system. 6 There are some air-operated flow control valves 7 which had a nitrogen flow station for each of the units. 8 That was divided and redundant for the two trains. There 9 was no diversity on the inlet valves. They were all AC. 10 Two of them are on each of the units. Two are DC and one is 11 AC. 12 The pressure reduction station was taken out, and () 13 the high pressure turbines and a trip and throttle valve 14 were installed in each unit within each train. 15 MR. EBERSOLE: Pardon me. You are telling me 16 that they are normally open? 17 MR. MC DONALD: Yes. 18 MR. EBERSOLE: How does one maintain aux 19 feedwater flow when you have a total AC power failure? 20 MR. MC DONALD: This was post-TMI modification. 21 (Slide.) 22 These are the normally closed. These are steam 23 inlet valves. DC, AC, and DC. So we have DC now, 24 post-TMI. Prior to that it was only AC operated, like a lot () 25 of the older auxiliary feedwaters. ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-6646
T- l 0170 08 09 103 i 1 MR. EBERSOLE: Are there any environmental or l (^JiDAVbur u 2 other problems with these auxiliary turbine driven pumps i 3 which require AC cooling or any other supportive function? 4 MR. MC DONALD: There are support cooling systems 5 which I didn't include as one of the things. But this 6 cooling system form has also been upgraded. 7 MR. EBERSOLE: Can these pumps run for an 8 acceptable period of time without any AC? 9 MR. MC DONALD: Yes, sir. 10 MR. EBERSOLE: Thank you. 11 MR. MC DONALD: As I mentioned, the only thing I 12 wanted to indicate, you can see the train, and now this is () -13 very busy and very difficult to follow. But you have Train 14 1 and Train 2. You can see the valving arrangement and the 15 piping arrangement has been modified, both on the inlet and 16 on the outlet sides. 17 So given a failure in any one train, in any 18 components within that train now, you do have a two-train 19 concept versus three shared, which is still a very 20 complicated system. The pumps, as I indicated, are now high 21 pressure. They have taken out the pressure reduction, which 22 has improved the reliability of the pumps themselves. They 23 have installed a trip and throttle valve, which is normally 24 open. It is 'actually used in the electronic overspeed: for l () 25 the turbines.
- ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 6646
0170 08 10 104 1DAVbur 1 Again, this is not part of TMI, but there were 2 installed the motor driven pumps with another source of 3 water, 500,000-gallon demineralized water storage tank. 4 (Slide.) 5 The reason I bring up the question about the lack 6 of diversity in the motive sources -- I have indicated there 7 is diversity in the steam inlet now, but with the 8 installation of the nonsafety grade steam generator feed 9 pumps we have a letter with a commitment from the licensees 10 to tech spec these pumps, provide surveillance to assure 11 that if the safety grade system fa.ils that one of these two 12 pumps will be adequate and available to provide a water
) 13 source for the steam generators.
14 I might mention they are nonsafety, but this 15 size is somewhat in need. There are five what we term Black 16 Star diesel generators. 17 These Black Star generators are used with two 18 fossil units that are onsite. These Black Star generators 19 are adequate to power these two pumps for exactly the same 20 manufacturing size as the safety-related diesel generator 21 onsite. 22 They have hardwired these in. They have taken 23 the breakers out. There are interlocks. They have a 24 procedure, and in about 15 minutes they can provide loss of
- 25 all offsite power and could provide a backup power to these a
ACE. FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-33M686
0170 08 11 105
>DAVbur 1 pumps and provide another method of getting water to the 2 auxiliary.
3 MR. EBERSOLE : I believe you said they were 4 radiator cooled, weren't they? 5 MR. MC DONALD: I believe they are. I am not 6 sure, Mr. Ebersole. 7 And I will get back on some of the actions that 8 we are taking in the proposed generic letter to address the 9 concern. There are some plants out there that lack 10 diversity. 11 MR. REED: You certainly overwhelm me with the 12 redundancy in the equipment for auxiliary boiler feed as a () 13 technique to remove decay heat. 14 Just another question: do they have diversity in 15 principle perhaps; what are the sizes of their PORVs; do 16 they have PORVs? 17 MR. MC DONALD: They have PORVs. I don't know 18 what the size are. 19 MR. REED: You don't know what their bleed and 20 feed capacity is? 21 MR. MC DONALD: They do have a procedure they 22 have already developed. It will be in process or in place 23 by the end of the month. 24 The staff has tried to separate on the secondary j 25 side feed and bleed, being our defense in depth concept. We ACE. FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6646
I 0170 08 12 106 T~DAVbur 1 still want to decide that the auxiliary feedwater systems G 2 and the diversity in it and use the feed and bleed as the 3 defense in depth. 4 In fact, I will go to the generic letter. There 5 are at least eight or nine plants that have been identified 6 that do not have the reliability that we defined, high 7 reliability, 10 to the minus 4 to 10 to the minus 5, that we 8 feel should go back and look at their systems and may 9 require some equipment modifications. 10 In addition, that generic letter, which is 11 getting put together for a CRGR package, will include the 12 concern about diversity, and I believe there are only four () 13 plants out there that do not have motor driven diversity in 14 the motive source for aux feedwater. 15 With a commitment from Turkey Point, all four of 16 these plants, similar to Turkey Point, have tech spec 17 procedures in place as a backup for their steam driven 18 systems. 19 I have kind of jumped ahead a little. I won't go 20 into the immediate action, but Bob mentioned there was a 21 procedural problem with oil gunk, also. We had to confirm 22 that all the components within the system were operable. 23 There were some problems with the air system. We 24 immediately had to verify on Unit 3 and also the operating 25 reactor that all inherent components were operating. (]) ACE. FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
E 0170 08 13 107 DAVbur 1 An information notice.
-(
2 There was a problem with dumping the oil, and 3 that is being developed by I&E. 4 As I mentioned, we have a commitment for the 5 nonsafety pumps tech spec'ing, and we have addressed the 6 general concerns expressed in your letter to the 7 Commission. 8 9 10 11 12 O ta -- 14 15 16 17 18 19 20 21 22 23 24 w O 25 ACE FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 336 6646
l 0170 09 01 108
'DAV/bc 1 I believe the sensitivity that staff has now is m
2 tne regulatory process breaking down. We have, as you are 3 aware, a Maintenance and Surveillance Branch now. And it 4 appe,ars that improper maintenance, not in itself but 5 collectively, with testing, training and all these things 6 together, seem to contribute to these events. 7 We have a team going out to survey several plants 8 in the area of maintenance and surveillance. And they 9 happen to be going to the Turkey Point site on December 5th 10 and 6th. This will be factored in to either requirements or 11 the way that we expect to look at these aspects. 12 In addition, they're going to have another
) 13 outlook at the unit and have a look at what is being done to m
14 upgrade the units. 15 As far as Florida Power and Light, they took care l 16 of the immediate problem but in the long-term, they have a 17 program in place which we call the PET Program, the 18 performance enhancement program. To get a feel for that, 19 it's a $100 million effort. They hired 140 additional group 20 that will be on the Turkey Point site. They have made 21 management changes, including a site vice president at the 22 Turkey Point site. 23 In this program, because of the time, I'll just 24 go through. It addresses design, operation, testing, 25 training, all these programmatic things, coupled back { }) ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwiie Coverage 800 336 6646
0170 09 02 109 ( '; DAV/bc 1 together and along with QA and QC, as we discussed earlier, 2 before the committee on the Rosemont. It's not just that 3 the QA program is there, but the people that are looking at l 4 it are taking what we term a vertical look, looking 5 indepth. For example, if it's a qualification question, as, 6 in fact, has been installed as the qualification requires. 7 If it's a test, is it tested to what it should be 8 for performance. FP&L is attempting to get people that are ; 9 licensed operators for the Turkey Point facility integrated 10 in all these areas. So they will get the operations 11 feedback through the checks and balance system. 12 Part of this PET program includes a plant () 13 specific simulator, new health physics, and some other 14 aspects. I'll touch on this team concept again. 15 As part of the staff's sensitivity to the 16 regulatory process, we'll be looking at the right things. 17 As I indicated, I&E even prior to Davis-Besse, and 18 subsequent to that, determined that possibly the inspection 19 modules were not adequate for picking out the interaction of 20 all these different items. 21 The special team included what we term PAT and 22 IDI performance appr'aisal teams, an integrated design 23 inspection concepts where they looked at both programmatic 24 and at the design from the initial design basis through all () 25 the changes. And is it still within its original design ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-6646
0170 09 03 110 basis given the modification changes, and also some
'DAV/bc 1 2 investigative techniques that were used in Davis-Besse.
3 In summary, in the long-term, we are looking more 4 at maintenance surveillance. They are looking at different 5 inspection methodologies to ensure that this is an 6 interactive thing. 7 The licensee program, which I haven't given as 8 much attention as I should, on the PAT program and the 9 maintenance, for example, they're looking for predictive 10 analysis, not just some sort of calendar schedule of 11 maintenance. They've got what they term MOVAT, 12 motor-operated valve analytical testing machines, which will
) 13 automate testing of MOV's.
14 So it's very comprehensive. It will be some time 15 before all the benefits are seen, but it is in progress. 16 MR. MCDONALD: Are there any questions? 17 Yes, sir? 18 DR. KEPR: I don't mean this to sound like a 19 critical question. It's not meant to be. 20 Have you looked to see if there are any tech 21 specs or regulations that are likely to interfere with 22 achieving a more reliable system? 23 MR. MCDONALD: Any tech specs or regulations? 24 I'm sure not aware of any. I can't think. () 25 DR. KERR: Has there been a systematic look to ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6646
0170 09 04 111
;DAV/bc 1 see if something about the way the regulations are written 2 would inhibit the program, which would otherwise improve 3 reliability of the system?
4 MR. MCDONALD: Related to reliability, I don't 5 know if there's been a systematic look at it. But we do 6 have a group that is looking at -- I don't know the proper 7 term -- tech spec betterment, or whatever. They are looking 8 at the tech specs per se. 9 DR. KERR: I applaud what you're doing. It seems 10 to me what you're doing makes sense, but we both know that 11 there is an extensive literature of tech specs associated 12 with the operation of any of these plants. And it,'s clear
; 13 to me that one might want to look and just make certain that 14 somehow this one is not overloaded.
15 MR. MCDONALD: I appreciate that. One thing I 16 did mention about Turkey Point specifically. It is an old 17 vintage plant, licensed in 1971 and '72, with plant specific 18 tech specs, which they're in the process of converting to 19 the standard tech specs. 20 But, in response to your question, we are looking 21 at the standard tech spec and I assume we're looking at any 22 effect it could have on reliability, or these other things. 23 But I can't specifically say. 24 MR. ROSSI: One of the things that's being looked f; 25 at in tech specs is trying to optimize things like ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-336-6646
0170 09 05 112
'DAV/bc 1 surveillance. And that sort of thing, to get good 2 reliability and make sure we're not degrading reliability by 3 doing more testing of systems than is necessary.
4 I would assume you're generally aware of the 5 fairly extensive tech spec work that's been underway for 6 some time. 7 MR. EBERSOLE: Mr. Mcdonald, I detect here that 8 now we at long last are in the process of doing field in' 9 situ examination of design adequacy. 10 MR. MCDONALD: We are in the licensee also. 11 MR. EBERSOLE: It's been long in coming. You 12 know, the old inspection routine never did that.
) 13 MR. MCDONALD: I think that's true. My v
14 experience has been, and I hate to use a cliche, but the 15 horizontal look is just like when we were talking earlier, 16 in Rosemont, if you've got a OA program, you should catch 17 these things; but sometimes they don't. 18 ' When you start doing the indepth, you find more 19 things; even though, as I mentioned, this PET program has 20 been in place for some time, I believe that what was found 21 in this inspection by the staff has sensitized them more for 22 the utility itself to look in more depth. 23 And they surely are. They've made several 24 commitments to us. ( ) 25 MR. EBERSOLE: This may be beyond existing ace-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
0170 09 06 113 DAV/bc 1 regulations and criteria. 2 MR. MCDONALD: I think the findings of the team 3 and the event at Turkey Point provided support for the 4 things the staff is looking at and doing in terms of these. 5 As a result of this event and the inspection on 6 Turkey Point, Florida Power and Light is going to be a 7 little ahead of the pack and have these all ready. 8 MR. EBERSOLE: Mr. Mcdonald, your point was 9 intended to give you all a warm feeling, and I hope you have 10 it. If you don't, speak up. 11 (No response.)
- 12. MR. WARD: Thank you.
13 Do you think we could skip the next item? s 14 MR. ROSSI: This is fairly short and I'm not 15 going to get up there to give it. It's recent reactor trip 16 breaker failure on the Westinghouse Plant on October 29th. 17 The D.C. Cook Unit 2 reactor tripped on a spurious, low flow 18 signal from the reactor component pump breaker contacts. 19 And there wasn't an actual loss of power to the pumps or a 20 loss of flow. 21 During the trip, reactor trip breaker A failed to l 22 open, and reactor trip breaker B accomplished the trip 23 function. As I'm sure you're aware, the Westinghouse plants 24 have two breakers, either one of which will accomplish the (} 25 reactor trip. These breakers are Westinghouse DP-50 type. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 80 4 336-6646
0170 09 07 114
;DAV/bc 1 Neither of the breakers on Cook Unit two had the 2 automatic shut trip installed as yet. Shunt trips, 3 automatic shunt trips, however, have already been installed 4 on Unit one. In evaluation of the event at Cook, we feel 5 that the A breaker had marginal, well, the undervoltage trip 6 attachment has marginal capability to trip the breaker.
7 On November 3rd, a few days after the event on 8 Oc'tober 29th, they did a bench test of the remaining seven 9 operating and spare breakers of Cook units one and two. And 10 during that, the breaker which had been successful in 11 getting the reactor trip on unit two on October 29th failed 12 the undervoltage trip attachment force margin test. 13 Now we have gone out on the 5th of this month
}
14 with an I&E bulletin. And that bulletin is addressed for 15 action to the licensees of operating Westinghouse reactors 16 that have not yet installed the automatic shunt trip feature 17 on the reactor trip breakers. And that bulletin will 18 require forced margin testing of the undervoltage trip 19 attachment within seven days of receipt of the bulletin. 20 And then, thereafter, it requires monthly forced 21 testing until the shunt trip has been installed into the 22 automatic protection system and our preliminary survey 23 indicates that there are now three operating plants that 24 have not yet installed the shunt trip into the automatic 25 protection system. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336 6646
0170 09 08 115 DAV/bc- 1 There are six other plants that have not yet 2 installed the shunt trip in the automatic system. But they 3 are currently shut down and we expect that they're going to 4 complete the installation of the shunt trip into the 5 automatic system before resuming operation. 6 That's all I have to say on that. 7 MR. EBERSOLE: Mr. Chairman, that's it. 8 DR. KERR: The three does not include two? 9 MR. ROSSI: That's correct. There are three 10 other than Cook II, but Cook II is now shut down and they're l'1 going to install a shunt trip before they go back up. And I 12 believe Cook II, I think they're included in the remaining {} 13 14 ones. Oh, one'further thing. We just received a call 15 today from Kewanee. They had started tP9ir testing as a 16 result of the bulletin and the B bypass breaker failed the 17 undervoltage trip attachment margin test. 18 The A bypass breaker there had been tested and 19 passed the test. The B bypass breaker failed the test. And 20 they had not yet tested the breakers that are actually being 21 used in operation. 22 MR. EBERSOLE: I want to thank you and your team 23 for making the presentation. 24 MR. WARD: Thank you. Let's take a break and {} 25 26 return at quarter to 5. (Whereupon, at 4:35, the committee recessed to go 27 into an unrecorded spsion.) REPORTERS INC ACE-rEDERAL , . 202-347-3700 Nationwide Coverage 800 336 6646
CERTIFICATE OF OFFICIAL REPORTER l') \J This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the matter of: .' NAME OF PROCEEDING: ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 307TH GENERAL MEETING E DOCKET NO.: 4 PLACE: WASHINGTON, D. C. O b DATE: THURSDAY, NOVEMBER, 7, 1985 were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission. (sigt) / v N (TYPED) DAVID L. HOFFMAN Official Reporter Ace-Federal Reporters, Inc. Reporter's Affiliation O
O ADVISORY COMMITTEE ON REACTOR SAFEGUARDS FULL COMMITTEE MEETING NOVEMBER 7, 1985 ARIZONA NUCLEAR POWER PROJECT
~
PRESENTATION O
, . _ _ _ - --=- " -
- i. ., - - . .
4,
- 1. l
}
emnnraow - - m , n og s D
. I
- O
.#. oJ "O
o
'. a E
i \
,, o
- w 't g t
- =
x - a O - l- > m :: 1 *. *. ... [ , g- . ** W
*ss
> m .
S
/p
- s -
l O *, P@ e e e e e e e e e e e l a e e ~ e e s e e *- -
,,-,_=,,--,.--.
. , . - , _ _ - , _ . . - ________-_2-
DEVELORENT OF PAT PRO @M SCEDULE ACTUAL TEST PR08M DWATION AT OTER CE PLANTS SCOPE OF TESTING F(R PWGS vs. OTHER & PLANTS ACTUAL TEST THE TEST PREDICTIONS vs. TEST RESULTS VERY GOOD #REEMENT BETWEEN PREDICTIDN/RESILTS O
'~^
NO IMXPLADED TEST RESULTS OR PLANT TRANSIENT RESPONSE PLANT TRIPS: 7 TRIPS OWO WHILE TESTING) TWO MAIN FEED RW TRIPS OhE CEAC CIRCUIT B0MD FAILlRE THEE LOSS OF PWER TRIPS (ONE MUI.E TESTING) OE S/G LW LEVELS TRIP (WHILE TESTING) O
-e , ._ _
- _ _ _ _ _ . _ _ +h_ mala.
MODIFICATION
SUMMARY
(~T v
- ADD SECOND, DIVERSE REFERENCE LEG TO VCT LEVEL TRANSMITTERS ADDRESSES ROOT CAUSE OF LOSS OF CHARGING EVENT REPLACES NEED FOR DAILY REFERENCE LEG CHECK PROVIDE POWER TO VCT OUTLET AND RWT GRAVITY FEED LINE VALVE FROM 1E MCC ELIMINATES NEED TO MANUALLY RESTORE POWER TO THE VALVES FROM OUTSIDE THE CONTROL ROOM FOLLOWING AN SIAS ADD AUTOMATIC ACTUATION TO GRAVITY FEED LINE VALVE ELIMINATES THE NEED TO MANUALLY OPERATE THE GRAVITY FEED LINE VALVE FROM THE CONTROL ROOM DURING LOSS OF 0FFSITE POWER CONDITIONS L ) LOCK OPEN TWO NORMALLY-OPEN VALVES IN AUXILIARY SPRAY FLO ELIMINATES POTENTIAL FOR ISOLATION OF GRAVITY FEED OR CHARGING LINE DUE TO SPURIOUS ACTUATION OR OPERATOR ERROR O
,w. - -. * -
e. s.+ ne.-, .s.
.O MUITIPIFXER SYSTEM o REMOTE OPERATION AND INDICATION OF NON-SAFETY EQUIPMENT, INCLUDING THE NON-SAFETY 13.8 KV SilITCHYARD BREAKERS o COMPUTER-BASED SYSTEM UTILIZING FIBER OPTICS OCTOBER 3 AND 7, 1985 EVENTS o SPUR-IOUS SIGNAL FROM THE PLANT MULTIPLEXER CAUSED AN INADVERTENT OPENING 0F THE MAIN FEEDER BREAKERS SUPPLYING 0FFSITE POWER O - -
CORRECTIVE MEASURES o OMPLETE HARDWIRE MODIFICATION FOR CONTROL OF 13.8 KV SWITCHYARD BREAKERS o PERFORM SURVEILLANCE TEST TO ENSURE PROPER BREAKER CONTROL . O
O ti,Ce:Si m 3aS1S PoR PAIL 0 VERDE 13 NIT 2 RESOLUT10N OF THREE REMAINIE ISS11ES POST ACCIDENT SAMPLING SYSTEM ECCS REAMALYSIS PRESSURIZER AUXILIARY SPRAY SYSTEM CERTIFICATION BY APPLICANT THAT DESliGN, CONSTRUCTION AND TESTING OF UNIT 2 HAVE BEEN COMPLETED IN CONFORMANCE WITH FSAR AN THER DOCXETED COPMITMENTS, O COMPLETION OF POWER ASCENSION TEST PROGRAM FOR UNIT 1 PRIOR T0 INITIAL CRITICALITY OF UNIT 2. I O k_ _ . _ . . . . _.
.~ .
O- .. o SDPLIFIED SOEMATIC 0F PALO WRDE CVCS O SKMNG SOLECES OF DmATED WATER fm AUXILIARY SmAY
- r. n n.iic n i.
MKMtte1 SNR II DU in /,
==ic a m H- %
A m.ea,m,s, nx tan en ionunsnem,as
.xiwr.
i itnt u m mus po-- b -N-b4-awr.nu n,r 3 l CH-ties o ,-sst Om itt ne-1 nm .,u., ;? anmu mu i smiemue neri MM-N- - -M,94-N--- "" >4.' % a'*0d N '"~" , 2,0,,,,,,, O-'"~"' w b _
-N-b4-ai-riw ""'"5a8 y anunc scio o 1
Wi. M*- . i N 'E "/, g'";,*.. maic we O,a,ru-su i
><- e- m r-m en WM N ->4- )
- "c+
mm -p4-d N wo - h . N-o<: :: :
'"M g ,,,.n.. .
l 30 8EP SEAf. i lNEClisisEAl DONada i J 0.ASSIFICATION OF .%)0R (UF0ENTS A31i ! SEISMIC N NTS (IDE CATEGORY
' ENVEONTNTAL
.IE PaER CUALIFICATION
' CHARGING PUNPS (3) III. CLASS'2 I~ A/B YES l CH-501 III. class 2 I A sTP.s-1 CH-536 III. class 3 I"'
,- A are.5-1 2 l
J 4 1
w 2 b O {.3 _ p' _ _ _ __ hh
'9 # s l t(-6 h
b e Y
~ ' -
3 ml
= <
Ng R .; i t_~ .. < a@ a o ni g*j i: ii =w . x:
~
MM 2' u! ? 5@" ----a O #> : j $i'as e i s; # g (i - 3 a y
- 3
.e g m m -- --
- = -d ta en - en en
, d dd m.
.e
~ : : :::
=
M p3 -w--=---==- _ = = = ; A 5. M 8 $ il 55 55 . O
~
^-
VALVE POWER SugY CHANGE . C EXISTING MODIFIED l- lE BUS lE BUS s
) SlAS TRIP ON
)
NON lE MCC IE MCC I I) I) CG l) i) CG 536 501 M M 5 01 536 CHANGE ENSURES OPERABILTY FROM CONTROL ROOM AFTER SIAS AND LOP SUCH THAT SUCTION COULD BE ALIGNED TO RWT FROM VCT. FIGURE 3
4 PRESSIRIZER VENT PATE O c od T^i4 Me KT ATMosNEW J L gc-lo9 gc-loo 3.i g6 g5 Vj w , i' w , itc lob h[x("
, c W +;
4.. T W103 r, gn f2 ys I [ 1I 8C lob
.321 "
O h5%dgtzg g I I I l O ASSIFICATION OF MAJOR COMFONENTS I ASME i SEISilC I I (IMPONENTS CODE I ENVECtNENTAL l CATEGORY t IE POWER CUALIFICATION l i J km. CLASS 21 l l' RC-103 i I I I I A i YES I IRC-105 I I l IIII. CLASS 21 I I B l 1 YES l I i ' ' IRC-106 IIII. CLASS 21 i I I A l l YES i i l IRC-108 i I I IIII CLASS 1] I I B l YES I I 1. I IRC-109 I O i lIII. CLASS l' i I B l i-YES I I i aa _ _ __ _ _ _ - --__11InR/RR.__. .-
b PRE-ACCIDENT IODINE SPIKE (S) THYROID DOSE-REM . 2 HOUR JOSES 8 HOUR DOSE EXCLUS10'i AREA LOW FUPULAl10N B0UNDARY ZONE (REM) (REM) PVNGS FSAR 200 41 APPENDIX 15A AUX-SPRAY AT 1015 SECONDS AUX-SPRAY AT 208 44 TWO HOURS PRESSURIZER 208 44 VENT AT TWO HOURS 10CFR100 DOSE LIMIT IS 300 REM
.GWS 11/05/85 O -
l l
~ . _. .
l l RADIOLOGICAL CONSEQUENCES OF n(E STEAM GENERATOR - RIBE RUPTURE WITH A LOSS OF OFFSITE POWER - l
-O. tuo rutLv SruCr Oetu iov
- APSS OPERATION AT 1015 SECONDS
* (PVNGS FSAR APPENDIX 15A)
~
l
, Location .
Offsite Doses. Rees GIS PIS -
- 1. Exclusion A'rea Boundary -
40 2k)0
- 0-2 hr. Thyroid
'! 2. Low Population Zone Outer 20 41 Boundary 0-8 hr. Thyroid ,- ,
RADIOLOGICAL CONSEQUENCES OF THE STEAM GENERNIOR, RJBE RUPTURE WITH A LOSS OF OFFSITE POWER AND FULLY S'10Cr OPEN ADV l APSS OPERATION DELAYED TWO HOURS
- I l
Offsite Doses. Ree Location GIS PIS 4 e
; 1. Exclusion Area Boundary 42 208 0-2 hr. Thyroid .
- 2. Low Population Zone Outeri 22 44 Boundary 0-8 hr. Thyroid 1 I
i l
- Results for SGTR with a fully stuck open ADV and using the pressurizer vent system are the same.
o ' 11/05/85
9- . O
~
i COMPARISON OF THE~RCS PRESSURE RESPONSE FOLLOWI
. AUX SPRAY INITIATED AT 2 HOURS vs PRESSURE VENT O AT 2 HOURS 2500 ,
i i i i OPERATOR TAKES CONTROL OF PLANT-OP EACH SG 2000 -
\ OPERATOR I ITIATES DEPRESSURIZATION l -jPERATORBEGINSTOTHROTTLEHPSIFLOW.
o , 1500 -f,
/\'
5 m
+
! 1000 -
~$
m t, - E
'./'s s
RCS REACHES'
% SHUTDOWN COOLING 500 - "
ENTRY CONDITIONS
'i.T
l ..,
---- \ -.
0 I I I I - 1 I O ~5.000 10000 15000 20000 25000 30C00 l TIME, SECONDS l SGTR WITH AUX SPRAY AT 2 HOURS , O ---SGTR WITH PRESSURIZER VENT AT 2 HOURS 1 t STEAM GEhT.RATOR TURE RUPTURE WITH 1488
' 0F QFFSITE POWER AND A FDILY STOCK OPEN ATMOSPMERIC DUMP VAZ.VE
O Agenda for ACRS Meeting on November 7, 1985 3:00 p.m. Room 1046, H Street RECENT SIGNIFICANT EVENTS Presenter / Office telephone a _Page Date Plant Event Rosemont Transmitter Improper P. Sears, NRR 2 10/10/85 Maine Yankee Installation (492-7458) 09/19/85 Dresden 3 Scram Solenoid Problem Induced E. Weiss, IE 6 RCS Leakage (492-9005) 7/22/85 Turkey Pt. 3 Problems with Auxiliary R. Baer, IE 15 Feedwater Following Reactor (492-4780) Trip (Followup) D. Mcdonald, NRR 16 (492-7363) 10/29/85 D.C. Cook 2 Reactor Trip Breaker Failure C. E. Rossi, IE No (492-4193) Slides O
MAINE YANKEE - ROSEMONT TRANSMITTER ! O IMPROPER INSTALLATION OCTOBER 10, 1985 (P. M. SEARS, NRR) l PLANT DOWN FOR REFUELING PROBLEM: MAINE YANKEE DISCOVERED THAT PRESSURIZER PRESSURE (4), PRESSURIZER LEVEL (2), AND STEAM GENERATOR LEVEL (12) TRANSMITTERS HAD NOT BEEN INSTALLED IN COMPLETE AGREEMENT WITH THE EQ TEST CONFIGURATION, THE INSTALLATION WAS COMPLETED IN 1982 AND THE TESTS WERE COMPLETED LATER. SAFETY SIGNIFICANCE: A STEAM LINE BREAK COULD HAVE COMPROMISED THE AB0VE SYSTEMS. DISCUSSION: UPON CHECKING THE EQ TEST REPORT, MAINE YANKEE FOUND THE FOLLOWING: g - SOME CONAX SEAL ASSEMBLIES (WHERE LEADS ENTER) WERE V NOT TORQUED UP AS TIGHTLY AS IN THE TEST-CONFIGURATION. SOMEROSEMONT/CONdXSEALCONNECTIONSNOTTORQUEDAS TIGHTLY AS IN THE TEST CONFIGURATION. SOME VENTING ENCLOSURES ON LIMIT SWITCHES NOT SCREWED DOWN AS TIGHTLY AS IN THE TEST CONFIGURATION. NO DRAIN HOLES IN THE BOTTOM OF JUNCTION BOXES. ROSEMONT COVERS NOT TORQUED DOWN.AS TIGHTLY AS IN THE TEST CONFIGURATION. THE COVERS ON ALL CONAX RTDs AND TCs (WHICH WERE ON THE EQUIPMENT QUALIFICATION LIST) WERE NOT TIGHTENED AS TIGHTLY AS IN THE TEST CONFIGURATION. FOLLOWUP: 1 MAINE YANKEE HAS CHECKED FOR LIKE PROBLEMS WITH OTHER QUALIFIED EQUIPMENT AND HAS FOUND NONE. ON SEPTEMBER 10, 1985 J. T. BEARD BRIEFED THIS SUBCOMMITTEE ON THE S/G PRESSURE C) INSTRUMENTATION THAT WAS COMPROMISED. TAKEN TOGETHER THESE EVENTS ARE BEING CONSIDERED BY NRR AS A CANDIDATE FOR AN ABNORMAL OCCURRENCE. - d
?
l .< .-
- O .
i. Manus,aSOUNTING 5AACMET 4
> e -
+
- f ,,
g . - - , - ,. , , - - - - kn counoir antav 3
..- . s , . . .. . . . . .
. \
h Y I- a
- l l
. t 3 4
) :,.. .
I O "'
,ROSEMOUNT liSSb l t
r i i . i - - c LONG / r . - - SHDRf PHiTAILS WIEF) , PIGie4.5 NF) NP1 . pMPY .. TAh j P 4 Il '
- 6. .
- - --I ." , g ' : . II - _ , ,- m
. __wser m -
. -orvics -
rgrovHnu sueesstueLY l O
><a' =aav' u m.ca ecanut.c . 1,h'T/L.tv Me0 LOCK C AP s .
c,o N AX 3 l . L---_..---__ , _ - . , - . _ _ _ _ _ _ _ . _ _ _ . - . _ -_ _ , _ . _ _ . _ . _ _ _ . _ , ._ - ._ _ _ _ _ - __ _-__., -______ .__-
__ _ --------w ,, 9 9
)
6
~
_= Blk
- - 't . /
/ _a_ AP; >
f a. y
&,-w~w. .> ,
am l m as s a p5 E d L q i .. . v 2 t'anQ
,: m-L -
- .-O' .
. vi l l ~~~~
e I (u!CO'd 3
%-ne s q 2
t Y
.,___%,_,, J
I. . s e o "g t ge. . . t-3
~
t $e .4
!U -
1 0 o t
- P i : . 3" 4 t- ! -
- .gH.
g
. - , - '. 2. l p
a O m. . 1 23 y
- j. ,
g.
- - I -
l L
- i, ; , .
s s n -
'{ : ,
\. ',
g 1h l ( _. !: . . _ _{ _g N. s u
.p ___
g y Q .' 8 ,t 1 yJ_; -
$i .
A . tJ' l l .. J .
~2.
0*E
~
O ,, B 'd . tM nth 5NINbtLA DIW016 33*WA
[; l DRESDEN 3 - SCRAM SOLEN 0ID PROBLEM INDUCED RCS LEAKAGE l O SEPTEMBER 19, 1985 (ERIC WEISS, IE) L PROBLEM - RCS LEAKAGE TO REACTOR BUILDING VIA LEAKING SCRAM SOLEN 0 IDS AND SCRAM DISCHARGE VOLUME VENT - AND DRAIN AND INTO VENT SYSTEM. i SAFETY SIGNIFICANCE - INTERMEDIATE AIR PRESSURE CAUSES PROBLEMS WITH SCRAM SYSTEM VALVES. POTENTIAL DEFICIENCY IN MODE SWITCH. SEQUENCE OF EVENTS , REACTOR OPERATING AT 80% POWER TURBINE CONTROL VALVE FAST CLOSURE CAUSED BY INSTRUMENT TECH PRESSURE SPIKE CAUSES SCRAM ON APRM HI OPERATORS ATTEMPT TO RESET SCRAM BUT ONLY CHANNEL A
- O RESETS (RESETTING CHANNEL A CAUSES SCRAM DISCHARGE VOLUME (SDV) VENT AND DRAIN VALVES TO OPEN)
LEAKING SOLEN 0ID VALVES CAUSE LOW AIR RESULTING IN
', PARTIALLY OPEN SCRAM OUTLET VALVE AND SDV VENT AND DRAIN VALVES STEAM IN REACTOR BUILDING AND RAD ALARMS ACTIVATE OPERATORS MANUALLY CLOSED SDV VENT AND DRAIN VALVES 3 INDIVIDUALS CONTAMINATED DURING EVENT l
- CAUSES MODE SWITCH NOT MOVED ALL THE WAY TO " REFUEL" (THIS PREVENTED RESETTING CHANNEL B) l 51 SCRAM SOLEN 0 IDS LEAKED (LEAKAGE PATH WOULD HAVE
- l. BEEN BLOCKED IF CHANNEL B HAD RESET) i O
6
2- ' O
- LEAKAGE PATH .
i FROM VESSEL THRU LEAKING SCRAM SOLEN 0ID VALVES T0 , j SCRAM DISCHARGE INSTRUMENT VOLUME LEAK LASTED FOR 23 MINUTES I i FOLLOWUP ACTION
- 1 4
PLANT SHUTDOWN 4 - LICENSEE EVALUATING PROBLEM AND REPLACING SEL"r.TED - VALVES REGION FOLLOWING 1 i e O < s O 1 1 t l 1 1 4 O R 7 a w w-- y N -, re-=~' W w w w- , nw---newm-~,~nm----w-n,- t e -e *ww-e-,- - s---~---~*-+s-*---r---~w~e~e= - - - - - m-*---s -
-*m-- =---* * *--
e 3 O DRESDEN 3 - SCRAM SOLEN 0ID PROBLEM 'l'0UCED LEAKAGE - FOLLOWUP SEPTEPBER 19,1985 (ERIC WEISS) PROBLEM RCS LEAKAGE TO REACTOR BUILDING DUE TO LEAKING SCRAM SOLEN 0 IDS. LEAKAGE PATH FROM SCRAM OUTLET VALVES TO SCRAM DISCHARGE V0 Life VENT AND DRAIN VALVES, . SAFETY SIGNIFICANCE DRESDEN PROBLEM PAY BE GENERIC TO BWR'S i PRELIMINARY CONCLUSIONS MOST BWR'S WILL HAVE A LEAK OF REACTOR WATER OUTSIDE DRYWELL IF: 1 THEY HAVE A HALF SCRAM CONDITION (A CH RESEF a B CH TRIPPED) AND O 2 THE HALF SCRAM FOLLOWS A FULL SCRAM THE SOURCE OF THE LEAK IS NOT OBVIOUS TO OPERATORS UNLESS TRAINED TO RECOGNIZE THE PROBLEM THE LEAK IS EASILY TERMINATED AT DESDEN WITH MANUAL CONTROL OF VENT AND DRAIN VALVES THE EXHAUST DIAPHRAM IN THE 117 VALVE IS NOT HOLDING PRESSUE WHEN AIR HEADER PESSUE IS SLOLY INCREASED THE LEAK WILL OCCUR WITH EWLY REBUILT VALVES. ALTHOUGH TESTS IN SITU SUGGEST THAT SYSTEM DESIGN IS ESPONSIBLE FOR THE LEAK OF REACTOR WATER, INDIVIDUAL BENCH TEST SUGGEST VALVE CONDITION LEAK t0E LIKELY, PAINTENANCE ECORDS SHOW THAT SCE OF THE WORST LEAKING VALVES WERE REBUILT RELATIVELY RECENTLY FOLLOWUP IE IS PURSUING MATTER WITH GENERAL ELECTRIC Q AN INF0PFATION NOTICE IS IN PREPARATION 8 1
1,lf fj fl f O hA f. r. 7 m,U V p S . A T E M u* L T AO HN o L O U O .ST" C A*
- l~* A" 6 D .EN~
N .LE o
- T" N
O Ca 0 2 A T* t
.iiV WE ,i to.
t s . 0 j E5 M ,noe
,octc _
f i gP LEESU' N ' NV OL P ,sieetn ci ns-r D E I LLOS e
. MACVEO AV V
- ,vis n
. 4 - A OELO ,oots
,s nn
- S P RNTCAT CO ,sneamio R SIRRVRTI A a
,utst r^
I E AAl N A i la .
. T LECAE aI ,tnns
,cooo c N T arcr 2 E V
O0nISnR N41AODA a , . a
~ a. ..-
. a .
/
7 u' ~ s
/erM < -
~ .- .
<M .
. [vU 7 -
4 "' T N
- wf** >-
T E N M 4 [ E - V E l yN, G Ag P p 3 N
. 5 V *
,O $. a 3
Cg ArA T N E r o t EA JR FR j r u" T o I A S O5 V E N R L MIE M E . f AO V U EE MM mJN V m' CsSLV O RNL CEA L O UU RL MLA s Aen V TO SV -
- S E SV 7 G R IN T M 3
8 - A RE A H ELS R C lTE t C L iUV OOL S D MM A
- M OA V A nn
. R FC _ C M s SA3An l WVCS 4 d 4 P 6 D n - C 0 A M T
- A ( A E c g]yRL r '
r CN - l-St h f Q- - e B G 9 - N I R t CE 02 RTA A 8 8 8 A Us t iM l O - C 2 j : rc o n i n A g ( : N A n c 4 ll lI (li' ( ;!'[; I lr
,i ,
hk ~ f.f - e I. O $
/ -
o, T' .s E M L I AD e RN
'n L
o Ur O i. C STA
- n D EN l- TM N LE -
e NO A sV r e OO - T awe
". j/* C,D
- a j, E5 M '
e
- il S
r p LEE U* N x d. RM INV5OL( L LO5 P E
*6 e
MACVVO E 8 4 b*
~
. In A AV 0E LO 5_ nN1 GAT O .
P CO JF y, C p R 5s aRVR Ts A
" f, . T E: AAla INAI LEcAE g,
T gAA , N E V T O OniRR NIS ADDA s ' l N
. 'I E ,
V - " d -
, a *
/
y, ' . s
.-, c n
/c< d a
= N .
[ef U (.# T T N ' T N . E e E N
., V /. f E 1j M V E g# P g G 0NA O. EE 1 j
. C 5' T T =
We ., e AP N k D itR g o R D5 V E J E T E 8 fRA
~
k i
=
f*
?
4 4E aOV A R NL CEA M U L N EE MM UU sC J 8 I. E L WLV CS SLV O IA A V RL
- AP A O
S TO V -
** .E SV C N SM R I T A RE A H ELS R C iT t
E C 15 TU OOL V S 0 MM A M A OAV T HH R FC
~ C ME S
. ~ hS ALI D U
[ C SO 7
~
- D n
M C A R a ( A M T E O . f l
*>" CSIN RL r . .-
C Q- - S G 9 -
- N I R GET 0 s R
AA 82 1 8 t _ L W O O 2 s : 2n c m c n " y : N A ii C O _.. [. i ?
d-x oV w .
/ T E M L AD U R N*
. T -
' CA* -
O ST
" 8 s *4" O .EN .
N .LE o 0 A es V so es e
.
- 2 i .
,vs
., A: . 8 TE '5 WE M ,oou
,oac l
,5P , LEESU' N c a.
S M
" R NV I L OL' PE LOS
,sae
..e
,s.s s
R MACVE AV VO ,
,oots v
,i A OELO . g ,a e sn a &. ;
S P RNTGAT CO ,vsawe
,a aa _
c
.,. R SI RRVR ,ur i
- TI A IA s g
,sans E AAllN ,arcr i.c ,. ,,
s AFTN N E T LECAE O0R4 5 I RR
,cooo * '
3 V N$ 5 ADOA T . N ' N E V b* - ."
?
s -
- a
- e
- y. '
t . r .
/c _
. [esN # -
f , T - 4 N w#* T N E M
. /f E V ) E J.
f G v N u- *A O.
$ T
! A CS AP T
N D E fRR R I L A s m E T e
~.
' M,D RE 5 V E M N EE
.~2[ R
) E m A OVNL U MM 1_~ j rV m' 4RCEA L O UU nLiA C5 3LVO RL t S V
. V -
AP TO
' b R S E SV G N T M
. R A
8 RE A t ELS R C i tite C 5 4 TUV OOL S D MM A
- M OA V A RR R T FC C ME S ALI R
A CU SO f e E s - - e , R O AV
,' AM
. T E .
- RL r g]F CN SS A .
,* S P .
- t . - _ P l G N" t-e Ce 8 O2 S w b 4 S GUho R" A* t s e L O a c M C G N h G 8 C 6 O N 2 J L J
>gt NlA l
l C
. ~
N
, , I ! ) I ll i < j/
r Y" o' W.- 7 d
~
*/ 7 7,
4
/ T .
E L'.MA D i
' T ~~ nN*
n
' U CA O -
" ST'
/n.
~
e O EN N- LE *
'o.
cO 0 A HV' e sn
'o x
vYc
- d.
>l s
u ps e r ' T E5 NVOL I L l'WE LEE SU M LOSE P N E g i ou
- n. 'Tc nc o?a n'
i n MACV VO $ a'.s . be a AV y t ReOELO s'#ts
- r. o e s r CDTCAT ( ta C n S4RRVA s
=' ao s f.a
,. . :1I s
AAl ANI A s. 'V sss I y ie f, c g,,,, T N E T LECAE O0R45RR l
.soo'~
. :rr T .
E V 1 N$ A0OA y ~ N E ? v ~
~
s .
, s s
j, . ' s
, r '
/e<
U ~
. [ef U' ~
T 4, N . E
- l. f' ^= g. .
M
.- . y y E o.
G
. Ohb g
v g N f # CS y T oA rR
- AP m D t M__ " B g y T E
8 rR iA~ 6 . E N n iE v *. t 1, M.3 AO S R N'
'" M U Et Mu ej pV
.~
1 CS S L CE* L O Uu Rt nLA AP R O* V TO o V
. - S E G
SV N S M I T f" .
. R A
H C RE ELS 6 lT TU V E R A C p' 3 D 4 OOL S MM A M OAV e A R T Rn FC L h C ME ALi R S f Cu SO t e k; s wIs s ~ e . t D s e ?. 'g# n C R -
~
[t . I 0 .
. & AE A 6 . l
, 7M AL CN SG r
5 c. Q- -
.Isl n
nn a, ,, o
!' Nn e i cE 02 h L ,
, nT A*
' 8 c9 l
n* O C C
, s : yi G
c N i O u ( : Nna ia C S
\P 4 .
o O- O
; wrom WEST ttCU*e C an TT *
" " " - - "" ; ; ;h;..
s ' fin
,: ;h ..
,,.m i #Vn
<r 4 r < r
, : ,, qr ir WEST SCRAM DISCH VOLUM J N k "
N N RWC.O ; r S HX (AST SCRAM DISCH VOLUME I [D ROOM ROOM [) i I I I Scr m Scram Scram Scram WEST EAST 8CPAM ( g, , go,,, , ( go, , ( go,,,, SCRAM INSTRUMENT INSTRUMENT VOLUME [ LE T LE T f LE ( LE }j ( Alarm VOLUME ( Alarm J [ Rod Block) ( (Rod Bloo l I . I I i RBEDT RBEDT NOTE: ALL VALVE 8 are shown p p la the normal UNSCRAMMED poellione M 8 CRAM 4- 4.- : 4- ' 8 f- INLET OUTLET 4 #[\IV i 640Ah " y 4: > t : ' ,: : : : : : : : : : :': : : : : : ':
^
l G --l ,
' 8ACMUP 8 CRAM RPS A
l I SCRAM r -208 --n8-$ -117 Q Q 54981'd 4: I : A DUMP: NE NE gj , m) , VALVE 8; P8 8 ' ' ' 8ACKUP .. -30A -118 ARI ^ PS PI VENT VENT CRD FILTER VALVE 8 VALVES -]g i r SCRAM
, 12 S v do , 3' " Pl
,ND12Svdo,. gD Vg g gINSTRUMEN - ND . ;ND .
.. ND AIR pE #hp# .Eh # ,E # #
# # 48Dh 180bT N' ' ~
6248 624A ISS ISA TEST VALVE - ~ To o er PRV 't y 8488 r 1r , i
, , y HYDRAULIC CONTROL UHF
. -r,- -
2 548A
' , c :2 r,-
1 N Figurel. Scram Discharge Volume (SDV) Q ,
. 4 l
. -~.s -- ** * *. * ^
I . O . SUPPLY PRESSURE P l P J d Iy SCRAM PiLOTVALVE { 30 e lis PRESSURE ,
.. OfAPHRAGM . i -
a RPS CHANNEL g P1 8 Q' Imkag,) f EXHAUST DIAPHRAGM E1
._ q i .
7* k 1 6 F ( 1 i 2~h
- f. *.
C Q:'t 7
. -Colt D_
E O . EXHAUST PRESSURE E PRESSURE , DIAPHRAGM P2 -
' [
SCRAM PILOT VALVE
# NC 305-117
" L y%
RPS CHANNEL A
. v 3 m.m EXHAUST DIAPHRAGM E2
'k N \
y .L 1 . z% 33 y, '
~ coln.
. t k., . ~
,i ,
\
SCRAM INLET
/ \ /
SCRAM OUTLET 1 VALVE DIAPHRAGM VALVE DIAPHRAGM Q Figure .2 serem Pi(or Valve Configuration 1 with 4 RPS Chennel Reset and B RP5 Channel Tripped Y
m i Robsrt Bs:r . TURKEY POINT AFW SYSTEM PROBLEM C l l
SUMMARY
OF EVENT
- Unit 3 Tripped from Full Power on July 21, 1985
- High level in SG caused MFW pumps to trip
- All 3 steam driven AFW pumps started and ran. One pump had problems with air operated AFW control valve MFW Pump restarted. AFW pumps shutoff
- AFW pump turbine governors reset locally. Steam supply valves to one turbine placed in off-normal configuration to facilitate trouble shooting of flow control valve.
- High SG 1evel occurred again with MFW flow
- AFW pumps restarted. Two tripped on overspeed. The AFW pump with the off-normal steam valve configuration ran, but flow oscilated Q Basic Cause Governor oil not drained properly when AFW turbines reset locally.
*
- Procedure riot sufficiently explicit ,
Generic Actions
- Have obtained more detailed information from Terry Turbine
- Plan to prepare Information Notice O
/C
[ DANIEL MCDONALD NRR-PM TuRxey e0 INT etANT O AUXILIARY FEEDWATER SYSTEM ORIGINAL SYSTEM AND DESIGN BASIS CURRENT DESIGN BASIS SUPPLY 373 GPM TO SINGLE UNIT WITH AC POWER AVAILABLE SUPPLY 286 GPM TO EACH UNIT WITH LOSS OF AC POWER SYSTEM MODIFICATIONS - POST TMI AUTOMATIC SYSTEM INITIATION REDUNDANT FLOW INDICATION REDUNDANT LEVEL ALARMS ON THE CSTS TWO SEPARATE FULL FLOW CAPACITY AFW TRAINS SUPPLIES TO FLOW CONTROL VALVES REDUNDANT N2 DIVERSE POWER SUPPLIES TO FLOW INLET MOVS (AC 8. DC) , HIGH PRESSURE TURBINES AND TRIP / THROTTLE' VALVES O
- NON-SAFETY SG FEED PUMPS TWO 850 GPM PUMPS 300,000 GALLON DWST t
AFW EVENT FOLLOW-UP ACTIONS - NRC STAFF IMMEDIATE CORRECTIVE ACTIONS DEVELOPING IE INFORMATION NOTICE FPL COMMITMENT TO TEC,H. SPEC. THE NON-SAFETY SG FEEDPUMPS. SURVEY OF FPL MAINTENA'NCE PROGRAM BY NRR TEAM PROPOSED GENERIC LETTER REQUIRING UPGRADE AT SOME PLANTS. LETTERS TO THE COMMISSION AND ACRS AFW EVENT FOLLOW-UP ACTIONS - LICENSEE IMMEDIATE PROCEDURE CHANGES INTERIM AND LONG TERM CORRECTIVE ACTIONS O - SPECIAL lE TEAM INSPECTION OF AFW SYSTEM
/6
AUXILIARY FEEDWATER SYSTEM O PRESENT DESIGN IWO SEPERATE'. FULL FLOW CAPACITY AFW TRAINS. CAPABLE OF SUPPLYING 373 GPM TO ONE UNIT FOR LOSS OF NORMAL FEEDWATER. CAPABLE OF SUPPLYING 286 GPM TO EACH OF TWO UNITS FOR LOSS OF OFFGITE POWER (RCP'S STOPPED).
- EACH UNIT MAINTAINS A SUPPLY OF 185,000 GALLONS OF DEMINERALIZED WATER IN ITS CONDENSATE STORAGE TANK.
SYSTEM AUTOMATICALLY INITIATES ON THE FOLLOWING z SIGNALS: _, O~ 1. SAFETY INoECTION
- 2. t 0W-LOW LEVEL IN ANY OF THE THREE STEAM GENERATORS
. 3. BUS STRIPPING ll . LOSSOFBOTH'FEEbWATERPUMPSUNDERNORMALOPERATING tbNDITIONS.
SYSTEM IS ' DESIGNED TO REMAIN FUNCTIONAL FOLLOWING A SAFE SHUTDOWN EARTHQUAKE AND MEET MISSILE PROTECTION CRITERIA. ' ')
'e fo '
17
AUXILIARY FEEDWATER SYSTEM
, O CuanenT ensinesRise ea0aeCTS
. REA-TPN 85-30 AUXILIARY FEEDWATER SYSTEM DESIGN REVIEW AND OPTIMIZATION
- 1. MINIMIZE AFW FLOW CONTROL OSCILLATION
- 2. AFW TECHNICAL SPECIFICATION REVISION
- 3. AFW GOVERNOR PERFORMANCE
- 4. UPGRADE OF AFW FLOW CONTROL VALVE NITROGEN SUPPLY LINES
! 5. AFW T&T VALVE SWITCHES AND LOCAL INDICATING LIGHTS
- 6. DESIGN BASIS DOCUMENT :
- 7. AFW-NITROGEN BACKUP SYSTEM
- 8. AFW SYSTEM ACCESSIBli.ITY
- 9. AFW SYSTEM STEAM LEAKOFFS
- 10. AFW-AFP RPM MONITORING
- 11. AFW-ELECTRICAL CONDUIT REROUTE
- 2. AFW-REMOVAL OF OBSOLETE EQUIPMENT O 13. SINGLE FAILURE ANALYSIS
- 14. MISALIGNMENT OF AFW
- 15. AFW-AUXILIARY OIL PUNP
- 16. AUXILIARY FEEDWATER SUPPLY FROM THE DWST L O
., g l
i i i i i i c 1 : : : -
<. . I l!{I/ - i{ 1{
! ! i '$
, I/.
1/. f3 ! !.
~'
j fg E! OE:! 6:! OE:16:! G:1 'M:! OE:! 05:1~ 4:1 GE:! OE:i i! y .i. .
.t.
.i. .
[y.i. .
.i. .
.i. .
lj,!- a5 ** ' IX !X g I{ !{ !{ !{ 1l - t J . l l : l l l l l l . m .. '
. ,, ,,, , J ' '.
- u}_
l db [ db l:
- db , ,
g r D ' O ' l [
'I :: h,
* [:,,
h 5 r' ! '
@I! o. . 63. lit -
,,,112 .-
d;112{ilti p!z 1%, O
;'% M' of-l 1
i X
! l ,
% ::,T l-il a t- o 1-
- . l l ll
-11I,1 T--!i -
T---64 3f,f,", , hij,5 0~ - g $$ I I - 4 11 /t
,i 1 i
-lill IX !X !X !X . !X IX
- :: . i rEl rE E 1 pEl rMT PEE! i g
Ilt iis ilt dit [-M:! mit fit I I I I ! ! 11li I i ll!i i * ;.
.j - ,i
- g g e -
l ;
'IIII! !Illi is f
* ; ri ; -
,is 8 $j1 El lh'i . i
!* li - :
!1 !XS EX5 yh !* !X !X' yg I
- h 3
[ 8 1j -> un j i t l t i i 3
- t. i i .
l l. H!r! 7 !E! =! H3'I' g !! H:! - H:! H:!
, , q., ti .i n . .I t! .i . t! .g.. ti ,!. . :l
!'i; f If nd t1 > ! 's ti !p!!r.. ia i
j i! 19 !! ks ut. ti . l
. i, n - ii !.!. -
1a If
- ) i' . . :
i ! : Q!:
- i
= -
p3 1 i [, , % r !
""1 ,
o I g g . i
-l
! = !0 is !p !!) i::= !fs10!p!!** t :
<, e, 9, I
s f ( g: ! a t b I- i ( I
.T
^
1:
.T
[4 !: v f4 1 'lpI " Q! j I !; , 3 ," !: T!]
!: T -
if
.it I! ! !! i i si If 11 ft 7
i . .I I. 7 i !!!j,J jf; l): 1 10 i I::= 3tip 1::= 1 10 ! l 1 5 !. fil ! II= E X' !::= ! iO !. I j I i I I!: :. i : kl ,l% .!Ill g .i. I ,f O O.F$
'l I'@ @!vrf a
d .
.. it El g; ,V. , tit / >I p 1
, 1 i 11 ,, .
1i e.= m} ti>!! . it t! (! w! (
!3: !3, !3, !3, !3 !33-A E-::1 E-::! E4! E4! E4! E4! :
(J 1p,:, Ep> Ep !p !p !p 61 l i
l Y
, k UNITED STATES NUCLEAR REGULATORY COMMISSION 5
h WASMHeGTON, D. C. 20586 OCT 2 9 Im MEMORANDUM FOR: Chainnan Palladino FROM: . William J. Dircks . Executive Director for Operations
SUBJECT:
SEPTEMBER 20, 1985 LETTER FROM ACRS ON TURKEY I'0 INT This is in response to the questions in your October 16, 1985 letter. (a) What actions the staff has underway with respect to the ACRS s letter? . Response: A reply has been sent to the ACRS, copy enclosed. ' (b) What priority and staff resources have been assigned to explore' the deeper implications of the event? -
..- Response: Innediate staff resources were assigned to explore the
~
deeper implications of the July 22 Turkey Point event within O a matter of days following the event. ACRS was infonned of the special inspection discussed in our enclosed responses - on September 13, the day that the exit interview was held. I note that the ACRS letter is dated one week after that. The actions taken are described in the enclosed letter to the ACRS. High priority and staff resources continued to be assigned in our evaluation of generic implications of the Davis-Besse AFW event as they may apply not only to Turkey Point but also to all pressurized water reactors. (c) When a report to the Connission and a response to the ACRS will
~
be available? Response: The enclosed response outlines the actions already taken and those still underway. I have requested a meeting with the President and Chief Executive Officer of Florida Power and Light to examine whether sufficient' management attention has been given to improve overall nuclear performance at, Turkey Point. CONTACT: J. D. Neighbors, NRR 492-4837 - O j l W l
Chairman Palladino Our recomeadation of the short term actions to be implemented regarding 'JW systems as a result of the Davis-Besse event will be availat,le by the end of this year; recomendations for the longer trrm actions by mid-1986. gapes W811am L Dints William J. Dircks Executive Director for Operations i Enclosure- 1 As stated l s cc: Comissioner Roberts -
~
Comissioner Asselstine . Comissioner Bernthal ' Comissioner Zech OGC OPE - -
, . SECY O
e MG 2
~. .
%Sa 9
O 22
4' p me.g EHCLOSURE 9, UNITED STATES
! o NUCLEAR REGULATORY COMMISSION
-@ I wasumaron, o.c.zossa Q
% ,,, , October 28, 1985 Mr. David A. Ward Chainnan
, Advisory Cormittee on Reactor Safeguards U.S. Nuclear Regulatory Comission Washington, DC 20555 i
Dear Mr. Ward:
1 I have been requested to respond to your letter to Chaiman Palladino dated September 20, 1985, in which you expressed the ACRS concerns relating to the recent post-trip failures in the auxiliary feedwater system (AFWS) at the .
- Turkey Point Plant Unit 3. The Comission and staff share your concerns and have placed a high priority on detemining the implications of the Turkey Point event. g s .
The auxiliary feedwater failNsonJul 22 1985 were reviewed by the staff imediately fo owingtheevent.:hes,ignificanceofthe' failures was recognized by the staff in reviewing the' licensee's restart plans and in terms of the generic implications of the event. Follow-up discussions were ' held between the staff and the licensee to detemine the cause of the failures and the need for corrective actions. These failures were broug1t to the attention of NRC senior management at the Operating Reactors Events ' Briefing of August 13, 1985. In order to assure that other licensees are awa're of this comon cause failure potential, an IE Information Notice, is being developed.
-Other staff actions related to the Turkey Point AFW system specifically and to AFW systems generally are:
. A special first-of-a-kind inspection was organized by the Office
~
of Inspection and Enforcement during the weeks of August 26-30 and Septembar 9-13, 1985 addressing the operational r=adiness of the Turkey Point AFW system. This special inspection was an NRC _" ~
. inspection initiative resulting from the June 9,1985 event at the Davis-Besse plant. Turkey Point was selected as the initial plant for this special inspection because of its recent AFW events. We are pursuing with Turkey Point the resolution of our specific concerns as outlined in our inspection report; however, the findings resulting from this special inspection will be included in the areas the staff has identified as necessary to pursue in evaluating potential generic short term and long term actions as a result of the June Davis Besse event.
In addition to the evaluation of generic short tenn'and long term actions as a result of the June Davis-Besse event, the staff is reviewing past and present requirements for AFW systems and evaluating the need for further action for those plants where the reliability of the AFW system should be improved. The staff O expects to submit a proposed generic letter to CRGR for review in early November which would require AFW modifications at some plants. 23
Q Mr. David A. Ward October 28, 1985 The team which perfomed the special ins during Perfomance Appraisal Team (PAT) pection and inspections utilized techniques Integrated employed Design Inspection (IDI) efforts to assess both the design changes and the other functional areas which could affect the operational readiness of the AFWS. The Inspection Report (50-250/85-32; 50-251/85-32, " Safety System Functional Inspection Report") was issued on October 7,1985, in which the staff addressed deficiencies in maintenance, surveillance testing, and design changes and modifications. Copies of the report have been sent to the ACRS. The licensee has already responded to our concerns relating to procedures and training and has comitted to provide Technical Specification requirements to assure operability of the two 100% capacity motor driven non-safety grade AFW pumps. These pumps are routinely used for start-up and shut-down and, if off-site power is lost, can be powered by any of the five non-safety diesel generators on site. The licensee has been requested to respond.to ' - all the items in the report and to provide details'of additional actio'ns to ' be taken to resolve our concerns. This matter has been referred to.the Chief Executive Officer of Florida Power and Light for his personal participation in resolving our concerns at Turkey Point. Inaddition,asurveyofthemaintenanceprogramatTurkeyPointwillb5
'.. carried out within the framework of the Maintenance and Surveillance Program in NRR. The survey is scheduled the week of December 2, 1985. The objective O of the survey is to review the overaii practices and orsaaizatioa of the maintenance programs. The survey will examine the organization and .
administration of the maintenance department, facilities and equipment. *
' technical procedures, personnel, and work control policies. Four such l surveys have already been completed at other plants. An identical protocol, l developed in the Maintenance and Surveillance Program, is being used for all site surveys. A copy of the Turkey Point sde survey report will be available in mid-January 1986.
~
We will continue to keep the comittee infomed of our efforts relating to i enhancement and improvement of AFW systems to provide increased assurance that 4 hey.will be available and capable of perfoming their critical safety function when needed.
'> Sincerely.
l w Harold R. Denton, Director . Office of Nuclear Reactor Regulation O 24 l , u
- - . . _ _}}