ML20198E330
| ML20198E330 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde, 05000000, 05000470 |
| Issue date: | 11/05/1985 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1465, NUDOCS 8511130299 | |
| Download: ML20198E330 (404) | |
Text
-
ORIGINAL OMSF&&5 q'-
UhllED STATES NUCLEAR REGULATORY COMMISSION
. i IN THE MATTER OF: DOCKET NO: ..
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON CESSAR/PALO VERDE NUCLEAR GENERATING STATION l RFA (E. 1 N}p) 6 @
o V .
LOCATION: WASHINGTON, D. C. PAGES: 1 - 326
?
DATE* TUESDAY, NOVEMBER 5, 1985 l a 30 XOi30m0Veir0m ACRS c ACE-FEDERAL REPORTERS, INC. #()'4J Uc qY hQ Official Reporters 444 North CapitolStreet 6
C),$. ]
Washington, D.C. 20001 B511130299 851105 (3 2) 347-3700 PDR ACRS PDR '
T-1465 NATIONWIDE COVERACE
CR24925.0 1 ;
DAV/0jg j I
, -y UNITED STATES OF AMERICA )
't
~ '
i 2
NUCLEAR REGULATORY COMMISSION 3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS l
,4 SUBCOMMITTEE ON 3
CESSAR/PALO VEPDE NUCLEAR GENERATING STATION 0 Nuclear Regulatory Commission Room 1046 7 1717 H Street, N.W.
Washington, D. C.
8 9
Tuesday, November 5, 1985 10 The subcommittee met at 8:30 a.m., Mr. Jesse C.
11 Ebersole presiding.
12 f_ ACRS MEMBERS PRESENT:
(/ 13 MR. JESSE C. EBERSOLE, ACRS Member MR. CARLYLE MICHELSON, ACRS Member 15 l
MR. GLENN A. REED, ACRS Member !
MR. DAVID A. WARD, ACRS Member MR. CHARLES J. WYLIE, ACRS Member 18 19 20 21 22 23 24 Reporters, Inc.
25
. a
-( ) PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSIONERS' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS TUESDAY, NOVEMBER 5, 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
l i
i l
l i
9250 01 01 2 1 PROCEEDINGS jggDAVbur 2 MR. EBERSOLE: (presiding) This meeting will now 3 come to order.
4 This is a meeting of the Advisory Committee on 5 P2 actor Safeguards, the combined meeting on the CE plants 6 and Palo' Verde.
7 I am Jesse Ebersole, Subcommittee Chairman of the 8 Palo Verde Subcommittee. Mr. Wylie will act as Chairman 9 during the discussion of combustion engineering plants.
10 The other ACRS members in attendance are 11 Mr. Wylie, Mr. Glenn Reed, Mr. David Ward, and I hope 12 l Mr. Carlyle Michelson will show up, m
13 The focus of this meeting is one to review the 14 startup test experience of the Palo Verde Unit 1, including 15 'recent trend events, and discuss the rapid depressurization 16 of the CE reactors, given recent Operating experience with 17 the repressurizer spray system on Unit 1.
18 Mr. Dean Houston, Mr. Paul Boehnert are the 19 designated ACRS staff members for this meeting.
20 The rules for participation in today's meeting 21 have been announced as part of the notice of this meeting, 22 previously published in the Federal Register on Wednesday, 23 October 30, 1985.
24 A transcript of the meeting is being kept and ,
) 25 will be made available as stated in the Federal Register 1
I l
l ACE-FEDERAL REPORTERS, INC. l Nationunde Coverage 800-336-6640 l 201-347 3700
p$2900102 3 ,
gAvbur. 1 notice.
2~ It is reques~ted that each speaker first identify 3 himself or herself and speak with sufficient clarity and
-4 volume so'that he or she'can be readily heard.
5- We have received no written comments from members 6 of the public and no requests for time to make oral comments 7 from m6mbers of the public either.
3 I have a few observations to make, and then 9 Mr. Wylie has also.
10 I am sure that all of you realize that the roots 11 of concern for this plant and this design is based on the 12 inability to directly depr6ssurize the primary coolant l() 13 syste.m and so obtain cooling by direct low pressure flow 14 across the core.
15 To compensate for this, a high numerical 16 reliability index has been created for the aux feedwater 17 system, although this is a rather standard design, and a Ic . modest increas9 in reliability of the secondary system l.9 blowdown is provided by qualifying the secondary system t
20 PORVs.
21 The primary system sprays have already given some 22 trouble at this plant. Of course, this is the normal 23 routine way of controlling primary system pressure. -
24 There has also been some trouble with the
() 25 multiplexer system, a system which is of course designed ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 9 6
, ,_ _ . - . _ _ . . __-__.,_,, . _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ . . _ _ _ . . . _ _ _ _ _ _ . . ~
l l
l l
1925010lf03_. 4
.to save wiring and to increase reliability. At that point I gAVbur ~1 2 am going to have Mr. Wylie make a few observations about the 3 multiplexer system.
4 MR. WYLIE: I note that this afternoon there is a 5 presentation on the switchyard and the control. I assume 6 during that presentation that the modification or whatever 7 to the control system to prevent the reoccurrence of what 8 occurred previously will be discussed and how this meets the 19 General Design Criterion 17.
1 10 I think that is about it.
I 11 MR. EBERSOLE: As I recall, the multiplexer 12 system has been used elsewhere in the plant. I seem to.
13 recall one that has to do with the interactive working of
({))
14 the coupling to the ultimate heat sink.
15 Ue will now proceed with the meeting. I will 16 call on Mr. George Knighton of NRR to start.
17 I would like to also comment this way. I am 18 introducing Mr. Kniyhton with the observation that NRC 19 rarely, if ever, challenges the concept, which is the root 20 issue here, but invokes various what I call patches to 21 whatever is of fered to their satisfaction. In a way, this 22 meeting is somewhat similar to that that we might have on a 23 Babcock & Wilcox plant, which has had some fundamental 24 problems because of the boiler design.
()
J25 Mr. Knighton, the meeting is yours.
i ACE-FEDERAL REPORTERS, INC.
202-347 3700 Natiotwide Coverage 800-336-6646
_--_ _ _ _ .____- ._ ~ - _ - - . _ _ _ _ _ -.- _ __ _
N
- 9250.01 04 5 qlgpAVbur 'l MR. KNIGHTON: Mr. Chairman, I would like to 2 Lthank the committee for meeting with us today.
-3 As I mentioned at your planning session last 4 month, it appears that we would be ready to license this 5 plant sometime during the month of November. We would 6 appreciate any comments from the ACRS or any concerns that 7 -they have prior to issuance of any license. So these 8 concerns will be addressed.
9 MR. EBERSOLE: Mr. Knighton, do we have an 10 accurate, updated, realistic startup time for this plant 11 that can be depended on?
12 MR. KNIGHTON: I think we are going to be talking
.() 13 about that.
14 MR. EBERSOLE: Okay. :
15 MR. KNIGHTON: I will have speakiny today Ermando 16 Licitra, who is- the project manager for Unit 1.
- 17 Dr. Wayne House, who is not here at the present
-18 time. He is AD with the Division of Safety and Systems 19 Interaction.
20 Jesse Crews of Region 5 to speak to the startup 21 experience.
22 Gary Ficrelli is here, the resident inspector.
23 With that I will turn it over.
24 MR. EBERSOLE: We have one little interruption.
() 25 I failed to recognize Mr. Reed and Mr. Ward. It is their ACE-FEDERAL REPORTERS, INC.
ll02 347-3700 Nationwide Coverage 800 3364646
9250 01 05 6
]lgDAVbur 1 prerogative to make comments here at this time.
2 Gentlemen?
3 MR. REED: We will surely do later. I think I 4 will wait to hear some comments.
5 MR. EBERSOLE : Thank you.
6 MR. LICITRA: I presume you can all hear me.
7 My name is Manny Licitra. I am the NRC licensing 8 project manager for Palo Verde Unit 1.
9 What I will be discussing at this time is the 10 item on the agenda that is called Palo Verde Review. In 11 fact, there are two separate entries under that heading.
12 What I plan to do is really discuss them as a package
]) 13 because I think they are interrelated.
14 The Palo Verde Unit 2 plant is really e duplicate 15 of Palo Verde Units 1 and 3. The three units were reviewed 16 as a package. An FSAR was submitted for all three units, 17 which incitaded the CESSAR FSAR.
18 We had supplements written, an SER and 19 supplements written, which provided a basis for licensing 20 Unit 1 and in fact, except for some minor exceptions, 21 provides the basis for licensing Units 2 and 3.
22 However, there are certain specific issues that 23 also needed to be addressed on Palo Verde Units 2 and 3.
24 They are being covered by a subsequent supplement,
() 25 Supplement No. 9.
ACE-FEDER AL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33(H5646
m 9250'01106 7
.[v DAVbur 1 I think at-this point I should put the viewgraph
~
-2 : on.
3' (Slide.)
4 At this stage of the-game there really remains 5 only three issues to resolve for the low power licensing of 6' Palo Verde Unit 2. They are the post-accident sampling 7- system,- ECCS reanalysis, and pressurized auxiliary spray 8- system.
9 In addition to the resolution of issues prior to 10 issuing the license, we do require from the applicant a
'll certification letter that the design, construction, and 12 testing of Palo Verde Unit 2 has -l>een completed in
() 13 conformance with.the FSAR and other document commitments.
14 MR. EBERSOLE: Mr. Licitra, I notice down there 15 you refer " prior to initial criticality of Unit No. 2." I 16 would like to ask you to defend the notion that you have to 17 stop criticality when in fact low power operation is a 18 nonhazardous mode of operation and would give the people 19 substantial flexibility without any significant risk at
-20 all.
21 I never have been able to appreciate identifying 22 initial criticality as a mode of operation that had any 23 potential for any trouble anyway except radioactivity 24 experiences.
() 25 MR. LICITRA: You are addressing the third point k ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverase 800 33H646
9250 '1 0 07 8 vbur l' on the graph?
2 MR. EBERSOLE: Right. I don' t see it as a 3 significant point in the evolution of the startup.
4 MR. LICITRA: It is j ust that we want to make 5 sure that the design of the plant is well understood during 6 the completion of the qual ascension program for Unit 1.
7 MR. EBERSOLE: 'At worrt you can just dirty up the 8 fuel a little bit and it has not really -- it has little, if 9 any, hazard potential up to low power. That is all I want 10 to do, is to challenge the reality. !
11 MR. REED: I would like to support Jesse's 12 comment. I think that criticality over the last 25 years
() 13 has been overplayed as an event. It is something carried 14 , over from the physicist world.
15 MR. EBERSOLE: It is an anachronism.
16 MR. REED: I don' t think it has anything to do 17 with decay heat removal, which is the real issue.
18 MR. LICITRA: Perhaps we are being a little 19 conservative here, but I understand -- I will have the 20 utility speak to this, but I understand that this does not 21 cause any schedule problems for Palo Verde.
22 MR. EBERSOLE: Well, fine. I would like to have 23 the utility really speak to the possible advantage of not 24 bothering with the criticality barrier.
() 25 Okay, thank you.
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
1 l
9250 01 08 9 jlgDAVbur 1 MR. REED: I have one point under your top 2 bullet, those three items. I am a little surprised not to 3 see the issue, PORV, listed unless it is hidden under 4 pressurized auxiliary spray options.
5 There are several ACRS letters since 1981 -- that 6 is three or fou" -- which address the PORV issue. How come 7 it is not the fourth item?
8 MR. LICITRA: Because the PORV~ issue is being 9 covered by the resolution of Task A-45, Unresolved Issue 10 A-45. It is currently not a licensing issue.
11 MR. REED: Then A-45 should be there.
12 MR. LICITRA: We will be discussing the status of 13 A-45 this afternoon, but we don't need a resolution of A-45 14 to be able to license Palo Verde Unit 2.
15 MR. REED: That may formally, officially be the 16 position, but I think from a judgment and technical point of 17 view we should continue to carry the PORV issue.
18 MR. WARD: Excuse me, Manny. Didn' t you misspeak 19 there when you said the PORV issue is not going to be 20 resolved as part of A-45 -- in my understanding -- but 21 rather the final resolution of the PORV issue is going to 22 await resolution of A-45?
23 I think that is a separate thing. Once A-45 is 24 resolved, then there is some sort of implementation plan,
') 25 then the PORV issue for these plants will be looked at ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336 6646
r I
) 9250 01 09 10 JggDAVbur 1 again with that background. But I don' t think you can look 2 to A-45 as providing a resolution for the PORV issue.
3 MR. XNIGHTON: I think you are right. The only 4 thing is A-45 should offer information to help us arrive at 5 the final point.
6 Currently, however, the licensing basis is that 7 both the auxiliary pressurizer spray and aux feedwater and 8 PORV provides sufficient reliability until we complete 9 A-45.
10 Could I address the last bullet that Mr. Ebersole 11 raised a question on?
12 Our concern is not the system and the fuel. Our 13 concern here has to do with operating two units at the same 14 , time. This is a new plant. We are trying to bring two new l
units up. We have new people running it, and it is our 15 l 16 concern and we want to be sure that they are ready and not 17 have two units at this point in time running some transient 18 operation.
19 That is the reason for the criticality decision.
20 MR. EBERSOLE: Criticality tests don't involve 21 transients.
22 MR. KNIGHTON: It involves individuals, involves 23 the method of operation.
24 MR. EBERSOLE: Right. Okay.
s
) 25 MR. REED: I am not sympathetic to your ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
i
~
)
9250 01 10 11 jggDAVbur 1 statement, having seen multiple units started up in sequence l 2 and these other kind of things have been involved, the kind i
3 of people that are involved in initial criticalities. It is 4 light operations and heavy reactor engineering, and I don' t 5 know that there would be people who would be there at all.
6 MR. KNIGHTON: Our experience would suggest that 7 we take this extra care, our past experience.
8 MR. EBERSOLE: Okay.
9 MR. LICITRA: That is all I intended to say. The 10 subject of operating experience is going to be covered later 11 on in the agenda.
12 So if there are no other questions?
- 13 MR. EBERSOLE
- Questions?
14 (No response.)
15 MR. EBERSOLE: Thank you.
16 Is that the end of the staff presentation?
17 The B topic under Item 2 is Unit 1 Power ;
18 Ascension Testing Experience, General Operating 19 Performance, and we have a Mr. Jerry Haynes on this.
20 (Slide.)
21 MR. HAYNES: My name is Jerry Haynes, Vice 22 President of Nuclear Production for Arizona Public Service 23 Company.
24 There are two areas of discussion for APS under I '; 25 Item 28. 1 and 2 I intend to cover: Itam 1, General ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
, 9250 01 11 12 jlgDAVbur 1 Operating Performance; then I will introduce the speaker for 2 the next item.
3 (Slide.)
4 The first portion of my presentation is going to 5 deal briefly with the schedule and how it was developed for 6 our power ascension test program.
7 The schedule was based on actual test durstions l
8 at other CE plants. The scoping of the testing at Palo 9 Verde versus other plants and then actual test time taken on 10 Unit 1 was factored into the schedule that we planned for 11 Unit 2.
12 So in summary, this testing schedule is based on
() 13 schedules experienced at other plants and the differences 14 between Palo Verde testing and the testing at other plants.
15 (Slide.)
16 The next slide shows our test schedule and how we 17 have done on the schedule. The slide shows power level over 18 on this side and time along the bottom.
19 The green portion of the slide shows our original 20 schedule. The black line shows the actual performance 21 relative to the schedule.
22 As you can see, early in the test program, back 23 in this area, we were well ahead of the program. 50 percent 24 testing was scheduled here. We were pretty well complete
() 25 with 50 percent testing for that point in time.
ACE FEDERAL REPORTERS, INC.
202 147 3700 Nationwide Coverage 800 336 4646
.9250 01 12 13 DAVbur~ l You can see a break in the schedule at that point 2 in time. That is the point in time where we discovered a 3 problem with our post-accident sampling system. It took us 4 approximately a month to perform modifications to resolv'e 5 that problem.
6 We then returned to approximately 50 percent 7 power and proceeded with our testing. At that point we were 8 behind our planned schedule.
9 Another areak occurred at this point. We will 10 discuss the reason for that and the corrective action we 11 have taken. That was the loss of power that resulted from a 12 load rejection test on September 12th. We will discuss what O 13 occurr a aa a e corr ceiv cetoa h v e
- a 1 e ta 14 the presentation.
15 We refer to that as the September 12th event. We 16 performed a corrective action and returned to 80 percent 17 power and proceeded with testing until this point, where 18 there is another break in the testing program.
19 That is associated with an event that occurred on 20 October 3rd. It was a loss of power -- it was mentioned 21 earlier -- associated with a fault in our multiplex system 22 which provided control of some breakers in the switchyard.
23 The fault resulted in both offsite power breakers opening 24 and an actual loss of power event. And we will discuss that O 25 ta a t it eaa-ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee 8003)H646
9250 01 13 14 1 During the troubleshooting associated with
~(}.DAVbur 2 correcting that problem, we experienced another loss of 3 ' pawer event. That also was associated with a f ault in the 4 multiplex system, not with a fault in the troubleshooting 5 sequence, as will be explained later.
6 MR. EBERSOLE: May -I ask a question at this time?
7 GDC-17 goes to considerable trouble to require 8 physical independence of the lines, and so forth, so you get 9 at least a duplex offsite power system before you have to 10 fall back on the diesels. So you invest a lot of money in 11 doing this sort of thing and bringing in the wires and 12 towers to do it.
() 13 Yet my impression is you basically convert to a 14 single multiplexer to control the operation of these two 15 expensive, physically separate systems.
- 16 Is that true?
17 MR. HAYNES: As I understand the operation of the 18 multiplex system -- and it will be explained later in more 19 detail -- it is separate. It shares some components, 20 however.
21 MR. EBERSOLE: It must have to do that.
22 MR. HAYNES: We were unable to find the fault 23 with the multiplex system. We took it out of that service 24 and hardwired the system. So the multiplex system is no
() 25 longer in question.
ACE. FEDERAL REPORTERS, lNC.
l l 202 347 3700 Nationwide Coversee 800 336 4 646
9250 01 14 15 jggDAVbur 1 MR. EBERSOLE: But is this true of other systems, 2 the one I mentioned in connection with the performance, the 3 control of the access to the primary heat sink?
4 MR. HAYNES: No, sir. I do not believe the 5 multiplex system is associated with that. The multiplex 6 system does supply some other components, but none of those 7 components are important for the mitigation or prevention or 8 coping with the results.
9 MR. EBERSOLE: Thank you.
10 MR. HAYNES: After removing the multiplexer from 11 the system and hardwiring the control breakers, we returned 12 to power and have been testing at the 80 percent level since 13 then.
14 On October 25th, during the performance of a 15 test, we again had a trip of the reactor and the turbine.
16 It was not what'should have occurred on the testing, and we 17 will also discuss that in some more detail.
18 At the present time we are out of service, 19 recovering from the trip that occurred.
20 On October 24th, corrective action for the reason 21 for that trip has been implemented. We are in the process 22 of returning to service with a 13.8 KV bus faulted. We are 23 now in the process of returning that fault.
24 We can discuss that in more detail, also, if you I
- 25 like. It is in the supply from the switchyard into the ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33 6 6646
9250 01 15 16 jlgDAVbur 1 plant. We are inspecting all of the other bus work of 2 similar design and for correct assembly and for performing 3 the corrections as needed. We expect to return to power 4 next week and complete the remaining 80 percent testing.
5 There is only one test left to perform at 80 6 percent power. That is the load rejection test.
7 He believe we understand well and have corrected 8 the reason for the unsuccessful completion of the test on 9 October 24th. We have no reason to expect -- that will not 10 result in tripping the reactor, so we will be able to come 11 back to power rather quickly and proceed to the 100 percent 12 level.
13 We have about three weeks of testing at 100 14 percent, and that will essentially complete all the NRC l
required testing with the exception of a natural circulation 15 l 16 cooldown, which we plan to do later when we have more decay 17 heat.
18 MR. EDERSOLE: Let me ask you a question on this 19 matter. When you finally have to resort to stop valve 20 closure, do you retain excitation on the main generator?
21 MR. HAYNES: No, we do not.
22 MR. EDERSOLE: It is zero excitation?
23 MR. HAYNES: I believe so, yes.
24 MR. EBERSOLE: If that is not true, we can
! 25 confirm it later.
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33H64
9250 01 16 17 JggDAVbur 1 Thank you.
2 MR. HAYNES: We are essentially four weeks away 3 from complehion of NRC required testing, with the exception 4 of the natural cooldown test.
5 We then also have some ASME performance testing 6 to be run. That is relatively straightforward.
7 So that is our status with respect to schedule.
8 (Slide.)
9 As to performance, I talked about some of the 10 events that have occurred, the primary event.s that have 11 occurred during the power ascension testing.
12 We feel we have had a successful program. If you 13 look at events such as reactor trips, we have had seven 14 trips. We would hope to knock out some, but if you compare 15 that to what has been experienced at other plants during 16 this same period of time, that is a pretty good record, I 17 think.
18 only two of the trips were associated with 19 tests. Generally, our test program has gone very well.
20 From the physics standpoint, the results have 21 compared excellently to the predictions.
22 From the plant transient standpoint, most of the 23 tests associated with transients have gone very well. As I 24 mentioned, only two resulted in reactor trips when that
[ ,< 25 wasn't expected. We also had two trips very early in the ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
19250 01 17 18 {
qlgpAVbur 1 game associated with the loss of main feed pump due to i
2 different reasons.
3 One of them was due to the need to do some 4 . fine-tuning of the recirculation flow on the main feedwater 5 pump and the condensate pump. That was accomplished, and we 6 proceed ed.
7 The second one was due to a suction strainer 8 which became plugged on the feedwater pump and resulted in 9 a low MPSH trip. This particular trip was associated with 10 control development of circuit board failure. It was pretty 11 straightforward. The board came back in service.
12 There were three trips associated with loss
() 13 power. We will discuss those in more detail. The first ynus 14 ! associated with testing, which :was a load rejection test, 15 which resulted in loss of power.
16 The other two were associated with the 17 multiplexer system.
18 One of those, by the way, was a trip with the 19 reactor suberitical. We had the shutdown rods on without 20 even approaching criticality. We were just sitting there 21 planning for the return to power in a couple of days.
22 MR. EBERSOLE: I wonder if you could give me your l
23 criteria for the use of MPSH trips on pumps. Of course, you l
24 have a lot of HE pumps. What about the other pumps?
l () 25 MR. HAYNES: We had an MPSH trip on the main ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 46d6 l
1 9250 02 01 19 yggpAVbur i feed pumps, of course. We have it on charging pumps. I do 2 not believe we have MPSH trips on any of the engineering 3, saf eg uard s.
4 MR. EBERSOLE: What is the rationale behind this?
s, 5 !!R. HAYNES: We are more interested in operation 6 of the pump than we are protection of the pump. '
7 MR. EBERSOLE: But you are interested in 8 operation over the long term and the prevention of damage in 9 the short term?
10 MR. HAYNES: Yes. There are, of course, two ways
, 11 to look at it.
l 12 MR. EBERSOLE: Yes. Significant damage, it would l () 13 make sense to have an MPSH trip, wouldn't it, but you are 14 not going to pump in the water anyway?
15 MR. HAYNES: That is absolutely correct.
16 MR. EDERSOLE: So I am trying to get at your root 17 logic for your MPSH trips, and I am not getting anywhere.
18 MR. HAYNES: Let me defer that question.
19 MR. EBERSOLE: We can defer it till later if you 20 wish.
21 MR. HAYNES: Let's defer that till later.
! 22 MR. EBERSOLE: All right.
23 MR. HAYNES: The last trip was associated with a l
24 fault in the steam generator level signal. That occurred on
(( ) 25 October 24th. I would like to discuss that in some l
l ace. FEDERAL REPORTERS. INC.
9250 02 02 20 qggDAVbur 1 detail.
2 We understand what occurred, and we have taken 3 corrective action to prevent that from recurring. So we 4 feel from a performance standpoint the unit has done very 5 well.
6 One thing I should note, none of these trips were 7 associated with operator error or plant stop error. They 8 were all associated with some kind of component type 9 failure or a failure sometimes much earlier.
10 But we are rather proud of the way t'he staf f has 11 performed.
12 MR. EBERSOLE: There has been no further with the i main coolant pumps?
13 l 14 l MR. HAYNES: No, there has not.
I 15 ' of course, you are aware of the problems that we 16 had with the hot function testing on Unit 1. Those problems 17 were corrected. We performed a demonstration test on Unit 1 3
18 and expected to thoroughly complete the work. The problems 19 were corrected.
20 We subsequently had hot function testing on Unit 21 2 and have inspected all the components associated with the 22 problems that occurred on Unit 1. We found no problems at 23 all. We found essentially no problems on hot functional 24 testing.
m
) 25 So we believe we have those behind us.
ACE FEDERAL REPORTERS, INC.
202 347-)?00 Nattoriwide Coversee 800136-M46
9250 02 03 21 1 The last item I would like to discuss is the
{~';DAVbur 2 schedule of Unit 1 versus the schedule of Unit 2.
3 (Slide.)
4 You see here in blue is the Unit 1 schedule. As 5 I mentioned, we have one test lef t to complete at the 80 6 percant power level. So we would expect to be complete 7 with 'an 80 percent testing here within about a week.
8 We are then going to the 100 percent testing. As 9 I mentioned, it is about three weeks of testing. So early 10 to mid-December we would expect to be finished with this 11 testing on Unit 1.
12 our schedule at this point -- and we are getting
() 13 close enough now I think we can assess the schedule pretty 14 well to your point, Mr. Ebersole -- we expect to have Unit 15 2 ready for fuel loading by the end of November. We have 16 assessed that from several viewpoints. :
17 one is to look at the number of items that are 18 open versus the number of items that were open at the same 19 point in tine prior to fuel load on Unit 1, and about a week 20 ago we assessed that. The result was that we believed we 21 would be ready for fuel loading.
22 MR. EBERSOLE: Now, this fuel load, when do you 23 expect to be full up and ready for the criticality test?
24 MR. HAYNES: Criticality will occur at the
() 25 completion of post-core hot functional testing. So that is ACE FEDERAL REPORTERS, INC, 202 347 3700 Nationwide Coversee 800136 4M4
I I 9250 02 04 22 gAVbur 1 late February.
2 MR. EBERSOLE : That is some time off, anyway?
3 MR. HAYNES: It is some time off. We don't 4 object to the staff's request. Of course, that assumes that 5 we will not have any significant long term problems on Unit 6 1. I think if we did we would want to come back and discuss 7 that.
8 MR. EBERSOLE: What would be your observation on 9 what we said earlier about criticality -- a barrier which is 10 of no particular importance?
11 MR. HAYNES: Criticality certainly does not 12 concern me from the standpoint of nuclear safety in and of 13 itself. I believe that we ought to control and we do plan 14 to control the activities that occur on Unit 1 and the 15 activities that occur on Unit 2 so that our plant staff is l
16 not overloaded.
17 We do have unitized staff. That is, we have a 18 staf f devoted to Unit 1 and a staf f devoted to Unit 2 and 19 another staff devoted to Unit 3. But some of the plant 20 management is common to all three units. We have done that 21 deliberately so we can transfer experience between units, 22 and I don't want to get them in the position of being 23 overloaded.
24 So we plan to control activities, at any rate.
[l 25 We believe we can control them without any real imposition ACE FEDERAL REPORTERS, INC.
202 347-37M Nationwide Coverage 800 33 & 6646
9250 02 05 23 1 of regulatory requirements to do that.
-JllDAVbur-2 MR. EBERSOLE: Thank you.
3 MR. HAYNES: That completes my presentation.
4 I would like to introduce Gary Waldrep, who will 5 discuss some of the events I referred to earlier.
6 Gary is the supervising STA, Shift Technical 7 Adviser, and has been fortunate enough or unfortunate 8 enough, depending on how you look at it, to have been on a 9 shif t associated with some of these events. So he should be 10 able to give you a good description of them.
11 MR. EBERSOLE: What is a supervising STA, since I 12 thought they were all pretty much advisory?
() 13 MR. HAYNES: There are a number of shift 14 technical advisers and we feel a need to have a supervisor 15 that looks af ter their performance and those kind of 16 things.
17 MR. WALDREP: Like Jerry said, my name is Gary 18 Waldrep. I am actually a lead shift technical adviser.
19 There's about six or seven STAS per unit.
20 I have been involved in all seven of the reactor 21 trips. I have been involved in the post-review process. I 22 was present in the control room for four of the seven 23 events, and I have had a major activity in most of the power 24 ascension tests that we have done.
() 25 I would like to first address the question that ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
-9250'02 06 24 l T' DAvbur 1 you had on the ESM pumps.
V
- 2. Although we don' t have an automatic trip feature 3 on them on low suction pressure, we do have other 4 indications such as low flow conditions that would alert the 5 operator that he had a problem, and then he could take 6 whatever action that lue deemed warranted.
7 MR. EBERSOLE: How much time does he have to do 8 that before damage occurs?
i 9 MR. WALDREP: If he was to get a low flow 10 condition?
11 MR. EBERSOLE: I am just, you know, trying to 12 rationalize whether you need auto trips.
() 13 MR. WALDREP: I don' t know how long the pump 14 could run at low flow conditions.
15 MR. EBERSOLE: Then I will postulate 30 seconds, 16 and we will argue about it.
17 MR. WALDREP: I would say longer than 30 l 18 seconds.
19 MR. EBERSOLE: So how long would you say?
! 20 Anyway, you.know, there is an equation here. If 21 it is too short, you have got to do something because it is l
22 the long term opera' tion of the pump that counts.
23 MR. WALDREP: Agreed.
i l 24 MR. EBERSOLE: Why don't you work up something, t
() 25 your rationale for automatic suction trips?
1 ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nadonwide Coverase 800 336-6646
[
~
L9250 02 07- 25 ,
p Vbur. >l MR. REED: I guess. Mr. Haynes said there are
- 2 automatic, suction trips on the ' charging pumps. You are 3 talking .about the primary chargers, . the reciprocators, is 4
4 that right?
.5 MR. WALDREP: That-is correct.
6 MR. REED:
I am just wondering on the-7 advisability. of that. Certainly, you have got your
? 8 so-called depressurization system tied off those charging 9 pumps and you put suction trips on those charging pumps, and 10 I don' t think that sounds like a good idea. Then the 11 charging pumps, reciprocators, a loss of suction. It would 12 be my opinion that the time to damage would be rather long
.()
13' and you would get into a lot of cracking.
14 MR. WALDREP: That is correct. I think we will 15 .probably -- in our presentation of the charging system in 16 aux spray we will probably get into that and answer that 17 question then.
18 MR. EBERSOLE: In that connection, let me take up 19 a more-general topic.
20 What about your main diesels? Suppose they go to-21' zero oil pressure. What happens then?
22 MR. WALDREP: The diesels have three main' trips.
23 If, for example, ~1oss of power occurred and the diesels had 24 to start, there are three essential trips that would trip
'( ) 25 the diesel, and most of the other trips that are associated s
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33M1446
9250 02 08 26 qlgpavbur 1 'with operation with operation of the diesel would be 2 overridden.
3 So unless it was a critical failure with the 4 diesel trip, one of them would be. low oil pressure.
5 MR.1EBERSOLE: So you see, you do provide trips 6 where the viability of the critical component is in short 7 term danger?
8 MR. WALDREP: That is correct.
9 MR. EBERSOLE: So this is on a par with what you 10 are talking about?
11 MR. WALDREP: My real topic of discussion today 12 is our experience with the reactor power cutback system.
() 13 (Slide.)
14 In this context, I will discuss our most recent 15 event of October the 24th.
16 Just a brief outline of what the reactor power 17 cutback system is. It is a control grade system which is 18 available to enhance the unit's availability. It is 19 designed in conjunction with the steam bypass control system 20 to drop selected CEAs, and it works on either a loss of feed 21 pump or a large load rejection. In those cases it will drop 22 CEAs to match steam bypass control capabilities with the 23 reactor power and thus keep the plant at power so that the 24 reactor doesn' t trip.
() 25 MR. EBERSOLE: That is an old system that used ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwkle Coverase 800 316 6646 I
, m . ~ . . _ . _ _ __. . . _ . _ _ _ _ . . _ . - . - _ _ _
~
\
9250 02.09 27 f~'DAVbur 1 to be called setbacks, or is it any different?
O 2 MR. WALDREP: I am not sure of that terminology.
3 MR. EBERSOLE: It is a rod setback' to reflect 4 changes in design.
5 MR. WALDREP: This works a little differently.
6 It does drive rods in; it actually drops rods. It selects 7 certain CEAs and drops those in order to' reduce power.
8 MR. EBERSOLE: I see. It doesn't have a name per 9 se? It is just a system at this time?
10 MR. WALDREP: That is correct.
11' MR. EBERSOLE: Thank you. I 12 I don't know how it can escape an acronym.
~
-() 13 MR. WARD: Wait a minute. He is calling it the 14 reactor power cutback system. Make up your own acronym.
15 (Slide.)
16 MR. WALDREP: I would like to go over the tests 17 we have done and the tests we plan to do on this system, 18 since it is. an integrated type system.
19 The first two tests that we have done is the loss 20 of: feed pump from.approxLnately 50 percent. That was 211 performed successfully the first time. We did not get a
- 22) cutback in the second feed pump. What we did is we tripped 23 off one feed pump. The second feed pump assumed the load.
24 i s The second test we did was a loss of feed pump
) 25 at 70 percent power. This test was also successful. What ACE-FEDERAL REPORTERS, INC. l 1 202-347 3700 Naionwide Covege 800 33H446
9250 02 10 28 1 we got there was the turbine ran back to 60 percent and no
(')DAVbur 2 actual reactor power cutback or no dropping of CEAs 3 occurred.
4 The third test we did was a loss of load from 50 5 percent. The initial test was unsuccessful and resulted in 6 'a reactor trip. So we reperformed the test, made the 7 correct modifications necessary, reperformed the test a week 8 later, and it was 100 percent successful.
9 I will discuss those when I get into that 10 specific reactor trip.
11 KR. EBERSOLE: These depended on bypass 12 functions?
() 13 MR. WALDREP: On the steam bypass function, yes.
14 It works hand-in-hand with the steam bypass.
15 MR. EBERSOLE: What kind of steam bypass do you l'6 have?
17 MR. WALDREP: 65 percent.
18 And the final test that we did, which also 19 resulted in a reactor trip, is the loss of load test'from 80 20 percent power. In this one, like I said, the initial test 21 was unsuccessful. This is the October 24th event.
22 We did expect a reactor power cutback actuation 23 to occur and actually drop CEAs. What we got was a reactor 24 trip.
() 25 Before I go into that event, I would like to ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33(H5646
925002 11 29 p Vbur 1 briefly talk about the tests that we are going'to do on the 2 system.
3 Of course, we are going to reperform the 80 4 percent loss of load test which was unsuccessful on October S 24th. That will be our first test when we get back up.
6 We will also do a loss of feed pump at 100 7 percent _ power and a loss of load from 100 percent power.
8 9
10 11 12 13 14 15 16 17 18
.L 19 20 21 22 l
23 I 24 O 25 l
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
. 9250 03 01 30
]lgDAVbur 1 MR. EBERSOLE: This high bypass -- it is a pretty 2 high bypass. Beaver Valley has 90 percent bypass. In case
. 3 of an inadvertent, or rather a reactor trip, you do have .
4 trip stop valves, don't you?
5 Mk. WALDREP: The turbine stdp valves, that is 6 correct. That is correc*., on the turbirie trip the stop 7 valves trip.
~8 MR. EBERSOLE: Does that directly disconnect the 9 excitation?
10 MR. WALDREP: No, it does not. In fact, when we '
11 1 .get into the trip you will see that it doesn't. The 12 excitation isn' t tripped off automatically until the voltage 13
~( ) on the bus decays to some preset level.
14 MR. CBERSOLE: Let me ask you a question.
15 Suppose you experience the usual challenge of a stop valve 16 closure and ycu had one of these rare and unfortunate events 17 where you have turbine runaway because it didn't flow.
la Those do happen once in a while.
19 The normal result of that is to expect turbine '
20 fragments all over the place. That is dealt with in a l 21 statistical manner, and you don't hit anything important. ~
s I
22 HoWever, if you carry away all of the cbunected loa <1 dist is 23 connected with it because you retairad excitation. As inu 24 run up, doesn' t that extend the damage to the paralle7.
('.) 25 connected load all over the plant? What do you d4 accut ACE-FEDERAL PM6KERS, INC.
202-347-3700 Natiewide Coverage 800-3 % 6546
1 1
- 9250 03 02 31
[';DAVbur - 1 that? Do you have excess frequency trips?
%J 2 MR. WALDREP: We have overspeed trips on the 3 turbine, both manual, mechanical, and electrical, 4 MR. EBERSOLE: Those haven' t worked , ohr4y.
5 MR. WALDREP: And we also have -- I will call l
6 it -- a quick transfer of th6 bus loads from the Generator.
7 MR. EBERSOLE: Not if it ia still carrying 8 voltage at frequency.
9 MR. WALDREP: If it senses a turbine trip, it 10 automatically swaps over to offsite power.
11 MR. EBERSOLE: On what signal? Isn't it a 12 voltage fregrency signal?
((} 13 MR. WALDREP: Noi that quick transfc;& not that 14 fast transfer. e 15 MR. EBERSOLE: kli I want to hear fou say is I 16 disconnect the ~ parallel load if I overrun and retain 17 excitation.
18 MR. WALDE3P: That is correct.
19 KR. EBERSOLE: You do that by some method you can 4
)
20 tell ne lat6r aboQt?
21 MR. WALCREPs I can.
22 MR. E9ERSOLL: Okay.
~
23 (Slide.)
z
- 4 MR. WALDREP: Now, to get specifically into the 25 Station
_() October 24th event. We were at 80 perceret power.
I ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide coversse 800 336-6646
9250 03 03 32 jggDAVbur 1 loads were being fed from offsite power. We initiated the 2 event with protective relays on the turbine.
3 Like I said, we expected the reactor trip not to 4 4 occur and that we would get actuation of the steam bypass 5 control system. What we got was a reactor trip concurrent G with this turbirca trip. In fact, within a second the 7 reactor power cutback system didn' t actuate. It didn' t have 8 time to actuate, and all the safety systems performed as 9 designed. ,
10 I would like to show you a time run of the event 11 and briefly hichlight what happened in the event.
12 (Slide.)
() 13 Like I said, we initiated it with the turbine 14 trip and we got the reactor trip concurrent with that. The 15 l steam bypass control valve opened to control pressure and i
16 the steam bypass control opening, which I will explain in a i ,
17 little bit more detail, caused us to cool down the RCS to 18 the point where safety injection actuation occurred.
19 I From'that point on the operator handled the event 20 in the manner that he should, with his procedures, with the 21 emergency procedures. All plant and equipment responded as 22 required. No real events of significance af ter that. It 23 stabilized and the operator could reset.
24 MR. EBERSOLE: With this high bypass and the
() 25 other opportunities for excess chilling during these ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646
o-
- 9250 03-04 l 33 1 various transients, what is your control for excess ggpAvb'ur
-2 chilling? Main steam isolation?
3 MR. WALDREP: We do have main steam isolation,
'4' that is correct.
5 MR. EBERSOLE: Is that the way you stop it?
6 MR. WALDREP: That is the way we stop it if it is 17:. induced by the steam bypass control system or something 8 downstream.
9 MR. EBERSOLE: Thank you.
'10 MR. WALDREP: I'would like to talk about the 11 specific events of the trip and our evaluation of those
. 12 events.
C 13 (S1ide'. )
14 The first one, like I said, was the reactor trip, 15 which we - didn' t expect to occur. This was caused by low 16 steam generator level signal, and what happened was the 17' .resulting turbine stop valves going closed caused a pressure 18 wave to be generated in the main steam lines, which the wide 19 range steam generator level sensed as a low pressure 20 condition in the generator and caused the reactor trip.
21 I would like to stress that the generators were 22 not at low levels. In actuality, the narrow range 23 i instruments were still on scale.
24 MR. EBERSOLE: Didn' t you just have a void t-I- 25 collapse?
L r
i ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 6646
l
, 9250-03 05- 34 1 MR. WALDREP: It appears to just be a pressure
' {v)DAVb ur:
2- wave induced in the upper reference leg in the steam space.
3 It only takes about four pounds dif ference to cause a 40 4 percent change in level. So that induced the reactor trip.
5 MR. EBERSOLE: Fine.
6 MR. WALDREP: The corrective measure is to 7 increase the time response in the reactor for the steam
'8- generator low level to the point where'this pressure wave 9 has time to dissipate before we get the reactor trip.
10 MR. EBERSOLE: In fact, it is true, isn' t it, 11 that due to the various voiding fractions the level rises
- 12. and falls pretty quickly?
13 MR. WALDREP: That is true. However, in this
(}
14 case the low level condition was sensed and was cleared
.15 within 300 milliseconds.
16- The second thing that happened was the cooldown 17 of the primary system, which we also didn' t expect. It was 18 caused by an incorrect setpoint in the steam bypass control 19 system itself.
20 Our corrective measure is to adjust the setpoint 21 in that steam bypass control system, and we did -- as a 22 protective measure -- did a reverification of all the' 23 setpoints in all our control systems to ensure they were at 24 the correct setpoint in agreement with the CE documents.
i
() 25 MR. REED: I am a little surprised that the ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 33(Hi646 t
9250 03 06 35 p Vbur 1 focusing on this kind of testing when you have not reached 2 'anywhere near, I am sure, equilibrium decay heat in the core 3 because primary system cooldown is a function of decay 4 -heating and you have not built up your decay heat and you 5 are doing this kind of complicated electrical power 6 continuity assurance testing in a period which surprises 7 me.
1 8 -
In fact, I am a little surprised that your focus 9 wouldn' t be on other kinds of testing that relate to basic 10 safety issues rather than the continuity of electricity 11 because you are going to have to go around and change this
- 12. later on when you have got equilibrium decay heat, in my 13 opinion.
14_ MR. WALDREP: I don' t think we are going to have 15 to change anything. I am not sure if I understand.
16 MR. REED: You are not going to get the primary 17 system cooldown, for one thing. You are not going to get 18 the safety injection after-you have got equilibrium decay i
19 heat.
20 MR. WALDREP: That is true. However, with the 21 systems operating as they should, we wouldn't get a cooldown 22 even now. We shouldn' t have got the cooldown.
23 .So with the correction in steam bypass control 24 system, we will not get the cooldown.
O 25 To de specific, what happened was the steam ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coversee 800 33H646
9250 03 07 36
.FDAVbur_ 1 bypass control system quick opened to control steam
\_/
2 generator pressure. It began to modulate closed. Then you 3 will maintain pressure. However, due to the incorrect 4- setpoint, a.second quick open of all the steam bypass
-5 control valves occurred. This resulted in, of course, the 6 cooldown.
7 MR. REED: I think the point that I am making --
8 and it relates to something Jesse was starting to say -- is 9 that the first thing it seems to me that is bnportant is do 10 your safety decay heat removal and all these fundamental 11 safety tests rather than the assurance of electrical 12 continuity.
p
() 13 But here you are focusing on assurance of 14 electrical continuity. Maybe all these other tests went 15 perfectly, so you don' t have to worry about it. But I might 16 be attempted to ask the question: did you do realistic 17 turbine overspeed tests?
18 That is a complicated question. Maybe your 19 layout doesn't require it.
20 Do you have turbine buildings perpendicular or in 21 line with the containments? What is the layout? I have 22 forgotten.
23 MR. HAYNES: They are perpendicular to 24 containment.
'( ) 25 MR. REED: So you don' t care if you have ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646
9250-03 08 37 h v b'u r l' -overspeed?
2 (Laughter.)
3 MR. HAYNES: We care.
4 MR. REED: Your diesels and all your equipment is 5 located in your fuel storage and your containment so that 6 :you have low pressure spindle rupture, disrupture, and you 7 don' t have a safety problem?
8 MR. WALDREP: That is correct.
9 MR. HAYNES: Ac I mentioned, we have performed a 10 number of tests, and we are really not focusing -- the-11 testing program is not focusing on these particular tedts.
12 It is- just that the reactor power cutback system is a rather
() 13 complicated control system, and so a problem has developed 14 with a couple of the tests that test that.
15 other testing has gone well, and you are not 16 hearing about that.
17 MR. REED: Would you say that this reactor power 18 cutback system is a nonnuclear safety-related activity in 19 tests?
20 MR. WALDREP: Yes, it is not.
21~ MR. HAYNES: It is classified as a 22 nonsafet?-related control system.
23 MR. EBERSOLE: In these tests is there any 24 compensatory allowance for the fact that, as Glenn says,
() 25 that when you adjust the fission rate in the normal full ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33(Hl646
I
-9250 03 09 38 T~^;DAvbur 1 power mode you don' t change the 5 percent outflow of decay V
2 heat which is always there? Do you account for the fact 3 that you don' t have any decay heat in biasing the the 4 setpoints?
5 MR. WALDREP: No, because the setpoints don' t 6 depend on what they look at. What is happening in the steam 7 generator, on that they base what they have to do.
8 So in a sense they do. It is an automatic type 9 control.
10 MR. EBERSOLE: Okay.
11 MR. WARD: Gary, one question. The setpoint of 12 the steam bypass control system that was incorrect, did you
() 13 determine the cause for it being misset?
14 MR. WALDREP: Yes, we did. The setpoint that was 15 in was a previous setpoint. When the revised CE document 16 came out, that particular setpoint was dismissed and wasn' t 17 updated . It was during the startup to operation phase of 18 testing, and that point was j ust missed.
19 MR. HAYNES: As Gary mentioned, we did go back at 20 this point in time, went back and checked all those 21 setpoints, and you can mention the systems, Gary, and verify-
^
22 that they were all corrected.
23 MR. WALDREP: Just to amplify, we are talking in 24 these systems hundreds of setpoints that have to be adjusted r
() 25 to for correct system operation.
ACE-FEDERAL REPORTERS, INC.
a 202-347-3700 Nationwide Coverage 800 336-66 4
l 9250 03 10 39 1 MR. WARD: So what you said was you had some JggDAVbur 1
2 incorrect documentation of combustion; was that it? l l
3 MR. WALDREP: No, I did not. The original 4 documentation had the setpoint at a certain value. A 5 subsequent CE document had it changed to a different value.
6 For some reason it was dismissed and wasn' t updated to the 7 corrected value.
8 MR. EBERSOLE: The system is generally a part of 9 the general plan of everybody to minimize scram frequency.
10 What do you expect of it? How did you manage to justify it?
11 MR. WALDREP: Unit availability.
12 MR. EBERSOLE: So it looks like it is worth j 13 money?
14 MR. WALDREP: We think so.
15 MR. REED: I am becoming more and more confused 16 about the need or the desire to spend great sums of money to 17 keep the unit online, shall we say, after it had a 18 transient.
19 Let me ask the question. I know if you have a 20 unit trip now there is red phone reporting, you have got to 21 convene your committees, you have got to do startup 22 procedure checkoffs, you have got to decide on the cpuse and 23 all these kind of things.
24 Supposing you have a partial trip, a cutback so 25 to speak or runback or whatever you call it, what are the
([ )
ACE-FEDERAL REFORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
9250 03 11 40 1 reporting requirements? What are the convening of committee jggDAVbur 2 requirements? What is the time allowed to restart?
3 Are you saying' those are all voided now?
4 MR. HAYNES: No. We would consider a turbine 5 trip with an actuation of our reactor power cutback system 6 internally the same way as a reactor trip. So we would go a 7 similar process for return to service as we would for 8 reactor trip.
9 MR. REED: So what you have saved -- you might be 10 sitting there two, six, or eight hours before you would be 11 allowed or would go back to power generation, so what you 12 have saved is a complete rod drop. That is all you have
(~ ') 13 saved.
A/
14 Is that worth that?
15 MR. HAYNES: We believe it is worth it, yes.
16 MR. REED: I would question it.
17 MR. EBERSOLE: There is something else you say.
18 If you have a trip, you have the usual costs associated from 19 the ill effects of the trip.
20 What trouble do you go to to retain main 21 feedwater flow without really challenging the aux feed 22 system, thus adding to its burden of reliability?
23 MR. WALDREP: On a normal reactor trip the 24 feedwater control system maintains its operation, the
(~') 25 condensate pumps continue to operate, and the feed pumps us (
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33MM6
9250 03 12 41
' 'DAVbur 1 continue to operate.
2 There's built-in control systems that will sense 3 the reactor trip and provide the sufficient feed flow to the 4 generators.
5 MR. EBERSOLE: So the reactor trip under the 6 circumstances is no more than a kind of benevolent exercise; 7 it doesn' t have any potential of quitting the cooling?
8 MR. WALDREP: It should not.
9 MR. REED: Just to summarize, for 10, 15, maybe 10 20 years now I have questioned whether the elaborate and 11 expensive investments in prevention of rod drop to keep the 12 reactor pushing out steam while the turbine is tripped, that
(} 13 these investments and complexities and time -- you just 14 spent a lot of time doing all this testing ,e is worth it at 15 all.
16 I think we should focus more on the basics and 17 stay away because we are stuck with reviews, reporting, and 18 all these things. I really don' t think we should be 19 spending our time and tying up our personnel and fooling 20 around with trying to save rod drop.
21 In fact, rod drops once in a while in the early 22 history of reactors are kind of good. You find out how many 23 go down.
24 MR. WALDREP: That concludes my presentation.
'\ 25 MR. EBERSOLE: Any other questions?
(J A.
ACE-FEDERAL REPORTERS, INC.
3)2 347 3700 Nationwide Coverage 800 336-6646
l l
9250 03 13 42 l T'~' DAVbur 1 (No response.)
V 2 MR. EBERSOLE: Thank you.
3 Well, we are right on schedule here. We have at 4 this time staff. comments on Unit 1 power ascension test 5 experience from Region 5 and NRR.
6 Mr. Crews.
7 MR. CREWS: I am Jess Crews, Senior Reactor 8 Engineer for Region 5.
9 I want to address the next item on your agenda, 10 or at least portions of that dealing with the NRC staff 11 comments on Unit 1 power ascension test experience.
12 In terms of the overall experience, the power
(; 13 ascension test program at Palo Verde Unit 1 commenced in U
14 early June of this year. We have had an opportunity ove'r 15 the past five months to observe the performance of people, 16 the operating crews 1* particular and the technical support 17 organizations for plant operations, the plant systems, their 18 overall performance, and the management systems for not only 19 the power ascension test program but the overall conduct of 20 plant operations.
21 From our observations and evaluations to date, we 22 have some comments in each of these areas.
23 The operating crews have been challenged rather 24 substantially, particularly in the last couple of months 25 with some of the operating events that the licensee has gone
(])
ACE FEDERAL REPORTERS, INC.
ll02-347 3700 Nationwide Coverage 800 33Mi646
_ _ = _
.... .__..- -. _ - _ _ - . _ . = _ _ . -_ . .-
-9250 03 14 43 TDAVbur 1 through, and in a somewhat different way than .one gets from N
2 a typical' simulator training scenario. ,
3 LSystems did not respond in the expected fashion 4 in some cases, and in some instances, particularly with the 5 . auxiliary pressurizer spray system, instrumentation failures 6 further complicated the operator's difficulty in responding ,
7 to the events.-
8 MR. EBERSOLE: Let me ask you this. The plant 9 does have a simulator, doesn't it?
10 MR. CREWS: Yes, it does.
11 MR. EBERSOLE: Are these events now programmed on 12 the simulator for reexercising?
13 You said, of course, there is a dif ference in
( )-
14- real life and a simulator.
'15 MR. CREWS: I can' t speak specifically to that. ,
16 There has been some training; some changes to procedures 17 have made as a result of some of these experiences. To my 18 knowledge, there hasn' t been a specific -- maybe the 19 licensee can address that.
20 We would expect that the experience is 21 encountered in the power ascension tesc program like other 22 . experiences.
23 MR. EBERSOLE: Well, it would appear that the
- 24 operator would be interested in reprogramming his real life 25 experiences on the simulator for occasional reworking of
.(]).
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
- 9250 03 115L 44 p Vbur 1 the scenario. ,
- 2 I guess we will ask you all later about that.
1 3 MR. CREWS: To .the extent that it has resulted in
! 4 . changes to-the emergency procedures and anticipated events
~
-5 the operator has to deal with.
6 MR. EBERSOLE: But the whole subject is an j 7 in'teresting one, wt$ ether or not real life experiences ought 8 ' to be always backprogrammed on the simulator in case they p
9 recur.
t
, 10 MR. CREWS: It is part of the requalification 11.' ' training program. This is done, and as I say, we would 12 expect these experiences, like others, to be factored into
- 13. the requal training program.
- 14 Anyway, under 'the circumstances we think that the i
15 operators have performed remarkably well in taking the 16 numerous actions necessary to return the system to ,
17 operability. It reflects well on their training and 18 knowledge of the systems, we think.
19 So overall, we score the performance of the 20 operating crews quite high.
21 MR. WARD: Jess, could I ask you a question about 22 that? When you say the operators have performed remarkably 23 well, are you talking about the operating crews? Who are 24 you including in that? Are you talking about shift crews O 25 xc1u tv 17 or o dro aer2 ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nanonwide Coverage 800 336-6646
9250 03 16 45 I jggDAVbur 1 MR. CREWS: In the context I just discussed it, 2 it is the operating crews.
3 MR. WARD: The shift crews?
4 MR. CREWS: That is right. I am going to talk 5 more about the technical support organization.
6 Yes, Glenn.
7 MR. REED: I guess what you are saying is -- you 8 made a comment that the simulators hadn't really predicted 9 the transients, and I can understand that. I find that 10 engineering at vendors' shops sometimes is misplaced or 11 decoupled from the real field.
12 I guess what you are saying is that preordained
~'
l ; 13 robots are not ready to take the place of genuine, selected, a
14 and trained operators?
15 MR. CREWS: That is true.
16 MR. REED: I am pleased to hear you say that.
17 ( Laugh ter. )
18 MR. CREWS: And we were pleased to see that the 19 operators coped with situations they were not specifically 20 trained to respond to.
21 The technical support systems overall performed 22 acceptably in terms of their post-trip and post-event review 23 and evaluation process.
24 The licensee management has taken steps to
() 25 improve upon the post-event and post-trip process.
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-136-6646
9250 03 17 46 T'IRVbur 1 In general, the staff has generated -- the
.v 2 : licensee's technical support staff has demonstrated their 3 technical capability to understand problems encountered and 4 recommend both design and procedural actions to correct and 5 prevent recurrence. In some instances, and generally as 6 appropriate, we think some of their expertise has been 7 called upon to better understand some of the problems that 8 they have experienced.
9 10
- 11 12 13 f]
14 15 16 17 18 19 20 21 22 23
.24 25 ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336 4 646
9250 04 01 47 1 MR. WARD: Jess, I want to go back over that jggDAV/bc 2 rather carefully in making these statements. I want to make 3 sure I understand what you're saying. You said the 4 technical and support organizations performed acceptably.
5 That's in contrast to remarkably well with the shift crews.
6 MR. CREWS: That's true.
7 MR. WARD: Then you went on to say there have 8 been some improvements in procedures. Is that right?
9 MR. CREWS: Right. And I will discu s and touch 10 upon some of those later on in the overall management 11 systems.
12 MR. WARD: And what you mean by that are 13 administrative procedures?
(
14 MR. CREWS: Information retrieval systems that 15 , allow a meaningful post-trip review, post-event revi.w. The 16 licensee has taken steps to expand the information retrieval 17 system capacity. They've been trading on the information 18 from the system. And the use of that information in the 19 post-trip / post-event review process.
20 We've also taken some organizational steps to
}
21 dedicate people who are predesignated to respond following 22 trip, or an unusual event. And these people have on call 23 other disciplines to participate in the post review / post 24 event process. We think those are steps that enhance the
() 25 ef fectiveness of those organizations.
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336 4 646
9250 04 02 48 1 In meeting - this training, . has the new system been gDAV/bc 2 much help as compared to the straight old board instruments-3' of the past?
4 MR. CREWS: I don' t think there's a great deal, 5 and perhaps the licensee can respond to that, but I don't 6 think there's been a great deal of utilization of that 7 system.
.8 MR. EBERSOLE: It's still logbook checking, and 9 so forth?
10 MR. CREWS: Well, computer data that's logged and 11 retrieved in sequence of the events, that sort of-thing.
12 MR. EBERSOLE: I was just noticing,.you know, the
(} 13. famous Davis-Besse event. They did have a display that was 14 very useful.
15 MR. CREWS: I don' t think that has played the
-16 principal role. Again, if'the licensee feels otherwise, 17 they can correct me.
18 MR. REED: In view of that " remarkably well" 19 statement for operating crews, I cannot resist making a 20 standard punchline of mine that the operating personnel at 21 APS Palo Verde were natural ability selected for their jobs, 22 according to information we were given.
23 MR. EBERSOLE: Give him a small flag to . wave.
24 MR. CREU3: Certainly we think the operators have
.(} 25 undergone extensive training. There's been a great deal of ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33tNi646
f 9250'04 03 49
~N 1 use- of the simulator and the power ascension test procedures
'[VDAV/bc 2 have specifically been used in the simulator. So we think )
3 _this has paid off.
-4 MR. WARD: Let me ask a more specific question of
-5' 'the licensee in response to Mr. Reed's comment. What 6 selection testing is used for your operating staff?
7 MR. HAYNES: Let me refer that to Bob Adni. He's 8 the superintendant of the unit two operating staff.
U 9 MR. ADNI: What we have is a standard industry 10 pause evaluation test; plus we require of all the operators 11 at least -two years of some type of . power plant process 12 experience prior to coming into an entry level job.
()' 13 So we fee 1' comfortable-with them having the 14 academic background and skills in a hands on mechanical type 15 background.
16 MR. WARD: Thank you.
17 MR. CREWS: In terms of plant system 18 ' performances, you've just heard from the licensee there have 19 been systems where experience during the power test phase 20 have not performed. of particular significance in this 21 regard are the experiences encountered with the auxilliary 22' pressurizer spray system and the difficulties in the gas-23 binding charging pump and the CBCS components in that 24 system.
() 25 One of the principal purposes of the power ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33H646
49250 04 04 50
~7 sDAV/bc 1 ascension test program is to put plant systems to challenges V
2 _necessary to demonstrate acceptable and reliable performance l~ 3 of the system.
4 And we try to keep the experiences that have been 5 encountered in perspective in that regard. Today, the power i 6 ascension test power program at Palo Verde Unit I has served 7 that purposes; not in all instances, I might add, in exactly 8 the manner the testing program was designed to do so, the 9 experiences encountered with the multiplexer system and the 10 electrical distribution system.
11 It's a particular example of.the latter that I 12 would point to. This feature of the Palo Verde plant is the 13 first application that we on the staff were aware of, at <
{ }~
14 leact where it's been utilized in a facility licensed by the 15 NRC. And I understand there is another facility that is not 16 yet operational, as a similar system. I am sure we will 17 focus on that system in handling that problem.
18 In terms of the overall management systems, the 19 overall effectiveness of the management systems for the 20 planning and conduct of the power ascension testing program 21 and the conduct of operations in general, it is our ,
22 assessment that these systems have proven to be generally 23 adequate as well.
24 There have been experiences, however, where
(} 25 needed improvements have been revealed and the licensee has ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33H646
l
~
- 9250'04 0'5 51 V/bc- 1 taken action to effect those improvements.
2 I mentioned post-trip / post review process. There 3 have been changes made in information retrieval systems and 4 training in retrieving this information and its use in the 5 post trip review and evaluation process.
6 I mentioned also-that they had dedicated people
--7 tx) respond for' the following reactor trips or unusual 8 ' events. Those people are predesignated. They represent the 9 reactor protection safety-related part of the plant as well 10 as the secondary plant. And then they have on call to them ,
11 other disciplines that are available with alternates to 12 respond.
13' The licensee mentioned in their presentation that
( )-
14 there was some difficulty. I guess I characterize the 15 difficulty as the control of other information, particularly 16 as it relates to the secondary side of the_ plant. The 17 licensee has taken steps, in fact, prior to this event, had 18 recognized the need to bring other information, even though 19 it relates to the secondary side of the plant. Under better.
20 controls, currently, that sort of information is under the 21 same controls that are applied to safety-related information 22 for the reactor safety systems of the plant.
23 This one did. The information received, as I 24 recall, was about February or so of this year. That was
() 25 about the time they were implementing the improved controls ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33H646
_ - - - - . .-~ . .. - - -._ .
9250 04 06 52
.of the systems, and somehow it dropped through the crack. I
. !m .JDAV/bc1
- 2 They just missed it.
3 But vendor information of that nature now is i 4 controlled under the design review process. That's applied 5 by the vendor.
-6 MR. EBERSOLE: As you know, for many years, the J 7 secondary side was kind of shunted off as being old art and j 8 science, engineering, and it was little recognized that it 9 was the home place of heat rejection, at least when you are 10 on primary system pressure.
11 This is particularly true of this plant where you 12 have very strange methods of getting down to low pressure,
- I) 13 except through the secondary side. Are they giving that
- 14 attention on quality control systems?
15 MR. CREWS: I think, generically, the staff has 16 really focused attention only recently on the quality of the 17 maintenance and the modifications to secondary plant and the 18 performance of secondary plant more in the context of the D 19 potential of malfunctions in this part of the plant to offer 20 unnecessary challenges to reactor protection.
21 MR. EBERSOLE: The old escape route used to be 22 that you could pour it down through the PORV's and get into 23 the old mode anyway.
24 MR. CREWS: I think there has been, particularly
() 25 in terms of the auxilliary pressurizer spray system with i
4 ACE FEDERAL REPORTERS, INC.
. ..202
_ 347
_ _ 3700 Nationwide _ .. _ __ . ___ . _ _800
_ _ ._ _Coversee
. _ . .__ ,. . _, __ 3 36-6646 , _ _ . _ _ . _ , _ _
9250 04 07 53 jlgDAV/bc 1 staff interaction, a greater sensitivity in terms of the 2 importance of that system and its dependence, its 3 reliability, dependence on nonsafety-related CBCS 4 components.
5 We've taken some steps procedurally and we'll be 6 discussing some design events that will improve the 7 reliability of those components.
8 MR. EBERSOLE: What will it do if you lock up 9 liquid solvent by some strange transient? Nothing, right?
10 MR. CREWS: I'm hard-pressed to think of 11 anything.
12 MR. EBERSOLE : Or maybe a leaky safety, Lj 13 whatever. Well, anyway, you know, it's kind of an open 14 question.
15 MR. CREWS: Right. I mentioned the control of 16 information even though it is information that relates to
, 17 recommended setpoint changes in secondary plant system, 18 steam bypass control, and that sort of thing.
19 We think they have a system for better control to 20 assure the implementation of this recommendation, since it 21 relates to the secondary plant. In terms of the review of 22 test procedures, we have seen an example from the September 23 12 event where the procedural step in initiating that test 24 was done in such a way that it precluded the auto transfer (nj 25 from occurring on the auxilliary loads from houseload to the ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
'9250 04 08 54 JggDAV/bc 1 offsite power system.
2 The licensee recognizes and corrected the 3 deficiency subsequently and, subsequently, successfully 4 performed the test. However, we do feel that that was 5 avoidable had the proper review and perhaps by the vendor in 6 this case, the turbine supply, would have recognized that 7 step. It would not have achieved what they had intended in 8 terms of initiating the test.
9 And they have committed to bringing their 10 expertise into review of their power testing procedures, 11 Also, we've seen and the licensee has been able 12 to control some of the maintenance activities again in the
()
13 secondary part, as I mentioned, after the experience, which 14 d revealed the failure to*mplement a setpoint change in the' 15 steam bypass control system.
16 They did go through a thorough checkout of the 17 setpoints and other secondary systems and found on the order i
18 of three or four other setpoints that were not properly 19 set. Those discrepancies were not from the same cause.
20 l Changes had been made during the maintenance of 21 the systems and they've taken steps to better control that.
22 MR. EBERSOLE: May I ask you a question about the 23 new importance of primary loop pressurizer level control?
24 I'm reminded of the fact that the boilers rest on the 25
(-)s viability of the boilers, which is on the dump water for l
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646
' Nh ;,
9250 04 09 55 x pDAV/bc 1 scram.
Gi 2 Here the CE plants must have an accurate and 3 continued representation of the level in the pressurizer to, 4 be sure that they can depressurize with a pressurized spray s
5 system.
6 Let me hypothesize that,they go blind on the 7 pressurizer level. I realize they have three channels. So 8 what-do they do? Don't.they charge it solid?
9 MR. CREWS: I don' t know. I really don't feel I 10 can comment on that.
11 MR. EBERSOLE: And further lock it up? We'11 ask 12 that-later.
{/ .
~
13 < MR. CREWS: That concludes the comments we have 14 on the experience to date.
., 15 MR. EBERSOLE: Thank you. I believe we have a 16 continuance of this until about 10 o' clock, by Mr. Licitra.
I' 17 (Slide.)
18 MR.LICITRA: I'm Manny Licitra. I'm Manny U 19 Licitra, again. The next item on the agenda covers the
, 20 question about, as I understand it, what was the basis for 21 being able to restart Palo Verde Unit I following the 22 issuance of a confirmatory action letter on September 17th.
23 A little background regarding that letter. What 24 had happened is, during the September 12th event, where we 25 had a turbine trip / reactor trip safety injection, et cetera,
- v.O ;
\
4 ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336 6646
9250 04 10 56 T)DAV/bc 1: we -learned there were certain aspects about the plant that
%.J 2 did challenge the auxilliary pressurizer spray system. For
- 3. example, there was a level indicator on the volume control 4 tank that went bad, and that contributed to the charging 5 pumps going dry.
6 So that was one of the problems. We also learned 7' that on safety injection, power supplies to motors for 8 certain valves got shed, which again affected the ability to 9 provide charging flow and, in turn, to provide, if required, ,
10 flow for the auxilliary pressurizer spray system.
11 Because of that, we did have discussions with the 12 utility and they recognized there was a problem. So they 13 took it upon themselves to keep the . plant out until they
(~/T.
x_
14 reached some resolution with the staff about restarting.
15 That was confirmed by letter on the 17th.
16 Subsequent to that letter, the utility did 17 -provide a response on the 18th and did meet with the staff 18 on the 20th, and went over the sequence of events, which 19 will be covered later. And, as a result, we did reach 20 resolution on what would be required to restart the plant 21 pending final resolution of the issue.
22 These were spelled out in a followup letter dated 1 23 the 20th which identified the compensatory measures. The )
l 24 first item is the daily monitoring of the reference leg of
( )' 25 the volume control tank level indicator.
l ACE-FEDERAL REPORTERS. INC.
202-347 3700 Nationwide Coverage 800 336-6646 1
I9250 04 11 57 l-_ MR. EBERSOLE: Is there only one indicator?
f~]DAV/bc v
2 MR.LICITRA: They have two indicators but one had 3 one reference leg.
4 MR. EBERSOLE: So it's one of the old cases where 5 we have ~ two indicators with a common mode failure at the 6 transducing level?
7 MR.LICITRA: That's correct.
8 MR. EBERSOLE: Should we fix that?
, 9 MR.LICITRA: They are fixing that, I believe. I 10 don' t know if it's fixed at this time, but they are fixing 11 it..
12 MR. EBERSOLE: Is it true of any other
(~') 13 transducers that they share a common hydraulic leg?
U 14 MR.LICITRA: I'm not aware of that.
15 MR. EBERSOLE: I'll ask the applicant that 16 later. That has happened in the past, so the failure of the 17 hydraulic leg or the impulser static line gives you common 18 mode failures.
19 Let me ask you this. I think we must take it for 20 granted that sooner or later we're going to lose the spray 21 mode of depressurization, whether it's by 22 auxilliary...auxilliary is one way.
23 MR.LICITRA: They have normal spray, they have 24 auxilliary spray.
25 MR. EBERSOLE: They have two ways of doing it, (a~)
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
l 9250 04 12 58 m
(_,DAV/bc 1 but let me say, sooner.or later, we' re going to lose that.
2 Does the testing program assure that if that occurs, they 3 can have an adequate depressurization system by invoking 4 secondary blowdown?
5- MR.LICITRA: I don' t believe the program covers 6 that where they lose ~both main spray and auxilliary spray.
L ,
7 MR. EBERSOLE: Isn't it me.ndatorar where the 8 applicant has claimed secondary great safety eg'.'.ipment? If i 9 we don't have a testing program for that, I'm goir.g to be 10 really critical. That's been the main clat n to fame, thnu 7 11 can always get down oy using secondary dep'ressurication. if <
12 we don' t test that, where are we?
) 13 MR.LTCITRA: They do have a program, the test, 14 the depressurization scheme, using the aux spray system-15 MR. 3BERSOLE : No, that's not what I said.
16 j MR. LICITRA.: I knotr. ,
17 MR. EBERSQLE: Well, we'll mark that question for >
18 , inquiry later on. Thank you. Go ahead.
19 MR.LICITRA: The next measure was to revise the .
20 procedures for assuring that the refueling water tank was on 21 line to ch.e charging pump section upon loss of offsite 22 power. Actually, the way the system is designed, the way it 23 was designed, and the way, even af ter they changed it, it's 24 still the came, if you have a loss of offsite power, you do C)
(_. 25 not lose power to certain valves.
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Na',ionwide Coverage __ , __
800 3 4 6646 _. _
s9250.04 13' 59
.c\
- ( ,,DAV/bc 1 The valve that closes the volume control tank and 2 the valve that opens up the flow from the refueling water 3 tank, d ose valves would still have power. But if you have
. 4 a safety injection, they would lose power. And when you S. lose powery you would f ail the valves as is.
6 Now, in anticipation of eventually getting safety 7 injection, a compensatory measure was put in so that as soon 8 as you had a loss of offsite power, you change. You change 9 over tn the refueling water tank.
10 So with the current design, if you've got a -
11 ' safety injection, you would still have flow to the charging 12 pump.
() 13 MR. EBERSOLE: This will even add to the chance
- 14 that you will go water solid.
15 MR.LICITRA: Excuse me?
16 MR. EBERSOLE: You lock in the charging method of 17 the charging pumps.
18 MR.LICITRA: You assure that the charging pmaps 19 bave a source of water.
20 MR. EBERSOLE: So you' re doing many things to 21 enhance the flow of charging water, which disenhances the 22 notion that you will go water solid. You're riding on the 23 level controls to stop it. Right?
24 MR.LICITRA: I guess I don' t understand the
() 25 question.
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6646
^
\
l
- - l
'9250'04 14 60 l 1 MR, EBERSOLE: I'm doing everything I can to gggDAV/bc 2 ensure charging flow. You can lock it on an emergency 3 transient. I'm trying to get the primary loop full of 4 water. Right?
5 MR.LICITRA: The charging flow is not the 6 mechanism for adding water to the primary system. If you 7 have a safety injection signal, they have a high pressure 8 safety injection signal.
- 9. MR. EBERSOLE: And that signal runs in overrides, 10 even the level controls. Right?
f 11 MR.LICITRA: I do not know.
12 MR. EBERSOLE: Is that right? It always does.
y 13 So inadvertent safety injection overrides level control, so 14 another way of killing the spray depressurization system is 15 inadvertent initiation of safety injection and overriding if 16 Palo Verde is going to stop it.
17 MR.LICITRA: That's the charging flow.
18 MR. EBERSOLE: Right. The system is anxious to 19 get full of primary water without limit. If the operator 20 doesn' t intercede, it will go solid. That kills the 21 auxilliary sprays and the main sprays. A troubleso.ne 22 situation. i 23 24 ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
"9250-05 01- 61
'4MImVbw- 1 MR.1KNIGHTON: At'the same time, however, that O, '
~2 you're using it, you' re also cooling down the primary system 3 ,MR. EBERSOLE: Sure. Oh, sure, until you get 4 full. Okay.- Go ahead.
5 MR. LICITRA: The third compensatory measure was
- 6. to institute pr'ocedural cautions on the restarting of the
'7 ' charging pumps.- The reason for that is, during the event 8 the charging pumps did go dry, because of other 9 ' considerations. The bond control tank went dry.
Therefore, 10 there was no flow to the charing pump. There was no path 11' for the fueling water tank to get water to the charging
'12 pumps. So they had to vent the charging pump line to make
- 13 sure they could get.
)
14 MR. .EBERSOLE:. Why did-they turn off the charging' 15 pump source of water, when they invoked safety injection.
~
16 Isn' t it- just another way of putting water in the primary?
17 MR..LICITRA: I think they manually tripped the 18 charging pumps, because they were behaving erratically 19 because of the gas binding.
20 MR. EBERSOLE: Okay. My question was, you know,-
21 .w hy do you deny the charging pump operation, once you go 1
22 into safety injection? Isn' t it just an additive way to get 23 water?
24 MR. RAYNES: We don' t deny it.
{} 25 MR. EBERSOLE: I thought you valved it out.
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
l i
i i
'9250 05 02' 62 (4
j DAVbw 1 MR. LICITRA: Valve it out? Oh, now. I'm
.2 sorry. Maybe I should.go back to the second point. The' 3 valving. arrangement was to assure that there is water to the ;
-4 . charging pumps.
5 MR. EBERSOLE: Even in the condition of safety 6 injection.
7 MR. LICITRA: Even in the condition of safety 8 injection.
] 9 MR. EBERSOLE: Okay. .I misunderstood. All i
10 right.
11' MR. LICITRA: What the safety injection signal 12 did is to strip a motor control center that provided power 1() 13 - to certain valves, and therefore, they did not have ' the 14 ability to realign the charging flow without going outside 15 the control room and resetting that motor control center.
16 MR. EBERSOLE: Okay.
17 MR. LICITRA: This compensatory measure allowed 18 them to switch over from the control room before they
! 19 actually get a safety injection signal, so that when you do 20 get a safety injection signal, the valve is already in line 2
- 1 properly. So they shed the power to the valves, they're in 4
22 the right modes.
23 MR. REED: Do you think you've cured all those i
j -24 problems now? I'm really disturbed by the fact that here
() 25 the system, the auxiliary spray system is concocted to 1
2 I ACE-FEDERAL REPORTERS, INC.
. - ._ . __ _M:Pl*3*._ N"DNME _. _ _ . M3M __._, ___ _ _ _ .
l 9250!05 03 63 DAvbw- 1 replace PORVs, which have a very important function, and new
{
2 the details of system design caused this system not to be 3 ' operable in the real world.
4 Now it bothers me that that kind of thing at the 5_ outset comes out in.the start up testing. It bothers me to 6 the extent that I wonder how thoroughly has this whole 7 auxiliary ~ spray scheme of depressurization been evaluated 8 anyway.
9 MR. LICITRA: I'd like to defer responding to 10 that question to later on in the presentation, when we go 11- into details about the auxiliary. pressurizer spray system.
12 It's a good question, but there'll be someone else who will
() 13 be able to respond'to that at the appropriate time.
14 The next compensatory measure was to assure that 15 the post-trip review process-evaluates events with respect 16 to potential safety significance. From what we saw of the 17 post-trip review report, it did not focus on the aux spray 18 ' system at all. It wasn' t recognized at that time that it -
19 was something that could not have worked, and we thought it 20 would be appropriated that from now on they should focus on 21 it. In.the post-trip review process, you focus on the 22 entire process, what happened and how it relates to any 23, safety significance.
24 MR. EBERSOLE: What bothers me most here is the
() 25 Staff's permissiveness in allowing them to rise to power ACE-FEDERAL REPORTERS, INC.
202 347 3700 ' Natinnwide Coversy 800-336 6646
i
'9250:05 04 64 1- without the presence of any test on secondary blowdown as a
([)DAVbw 2 means of primary. pressure control and'the recognition, I 3 think, that' sooner or later we're going to lose spray 4 control pressure.
Though we have not required a test to 5 show how one does this through this newly qualified 6 secondary blowdown system, which is the only way you're 7 going to get access to low-pressure system.
8 MR. LICITRA: George?
9 MR. KNIGHTON: I guess my question is, I am not 10 familiar with your qualifyir:g secondary blowdown.
11 MR. EBERSOLE: That's one of the claims to fame.
12 MR. KNIGHTON: This is where I personally don' t
() 13 understand it. We might want to look into this.
14 MR. EBERSOLE: That's why I went to the modest '
15 trouble to qualify the secondary blowdown and upgrade, at 16 least numerically, if not really.
17 MR. KNIGHTON: You mean, in terms of cooling down 18 the primary?
19 MR. EBERSOLE: To get it on RHR; yes.
20 MR. KNIGHTON: Okay. But the depressurization of 21 the primary is accomplished on the primary side. Cooling 22 down the primary is accomplished on the primary side.
23 MR. EBERSOLE: Depressurization is cooling down 24 or rather cooling down on the secondary, accomplishes (O,j 25 primary depressurization. That is the escape route for CE, ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-33MM6
9250:05 05 65 jlhDAVbw 1 if the cooling spray system doesn' t work.
2 MR. KNIGHTON: To some extent, you can only 3 depressurize down to RH.'t.
4 MR. EBERSOLE: That's the whole point.
5 MR. KNIGHTON: Using the primary system, spray 6 system, either with the reactor coolant pumps running --
7 MR. EBERSOLE: Wait a minute. Hang on. You're 8 saying something that bothers me more. You're telling me I 9 cannot use secondary blowdown to get down to 400 psi on the 10 primtry; is that true?
11 MR. KNIGHTON: I would personally have to turn to 12 somebody who is more technically competent.
I) 13 MR. R2ED: Jesse, I have heard that remark about 14 secondary blowdown, secondary cooldown being able to 15 depressurize primary, and I never understood it from the 16 beginning.
17 MR. EBERSOLE: Is it true that secondary 18 blowdown, which has been enhanced by upgrading the secondary 19 valving, will get you down to primary coolant conditions?
20 MR. HAYNES: Perhaps CE can address that. To my 21 knowledge, at Palo Verde, we've taken no credit for 22 secondary blowdown, for cooldown, nor have we upgraded any 23 valves on the secondary blowdown.
24 If I understand yoa correctly -- you mean steam
() 25 generator blowdown?
l i
ACE-FEDERAL REPORTERS, INC. l 202 347 3700 Nationwide Coverase 800 336 6646 :
9250 05 06 66 dlhbAVbw 1 Is anyone from CE able to address that?
2 MR. TURK: Rick Turk. I think what you're 3 referring to was a study that was down regarding not the aux 4 spray but the aux feed. That study showed that in the case 5 of unavailability of the high hand aux feed pumps, the 6 generators could be depressurized through the atmospheric 7 pump valves to allow feed from the condensate pumps.
8 Secondary blowdown is not any kind of backup for 9 auxiliary spray. Auxiliary spray just is required to remove 10 heat from the pressurizer, not cool the plant. Cooling the 11 plant is the job of the steam generators.
12 MR. EBERSOLE: So you do not have access to low 1J 13 pressure cooling via cooling down the primary loop to a 14 pressure level, that of 400 pounds, which I take it is the 15 uppermost RHR pressure. You cannot use the secondary 16 blowdown as a means to depressurize the primary.
17 MR. TURK: The cooling associated with heat 18 removal from the generators, whether it's using the aux feed 19 pumps or whatever other means will result in reducing 20 pressure in the primary.
21 MR. EBERSOLE : Can' t you get it down to a 22 saturation pressure below 400 pounds pressure, at which 23 point you can invoke RHR cooling?
24 MR. TURK: Via cooling from the steam generators?
q
_) 25 Yes, you could do that. You also have to remove heat from ACE FEDERAL REPORTERS, INC. ,
202 347 3700 Nationwide Coverage 800 336 6646 )
---9250 05 07 67 h Vbw . . l' the pressurizer, which is the job.of the auxiliary spray or 2 ' main spray.
3- MR. EBERSOLE: - So you' re saying the pressurizer 4 will remain locked up at high temperature, even af ter you 1
5 cool the main loop. For how long?
6 MR. TURK: It will, if level is maintained in the
, 7 pressurizer.
8 MR. EBERSOLE: So you are really hung up at high 9 pressure, even though you couldn't cool down the primary.
10 MR. TURKS. When you say " hung up," this is a 11' process that has to be accomplished over .several hours, and g 12 it's a matter of ensuring that there's sufficient auxiliary
- 13 . feedwater supply to allow the. pressurizer to cool off.
14 MR. EBERSOLE: But there is no mode of' cooling of 15 the pressurizer per se.
16 MR. TURK: There are several modes of cooling.
17_ There's main spray.
18 MR. EBERSOLE: I'm not talking about that.
19 MR. TURK: Aux spray. It cools the atmosphere.
20 There are small vent valves that can be used.
21 MR. EBERSOLE: Oh, there are small vent valves.
22 MR. TURK: I think you'll hear about those later 23 in the presentation. Those should not be < >nfused with the 24 PORV, which is a massive release.
() 25 MR. EBERSOLE: But you do not claim secondary ACE FEDERAL REPORTERS, INC.
FM347 3700 Nationwide Ce 800 336 4646
I 9250 05 08 68 jggDAVbw 1 blowdown as a means to pressurizer the primary system.
2 MR. TURK: We do not claim secondary blowsdown as 3 a means of controlling temperature and pressure in the 4 pressurizer.
5 MR. EBERSOLE: Okay. So that is passe, right, 6 because mainly it's - not part of the heat transfer path.
7 It's locked up, outboarded on its own; right?
8 MR. TURK: Correct. But the primary concern is 9 the heat removal from the primary system, and the ability to 10 depressurize to shut down cooling is a matter of time in 11 getting into that. There is nothing inherently unsafe about 12 being at that high pressure and removing heat through the
(') 13- steam generator.
14 MR. EBERSOLE: How long. would it take the 15 pressurizer vessel to be cooled by these small vents that 16 you have?
17 MR. TURK: Within the time frame of the 18 condensate storage interval maintained in the plant. This 19 is the RSB 5-1.
20 MR. EBERSOLE: Can you tell me, in hours?
21 MR. TURK: Poughly 10 to 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
22 MR. EBERSOLE: 10 to 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />, you're locked up
~
23 at a high temperature, even though you cool down the 24 primary.
()
~
25 MR. TURK: You may be gradually depressurizing.
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 6646
9250-05 09- 69 h Vbw . l' MR. EBERSOLE: Thank you.
2 MR. REED: Manny, this seems like a good time to 3 go back to-your earlier presentation and just make that
~4 point again about PORVs that were not listed. You had'three 5 items you didn' t have PORVs, and I'd like to point out- that 6 on May 13, 1985, there was an ACRS letter that said, okay, 7 ~ we've looked again at Palo Verde, and I guess System 80 is 8 implied and subject to our previous letters and what they 9' said, everything looks fine. But if you go back to the 10 letter of October 1983 from the ACRS, there are some words
'll
.intherethatsaysomethingaboutimprovementsaresortof 12 expected from A-45, within one year, that will relate to O 13 verhaes the backfits necessarv.
-14 Now I'm beginnir.g to wonder if -- 1983, of
.15 course, plus one, is 1984, and here we are in 1985, which 16 -shortly is 1986, and this . issue is still floating around.
17 With respect to the PORVs, I am beginning to think- that 8 'under your items in your initial presentation, you should 19 have the PORV. issue there. I think it is alive, because 20 it's not been resolved, as far as the ACRS letters are 21 concerned.
22 MR. KNIGHTON: Again, our licensing position 2J right now, which we mentioned earlier, doesn' t have us 24 condition a license on putting PORVs in. That is what I h 25 understand you're saying. ton't license Unit 2 until A-45 ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33HC46
9250 05 10 70 Vbw 1 is completed.
2 MR. REED: I think you should do a careful 3 rereading of the May 1985 letter. The October 1983 letter 4 and the 1981 letter, to decide whether or not the ACRS has 5 really signed off. I don't think they have.
6 MR. KNIGHTON: I think it was the '83 letter.
7 You made a point, I believe. The strong point in there was 8 to get through.and finish A-45.
9 MR. REED: Within one year.
10 MR. KNIGHTON: That's the way I read your 11 letter. I did not say that Palo Verde was unsafe because it 12 didn' t have PORVs.
13 So that's where we are right now.
( ([
14 MR. REED: Would you like me to read the 15 paragraph? I can find it.
16 MR. KNIGHTON: My only point, I was trying to 17 reflect my reading of it, but we will go back and look at 18 your letter.
19 MR. REED: We appreciate it.
20 MR. EBERSOLE: Let me rescind what I said then 21 about testing the secondary side or its ability to control 22 primary pressure, since it cannot. I have been under a 23 palpable delusion that it might be useful in that aspect, 24 thinking that we might shrink the primary and eventually get 25 the hot water out, but I'm wrong. Are there any other ways
(}
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage M3364646
9250 05'11 71 t
p Vbw 1 of getting water out of the primary, and so discharging that
! 2 hot vessel-full of water, which is our primary source of 3 trouble.
4 If' we could get it in the circulatory system, can-5 we discharge coolant anywhere else?
6 MR. LICITRA: I believe the utility is going to 7 present an analysis which shows, using another me'thod, they
- 8 have the capability for mitigating, for example, a steam 9 generated tube rupture accident, which --
i 10 MR. EBERSOLE: Our trouble is that great volume L
! 11' of hot water that we can' t cool down, except by spray. One 12 way to cool it down would be to suck it down to where we l-O ta cou1d circ =1 e te, atca t 1 o o
- e tac 1 tory, 14 since you' re losing primary.
15 Anyway, let's go on. I now have a better feel 16 for a less desirable situation than I thought otherwise. Go i
17 ahead.
18 MR. LICITRA: The last two items on the Vugraph, 19 which are finding compensatory measures for restart, were 20 one, to assure that appropriate vendor reviews have been 21 made for the remaining power ascension test programs and, 22 two, the review shift complement, the remaining power 23 ascension tests to see if additional people on shif t would f 24 be wattranted.
O 2s wien eho e co ven tory mea ure , the ARC did i
i ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 136 4646
) 9250 05 12 72 i C~'DAVbw 1 issue a follow-up letter, saying that we believa these are f U-2 sufficient for restart, and pointed out that we did want to 3 see more on the final resolution of the issue.
4 MR. EBERSOLE: I believe before we have a break, 5 we have some remarks by Mr. Crews.
i 6 MR. CREWS The next item on the agenda deals
, 7 with the comments we have regarding either performance 8 appraisal of the Staff evaluation of Palo Verde and our 9 SALP. There have been no performance appraisals or staff 10 evaluations at Palo Verde, and our SALP period ended the 18 11 months ending September 30. We are convening a SALP board 12 next week, and so we dor.' t have any current SALP comments to 13 make.
( 14 MR. EBERSOLE: Is that all you're going to say 15 then? 16 MR. CREWS Yes, that is. 17 MR. EBERSOLE: Any questions up to this time? 18 (No response.) 19 If not, we'll have a break until 25 after. 20 (Recess.) 21 MR. EBERSOLE: Let's reassemble and get this 22 meeting back on the road here. 23 We are now on Item C, somewhat behind time, and , 1 24 we have scheduled the APSS design by T. Quan. ] () 25 (Slid e. ) ACE. FEDERAL REPORTERS, INC. i 202 347 3700 Nationwide Coversee 800 336 6646 I 1 w_
9250 05 13 73 1 MR. EBERSOLE: Mr. Quan, I understand that gggDAVbw 2 Arkansas Nuclear 2, for reasons I have never seen put down 3 on paper very well, adopted an expedient of putting in a 4 valve of willy-nilly on the pressurizer to get it down, 5 irrespective of whatever you-all do. 6 Could you comment, in the course of your 7 presentation, on your position on the presence of that 8 valve? I know these utilities. They' re like the medical 9 profession. They don' t like to advertise good things they 10 do, because it brings down the wrath on the others, but I'd 11 like to know whether you could put in that valve, or what 12 you think of it. f) 13 MR. QUAN: What we may do is defer that question xs 14 to CE. 15 MR. EBERSOLE: That'll be all right. 16 MR. QUAN: Good morning. My name is Terry Quan. 17 I am the licensing supervisor for Arizona Public Service. 18 Before I get into my portion of the presentation, 19 what I'd like to do is explain what we'd like to do with 20 Section 3(c) of the agenda. There will be several of ua up 21 here. I'd just like to try to tie the pieces together. 22 First off, I will start off discussing the design of the aux 23 pressurizer spray system, including the classification of 24 major components, the ASME code and the environmental (]) 25 qualification IE power, the seismic design, and so on. ACE. FEDERAL REPORTERS, INC. I 202 347 3700 Nationwide Coversee 800 336 6646 s
9250 05 14 74 gggDAVbw 1 After my presentation, Gary Waldrop will be back 2 up to discuss the September 12 event, where we had a loss of 3 charging pump sunction. He will also include a discussion 4 of the root cause evaluation of that event. 5 After Gary, Mark Radspinner from our Nuclear 6 Engineering Group will be up here to describe to you design
'7 enhancements for making to the aux spray and charging 8 portions of this CBCS system, which came out of our review 9 of the event, which we believe enhanced the reliability of 10 the aux pressurizer spray capability.
11 Finally, Jerry Sowers, from our Nuclear Analysis 12 Group, will be up to discuss steam generator tube rupture 13 events, particularly how the results are affected by the 14 delay of aux spray initiation and use of the pressurizer 15 vent system. 16 (Slide.) 17 To start off my presentation, I'd just like to 18 briefly summarize the purpose of the aux pressurizer spray 19 system, which is to provide a means of reactor coolant 20 system depressurization when the main pressurizer sprays 21 are not available. As far as design criteria for the 22 system, the critoria used were the criteria of Branch 23 Technical Position RSB5-1. This Branch Technical Position 24 for a Class 1 plant, which encompasses both the Palo Verde
^', 25 design and the CESSAR design, allowed operator 1ctions to ACE. FEDERAL REPORTERS, INC.
202 3474 700 Natsonwide Coverage 800 336 M46
9250 05 15 . 75 7' oAvbw 1 be taken outside the control rcom to overcome a single V t 2 failure and allowed operator actions inside containment 3 after the safe shutdown earthquake. It also remaining on 4 hot standby until manual action of repairs.are completed. 5 MR. REED: Just to keep ourselves clear on this 6 path, I'd like to point out, and I believe this is true, 7 that the main pressurizer sprays are frequently not 1 8 available, and the reason for that is that loss of main AC l 9 or outside AC, motor control centers or whatever that kills 10 the pumps, kills your so-called " main spray," nothing more l f 11 than a normal plant operating spray from scavenge to main 12 cooling pump head; is that correct? f) 13 MR. QUAN: That's correct. Any time you lose a v 14 reactor coolant pump, including loss of of f-site power; 15 that's correct. 16 17 18 19 20 21 22 23 24 25 ace. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coversee 800 ))6 6646
9250 06 01 76 1 MR. REED: So auxiliary spray functioning is not JggpAVbur 2 an infrequent requirement; it can happen frequently? 3 MR. QUAN: Whenever you use reactor coolant 4 pumps. 5 MR. EBERSOLE: Mr. Quan, also, I take it that l 6 your presentation wilA be based on the presumption that you 7 always have visible pressurizer level so you know where you 8 are, and I would like to have you consider a case where for 9 some magic reason that I won' t define that you no longer 7
- - i l 10 have visible pressurizer level and you resort both l
i 11 automatically and manually to drastic means to be sure that 12 the system is full, which inevitably will culminate in 13' a solid system. u(]) 14 MR. QUAN: We are currently discussing your 15 concern about operating the plant when we have a condition 16 where we have a solid pressurizer. We will be able to get 17 back to you after lunch and adoress your concerns. 18 MR. EBERSOLE: Thank you. 19 MR. MICHELSON: As long as you are interrupted 20 and since I haven' t necessarily followed the details of some 21 of the Palo Verde arrangements, could you tell me just 22 briefly the spray sparger arrangement and the pressurizer 23 for both the normal spray and for the auxiliary spray? 24 MR. QUAN: It is a spray nozzle, not a sparger 25 ring. (]) ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverses 8043364H6
9250 06 02 77 Vbur 1 MR. MICHELSON: It is a single nozzle at the very 2 top? The same nozzle for both functions? I 3 MR. QUAN: That is correct, and I will get into 4 that when I discuss the simplified schematics of the 5 system. l Thank you. 6 MR. MICHELSON: 7 MR. QUAN: Which is where we are now. 8 (Slid e. ) 9 This is the single pressurizer nozzle. As you 10 can see, the main spray would come in here. The aux ! 11 pressurizer spray is right here. So they do share a common 12 nozzle at the top of the pressurizer, to address your (~') x_- 13 question. l 14 MR. MICHELSON: What type of nozzle is this? l 15 MR. QUAN: The type of nozzle? l 16 Mike. l 17 MR. CROSS: I think it is just a straight pipe. 18 Mike Cross, from Combustion Engineering. 19 I just think it is a straight pipe. 20 MR. REED: I doubt very much it is a straight, 21 open-ended pipe. 22 MR. MICHELSON: I doubt it, too. That is what we 23 want to find out. 24 What is it? l () 25 MR. EBERSOLE: Will it blow the sprays out? l L ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coversee 800 336 4 & 6
r. l;9250 06 03 78 1 MR. MICHELSON: I am not alluding to anything. I
.jggp4Vbur 2 am just trying to get some information.
i 3 MR. QUAN: Give us some time. We will find an 4 answer for you. 5 MR. REED: While you are doing that for him, I : l 6 think we have to think about the reliability of this spray i 1 7 nozzle. I think that is what Carl is leading up to, and is i i 8 it equivalent to piping because it is a component really. 9 Thank you. l-10 MR. MICHELSON: What I would eventually want to l 11 find out a little bit about is in the unlikely event that l 12 there was a failure of this nozzle, what kind of spray t 13 configuration would remain and how would it be judged to be
.(])
14 adequate to do the spray job?
^
15 MR. QUAN: To refresh your memory on the aux L 16 pressurizer spray, I will start from th'e pressurizer nozzle I 17 and work back towards the refueling water tank. t 18 As I said, we have the aux pressurizer spray. It 19 is parallel solenoid operated valves to divert charging flow l 20 from the normal charging to the loop path. We require 21 isolation by valves 239 or 240. The flow coming back. i 22 CH-524 is a containment isolation valve, normally L 23 open. A fail as-is motor-operated valve. It can be ; 24 controlled from the control room. It does not receive the l () 25 containment isolation actuation signal. ACE FEDERAL REPORTERS, INC. ; 202 347 3700 Nationwide Covwage 8(4 336 4 M (
, - . . . = - - . - - _ _ _ .. _ _ . . _ _ _ _ . _ _ _ _ . _ _ _ _ _ _ - - .
, 9250 06'04 79 vbur ~1 This table here gives you a summary of the 2 classification,.if you will, of each of these major-
'3 components I just pointed out.
4 As you can'see, all of them are Section C, either 5 . Class 2 or. Class 1. All of them are Seismic Category 1. 6 CH-139 and -240, since they are isolation valves, 7- they are fail' closed valves. We have nonlinear power to i. 8 those valves. ? 9 CH-203 and -205, the redundant auxiliary spray
'10 valves, receive power from redundant I-E powar sources.
I 11 All of these valves that are actuators are l 12 environmentally qualified. { 13
'14' air system?
MR. EBERSOLE: Are they connected to the common 15 MR. QUAN: These valves here, 239 and 2407 Yes, 16 they are tied to the instrument air system. 203 and 205 and 17 524 are not tied to the instrument air system.
- 18 MR. EBERSOLE: What are they, just direct i
19 solenoids? 20 MR. QUAN: Yes. l 21 MR. EBERSOLE: Thank you. l 22 MR. REED: I would like to hear some more about 23 the regenerative heat exchanger. You realize an auxiliary l- 24 spray is not something new and novel; it exists on some
] '25 plants. And in those plants that I am familiar with, the ACE FEDERAL REPORTERS, INC.
202 341 3700 Nationwide Coverage 800 33H646
9250 06 05 80 p Vbur ~ 1 auxiliary spray line does not come off, shall we say, 2 downstream of the regenerative heat exchanger. It comes off 3 down in the solid piping, and it do'esn't use the 4 regenerative heat exchanger as a component to'go through. 5 You may say, well, that gives you a . lot colder 6 water and you get a more thorough shock and that put more 7 stress on the nozzle. And of course you are going to have 8 stress on the nozzle, anyway, because you can have no 9 letdown flow and have thermal shock. l 10 One of the problems with components, of which the 11 regenerative heat exchanger is a component, is such' a thing 12 , as tube failure. () 13 I am just wondering if in your design analysis 14 you thoroughly went through " Sat happens with the L 15 regenecative heat exchanger tube failure because it is going 16 to see lots of transients. In this auxiliary spray scheme 17 you are going to see more transients on the regenerative 18 heat exchanger than you would on a normal PORV type of 19 pressurized water reactor. 20 Have you gone through all the happenings and 21 their impacts with respect to that regenerative heat l 22 exchanger on the availability of spray and whether or not l L 23 you can get spray to a depressurizer? 24 I am pointing out that certain tube values can (]) 25 get you around in a circle. You are charging pump flow ACE. FEDERAL REPORTERS, INC, 202-347 3700 ~ NationwWe Coverage 800 33 H 646 (
t 9250 06 06 81 pAVbur 1 right back to the letdown line and you get nothing coming 2 out in the spray. 3 MR. QUAN: I will defer the question to Rick Turk 4 from Combustion Engineering. 5 MR. TURK: Rick Turk. I guess I would point out 6 that on the 'CE plants, the earlier plants that do have s 7 PORVs, they also have an aux spray system. 8 We would not expect Palo Verde to use their aux 9 spray system any more than those other plants because they
~
10 would not open a PORV just to cool down the pressurizer to 11 shut down the coolant. 12 Aux spray has always been part of the CE design,
') 13 and the thermal transients associated with aux spray have 14 always been considered, and in f act in the 14 questions we 15 got from the NRC on'the PORV issue, one of the very specific 16 questions we went into detail on in the report, CEN-259, was 17 the thermal transients associated with the aux spray.
18 So, yes, that has been looked at, it has been
-, 19 considered, and I wouldn' t expect it to be any dif ferent in 20 Palo Verde than any previous design.
21 MR. REED: You are saying, I guess -- you didn' t 22 really answer the question, but you are saying this, that 23 the transients the regenerative heat exchanger may receive { 24 are not going to be any more than with a PORV PWR? (} 25 MR. TURK: In other words -- ACE FEDERAL REPORTERS, INC.
- 202 347-)?00 Nationwide Coverage 800 336-6646
9250 06 07 82 JggDAVbur 1 MR. REED: In other words, your long history 2 shows that the spray valves have never opened; is that what 3 you are telling me?, 4 MR. TURK: Not at all. I am saying that no plant 5 with PORVs and aux spray has ever used the PORV in lieu of 6 the aux spray. The aux spray has a specific purpose and 7 always has. 8 MR. REED: In your design you always had aux 9 spray? 10 MR. TURK: Correct. 11 MR. REED: And it has always been piped that way? 12 MR. TURK: It has always been piped that way. () 13 MR. REED: So you have a single component here 14 that could fail and fail the auxiliary spray? 15 MR. TURK: The nozzle? 16 MR. REED: No, the regenerative heat exchanger. 17 MR. TURK: Due to a tube leak? 18 MR. REED: A rupture, yes. A heat exchanger 19 rupture or tube leak, whatever you want. 20 MR. TURK: I guess I will have to think about the 21 tube rupture case. 22 MR. REED: You mean you haven' t thought about it 23 before? 24 MR. TURK: I personally haven' t thought of it in () 25 this context right here, no. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336 6646
9250 06 08 83 jlhDAVbur 1 MR4 REED: I think, like the nozzle, it can 2 jeopardize the success of the auxiliary spray. 3 MR. QUAN: I would like to go on to the 4 simplified scheraatic of the sources lof water. i
, 5 MR. EBERSOLE: Can you go back to that drawing l
- 3 6 just a moment, please?
7 I see two inputs to it. Tell me now,'the inp'uts 8 to the spray nozzle up at the upper right, I have reactor m 9 coolant main spray valves. Is that from the main coolant 10 p'wnp? 11 MR. QUAN: Yes. 12 MR. EBERSOLE: That is the one you lose when you () 13 trip the main coolant pumps. 14 Now, what 'do I have down at the bottom, down 15 there. Normal charging to Loop 2-A? 16 MR. QUAN: The flow is in this direction 17 We normally, and that is the gormal loop charging line. 18 isolate this line by closure of one of these two valves to 19- divert charging flow to the aux spray valve. 20 MR. EBERSOLE: Those are non-1-A valves. So as 21 far as I am concerned they can be hooked into the same 22 supply?
'23 MR. QUAN: They are fail close, air-operated 24 valves. So the velocity of either the power or air would
()- 25 close one of those valves -- either one of those valves. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
9250 06 09 84 jlgDAVbur 1 MR. EBERSOLE: Now, can I be assured that they 2 guarantee the loss on a non-1-A circuit? 3 MR. QUAN: We do have overvoltaga protection en 4 those circuits to assure that we won't get faults in the 5 power system. 6 MR. EBERSOLE: You have what -- are those 7 manually controlled? 8 MR. QUAN: Those are controlled from the control 9 room in normal operating conditions. 10 MR.'EBERSOLE: By manual remote independent 11 switches? 12 MR. QUAN: Yes. () 13 MR. EBERSOLE: So you could deenergize them from 14 the control room and they would clo'se? 15 MR. QUAN: Right. 16 MR. EBERSOLE: If the air system is working 17 right? 18 MR.-QUAN: If the air system failed, the valves 19 will also close. 20 MR. EBERSOLE: We just yesterday heard of cases
- 21. where air systems fail at pressures midway between normal 22 and high, in which case they don' t work at all.
23 What happens when that takes place? 24 MR. QUAN: We will have to take a look at that () 25 and see what pressure degradation takes place. t ACE-FEDERAL REPORTERS, INC. 202-34' 3700 Nationwide Coverage 800 336 46M
I 9250 06 10 85 :
%lgDAVbur 1 MR. EBERSOLE: This happened to be on some scram 2 systems on boilers. When they failed intermediate to low 3 pressure because of leakage or incapacity of the air 4' systems, then they float between operating and 5 nonoperating. They don't work at all.
6 So there is another mode failure in which they 7 ' stay open. 8 If they stayed open, so what difference does it 9 make? You still have the spray from the main coolant pumps? 10 MR. QUAN: Right. 11 MR. EBERSOLE: And the check valve will prevent 12 loss of that, okay. ( 13 Thank you. 14 (Slid e. )
'15 MR. QUAN: I would like to go on to the borated 16 water source slide. <
17 Again going back from the pressurizer, three t 18 charging pumps. It takes normal suction. Normally, it 19 takes suction off the volume control tank. If the volume 20 control tank is not available and offsite power is 21 available, charging suction will be through the boric acid 22 makeup pumps, opening of CH-514 through that check valve. 23 If offsite power is not available, charging 24 suction is provided by gravity feed line closure of 501 and s () 25 opening of CH-536, taking suction off the refueling water ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
9250 06 11 86 1 1 ' tank through this check valve, providing suction to the 1
'{}DAVbur 2 charging pumps.
3 This is the preferred gravity feed line. There 4 are several others if this one is not available. 5 For example, a suction line can be established l 6- off the safety injection suction line, opening of CH-327, 7 and valves 757, 756, or 755, which are associated with each 8 of the 'three charging pumps.
- 9. And there are also other gravity feed lines that 10- are available which can be manually aligned.
11 MR. REED: It seems to me in the situation where 12 you have to be concerned, you have a single valve - I guess (]) 13 it is 536 -- in line to provide you with the fluid that you 14 want for spray, auxiliary spray. It looks to me like your 15 other lineup are manual valves in the closed position, is 16 that correct? 17 MR. QUAN: That is correct. 18 MR. REED: How.do you feel about that? You have 19 a single valve, and the valve reliability is not that good.
- 20 MR. QUAN
- We feel that there is ample time, if 21 this valve were not available or were not able to open, in 22 which to _ establish a gravity feed line to the charging 23 pumps.
24 MR. EBERSOLE: Do I understand that you have lost 25 ( )~ gravity to the charging pumps already? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
l { 9250 06 12 87 T' DAVbur 1 MR. QUAN: Yes , we d id . EV 2' MR. EBERSOLE: How can you do that with so many 3 suction leads?- 4 .MR. QUAN: What it is related to -- and Gary
' 5- Waldrep will get into this in quite a bit more detail -- but 6- it had to do with the volume control instrumentation. We 7 had a faulty reading which misled the operator and also did 8 not close this valve when the actual level was at the low, 4
9 low level setpoint. 10 MR. E3ERSOLE: Has there been a mini PRA done on 11 this whole system, and was this contradictory to the 12- findings of that? () 13 . R. QUAN: M We did do a reliability analysis, both I 14- in CM-239, the rapid depressurization and decay heat removal , 15 study that the CE Owners Group put together; we also did a t 16 ' reliability study af ter the September 12th event to identify 17 those areas in which we could improve reliability. 18 MR. EDERSOLE: Was this event predicted in the 19 PRA?
.20 MR. QUAN: The volume control tank level was not 21 specifically modeled, if you will, but the'PRA work that CE 22 -did do did allow for a certain probability of losing suction 2,3 to the charging pumps.
24 KR. EBERSOLE: Was that consistent with it
.( )
25 actually having been lost? ACE-FEDERAL REPORTERS, INC. Nationwide Coverage 800 33H646 b 202 347-3700
'9250 06 13 88 lf~'DAVbur 1 MR. QUAN: As far as'I know, yes. So we were not
-v 2 surprised.
3 Rick Turk will be getting into this a little bit 4 more this af ternoon, concerning the effect of the September 5 12th event on the CE work that was done concerning PORVs. 6- MR. EBERSOLE: Thank you. 7 MR. WYLIE: Let me ask a question. You normally
,- 8 align to take suction out of the volume control tank, right?
9 MR. QUAN: Under normal operation, that is
-10 correct.
11 MR. WYLIE: Then when you need the' auxiliary 12 . spray, you pull suction automatically out of there? 13 MR. QUAN: That is correct. (]) 14 MR. WYLIE: How much storage do you have there?
'15 How long do you run that way?
. 16 MR. QUAN: Normally, I think it is on the order 17 of 12 to 15 minutes. 18 MR. WYLIE: Then you transfer to the refueling 19 within 12 to 15 minutes? 20 MR. EBERSOLE: What is the point of taking it 21 from a little tank like that when you could take it from the 22 fuel in the first place? 23 MR. QUAN: Under normal operating conditions the 24 volume control tank blends both borated water through this 25 (]) path and unborated water through this path. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
9250 06 14 89 JggDAVbur 1 MR. EBERSOLE: So you get the blending? 2 MR. QUAN: Correct, under normal operating l 3 conditions. 4 Abnormal conditions, in which aux spray or 5 charging is required as a compensatory measure for the short 6 term, the operators are instructed or have instructions to 7 isolate the volume control tank or close CH-501 from the 8 control room and, by opening CH-536 also from the control 9 room, to align suction to the refueling water tank. 10 MR. EBERSOLE: But they didn' t do that? 11 MR. QUAN: This was a measure that was 12 bnplemented in procedures af ter the September 12th event. r (m) 13 Also, later in the presentation Mark Radspinner 14 will be talking about design enhancements that we intend to 15 implement, which will provide an automatic switching, if you 16 will, of the suction line. 17 MR. REED: I am not so sure that your change in 18 procedure hasn' t reduced your ability to spray. What you 19 are saying now is you have changed the procedure to shut off 20 the volume control bank as a source of water, be it borated , 21 unborated, or whatever. But it is still good for killing 22 steam bubbles, spray water. You are going to shut that off 23 now, yet it was a parallel valve and pipe system to give you 24 suction to the charging pump suction header, rm () 25 Now, you have turned yourself -- by operating ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-6646
9250 06'15 90 1 procedure, you are going to shut that valve off the volume ()DAVbur 2 control tar'. and say that a single component, the 536, is 3_ going - to open in this timeframe and your backup is to run 4 ~ around and open four .of the valves that are manual valves. 5 I am not so sure if you did a PRA which one has 6_ got more reliability to deliver some kind of water to the 7 top of the pressure.
-8 MR. QUAN: The preferred suction source under 9 abnormal conditions for the charging pumps would be the 10 refueling water tank.
11 MR. REED: 1 can understand because it is 12 borated. (). ?l3' MR. QUAN:- Correct. 1so, it is a tech spec 14 tank. There is a considerable amount of water. The volume 15 control tank is a normal operating piece of equipment. The 16 level can be fluctuating. 17 MR. REED: Understood. But af ter your design was ~ 18 complete and everything, your first patch -- we will call 19 it, to use Jesse's words here -- your first patch is to 20 change the operating procedure and have the people shut off 21 the volume control even though .i t might be a viable source 22 of spray. 23 Now, you turn over and go to a single valve, 536, 24 and say "Thou must open to give me that spray I just shut () 25 off." It is better water in that it is 2000 ppm, perhaps, ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-3364646
'9250 06 16 91 h Vbur 1 of borated water. 2 Do you feel that single valve is going to open 3 reliably? 4 Of course, you have got the backup of sending 5 people around to these manual valves. I am just wondering 6 if you have completed your improvement. Maybe all these 7- manual valves now have to be motor-operated valves, but I 8 have a better idea than that. 9 MR. QUAN: This procedure change in which the 10 operator can close the CH-501 and open 536 --
~ 11 MR. REED: Tries to open 536.
12 13 14 15 16 17 18 19 20 21 22 23 24 25 ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(H5646
07 01 92 1 DAVbur 1 MR. EBERSOLE: That provides a deborated solution 2 to the reactor when you go up? 3 MR. QUAN: Or a more controlled boration source. 4 MR. MICHELSON: Why are you closing Valve 5017 5 MR. QUAN: We would like to isolate the volume 6 control tank for the reason that there is a hydrogen 7 overpressure. So we would like to isolate that. 8 MR. MICHELSCN: In the unlikely event that 501 9 fails to close, which is another possible valve failure in 10 some position, what would happen? 11 MR. QUAN: 501 is a valve which is normally
' ~' If it were to fail to close in the control room,
; 12 operated.
+/
13 the operator can be dispatched to close that valve 14 manually. 15 MR. MICHELSON: That is a single indication? 16 MR. QUAN: That again takes time. 17 MR. MICHELSON: In the meantime, the hydrogen 18 might be entering the charging pumps already. So it gets to 19 be a race of time in all of this. 20 Have you figured out these time raises and 21 assured an adequate margin for these manual operations? 22 MR. EBERSOLE: Carl, it occurs to me that what 23 has happened here is several things. They have invoked the 24 charging pumps, the safety pumps. Normally, they are just a (a~') 25 rcutine makeup pump. They have now attempted to disregard ; 1 ACE-FEDERAL REPORTERS, INC. I 202-347-3700 Nationwide Coverage 800 336 6646
l l 9250 07 02 93 1 the various ramif.ications of design that prevent the }gpAVbur 1 2 charging pumps from being safety grade in its integral 3 sense. 4 One of these is the thing that you just 5 mentioned, the single suction intake, which can instantly, 6 as f ar as I am concerned, gasify the charging pumps and 7 destroy a now newly existing safety function which wasn't 8 there before we had this auxiliary charging lead. 9 MR. MICHELSON: I don' t know how soon this can 10 happen, but I gather earlier you said in 12 minutes you 11 needed to make the transfer. But I am not sure how close to 12 sucking the gas you are at that point. 13 MR. EBERSOLE: There has been a subtle change ([ ') 14 here that changes the charging pumps to safety grade 15 leads. 16 MR. MICHELSON: Oh, it is not subtle. 17 MR. EBERSOLE: Well, I know, but originally, you 18 know, these things pump in a little water to make it up and 19 they really don't make up the safety grade flow. They have 20 deborated water. They have kind of been in the shadows. 21 MR. QUAN: There's two methods in which 501 will 22 close. There is an actuation off the volume control tank 23 level instrumentation, which Mark Radspinner will also 24 discuss a little bit more. Or regardless of volume control () 25 tank level, the operator can close that valve from the ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
9250 07 03 94 P DAVbur. 1. control room. V
.2 MR. MICHELSON: In the unlikely event-he has to
'3 go to the remote location to close it, how much time does
'4 that take and how does that compare with the possibilities 5~ of drawing gas as a suction to the pumps?
6' MR. QUAN: I think the situation you are 7 describing we did go through on September 12th, and I think 8 if you will wait a few minutes and let Gary go through the 9 scenario, I think you will understand what the operators 10 have to go through to restore the charging system; also, our
-11 ability, as demonstrated , to do that.
12 MR. MICHELSON: What size valves are these that 13 are involved -- like 501 and 536?
-(}
14 MR. QUAN: I believe thpse are four-inch valves. 15 MR. MICHELSON: And they are Limitorque l 16 motor-operated? 17 MR. QUAN: Correct. ; 18 MR. REED: I might wonder if the situation in
- 19 startup testing is going to be the same as in the normal 20 run, a little bit of a fuel-fail situation. I assume the 21 501 is in the high rad area and it is a lot lower area. I 22 am just wondering about the access to that valve on the 23 timeframe that you are talking about.
24 MR. QUAN: You are correct. It is behind a 25 locked door. ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800-336 6646
I , 9250 07'04 95 1 MR. MICHELSON: I would guess the defense would gDAVbur 2 be: "So what is the barrier, anyway?" 3 Is that right? 4 MR. QUAN: That is part of our defense. That is 5 correct. 6 MR. WYLIE: Let me ask a question. I believe 7 your fix earlier, you didn't have those two valves on a 8 non-1-E source and now you have moved them over on the 9 safety-related motor control center Class 1-E. But you have 10 got them in parallel off the same breaker, is that correct? 11 MR. QUAN: That is correct. Mark Radspinner will 12- have a diagram, and he will discuss that. 13 MR. WYLIE: I have seen that. But in the event (]) 14 536 is frozen, then it trips it and then you don' t open 15 501 -- or you don' t close 501. If you try to open 536 and 16 it is frozen, then that trips out both figures? 17 MR. QUAN: Not necessarily. There is a 18 protection relay between the bus and the motor actuator and 19 between 536 and 501. 20 Any other questions? 21 (No response.) 22- MR. QUAN: If not, I will go ahead and P. urn the 23 podium over to Gary Waldrep, who will go into detail on the 24 ' September 12th event. 25 MR. MICHELSON: Before you leave, though, a (]) i ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 33HM6 !
- 9250 07f05 96 vbur 1 clarification.
2 When you talk about the Seismic 1, the entire 3 piping system is Seismic 1, isn' t it? 4 MR. QUAN: That is correct. 5 MR. MICHELSON: Thank you. 6 (Slid e. ) 7 MR. WALDREP: I would like to go through now and 8 present the event which occurred on September 12th and go 9 through basically the initial conditions and then a tLae 10 line of the sequence of ever".s of that, and maybe you will 11 better understand where we are coming from and what the 12 results were. {} 13 We were prepared to do a load rejection, nonclass 14 loads, on the main generator. We were at approximately 50 15 percent power. We initiated the large load rejection. 16 What we expected to see in the test was not to 17 get a reactor power cutback actuation and not to get a 18 . reactor trip. 19 What we actually got was a reactor trip occurring 20 and again the reactor power cutback, since we did not 21 actuate the drop rods. 22 (Slide.) 23 So if we look at a time line of this event, we 24 see that we opened the generator output br.eaker to initiate (} 25 the event. This caused the turbine stop valves to close. ACE-FEDERAL REPORTERS, INC. 202247-3700 Nadonwide Coverage 800 33(H5646
9250 07 06 97 AVbur 1 clarification. 2 When you talk about the SEISMIC I, the entire 3 piping system is SEISMIC I, isn't it? 4 MR. QUAN: That is correct. 5 MR. MICHELSON: Thank you. 6 (Slide.) 7 MR. WALDREP: I would like to go through now and 8 present the event which occurred on September 12th and go 9 through basically the initial conditions and then a time 10 line of the sequence of events of that, and maybe you will 11 better understand where we are coming from and what the 12 results were, f 13 We were prepared to do a load rejection, nonclass v 14 loads, on the main generator. We were at approximately 50 15 percent power. We initiated the large load rejection. 16 What we expected to see in the test was not to 17 get a reactor power cutback actuation and not to get a 18 reactor trip. 19 What we actually got was a reactor trip occurring 20 and again the reactor power cutback, since we did not 21 actuate the drop rods. 22 (Slide.) 23 So if we look at a time line of this event, we 74 see that we opened the generator output breaker to initiate ( ') 25 the event. This caused the turbine stop valves to close. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
9250 07 07 98 qlgDAVbur 1 At this time, as I mentioned in the initial 2 conditions , the house loads or the nonclass house loads are 3 still on the main generator. So with no more steam 4 available to the main generator, it begins to decrease in 5 speed. As it does, the reactor coolant pumps also begin to 6 decrease in speed. 7 MR. EBERSOLE: Pardon me if I refresh my memory. 8 Do the critical house loads -- are they normally run on 9 nongenerator feeds? 10 MR. WALDREP: Normally, the nonclass loads are on 11 the generator. 12 MR. EBERSOLE: I am talking about the class () 13 loads.
'14 MR. WALDREP: Class loads are provided from 15 offsite power.
16 MR. EBERSOLE: Thank you. 17 MR. WALDREP: As I said, as the reactor coolant 18 pumps began to decrease in speed due to the decrease in 19 speed of the main generator, we got the reactor trip and, of 20 course, our reactor coolant pump trip, which began the 21 natural circulation process. 22 The next major point right here is tne 23 restoration of station loads. This involves the operator 24 bringing back all the station loads through the offsite ( '; 25 power network. ACE FEDERAL REPORTERS, INC. 202-347-3700 Natsonwide Coverage 800-336-6646 l
19250 07E08 99 17 DAVbur 1 MR. EBERSOLE: What is the main feedwater system - (_) 2 doing in the meantime? 3 MR. WALDREP: The main feedwater pumps tripped on 4 low suction pressure because the condensate pumps tripped 1 5 off because they lost power. 6 MR. EBERSOLE: So you always lose the main feeds 7 because of loss- of the- booster. pumps? 8 MR. WALDREP: That is correct, and with that loss 9 of main feed the operators started the third auxiliary 10 feedwater pump to maintain steam generator levels. 11 So as I was going to say, the restoration of 12 station loads caused the main steam line drain valves to 13 open, or to reopen. We explain that in just a minute. ({} 14 These are eight two-inch drain -line valves that 15 open on a turbine trip to drain any condensate that might 16 have formed in the steam lines. It is a protective measure 17 to protect the turbine. 18 The operator, as you see up here, two minutes 19 into the event, had already closed those valves. When the 20 valves were repowered, they reopened. 21 MR. EBERSOLE: With zero af ter heat, doesn' t this 22 invite a chilling effect? 23 MR. WALDREP: That is correct, and you will see 24 the next step after that is that we got the safety 25 injection. So you are absolutely correct. (]) ACE-FEDERAL REPORTERS, INC. 3T2-347-3700 Nationwide Coverase 800-336 6646
50 07:09' 100 1 . DAVb r - 1 The opening of those valves caused the cooldown 2 of the RCF, which got us into the safety injections. 3 MR.-EBERSOLE: That threatens the level function 4 in the_ pressure loop, which you have to stop by hand, right? 5 MR. WALDREP: It didn't really threaten the
'G pressurizer level because we maintained pressurizer lev.el.
7 MR. EBERSOLE: But you did it by looking at it 8 and turning off the 3afety injection? The motor doesn't do 9 that, does it? 10 MR. WALDREP: We didn't have to take any manual 11 action in order to maintain steam pressurizer level. 12 MR. EBERSOLE: What about safety injection?
,]
13 MR. WALDREP: The safety injection pdmps ! 14 ' started. The pressurizer level is 50 percent to start-with, , 15 and it decreased when we had the reactor trip to 20 percent 16 or so. 17 HPSI came on, and HPSI began to inject water into i 18 the system.- The pressure of the primary system right now is 19 --2250 pounds, a little more, due to the trip. The HPSI pump 1-20 discharge pressure is around 1850 or so. 21 So with the pressure below the setpoint of HPSI, l 22 it injected some water until it was into the deadhead. 23 MR. EBERSOLE: HPSI cannot buck 2250 pressure? l 24 You have' no pressures to induce water in the primary other 25 than the charging pumps at standard pressure? ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 33H646 [
l 1 1 9 50 07 10 101 1 DAVbur 1 MR. WALDREP: That is correct. 2 MR. EBERSOLE: So you depend on depressurization 3 even to get safety injection? 4 MR. MICHELSON: You said the limit was 1800? 5 MR. WALDREP: 1837 is where the HPSI comes in. 6 MR. EBERSOLE: So the safety of the plant then 7 rides on the charging pumps? 8 MR. WALDREP: I don't know if I understand. 9 MR. EBERSOLE: They are the only pumps that can 10 get water in the plant as long as it is at standard 11 operating pressure? 12 MR. WALDREP: That is correct.
~'
13 MR. EBERSOLE: Ye t they are really fundamentally 14 not safety pumps, or classically they aren't; they are just 15 makeups. 16 We have some interesting philosophical 17 considerations here. 18 MR. WALDREP: They are Class 1-E power. 19 MR. EB ERSOLE : Oh, I know, but they have these 20 curious ways of going dead, like gasifying. 21 MR. WALDREP: Let me go through that, and we will 22 explain the gasifying. 23 MR. MICHELSON: While we are on the charging
, 24 pumps, another clarificaticn. These are the piston type?
L) 251 MR. WALDREP: Yes. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80 4 336-6646
i 4. c i 9250-07'11 .102 DAVbur 1 MR. MICHELSON: Were they originally put in as L2 :nonsafety and. then later upgraded to SEISMIC I, and so 3 .forth? They have always been SEISMIC and the have always 4 -been OA control; they have always been considered safety 5 grade, is that right? 6 Thank you.
- 7 MR.' WALDREP
- Getting back to the scenario, so we
- -8 ,had the main steam line drain valves reopen, which gave us 9 the cooldown and subsecuent safety injection signal. 10 We got HPSI. All the safety injection pumps 11 started operating as required. They injected water into the. 12 primary and brought the pressure back up to the setpoint, as O,, 13 they should. 14 All these systems functioned entirely as 15 designed. l 16 Okay, at this point we began to lose the VCT 17 level. Now, as we bought out before, the VCT only had one 18 reference leg, which had given the operator the incorrect 19 reading of the 20. percent level. I 20 MR. EBERSOLE: Right. 21 Well, go ahead. I will' wait till you get to the 22 end of this scenario here. 23 MR. WALDREP: The operator thought he had the 20 l 24 percent level, and that is what it read throughout the O 25 event. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 33 H 646 J ...
9250 07 12 103 11 . The VCT level was lost, which resulted in a loss gggDAVbur 2 of suction water to the charging pumps. The operator 3 realized. this by the erratic operation of the pumps at the 4 discharge pressure and the lack of discharge flow. 5 -So he then. secured all three of the charging 6 pumps. He then proceeded to line up the gravity feed, as we 7 looked at earlier, from the refueling water tank. 8 MR. MICHELSON: How long did that take?
-9 MR. WALDREP: On the order of -- in this case, we 10 had lost offsite' power, and then we had a safety injection.
11 So that stripped those valves that are controlled from the 12 . control room. Then operation was done locally. 13 MR. MICHELSON: How long did that take?
- ( )
14 MR. WALDREP: 10 to 15 minutes, something on that 15 order. 16 MR. REED: But it says an hour and a half before 17 you started the charging pumps B and C. 18 MR. WALDREP: That is correct. Like I said, they 19 tried to line up the gravity feed. In fact, they did line 20 up the gravity feed fran the refueling water tank and then 21- bled off the pressure on the section of the charging pumps 22 in an attempt to regain suction water flow from suction 23 water to the charging pumps. 24 We didn' t have any luck in doing that. So they 25 (]) ' realized that the hydrogen overpressure must be what is ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 6646
9250 07 13 104 1 causing them -- you know, the gas in there was voiding that qggpAVbur 2 space so they couldn' t get water from the refueling water 3 tank. 4 In fact, that is what happened. 5 MR. MICHELSON: Why couldn't it be vented off? 6 Because it is not in the right places? 7 MR. WALDREP: It could have been vented off the 8 regulator that admits hydrogen into that VCT tank. 9 MR. MICHELSON: You are trying to vent the pamps 10 themselves, aren' t you? 11 MR. WALDREP: In the suction line of the pumps. 12 MR. MICHELSON: You did that, and it was ( ', 13 unsuccessful, I thought? L! 14 MR. WALDREP: That is correct. 15 MR. MICHELSON: Why was it unsuccessful?. 16 MR. WALDREP: At that point they had already 17 closed the 501 valve, isolating the VCT. So there is no 18 more hydrogen getting in. However, there was higher 19 pressure hydrogen in those suction lines than there was head 20 of the RWT to push that. 21 MR. MICHELSON: So we are still blocking it, and 22 you couldn' t vent that particular part of the hydrogen 23 because your vent lines weren' t in the right places for that 24 purpose?
/~')/ 25 MR. WALDREP: For whatever reason.
\
i ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800 3364646
. 1 9250 07 14 105 yllpAVbur 1 So what we did is we started our boric acid 2 makeup pumps up. After the power was restored to those, 3 this allowed us to overcome that pressure. We vented the 4 system, and then we were able to start two charging pumps. 5 MR. MICHELSON: You are going to tell us later 6 what you are going to do to be able to vent in the future 7 without this difficulty? 8 MR. ADNI: Bob Adni, from Arizona. 9 What had occurred is they had isolated 501, the 10 outlet of the VCT. They sent an operator down to vent the 11 line. They weren' t familiar with how long it would take. 12 They were working on two parallel paths on getting the power
) 13 restored to get the normal boric acid makeup pumps running 14 and the charging line, the suction line.
15 The operator didn' t realize how much pressure was 16 in the line. He didn' t have the valve open far enough to 17 vent, and they thought it would take an extremely long 18 period of time to vent that particular portion of the line. 19 So they concentrated their efforts on restoring 20 power and starting the boric acid makeup pumps to supply the 21 suction. 22 MR. MICHELSON: I guess sometime today you are 23 going to tell us why this is no longer a problem in the i 24 future? I s q,) 25 MR. ADNI: Right. l l ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
L 9250 07 15 106 DAVbur 1 The operator had just cracked the valve. He was
}
{ 2 a little wary of the situation. 3 But there is a fairly short run of pipe that is
- 4 pressurized, and they have three-quarter inch vents on l
l 5 them. There is no usually no problem venting that pressure l 6 off the line. 7 MR. REED: Where does it vent to? Is it piped to i 8 a safe place, or is it locally hydrogen vented? l 9 MR. ADNI: There's two types of vents, one that I 10 would vent right to a drain tank, and there's other vents on 11 the line that vent right to the atmosphere. 12 MR. REED: I can understand why he was a little (^') 13 cautious. x_/ 14 MR. WARD: Mr. Adni, did the operator have a t 15 procedure to guide him in this' step? 16 MR. ADNI: Not in that particularly evolution. 17 MR. WALDREP: To finish up the time line, we now 18 had the charging pumps restarted. The plant was stable, and 19 we went about our resetting of our safety injection and 20 plant stabilization. 21 MR. EBERSOLE: Mr. Waldrep, let me ask you to run 22 through a little bit of a hypothetical action here. 23 I have a primary system leak somewhere low in the 24 system. So I lose primary fluid. I don't lose primary 25 pressure because I have got this hot leg out there that q) ACE FEDERAL REPORTERS, INC. 2 2 347 3700 Nationwide Coverage 800 336-6646
9250.07 16 107 7^'DAVbur 1 keeps restoring it. So I continue to lose inventory, the G 2 level comes down, but I can' t make up except through just 3 the charging pumps and they are not big enough to make up 4 much of a leak. S I suppose it is quite small, their capacity to 6' restore leakage? 7 MR. WALDREP: Three pumps 44 gallons a minute. 8 MR. EBERSOLE: That is not much of a leak. 9 So my inventory must fall because I haven' t got 10 any way to make that up except the charging pumps, and I 11 have exceeded their capacity. 12 So what has to happen then, the level has to go 13 down along with everything else, and it comes on down. But (]) 14 I still have this hot plug of water and I hold this pressure 15 until I get down into and begin to really suck out the 16 pressurizer hot water into the full field of cooling as a 17 result. 18 In short, I lose my hot leg. At that point, 19 somewhere along the line, the high pressure charging pumps 20 then begin to take over, is that correct? 21 MR. WALDREP: That would be correct. 22 MR. EBERSOLE: How far dokn is that, and how 23 excited would you get as you lose level? 24 HR. WALDREP: The setpoint for safety injection () 25 is 1837. ACE. FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
I l 9250 07 17 108 jlgDAVbur 1 MR. EBERSOLE: Where is my level at this point? 2 It is gone, isn' t it? I can' t see it any more? 3 MR. WALDREP: Depending on the scenario, how fast 4 it happened. It could be low, I am not sure. I don' t have 5 a feel for it. 6 MR. EBERSOLE: So you run straight blind into 7 this and hope the pumps come on and take you out? There is 8 a moment of invisibility where you are? 9 MR. MICHELSON: You are in a lit':le bit of a 10 race, also. If you really had decay heat in quantity, you 11 would be in a race as to whether you take the heat out 12 faster, whether you are cooling down or pressuring.
) 13 MR. EBERSOLE: Whatever I do to the primary loop, 14 in a temperature reduction sense it doesn' t affect the 15 primary pressure until I get the primary pressure volume 16 down to where it is a part of the process.
17 MR. MICHELSON: I think you can, depending -- 18 MR. EBERSOLE: I am still holding pressure up 19 because of my hot water boiler and I don' t have any way to 20 break it up until I drop to 1800, and by that time the 21 pressurized volume has had to go down and get into the 22 primary flow process, hasn' t it? 23 MR. MICHELSON: I don't think it is possible to 24 drop -- for that level to drop way down and even be at high () 25 pressure yet. ACE-FEDERAL REPORTERS, INC. 202 347-1700 Nationwide Coverage 800-336-66 4
8' 5
,* A LA250'07 01 . 109 Vbur ~1' MR. EBERSOLE: That is what I.am saying, 2 precisely.
.3 MR. WALDREP: So you are talking basically the
+
y /
~
s 4 small break LOCA? N 5 , MR. EBERSOLE: In the context that cooling the e 6 per.fary does not reduce pressure. 7 8 9 10 11 , 12 / h ' 13 14 s .
^
,- 15 .; .
., 'C ,,
^
16
.s,..,-
w r p s 17 . . . 4 di 18
'l 19 s
~
,- 20
- 21 '
. > ~
. . 22.. :.+ ,
,n 23 ,.
< y,
~
.- r .
24 ^ ^ ' O 25 ACE-FEDERAL REPORTERS, INC. j 202-347 3700 , Na@nwide Coverase 800 336 4646 l
9250 08 01 110 1 MR. ADNI: Once again, we're talking about a jggDAV/bc 2 small break LOCA making up. We have 120 GPM from the 3 charging system and if the break is 150 GPM, or whatever, 4 it's beyond the capacity of that. You' re right. We're 5 going to be losing inventory and going to be concentrating 6 on maintaining inventory in the primary. 7 We' re also going to be cooling the system. 8 MR. EBERSOLE: But cooling it is not reducing. 9 You're enhancing the shrink. 10 MR. ADNI: That's correct. That's going to drop 11 the pressure. 12 MR. EBERSOLE: And you must go into a blind mode () 13 of operation.
- 14 MR. ADNI: I wouldn't necessarily call it a blind 15 mode.
16 MR. EDERSOLE: You don' t know where the inventory 17 is except for the new system y\ou put in. 18 MR. ADNI: Or VLMS. You have indications of 19 subcooling. 20 MR. EBERSOLE: That comes into its regime at that 21 point then. 22 MR. ADNI: That's correct. If we lose control of 23 the pressurizer level. l 24 MR. EBERSOLE: Which is what you will do. f) xs 25 MR. ADNI: That's correct, but you still maintain l l ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33H646
L9250 08 02 111 sf sDAV/bc 1 subcooling in that particular scenario. That's really v 2 what's the paramont issue. 3 MR. EBERSOLE: But you really do have a viable 4 use for motion of the level system in the vessel, which was 5 not there for all these years. 6 MR. ADNI: That's correct. We have two trains. 7 MR. REED: These discoveries of the frailties of 8 the auxilliary spray system were made, I _ guess, incidental 9 to a test for electrical power continuity, or something like 10 that, one of your loss of power trip tests.
' l f,. Did you have a test or tests at the outset that 12 were going to shake down 'the auxilliary spray system for its '
13 prooftesting?- (~} x_-~ 14 MR. WALDROP: We have. I'm not sure what the 15 status is or when those were completed. 16 MR. QUAN: The aux spray system will be used 17 during the natural circulation cooldown test to quantify its 18 capability. ; 19 MR. REED: Let's assume that this incident hadn' t 20 occurred and you hadn't incidentally discovered these 21 frailties. Would the tests that you were placring, would 22 they have been realistic enough to have disclosed these 23 problems?
~ 2'4 MR. QUAN: I'm cot sure what you' re getting at.
25 MR. REED: Well, the fact that there are valves
-{}
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwid'eCoverage 800-336 6646
9250 08 03 112 I I Dhv/bc 1 that didn' t have the proper electrical supply and the 2 charging pumps got hydroger down, and so or and so forth. 3 The test that you plarred, the ore fcr prooftesting 4 ~ auxilliary spray, would they have disclosed these problems 5 or did you discover -these things ' accidentally and 6 incidentally-by this. electrical test? 7 MR. QUAN: I'm' rot quite sure.if the problems 8 would have been found with_ testing. 9 MR. REED:- Well, you could evaluate that by 10 looking at your procedure, I guess. You must have a.writter 11 procedure for this test. And looking at it and deciding 12 . whether or not you have discovered these problems. w, 13 MR. HAYNES:. Well, Glern, I think the test as
}
plarred would not have revealed these problems, because 15L there was no intent to lose volume control tank level in 16 gasifying the pumps. That is not a plant test.
- 17 MR. REED: Well, that's interesting. You very
.18 . forthrightly made the statement, but it always puts me back 19 to the issue that when we are dealing with decay heat i
20 removal, there can be subtleties. The simulators.and the l-21 engineers haven' t figured it out all the way, and, 22 therefore, certair diversities, in my opinior, are very, 23 very helpful. 24 By that, I mear PORV diversity. 25: MR. HAYNES: I think I should go or to point out, i= A . ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nadonwide Coverage 800-336-6646 L_ s_ - _ __
9250 08 04 '113 l l' ever giver the condition we found ourselves ir, we were able
. ggDAV/bc 2 to recover from it and we were able to restart the charging j l
3 pumps. And would have been able to deliver. ' 4 MR. REED: You were able to recover in an hour
'S and a half without an incident or the primary system? I 6 think that's a lot different. You were doing something in 7 startup days when it's nice to do that. But wher you've got 8 the decay heat there and you' ve got another initiating 9 irciden t, it might not have worked out.
10 (Slid e. ) 11 MR. WALDROP: To get back to the specific evert 12 in the root cause evaluation that we did, we identified ('T 13 three major events or situations. One was the cooldown of
\_/
14- the -primary system, which, as I mentioccd, was caused by the 15 reopening of those main steamline drain valves. 16 The corrective measure to remove those, the 17 automatic opening of those valves or a turbice trip. It was 18 determired that they were not recessary to oper or a turbine 19 trip and we proceduralized it such that they will be 20 reopened prior to turbine startup. 21 We also lost power to our coressential station , 22 loads. And the cause of this was the inability of the 23 turbire generator system to maintain houseloads af ter we 24 initiated the trip. S 25 Our corrective measures for this are two things. NJ ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
1 1 l l 9250~08~05 114 I 1 One, some minor modifications to the turbire supervisory gggDAV/bc
'2 system for control of the stop valves and control valves to 3 the turbines, and to modify thef turbire EHC system to better 4 echance its control of those same valves.
5 MR. EBERSOLE: Had you had 5 percent decay: heat, 6- you would have had enough steam to maintain houseloads, 7 wouldr' t you? 8 MR. WALDROP: We did have sufficient steam to 9 maintain houseloads.. The problem was in the way the steam 10 closed the valves. 11 MR. HAYNES: The problem was with'the turbice
, 12 hydraulic control system. It did not respond quickly enough 13 to bring the valves back oper and maintain hoiseloads.
f')' 14 MR. EBERSOLE: You did have enough storage steam? 15 MR. MICHELSON: I gather this turbice is. stable 16 ther or houseload alore? 17 MR. HAYNES: Yes.
-18 MR. MICHELSON: Will you eventually run it that 19 way? You eventually tested it that way?
20 MR. ADNI: We'd run that way, but it's for a 21 short period of time. You can' t sit for hours on 22 houseload.
- 23. MR. WALDROP: You will experience some 24 overheating.
() 25 MR. MICHELSON: But it is stable for a shorter t ACE-FEDERAL REPORTERS, INC. ~ 202-347-3700 Nationwide Coverage 800-336 6646
.w 9250 08 06 115 '9,J O DAV/bc' 1 time, like 20 or 30 minutes? ,
2 MR. ADNI: Sure. 3 MR. MICHELSON: You've already operated it at 4 that? 5 MR. ADNI: That was part of the intial testing. 6 MR. EBERSOLE: But you're sure, if you have 7 stop-valve closure, you'11 kill the excitation? , 8 MR. WALDROP: To answer the question you brought 9 up earlier, what we have is, in the EHC system, there is a 10- volts per hertz trip. So if it looks at the voltage of the 11 frequency, it fixes the excitation accordingly. 12 MR. EBERSOLE: Or over-frequency? O 13 aa w^toaoe= ra e' riene-14 MR. EBERSOLE: On excess frequency. Watch it 15 row. 16 MR.-WALDROP: I'm not real familiar with that 17 system. 18 MR. ADNI: Is your question discussing stop valve 4 19 closure? 20 MR. EBERSOLE: I dor' t want to marry the station 21 auxilliaries with the exposure of the turbine. 22 MR. ADNI: What happers is, when the turbire 23 trips, excitation remains on the machine for a giver period 24 of time. h 25 MR. EBERSOLE: Then it will carry with it the ' ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-6646
~
9250 08 07- 116 qggpAV/bc 1 houseloads. If you happer to have an overspeed accident. 2 MR. ADNI: The unit is still tied to the grid.
; 3 MR. EBERSOLE: I'm discorrected from the grid for 4 this mode. I'm carrying houseloads and that's what I have 1
5 my turbine overspeed. Then I carry houseloads up to -- wher 6 did the turbine come apart? 7 MR. HAYNES: If I understand the question, the 8 question is: Does the excitation trip, giver a charge in 9, frequency -- 10 MR. EBERSOLE: Above normal. 11 MR. HAYNES: Right. And I think the answer was: 12 Below normal, it does. John, do you know the answer to the 13' question or the above normal? (]) _ 14 MR. BARROW: John Barrow, APS. There's only one 15 trip that will trip or frequency and that's a volts. per 16 hertz trip. You have to look at the setting or the volts 17 per hertz relay to determine that. The excitation and the 18 500 Kv trip in the switchyard are both set up to trip or 19 turbire sequential trip, but they have a reverse power relay 20 in the circuit. If you overspeed, you don' t see the reverse 21 power. 22 So that neither the excitation nor the 525 feet 23 per generator would trip, and you remain corrected to the 24 grid as long as you have excitation. We postulate that () _25 you'd have to lose the 525 KV correction. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-6646
9250 08 08 117 JggDAV/bc 1 MR. EBERSOLE: Are you satisfied that you carrot 2 carry in parallel or overspeed condition? 3 MR. BARROW: Yes, as long as you're corrected to 4 the grid. However, your reactor coolant pumps drip or o 5 overspeed. So the rest of your houseloads will be put 6 overspeed. 7 MR. EBERSOLE: I don' t care about the mair 8 coolant pumps anyway so much as other critical pumps like 9 service water, safety injection. 10 MR. BARROW: Yes, but safety injection would not 11 overspeed because we' re supplying of fsite power for safety 12 injection. 13 MR. EBERSOLE: Oh, I see, your normal load is (]') 14 offsite power supply anyway, so this would not be reflected 15 in the critical motors? 16 MR. BARROW: That's right. 17 MR. EBERSOLE: It would be in the concritical 18 motors that might give you coolart trouble. Thark you. I 19 had forgotter that. 20 MR. WALDROP: So the third major evert in this 21 scenario was the loss of charging flow. s we previously 22 discovered, the root cause of this was the root cause of 23 this was the inaccurate VCT level irdication. 24 MR. EBERSOLE: Let me go back to this. On the () 25 other hard, it would be urforturate to run the main ACE-FEDERAL REPORTERS, INC.
*02-347-3700
. Nationwide Coverage 800-33H646
- 9250 08 09 118 1 coolant pumps at-80 percert, and what would happer to them
'qggpAV/bc 2 with the flywheel, and so forth..
3 MR. BARROW: The main coolant pumps would cool on 4 overspeed within a few percent of normal rated speed. 5 MR. EBERSOLE: Are there any other big , 6 concritical pumps that could provide disruptive results as a 7 result of overspeed? 8 MR. BARROWS I can' t answer that. 9 MR. EBERSOLE: Like corderser circs, and so 10 forth? Oh, ventilation things. How about them? 11 MR. BARROW: Say that agair? 12 MR. EBERSOLE: You knew the problem. You don't (} 13 wart to marry a turbire overspeed condition to disruption by 14 carrying out the corrected houseloads with it. So that's 15 the gist of it. 16 MR. BARROW: All I can tell you is that we don' t 17 have overspeed trips or any other large motors in the 18 conclass 20 system. 19 MR. EBERSOLE: Why don' t you j ust have excess 20 frequercy trip? 21 MR. BARROW: Because we have variations in excess 22 frequercy. Normal variations in the grid to 5 percent. 23 MR. EBCRSOLE: No, I'm talking about 20 percent 24 or something, something you car safely say that's a cutoff. 25 MR. BARROW: I don' t know why we don' t have that. f]) ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 33G6646
- 9255 08'10-
~~
119 V/bc- 1 MR. EBERSOLE: Why don' t you kind of look at that' 2- matter a little bit? ' 5
/- 3 MR. WYLIE: Let me ask you a question. I got the 4 impression that this unit was designed to separate the
-5 system or grid rupture and continue to carry the house 6- auxilliary. Is that correct? .
.7 MR. BARROW: That's the function of the reactor 8 power cutback system.
9 -MR. WYLIE: The whole plant is designed to 10 separate from the grid and carry its own auxilliaries. Is 11 that correct? 112 MR. WALDROP: That's correct, or a generator 13 .. dif ferential, let's say, trip. 14 MR. WYLIE: I'm not talking about a differential 15' trip. I'm talking about grid trouble. Is that cchrect? 16 MR. ADNI: You can run in a normal configuration, 17 keep the safety loads on the switchyard,' the coressential 18 loads on the auxilliary transformer. 19 MR. WYLIE: That's not my question. The question is, is the plant designed in the case that you have -real 21 trouble, you're going to have a grid blackout? It's goteg 22 to be a degraded grid? To separate the grid and continue to 23 operate this plant. 24 MR. BARROW: Well, that's not a design basis. 25 MR. WYLIE: Okay, thank you. That answers my ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 33H646
19250'08-11 120 V/bc 1 question. 2 MR. WARD: And the safety loads all come from the 3 grid? 4 MR. WYLIE: I understard but my point is, if you 5 did design to separate from the grid to carry houseload, 6 you'd also have to take your other auxilliaries with you. 7 Then you'd be back. Thank you. 8 MR. WALDROP: As I said, the loss of charging 9 flow, the root cause of which was the inaccurate VCT level 10 indication, as previously discussed, our corrective measures 11 were, ore, to institute a daily surveillance of the 12 reference leg to ensure that it was adequately filled and 13 resporded properly. We revised our procedures for the f 14 lineup of the charging pump suction such that if we got a 15 loss of power situation, we would promptly lire up suctior 16 to the RWT to ersure that we didn' t experience the problem 17 that we did with this evert. 18 We also modified our procedures to ware the 19 operatcrs of the potential for this type of a failure to the 20 charging pwmps. And we also did various other 21 administrative charges and echancements in some of our 22 systems, including the post-trip review. 23 Finally, we began to pursue design echancements 24 .to the charging system. 25 With that, I'll ture it over to Mark Radspierer
-(}
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
. 9250 08 12 121 1 to discuss those desige enhancements.
}gpAV/bc 2 MR. RADSPINNER: My came is Mark Radspierer. I'm 3 a mechanical engineer in the Nuclear Engineering 4 Departmert.
5 (Slid e. ) 6 As Terry Quar indicated in his introduction, I'll
~
7 be describing some modifications we've been making to the 8 Palo Verde plant design in response to the temporary loss of 9 ta rge t , low suction event. 10 MR. EBERSOLE: You call them design 11 echarcements. I regard that as a fancy phrase for 12 patchwork. 13 MR. RADSPINNER: The first of the four
)
14 modifications we will be making addresses the root cause of 15 the evert, which, as Gary irdicated, was the inaccurate 16 volume control tank level indication caused by a partially 17 drained common wet reference leg to the two existing 18 trarsmitters that we do have on that tank. 19 To address the problem, we will be providing a 20 separate reference leg, as was referred to earlier. We'll 21 make ore of those reference legs wet and one will be dry, 22 p avidicg some diversity there. Ard we will provide ar 23 additional control room alarm that actuates when the two 24 transmitters indicate a dif ferent level in the tack. 25 This modification that eliminates the possibility (}) i ACE-FEDERAL REPORTERS, INC. I 202 347-3700 Nationwide Coverage 800 336 6646
-9250 08 13 122
- PmDAV/bc- 1 of a similar event from occurring because the partially
- \_[
2 drained reference leg would cause that transmitter to be 3 different.than the dry-based transmitter, actuating the
'4' control' room alaos. And effective action car ther be 5 taker.
6 That modification ther replaces the need for one
.7 of the corrective measures that Gary mentioned, that being 8 the one-involving the daily inspection of the reference 9 leg.
1C MR. EBERSOLE: If one of the reference legs gets 11 into disarray, ther you have the check system that says 12 they're no longer in balance. Right? 13 MR. RADSPINNER: That's right. L(]} 14 MR. EBERSOLE: Is that in any way dependert or 15 the condition of the individual reference leg so there could 16 be a coincidence f ailure? 17 MR. RADSPINNER: No. If the wet reference leg j 18 loses volume, it indicates high. If the wet reference leg 19 loses volume, it indicates low. So it will go in opposite 20 directions on any type of common interaction. i 4 21 MR. EBERSOLE: Okay. 22 MR. RADSPINNER: The secord modification has also 23 been alluded to, provides power to the volume control outlet 24 valve and the RWT gravity feedlire valve directly from a lE 25 motor control center. Due to that provision that's been {" } ACE. FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 6646
.9250 08 14 123 1 described, that trips all convital loads from the lE bus. fAV/bc 2 The subsequent operator action would be required to manually 3 restore power to those valves, outside control. 4 With our modified design, we will re route power 5 to those valvos directly from the lE MCC, which is of course 6 corrected to the lE bus that MCC would to longer would be 7 shed or the safety evacuation. 8 MR. EBERSOLE: Do these gravity feedlire sources 9 provide adequate suction if they' re oper and multiple? 10 Or... 11 MR. RADSPINNER: If several gravity feedlines are l 12 corrected to pump suction, it wouldn' t result. r~ 13 MR. EBERSOLE: Okay. It wouldr' t be that the Q) 14 hydroger system from the volume cockrol tank would override 15 in suction mode ever though the others were oper? 16 MR. RADSPINNER: If the volume control tack 17 suction was aligned? 18 MR. EBERSOLE: Wouldn' t it override the other 19 suction systems? 20 MR. RADSPINNER: If we did not isolate the volume 21 control tank. 22 MR. EBERSOLE: That's a single mode isolation. 23 MR. RADSPINNER: That's correct. 24 MR. EBERSOLE: So we already have a single mode 25 failure, which is failure of that to close because it O~s ACE FEDERAL REPORTERS, INC. 3)2-347-3700 Nationwide Coverage 800-336-6646
. - . . . - . . -~ ._ - . . . - _ - - . - - -. ..- . . ,
?!
~
j 3 9250f08 15. 124- ! 1 overrides the suction from the other valves. I presume that gph'J/bc - gf 2 thl6 pressure -- l '3 MR. RADSPINNER: That's right. If we are in a ' t 4 loss of offsite power- corditioe where we doe' t have a normal t 5 alternate feedpath oc the- RET, which is the boric acid
- 6_ makeup pumps, then that cordition happers.
7 MR. EBERSOLE: Mow can I rationally ther only 18 have one valve there consistent with your having to have 9 random failure? 10' MR. RADSPINNER: Because in that unlikely evert, 11 we still have adequate time to take the macual actions 12 -required tu restora the system. j( ) 13 MR. EBERSOLE: You'll force it shut? 14 MR. RADSPINNER: I ear either force it shut or 15 they car ture off the target pumps and ther close these 16 three valves thshe. 17 (Slide . ) 18 And oper the gravity feedlice from the RNT that 19 Terry explained. 20 MR. EBERSOLE: There you invoke the fact that I'm 21- rot there anyway. 22 MR. MICHELSON: When you say you have adequate 23 time, have you doce some kird of a time and motion study 12 4 that's documented that shows how long it takes to do these { .25 various operatiors? ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverass 2 336 6646
'9250'08 16- 125 V/bc 1 MR. RADSPINNER: We haver't doce a time and
'2 motion study. We've experienced ar actual evert where they 3 did allge that path up well within the two-hour timeframe.
4 MR. MICHELSON: I was under the understarding
.5~ that 501 had closed under your evert conditions. Now, in
'6 the unlikely evert you car' t get 501 closed, you ther 'said
; 7- .you would go back to the_ individual charging pumps, et 8 cetera.
9 Have you doce a study that backs up that 10 statemen t? 11 MR. ADNI: We haver' t performed a study but the 12 location of the valves, these particular valves, all the
'j].
13 suction lines are in remote operated valves. They're easily 14 accessible. The same for our CH valves. 15 3 16 . 17 18 19 20 21 22 23 ,_ 24 s I 25 ACE. FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 6646
,0250.09 01 126 MR. MICHELSON: I'm sure there are, but there are j DAVbw 1 2 a limited number of operators trying to do things that would 3 be nice to see such a study.
4 MR. EBERSOLE: And they have suction trips, these 5 pumps, don't they? In case the operator gets excited and 6 closes all the available suction before he opens the other. 7 MR. RADSPINNER: Yes. I should mention that we 0 may not a documented time and motion study, but we've asked
-9 the operators to simulate that action, actually run down 10 there and do those actions and report back to the control 11 room with a further accurate time.
12 MR. MICHELSON: How much time did it take to go () 13 down and find out that 501 didn't close, and therefore, you 14 had to go to the individual charging pumps. 15 MR. RADSPINNER: There was an indicating. light on 16 the control room panel, which, if 501 did not close -- 17 MR. MICHELSON: But how long a time did it take? 18 You said you went through this exercise anyway, I thought. I 19 How long did it take to run the exercise wherein you went to 20 501, couldn't get it closed and had to run over to the 21 individual charging pumps, wherever they were and get those 22 closed individua11. How long did this take? 23 MR. RADSPINNER: T don't have the exact number, 24 but the whole scenario took an hour and a half. {} 25 MR. MICHELSON: I know that, but it bothered me ACE FELGRAL REPORTERS, INC. ll02 347-3700 Nationwide Coverage 800 336 4 646 j
9250 09 02 127 I a little bit. I am sure this didn't take an hour and a jggDAVbw 2 half. 3 MR. ADNI: We don't have the time available for 4 that. 5 MR. EBERSOLE: I'm going to postulate that the 6 reason you're going to be in a hurry is that you've had a 7 severe steam rupture. The steam is spewing into the 8 atmosphere, and it's going to go to the primary, and you 9 can't get the pressure down. 10 MR. RADSPINNER: Gerry Sowers will present some 11 analysis after I'm finished, which establishes the two-hour 12 time frame for that event also. 13 MR. REED: I assume that you're a designer; is 14 that correct? 15 MR. RADSPINNER: That's correct. I'm in the 16 Engineering Design. 17 MR. REED: I've been a designer and been an 18 operator, and I've always leaned to the operators and the 19 burdens that they sometimes get heaped upon them. And I'm 20 looking at your modification summary, and I find your 21 modifications are pretty lean. What yoit're talking about is 22 mostly burdening the operators. Yott give them a whole bunch 23 of options to look at. They've got to figure out a whole 24 bunch of things and then run around all over the place. You
' ), 25 know, motorizing valves where you have these single ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 33M/A6
9250 09 03 128 1 component failure issues. I know that's the tendency with a gggDAVbw 2 built plant. You don't want to change it, and the designers 3 have a certain prestige they have to maintain, I guess. 4 They don't like to get operator criticism. 5 Is everything in harmony between your operators 6 and the designers here, that they're just loving the idea of 7 all these options? I've looked at one of your procedures 8 having to do with the use of auxiliary spray in accidents 9 and incidents, and so on, and I found it very complex for 10 operators to sort out the options and deal with this. 11 MR. EBERSOLE: Are you asking if there'are warm 12 relations between design and operators? MR. REED: .( ) 13 Yes. I want to know if the operators 14 ha.ve the right of refusal or override. 15 MR. HAYNES: I'd like to ask Bob Adni to speak to 16 that. He's the Unit 2 superintendent and author, or at 17 least lead author of most of our operationg procedures. 18 MR. ADNI: Yes. We've been in quite some heated 19 discussions on particular points of plant design. If we 20 firmly believe that the repairs, the fixes, the 21 enhancements, per se, are not well within what we would deem 22 a reasonable amount of time for the action for the operator 23 to perform, I believe we've made our point. These actions 24 -- like I said, we have walked them down in the plant. We ([ ) 25 have reviewed them. We're involved in the initial ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33H646
9250 09 04 129 1 suggestions on the team that put the recommendations gggDAVbw 2 together. We believe it was certainly a viable option. 3 MR. REED: It's a good thing I'm not an operator 4 at your plant. 5 MR. WYLIE: Let me ask a question. I asked the 6 question earlier on your second bullet up on providing power 7 to the two valves. 8 Is that two separate circuits, two starters from 9 the motor control center, or what is that? 10 MR. RADSPINNER: Your question is on the modified 11 design? 12 MR. WYLIE: Yes. You've got one circuit
~
4 13 supplying two valves, is what you've got. 14 MR. RADSPINNER: There is a single circuit 15 supplying power f' rom the bus to the MCC and there is a 16 single circuit from the MCC. Well, from the MCC, there's
'17 actually two separate circuits to the valve.
18 MR. WYLIE: So I go back to what I said a while 19 ago. If you had to open up 536, and it doesn't open, it's 20 frozen, and it trips out the whole circuit, and you don't 21 close 501. 22 MR. HAYNES: No. Mark, would you step back. If 23 you see, there are breakers at each of the valves, and the 24 trip functions are synchonized so that the breakers let the
; 25 valve go on a fault on the valve before the two in-line ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 900 336 6646
9250'09 05 130 DAVbw. 1 breakers trip. 2 MR. WYLIE: How do you do that? 3 MR. HAYNES: By synchronizing the point at which t
'4 the trip.--
5 MR. WYLIE: Aren't all of those motor case 6 breakers?- As far as I know, there's no selectivity on motor {
-7 case breakers.
8 'MR. RAYNES: Let me ask our electrical designer 9 to address that.
'10 MR BARROW: John Barrow frma APS. I can't 11 address right here as to whether or not these are motor case 12 breakers, but if they were motor case breakers, motor case 13 breakers may be selected to trip at certain ranges.
(} " 14 MR. WYLIE: On short circuit?. 15 MR. BARROWS Yes. 16 MR. WYLIE: I've never know that. 17 MR. BARROW The breaker is upstream from the 18 motor case breakers. The motor case breakers are 19 coordinated to the upper range, which means they should pick 20 up the fault. The only reason they would not pick up the 21 fault is if one of them failed. 22 MR. WYLIE: 'I'd agree with you on overcurrent, 23 but as far as overcurrent for a locked rotor or short 24 circuit, I didn't know you could coordinate motor case 25 breakers. ({} ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33H646
~9250:09 06- 131 1 MR. BARROW I'm coordinating those with the jgpAVbw 2 upstream breakers.
I 3 MR. WYLIE: You mean those are not motor case 4- breakers there?
.5 MR. BARROW: I do not believe.they are. I'll 6 have to check that.
l 7 MR. WYLIE: All the motor control centers I ever l ! 8 saw had motor case breakers on them. That's the name of 9 game. ld MR. RAYNES: Let us respond to that. l- 11 I believe that what I said is correct, but you 12 may be right.
' L'
/'T 13 MR. WYLIE: Why don't you run two circuits like 14 you had before?
15 MR. RADSPINNER: Let me explain what we actually l l 16 did. Maybe it'll answer the question. At present, there is i 17 a 1E MCC located in there in proximity to this one. We 18 brought out the cable then over to here and nipped the u 19 diesel off the IE MCC and tied them right there. 20 MR. EBERSOLE: And why do you have all those l 21 breakers there? I 22 MR. RADSPINNER: This schematic was to i 23 demonstrate that the individual breakers are still in l-l- 24 exiscence. 25 MR. EBERSOLE: Just because you left them there, i (]} l l ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336 4646
9250 09 07 132 1 rather than that they were needed. gDAVbw
.2 MR. RADSPINNER: They are there.
3 MR. EBERSOLE: What's the purpose of all the 4 breakers in series over there on the right side? 5 MR. RADSPINNER: These are the IEEE 279 required 6 separation between the safety-related MCC and the 7 nonsafety-related valves. 8 MR. MICHELSON' I'm beginning to get a little 9 concerned on your electrical. Which parts of this are 10 seismically qualified? 11 MR. RADSPINNER: Everything. 12 MR. MICHELSON: Up to that point, it certainly. ('y s_ 13 It was IE; it still is. Your non-IE was not and still'is 14 not. You're saying, really, that this is a nonsa'ety 15 function you're now dealing with, as far as these breakers 16 or as far as these valves are concerned. This is a 17 nonsafety arrangement. 18 Therefore, in the seismic disturbance, all these 19 things might not perform, if you had a need for them, which 20 you might have. 21 MR. QUAN: If there were a seismic event and 22 these valves did fail, these valves do have hand reels, so 23 that we can rely on manual operation to operate these 24 valves. The valves are seismically qualified. 25 MR. EBERSOLE: This is typical, then of any ( }; ACE FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 33H646
9250 09 08 133 jlhDAVbw l' nonrunning connection to a lE system; right? Double 2 breaker disconnects. And they have, what, short-circuits? r 3 MR. RADSPINNER: This has been used before. 4 MR. EBERSOLE: So they don't challenge the lE 5 motor. You have a guaranteed disconnect. 6 MR. RADSPINNER: Right. 7 MR. MICHELSON: You're also assured that in a 8 seismic unit, you may have to revert to manual operation. 9 You haven't really qualified the system. l 10 MR. QUAN: That's right. We may have to revert 11 to manual operation. 12 MR. MICHELSON: The plant state is a little () 13 different after an earthquake and the state of the operators 14 is a little different after the earthquake. All this time 15 in motion, it's certainly going to take longer rather than l 16 shorter periods. l 17 So I hope there's lot of margin, and they can 18 take care of the unlikelihood event of an oarthquake. 19 MR. EBERSOLE: That left-handed existing system, 20 you didn't comply with IEEE requirements. 21 MR. RADSPINNER: This is an alternate acceptable 22 means of applying that separation. l 23 MR. EBERSOLE: Why should that be deprived of the 24 second breaker? Does IEEE say that you can go into the () 25 remote control center, but you can't go into an individual l ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-66M [
9250'09 09 134 h Vbw 1 labor of double disconnects? That seems to be an 2 incongruous requirement. 3 MR. RADSPINNER: John, do you know? k 4 MR. BARROW: John Barrow. We switched from the 5 triple SIAS to the double breaker trip, due to discussions 1 6 we had with the NRC, on which was perceived to be the most 7 appropriate solution to this problem. The two breakers 8 below the lE MCC on the right-hand side are both IE 9 environmentally and seismically qualified breakars. 10 MR. EBERSOLE: Right, but on the left hand side, 11 I see one breaker, it appears between a 1E bus and a non-lE 12 MCC, which has presumably all sort of loads on it. Yet it 13 has only a single disconnect. Was that exempted from 14 compliance with IEEE7 , 15 MR. BARROWS Yes, it is, because it is a 16 qualified breaker. It trips on SIAS. 17 MR. EBERSOLE: So IEEE permits a single 18 disconnect between a lE bus and a non-lE bus of i 19 indiscriminate character; is that correct? 20 MR. BARROWS Exactly, with a triple SIAS. 21 MR. EBERSOLE: It still trips the same mechanism, 22 for goodness sake, whatever the control circuit is. So they 23 permit, then, the hypotheticci failure of that breaker to 24 trip, and then the cascading failure of the IE bus. O 25 MR. ARROWS Thae ereeker is seismica11r and ACE FEDERAL REPORTERS, INC. 3)2 347 3700 Nationwide Coverase 800 33M646
9250 09 10 135 jllDAVbw 1 environmentally qualified. 2 MR. EBERSOLS: I understand. It's still one 3 breaker and it doesn't suffice to escape the random failure 4 hypothesis and that is an accepted IEEE solution to that? 5 MR. BARROW: We understand it is. 6 MR. EBERSOLE: And that can be done in multiples, 7 so I can have a dozen of them. It seems like IEEE is -- 8 well, isn't that interesting. 9 MR. WYLIE: That's true. What he says is true. 10 It is permitted. You can connect nonsafety loads to a Class 11 IE bus, as long as you use a class -- safety grade Class lE 12 device to do this operation, and it's qualified. 13 MR. EBERSOLE: I can get 50 of those, Charlie. 14 MR. WYLIE: You do have 50, but they're supplying 15 the same -- 16 MR. EBERSOLE: No, no. Not safety loads. 17 MR. WYLIE: There's no difference in supply 18 safety loads. 19 MR. RADSPINNER: That then becomes a single 20 failure. 21 MR. EBERSOLE: True, but the safety loads are at 22 least hypcthetically protected from multiple failures. 23 Okay, thank you. 24 MR. REED: Jesce, you have repeated something a 25 couple of times, and I want to make sure that it's still 1 l l ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 3364t6
9250'09'11 136. h Vbw 'l. 'on the agenda. You've repeated the fact that these valves . 2 don't have to act quickly. They don't have to open 3 reliably, and you've said that the licensee has said there's 4 lots of time to depressurize. 5 I'd like to say that I'm going to challenge that 6 phrase, " lots of time to depressurize," when we get.around 7 to steam generator tube rupture. 8- MR. EBERSOLE: Yes. I think that's the crux of 9 it, Glenn. How much time do-you have with a major steam 10 generator tube failure? I think you've be glad you're in 11 the desert. 12 (Laughter.) O 13 (Stide.) 14 MR. RADSPINNER: The third modification involves 15 the addition of a diverse reference link. As Terry b 16 described, the IWT is automatically aligned to;the charging 17 pump suction on the lower volume control tank level via the 18 boric acid makeup tank flow path pump. If this normal path 19 is unavailable due to a loss of offsite power condition, l 20 then an operator action would be required to manually open L 21 the gravity feed line valve from the control room, to align l 22 his gravity feed line suction path to eliminate reliance on 23 that operator action. And we are adding a control feature, 24 which when the loss of of f-site power condition exists, the O 25 1ow votome contro1 siona1 wi11 ehen aueomatica11v egen the ACE. FEDERAL REPORTERS, INC. L- 202 347 3700 Nationwide Coverage 800 33H646
, ., n q r
- s. <
l 9250'09 12 137 1 gravity feed line valve, and thereby automatically align ggpAVbw 2
- that gravity feed line path.
,{' ,. '[ MR. .WYLIE: ihkt automatically precludes the
'4
- valve to the volume control tank?
s 5 MR. RADSPINNER: Yes, that interlock is still 6 .present. 7 MR. MICHELSON: Is this going to be lE or non-lE? 8 MR. RADSPINNER: Is what going to be? 9 MR. MICHELSON:' The control circuits. This is [ . 10 a non-lE transfer system.
- w .
% 11 MR. RADSPINNER: It has certain features. The 12 transmitters control circuits are non-lE. They have a
/~l
\J 13, non-1E backup power supply.
14 MR. MICHELSON: How do you prevent this non-lE 15 system from giving you incorrect actuations like a transfer s
. 16 back to the storage tank at the time you're running off of
.17 the refueling tank, and the storage tank being already 18 empty, it sucks high energy into it in a hurry.
19 MR. RADSPINNER: The control circuits just don't 20 work backwards. 21 MR. MICHELSON: I ddn'teknow. I haven't seen 22 your control e,ircuits yet. You h'aven't designed it yet, but h 23 you*tell me it's 9 lE, so it can'most anything, unless there 24 are special criteria applied. M' s 0j +. 25 MR. RADSPINNER: The criteria we've applied are i 202 347 3700 ACE FEDERAL REPORTERS, INC. Nationwide Coverage 800 33MM6
o
.9250t 09 13 138 vbw _ 1 two.
2 MR. MICHELSOh; It's unidirectional and it's 3 . designed.so that it cannot even under seismic conditions 4 ~ flip back'to the previous-position? Is that_what you're 5 saying? It's lE in one. direction, but'not in the other? I
'6- guess ,it's a little premature, but clearly, when you get
_7 done designing this system, if you're not going to qualify 8- it, then one has to ask about its failure modes and-9 effects. And you'll work through a failure mode and a fact 10 analysis and provide that as part of your licensing 11 documentation, I assume, because clearly you worry about 12 failure modes in this system. 13 Getting you back on a tank that you've already (~]N u 14 emptied and transferred from, and now we don't want to 15 ' suddenly start throwing hyd'rogen before you have a chance to 16 get it under control, and you have no time now to.run down 17 and fiddle with the valves. So.it's just, I think, 18 something certainly I would be interested in seeing in your 19 analysis. 20 MR. EBERSOLE: Let me ask, all these generations 21 we're talking about,-rewiring, and what do we call it, 22 . enhancements, that great word we use for patching, are they 23 reflected in considerations of Appendix I and the alternate 24 shutdown time, where I think there is a requirement that you {J 25 go to cold shutdown? ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverase 23Wi646
9250 09 14 139 1 MR. RADSPINNER: That's right. The two {J~lDAVbw 2 particular valves still have-the remote disconnect switch, 3 so they can be operated at the MCCs. J 4 MR. EBERSOLE:, So they're reflected in the 5 operation of that auxiliary center. 6 MR. RADSPINNER: That's correct. 7 MR. MICHELSON: I think there's even a more 8 encompassing question, Jesse. You will do the Appendix R 9 type analysis or really the final analysis on this entire 10 system to determine the possibility of local fires, getting 11 into difficulty, particularly post-earthquake, when you Il12 might need this system, and then you may have some local
/J 13 fires as a consequence of the earthquake.
14 MR. EBERSOLE: I don't know. You said that the 15 depressurization function is a safety function? 16 MR. RADSPINNER: By the issuance of R35-1, that 17 made it a safety-related function.
.18 MR. EBERSOLE: So I'm entitled now to see in all 19- the paperwork of testing and test specs and so forth; is 20 that right?
21 MR. RAD 3 PINNER: I believe that's correct. 22 MR. MICHELSON: They're not providing 23 safety-related circuitry to perform the safety-related 24- function. For instance, this transfer aspect is not
]
i 25 safety-designed. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-6646
9250 09 15 140 1 MR. EBERSOLE: I don't know where that breaks qggpAVbw 2 off. 3 MR. MICHELSON: I don't either. That's why I'm ) 4 pursuing this question. You just told me it's 5 safety-related, yet you also have told me recently that 6 these aspects are not in Class I seismic, et cetera, so 7 what's your rationale, although the function of auxiliary 8 spray is safety-related, you're not going to design a:1 of 9 the circuitry to be safety-related. 10 MR. EBERSOLE: Carl, I think you could design a 11 safety-related system out of a bunch of old lumber, as long 12 as you could patch it fast enough. ( ') 13 MR. MICHELSON: That could always be an answer. 14 That's one of the answers we've been pursuing, is manual 15 operation. But for instance, on the automatic transfer, I 16 don't manual operation will get you out of trouble fast 17 enough. 18 MR. EBERSOLE: That's to be investigated this 19 afternoon, isn't it? 20 MR. MICHELSON: Right. 21 22 23 24 I ') 25 v ACE. FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
I 9250 10 01 141 1 MR. RADSPINNER: The final modification is gggDAV/bc 2- administrative control enhancement in which we will be 3 locking open two valves in the primary auxilliary spray 4 path. In the RWT isolation valve, which is on the high 5 point suction off the RWT, also the charging line 6 containment isolation valve, both of those valves are 7 normally open. 8 The RWT valve is a fail open air operated valve. 9 The containment operated valve is a fail as is 10 motor-operated valve. As Terry mentioned, neither valve 11 receives an automatic actuation signal and, thus, can only 12 be closed by the operator from the control room. . ~.s ("') 13 However, to prevent any possibility of 14 inadvertent isolation of that flow path either by spurious 15 signals or by operator error, we will lock open and remove 16 power to'the two valves. 17 MR. MICHELSON: Which valves by number are these? 18 MR. RADSPINNER: Valve 532, which is that valve 19 right there. This is an error-operated fail open RWT 20 isolation valve. 21 MR. MICHELSON: Where is the other one? s 22 MR. RADSPINNER: The other one is the charging 23 line isolation valve, 524. 24 MR. EBERSOLE: We'll talk this afternoon about (} 25 Glenn's problem with the tube failures in the heat exchanger a ACE-FEDERAL REPORTERS, INC. I 202-347-3700 Nationwide Coverage 800 236-6646 2
L9250!10'02 142
~
and the steam' generator-tube failures. gAV/bc 1 2 MR. RADSPINNER: -This afternoon, when we get to 3 that -- .
~4- MR. MICHELSON: Is 2524 considered to be a.
5 containment isolation valve? 6 MR. RADSPINNER: It is a containment isolation 7- valve. 8 MR. MICHELSON: You lock it open and remove the 9 ' power? 10 MR. RADSPINNER: Correct. 11 MR. MICHELSON: So what is now your containment , 12 isolation valve? You've eliminated a containment isolation 13 valve. ;Is there some logic that says that's okay?
-{
14 MR. HAYNES: It's a containment isolation valve 15' but it does not receive the containment isolation valve 16 circuit. 17 MR. MICHELSON: I realize that but it's now no- i
~
-18 longer a containment isolation valve, is it? It's locked, 19 open and depowered. So it's only available under the 20 unusual circumstance of going to the breaker and closing the 21 power on it. I don't know what " locked open" means.
22 Locking the valve itself, or what you're doing. But you
- 23 have to go through several steps in order to make it a 24 contaimment isolation valve.
25 Do you feel...really, you're saying that there's
'({}
p ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
9250-10 03 143 7"'DAV/bc 1- really very little if any need for that valve. (/ 2 MR. HAYNES: That's correct. And the staff 3 agrees with that. And, in fact, proposed this. 4 MR. MICHELSON: And that's a four-inch line? 5 MR. REED: I think that would be classified as an 6 ingoing line to containment. And I believe the criteria
? calls for only two checks. I guess.I see two checks,
'8 although I don't see them where they're considered 9 traditionally, one inside and one outside.
10 MR. QUAN: Again, this is a simplified schematic, 11 not a detailed PNID. But there are two check valves. 12 MR. RADSPINNER: With that, I'd like to turn the (} 13 mike over to Jerry Sowers, who will discuss steam generator 14 tube rupture. 15 (Slide.) 16_ MR. SOWERS: My nama is Jerry Sowers, the Safety 17 Analysis Supervisor with APS. I would like to discuss our 18 steam generator tube rupture analysis. 19 I'd really like to do two things this morning. 20 First, I would like to explain the use of the auxilliary 21 pressurizer spray in the current anlysis; second, 22 demonstrate that the dose consequences for the Palo Verde i 23 steam generator tube rupture analysis are relatively 24 insensitive to the top frame that auxilliary pressurizer l 25 spray is initiated. (~3 s/ i I l i ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646 {
9250'10 04 144 l I
?11NV/bc -1 And third, I-would like to show that an !
, \_) 2- alternative' system exists to facilitate depressurization
- 3. following a steam generator tube rupture.
4 (Slide.) 1 5 To do this, I'd like to talk about three 6 analyses. The first analysis is currently in the Palo Verde
- 7. FSAR, Appendix 15A. It's a steam generator tube rupture l
8 followed by a loss of offsite power.
- 9. We were asked to analyze this accident by taking 10 into account some of the actions that the operator would 11 possibly take during that event. One of the actions that he
- 12. would possibly take is to open atmospheric dump valves for ,
\
- 13 f]) decay heat removal.
14 The asumption:is made in this analysis that the' i 15 dump valve, the atmospheric dump valve on the affected steam , 16 generator opens fully open, wide open, and sticks open for ! 17 the duratien of the event.
- 18. In-this analysis also in recognition of what the 19 emergency operating guidelines provide, the auxilliary
-20 pressurizer spray was assumed to be used at about 17 minutes 21 into the accident.
22 In order to illustrate the effect of delaying 23 that aux spray actuation, we did a second analysis. It's 24 ossentially identical to the first one except that the j ]) 25 auxilliary pressurizer spray operation is delayed until two ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800 33(Hl646
9250 10 05 145 f~^DAV/bc 1 hours after the tube rupture. -%) 2 MR.'EBERSOLE: My impression is, however, that if 3 you don't get auxilliary spray, it may be like 10 hours, or 4 a long time before you get the pressure down. 5 MR. SOWERS: We chose two hours because our 6 recent experience had indicated that two hours was a 7 sufficient time to take the manual actions that would be 8 required in case of failures to restore the spray. 9 MR. EBERSOLE: At this time, what is your option 10 to restore spray if because of the emergency situation you 11 have gone water solid? 12 MR. SOWERS: I'm not sure how to answer that
~
13 because in this analysis, you never go water solid. (} 14 MR. EBERSOLE: Okay. 15 MR. SOWERS: The third analysis that we did to 16 illustrate the alternative to the auxilliary pressurizer 17 spray, essentially, the same ground rules, except, instead 18 of actuation of the auxilliary pressurizer spray two hours 19 after the initiation, the operator was assumed to open the 20 pressurizer vent at two hours. 21 MR. EBERSOLE: What is this vent? 22 MR. SOWERS: I'll get into a description of the 23 gas vent system before I go through that analysis. 24 MR. EBERSOLE: Is it sort of a miniature valve? (} 25 Well, you'll get into it. 1 ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800-336-6646
9250 10 06 146 JggDAV/bc 1 (Slide.) 2 MR. SOWERS: Looking at the first two analyses 3 here, what I have here is a graph of the RCS pressure as a 4 function of time. The solid line is the FSAR analysis that 5 assumed the initiation of the auxilliary spray, about 17 6 minutes into the event. 7 The dotted line is the analysis that delayed the 8 initiation of the spray until two hours into the event right 9 here. If we can just go through this, I can illustrate the 10 major events that take place on this graph. Of course, 11 first, you get the rupture, the trip. You get an immediate 12 depressurization just due to shrinkage. The operator opens 13 his ADV. On each steam generator in this analysis, the one [v) 14 on the affected steam generator fails wide open. 15 As soon as it fails wide open, you've got a lot 16 of decay heat removal capability. 17 MR. EBERSOLE: But I don't have pressure removal. 18 MR. SOWERS: Well, this is a pressure plot. And 19 we get a significant depressurization. 20 MR. EBERSOLE: Right, but that's if you get the 21 spray system going. 22 MR. SOWERS: If you get the spray. You see, you 23 get this much right here down to below 1,500 psi. It's 24 strictly due to the ADV. It opens the ADV here and it gets 25 down that far. ( }) ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
9250 10 07 147 ' .M'4DAV/bc 1. At this point, in this analysis, the spray is
.V 2 actuated. So through here, the depressurization is from two 3 effects -- the spray and the decay heat removal. Just the 4 heat removal from the system here, you're cooling down 5 rather rapidly through tnis part of the analysis.
6 MR. EBERSOLE: That's the other curve I 7 ' mentioned. That's the one when you don't get spray. 8 MR. SOWERS: We'll get to that. Okay? 9 Lcstly, the operator starts throttling the HPSI's 10 on the system when the conditions have been met to allow 11 them to do that. We look at the second curve here. You can 12 see what happens if you don't use the spray. (~ 13 MR. EBERSOLE: You get down to what? Fourteen (>Y 14 hundred now? 15 At that point, HPSI doesn't put any water in. 16 MR. SOWERS: It's about 1,570. 17- MR. EBERSOLE: HPSI's are now putting water in? 18 MR. SOWERS: ~Yes, they are putting water in. 19 What's happening, during this time right here, is that you 20 'can see you've reached essentially a steady state pressure 21 in the RCS. 22 MR. EBERSOLE: They pushed the water back into 23 the pressurizer? 24 MR. SOWERS: You do get a little bit of water (} 25 back into the pressurizer at this time. That steady state ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(Hi646
9250 10 08 148 qggpAV/bc 1 is set up by the pressure where the HPSI flow rate and the 2 leak flow rate are equal. 3 You can see if the leak flow rate is greater at 4 any time. You'll depressurize some more. If you 5 depressurize, that helps the HPSI's pump more water in. You 6 reach an equilibrium. If you're at the other side, where 7 your HPSI flow rate is too high, higher than the leak rate, 8 you'll start pressurizing the system, which is what's 9 happening right here. And you'll pressurize the system. 10 The leak rate will slightly increase and the HPSI 11 rate will be cut down and you will reach an equilibrium 12 eventually. That's what happens if you never touch the (') v 13 sprays. 14 As an incidental, the pressurizer level through 15 there is about 20 percent. Okay, it's not up, it's not 16 down, it just sits there. 17 MR. EBERSOLE: It's still visible on the 18 indicator, is it? 19 MR. SOWERS: I believe it is, 20 percent, yes. 20 It has gone down and come back up. 21 MR. EBERSOLE: Is this with one tube failure full 22 open? 23 MR. SOWERS: This is a guillotine double-ended 24 tube rupture. () 25 MR. WARD: One tube? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
9250 10 09 149 l l DAV/bc 1 MR. SOWERS: One tube. 2 MR. REED: I've got some concerns here. Now this 3 is some plots on the various conditions of the pressure and 4 what happens to pressure based on spray use and HPSI use. 5 I notice those pressures fall perhaps below that 6 which would support natural circulation heat removal from 7 the steam generators, unless you're asuming that the leak 8 takes care of the heat removal, that the ruptured tube leak 9 will take care of the heat removal; because you admit they 10 have overpressure on the primary in order to remove heat by 11 natural circulation. 12 I'm also concerned that going on at the same time () 13 as these curves are drawn is the pucential for steam 14 generator overfill, because you either are or are not going 15 to maintain a substantial differential to drive fluid from 16 the primary to the secondary through the ruptured tube. 17 And I have to think about overfill and I have to 18 think about the cost of radioactivity. But I guess you're 19 saying your site location helps you on the radioactivity 20 release? 21 MR. SOWERS: I'll talk about the dose results 22 from these two and where the dose comes from. All these are 23 legitimate concerns with the tube rupture. 24 In this analysis, in fact, in both of them, you 25 never get to the point where the steam generator gets near (G~'s ACE-FEDERAL REPORTERS, INC. I 202-347-3700 Nationwide Coverage 800-336 4646
9250 10 10 150 J overfill because of the leak rate, because at the same time } DAV/bc 1 2 on that same generator, you've got the fully stuck open ! l 3 EDV. 4 MR. MICHELSON: What would happen if you did not 5 have a stuck open relief valve? 6 MR. SOWERS: This gets a whole lot easier to 7 analyze, is what happens. 8 MR. MICHELSON: What do the curves look like? 9 MR. SOWERS: I really couldn't compare them right 10 here. That is one of the analyses that is in the FSAR. 11 MR. REED: Supposing out of those things that you 12 have mentioned, you addressed the heat under natural (~ ; 13 circulation? Lj 14 MR. SOWERS. Natural circulation really depends 15 on the temperature difference that you establish between the 16 steam generator and the reactor core, because it depends 17 really on the density difference between the cold leg and 18 the hot leg; the fact that you're cold in the hot leg and 19 it's heavier and wants to go down doesn't really depend on 20 the overpressure you have on the system, as long as you 21 maintain the temperature difference between the core and the 22 steam generator, which is up here with your stuck open ADV 23 helping you cool down your steam generator. You really 24 don't have any problem maintaining natural circulation. l) 25 MR. REED: I'm not so sure that I'm easily
\_/
ACE-FEDERAL REPORl' ?S, INC. 202-347-3700 Nationwide Coverage 800-336-6646
9250 10 11 151 1 convinced. This is a U-tube steam generator system. Right? {v']DAV/bc 2 The driving head comes from the core only. There is no 3 driving head from the steam generators in a low flow 4 condition. 5 MR. SOWERS: I guess in any closed loop system, I 6 always think that the driving head comes from the total 7 differences in densities in both legs. 8 MR. REED: You have to look at U-tubes and where 9 the cooling takes place. If it takes suddenly, as I would 10 expect it would, you might have the'same density. 11 MR. SOWERS: Of the U-tube, you're talking of the 12 steam generator itself, just that part.
'/~T 13 MR. REED: So I have some concern about whether - V 14 or not, whether this pressure of decay that you have, you 15 would maintain decay heat removal.
16 Also, I think you have accumulators on this , , 17 system. Right? Or nitrogen? Potential nitrogen injectors? 1 18 Nitrogen pressurized accumulators? 19 MR. SOWERS: Yes. 20 MR. REED: Will they be allowed to inject along 1 21 this trail of pressure? 22 MR. SOWERS: Not until you get down to 600 23 pounds. 24 MR. REED: So they will be injecting and _ (} 25 contribu, ting their potential for a vapor lock or gas block. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
9250 10 12 152 1 MR. ADNI: The procedures specifically state when }gpAV/bc 2 to isolate. 3 MR. REED: Now, wait a minute. You're telling me 4 you can isolate your accumula' tors? 5 MR. ADNI: Absolutely. 6 MR. REED: You do not have racked out breakers on 7 your accumulators? 8 MR. ADNI: They're locked and open but the 9 procedures at the particular point in time. 10 MR. REED: In other words, these POS selected 11 test operators, they're really good, I agree with you 12 there. But they have another burden here while they're (~') 13 doing these other things with the auxilliary spray system. v 14 They're going to go unrack certain breakers and keep the 15 accumulators from injecting. 16 MR. ADNIr Right. Many hours into the incident. 17 It's a long drawnout process. It's nothing that has to be 18 done within five, 10, 15, 20 minutes. We're talking several 19 hours. 20 MR. REED: I know the POS selection tests will do 21 great things but what I'm worried about, the decay heat 22 removal and all the other things that are taking place, I 23 assume that you people have done not only pressure curves 24 but you have looked at all these other injections and their (~' 25 influences? V) ACE-FEDERAL REPORTERS, INC. I 202-347-3700 Nationwide Coverage 800 336 6646
I l I 9250.10 13 153 DAV/bc 1 MR. SOWERS: Yas. And I think that that 2 capability will be tested during the natural circulation 3 cooldown test. 4 MR. REED: Realistically, are you going to let 5 the accumulators inject? I'd like to see that. 6 MR. ADNI: No. What we're going to do is follow 7 the procedure. 8 MR. REED: Keep in mind now that when you rack in 9 the single valva, you're back to another single component to 10 prevent accumulator injection. 11 MR. SOWERS: Okay. Going on looking at this, 12 just a couple of other points. You can see where the (} 13 operator initiates the spray. You get depressurization 14 through this period solely because of the spray initiation. 15 At the same time, he's recovering level in the pressurizer 16 to the point where he will meet the criteria for throttling 17 the HPSI flow. And from here on, the depressurization 18 results from a combination of those two ef fects. 19 MR. EBERSOLE: In the meantime, the secondary has 20 been getting a discharge from the ruptured tube, mixing with 21 whatever water is in the secondary and it's going out. 22 MR. SOWERS: Right. All during this time that 23 ADV is stuck open, eventually, you get to the point where 24 you're essentially on the same path that the original (~) v 25 analysis had you on. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(H5646
9250 10 14 154 Iv. IDAV/bc 1 (Slide.) 2 If you can just lock at the dose results from 3 these analyses, the FSAR analysis which assume the aux spray 4 initiation at 17 minutes, the-two-hour dose at the 5 exclusionary boundary was 200 rems thyroid doses. The 6 .eight-hour dose, and I might mention that it's eight hours, 7 because in eight hours you reach. shutdown cooling a 8 conditions; entry conditions is 41 rem. 9 When you delay the aux spray operation until.two 10 hours into the event, your two-hour dose increases to 208 11 rom. And your low population zone dose to 44 rem. 12 MR. EBERSOLE: If, for whatever reason, I can't (.) ( . 13 get the aux spray going -- for instance, I've solidified the 14- primary system because of inadvertent solidification due So 15 high pressure injection, but you did have access to high 16 pressure injection and you got that load in there? 17 MR. SOWERS: Unless you have a hole in the top of 18 your pressurizer, I would doubt that you would get 19 solidification on high pressure. 20 MR. EBERSOLE: There'll just be a higher pressure 21 steam condition? 22 MR. SOWERS: You'll reach the shutoff head of the 23 HPSI's. 24 MR. EBERSOLE: With the steam still up there. () 25 MR. SOWERS: With a leak in the system, you don't i 1 ACE-FEDERAL REPORTERS, INC. ' 202-347-3700 Nationwide Coverage 800 336 6646 I
9250 10 15 155 gAV/bc 1 even reach that kind of an equilibrium. 2 3 . 4 5 6 7 8 9 10 11 12
; ') 13 v
14 15 16 17 18 19 20 21 22 23 24 7 's 25 L) ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 4646
9250 11 01 156
'UJDAVbur 1 ~ MR. EBERSOLE: You would just really compress the 2 steam level and hold it to the close-off pressure?
3 MR. SOWERS: Not quite at the close-off 4 pressure. You would hold it at the pressure where the HPSI 5 flow rate equals the leak flow rate. That stabilizes the 6 level. 7 MR. EBERSOLE: I would argue that if you can 8 spray you can depressurize. 9 MR. SOWERS: At'that point, yes, you can 10 _depressurize. 11 MR. REED: Could you tell me the size leak 12 assumed at, let's' say, 1600 psi or wherever? You say O
- ' 13 single' tube flow. What is the gallons per minute?
14 MR. SOWERS: At the point where the pressure is
~
15 about 1600, I think it was about 60 to 70 gallons per minute 16 through the tube. 17 MR. REED: That is an awfully low number to 18 assume, isn't it? 19 MR. SOWERS: I am fairly sure that is the 20 number -- per second , excuse me, 60 pounds per second. Tha t 21 makes more sense. 22 MR. REED: Pounds per second. 23 MR. EBERSOLE: Let me hypothesize that you would 24 do a very good job of spraying, indeed, and you can turn it O 25 off, and in fact now you achieve a loss of the steam j ACE. FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-66 4
9250 11 02 157 DAVbur } 1 bubble. You have reduced it down. 2 At that point, then, the high pressure injection 3 pumps can cancel, can't they? 4 MR. SOWERS: I am not sure. I haven't thought 5 that through. 6 MR. EBERSOLE: You know what I am trying to do is 7 kill your spray. 8 MR. SOWERS: I might talk about these doses just 9 for a second. 10 90 percent of that dose occurs really fairly 11 early into the event. When you open the ADV, you start 12 losing inventory from the steam generator, and you uncover ('~ ; 13 the U tubes. For a period the U tubes are uncovered.
~-)
14 : One of the reasons why that dose is as high as it 15 is, is during that period when the U tubes are uncovered it 16 is conservatively assumed that all of the activity 17 associated with that leak is released to the atmosphere. So 18 we don't try to figure out what part flashes and what part 19 doesn't and where the water goes. 20 MR. EBERSOLE: And this is at the standing 21 concentration of fission products in the coolant, the tech 22 spec concentrations? 23 MR. SOWERS: This particular analysis assumes the 24 preaccident iodine spike in the system. (; MR. EBERSOLE: 25 So it is pessimistic, really? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
l 9250 11.03 158 T DAVbu r. . 1' 'MR. WARD: What meteorology is assumed for these
.V 2 numbers?
3 MR. SOWERS: Accident meteorology. 4 MR. PAJUMBA: I am Dave Pajumba, from Combustion 5 Engineering. 6 What we use is the specific site meteorology that 7 we. find in Chapter 2 of the SARP. ; 8 MR. WARD: So it is 99.5; that is what the 9 specific numbers.are based on? 10 101. PAJUMBA: I am not sure exactly.
~
11 ~ MR. SOWERS: We can check. Meteorology isn't the-12 strong point. 13 MR. MICHELSON: Is your atmospheric dump valve ({} 14 upstream of the MSIV? 15 VOICES: Yes, it is. 16 MR. MICHELSON: Now, in the unlikely event that 17 this stuck open ADV can never be closed, which is, I guess, 18 a possibility -- the reason it stuck open is that it has got 19 a bent stem or whatever -- how is this event finally 20 terminated? 2;l MR. SOWERS: .You would get under shutdown cooling , 22 and cool the whole system down below 210. 23 MR. MICHELSON: Would this before you have 24 completely filled the steam generator? (). 25 MR. SOWERS: Yes. 1 ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6646
9250 11 04 159 1 MR. MICHELSON: And your calculations show that gggpAVbur 2 is the case? 3 MR. SOWERS: By the end your leak rate has gone 4 down as you depressurize the primary. The leak rate has 5 gone down substantially. 6 MR. MICHELSON: But you are never able to 7 terminate this until such time as you do get to essentially 8 nearly total depressurization? 9 MR. SOWERS: Yes. 10 MR. MICHkLSON: And that takes many hours. That 11 is what I wanted to find out. And you have done that 12 calculation. (} 13 And how far up in the steam generator does the 14 level finally terminate? 15 (Slide.) 16 MR. SOWERS: Let me talk a little bit about what 17 we assumed for this cooldown rate. 18 MR. MICHELSON: That won't answer my question. 19 MR. SOWERS: It addresses it indirectly. 20 We chose a very conservative cooldown rate during 21 that period in order, really, to maximize most of the bad 22 things that can happen. 23 Your depressurization during that period, of 24 course, is limited by your cooldown rate. You have to (} 25 maintain subcooling. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
9250 11 05 160 1 We could actually have cooled down much faster
- { ' DAVbu r -
2 during that last period and in fact reached the point I just 3 talked about at the end of that eight hours. 4 So it is another conservatism we put in. 5 MR. MICHELSON: Now, how about the answer to my 6 question? Where does the water level finally end up in, 7 let's say, your most pessimistic of calculations? 8 MR. SOWERS: In this calculation, Dave, where did 9- it end up? It hit a maximum. 10 MR. PAJUMBA: The tube leak rate at the end of 11 the-transient was less than the amount of flow that was , 12 going out the ADV. So actually the operator takes action in
-( ) 13 this analysis to cover the tubes, and it is about 70 percent 14 . level.- So you are at 70 percent indicated level at the end 15 of the transient.
16 MR. MICHELSON: In the steam generator? 17 MR. SOWERS: In the steam generator. 18 MR. EBERSOLE: May I take the case of maybe less 19 than a full break of the steam generator tube, a split which 20 is so unfortunately sized that it never permits i 21 depressurization to occur down this far, and you simply sit 22 there with the charging pumps hopefully making'up high 23 pressure? 24 MR. SOWERS: You always depressurize to the point - () 25 where the HPCI can make up some flow. If you don't use 1 ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-3364M6
9250 11 06 161 1 spray, you will reach for that equilibrium. If it is a jggDAvbur 2 smaller break, it would be a smaller leak, and with a 3 smaller leak area you would most likely settle off at a 4 higher pressure. 5 MR. EBERSOLE: Would it be above the closed 6 pressure? 7 MR. SOWERS: By definition, it has to be below
'8 because the HPCIs will always try -- the RCS will reach the 9 pressure where the HPCIs can supply just as much flow as is 10 going out the leak.
11 If you are above that, then the leak will cause 12 you to depressurize when you take away the HPCIs. (~) L.) 13 MR. EBERSOLE: What I am trying to find, though, 14 is a leak which causes you to lose primary inventory and 15 thus change water to steam below the pressurizer. But it 16 does not permit the injection of water except at the very 17 high pressure that you have available from the charging 18 pumps. 19 This gives me a picture, you know, of the 20 pressurizer with its volume of water being held up above a 21 stea.n-filled system. 22 Is the system basically designed to drain in that 23 case? . 24 Some pressurizer connections, I understand, will 25 be held up even though there is nothing but steam below (~)3 m ACE-FEDERAL REPORTERS, INC. ) 202-347-3700 Nationwidc Coverage 800 336-6646 l
\ ~ 9250 11 07' 2 162 them',:if they.will not intrinsically drain.
gAVbur 1 2 MR. SOWERS: I would like to look into that. 3 MR. EBERSOLE: I guess the real heart of this I 4 thing, if you could call it that, is a steam filled primary
- . 5 with a water filled pressurizer holding the pressure up so _
6 you can't do anything about it. 7 MR. SOWERS: I am just not sure you can get to 8 that. 9 MR. EBERSOLE: It.may not be. It may be that the t 10 pressurizer will drain on down and everything may be over. 4
'll MR. SOWERS: It will drain first.
4 12 MR. EBERSOLE: We know that? I am very 13- suspicious about pressurizers'that don't drain.
]}
14 MR. SOWERS: With that pressurizer, it drains. 15- MR. EBERSOLE: Great. You might have a look at 16 that. 17- MR. SOWERS: I remember the geometry fairly 4 18 clearly. 19 MR. MICHELSON: Let me finish the question of 20 steam generator level after the transient is over for the 21 case where tne system is absolutely cold and the RHR pumps 22 are running, circulating the primary water, and everything 23 is under equilibrium hydrostatic conditions. 24 Where is the water level? Is it running out on (J 25 the ground through the open ADV valve under absolutely cold ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 4646
l 9250 11 08 163 pAVbur_ 1 conditions and just looking at hydrostatic differentials 2 around the loop? Is the water running on the ground? 3 This is what you ultimately have to get to. You 4 have got to keep that decay heat removal running for many > 5 days, and you are getting the system colder and colder, and 6 you are getting it down to equilibrium hydrostatic 7 conditions. 8 Are you assured that that is not a high enough 9 head generated through the break into the steam generator to 10 lift the water and pump it out onto the ground? Not at a 11 very big rate or anything, but would you reach that 12 condition? Are you assured there is enough hydrostatic head (} 13 in that system out through the ADV that their water will 14 never come out? 15 It is a standpipe now on the steam generator. It 16 is an open hole on the steam generator. It has never been 17 l closed. 18 Does water eventually contaminate? Does water 19 eventually run out on the ground through that hole? 20 MR. PAJUMBA: Rick brings up a good point. The 21 generator is the highest point of the RC.
~ 22 MR. MICHELSON: But the decay heat removal is now 23 running and generating differential pressures around the 24 loop.
25 Have you looked at those pressures to see that
)
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(H5646
'9250 11 09 164
'NDAVbur 1- they are not adequate to lift the water right on up?
)
2 I don't know, I am asking you the question. I aa l 3 'not saying this happens at all. I just wondered. 4 You certainly would want to do the calculations 5 to show that you don't eventually end up with enough
'6 differential pressure created by decay heat removal to pump 7 the water all in the ground.
8 MR. PAJUMBA: I would speculate, once you go into 9 shutdown cooling you can control your RCS pressure to 10 minimize your leak. 11 MR. MICHELSON: You can't control it unless you 12 get this under control. Eventually, you would like to get f') 13 everything down to cold conditions with only equilibrium
's /
14 hydrostatic pressures, judging where the water levels 15 finally end. 16 That is wha't I wondered. What does that 17 calculation show? What is the final equilibrium hydrostatic. 18 water level in the steam generator or piping or whereve- it 19 finally gets to? 20 I have never done the calculations. 21 A simple question. If you don't have the 22 ' answer -- 23 MR. DAVIS: Excuse me. My name is George Davis, 24 from Combustion Engineering. Let me try to take a stab at (} 25 answering that. ACE-FEDERAL REPORTERS, INC. , 202-347-3700 Nationwide Coverage 800-336 6646 l
4 y.
?
<9250 11 10 165
'[^]DAVbur'Nl~ During the event analyzed, we are taking the
>~s .
~
-2 system into shutdown cooling. Once you'are on the shutdown 3~ ; cooling system, you can just go ahead and continue to cool
,- 4 down until you are atmospheric pressure, and at that point, 5 because the steam generator is going to be higher than the a "
6 rect of the reactor coolant system, you have gotten into 7 , shutdown cooling. You are at 70 percent level in the steam f C ' generator, and --
,* 9 MR. MICHELSON: Itthink'you still missed my
+4 ~,
10 point. 11 Your loop going around through the U tube is not
,, 12 at atmospheric pressure. It is at an artificially high 13 pressure created by the decay heat removal circulating (J'T 14 system. ,
15 At the b'reak point there is still pressure, and 16 the pressure is_ forcing water still out of the break at a S 17 very'small rate until such time as there is enough headwater m7
)g 18 above that break to go into equilibrium with whatever the
^"x "' 19 pressure is.at that point. QY,i 20 And I am just asking, what is that elevation? c,. 21 MR. DAVIS: Even as in normal shutdown cooling, 22 you start out with a full system and yo,u bring the water 23 level down so that you can eventually'take it off through 24 that vessel head. () 25 So once you are in shutdown cooling and you are
/
W ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide coverage 800-336 6646
.9250 11 11 166 1 completely depressurized, you can control that water level.
jggDAVbur 2 MR. MICHELSON: But you don't have a hole in your 3 tube under those circumstances. 4 MR. SOWERS: I think the answer is that we 5 haven't looked at that specific circumstance. 6 MR. EBERSOLE: Wait a minute. The suction for 7 RHR is at the high point of the vessel, isn't it? 8 It is the normal discharge point for circulatory 9 system discharge into the steam generator. That is where 10 you take suction RHR to provide a flow upwards in the core, 11 and then it is a closed system. So the return flow is the 12 other side, and there can be net zero pressure on the pumps (y 13 at that point. 14 MR. MICHELSON: There can be, that is right. 15 Under these circumstances, are they? 16 That was my question: 17 Under these circumstances, what is the final 18 equilibrium hydrostatic pressure at the break, which is very 19 close? 20 MR. EBERSOLE: Eventually, they have got to take 21 the lid off. 22 MR. MICHELSON: Oh, yes, but that is a whole 23 different kind of operation, the throttle pumps and so 24 forth. Now, I assume we are running these DHRs to get a lot () 25 of heat removal. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
9250 11 12 167 AVbur 1 MR. SOWERS: We can look. 2 MR. MICHELSON: It is a simple question. l 3 MR. HAYNES: Let us take a look at it. We will 4 respond to it this afternoon.
- l 5 MR. REED: I would like to make a point that goes 6 beyond this particular PWR. It goes to'all of them, but 7 related to this PWR.
8 This PWR, because of the size of the tube ring 9 that you have assumed, cannot transfer the heat by water 10 injection from safety injection pumps through the break into 11 the secondary side of the steam generator, assuming out in 12 the accident that that heat would not take care of the decay
~
(T u 13 heat at the core. So you must in these scene maintain j
- 14 natural circulation through the steam
- generators to transfer
{ 15 the heat, the total mcay heat. 16 It al'. .y r )others me to see these complex scenes 17 where, you know, we hear more and more about bleed and feed 18 as an option and an alternate to decay heat removal. Here 19 you have not even a single valve or small valves or an i 20 attempt at bleed and feed. You must go the natural 21 circulation decay heat from the steam generator. 22 (Slide.) 23 MR. SOWERS: Just to revisit for a second the 24 depressurization during a steam generator tube rupture, and 25 this really after you set up that equilibrium between the (]) ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
9250 11 13 168 (N V DAVbur 1 HPSI flow and the leak rate. 2 You depressurize primarily by throttling your 3 HPSI flow. If you-can't throttle your HPSI flow, you won't 4 .depressurize. 5 _As you throttle the HPSI flow and decrease the 6 HPSI flow rate just'a bit, you will be in the circumstance 7 where the leak rate will exceed the HPSI flow rate. .Okay, 8 you will drop pressure until those two reach equilibrium 9 again. 10 Obviously, you don' t do that until you have level 11 'in the pressurizer. You can't throttle the HPSIs until you 12 have recovered level in the pressurizer. The auxiliary
'I'-
13 pressurizer spray restores the level, so you can throttle 14 the HPSIs. 15 And what I would like to talk about now is 16 another system that you can use to accomplish the same 17 thing: restore level in the pressurizer so you can throttle 18 the'HPSIs and depressurize.
-19 (Slide.)
20 The system is the pressurizer gas vent system. 21 There.is just a small diagram here to illustrate it. 22 The pressure comes off the top of the 23 pressurizer. There are two separate flow paths, one to 24 these two valves up here. The lower flow path goes through l lf'#'% 25 an orifice that is 7/32nds of an inch. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-33(H5646
1 I k?50 11 14 169~ l 1 Ultimately, you discharge out to the containment gggpAVbur 2 or.to the reactor drain tank. 3 This system is a safety grade system, and it is 4 single failure-proof. S- MR. EBERSOLE: So it is, in essence, a bleed 6 valve? 7 MR. SOWERS: Yes. What we did was we postulated 8 the circumstance, okay, you have lost your sprays, you don't 9 have them. Okay, but you still want to accomplish 10 depressurization-during a tube rupture. And we analyzed 11 doing that with a gas vent system. We also used the most 12 limiting flow path through that system. () 13 Of course, if the operator really was trying to 14 do this, preferably he would want to use the spray first. 15 That is his first choice. 16 Secondly, he would want to use this line because 17 it has a higher capacity. , 18 His third choice would be the bottom line. 19 So what we actually have analyzed now is his 20 third choice for facilitating depressurization. 21 MR. REED: The third choice is the Three Mile 22 Island 2 backfit hydrogen gas phase, right? You have added
.23 a parallel line of one inch or three-quarters of an inch 24 type of double series valves, double and triple discharge?
()
~
25 MR. SOWERS: Yes. ACE-FEDERAL REPORTERS, INC. 202-347-3700 ~ Nationwide Coverage 800-336-6646
-9250 11 15 170
- f). .
V DAVbur -1 Okay. 2 (Slide.) 3 What I have shown here is again a comparison of 4 two results. The solid line is the result that I was just 5 talking about. Wait for two-hours, then turn on the. spray, 6 with the same actions opening the ADV. One of them sticks 7 open here at.two hours. And the solid line initiates 8 depressurization by turning on the aux spray. It throttles-9 HPSI flow. 10 In this analysis, he opens the gas vent here. So 11 you don't really get much depressurization from the vent 12 system. What you do is you recover level. When you recover l 13 levebr-you throttle the HPSIs, and throttling the HPSIs 14 gives you your depressurization. 15 Now, this kind of a process really is 16 characteristic of a system that has a hole in it. Okay, a 17 tube rupture. 18 This is where you get your ultimate 19 depressurization. You will always come off , you know, hands
.20 off, and you will get to that steady state where things 21 balance, and you have to ultimately get away from that.
22 The only way to do that is to throttle the 23 HPSIs. So that sends you down on a gradual 24 depressurization, okay? 25 Eventually, you get down to the same place where ; ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336 6646
'9250 11 16 1
171 DAVbur' 1 you were before. 2 MR. EBERSOLE: It is throttling the HPSIs along 3 .with this depressurization?
- 4 MR. SOWERS: It is throttling the HPSIs, and it 5 is really illustrated here that ultimately gives you the-6 .depressurization.
7 MR. EBERSOLE: Wait a minute, if I just throttle 8 the HPSIs -- 9 MR. SOWERS: What happens when you throttle the 10 HPSIs? 11 You have got a level in the pressurizer, okay? 12 You throttle the HPSIs. Now, when you do that, you are 13 going.to be in a circumstance right after you throttle it d 14 .where your HPSI flow rate =is going to be less than the leak 15 rate. The pressurizer level drops. You expand the bubble. 16 The bubble. expansion drops the pressure, okay? 17 18 19 20 21 22 23 24 0: 25 ACE. FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 e646
19250 12 01 172 ()DAV/bc 1 MR. EBERSOLE: When you start the HPSI's, what 2 you do is sustain the hot water in the pressurizer, which
-3 holds the pressure up. You keep the pressurizer from 4 draining, then you won't have to leak.
~5 MR. SOWERS: I still don't understand.
6 MR. EBERSOLE: If I don't start the HPSI at all, 7 the level will fall and I will eventually lose that bubble 8 of hot water which is in the pressurizer and will have gone 9 into the secondary. 10 MR. SOWERS: If you don' t start the HPSI's in the 11 tube rupture, I'll guarantee you, you depressurize. 12' MR. EBERSOLE: That's wha t I 'm saying . And then, O (_) 13 having depressurized, the story is all over. I can now have 14 free access to the primary loop in whatever way I want. I'm 15 not so sure but what that night not be a better way to go. 16 Do you follow me? 17 MR. SOWERS: I'd rather have the HPSI's on. 18 MR. EBERSOLE : What you're doing is you're 19 sustaining the high pressure bubble. 20 MR. SOWERS: But I'd rather keep the core 21 covered. 22 MR. EBERSOLE: Well, presumably, you didn' t get 23 that f ar down. _ 24 MR. SOWERS: Not here. In any case, you do
-O(_s 25 effect that depressurization.
ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336 4646
9250 12 02 173 ! rs (J DAV/bc 1 MR. EBERSOLE: When you just use the HPSI's and l 2 17ou don' t use this one little on-inch or three-quarter line, 3 you hold the pressure up rather substantially high for a 4 long time. So it's 1,600 and it's a long way from cold 5 excess. When you open the little valve, you accelerate it 6 so that in about eight hours or so, you get a large R. In 7 the meantime, you're leaking. 8 So we need a dose level at eight hours instead of 9 two hours. 10 (Slide.) 11 EMR. SOWERS: For two hours, in fact, the doses
,12 by definition are identical because the events are identical O
N- 13 up to that. As it turns out, if you look at the eight-hour 14 doses, which is the time when you enter shutdown cooling, 15 there is no impact on the dose. So it doesn'.t really matter 16 whether you use the auxilliary pressurizer spray or the 17 pressurizer vent. Initiating them both at two hours, you 18 get the same offsite dose result. 19 MR. EBERSOLE: Even though you depressurize 20 quicker with a vent? 21 MR. SOWERS: Even though you depressurize more 22 slowly with the vent. 23 MR. EBERSOLE: Wait a minute. I don' t follow 24 that. 25 MR. SOWERS: Okay, I can explain. Go ahead and ACE FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336-6646
o i
.9250 12 03 174
)DAV/bc. 1 put the other one up.
2 (Slide.) : 3 This is the spray. This is the vent. During 4 this time, let's just- kind of review where you're at right 5 here at the two hours. Okay, you are pressurized. Yo u ' ve 6 got a leak. It's that 16 gallons per second we were talking 7 about. Your steam generator level is up above the top of 8 the tubes and the dose when the leak is below the water 9 level comes from the fraction of the leak that flashes to 10 steam and goes out the ATV.
'll MR. EBERSOLE: Let me first characterize the 12 curve here. When you say, first initiates depressurization,
( )1 13 you're talking through the use of the spray process? 14 MR. SOWERS: When I say " initiates 15 depres suriza tio n" , it means from this plot, using this 16 spray, throttling HPSI's, at this point, he opens the vent. 17 MR. EBERSOLE: But he loses the spray. 18 MR. SOWERS: He does not have the spray, 19 correct. In the dotted line analysis. 20 MR. EBERSOLE: This then gives him a somewhat 21 longer time to depressurize without the spray? 22 MR. SOWERS: Without the spray, yes. 23 MR. EBERSOLE: As a matter of fact, it's about a 24 30 percent increase in time at high pressure. So you get a () 25 dose or the case where you use the vent of about -- what ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverase 800-33A'4
() Laj0 12 04 175
-1 DAV/bc 1 'will I say? -- six hours or thereabouts. Certainly, it 2 persists at a higher-leakage rate for a long time.
3 MR. SOWERS: Yes, it does persist at a higher 4 leakage rate; at the same time, you have safety injection 5 making up primary fluid at a higher rate, and you are 6 -lowering the enthalpy of the primary system. Yo u ' re 7 cooling it down just due to safety injection flow. 8 When it goes through the tube, a smaller fraction 9 of it flashes to steam. So a smaller fraction of the leak 10 goes out the top and contributes to the dose. 11 MR. EBERSOLE: On the secondary?
' 12 MR. SOWERS: Right, because of this , the dose 13 results end up staying the same. The leak gets larger, 14 yes. The dose stays the same.
15 MR. EBERSOLE: I see. It's an interesting set of 16 conditions. Thank you. 17 (S lide . ) 18 MR. SOWERS: Just in summary, the dose results 19 for this analysis are relatively insensitive to the time 20 when you initiate auxilliary pressurizer spray during the 21 first two hours, any time in there. I'd also say that 22 nobody is going to wait for two hours if they've got a 23 choice.
-) 24 MR. EBERSOLE: Are these valves remote manual?
25 MR. SOWERS: Yes , from the control room. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coversee , 800-33(H5646
9250 12'05 176 ()DAV/bc l' MR. EBERSOLE: They are not safety grade? 2 MR. ADNI: Yes. 3 MR. EBERSOLE: They are safety grade. So you do
~4 have valving discharge from the primary system pressure.
5 It's just that they' re tiny? 6' MR. SOWERS: They're small. The doses are within 7 the dose limits even when you delay actuation until a few 8 hours, and we do have the pressurizer gasified system, which 9 can accomplish the same purpose to f acilitate a 10 depressurization during a steam generator tube rupture. 11 MR. EB ERSOLE: Tell me, there is sort of a 12 philosophical objection on the part of the designer I think /~%) (_ 13 to any kind of a system of depressurizing the primary. Yet, 14 you have one. 15 What happened to the objections to having that? 16 MR. SOWERS: I'm not sure there's a philosophical 17 objection to any method of depressurizing a system. 18 MR. EBERSOLE: Just the PORV type. I agree with 19 you, tha t's the worst kind. You know, there have been other 20 valves used. And the valves you use, I guess, are good 21 valves. There's no real potential for spurious opening. 22 And even if you did have spurious opening, so what? i 23 MR. SOWERS: It's redundant. You either have to 24 open two valves, three valves or, well, open three valves or () 25 go through two valves and an orifice. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage M336-6646
-9250-12 06 177 ,- ~ .
n' DAV/bc~ l MR. EBERSOLE: And the frequency of that 2 occurring should be almost zero. 3 MR. SOWERS: We'd hope so. Just to tie up... 4 MR. REED: I'd like to ask a question. This 5 three quarter inch parallel line with the TMI-2 backfit, is 6 that a recent innovation? That three-quarter inch line? 7 Does this belong on all System 80 's or just on Palo Verde's? 8 MR. SOWERS: This is a Palo Verde design and it 9 existed from the beginning. 10 MR. REED: I don' t know how you answered my 11 question. 12 MR. SOWERS: I'll let George Davis speak to it. ('T 13 MR. DAVIS: , I'm George Davis of Combustion. That 14 vent line was with the 736 orifice has been in all our
~
15 plants. It was usually attached to a manually, 16 locally-operated valve. 17 MR. REED: I'm aware of the 736 history. I'm 18 talking about the one in three-quarter inch line. 19 MR. DAVIS : Okay. On the other plants, when it 20 came to implementing the post-TMI requirements for reactor l 21 gas venting systems under the high points, most of the 22 plants just used another line of the same size. 23 Arizona, in this particular case, used a larger 24 line. Most plants, you would find the 736 orifice in both. 25 MR. REED: So you' re saying that Arizona has a ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverase 800 336-6646
9250:12 07 178 rm .(_)DAV/bc 1 slow ~ alternate means of depressurization but the other 2 plants have an even' slower? 3 MR. SOWERS: No. Remember the analysis we did 4 use the'732 second inch line? 5 MR. REED: What you put on the board, was that 6 730~ seconde' 7 MR. SOWERS: Yes. 8 MR. REED: I guess there's a difference between 9 slow, slower and slower. 10 MR. DAVIS: As a clarification, when we did the 11 studies, CEN-239 for the rapid depressurization issue, we 12 put some members in that talked about the flow path to the (l (> 13 larger pipe there. 14 You got a depressurization rate that was pretty. 15 much on the same order of magnitude as using the aux spray. 16 So if you used that path, you would have a result that's 17 similar to using the aux spray. If you used the 730 seconds 18 orifice, you'd get the results that Jerry has presented. 19 MR. EBERSOLE: I guess it might be important 20 before you sit down to mention, I believe ANO-2 has a bigger 21 pipe. Righ t? , 22 MR. DAVIS: ANO-2 has, I thinks a three-inch pipe 23 off of theirs , but that was added for long-term cooling 24 considerations plus LOCA. 25 MR. EBERSOLE: You mean, feed and bleed as well? l ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-33(H5646 l
9250 12 08 179 (,)DAV/bc 1 It makes possible feed and bleed? 2 MR. DAVIS: It wasn't put there for the purpose
.3 of the feed and bleed operation. It was put there'for a 4 post-LOCA long-term cooling providing in the flowpath.
5 MR. EBERSOLE: Where does it go? Where does it 6 -teominate? Is it a closed path? 7 MR. TURK: It goes to the quench tank. It's part 8 of a system that was put on to meet the requirements of a 9 low temperature overpressurization, or cold 10 overpressurization. That consists of two spring-loaded 11 safety valves that are set around 400 pounds and normally 12 valved out, but valved in for water solid operations. f~T
'( / 13 This is just a bypass around those valves.
14 MR. EBERSOLE: I see. So it would go to the 15 quench tank. It could be picked up in the sump if you 16 wanted to use it that way. 17 MR. TURK: Right. The rupture disk would blow 18 and it would go into the sump. 19 MR. EBERSOLE: And that is a low pressure, direct 20 water flow cooling? Well, still, this just depressurizes 21 but it provides no access for the feed and bleed, whether 22 tha' comes or not? 23 MR. SOWERS: That's true. Any other questions? 24 (No response.) ( 25 MR. EBERSOLE: Thank you. Let's see. We're 15 ACE-FEDERAL REPORTERS, INC.
!02-347-3700 Nationwide Coverage 800 33M646
9250 12 09 180
,f 3
'Li'DAV/bc 1 minutes behind. Mr. Houston, can you give us your talk here
-2 before a break for lunch?
3 MR. HOUSTON: I'll try to keep this as brief as I 4 can , given the hour. I'd like to begin', however, by making 5 a point. In liau of the events of the past few months with 6 respect to the Palo Verde plant, I must say it comes as 7 somewhat of a surprise to most of us on the staf f that we' re 8 here discussing the subject. We thought, with the licensing 9 of unit one on Palo Verde, the CESSAR System 80 design 10 approval, that we were dealing with what we might call a 11 fully safety grade auxilliary pressurized spray system. 12 In retrospect, the staff had not carefully (
\- 13 examined all aspects of the system, including that part of 14 the system which provides the water to the APSS.
15 In retrospect, this could be said to be an 16 oversight on the part of the staff. We did have 17 representations from the applicant that what we understood 18 to be the system was fully safety grade. 19 It turns out we had a miscommunication, 20 apparently. When we talked about the system, we were 21 thinking of it as a functional system. And, apparently, 22 when the applicant was talking about it, he was talking 23 about a physical system which was added to essentially a 24 preexisting charging system. (~~)
\- 25 In a minor respect, we were also caught because ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
I s I9250112 10 181
~
l DAV/bc 1 -of . divisions within ' the staff on who reviews what parts of 2 :what systems.
~3 With ' that out of the way, however, we came to the 4 realization on examination of -the September 12th event that 5 -there were significant single failure vulnerabilities in the 6 APSS system. And our knee jerk reaction was that'it should
-7 be made fully safety grade.
8 With plants that are essentially built, however, 9 in unit ~ one and unit two, I'm not' quite sure how far along 10 unit. three is concerned, we did cause to recognize that this 11 may be a somewhat unrealistic and possibly even unnecessary 12 hard position for us to take. LOL 13 So we have had a number of meetings with Arizona 14 Power and as recently as last week, we did pose to them what , t 15 essentially is a position, at least within our division. I 16 might misrepresent it totally as a staff position at this 17 particular point in time 'because it's not yet fully 18 resolved. 19 With respect to the handout that I gave you, it 20 got the statement together in improprer order. If you'll go 21 to the third page, I would like to start with identifying in
- 22 the auxilliary pressurizer spray system the basic reasons 23 why we think it's important to say -- the first one, the 24 very important one is that the APSS has been identified as O 25 required to function for plant shutdown in order to meet the ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33MI646
19 50l12'11: 182
.DAV/bc 1 requirements of the famous branch technical position, l
2 RSP-5-1 on decay heat removal. (
~
3 Again, I would like to point out in this respect
.4 :that Palo Verde is a class two plant, and this branch 5 technical position does not require that all:of the 6 components of the system be fully. safety grade.
7 MR. WARD: What does class two mean? 8 MR. HOUSTON: If I recall correctly, the plants 9 :that do not yet exis - were separated for plants under
'10 construction and were separated from operating plants, so it 11 was a chronological cutof f in terms of the extent to which a
.12 position would be backfitted.
.O
-13 So class two fell in the middle and they were 14 required, the application of the position required some 15 things but not the full safety grade application.
16 It was also the staf f's understanding that the 17 applicant was taking credit for the auxilliary pressurizer 18 spray system to function in order to mitigate the postulated 19 steam generator tube rupture event. 20 In addition, as you probably well recall, the 21 function of that system was factored into and was a 22 significant part of the staf f's conclusions on the 23 combustion engineering PORV study, the conclusion of which 24 was sort of lef t hanging in the sense that A-145 is
.O 25 finished, we'll have an answer to the question.
ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coversee 800 336 4646
9 50 12 12 183 DAV/bc 1 Hopefully, we'll hear more about that this 2 afternoon. The staff understands that the current, and by 3 that, we mean prior to any of these enhancements that have 4 been talked about this moraing, design, meets the 5 requirements of branch technical position 5-1. 6 However, the design of what we now will identify 7 as the auxilliary pressurizers, spray system, water supply 8 system, does not meet the safety grade requirements. 9 on the next page, we list some of the water 10 supply system's vulnerabilities. Many of these have been 11 identified to you in the context of where enhancements have ,s 12 occurred. The level instruments on the volume control tank ~ 13 relative to a single failure, some improvement has there i 14 been made. Isolation valve 501, which is the isolation 15 valve right below the volume control tank, is vulnerable to 16 single failure. 17 The water supply line from the refueling water 18 tank by valve HV-536 is vulnerable to single failure. 19 The gravity water supply line valve, and here we 20 have an interchange on the valve numbers on these handouts, 21 I believe that should be 532, and the containment isolation, 22 valve, which should be 524, are vulnerable to single failure 23 power supplies to the volume control tank 501 and HV-536. 24 On the handout here, it's indicated that they are x ,/ 25 not reliable. They were furnished with class lE power, but ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Com 80 4 336 6646
'9250 12 13 184 DAV/bc. 1 :through .the nonlinear motor control center which was knocked 2 out 'and isolated on the ' safety injection, as they .found out
-3 -~during that event.
4 -Finally, the charging pumps are vulnerable to gas-5 binding due to failure of VCT isolation. Now if we go back 6 . to the first page, the staff position, and there are two
' .7 aspects here of 'it, that we have tentatively taken -- ' that 8 we '.ve taken wi thin the past week -- the first one, which 9 relates to the last analysis which the applicant has just 10 shown us, the staff felt that the steam generator tube 11 rupture event should be reanalyzed. That is, reanalysis 12 compared to what's in the FSAR.
- 13 Assuming the worst single active failure and-14 using only safety grade systems and components for RCS 15 depressurization following the event, and it is response to e
16 that statement, as I understand it, that the applicant has i 17 4one the analysis in which the assumption is made that the 1 18 APSS is not available. 19 I might say that we just received that analysis 20 late yesterday af ternoon and we have not really had a chance 21 to review it. The staff took the position that certain 22 system enhancements and verifications were required. - 23 Most, if not all of these, at this point in time I 24 have been addressed by the applicant and we have just seen O 25 these providing power to the two valves from a lE motor l s. ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 3364646
)
il
9250 12 14 185 r M DAV/bc 1 control center. So that following either loss of offsite 2 power or safety injection actuations, power would not 3 automatically be removed. 4 MR. EBERSOLE: Let me ask a question. Everything 5 seems to be driven about your concern about a reactor or by 6 a tube failure in the steam generator. 7 Now, OR-81 is in place. How long are you willing 8 to believe that you'll want to stand that at high pressure 9 before you can get to low pressure ECCS? You don' t mention 10 this as a problem or a disadvantage. Everything seems to be 11 driven by your concern for the main steam generator tube 12 failure. You' re relegating that to A45, I take it, and the G U 13 combination of how-long I have to be on high pressure until 14 I can get on low pressure cooling. 15 16 17 18 19 20 21 22 23 24 (O 'd 25 ACE FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 900-336-6646
,9250 13 01 186
_p (_ DAVbw 1 MR. HOUSTON: As a broad decay heat removal 2 problem. 3 MR. EBERSOLE: If I could never get on low 4 pressure cooling as to how many hours I'd have to stay up 5 is not mentioned. 6 MR. HOUSTON: You're right. It is not 7 mentioned. I'm not clear that has been a significant Staff 8 concern. 9 Can you add anything to that? 10 MR. EBERSOLE: You can almost see you're willing 11 to stay up indefinitely at high pressure. 12 MR. HOUSTON: I think the Staff, there's probably
.I') 'x s 13 at least a moderate Staff consensus that where we would like 14 to be today, but are not , would be to have taken sometime 15 ago the position that applicants must demonstrate the 16 ability to go to cold shutdown through a safety grade path.
17 A management decision was made several years ago that we 18 would not require that. That management decision still 19 exists today. 20 Not every Staff member agrees with that, but it 21 was made. 22 MR. EBERSOLE: But with the pressure of the steam 23 generator tube failure, you intend to be able to do that 24 here. b ss 25 MR. HOUSTON: Following the standard review plan, ACE FEDERAL REPORTERS, INC. 202 347-3700 Nationwidc Coverage 800 336-6646
9250 13 02 187 ()DAVbw 1 this is one of the so-called Chapter 15 events that are 2 analyzed. The decay heat removal problem is the RSBS-1 3 position, which we _ attempt to implement. With respect to 4 posulated accidents, it is the steam generator tub rupture 5 event. It's the only event, I think, in Chapter 15 that 6 relates to the prospective use of the auxiliary pressurizer 7 _ spray system. 8 MR. EBERSOLE: Thank you. 9 MR. HOUSTON: -This is where we have a leg to 10 stand on. Continuing with the enhancements we felt were 11 necessary and some of which I think prior to last week, 12 Arizona Public Service had already agreed, I think, to carry O)
\_ 13 out, particularly, the reliability of the VCT level 14 indication, which we've heard about, the realignment for 15 automatic actuation of valves 501 and 536, on low, low VCT 16 level and the lost of off-site power, in order to align the 17 charging pump to take suction from the f uel and water _ tank.
18 MR. MICHELSON: Excuse me. What is the branch 19 position then concerning the requirement for safety grade 20 transfer from the volume control tank to the fuel and water 21 tank? Is that clear from the branch technical position as 22 to how you j udge? You said not everything needed to be 23 safety grade. Does that particular feature need to be 24 safety grade? ( 25 MR. HOUSTON: I don' t believe the Staf f has taken ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336 6646
-9250'13 03 188 \ 'DAVbw 1 a position that it does need to be safety grade to comply.
2 MR. MICHELSON: Have they looked at the 3 inadvertent actuation of the system after having valved off 4 from an empty volume control tank that would now and up 5 valving back into it. 6 Is that considered a credible single failure 7 event? 8 MR. HOUSTON: I don' t know that we've considered ; 9 that specifically. 10 Chu, could you speak to that? Chu Yang. 11 MR. YANG: I'd.like to add something on their 12 system design. The system, as currently designed, does meet (\ ') 13 the condensation allowing for Class I plants to having a 14 manual operator action in the control room to do some 15 realignment to collect a single failure 16 In this system design, in case of single failure 17 501 or 536, they do have a local manual which can be 18 realigned, so that water will be available from the safety I 19 injection line. 20 MR. MICHELSON: I understand all of that, I 21 think. What I didn't understand is , if you decide to make 22 it automatic, which I gather they' re deciding to do, make ; 23 the transfer automatic, now don't you have to deal with 24 single failure being the failure of the automatic system O l> 25 reverting you back to the original alignment? ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336 6646
a 9250 13 04 189 ',)DAVbw 1 MR. YANG: The last manual realignment -- 2 MR. MICHELSON: No. The question is, if this 3 inadvertent realignment occurs , do you have time for the 4 last ditch realignment. The tank is already empty. It's 5 f ull of hydrogen. The pumps are sucking on it. 6 MR. YANG: They did mention they had a provision 7 to vent the hydrogen. 8 MR. MICHELSON: Yes. You have to now go through 9 and do all your time and motion study and damage studies to 10 assure yourself that you can rescue yourself from this 11 situation; that's right. And if you can show that's the 12 case, then I think it would be all right. I just was trying s (_,) 13 to find out the basis now for your review, and you're saying 14 it doesn' t have to be safety grade. Therefore, I would 15 expect to see this time and motion study as a part of the 16 submittal. 17 MR. .. SNG : In their current analysis for 5-1, 16 cool-down analysis, the time to use the auxiliary spray 19 cystem is at a very late stage. 20 MR. MICHELSON: Keep in mind now, I'm using it. 21 I made the automatic transfer, and I experienced a single 22 failure in the transfer system, and retransfer me back to 23 the original alighment with an empty tank. Okay. Now it 24 binds up the pump and whatever. If you can show that this ( 25 doesn' t damage it, if you can show that this doesn' t damage ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coversee 800 336-6646
~
9250 13 05 190 DAVbw 1 it. If you show you have plenty of time to recover, then I 2 think you're okay. 3 MR. EBERSOLE: Well, they have NPSH protection on 4 the pumps. So they function properly, and you align the 5 system an hour or two later, and you don' t care, do you? 6 MR. MICHELSON: I don't know. I haven't seen 7 that study. Hhaving already required auxiliary spray, which 8 is the situation you're in, how long can you now divest 9 yourself of auxiliary spray, before you have to reestablish 10 it? 11 MR. EBERSOLE: My question is, it's not
- 12 critical.
13 MR. HOUSTON: It's preferred, but not critical. 14 MR. MICHELSON: In other words , you're saying 15 it's never critical saying it. 16 MR. HOUSTON: I'd hate to say never critical. 17 MR. EBERSOLE: That's a long time. 18 MR. MICHELSON: You picked the worst time for 19 this to happen, wha tever you name. 20 MR. HOUSTON: I think you've identified something 21 we perhaps need to look into. 22 MR. YANG: But the current engineering judgments 23 and thinking, there is suf ficient time during the cooldown,
,_ 24 because they have suf ficient water available for hot standby 25 for four hours and then cool down the plant. So we think ACE FEDERAL REPORTERS, INC.
e 202 347 3700 Nationwide Coverage 800 336 6646
l 1 L92501 13 06' 191 f~ . L-[DAVbw 1 there's plenty of. time. l
- ,' 2 MR. EBERSOLE: When'you invoke safety functions 3 ' using nonsafety grade components, do -you establish 4 disciplines of use of this nonsafety grade equipment? You 5 know, periodic tests of elevation by toch specs, and so g 6 forth, which is inconsistent if it is nonsafety grade, but o' < 7 it is nevertheless an important consideration to be sure 8 it's there.
9 MR. HOUSTON: Normally, it is not our practice to
- 10 toch spec things that are not safety grades.
11 M R. EBERSOLE: Yes, you keep them greased and in 12 good shape and everything. ls
.13 MR. : HOUSTON: There are exceptions to this in the 14 toch specs. I think it would be very difficult for me to 15 give you a generalized answer to that questions with the 16 particular case at hand. There are proposed -- and I think 17 we are essentially prepared to accept on Unit 2, the 18 proposed . tech specs on the APSS, in spite of the . fact that 19 it is not fully safety grade.
20 MR. EBERSOLE: It would seem reasonable to invoke 21 some special requirements on nonsafety grade equipment just 22 'to be sure it's in good shape. 23 MR. HOUSTON: I could quote some other examples 24 to that, but I won' t take the time .
.O 25 MR. WARD: Well, there's some precedent for that m
ACE. FEDERAL REPORTERS, INC. 202 347 3700 - Nationwide Coverage 800 33H646
p , _b 4
?9250 13 07- >
192 y ]y DAVbw- 1 with the ATWS.-
'2 MR. - HOUSTON: That's correct.
3 MR. WARD: I don't know if it's a very good 4 precedent.- 5 MR. HOUSTON: Running down 'the final list on the
;6 items that reflected the Staff position, on the next page, 7 on the handout, maintain valve HV 532. in locked open
.8 ' position. This is the valve in the gravity feedline from-
- 9 the reactor water tank. The Staff was a little' bit 10 concerned because of the location of the tap on the RWT, 11 whether there's enough water above it to be available for 12 APSS use. So we asked them to verify that fact.
13 We've asked 'them to remove the power lock open
~
14 valve 524, which is the containment isolation v,41ve at the 15 charging line to prevent a spurious valve motion or operator-F 16 error from defeating the f unction of the aux pressurizer 17 spray system. We also asked them to verify that the Palo
. 18 Verde equipment. qualification, to assure L that the components 19 associated in the flow path of the APSS properly and
- 20 environmentally qualified. They've responded to that , =and 21 we also still have to deal with the question. We have some 22 questions in our mind as to whether the current toch specs 23 are appropriate for the system as we now understand it.
24 I might add to this list one other thing that O- 25 .the Staff is concerned about, since, if we are able on l ACE. FEDERAL REPORTERS, INC. 1 l102 347 M00 Naionwide Cowwege 800 3M 4646
3 l Tl ~ l
~r ..
9'250 13 08 193 (M. .
\>IDsVbw I review of the analysis presentsd here, dealing with the
. ~ ,,'f 2 steam generator tube rupture event, which does not make use 3 of the APSS, it would raise the question in our mind as to 4 whether or not there ought to-be identified in the emergency
,47 ,
5 procedure, the guidelines for the use of these event systems y 6 as a backup system, si'the operator would know it is
, 7 available to him. #i ,
8 MR. EBERSOLE: I presume you would have that, 9 surely, in the emergency procedures. You say you might not?
/
Ns.
,, 10 MR. HAYNES : It's already in the emergency 11 procedures.
12 MR. HOUSTON: I'm not' aware that it is. It's an [ 7, \ 13 issue which we had not specifically taken a position, but we 3-14 have suggested that we would require that. "~ ! " MR. MICHELSON: In your next to the last bullet,
.c - - J 15 ,
'#V 16 when you talk environmentally qualified, I assume you 17 include seismic?
18 MR. HOUSTON: There are seismic. 19 MR. QUAN: We also address seismic 20 qualification. p 21 MR. MICHELSON: Let me make sure you understand. 22 Your automatic. transfer circuits, all the bhsic components 23 needed for the auxiliary pressurized spray are seismically 24 qualified.
*Q.
/ 25 MR. QUAN: Not necessarily.
j' ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33H646
l 9250 13 09 194 DAVbw 1 MR. MICHE LSON : This statement that they are all 2 environmentally qualified, and you now define environmental I 3 to include seismic, so I'm a little confused. 4 MR. HOUSTON: I think there's a little 5 miscommunication, possibly. I'm not sure. Again, if we 6 have somebody here from the Staff, he can correct. I ' m no t 7 sure, as proposed, we included seismic. That's something we 8 think of as safety grade. What we are thinking of is 9 qualified to function in the atmosphere that may exist in 10 the vicinity of a component of the system. 11 MR. MICHELSON: Some people, most people, I 12 think, when you are talking about qualifying equipment, m 13 environmentally qualifying it, you mean the seismic as well, 14 but you're saying no. 15 MR. HOUSTON: I'm saying, I'm not sure. I think 16 we did not necessarily mean seismic to be included. My 17 question is that much of it is seismically qualified 18 anyway. 19 MR. QUAN: Just a point of clarification. There 20 were a couple of components that are inside the containment 21 to regenerate a heat exchanger and the loop isolation valves 22 that we were asked to look at as f ar as environmental 23 qualfication. We also included seismic as part of our 24 review, and they are seismically qualified. 25 MR. MICHE LSON : But just inside of containment. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336-6646
9250'13 10 195 DAVbw 1 Outside of containment, how about the components? Some are 2 and some aren't. You listed some earlier. 3 MR. QUAN: Right. Some are and'some aren't. 4 MR. MICHELSON: So we really can't make a 5 generalized statement about the seismic qualifications of 6 the system? 7 MR. QUAN: Correct. 8 MR. MICHELSON: There fore,--I think we have to 'say 9 it's notqualified from the point of vied of functionality, 10 because not all thexfunctional components are seismically 11 qualified, therefore, this is not a seismic system, from the { 12 viewpoint of f unctionality. From the viewpoint of pressure f b's- 13' boundary, perhaps-it is. I ' don' t know. 14 MR.' QUAN: Tha t's correct. 15 MR.'EBERSOLE: Any questions? ( 16 MR. WARD: Le t's see . How is the Staff position
'17 going to be documented?.
t 18 MR. HOUSTON: Presumably, we'll have to write an 19 SER. What we haven't had right now is a letter, actually a L '20 series of letters from the applicant. It's not, as I
'21 understand,: an FSAR revision at this point in t ime . But 22 we'll have to write an SER.
23 MR. WARD: There will be an SER. What's the
-24 schedule - for that?
O 25 MR. HOUSTON: I'm not sure. You'll have to go to ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage ' 800-336 4646
9250 13 11 196 ()DAVbw 1 the table over here. 2 What's the schedule for an SER? 3 MR. KNIGHTON: Definitely prior to licensing. 4 (Laughter.) - 5 MR. HOUSTON: Within a few weeks, I think. 6 MR. KNIGHTON: The schedule is mid-November. 7 MR. MICHE LSON: Can Staff tell me quickly what 8 the seismic situation is at Palo Verde in some manner, the 9 nature of the anticipated earthquakes? 10 MR. LICITRA: I believe the design basis for the 11 Palo Verde plant is .2G, but it includes the seesaw System 12 80 design, which has a higher G value. So there are ("T V' 13 conservatives. 14 MR. MICHELSON: As far as the geology and so 15 forth, what's the property of experiencing the SSE? In 16 other words, simply, is this a high seismic or a low seismic 17 site? 18 MR. WARD: Most SSEs, it's a return period of 19 10,000 years or something like that. 20 MR. MICHELSON: But some places it's not that. 21 MR. QUAN: As far as our opinion, I think the 22 answer to that is relatively low probability. 23 MR. EB ERSOLE: I suppose in this connection, I 24 think the SONGS, it will be this afternoon, we'll hear the 25 CE speech on how much of this is representative of the CE ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6646
9250 13 12 197 (,) DAVbw 1 design and how much is simply with this project. 2 MR. HOUSTON: In light of that comment, and in - 3 light the discussion that immediately preceded me I'd like 4 to point out that one of the significant differences here, 5 - vis-a-vis the stuck open ADV is because the Palo Verde 6 design does not have block valve, whereas the CE System 80 7 designs, I understand i~t does as interf ace condition. 8 MR. EBERSOLE: Wait a minute. Where are these 9 block valves? 10 MR. HOUSTON: Upstream of the 80 Vs. 11 MR. EB ERSOLE: Palo Verde does not have block 12 valves. - g/ A- 13 MR. HOUSTON: As I understand, presently, the CE 14 System 80 plan has a requirement for block valves to be 15 installed there. It has an interface condition. You can 16 clarify that this afternoon. 17 MR. EBERSOLE : On what, the front side of the 18 ADVs? 19 MR. HOUSTON: Upstream of the ADVs; yes. 20 MR. REED: I notice you didn' t mention these 21 other two components, single components, the regenerative 22 heat exchanger, through which the auxiliary spray must fall 23 and the spray nozzle and the pressurizer, which begins to 24 add up the number of single component f ailure potentiality. 25 MR. HOUSTON: If one considers passive failures, l l l i ACE-FEDERAL REPORTERS, INC. 202-347 3700 Natl _. wide Coverage 800-336-6646
9250 13'13 198
~L,O)DAVbw 1 tha t's correct.
2 MR. REED: Why do you consider those passive? 3 They're not piping. They' re components , aren' t they? 4 MR. HOUSTON: They're components, but in order ; 5 for them to fail, you have to have a rupture in the tube, 6 iln the heat exchanger, and some plugging or something, of 7 the spray nozzle , so it would be what we would call, I 8 think, a passive failure, but normally, we can treat single 9 failures as active failures. , 10 MR. EBERSOLE: There's no relative motion in the 11 parts. 12 MR. HOUSf0N: Yes.
-(~) .
V' 13 MR. EBERSOLE: It's a static f ailure. Tha t 's 14 beer longstanding Staff practice. 4 15 MR. REE D: I don' t the. official ways, but 16 certainly, they are vulnerable components. 17 MR. HOUSTON: We recognize that, yes. 18 MR. EBERSOLE: It could be that the defuser or 19 whatever it is in the spray system has a small margin of 20 robustness or whatever, I don' t know. It would be 21 appropriate to know if that's not the case, you know. It's 22 not something that can be blow out. I gather a rodlike flow 23 is .not a depressurization spray. 24 MR. HOUSTON: I assume it would depressurize 25 very,-very slowly. ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33HM6
9250 13 14 199
^
r %- l (_)DAVbw 1 MR. EB ERSOLE: Any other questions? 2 MR. REED: It seems to me that in view of the 3 regenerative heat exchanger and nozzle, the other 4 components, the manual operation required in all these 5 things, that for depressurization on this unit, you have to 6 also look at the three quarter inch vent as a component that 7 is needed to effect slow, I'll call it, rather than rapid, 8 _ slow depressurization, according to the curves that were 9 presented. 10 Do you agree to that? 11 MR. HOUSTON: We've just been presented with the 12 results of the analysis yesterday afternoon. I think we'll ,, ~ , ^) x 13 develop some questions on that analysis, so we really have 14 not -- I can' t say, give you a yes, a direct yes answer 15 -right now. We have not had time to really f ully review 16 tha t . 17 MR. REED: I might make a comment that I made 18 af ter the trip out to Palo Verde. In looking at their 19 drawings some months ago, I was surprised at the number of 20 closed valves that they have in their systems , the ir 21 safeguard systems, where it appeared to me that many of 22 these valves could be reviewed from being in the open 23 position. Now you're talking about leaving a couple of 24 things open. That's in the right trend direction, but I (3
*) .. 25 think you have to reflect on with respect to this vent ACE-FEDERAL REPORTERS, INC.
3 2-347 3700 Nationwide Coverage 800-336-6646
9250 13 15 200 ,~ (_1/ DAVbw 1 system, because you see again lots of closed valves. They 2 don't'all have to be closed, I'm sure. And those certainly 3 are not passive. You can have all the electrical supplies 4 you want, but stems, galling, and these kind of things, the 5 failure to move still exists. 6 MR. HOUSTON : For example, the valve to the 7 quench tank could be open. 8 Is that the kind of thing you mean? 9 MR. REED: There are two solenoid valves on one 10 ~ vent line. They're both in a closed position. There's 11 backup valves that are closed also. The third valve down 12 the line. n k-)- 13 I saw this throughout the safeguard systems, and 14 I have not seens that number of valves in the closed 15 position. 16 MR. EBERSOLE: Also, there's a matter I'd like to 17 rein force . This business of having an air sysem as a power l 18 supply source to a valve function. These interesting 19 experiences with the part pressure operation of these thing s 20 need to be taken into more engineering consideration than i 21 j ust how to fix the GE scram system. It appears that j 22 pneumatically operated valves under part pressure conditions 23 appear to stick in the medium and don' t do anything they're 24 supposed to. It either requires a sharp cut-off to zero n
~) 25 pressure or another trip, which recognizes the gradual ACE-FEDERAL PJPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6646
9250 13L16 201
- (:x,)DAVbwL
-1 decrease of air pressure. '
^
2 So that might be something that is pertinent to 3 'the system. We ' re ' not s ure . . ; 4 ., Any'further. questions? ; 5 (No response.) 6 MR.. REED: Just a couple of quickies. 7 I didn' t realize you didn' t have backup manual 8 valve on the atmospheric dump. This seems rather strange. 9 My history with atmospheric dumpe , their failure to function
- 1. 0 properly is quite normal, and I'm just surprised.
11 Did you have any failures in all your startup 12 tests on the atmospheric dumps to close or open?
)( b 13 MR. .KAYNES: We did have some problems with the 14 a tmospheric dump valve. It was before my time,'so I can't 1
15 address them specifically, but if you're interested, we can 16 do tha t. 17 MR. REED: That's not surprising. I'm surprised 18 .that you would accept the design from the vendor that did 19 not have, backup manual valves. 20 One other point. I was worried about your safety 21 valves and their potential for leakage. Now you've run Unit 22 1. quite a while, you should have accumulated a hydrogen 23 block under the valves. Do you have safety valve leakage 24 : ye t? . , .( :25 MR. HAYNES: No. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nanonwide Coverage 800-336 6646
202 9250 13 17 d, , DAVbw 1 MR. REED: You have no evidence.of safety valve 2 leakage? 3 MR. HAYNES: No, we do not have any. 4 MR. REED: Not yet. Thank you. 5 MR. EBERSOLE: Are you goitig to say you will? 6 MR. REED: I stil) . predict they will. 7 MR. EBERSOLE: Well, let's come back at 2:15, 8 please. 9 (Whereupon, at 1: 15 p .m. , the hearing was recesed 10 'to reconvene at 2:15 p .m. , this same day.) 11
.- 12 r~. .U 13 14 15 16 17 18 19 20 21 22 23 ,
l 24 25 ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(H5646
19250 14 01 203 DAVbur' -1 AFTERNOON SESS.)N (2:15 p.m.) 2 MR. . EBERSOLE: ,I request we reassemble and get 3 this on the - road.
'4 We are 45 minutes behind. I can' t say that we 5 have wasted the time, but we are 45 minutes behind. If we
'6 .could pick up any in the remainder, we would appreciate your
- 7 doing so.
8 I want to get going immediately with Mr. Waldrep 9 on the Of f site Power Reliability and the October 3rd and 7th , 10 events. 11 (Slide.) 12 MR. WALDREP: I would like to begin the topic of 13 Offsite Power Reliability by discussing the two recent trips 14 .tha t we had on October 3rd and October 7 th. 15 Just briefly, on October 3rd the plant .was at CI: 16 percent power and the station loads were aligned to the 17 offsite power source. At the time , there was no testing
.18 occurring within the unit, and we experienced a loss of-19 power due to the opening of the two 13.8 KV breakers
-20 supplying offsite power to the nonclass loads.
21 This is a time line of the event on October 3rd. 22 (Slide.) , l 23 As you can see, we lost power, as I discussed 1 24 earlier. The loss of power caused a loss of power to the i
^pd 25 reactor coolant pumps, which caused the reactor to trip.
\
l
\
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 33(Hl646
~ l
# 4
'9250 14 02 204 :
g ~ We did have diesel start and loadup.on the ESF O DAVbur 'l ; l
.2 buses, as they are designed to do. We had three main' steam'
- 3 safety . valves open to control steam generator pressure. - The 4 operator then used ADVs.to control pressure.-
5 MR. EBERSOLE:- I understood that you carried the
'6 safety limits .on the offsite power as the normal mode of ,
7 operation? r 8 MR. WALDREP: That is correct. : L9 MR. EB ERSO LE: You were not aligned that way in 10 this particular case?. 11 MR. WALDREP: We were, the.offsite power of both 12 the nonclass.and the class loads.- There fore , when the two 13 feeder - break'ers_ opened, we lost all the class and nonclass 14 load's until the diesels picked up the- class . buses .-
. 15 MR. EBERSOLE: Why did these 13,8 KV -- well, 16 . they lost the main coolant' pumps , didn' t they?
That is why-17 the. reac tor tripped? A
- 18 MR. WALDREP: That.is correct.
19 Like~I said, the operator used ADVs,-the main > 20 steam safety valve receded, we regained offsite power and : 21 then restarted the reactor coolant pumps. 22 ' MR. . EBERSOLE: You have how much bypass? 23 MR. WALDREP: On the steam bypass control system? 24 -That is 65.
- ~25 MR. EBERSOLE: Why didn't it suffice?
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coversee 800 336 4646
a l l 9250Ll4 03- 205 'I DAVbur 1 MR.-WALDREP: The steam bypass control system is 2 a control grade system. It is.not safety related. When we 3 _ lost it in the nonclass loads, . we lost part of the control 4 power to the steam bypass control system.
-5 (Slide.)
6 MR. EBERSOLE: I am trying to get in my head now 7 the merit of having a bypass that doesn't work and being on 8 coolant pumps that were not on the unit. You lost offsite 9 power and you lost the. coolant pumps as a result? 10 MR. WALDREP: That is correct. 11 MR. EB ERSOLE: Had you had the coolant pumps on
'12 the generator output, would you have tripped? You have unit-( ), 13 station service requested to do that, don't you? Had you 14 1 been on them, would you have been able to continue it online i i
15 with the~ bypass? 16 MR. WALDREP: Correct. However, we would have 17 lost: our two of fsite power supplies 'which ' are required with 18 tech specs.- So we had to take the appropriate action 119 because of that. 20 MR. ADNI: I want to make one point clear. We 21 lost of fsite power. We lost our 13.8 buses, and we lost the 22 circ system. We lost the condensing. That is why we really 23 didn' t have these other steam bypasses. 24 MR. EBERSOLE: Again, the circ system, had it
.. .25 been on the main generator output, you wouldn't have lost ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nanonwide Coverage 800 33H646
E
'9250114-04. 206 9"'u ' b DAVbur 1 it, right?
2 MR.- ADNI: That is right. 3 MR. EBERSOLE: So it was just due to the l 4 peculiarity of the lineup that you got into this? 5 MR. WALDREP: That is correct, and that lineup was 'done in support of some additional testing that was
'6
-7 going to be done.
l, 8 MR. EBERSOLE: It was not a standard lineup, was 9 it? . 10 MR. WALDREP: Normal plant operation will be on
-11 the generator, that is correct. This lineup was .done to 12 support some testing that was going to be done later on.
i s 13 MR. EBERSOLE: I have got it. ,
- . 14 MR. WALDREP
- Our evaluation of this event 15 determined the root cause to be a spurious signal from the 16 plant multiplexer system which caused the inadvertent 17' opening of these two 13.8 KV of fsite feeder breakers.
18 Just briefly, the multiplexer system just allows , 19 remote operation of these breakers in the switchyard from 20 the control room. The control room initiates the action to 21 Lelose, open, or trip the breakers, and the multiplexer i ,. 22 system is just a computer-based system using fiber optics i' -23 which trausmits the signals from the control room to the 24 breakers to actuate them. t . 25 MR. WYLIE: Is that just for the 13. 8 breakers , k ACE-FEDERAL REPORTERS, INC. l 202-347-3700 - Nanonwide Coverage 800-336 4646 l
9250 14 05 207 DAVbur 1 or is that all for all switches? 2 MR. WALDREP: Just for those. 3 As soon as I am done, we will have a presentation 4 on the offsite power systems and exactly what it controls, 5 but it is only for the 13.8 KV breakers. It is only for 6 specific breakers. 7 So at this point the reactor shut down and we are 8 in the phase where we are testing the multiplexer system, we 9 are doing some troubleshooting, looking for the problems and 10 the reason that is causing these spurious signals to be 11 generated.
~ 12 We had the plant shutdown. We had the shutdown I )
13 groups withdrawn, and at that point we got another spurious 14 signal, very similar to the one that happened on October j 15 3rd, where those same two main feeder breakers opened. This 16 again causes the same sequence of loss of power to occur to 17 the unit. 18 Our corrective measures for the event was to 19 complete the hardwiring modification for control of the 20 breakers from the control room, such that we were taking 21 PMOCs outside of the system. . 22 MR. EBERSOLE: You sound like you just shook your 23 heads and fixed it. Did you find the root dif ficulty in th9
,_ 24 multiplexer design? '~'
i 25 MR. WALDREP: We did not find the exact failure ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
9250 14 06: 208 ( )DAVbur 1 in the multiplexer design. 2 MR. EB ERSOLE: You just fixed it in the 3 hardwiring. 4 Now, do you intend to go and find out why the 5 multiplexer didn't us-k? , 6 MR. WALDREP: We are still having a 7 troubleshooting phase looking at it. That is why we are in 8 here. 9 MR. EBERSOLE: Do your multiplexers have any 10 critical applications anywhere? 11 MR. WALDREP: The only other operation of the
-12 multiplexers is in the cooling tower fans for the circ water p)'
s 13 system. 14 MR. EB ERSOLE : And that is not a safety function? 15 MR. WALDREP: No. 16 MR. HAYNES : That is not the only other 17 application of the multiplexer system. The multiplexer is 18 used in a number of other applications, but none of them in 19 critical canponents. 20 MR. WYLIE : Have you had trouble with those? 21 MR. HAYNES: No, we have not recently. However, 22 in the distant past we did, when -the multiplex system was 23 first being started up. 24 MR. WALDREP: Associated with this hardwire 25 modification, we performed a surveillance test to verify ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 4 646
m
'9250 14 07 209 5 DAVbur 1 that we could adequately swap over from various sources of 2 offsite power to ensure reliability.
3 With that, I will turn it over to John Barrow to 4 discuss further our offsite power distribution system. 5 MR. EB ERSOLE: Thank you. 6 -(Slide.) 7 MR. BARROW: I am John Barrow, the Electrical 8 Engineering Supervisor for APS in the Design Group. 9 I am here to talk in light of the LOPS recently 10 discussed and the questions raised about GEC-17. My 11 presentation is to reflect on the design of the 12 onsite/of fsite interface. 13l If you recall from the previous hearings on Unit 14 1, we talked pretty much about of fsite power and we really 15 didn' t address the in ter face . 11 6 I also want to talk a little bit about how we
~ 17 handled the interface to avoid these kind of LOP problems.
18 First of all, the switchyard right here, 19 represented by these . two buses, is a breaker and a half 20 scheme; that is to say, _ for every two elements in the yard 21 there are three breakers, one between the elements and one 22 between each element and the bus . 23 There's advantages to this. Specifically, with 24 the breaker and a half scheme if you have a bus fault on ('- 25 either bus , all the breakers connecting to that bus can be ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33Mi646
H i e 9250.14-08 210
)DAVbur .1 tripped and you can maintain of.the operation of the plant 2 and the grid with one bus completely missing.
3 Also, in a breaker and a half scheme you have two-4 breakers between each transmission element; that is, the
~
5 line comes in, goes through' a breaker, there is another li ' breaker to another line coming in. You can lose one breaker 7 and you will only lose one line instead of two. 8 Also, you have a breaker connection between each 9 generator 'and startup element and the nearest bus to 'it, 10 which means'that between each there is'a startup element, 111 there is a generator -- or, rather, I have it backwards -- 12 there.is a generator element, then there is a startup f~) (_f 13 element, a generator and startup and a generator and 14 startup.
, 15 You can lose any one of these breakers, and.you 16 .still do not lose both elements. You will still lose just 17 one element that is connected.
18 Following that, following these connections-down 19 to the startup transformers, you will see that you have got 20 three startup transformers. Each of them has two secondary 21 windings. .Each secondary winding is sized to be able to 22 . operate the startup loads of one-half'of the unit's 23 nonessential load groups; that is, the normal operating 24 loads for startup, plus each of these can operate one ESF
/ 25 load group of a unit. That is to say, each unit has two ESF l l
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
9250 14 09 211 <~ ! ( ,x)DAVbur .1 load groups. So each unit can operate two ESF load groups. j 1 2 Now, the startup transformers feed the six buses, 3 which are' called the startup buses. Each one of them is fed 4 through normally opened and normally closed breakers. 5 As you can see, we have kind of a complex 6 scheme. If you will notice for Unit 2, we come off of the 7 Startup Transformer No. 1 and the Startup Transformer No. 3 8 through normally closed breakers going to Buses S0-5 and-9 S0-6 and then through the 13.8 KV lines to intermediate 10 Buses So-4 and S0-3. These two buses, S0-3 and -4, are the 11 buses which have the Class 1 load groups connected to them.
-12 'In addition, we have these normally open
(~)
- y 13 breakers connecting them to S0-1 and S0-2. Those are the s
14 breakers that allow us to feed startup power through these 15 startup buses through .these intermediate buses to the plant 16 to start up the plant loads before we synchronize them. 17 It is the condition we were in when we had these 18 LOPS -- was when we had the condition of supplying startup 19 loads to these buses. . 2-0 Now, as Gary said, these breakers here were all 21 controlled by the multiplex system. The switchyard breakers 22 in this position, 500 KV breakers, were all hardwired. 23 We chose the multiplex system here to allow for 24 simplicity in the duct system and wiring system, and the () 25 multiplex system is a dual computer-operated system that ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
9250 14 10 212
\,)DAVbur' 1 talks to each other. That is why we are having so much 2 trouble figuring out why we had those LOP trips because 3 theoretically they shouldn' t have happened. -
4 What we have done, though, is we have hardwired 5 them, which is to say through multi-conductor cables we have 6 run back from these buses to the control room, connected. 7 them right hardwired to the control switches so that now
'8 there is no computer interface between the operators and 9 buses.
10 MR. WYLIE: Are you going to keep it that way? 11 MR. BARROW: Yes. In fact, we have got a design 12 change in to make Unit 3 that way, also.
'13 I want to address the 13.8 KV lines. The feed 14 between these buses and these intermediate ' buses are 13. 8 KV 15 overhead lines which have f ull lightning protection.
16 The reason we chose to make them overhead rather 17 than underground was because of the stage construction 18 interface. We wanted to make sure they were protected 19 against construction outages, heavy equipment damage, and 20 that kind of thing. 21 The whole operation of this system means the 22 intent of GEC-17, in that we do have offsite sources of 23 power. 24 We thought that the multiplexer indeed would meet 25 that, and we are quite surprised to find that it didn't. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
9250 14 11 213 7 (_)DAVbur 1 That is being corrected. 2 Are there any questions? 3 MR. WYLIE: Your 500 KV breakers , what are these? 4 SF-67 5 MR. BARROW: They are SF-6.
-6 MR. WYLIE: Live tank breakers?
7 MR. BARRON: They are dead tank breakers. 8 They are Westinghouse breakers. The upper 9 structures are live, 1110 MR. WYLIE: Okay, live tank breakers, yes. The 11 cts are separate? Oil field cts?
.12 MR. BARROW: Yes.
(#3 S/ 13 MR. EB ERSOLE: Let me just make a shot in the
- 14. dark hero.
15, Your cooling water pumps, are they critical in 1 16 this context, in this plant? 17 MR. ADNI: No.
- 18 MR. EBERSOLE: You don' t need them post-accident 19 or _ pre-acciden t?
20 MR. ADNI : _ You need them to run the plant but not 21 for an accident scenario. 22 MR. EBERSOLE: You need them just to reject after 23 heat? 24 MR. ADNI: Our component cooling systems just bN/ 25 - supply cooling to the reactor coolant pump motors, those ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
l 9250 14 12 214 j
~ / T' l V DAVbur 1 types of components.
2 But as far as rejecting heat out of the shutdown 3 cooling heat exchanger, we have another system for that , 4 called the essential cooling water system. 5 MR. EB ERSOLE: And it is in a-closed loop? 6 MR. ADNI: It is a closed loop system, but it is 7 not tied to where you have a segregated valve or 8 essential / nonessential loop. 9 MR. EB ERSOLE: But it is an interposed loop 10 comparable to what is called a component loop in other 11 designs? 22 MR. ADNI : That is correct. 13 MR. EB ERSOLE: That loop, whatever you_ call it, 14 the service water and envir'onmental cooling and so forth, 15 that is normally carried by of f site power? 16 MR. BARROW: That is carried by S0-3 and S0-4. 17 MR. EBERSOLE: If you had a shutdown and there is 18 no transfer of power necessary or anything associated with 19 supporting the safety f unctions of the plant on my unit, 20 which is guaranteed to go dead? 21 MR. BARRON: That is right. There was an 22 alternative when we designed the plant of either having the 23 offsite sources immediately available or having them there, 24 and we chose to have them there.
<m -) 25 MR. EB ERSOLE: I thank you. I am glad to hear ACE-FEDERAL REPORTERS, INC.
M-347-3700 Nationwide Coverase 800 336 6646
19250'14 13- 215 i> DAVbur 1 that. 2 Okay, thank you. 3 Any other questions? 14 (No response . ) 5 MR. EBERSOLE: If not, thank you. 6 I believe the next -- go ahead. 7 MR. QUAN: Exc use me , - Je sse . Is the staf f going 8 to say anything? Item D-3? 9 MR. EBERSOLE: Next we have Mr. Houston. 10 MR.-HOUSTON: This will be very brief. 11 I would just like to add that the members of the 12 staff did visit the site for a couple of days about a week
'U 13 af ter the October 3rd event. They have reviewed the 14 modificacions ~made by Arizona Public Service with respect to 15 the hardwiring of the main breakers, 16 They find that thia enhances the offsite power 17 capability and particularly lowers the chance of losing 18 simultaneously preferred AC power sources for both emergency 19 buses.
20 So the staff is in agreement with the corrective 21 measures the applicant has taken. 22 MR. EBERSOLE: So everybody is happy with the 23 hardwiring. 24 Any gyestions? 25 (No response.) ACE-FEDERAL REPORTERS, INC. ll02-347-3700 Nationwide Coverage 800-3HWi646
19250 14 14 216
- - L/ DAVbur 1 MR. EBERSOLE
- The next presenter is A. Krainik, o
I i 2 I-believe. ' 3 MS. KRAINI_. My name is Angie Krainik. I am a !- 4 mechanical design engineer for Arizona Public Service. 1 5 (Slide . ) 6 The purpose of my presentation today is to review 7 the reliability of the aux feedwater system, especially in l'
- 8 light of the Davis-Besse event June 9th, which is the loss l'
I 9 .of all feedwater. . i
- l. 10 We feel that due to operational and design 1
11 dif ferences between Palo Verde and Davis-Besse that the 12 events that occurred there do not challenge or affect the
~
13 operability or reliability of the Palo Verde system. 14 To support this information on this topic, I
- 15 would like to first talk about the design of Palo Verde as 16 well as review the comparison between Davis-Besse and the 17 Palo Verde components with special regard to the manual 18 initiation of their aux feedwater systems and, finally, then 19 go through the Davis-Besse event analysis and show how the 20 same sequence of events would not lead to the same overall 21 event that happened at Davis-Besse.
22 (Slide . ) 23 First, then is an overview of the Palo Verde aux 24 feedwater design. A 25 We have three 100 percent capacity, independent ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 33(H5646
9250-14 15 217
,3 i
~( / DAVbur m 1 pumps that-all take suction from independent lines. in the 2 condensate storage tank. Any of the three can feed either
-3 or both steam generators.
'4 In addition, each unit at Palo Verde has its own 5 separate independent aux feedwater system. The two 6 emergency tripping pumps are able to be actuated off of it, 7 the emergency feeawater actuation signal, and provide o.f 8 course water to either or both steam generators.
9 In comparison, then, with the Palo Verde 10 design -- 11 (Slide . ) 12 -- I would like to bring up the salient points. MR. REED: 3 -- -- Just to clarify, back to your drawing 14 a minute, as I recall, you have one steam driven pump, one 15 electric driven pump which is safety grade, and one other 16 pump, a third pump, which is not in a secure area. It is 17 not. safety grade , and it is used primarily for startup type-18 activities? 19 MS. - KRAINIK: That is what it is designed for. 20 o It is also a'le, though, to nrovide water, and we take 21 credit for it. It can provide water to the steam 22 generators during a hot shutdown standby and during startup, 23 also. 24 MR. REED: To try to clarify that with respect to D U 25 accident protection, don't we have to assume that startup ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 336-6646
1-9250-14 16 218
- p.
- C.DAVbur: 1 pump,1so-called, is not a . piece of safety equipment; we have
~
, "i to say. that basically you have one steam driven pump and- one 3 electric driven pump?
-4 The reason I want to make this clear is I am
_ _5 going-back a'nd thinking about the steam generator tube 6 . rupture incident which was . presented earlier, and I think 7 about the decay of pressure and the fact that the steam 8 driven ' pump after a while doesn ' t do a very good job with 9 respect to removing decay heat through the steam generators,
~
10 and now we fall- down to one safety grade electric driven 11 pump for the heat removal path. 12 Is that correct or not?
~
13 MS. KRAINIK: The turbine. driven pump is-just as 14 able to feed the steam generator as is the electric driven d 15 pump. 16 17 18 19
-20 21
. 22 23 24 l 25 ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwule Coverage 2 3364646
. 9250 15 01 219 .7
(,/DAVbur 1 MR. REED: With 300 pounds pressure in the 2 primary? 3 MS. KRAINIK: Yes. That is what I was saying. 4 We didn' t have one electric driven decay heat removal and 5 one pump. 6 MR. QUAN: I am not quite sure where your 7 question is leading, but as far as meeting the requirements 8 for an aux feedwater system, the reliability requirements, 9 that third train, we are able to operate it, providing power 10 for a 1-E source to the pump and the associated valves so 11 that on a loss of of f site power we are able to use that 12 pump. sm I) s 13 All the controls that Angie was going to provide 14 in her presenation, all the controls for the appropriate 15 valves and the pump are the control room. So the control 16 room operator -- 17 MR. REED: What I am leading to is this. I know 18 you meet the requirements with one electric driven pump and 19 one steam driven pump. You meet the requirements. 20 But in this particular case, with the Combustion 21 Engineering system, all the decay heat must go through the 22 path of the steam generators to the aux feed system and 23 hence to the ultimate sink. You don' t have any option of 24 bleed and feed or another option of heat removal. You have (x'_N) 25 this path and this path only. ACE-FEDERAL REPORTERS, INC. 3)2 347-3700 Nationwide Coverage 800 336-6646
i 9250 15 02 220
/~T
(_)DAVburl 1 And so I know you meet the requirements, but I 2 have a longstanding argument going on PWR decay heat 3 removal, and I wanted to point out for the rest of the 4 subcommittee's benefit that when you get into this steam 5 gelerator tube rupture incident and your pressures are going 6 down, your steam availability is going down, your steam 7 capacity is going down, and you ultimately get down 8 somewhere before you get out of this to one electric driven 9 safety grade pump. 10 Now, the other one may very well be operable -- 11 it is not in a secure area -- to meet the requirements. 12 MR. QUAN: I think the third train is in a-secure O)
's 13 area. It is in a nonseismic building, the turbine 14 building.
15 I am not quite sure what you moan. 16 MR. REED: It is not in a vital area. It is not 17 locked up. There is access to its valves. There is access 18 to the pump. 19 MR. ADNI: It is not in a vital building. 20 MR. EB ERSOLE: What are the elements of design 21 and function of this nonsafety grade pump which differ in 22 the context of it being safety or nonsafety? Are they very 23 few? 24 MR. QUAN: It is not seismic. () 25 MR. HAYNES: We will get to that specifically in ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(H5646
9250 15 03 221
' DAVbur 1 the presentation.
,2 MR. EBERSOLE: What is it that keeps it from 3 being safety grade?
4 MS. KRAINIK: Let me go ahead and go through. s 5 MR. WARD: Will you answer the question about how 6 f ar down pressure can go for the turbine driven pump for it 7 still to have its full capacity? , s
~ '
8' MR. EBERSOLE: How far does the pressure in it 9 decline? 10 MR. WARD: If you want to answer that later, I 11 will wait. 12 MR. EBERSOLE: How far can the turbine driven
.) 13 pump pressure go down before the turbine driven pump ceases 14' to function?
15 MS. KRAINIK: We will go ahead and get t.ha t 16 in fo rma ti,on . 17 MR. EBERSOLE: Be fore I\ fo'rget it , I was at 18 Beaver Valley the other day. I noted they had run-out flow 19 protection on their pumps. 20 r Do you have that? 21 MS. KRAINIK: We have run the pumps during a test
, 'a 22, dt run-out flow.
23 MR. EBERSOLE: Can they depend on that flow? . 24 MS. KRAINIK: Yes, they can. (- ' 25 MR. EBERSOLE: You didn' t have to put arpses in? (_/ ACE-FEDERAL REPORTERS, INC. E-347-3700 Nationwide Coverage 800 336 6646
.9250 15'04' 222
. DAVbur! 'l' = MS. ' KRAINIK: No.
~
2 Before we go back to the comparison between 3 Davis-Besse and Palo. Verde, let me go through some
~
4 significant points. ~ 5 The first point is the diversity that Palo Verde
-6 employs ~ in theirLemergency feedwater pumps, one turbine
- 7 driven, one electric driven, both being 100 percent capacity 8 .and able to hold both steam generators.-
9 . Davis-Besse, on the other-hand, had no diversity , 10 .in their emergency . feedwater pumps. They were both stemn 11 turbine driven. 12 Another point then is that the condensate storage A ' tank, the place where we receive _the- water. that goes in the ( /. 13 14 steam generator, goes -through three independent lines to all o 15 of ~our feedwater pumps . 4: 1
-16 In comparison then-to Davis-Besse, all.three of 17 their pumps at one point go through' a common suction line.
.18 An event of importance, then, is the third pump-19 in both of the designs.
I 20 At Palo Verde, our third pump is 100 percent 21 capacity, has diesel generator backup ' power supply. to it. 22 It is fully- operable - from the . control room, and also the
'23 discharge valves also have backup air supply to them.
24 Contrast that with Davis-Besse. Their pump is z (} 25 only.a partial capacity pump. It requires local manual ACE FEDERAL REPORTERS, INC. ; 202-347 3700 Nationwide Coverage 800-336 6646 l
a q
, , s
,. 4.
A %' 9250;15 05 I 223 DAVbur 1 initiation for-flow.an'd also has no backup capability for
/: y 2 their ;. discharge valvss. .
'3 l Another. point with . regard . to the turbine itself 4 is - that the- length. of piping from- the steam generators to -
1 5 the turbine-itself is a significant point with regard to the 6 condensation'into the turbine which can cause overspeed. 7 At s Palo Verde the length of 'line has been
'8 minimized -so that we are not as susceptible to it. l 1
9 9 o H,owever, in Davis-Besse the lines are much longer J10 and contnibuted to the overspeed trips that we did see. 11 Another point then is that the steam generator at 12 Palo Verde is a Combustion Engineerinq_U-tube design in l
,l s '13 comparison with- the B&W once-through steam generator, which 14 -Davis-Besse employs, which.is more susceptible.to the dry-up 15 -condition they experienced.
16 Another point, then, which is a significant point 17 'in'the event, is the point at which the manual initiation to L18 -their steam feedwater rupture control system, which is 19 parallel to the Palo Verde aux feedwater actuation system.
.20 (Slide . )
21 What I would like to go ahead and do is show a 22 schematic 'of these drawings and do a comparison of them. 23 The schemes take both steam, generator pressure 24 and level and work through the steam generator rupture logic
~k -
25 to assure that the ruptured steam generator is not fed and l ACE-FEDERAL -REPORTERS, INC. 202-347-3700 : Nationwide Coverage 800-33Mi646
. _ . ._.--.~---.-.m,__-.._ - - - . - . - . -- . , , .
9250115.06 224 . ,en .
- 3. ) DAVbur 1 then will initiate aux feedwater automatically.
2 The point to note is the point at which the 3 manual initiation enters both schemes. 4 At. Davis-Besse~the operator must input the 5 parameters, steam generator level or pressure that must be 6 . input, and work through the logic before he gets an 7 : actuation signal. 8 However, at Palo Verde the design is dif ferent. 9 Our operators , because the manual initiation point is at the 10 end of the logic , is able to directly input . aux feedwater 11 actuation and therefore has a positive means of getting 12 out. They are not required to make a parameter decision, as O-ls ,)l 13 .the Davis-Besse operator did in the event. 14 At1this point I would like to work through the 15 event itself and how it happened. 11 6 MR. EBERSOLE: May_ I ask this question? 17 Picking Davis-Besse, of course, you picked 1.8 probably the worst plant you could find on this comparative 19 basis. I mean, yoa know, they are all turbines. They were 20 really a setup for the accident that happened.
-21 What if we pick instead a good Westinghouse PWR 22 .that has got two qualified electrics and one qualified aux 23 feed and still has the prerogative of blowdown?
24 That is the real comparative basis. 25 I don't know, what are the reliabilities that are ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
m.:-
'9250 15 07 225 k )'DAVbur 1 being clatmed that are characteristic of the conservative
~
2 Westinghouse design with two electrics and one feed, back of-3 _ which of course ' is the prerogative to blowdown? Do you have L4 ; any_ feel for the average that we claim for these good
~5 systems, not the worst system?
6 MS..KRAINIK: No, I' don't. 7 MR. EBERSOLE: Does the staf f have any feel for 8 this? What are the current PRA reliabilities springing from 9 the conventional excellent systems? 10 MR. GORMIEL: My name is Jerry Gorm'iel. I am
'll with the Auxiliary Systems Branch with the staff.
12 From our review of aux feed reliability studies, i
)t 13 we find the best Westinghouse plants are on the order
'14 _ approaching 10 to the minus 5, using the data base that we 15 have in our selected studies in the past.
16 MR. EBERSOLE: You say that is the sort of an 17 - average for the full array of the steam power and the two 18 electrics?
- 19 MR. GORMIEL: That is correct.
20 MR. EB ERSOLE: And on top of that, they have the 21 prerogative to. blowdown? 22 MR. GORMIEL: Most of your newer Westinghouse 23 plants , right . 24 MR. EBERSOLE: So you are 10 to the minus 5, and ( -25 you are claiming 10 to the minus 4, or a factor of 10 less, ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverpse 800-3?/ 6646
}
r 9200 15 08 - 226
.(3 '(,jDAVbur 1 which I guess is reflected in your .nonsafety grade electric 2 pwnp?
3 I really don' t know- what the basis is, but 4 presumably that is it.. 5 .Thank you. 6 MS. KRAINIK: At this point I would like to work 7 through the event and show the effect of each event as it
-8 could be on the Palo Verde.
9 Their initiating event was the main feedwater
'10 pump trip. The possibility of that is possible at Palo 11 Verde. However, we have successfully run main feedwater 12 pump trips without bringing the prbmary side down.
A_/ (3 13 MR. EB ERSOLE: You anticipate, I guess, a 14 substantial reduction in main feedwater trips with this 15 setback system you have, do you not? You know, the partial 16 rod drops? That will enhance your chancu to stay online 17 with the main feedwater pumps? 18 MS. KRAINIK: Yes. 19 The next event, then, due to their main feedwater 20 pump trip, they had a reactor and a turbine trip at Palo 21 Verde. On a main feedwater pump trip, we also have control 22 systems that are designed to prevent a reactor trip at Palo 23 Verde. 24 Their next event was the steam and feedwater 25 rupture control system. It blows their main steam isolation l ACE-FEDERAL REPORTERS, INC. l 202 347 3700 Nationwide Coverage 800-336-6646 f I
l ~9250 15 09 227 t) DAVbur. 1 valve' and isolates steam to the main feed pump turbines. At 2 Palo ' Verde, we did have an event where a turbine trip did 3 eventually result in c main steam isolation valve. However, 4 modificatiora are being implemented to prevent recurrence. 5 At the next po in t -- 6 MR. REED: Take a minute to straighten me out. I 7- assumed the steam driven pump, however, takes its steam from 8 upstream of the main steam isolation valve? 9 MS. KRAINIK: That is right, the aux feedwater 10 pumps do.
.11 MR. REED: So what difference does it make if the-12 ' main steam isolation valve goes closed?'DAVbur 1 is?
2 MS. KRAINIK: There is a length of cold piping. 3 MR. REED: How long is that length of cold F 4 piping? 5 MS. KRAINIK: Approximately 86 feet. 6 MR. REED: Wow. Why can't you go up to the 7 throttle valve or the shutof f valve at the turbine, or don' t 8 you have enough interrupting valves right up tight to the 9 turbine? . , 10 MS. KRAINIK: It is to guard against the effects 11 of a high energy line break. 12 MR. REED: So in a compromise situation with (~)#
'- 13 respect to high energy line going through someplace, then 14 really you are not much dif ferent from Davis-Besse on that 15 steam feed to the turbine?
16 MR. EBERSOLE: Are there drains in that that keep 17 it hot and dry? 18 MS. KRAINIK: There are drains that keep it 19 drained. i 20 MR. EBERSOLE: On the back side of the valves? 21 MS. KRAINIK: That is correct, the downstream-22 side of the valve. l 23 MR. EBERSOLE: So you keep it hot and dry? 24 MS. KRAINIK: It is kept dry because there are O-25 drain valves as well as steam traps in there. ACE. FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverase 800-336 66 4
'9250 1.5>12 230
)DAVbur 1 MR. EBERSOLE : I see.
2 MS. KRAINIK: The next point, then, is that in 3 the event at Davis-Besse another operator realized that the 4 steam f feedwater rupture control system had been ' improperly 5 . inputted with low pressure, went back and reset it with low 6 level. 7 However, there are steam generator isolation 8 valves, one for each steam generator, and they did not 9 reopen. In fact, the valves were not specified to open 10 against such conditions. 11 The design at Palo Verde is sufficiently 12 dif ferent that one valve will not isolate the steam 13 generator from aux feedwater as well as the isolation 14 valves. We have one isolation valve and one reg valve on 15 each line. Those valves are designed, and have be.en 16 successfully tested, to open and operate against such 17 conditions. 18 MR. EBERSOLE: I am sure I must have asked this 19 before, but let me just do it again, anyway. 20 The turbine pwmp is in no way dependent on the 21 electrical function of anything, including environmental 22 controls, and so forth, driven by AC power, is this correct? 23 MS. KRAINIK: That is correct. 24 MR. EBERSOLE: The line rupture problem, which is A) m 25 usually monitored by temperature measurements, is not ACE-FEDERAL REPORTERS, INC. , 202-347 3700 Nationwide Coverage 800-336-6646 1
w
- 9250 15 13 231 g
V DAVbur 1 compromised by loss of ventilation? 2 You know, you measure your temperature in a steam . 3 chamber someplace, and you interpret that as being evidence 4 of line break, then aux feedwater cemperature, steam supply 5 line, loss of feedwater will occur and it will be inferred 6 as a steam line break. 7 You don' t have that vulnerability, or do you? 8 MS. KRAINIK: I don' t believe so.
.9 MR. EBERSOLE: How do you detect a turbine steam 10 feed-line break should one occur? Do you do it by 11 temperature or flow measurement or both?
12 MS. KRAINIK: Do you mean a main steam line q O 13 break? 14 MR. EBERSOLE: No, a feed line to the aux feed 15 pump turbine, in which case you want to blow that isolation 16 valve. 17 You know, you have a line to supply the 18 aux -- steam driven aux feed pump. It is at full pressure 19 up to the isolation valve. That valve is normally closed, 20 isn't it? 21 MS. KRAINIK: That is correct, it is closed. 22 MR. EB ERSOLE : What opens it? 23 MS. KRAINIK: An aux feed actuation signal. 24 MR. EB ERSO LE : What is the energy that works the
~
N/ 25 valve? ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverase 800-3M-6646
l
./
I
-9250-15 14 232 . o-V DAVbur -1 MS. KRAINIK: 'DC power.
2 MR. EBERSOLE: For a line break upstream to that, 3 .there is another valve in the line, isn't there, ahead of 4 tha t? .
.5 MS. KRAINIK: No, there is not.
6 MR. EB ERSOLE: There is only one valve? 7 MS. KRAINIK: The inlet valve that opens on an 8 aux feed actuation signal, that is the first valve in the 9 line. 10 MR. EB ERSOLE : I see , all right. 11 So that valve is normally closed. You open it. 12 Once you open it, you invoke that it will not fail . in a line A
\-) 13 break sense, is that correct?
14 MS. KRAINIK: That it will not fail shut. 15 MR. EBERSOLE: There will not be a line r'ailure 16 coincident with use of the aux feed pump? 17 MS. KRAINIK: That is correct. 18 MR. EB6RSOLE: Thank you. 19 MR. MICHELSON: A clarification on your last 20 item, the failures at the 599 and 608 were not designed to 21 reopen against the high pressure. 22 What was the source of that information? 23 MS. KRAINIK: The source of the information? 24 MR. MICHE LSON: How did you know that? (')N 7
- k. 25 MS, KRAINIK: We got that from the informa tion i
ACE. FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 8t")-33(Hi646 L--
- 9250 15 15 233
;~,, .
L JDAVbur 1 'we have looked at on the Davis-Besse event. 2 MR. MICHELSON: My vague recollection -- and 3 maybe it is vague -- was that it was the improper adju'atnent 4 of the valves, not the fact that the valves didn' t have 5 adequate torque. 6 MR. WARD: That is what I recall, too. 7 MS. KRAINIK: There were torque switch setting 8 problems, also, on the motor-operated cell. 9 MR. VICHELSON: I had never heard, and that is 10 why I was curious, that the motor-operated was inadequate 11 and that if it had been properly adjus?.ed -- 12 MS. KRAINIK: What I was trying to say was the (% (_) 13 valve design itself. 14 MR. MICHELSON: It was my understanding it was
- 15. adequate against cold differential pressure.
16 MS. KRAINIK: We understand that it wasn' t spec'd 17 to open against these conditions. 18 MR. MICHE LSON: Tnat is a little different 19 statement, although I understood it was designed to open. 20 But that is neither here nor there. 21 Are your valves air-operated that have to open to 22 actuate or to supply steam to the turbine? Are they 23 mo to r-opera ted? 24 MS. KRAINIK: Ye s . O
\_/ 25 MR. WARD: DC, you j ust said?
ACE-FEDERAL REPORTERS, INC. g 202-347-3700 Nationwide Coverage 800-3364646
~
9250 15 16 234 n)DAVbur 1 MS. KRAINIK: Correct. 2 M R. MICHELSON: Okay, thank you. 3 (Slide.)
'4 MS. KRAINIK: The next event that happened at. ,
.5 Davis-Besse was a further complication of their event.
-6 Their atmospheric vent valves opened automatically and 7 depressurized steam generators.
8 In contrast, at Palo Verge our atmospheric dump 9 valves are not automatic and are only actuated by operator 10 action. 11 MR. EBERSOLE: .May I just ask? 12 The at:noopheric dump valves, one reason they are A L) 13 there is to prevent the safety valve from starting its own 14 turbine trip , right , to protect against the safety valve 15 lifting and the consequent probable. leakage? 16 MS. KRAINI K: They can be used that way. 17 MR. EBERSOLE: Isn't that required that they be 18 in an autcznatic mode? 19 MS. KRAINIK: We have the steam bypass control 20 system, which is designed in that instance. 21' MR. EBERSOLE: Do you use steam bypass and thus 22 avoid having automatic PORV on the secondaries? 23 MS. KRAINIK: That is correct. 24 (O/ 25 ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
9250 16 01 235
?"jDAV/bc 1 MR. EBERSOLE: You use that as a useful way of V
2 avoiding automatic dump valve action. Is that correct? 3 MS. KRAINIK: It's a steam bypass control valve 4 system. 5 MR. EBERSOLE: Which negates the need for PORV 6 secondary action. 7 MS. KRAINIK: Yes, we just have a dump valve on 8 automatic action. 9 MR. MICHELSON: MSIV isolation then, I guess, you 10 have to open all the safeties, is-that it? 11 MS. KRAINIK:. If you have a main steam isolation 12 valve closure, you can open your atmospheric dump valves. (} 13 MR. MICHELSON: You're not that fast though. On the sudden closure of MSIV's, No 14 operator is that fast. 15 they're going to fly open or blow up the plant, one or the 16 other. 17- MR. ADNI: That's correct. You're going to have 18 to go and open -- 19 MR. MICHELSON: So you face the opening of 20 safeties on an inadveretn MSIV isolation. 21 MR. ADNI: That's correct. 22 MR. EBERSOLE: Not all of them, I guess. Most of 23 them. 24 MR. ADNI: Probably. It depends on the power 25 level. ACE-FEDERAL REPORTERS, INC. , 202-347-3700 Nationwide Coverage 800-336-6646
,9250 16LO2 236 jllDAV/bc 1 MS. KRAINIK:~ The next point then in the 2 Davis-Besse event'was that they chose to use the third train 3' of their startup pump'to supply feedwater to their steam
<4 generators.- This required local manual initiation in order
'5- to get flow to'the generators.
6 At Palo-Verde, however, our design is different.
~
7 The whole' train, our third train, is able to be operated
'8 from the control room and also has backup power from the A 9 diesel generator.
10 .The-last point under consideration is the local 11 manual reset that was required for their turbine-driven 12 pumps. Palo Verde has a similar valve. However, it has a _() 13 motor operator. So the operator from the control room is 14 able to open the valve and the only local manual initiation m 15 that's required by the operator is to reset the trip 16 mechanism itself.
- 17. MR. EBERSOLE: He has to go down there to do o
IP that? 19 MS. KRAINIK: That's correct. 20 MR. MICHELSON: What kind of governor is on it?
- 21. Is this a Terry turbine?
22 MS. KRAINIK: That's correct. 23- MR. MICHELSON: What kind of governor are you 24 using? () 25 MS. KRAINIK: It's a Willard. ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336 6646
9250 16 03 237 DAV/bc 1 MR. MICHELSON: But which one? There are two or 2 three models of that. Is it an all-electronic governor? 3 MS. KRAINIK: A hydraulic governor. 4 MR. MICHELSON: So it's not the electronic 5 governor? o MS. KRAINIK: That's correct. 7 MR. EBERSOLE: Do you contemplate a full span, 8 full load test on the turbine of the aux feedwater pump 9 whereby such things as water hammer will be tested? 10 MS. KRAINIK: We have done tested on our 11 turbine-driven pump. 12 MR. EBERSOLE: Have you tried to invite water ,c3 \ ) 13 hammer to the extent that you don't expect to see it as a 14 result of your test? . 15 MS. KRAINIK: We talked earlier that we have done 16 testing. As a result of that, we have made sure that the 17 drain valves down in the room itself are open to assure 18 there is drainage of any condensate that might be in the 19 line. 20 MR. EBERSOLE: Are they up on time? 21 MS. KRAINIK: They are cracked open. 22 MR. EBERSOLE: Are they closed when you admit 23 steam or do they just stay open? 24 MS. KRAINIK: They stay open. r-) J 25 MR. EBERSOLE: Thank you. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-6646
l .9250 16 04 238 jlhDAV/bc 1 MR. REED: You say you have some open vents on 2 the pipe? Is that what you're doing? 3 , MS. KRAINIK: That's right. 4 (Slide.) 5 The Palo Verde design here is that these, from 6 either steam generator, these are the isolation valves here, 7 -and these should be the turbine itself. The valves tha~t are 8 open are down by the turbine to assure that the lines are 9 sufficiently drained. 10 MR. MICHELSON: You say they're cracked open. 11 How do_you know how far open they are? Is there visible 12 steam coming out? Or, is this a closed drain system, or ( ) x_e 13 what? 14 MR. ADNI: The line is not charged normally with 15 steam. 16 MR. MICHELSON: I thought you heard earlier you 17 had a warming all the way up. 18 MR. ADNI: No, no. 19 MR. MICHELSON: It's cold all the way up so that 20 valve, you say it's cracked open. I'm saying, somebody may 21 not have cracked it wide open and you wouldn't know the 22 difference, would you? 23 MR. ADNI: No. 24 MR. MICHELSON: Until after you started the () 25 turbine. Then you'd know the difference. And you don't ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 336 6646
-9250 16 05 239 -T^1DAV/bc 1 worry about that? Any valve that's supposed to be cracked R.,
2 open, a guy can go down and make it wide open, thinking 3 that's what you meant by " crack open the valve". 4 MR. EBERSOLE: Would these leakoffs show you that 5- you can go wide open instead of cracked? 6 MR. MICHELSON: There's nothing in it until the 7 turbine starts. 8 MR. EBERSOLE: The valves are cracked? 9 MS. KRAINIK: That's correct. Like the steam 10 trip valves.
~
11 MR. EBERSOLE: Okay. 12 MR. ADNI: The AC drainline in the turbine we're
13 talking about.
(T. / MR. MICHELSON: 14 And when that turbine starts up, 15 if it isn't just cracked open but is wide open, you've got a 16 steam blow into the room, if I understand the arrangement 17 correctly. And you don't worry about that? 18 MR. REED: Earlier, I mentioned, in my trip to 19 Palo Verde, I was surprised to find large numbers of closed 20 motor-operated or air-operated valves in the safeguard 21 systems, which I wonder in my own experience why they were 22 not in the open position. 23 I notice here again you look at your auxilliary 24 feedlines here and you have a series of motor-operated ('j) 25 valves, all in the closed position, and closed pumps. Both
~
ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nadonwide Coverase 800 336 6646
9250 16 06 240 jggDAV/bc 1 of your safeguard pumps. I wonder why you do that. What's 2 the reason for that? 3 MS. KRAINIK: The valves themselves have been 4 specified, designed and successfully tested to open against 5 the pressure of the pumphead itself. 6 MR. REED: That's fine and dandy, all those 7 things, but valves have a certain unreliability and that 8 unreliability will be added to the ability.to initiate the 9 system. The valves, you can have the right power supplies, 10 we've said before. You can have all kinds of things. But 11 you can have cockglands and C-stems and broken bore 12 operators, or whatever.
/
) 13 Is it necessary to have all these in closed 14 positions? I've seen many systems where they rely on 15 well-maintained checks. They've got two checks and two 16 motor operators on each one. And the motor operators are 17 closed.
18 MR. QUAN: If I understand though, I can't recall 19 which plant it is that has this problem, but if you only 20 have check valves, don't you run into some problems of steam 21 binding of these pumps? I recall that was a concern, too. 22 MR. REED: Not if the check valves don't leak and 23 your operator is around checking the line temperatures once 24 in a while. All they've got to do is feel the line. It's [) 25 supposed to be an uninsulated line. It doesn't have to be 1 1 1 ACE-FEDERAL REPORTERS, INC. ! 202-347 3700 Nationwide Coverage 80 4 336-6646 I
s - 9250 16 ! 07 241 DAV/bc. 1 insulated. . You've got :all those POS selected operators 2 there.-
- 3 All I'm saying is that right from my first trip
~
4 to Palo Verde, I've b0en concerned about more. valves in the 5 closed ~ position than nead to be, and I don't.know why you do 6 it.
.7 MR. EBERSOLE: -Well, presumably, you do it
~ ~
8 basically to avoid the thesis that you've got pressurized
'9 lines scattered all over the place, which get worse, you
- 10 know. Is that right?
- 11 MS. KRAINIK: That's part of it, yes.
' 12 MR. EBERSOLE: That's an unlikely mechanism of
. 13 failure. I think he's arguing about the fundamental logic H( )
14 of what gives you the most trouble, line breaks or opening 15~ of the valves. 16 May I ask this? What kind of a DC train design 17 do you have? Is.it two trains? 18 MS. KRAINIK: Diesel generators? 19 MR. EBERSOLE: No, no, direct current. Are there 20 two direct current trains? 21 MS. KRAINIK: There's four trains. 22 MR. EBERSOLE: In the matter of feeding the DC 23 necessary for the operation of these working in the valves 24 and working the control functions of the steam 25 turbine-driven pump, are the DC sources independent to the (]) ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverase 800 336 6646
9250 16'08 242
,~ .
( , .s)DAV/bc 1 three trains you show up there? 2 MS. KRAINIK: DC power is supplied only to the 3 turbine-driven train. 4 ?:R. EBERSOLE: That's off of one DC source? 5 MS. KRAINIK: Yes. 6 MR. EBERSOLE: If I lose that DC source, then I 7 lose that pump,. don't I? 8 MS. KRAINIK: There's four DC trains and there's 9 two trains that feed each one. 10 MR. EBERSOLE: Sometimes, you find that you have 11 a dozen DC systems but if you fail two of them, you have a 12 lot of trouble because of tha peculiarity of their
-n 's_f 13 connections.
14 So if I lose the particular DC system that 15 handles, let's say, the turbine-driven-pump, I presume that 16 may well produce a transient. Maybe a turbine trip. I'm 17 not sure, because I don't know what else is connected to 18 that DC source. 19 Then that throws me into a transient that may 20 well have lost the main feed, or whatever. And I need the 21 aux feeds. What then gives me the DC source power to the 22 other two electric pumps? 23 MS. KRAINIK: These are both AC-driven pumps and 24 have backup from the DC generators. () 25 MR. EBERSOLE: But even there the breakers are ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-33Mi646
9250 16'09' 243 DAV/bc .1 -. commanded to function by direct. current to circuit breaker 2 trip supplies. And where do'they come from? 3- R.-QUAN: Our recollection is that the two 4 motor-driven pumps'and the related equipment is all from the SL lE AC' sources.
-6 MR. EBERSOLE: But which one of the sources?-
7l Give me a train designation. 8 MR. QUAN: Off the diesel' backed buses for the 9 safety-related. aux feedwater train. It's the B-diesel. For 10 the third train, it's the A diesel. 11 MR. EBERSOLE: And the corresponding DC system?.
- 12. MR. QUAN: There's no alliance of the DC system.
13 MR. EBERSOLE: How do I close my circuit
/]).
-14 breakers? Do you use AC operated circuit breakers?
15 MR. QUAN: We'll have to find out for you.
-16 MR. EBERSOLE: -Why don't you tell me? It's 17 always a problem when you have a train of equipment which 18 .has a larger number than the DC source system. For 19 instance, you've got three pumps here. How do you divy up 20 the DC source, control sources? Presumably, one pair is 21 going to be on one DC supply.
22 You can always invoke transfer at risk to give 23 each critical load to DC but you'll have trouble with that 24 because of the cascading failure of both DC sources. q(]) 25 So why don't you investigate what the DC ACE-FEDERAL REPORTERS, INC, 202 347 3700 Nationwide Coverage 800 336 4646
7 9250L.16-10 244' hAV/bc 1 rationale or logic.is for functional trains larger in number 2 than a number of DC~ sources?
- 3. MR. QUAN:,'We'll take a.look at that.
< 4 MS. KRAINIK: Any further questions? , 5 MR. EBERSOLE: Any further questions on.this?
'6 -(No response.)
7 _MS. KRAINIK: I'd like.to introduce. Gary Waldrep, 8 who will talk about the operation of the auxilliary main 9- feedwater systems. l 10 MR. EBERSOLE: Thank you, Ms. Krainik. 11 MR. WALDREP: I'want~to talk a little bit about 12 our experience with our main and auxilliary feedwater O 13 systems and how eher responded ehrouoh eh power ascension 14 phase.
^
15 (Slide.) 16 I'd like to first begin by briefly describing how 17 we utilize the system during normal operation. As everybody 18 knows, we use the third auxilliary feedwater pump to L 19 normally start up the plant to approximately 3 percent 20- power. 21 At that time, we. switch over to one of the main 22 feedwater pumps and maintain operation with one main 23 feedwater pump until approximately the 50 percent power 24 area, when we put on the second main feedwater pump. O 25 aust to review, we do have an auxilliary ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-336 6646
1 29250 16'll' J 245 )
~ ,
!( 3,DAV/bc 1 feedwater system in emergency condition which will actuate
~-< ,
I 2- 'on an~AFAS signal. It will automatically actuate'at 25.8; e 3 l percent. wide range steam generator level; at 40 percent, it 4 will reset'to close the valve to secure flow. 5 Just to review the AFAS, it starts both the motor 6' and the turbine-driven auxilliary feedwater pumps and 7 - operates the valves according to whatever generator -needs to - 8~ be failed. 9 (Slide.) cl0 Now I'd like to talk a little bit about our main 11 _feedwater system and the way it performed up to date. 12 During .the initial operation of the system and up
.g)
( 13_ to about 20 percent power, the system, the control systems-14 for the main feedwater system were operated'in the manual 15 mode. Our control system involves two separate feedwater 16 control systems. Each control system controlling a feedpump
-17 and the speed of the feedpump and its associated downcomer 18 and economizer isolation valves.
19 Like I said, the initial power operation up to 20 about 20 percent was done with the systems in manual; as the 21 control systems were checked out and made sure they 22 _ responded accordingly, the initial swapovers in the 15 23 percent and 50 percent swapover areas were also done in the 24 manual mode to ensure the control system provided
) )- 25 appropriate signals prior to putting the systems on 1
ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336 6646
9250 16'12 246 gDAV/bc 'l automatic. 2 All of our significant changes or modifications 3 to the setpoints in those systems were completed by about 4 the 20 percent plateau. And we did our final checks on the 5 major modifications or whatever modifications we had to make 6 by-the 50 percent plateau. 7 From that time period on, we've operated the main 8 feedwater system in the fully automatic mode. We've had no 9 problems and the system really has performed exceptionally 10 well, with no discrepancy identified at the control 11 systems. 12 And all our subsequent startups and power () 13 escalations have been in the automatic mode. I might also 14 add that we operated the plant up to 70 percent power on one 15 main feedwater pump, you know, with three condensate pumps 16 and two heated drain pumps and still had a little more 17 excess speed available on that pump. 18 MR. WARD: Would you repeat, when you had a 19 reactor trip, how far down the decay heat curve do the main 20 feedpumps stay on? 21 MR. WALDREP As far as for steam generator 22 pressure, how long can steam generator pressure go and still 23 maintain feedpump operation? 24 MR. WARD: When does the system switch over to () 25 the auxilliary feed? ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-6646
c: 9250 :16 13 247 qllpAV/bc 1 MR. WALDREP: Maybe Bob could answer that. 2 MR. ADNI: About 3 percent power is really where 3 we start up their main feedpump and bring their main 4 feedpump on line. Three percent power is where we normally 5 run. 6 MR. WARD: I'm saying, going in the other 7 direction, if we have a reactor trip. 8 MR. EBERSOLE: You don't ever have to turn down 9 main feedwater, do you? 10 MR. ADNI: That's right. We've got decay heat, 11 but there's going to be some point where you don't need it 12 or you don't want it. You're going to have decay heat. () 13 MR. WARD: If you're sitting there in hot 14 shutdown, are you sitting there with a main feedpump 15 providing feedwater? 16 MR. WALDREP: Normally, no. In our experience to 17 date, obviously, we don't have a lot of decay heat. So I'll 18 explain in a minute in our reactor trips we've done so far, 19 we have utilized the auxilliary feedwater system after the 20 reactor trip. 21 MR. EBERSOLE: But if you need it, you could have 22 it if you wanted it? 23 MR. WALDREP: Main feedwater? Yes. 24 MR. EBERSOLE: So there's no intrinsic reason () 25 that you can't use it at 100 percent power, is that right? ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
9250 16 14 243 ygDAV/bc 1 MR. WALDREP: That's right. 2 MR. EBERSOLE: You have full low flow 3 competence. , 4 MR. WALDREP: The feedwater control system will l 5 actuate. i 6 MR. EBERSOLE: It tapers down to virtually zero. 7 Good. Let me ask a question. In view of the critical need 8 for aux feedwater at the CE plants, what was the rational ! 9 argument that gave you turbine-driven main feedwater pumps l 10 instead of Beaver Valley, which has two Tandem 4000's 11 driving the main feedwater pumps and, thus, is not dependent 12 on the circulating water system and the other accessory 13 apparatus? 14 MR. WALDREP: I'm not sure I can answer why we ! 15 chose that. 16 MR. EBERSOLE: What are the arguments that led to 17 turbine-driven or electric feedwater pumps? 18 MR. WALDREP: Have we got anybody? 19 MR. EBERSOLE: Are there any arguments at all? 20 Can somebody kind of raise his hand and say he put on 21 turbine-driven? Is it cheaper? 22 MR. HAYNES: I can't address it specifically here 23 but, in general, it's the size of the pump and economics. 24 That size pump, the economics usually drive you to a steam-
) 25 driven pump.
ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33M646
6r -
, , .c.
<\p
/ t>, t 1
.gy .
- 9250'16 15 , %i.
C . 249
'p i 1 AR. EBERSOLE: It 's' my -(inpression that's ' the.
v.DAV/bc: -
\
2 case, but reliability doesn't get into the picture very .
'3' much, does'it? , g i. \ *l -
u4 .MR.sHAYNSS: , p 5, Reliability is considered. And I
; .5 bell' eve,'although perhaps Combustion should-address this, in
.6 :my experience, . tho' gliability of the large turbine-driven ~
j 7 pumps is about as good as the'large motor-driven pumps. s, 2- , 1 \ .Mg. EBERSOLE: It can't be because they hang on ! s 1 9 i the condenser and all~of its associated functions. So
.> n -- ,
\
- 10. ,
there's'a-supportive syscem much more complicated. 4' 11 , MR. HAYNES: I agree it's more complicated, 12 certainly.
-f J' 13 i MR gpERSOLE: I It's just.a back'siew as to how
, r n e, you_got who're you are. We can take that 'p u some other
~
14 15 time. Go ahead.
, 16 MR. WALDREP I'd like to briefly discuss two of JL a ,.u 17 -the'rnactor trips, which I think Jerry mentioned in the ,
\
'I.
i
'T' i4 .
18 beginni{ag this morning, that have been involve'(in the main 19 feedwater' system. The trip that we had on June 14th, which t 20' wa,s our first reactor trip, the reactor was at. about 20 t f 21 percent power and we were swapping over from one main ; 22 feedwater pump to another one. I J 23 When we did that, we got a recire valve open on 24 the feedpump and due to some -- I don't want to say O 25 im, roper, bu,3 de ,.eo ehe see,oines in the ceneret syseems l l ;
\.
ACE. FEDERAL REPORTERS. INC. 202 347 3700 ' Nationwide Coverage 800 336 6646 7/ a,_
m "9250.16 16 < 250 jlDAV/bc 1 for.thefrecire valves and the. condensate pump, and in'that
- 2. particular feedwater pump, this resulted in too much bypass
.3 flow,.which; caused a below section pressure trip of the 4 feedwater pump..
/ ' ;5 The system responded as designed without any
'6 additional problems.
7 The'second-trip that occurrec, on July'the 1st, 8 was at about 50 percent power. It was at.43 percent power. 9' And we were:taking the plant from 40 percent initially to 10 the:50 percent plateau. At that time, we experienced some 11 -load swings in turbine load and some levels of changes in 12 'the. steam generator level.
;(f. 13 So we stopped our power increase and began to try
- 14 . to stabilize,the plant. The operator, we took manual
, 15 control of the feedpump in order to try to maintain steam 16 generator level.
17 This caused a low suction pressure trip of the
.18' feedpump and, of course, tripped it off. And the reactor 19 tripped on high pressure.-
20' Our analysis revealed there was an unusually high
'21 DP across the feedpump suction strainer. We removed the 22 strainers in both the feedpumps, found them to be clogged to - i l
23 about 30 percent. So we replaced the strainers with a 24 little bit coarser strainer and had no problems since with () 25 main feedpump sur: tion. ACE FEDERAL REPORTERS, INC. 202-347 3700 . Nationwide Coversee 800 336 6646
9250 16 17 251 jlhDAV/bc 1 MR. WARD: You just left that stuff through? 2 (Laughter.) 3 MR. WALDREP: The fine strainers of course were 4 in there for the startup phase. 5 (Slide.) 6 For auxilliary feedwater, we've used it for about 7 the 10 odd startups that we've done to date. There have 8 been no problems with the auxilliary feedwater pump used in 9 that capacity, nor in the transfer of auxilliary feedwater 10 to main feedwater. 11 We've also had a couple of different areas where 12 we've used auxilliary feedwater -- one in the second event () 13 that I just explained. We got down on steam generator no. 2 14 to the point where an AFAS 2 did occur. The steam generator 15 levels were low because of the transient I was explaining 16 and, as a result of that, in the loss of the feedpumps we 17 did experience a low level condition in the generator to the 18 AFAS secpoint. AFAS 2 to steam generator no. 2 was 19 initiated. Both the motor and the essential auxilliary 20 feedwater pumps started, all the valves operated, and the 21 system operated as designed. 22 We refilled the generators to the reset set 23 point. At that time, the operator then took control of the s 24 motor-driven pump and secured the turbine-driven pump.
,cx \
Q ,) 25 ACE-FEDERAL REPORTERS, INC. 202-347 3700 Natio:iwide Coverage 800-336-6646 l l
19250 17'01 252
.O\
r# DAVbw' 1 MR. EBERSOLE: In this cavitation condition, is 2 there a tendency for the' turbine-driven pump to go to an 3 off-speed condition? 4 MR. WALDREP: We haven't seen anything like
-5 that.
li MR. EBERSOLE: If you choke suction, something is 7 going to happen to the speed. I guess your. sunction trip 8 intercepted what happened. I think it might be interesting 9 to find out what would happen. You!ve got the standard
-10 overspeed protection on the turbine, I guess, and you hope 11 that will work.
12 MR. WALDREP: I don' t have any more information 13 on it.
~
14 The other two situations that we have, of course, 15 or the other reactor trips, two of which, after the reactor 16 trip, as I mentioned before, we started the third auxiliary 17 feedwater pump to maintain steam generator. levels.- In two 18 cases where weigot to the safety injection point, we used 19 ~ the essential motor-driven pumps to maintain level in those 20 ' case s . In .all those cases, we had no problem maintaining 21 steam generator levels or in the operation of any of the 22 auxiliary feedwater or main feedwater systems. 123 MR. MICHELSON: Just a clarification. When you 24 found that the strainers had become clogged, was this a " 'O " 25 sudden revelation or had you seen the signs of this for ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage M336 6646
9250 17 02 253
/T is/ DAVbw 1 .several weeks or months? And. if it was a sudden revelation, 2 why did he material suddenly clog the strainers?
3 MR. WALDREP: For one , we don' ' t have permanent 4 plant instrumentation to the control room on what, say, the 5 DP across the strainer like that would be. After the first
,6 event, we did add some local instrumentation in the
'7 condensate and the feedwater lines to try to see in the 8 feedwater heaters, to try to really look and see what was 9 really going on in there, to see if there was anything else 10 that was causing some problems in our system. That's when I 11 guess we realized af terwards that the feedwater strainer did 12 have an unusually high DP, so we pulled it out.
O - 13 MR. MICHELSON: What did you say was the 14 indication of trouble to begin with? How did you know you 15 even had a problem? 16 MR. WALDREP: I'm not sure at what point you're 17 talking about. 18 MR. WARD: Why did you go in and measure the-19 delta Ps? 20 MR. WALDREP: Just as part of the startup phase 21 of the secondary performance group. They had totally pretty 22 much instrumented this second system, looking just at system 23 response. . So it's a normal thing in the power ascension , 24 process for us to do, not looking at any specific problem, O 25 but just looking at overall system performance. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
-9250 17 03 254
, DAVbw 1 MR. MICHELSON: You're saying the system itself 2 seemed to be working fine, but you discovered that it had a j l
3 low suction' pressure; is that what you're saying? 4 MR. WALDREP: During normal operation we didn't 5 experience a low suction pressure, everything worked fine. 6 Af ter the second trip, we started looking very hard, as 7 where we might be getting a little more DP than we expected 8 to see. And w$e said, well, it would be maybe a little more 9 in the strainers. So we pulled the strainers out, checked 10 them, and sure enough, they were clogged to about 30 11 percent. 12 MR. ADNI: I want to make another point. This is
?
(}_/ 13 the first time we could actually get velocity in certain 14 sections of the line to carry anything. We went through a 15 flushing program and ran all the pumps, and we'd not had 16 adequate velocity. 17 MR. MICHELSON: To get you the large delta P? 18 MR. ADNI: Correct. 19 MR. EBERSOLE: May I ask, do the main feed pumps 20 exhaust into the main condensers? 21 MR. WALDREP: Are you talking about the recirc on 22 the feed or the excess steam? 23 MR. EBERSOLE: I don't know, the discharge, the 24 exhaust from the main steam pumps. Do they exhaust into the 25 main condenser? ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336 6646
i 9250,17 04- 255 l s 1 L DAVbw 1 MR. WALDREP: Steam; yes. , 2 MR. EBERSOLE: Do you have a condensable vacuum 3 trip on that? 4 MR. WALDREP: Yes. Oh, the condensor on the 5 pumps?' A low vacuum trip on the pumps?
-6 MR. EBERSOLE: At least hypothetically, at very 7 low loads, they could go on exhaust atmosphere.
8 MR. WALDREP: I'm not sure. I don' t know. 9 MR. EBERSOLE: Well, it's degrading influence. I 10 would suspect it would be set quite high, if you have it, 11 and you must have it. Otherwise you might lose the circ 12 water pumps and close the condsasor, which might be a good 11 3 thing?
- 14 MR. WALDREP
- I'd have to do some looking. I.
15 don't know for sure. 16 MR. HAYNES: Bob can address that. 17 MR. ADNI: We have a ruptured disk on the feed )~ 18 pump and the set point for the main turbine trip would be 19 about 23' inches back pressure. Off of the main. feed pump, 20 I believe it's about 21, 17, in that neighborhood. So we 21 maintain the condensor for feed pump availability. 22 MR. EBERSOLE: You really don't need a vacuum 23 . exhaust, however, for the functional running of it, do you?. 24 I'm saying,'you couldn't run it with a blown diaphrame. O~ 25 MR. ADNI: Sure. ACE-FEDERAL REPORTERS, INC. ! 202-347-3700 Nationwide Coverage 800-336 6646 . _ , . . - - . - . _ _ _ _ _ _ _ . _ _ . ~ . . , _ . , ___ _ . . , ._ _ , . . _ . _ , _ _, . _ . - _ . - _
- a. o u .a - - . - - .
I
'l 256 l
.;9250 p 17 05
\< DAVbw. 1 MR. EBERSOLE: It might kind of make a mess of
~
2 the turbine haul, but if your back was up against the wall, 3 you could do it. Okay. 4 MR. WALDREP: That's all I had. 5 MR. EBERSOLE: Any questions? 6 '(No ' response . ) 7 We have Mr. Houston from the Staff . 8 MR. GORMIEL: My name is Jerry Gormiel again. 9 Mr. Houston has asked me to speak for the Staff on the 10 offset power reliability. We've reviewed this and satisfied 11 the reliability criteria of the standard review plan. It's 12 acceptable for that reason, and we contemplate no further , .O 13 action at this time. 14 MR. EBERSOLE: You had the same words on 15 Davis-Besse, and I don't have much regard for them. 16 MR. GORMIEL: You certainly shouldn' t have, 17 because it wasn' t reviewed against the standard review 18 plan. 19 MR..EBERSOLE: It satisfied all regulatory 20 requirements. 21 MR. GORMIEL: It may, indeed, satisfy the GDC, 22 so it wouldn't satisfy the reliability criterion. 23 MR. EBERSOLE: You have a reliability criterion? i-24 MR. GORMIEL: For one and the same review plan. O 25 MR. EBERSOLE: What is that, in so many words? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
9250'17 06 257 YJ DAVbw 1 MR. GORMIEL: 10 to the minus 4, 10 to the minus i 2 5. j 3 MR. EBERSOLE: All right. I'm going to ride 4 through the break on the schedule, because we took a late , 5 break, and ,we'll go straight into Andy Marchese's discussion 6 of A-45. 7 (Slide.) 8 MR. MARCHESE: My name is Andy Marchese , NRC 's 9 task manager for Unresolved Safety Issue A-45, entitled 10 " Shutdown Decay Heat Removal Requirements." I was asked to 11 come here to give you a presentation on where A-45, what the 12 status of feed and bleed is in the A-45 program, as well as
'# 13 a summary of what we know about foreign positions on the use 14 of bleed and feed or feed and bleed.
15 ( Sli.de . ) , 16 First of all, let me say that feed and bleed is 17 one alternative among several that we're looking at as part 18 of the A-45 program. We're not in a position as of yet to 19 say whether or not we're going to recommend feed and bleed, 20 as the proposed resolution of A-4 5, because we ' re still ir. 21 the process of evaluating both its benefits and the impacts, 22 that is, the costs and the benefits. 23 So what I'd like to do this af ternoon is give you 24 an overview of where we stand on feed and bleed. Then in
.(-)- \'")
25 about a month, we'll be meeting with the subcommittee on ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
9250 17 07 258 DAVbw 1 decay heat removal, and we'll have a better feel of how the 2 7ther alternatives are stacking up, relative to feed and 3 bleed. 4 This slide outlines the approach that we're 5 taking of bleed and feed. Basically, we're assessing it as 6 part of the PRA process, that is, the value and the benefits 7 will be estimated, quantitatively, by means of PRA. 8 Then we will compare the feed and bleed 9 alternative with others that are being looked at in the 10 program. 11 I've presented these other alternatives before,
~~ 12 but basically, we're looking at add-on trains of aux I i
'~'
13 feedwater, high pressure injection. We're looking at the 14 dedicated feed and bleed systems, such as the one that's 15 been proposed by Mr. Reed, which is a closed cycle bleed and 16 feed system. 17 Just about everything I'm going to be talking 18 today is about open cycle feed and bleed, and we basically 19 use existing equipment and blow it down into the containment 20 building. 21 We've done a lot of thermal hydraulics analyses 22 on bleed and feed at Los Alamos, and I wanted to talk to you 23 a little bit about this. This is basically to confirm that
,_s 24 the thermohydraulic analyses support that you can, in fact,
( ) 25 bleed and feed. As part of this, we have looked at l ACE. FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-33H646
9250117 08' 259
-k DAVbw 1 the key parameters that result in a successful feed and 2 bleed operation. That is , valve size of the relief valves ,
3 safety injection, pump head and the timing, which relates to 4 steam generator water inventory. 5 (Slide . ) 6 Now the status of the analyses that we've 7 examined . Two plans to date, both Point Beach and Turkey 8 Point -- we have in the program, as you know, seven plants 9 that we're analyzing in detail and subjecting them to a full 10 . value impact analysis. Five of these are PWRs and two are 11 BWRS. 12 Of the five PWRs , we finished Point Beach and 13 Turkey Point , and we're in the process of evaluating ANL 1, 14 Trojan and St. Lucie. I-15 At Los Alamos, we've analyzed four plants in 16 detail, using the track code simulation. The track code , as 17 you know, is a very detailed systems thermohydraulics cost, 18 very, very detailed, and these four plants were picked. 19 They are basically Westinghouse B&W, CE, and H. B. Robinson 20 is a Westinghouse, but with a low heac safety injection 21 system. Those four plants were picked mainly because-there 22 were detailed decks already available, which is 75 percent 23 of the work, at least, is involved in setting up the 24 computer decks and confirming that the systems information O 25 is correct. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
I l 1 1 l
- 9250 17 09 260 '
k_/DAVbw 'l MR. REED: Are any of these units, CE System 80? 2 MR. MARCHESE: No. 3 And then from these calculations, Los Alamos was 4 able to provide extension statements to other plants. That 5 is , given the results from those four plants and comparing 6 the key parameters, such as relief capacity, HPSI pump 7 capacity, steam generator inventory, they were able to draw 8 some conclusions as to whether or not bleed and feed would 9 be successful on other plants. 10 I am going to talk a little bit about that later 11 on. 12 (Slide.)
.(-
' 13 This is the results that we've seen so far from 14 Point Beach, insofar as the value of bleed and feed. When I 15 talk about value here, now we're using in cero melt 16 frequency as one measure of characterizing value, and chere 17 are others, some of which are qualitative, as part of the 18 overall evaluation of what we're doing now, but in terms of .
19 quanitative value indication, the first number up here is 20 the core melt frequency for internal events only. This does 21 not include special emergencies, such as fire flooding, 22 seismic or whatever. This is strictly internal, and it is 23 those internal things that are being looked it. It's 24 basically the response of systems to small break LOCAs and
'- 25 transients.
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33 H 646
- 9250 17 10- 261 gm
\-)DAVbw' 1 The base case now for Point Beach, was feed and 2 bleed available, because the plant has procedures in place. 3 Some of the plants have procedures in place; 4 others don't. Where they have then in plaga, we- sed bleed 5 and feed as the base case. Where they did not have them 6 available, we said that was not the base case and looked at 7 it then with adding feed and bleed. 8 The core melt frequency that we saw with the 9 bleed and feed is like 1.7 times 10 to the minus 4. ,Then we
'10 - go in and take out the bleed and feed operation, and we 11 evaluate all the sequences involved in getting to core melt, 12 come up with a new frequency without feed and bleed and then 's-(d 13' determiae what the delta is.
14 Now that is & pretty significant change, the 2 15 times 10 to the minus 4. Any change on the order of 10 to 16 the minus 4 is significant. 17- Now we need to determine what that is in risk 18 base, that is, determine what that buys you in terms of 19 what that saves your in person-rems and then -calculate in 20 terms of dollars what that buildup would be worth. 21 But we're in the process of doing that. ' 22 MR. WARD: Andy, those numbers, as I recall, are 23 radically different from the initial preliminary ones you 25 MR MARCHESE: Yes. The reason for that -- and
~
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336.e646 l
./9250 17 11- 262
- Ls
(~~. kDAVbw' 1 we ' re _ going to ' get ' into this when we meet with . you next 2 month -- is that we've done a lot more detailed analyses on 3 common mode failures, particularly_ in the auxiliary 4 feedwater system. Before we did not. do such a good job on 5 common mode failures in the aux feedwater system. 6 Anyway, when you put in all' these common modes in
'7 the aux feedwater systems, the value of feed and bleed goes 8 up significantly.
9 What that is saying is, if you don' t have a
-10 reliable an aux feedwater system, the value of feed and
'll -bleed is. enhanced.
-12 (Slide.)
O'- 13 For Turkey Point, we've done the same thing. 14 Now Turkey Point, the base case does not include 15 feed ~ and bleed, because the plant does not include feed and 16 bleed, because- the plant does not have procedurs, , and they
/
17 specifically told us they don' t intend to bleed and feed.. 18 So the base vase here is that without feedrand 19 bleed, 5 times 10 to the minus 4-is just internal events, 20 just for decay heat removal, small LOCAS and transients . 21 Then we go in and we put in bleed and feed. We 'T 22 recalculate and reanalyze all the sequences. The reduction c- 23 is even more significant here. It's like a factor'of 5, 24 where the 4 was a f actor of 2, coming up with quite a
< b 25 substantial delta, in terms of changing core melt t_
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 8003Wi646
9250 17L12 263-(~} ; 1s DAVbw' l frequencies. 2 Now again, we have to extrapolate this out to 3 person-rems to determine a valve, in terms of reduction in
~4 person-rem and what the -value of that is, in terms of
~5 dollars.
6 (Slide.) 7 -Now I mentioned conclusions to date in terms of 8 the PRA analyses that are being done. Feed and bleed-9 capability does provide some risk reduction. That is, we're 10 seeing significant changes in core melt frequency; however, 11- the timing of feed and bleed initiation is very critical, 12 for some plants and a decision has to be made less than 20
'i
.(d 13 minutes into an J incident.
14 MR. EBERSOLE: Is that for-B&W? 15 MR. MARCHESE: It's even shorter for B&W. This 16 Jis for. CE plants , Calvert Clif f s is about 20' minutes. In 17 fact, I'm going to cover that right now. This slide
-18 .probably should have been presented first.
19 (Slide.) j- 20 But anyway, these are the results that we're 21 'seeing out of the Los Alamos detailed thermohydraulics i 22 calculations. Feed and bleed was generally successful, if 23 initiated prior to steam generator dry out. , 24 Now the time for steam generator dryout is very () 25 plant-specific, and it is very dependent also on the time ; ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 33(Hi646 j
, ,~
- 9250 17 13 264
'^3 t \J-DAVbw 1 in which you scram the reactor.
2 These times were done for a rather quick scram, 3 less than 20 seconds. If you wait longer than that, these 4 times go down significantly. 5 For Oconee, we're seeing that in order to.
- . -6 successfully feed and bleed, that is, to feed and. bleed 7 successfully at Oconee, you have to do it before the steam 8 generator dries out. That's like around 8 minutes. For 9 ,Calvert, the calculation indicated about 20 minutes, because 10 there is more inventory for Zion and Robinson, which~ had 11 even' larger inventory in the steam generators, the results 12 were like 40 minutes.
13 We also did some calculations to see what happens 14 if you wait longer than the steam generator dryout time, and 15 what we found is, if you wait till this primary system
~16 becomes saturated, in some plants it was successful, and in 17 some it was not.
.18 For Calvert, if you waited beyond the 20 19 minutes . That is, if you went to 27.5 minutes, I think it 20 was, seven more minutes beyond 20, when the primary system
' 2:1 became saturated, you could not successfully bleed and 22 feed. . Cladding temperatures starting going on an excursion; 23 however, for Oconee and Zion, which, as you know, have very 24 high head HPSI pumps, you could wait as long _ as primary
- O
-25 system saturation, but if you wait too long, you're not ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646
9250-17 14 265 d,,c I:aVbw 1 go'ing - to be able to recover, and you're going _to start 2 damaging fuel. 3 4 5 6 7 8 9 10 11 12
-13 14 15 ,
16
'17 18 19 20
-21 22 23 24 25 ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 4646 l
I l l
~9250-:18 01L 266
- (~)'
% DAVbur1 MR. EBERSOLE: This means in fact that you have
.2 uncovered fuel or will uncover it, is that right?
3 MR. MARCHESE: The code actually calculates core 14 .uncovery. You may still not be going on a cladding
-5 temperature excursion when you have core uncovery. Then 6 eventually you get to a point where you uncover it to such 7 an extent you start drawing out pins and you start seeing 8 the temperature excursion.
9 MR. EBERSOLE: At the upper level, okay.
~10 MR. MARCHESE: So the code does account for 11 that.
12 M R'. EBERSOLE: Isn' t there a critical point -in 13 the power level there where you don' t get sufficient 14 frothing, even at higher. levels? 15 At lower power levels it may come even more , 16 quickly because you don' t get the throw of the liquid onto 17 the upper level of the fuel. 18 So I really don' t know how that works out, but 19 I know there is some challenge about one-third core height 20 if you are at low power. 21 MR. MARCHESE: One-third core height is about 22 when they - start . 23 MR. EBERSOLE: ' You can get a throw of water at 24 one-third core height under the upper one-th(yd, but at
'O 25 low power it will be dry.
ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 2 336-6646
9250 18 02 267 (sj'
-~
~ 'L/ DAVbur 1 MR. MARCHESE: Unfortunately, Jesse, I don't have 2 the thermal hydraulics analyst here. 3 MR. WARD: Andy, your analysis or the analysis 4 'that Los Alamos is making really isn' t that sophiJticated, 5 is it? It is just making some assumptions about core 6 .uncovery? 7 MR. MARCHESE: It does have models on this 8 frothing business. 9 MR. HOLTZ : Howard Holtz, Generic Issues Branch. 10 The way TRAC is set up it actually calculates the 11 cladding temperature from the 2200 degree cutoff. So it 12 tells the truth. () 13 It isn' t a question primarily of where the water 14 is when the plant temperature reaches 2200. 15 MR. MICHE LSON : That is dependent upon the 16 frothing actions and the steam cooling actions? 17 MR. HOLTZ: Yes. It includes those. 18 MR. MARCHESE: I think that is modeled in the 19 code. 20 MR. WARD: If you assume TRAC is doing it right? 21 MR. HOLTZ: Right. It is supposed to be a 22 validated code and all that jazz. Well, you know the 23 story. 24 But it is based on the calculations of what the O 25 clad temperature is. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
9250 18 03 268 DAVbur 1 MR. MARCHESE: That is all I have to say on the 2 status of bleed and feed. I was going to get into the 3 foreign positions. 4 Basically, the bottom line is we are still not in 5 the position yet to determine whether or not we are going to 6 be recommending bleed and feed, but we will certainly have a 7 lot of work that has been done. Within the next six months 8 we hope to be in a position 'o c say "yes" or "no" on feed and 9 bleed. i 10 MR. EBERSOLE: You are talking about the 11 cost / risk aspect? 12 MR. MARCHESE: Right. ..' { ' 13 MR. EB ERSOLE: It is a low probability to begin 14 with, so if- you doubled it that doesn' t change things a 15 lot. So I don't know where you can go. 16 By the way, in doing this analysis, Andy, you 17 take into consideration whether you are talking about old 18 plants that have to be heavily modified versus new ones, I 19 presume, where it would be no big thing to provide auxiliary 20 heat removal systems? 21 MR. MARCHESE: Right. We are looking at it both 22 ways. , 23 MR. REED: A couple of points. 24 The first point is I wish you would consistently 25 use the term " bleed and feed" rather than " feed and bleed" ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
9250 18'04 269 DAVbur 1 because it relates to the safety injection pump shut-off 2 pressures, and there is no reason -why 1500 pound pumps and 3 1800-pound pumps can' t do the job nearly as well as the 4 higher ones, and in fact ,they reduce the problems of 5 pressurized thermal shock. 6 So I wish you would consistently say " bleed and 7 feed," " bleed and feed," get the ' b' before the 'f." It is 8 also that way in the alphabet. 9 MR. MARCHESE: I stand corrected. 10 MR. REED: . Another point. 11 You said that recently with detailed work on 12 common mode failure you are finding more benefit in bleed ( 13 and feed. 14 Did you get Davis-Besse in there to review common 15 modes? 16 I think you can say, well, we have to think about 17 many plants in these common mode failure scenes. I really 18 think when you finally get a good handle on common mode and 19 the practices of providing auxiliary feed pumps that.are 20 identical and motors that are identical and valves that are 21 identical, you will see the real gain from diversity in 22 - principle, getting over to another principle of decay heat 23 removal. 24 So your summary was good. I liked it.
- A ') 25 MR. EBERSOLE: Andy, I wonder if you would agree ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33 H 646
9250 18 05 270 (~>') .
\- DAVbur 1 with me that this meeting here and all this discussion about 2 the practicality and reliability of going to low pressure o
3 really hasn't changed the picture on bleed and feed very 4 much atiall, except I guess we have enhanced the picture, 5 that we think a little bit better that we can' t get to low 6 pressure than we thought previously, what with the 7 revelation of the small events and other ways of doing it? 8 But basically, this meeting has not contributed 9 much toward the issue, am I correct -- that root issue? 10 MR. MARCHESE: I don' t think so. 11 MR. EBERSOLE: It is still with us just as it 12 always was. We have enhanced the picture. We have a 3
\- 13 somewhat better access to the conventional method of low 14 pressure cooling, but that is about all.
15 It is still a complicated system, and low 16 pressure cooling involves three trains of water -- the 17 primary loop, the interposed loop, the service water loop or 18 wha tever -- and it is still just as complicated as ever and 19 has the same number of interdependencies, and that hasn't 20 changed as a consequence of this meeting. We just think we 21 -might get down there with a little more confidence than we 22 had earlier. 23 At least that is my impression. 24 MR. MARCHESE: Right. b N/ 25 MR. EBERSOLE: Any other questions here? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
9250 18 06 271' MR. WARD: !
. DAVbur 1 Yes.
l 2 Andy, the numbers you showed for the contribution 3 of bleed and feed.to preventing core melt. To what extent 4 have you considered the equipment performance for bleed and 5 feed? 6 I mean, have you just assumed -- I mean, how aid 7 you handle the probability that the PORV will open and stay 8 open and that the pumps will supply the flow and will
-9 continue supplying the flow, particularly in light of 10 perhaps the changing atmosphere in the reactor building?
11 Is that sort of thing factored in? 12 MR. MARCHESE: Let me say a couple of things. [] N/ 13 First of all, the reliability data base in terms 14 of actuation of PORVs was taken from the IRA data base, and 15 insof ar as environmental qualification issues are concerned,
-16 you know, the PRA doesn' t really do a really great job on 17 that. In fact, it doesn' t really account for it, whether or 18 not the equipment is going to operate in the various 19 . environments.
20 Now, one.of the concerns that people have on . 21 blowing down in the containment in the bleed and feed . 22 process is whether or not all of the equipment 23 instrumentation to successfully bleed and feed is going to 24 operate in that environment. ("/ s-25 MR. EBERSOLE: Wait a minute. I felt certain you ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverase 800 336 66 4
9250'18 07 272 h2 DAVbur -l would be dealing with bleed and feed in a cleaned-up system 2 designed for the purpose, not this hashed-up system which is 3 vulnerable to its own environmental complications. 4 When you take on feed and bleed, do you not do it 5 .with a system design that is improved to resist the 6 consequences of feed 'and bleed. - 7 MR. MARCHESE: We are doing it both ways. 8 MR. EBERSOLE: What is the difference in 9 reliability as a result of doing it the right way? 10 MR. MARCHESE: The PRA analysis techniques are 11 not sophisticated enough to account for this environmental 12 qualification. f'h 13 MR. EBERSOLE: With the PRA-wh-ich uses the 14 current PORVs , which. are vulnerable to their own discharge, 15 I would discount bleed and feed entirely. 16 MR. MARCHESE: The systems were never designed to 17 bleed and feed. 18 MR. EBERSOLE: That is not what you are using to 19 come up with these numbers , is it? 20 MR. MARCHESE : As I mentioned, the PRA models 21 don' t really account -- 22 MR. EBERSOLE: Come on, they must be better than 23 that. 24 MR. MARCHESE: Environmental qualification is not 25 considered in the data base. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
9250 18 08 273 i / DAVbur 1 MR. EBERSOLE: In the PRA? For heaven sake, 2 well, that really discounts it entirely as far as I am 3 concerned. 4 MR. WARD: You might say that the numbers he 5 showed might be valid, but when you look at the costs to 6 protect the equipment to deal with it, you haven' t put that 7 in the analysis yet. 8 MR. EBERSOLE: But for heaven sake, you need 9 motor-driven ball valves . 10 MR. MICHE LSON: Which aren' t even necessarily 11 reflected in -nuclear applications. In some cases the 12 valves' reliability numbers come out of industry -- f) x- 13 petrochemical, et cetera -- and don' t necessarily 14 reflect.... 15 MR. EBERSOLE: Tell me this, do you have some 16 numbers which incorporate the bleed mechanism using high 17 quality,- if necessary, petrochemical type ball valves with 18 motor-driven, hermetically sealed drives , guaranteed to 19 stand the environment? 20 I want to be sure when I am going to open up. 21 MR. MARCHESE: We are using the data base that 22 has been developed for doing PRAs, and it has failure rates 23 of valves. And as Mr. Michelson indicated, those valves 24 could be any number of different types of valves. (n)
%> 25 MR. EBERSO LE: Have people been claiming that ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646
9250'18 09 274 y, S t DAVbur 'l they can feed and bleed with valves which are not 2 environmentally qualified? 3 MR. WARD: . Sure, they are. 4 MR. EBERSOLE: On what reasonable basis can they 5 do that? 6 MR. MARCHESE: Some people contend that if it 7 works in a large LOCA environment it ought to work in a feed 8 and bleed environment. 9 MR. EBERSOLE: But it,doesn' t work in a large 10 LOCA enviromr. ant; it doesn' t have to work.
~11 MR. MARC HESE : Yes, I know. That other stuff 12 that has been qualified.
'r
' 13 MR. EBERSOLE: But this is not qualified.
14 MR. HARCHESE: I did have a slide to indicate one 15 of our main considerations is environmental qualifications 16 if you are going to feed and bleed on an open system. 17 MR. EBERSOLE: I wouldn' t even approach it in any 18' other context. 19 MR. REED: I think that we first ought to keep in 20 min: that bleed and feed, if utilized, is a last ditch 21 backup thing. Your auxiliary feed system has let you down. 22 You are into bleed and feed because you are going to try to 23 save the core. 24 Now, I think there are three ways that you might s (~'#
's 25 bleed and feed. If you have been reading my paper, you ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646 j
m 9250 18 10 275 ( DAVbur- 1 have read the three ways. 2 The dirty system. There is a slight upgrading of 3 the present PORV system. 4 The second system is the Cadillac, the outside
'S flash tank system.
6 Then you have what I call the containment 7 supporting system where you do partly take it out, but you 8 do mess up the containment. 9 But I really think you have to keep in focus here 10 .that your auxiliary feed has failed. So you ought to do 11 something now before core melt. 11 2 MR. EB ERSOLE: The least expensive thing to do is 7 k/ 13 to qualify the PORVs and say that they will open. 14 MR. REED: Sure. 15l MR. EBERSOLE : That is not a world-shaking -- 16 M R. REED: When A-45 is over, I would think that 17 that is what has happened. 18 MR. MARCHESE: If we decide bleed and feed is the 19 way to go, I think environmental qualification would have to 20 be addressed. 21 MR. WARD: That has got to be part of the cost, 22 the cost and benefits. , 23 MR. MICHELSON: The qualification process, 24 :though, doesn' t necessarily answer the reliability question 25 concerning the valves, and the numbers you are using, I ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-33(H5646
-9250 18 11' 276 DAVbur 1 suspect, are not' at -all based on the reliability of the kind 2 of valve we are - talking about because we haven' t had pe haps 3 enough experience yet to collect a good data base 4 So the' numbers, it .is questionable how meaningful 5 they really are when you are talking factors of 2. - I don' t 6 believe I can put any weight on a f actor of 2.
7 MR. MARCHESE: I agree, but to do a quantitative 8 cost / benefit analysis , which is what we have to do to go 9 through a CRGR process and everything, we have to use 10 something. 11 MR. EBERSOLE: Andy, isn' t it just logically 12 common sense to say bleed and feed has zero reliability r~g AJ 13 unless we have environmentally qualified PORVs? 14 MR. MARCHESE: I think that is going a little too 1 15 far. I think there is some chance if the valves are not 16 qualified -- and nothing is qualified as regard to feed and 4 17 bleed -- there is still some chance it would operate 18 successfully. 19 MR. EBERSOLE: Remember now, they have to be held 20 on voltage,' hot solenoids, with every opportunity to go bad 21 in the present of a hot, moist adversary, as well as the 22 motors, which may be unable to work even more backward. 23 Any other questions or comments? 24 (No response.) 25 MR. EBERSOLE: If not, we are still short. 4 ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(H5646
, _ _ . . . - -.- - .. . . .- - . _ . . _ . , . .=,
N 9250 18 12 j 277 f - D DAVbur 1 MR. WARD: He,hns got another chart. 2 MR. EBERSO LE': Go ahead.
' 'y "
3 MR. MARCHESE: I1was going-to give you some N 4 information on foreign positions. .
' < i;c m
-5 (Slide . )
~
.6 This is the list of references that' we used to
-7 characterize what we know about foreign positiors on bleed
- t . !
F c .t JG 8 and dee.d. t 9
' Let me just say a few things. This is a dynamic 10 situation, hnd these positions could change.
11 For exampl,e, about two years ago when(ve talked 12 to the Germans, they were very negative on bleed and feed. 13 They didn' t even want to talk about it. They just 14 completely had no iry tioh of even analyzing it., s. 15 But whens we talked to them about a ' year ago, they 16 were examining ble'ed and feed in terms of their severe 17 accident policy issue, and they were doing specific 18 calculations. 19 That informatf cn is a year old, and you know 3 s g' 20 things change in a year. So I want to caution you that what 21 I*am going tg present is about a year old, at least a year 22 old, and it mdy have changed. J I' 23 2 If you all want to send me over there to get the j s 24 la te st , just write a letter to Mr. Dircks. 25
..(.cughter.)
I ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 8tu 31'4646
.. u I
n 9250:18'13 278 g T_) DAVbur 1 MR. REED: I think that there may be a' trend 2 toward endorsement of bleed and faed. That is a good one to 3 make. 4 It seems to me I have read a memo by somebody 5 lately that says that sotne foreign country is waiting to see 6 what we are going to do from A-45. 7 Also, there is at least this fellow Petrocelli, 8 ' with his core rescue system, and one other person -- I
-9 believe he is Swiss -- who have written papers endorsing 10 this alternative.
11 MR. MARCHESE: I am sure that is part of it. I 12 think a few -countries are definitely waiting to see what we 13 do. 14 (Slide.) , 15 I am going to present information on about eight 16 countries , but this is just to start out with a summary. , 17 Generally, the eight countries that we have 18 information are generally not in favor of bleed and feed. 19 However, there is some variation from country to country.- I 20 mm going to get into that. 21 Those that are analyzing it all feel that it is a 22 last' ditch means of decay heat removal. Their primary 23 concerns are PORV and relief valve operation. That is, you
-24 are operating in several different flow environments , such
(-). 25 as single phase liquid, two-phase, and then single phase ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33WA6
Af .
-9250 18 14 279 p)
( .DAVbur l' steam, and whether or not the valve will operate reliably in
'2 all those modes .
3 It is questionable. 4 ' M:R. EBERSOLE: These countries, though, are their , 5 opinions based on their trying to drive these functions
,6 through the same PORVs ithey have already got?
'i' ', '
7 MR. MARCHESE: Kind of yes,'except the French. 8 The French feel they have a more reliable valve, the Siben 9 relief valve. I tdink they are about the only ones that 10 have bought that mobe reliable type of valve. 11 But basically it is prettf much the standard. 12 MR. EB ERSOLE: How do the French feel about bleed l ("% (-) 13 and feed? 14 MR. MARCHESE: They have done calculations to 15 indicate they can do it. They have developed , procedures , 16 but they feel that the procedures need to be confirmed. 17 And I have a slide on that.
-18I They are also concerned about the prompt operator 19 action time. That is, you have got to t,ake action very 20 quickly, of the order'of 20 minutes.
21 They are also concerned about the adverse
~
22 environment created inside containment. 23 MR. EBERSOLE: Why would they be concerned about 24 that when they must not be concerned about it for the large 0) s 25 LOCA? ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-33(H5646
t
.9250cl8 15 280 k DAVbur 1 MR. MARCHESE:- What you find is thatiduring a 2 bleed ~ and feed operation the temperatures and pressures, 3 even though they may. not be as high as a large LOCA, they 4 hang up for a longer period . of time, and so what you - have is 5 a hot . atmosphere , stretching _ out a long time.
16 In our case , the equipment -- some of the 7 equipment has not been qualified for that environment. 8 (Slide . ) 9 : Okay, what I was going to do is -just march 10 through these eight countries, if that is what you want to 11 do. 12 In West Germany, feed and bleed has not been -- .13~ dealt with as a licensing issue. They feel that they have a 14 .very reliable aux feedwater system, and in addition to that 15 they have special emergency systems , and bleed andifeed, at ' 16 least in terms of licensing space, is not only unnecessary 17 'but undesirable. 18 Now, the means that they prefer for rapid plant 19 depressurization is they prefer to use their secondary side 20 cooldown through atmospheric dump valves coupled with the 21 use.of safety grade spray systems. As you know, they have. 22 large ADVs, highly redundant. Their steam generators are 23 designed to cool down at 100 degrees Kelvin per hour, and 24 the steam -generators can take it.
- 25 MR. EBERSOLE: Pardon me j ust a minute. You ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 300-336 6646
9250-18 16 281 c /m . I ) DAVbur m 1 recall, we just got through saying that plant 2 depressurization is not accomplished by secondary blowdown. 3 MR. HARCHESE: With aux spray. I am saying they 4 do it with spray. They have the spray. 5 Now, the Convoy design has small relief valves, 6 but they are smaller than. typically-what we have-in the 7 States here, and they just basically were put on there to 8 minimize spurious actuation of the safety valves. .But they 9 are smaller than what we have. They have about four of them 10 on their newer KWU design. 11 They are doing calculations to see just what it 12 would take for bleed . and feed as part of their severe (- X.I 13 accident policy.
. 14 MR. EBERSOLE: So can you open them?
15 MR. HARCHESE: Yes. 16 Now, some of the older KWU designs, the Gosken 17 facility in Switzerland, which is one of the older KWU 18 designs, does not even have a PORV on the pressurizer. It 19 is kind of a mixed bag. 20 21 22 23 24
~ - '25 ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646
E' 9250 19.01 282 DAV/bc- 1 MR. REED:' On that West German one, when did you
~2 make your last contact?
'3 ' MR. MARCHESE: This is about a year old.
4 , MR. REED: Of course , we members of the ' ACRS had 5 contacted in May of this year. I would say there is a trend
'6 to think differently. You might want to make that trip.
7 MR. MARCHESE: I'm r3ady. My bags are packed. 8 _MR. REE D: I'm trying to think of the gentleman 9 you' re trying to talk to. Is it Meinger? 10 (Slides.) 11 MR. MARCHESE: Switzerland. The Swiss have no 12 plans to use feed and bleed. They are I think one of the t' - 13 most negative on it. 14 MR. EBERSOLE: The second sentence expla* ins that.- 15 MR. MARCHESE: Exactly. They prefer steam 16 generator reliance and bunkered special emergency systems. 17 I might add the Swiss, Germans, and some of the other 18 countries, have a lot more redundancy and diversity than we fl9 do in their safeguards systems. 20 For rapid plant depressurization, again, they l 21 prefer to'use_the rapid secondary side cooldown coupled with 22 aux spray to depressurize the primary and these systems are j i 23 all safety grade. 24 Again, their primary concerns are feed and bleed, 25 adverse environment and the effect on essential operating I ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800-336-6646 l
---- ~- -. - -. - - ...--.- - - - -- - ,
, ...- \
!9250'19-02 283 !
DAV/bc 'l equipment. 2 (Slide.) j 3 Belgtum now, as you know, basically, this is a , 4 Westinhouse Framatome sort of design. In addition to that, 5 they have added on a lot of special features for their newer 16 plants. 7 Now they have some older Westinghouse PWR's and
-8 they're examining feed and bleed for the older plants. I 9 think they' re one of the countries that are anxiously 5
.10 : awaiting to see what we do-0cn feed and bleed, particularly 11 for older plants.
12 But for their newer plants, the PORV's, the steam
- 13 dump valves , and aux spray are all safety grade. Again, 14 .they prefer reliance on the steam generator. and their 15 bunkered feedwater systems.
16 Also on this last bullet, I think, kind of 17 applies across the board to many of these countries, they 18 feel that they have more redundancy, more diversity in their 19 safeguard systems than we do in - the States. 20 Therefore, bleed and feed is not that much for 3 21 them. 22 (Slide.) ' 23 The Netherlands, they have an older KWU design 24 plant there called the Borsela. Their final position on 25 that has not been made; however, they do not think feed and l - ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 33H646 .
- -. -..~ .
l l l l 19250'19-03 284 DAV/bc 1 bleed is a good approach to emergency decay heat removal. 2 Now they're adding -a special emergency dedicated system 3 which is basically an add-on aux feedwater system as well as. 4 a' high capability to supply makeup to the primary system. 5 And, again, they would prefer using the aux 6 ' feedwater to cool down through the steam generators. Again, 7 their concerns are of course adverse environment
'8 instrumentation and control, and operability of the PORV
-9 during this kind of environment.
10 -(Slide.) 11 The Italians, the regulatory agency, ENEA, is 12 evaluating the effectiveness of feed and bleed for their 'O\"' 13 advanced PWR design. This is the PUN design. They're 14 actually doing.some specific calculations. They haven' t 15 arrived at a final position, but there again, they still are 16 concerned about 'the timely operator action that's required 17 and whether or not the instrumentation is going to operate 18 in an adverse environment.
-19 Now, those five countries, we talked to about a 20 year ago. The next three -- France, England and Japan --
21 the information is about -- my information is about two 22 years 1old. You may have some more recently. 23 (S lide . )
- 24 The French are now more positive on bleed and 25 They've analyzed it, they've done calculations , they feed.
1 ACE-FEDERAL REPORTERS, INC. 102-347-3700 Nationwide Coverage 800-336-6646
]
i 9250 19 04 285 h A_j DAV/bc 1 feel they have the capabilities. They have procedures 2 implemented but they're still concerned. at least the people 3 we talked to, that even though you have procedures in place, 4 that does not ensure that they are going to be successful 5 and that the operator is going to take the right action. 6 But they plan to use it if they ever need it. 7 (Slide.) 8 England. 9 MR. REED: I'm listening to these rundowns. I 10 wonder if, in your accumulation of information, the issue of 11 steam generator tube rupture is raised in its relationship 12 to bleed and feed cooling, and more rapid depressurization. / )
's / 13 Now, you take the Palo Verde. Here's a plant way 14 out in the wilderness, way out in the desert; I guess , large 15 radioactive release is not significant there.
16 But a lot of these European plants , radioactive 17 release is a much more significant issue. And if you do 18 incorporate and include some aspects of design into bleed 19 and feed, you.can turn around steam generator fillup 20 radioactivity release, and these kinds of things, rather 21 neatly. 22; Was that ever discussed, or are you just talking I 23 ~ decay heat optional techniques for removal, as with regard 24 to bleed and feed? 25 MR. MARCHESE : We talked about steam generator ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
L9250'19'05 286 (q L /DAV/bc 1 tube ruptures and, basically, I would characterize our
, 2 philosophy as they isolate the damaged generator. Their 3 plants have three or four generators, so if they get a' tube 4 rupture in one generator and they isolate that . generator, 5 .they cool down with the others.
6 MR. REED: But, wait a minute. You have to 7 assume safety valves dislodged or open. Are you telling me 8 they have loop isolation valves? Some do over there, I'm 9 _sure. Do they have primary loop isolation valves? Or just 10 the secondary bottling up? 11 That doesn' t snuf f out radioactivity discharge or 12 steam generator overfill, so long as you've got sa fe ty 13 injection pumps higher pressure and they' re pumping right 14 through the tube hole. 15 MR. EBERSOLE: What I heard you say is that they 16 can isolate steam generators. 17 MR. REED: On the primary side?
~18 MR. WARD: I don' t think so. But they come out 19 on depressurizing fairly rapidly on the secondary side, 20 getting the primary pressure down .
21 MR. REED: I think they're going to depressurize 22 cn the primary side. Probably better auxilliary spray. You 23 can' t get the primary pressure on the secondary side. Heat 24 removal. 25 MR. WARD: You've got to have both. You can't ACE. FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
s
'9250fl9 06 287.
DAV/bc. .1 -get it down through the spray either if you . don' t cool it 2 down.
'3 MR. MARCHESE: You've got1to have both.
4 14]R. REED: You've got to assure natural 5 circulation, too,'but I wondered if the steam generator tube 6 trupture thing is carried along in the discussions. 7 MR. MARCHESE: I think it is. Okay,.where.am I?
'8 England. The Siz-awell design. Feed and bleed is a diverse 9 .meth'od of decay heat removal. The design has the capability
'10 to bleed and feed. No credit was taken for it in the
'll Sizewell' safety analysis.
- 3. 12 However, the Sizewell owners', the safety O
'As 13 =autho.rities, require that the Sizevell owners have to adopt 14 bleed and feed procedures.
So they supposedly had the 15 . capability to -bleed and feed at Sizewell. 16 (Slide.) 17 Japan. Their position, and this is at least two 18 years old, feed and bleed is not required based on what they 19 feel is a highly reliable aux feedwater system. They're 20 very confident of their steam generator tube integrity. As 21 you know, they have a lot of inspection techniques going on
.22 over there. And annual inspection of the steam generator 23 tubes is performed.
24 They're very confident on their aux feedwater 25 system. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 6646
9250 19 07 288
. /-
(_/ DAV/bc. 1 Okay, that's all I have. , 2 MR. WARD: Ycu know, one thing, as I recall, the 3 KVU plants', they claim in both Germany and Switzerland, they 4 claim that the steam generator tubes are thicker than. in the 5 U.S. The U-2 generators. Because of the different design. 6 Something about their design for some stress limit based on 7 imposition of both maximum delta T and delta P at the same 8 time, or something like that. 9 But they end up with a little thicker steam 10 generator tubes. That makes them feel better, I guess, 11 about both failures and also about the ability to tolerate 12 this rapid cooldown through the secondary side. (O _) 13 MR. REED: In fact, that's what I talked about to 14 Dr. Meinger. He made that point -- well, we have thicker 15 tubes. We can take the full differential of this high 16 pressure safety injection approaching zero on the secondary 17 side in a hot condition. 18 And I don' t know how well our tubes, based on the 19 criteria that we have for rejection and plugging, can do 20 that. 21 So they have an advantage in some of their steam 22 generator tube analysis. 23 MR. EBERSOLE: Further questions? 24 (No response.) ^ 25 MR. EBERSOLE: I believe I must call a 10-minute ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-66E
, m _
E '9250 19.O'8
.289
- fm ,
5 ) DAV/bc 'l break. We 'll come back at 20 af ter 4. 4 2 (Recess.) 3 MR. EBERSOLE: I'd like to reconvene here. We 14 have Mr. Rick Turk from CE on the last topic of the 5 evening. . -Then we'll have a general discussion of what to 6 - bring before the f ull committee. And have a word about what 7 to do with Commissioner Asselstine's request to us. 8 Rick, it's yours . 9 MR. TURK: I'd like Mike Meisner from Louisiana 10 Power and Light to make a couple of comments regarding the 11 CE owners group. 12 MR. MEISNER: Good afternoon. I'm Mike Meisner.
~
113 from Louisiana Power and Light. We brought several of the 14 CE' owners group representatives with us today. As you 15 recall, we were before you a couple.of years ago on this 16 PORV issue. What we wanted to do today was address the last 17 item on your agenda as it relates to what happened at Palo 18 Verde on September 12th. 19 To that extent, Rick will be making a 20 presentation and I'm available to at least try to answer any 21 questions. For Waterford III, we have Dennis Cox from 22 Southern California Edison. We have Mike Keller from 23 Washington Public Power Supply System; and Terry Quan from 24 Arizona. 25 MR. EBERSOLE: Do any of you wish to make your ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 336 6646
I 9250-19:09 290 f'l l j DAV/bc 1 own' independent statements, or are you going to. leave it to u 2 ' Rick? 3 MR. MEISNER: ' We hadn' t planned on making 3-4 independent statements but.we've anticipated perhaps you i ! 15 would have plant-specific questions after going through Palo
'6 ~
Verde. 7 MR. EBERSOLE: Thank you. Okay. 8 - ( Sl ide .') 9 MR. TURK: Before I start the presentation 10 regarding the auxilliary spray and how it related to the I 11 work we did a few years ago, I'd like to see if I can' t 12 . answer a couple of the questions- that came up this morning r A' 2 \/ 13 and in the early afternoon discussion. 14 Mr. Reed is not here but there was a question } 15 ~ regarding the sprayhead assembly and what it looked lik. 16 -inside the pressurizer. Essentially, it's a four-inch 4 17 nozzle with a thermal shield that extends down through the 18 nozzle. Then there is a sprayhead assembly which I really 19 don't have the details as far as its connection and 20 construction, but it's essentially a diffuser with
- 21 perforations to atomize the spray flow.
22 I think Mr. Reed's question in regard to failure 23 of that as a component as well as backdown in the upstream 24 side, the regenerative heat exchanger, another component. k/ 25 As the staff said in the basic analyses of these i ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33 4 646
9250 19 10 291 ' f]DAV/bc t- 1 systems, these' are considered passive components.- We'd be 2 considering things like ruptures in the tubes or somehow 3 breakage of these components. 4 'I think it's important to remember that -these are 5 systems that are on line during normal operation. So if you 6 develop a tube leak in the regenerative heat exchanger or 7 you saw degrading of the spray efficiency through normal 8 spray, it would become apparent to you as an operator. 9 The other .g I would point out, with regard to 10 the nozzle , normally .t will see flows of up around 300-350 11 gpm from the reactor coolant pumps; whereas, under the aux 12 spray condition, we're talking about flows on the order of , . 13 maybe 30 or 40 gpm. 14 MR. REED: That's exactly part of my point. When 15 you reduce the ficwa, the atomizing action becomes more
'16 necessary. When you put a great big squirt in there, you 17 perhaps can get it just from an open-ended pipe. I might 18 point out, I believe, in ancient history, there is a case of 19 the spray nozzle coming off and being discovered in an 20 inspection, a yearly inspection. I don' t know whether it 21 was bolted on or where the reactor was -- whether it was 1
22 screwed on or whether it was welded on. 23 MR. TURK: A good point. The other thing that I 4 24 guess I would like to reinforce is the fact that this is not
~ 25 a unique feature to the System 80, or not a unique feature ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
L9250-19 11 292 C) T'v' DAV/bc 1 to the plants without PORV's. This is essentially the same
, 2 spray system that, as far as I know, is on all PWR's 3 regardless of who the vendor is.
4 There are some slight variations, but I think
. 3 they all have basically an auxilliary spray to perform that 6 function of not really decay heat removal but it's the 7 function of 7etting from one decay heat removal mode to the 8 other decay heat removal mode.
9 MR. EB ERSOLE: Rick, in reference to this 10 business of the large flow versus the small, and to the 11 applicant, is it required that they periodically investigate 12 the auxilliary spray mode rather than the full spray mode
\#
13 from the main coolant pump to confina that diffuser is still 14 functioning?
.15 MR. , TURK: It's not required but I think it 16 happens. I think it happens in virtually every shutdown.
17 You eventually get to a situation where you wind up stopping 18 the reactor coolant pumps. That's really the reason the 19 system is on there. It's for the latter stages of cooldown, 20 where you're stopping the reactor coolant pumps. Pressures 21 admittedly are down and you go on to that system. 22 So it gets exercised with some -- 23 MR. EBERSOLE: With routine operation. 24 MR. TURK: Routine is probably not the right 25 word, but certainly with some periodicity. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
9250 19 12 293 p i / AV/bc D 1 I guess, the other point that was .':ind of lef t 2 somewhat. unanswered was the question about pressurizer level 3 and situations where the plant might be water solid or the 4 pressurizer water-solid. 5 As Mr. Ebersole pointed out, there are redundant 6 level indicators which give virtually full span indication 7 in the pressurizer. I think one thing that came out during 8 the discussion was the fact that we are using HPSI pumps 9 with shutoff heads, that are around 1,800 pounds. 10 So that generally we won' t be in a position where 11 those pumps will wind up charging the plant solid. Now, 12 there are provisions in the emergency guidelines that are i As 13 reflected in the procedures that discuss the various success 14 paths under a situation where , for maybe some reason, you 15 have lost level indication, all your level indications, 16 where you might choose to pump the plant to solid. 17 This would be a situation where you are shut down 18 and if you do that and go to a solid plant mode, which, 19 again, is not something that's totally unheard of in 20 shutdown modes , cold shutdown modes , you'll wind up with a 21 situation where plant pressures can change very rapidly, 22 either up or down due to temperature changes and mass 23 additions or mass withdrawals from the system. 24 So you essentially wiad up in a situation where (~~\ l
\' 25 pressure would be being controlled by that HPSI shutoff l l
l l ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
9 50-19 13 294 E~DAV/bc 1 head. You essentially have that pump on the system.
-2 Then, by removing heat-through the steam 3 generators and throttling on the pump discharge, you could
~4 reduce system pressure.
5 MR. EBERSOLE: The fuse at the bottom, does that 6 permit a full tank drain if there's superheated steam below? 7 MR. TURK: It offers resistance obviously to 8 drainage. I~know there are dynamic LOCA situations where 9 you wind up -- you may wind up with two-phase in the loop 10 situation and subcooled water in the pressurizer that have 11 .been analyzed in detail through the LOCA scenarios. 12 I'm really not prepared to discuss those. I
\/
13 think, for the kind of decay heat removal scenarios that , 14 we're talking about, where you'll be cooling down the plant 15 and draining it, there's no question that it would just 16 drain down. And we're talking about a process that you do 17 every time you take the plant to cold shutdown. 18 19 20 21 22 23
.f g 24 U 25 ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336 6646
~
9250 20 01 295
- { DAVbw 1 MR. EBERSOLE
- With the cold shutdown, you would
. 2 always maintain pressure on the level, because you're not in 3 a pressure now, or you could keep it under conditions where 4 you have a leak of some sort in the primary system below the 5 pressurizer,: through which you could leak water and rate in
-6 excess of-the reciprocating pump makeup. At least, 7 . hypothetically, you could go - to a steam condition below the 8 pressurizer and have a steam-cooled primary loop with a 9 liquid-filled pressurizer and still show-level.
10 MR. TURK: A steam-filled -- 11 MR. EBERSOLE: -- core, if this doesn't drain 12 down. f'Y sj 13 MR. TURK: If it, for some reason, was blocked 14 and isolated from the system, for instance, if yr2 had an 15 isolation valve. 16 MR. EBERSOLE: I understand there are designs in 17 the field which don't require a valve type isolation, but 18 rather just the steady pressure, will hold water flow. The 19 orpis design or screon design is such that differential 20 pressure will hold, and it won't come down. 21 MR. TURK: I don't know of any any static 22 situation that would do that. The only thing I can think of 23 would be a dynamic blowdown effect where you've got lots 24 amount of two-phase in the loop, due to a large LOCA that () 25 would be holding up. ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-6646
-~9250'20 02 296 p)DAVbw-(, -1 MR. EBERSOLE: Carl, can you comment on this?
2 MR. MICHELSON: No. What I know of is where the 1 break is at the top of the pressurizer, and then you have a 4 ' problem. 2 5 MR..EBERSOLE: Of course, it's clear in that
-6 case.
7 MR. MICHELSON: But if the break were low on the 8 surge line, for instance, and the pressurizer is still 9 attempting to control it, all it did was proceed to drain 10 itself, because it's a high pressure point in the system. 11 MR. EBERSOLE: We can then afford to say that it 12 will always drain; correct?
,a
(_/ 13 MR. MICHELSON: I think that's correct for a 14 break down below the surge line. 15 MR. EBERSOLE: Okay. 16 MR. TURK: I think those -- 17 MR. MICHELSON: I'1,1 hasten to qualify,s of
.18 course, that depends on the size of the break and what the 19 scenario will be. It takes time, because of the diffuser-20 screens, to drain the~ pressurizer, but given time, it will 21 drain.
22 MR. TURK: One fact that I do know. I don't know 23 that it will help. The total flow areas of that diffuser 24 screen is, in fact, larger than the surge lite pipe.
) 25 MR. EBERSOLE: Fine. So if you go to some ACE-FEDERAL REPORTERS, INC.
_ . _ .,-, JD2-347-37{_ _ _ Nationwide Coverage 800-33H646
, ~ l l
~9250 20 03 297 jlgDAVbw 1- saturation pressure down below, the water will come downm.
2 MR. TURK: Yes.
-3 MR. EBERSOLE: Of course, it will'.
4 MR. TURK: If there are other questions that I've 5 left unanswered from earlier in the day, think of them and 6 bring them up. 7 (Slide.) 8 But I'd like to proceed now with the planned 9 presentation to discuss the review we've made-of the owners 10 group evaluation of a year and a half ago on the subject of 11 rapid depressurization, and that review was effectively 12 directed, as Mike Meisner said, to what we learned from the () 13 auxiliary spray issues at Palo Verde and how they affected,
'14 first of all, the previous depressurization evaluation, and 15 I think the tone for that response has pretty well been set 16 during the day today, as we realize that auxiliary spray was 17 not a big factor in the PORV feed and bleed ' issue. It was a 18 matter of proceeding from the one mode of heat removal to 19 the other.
20 So I don't thinP you'd be surprised to see that 21 we don't feel the issues have a large effect on our 22 conclusion that we couldn't recommend backfitting PORVs. We 23 do also want to cover, though, the specific differences 24 between the Palo Verde system and the other three plants () 25 that were involved in the study and those Mike mentioned are ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 33H646
d 9250 20 04 298 l Waterford Unit 3, San Onofre Units 2 and 3 and the
-v!'DAVbw 2 Washington Public Power Supply Unit 3.
3 So to do that, again, and to kind of reenforce 4 .the conclusion, I think, that kind of came out of the rest 5 of the day's discussion. 6 (Slide.) I 7 It's convenient to go back and look at how that 8 study took place over the last two years or so. If you 9 recall, it was a substantial effort that was done to respond 10 to basically 14 questions from the Staff regarding the 11 potential uses for PORVs. The purpose of this slide is 12 really just to show that there was a large number of aspects () 13 and issues that we looked at. There were really only a few 14 that bore on the auxiliary spray. We do have a question 15 about that auxiliary systems capability and how it is used 16 to get to cold shutdown, which is obviously impacted by i 17 these issues. 18 We also did some probabilistic evaluations of the
. 19 ability to prevent core damage, both with and without PRVs, 20 and certainly, there was some implication of the auxiliary 21 spray in those evaluations, and I want to address those.
22 MR. EBERSOLE: Are the other plants affected by 23 the findings at Palo Verde? 24 MR. TURK: Other affected, in essence, no, for () 25 two reasons. One, because the auxiliary spray really had ACE FEDERAL REPORTERS, INC. Nationwide Coverage 800-33Mi646
, . _ -.202 347 3700 . - - _ - , . .. -- --_ --
1 l l 9250-20 05 299 7"iDAVbw 1 very little bearing on the conclusion not to pursue PORVs, l
\m/
2 two, because their design on the suction side of the j 3 charging pumps for other reasons are different, and it 4 results in a situation where they do have emergency power to 5 the various valves. They do have remote control room 6 control of those valve and suction sources, which we'll talk 7 about. 8 MR. EBERSOLE: By the time the modifications are 9 made to Palo Verde to fix it up against these findings, will 10 the other plants, will they be equivalent to or better? 11 MR. TURK: They already are equivalent to, is 12 what I'm saying. 13 MR. EBERSOLE: So then Palo Verde got into its { 14 situation through unique design features, not a standardized 15 design, not a CE design? 16 MR. TURK: It was, in fact, a CE designe but it 17 was a change in design that was made, and I'm jumping ahead 18 a-few slides. 19 (Slide.) 20 There were some functional differences between 21 the System 80 CVCS design and the 3410 megawatt CVCS 22 design. 23 In both cases, the scope of the system was within 24 CE, but the functional differences that we made between {} 25 those plants and what we were trying to do at the time was ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverase 800 336-6646
1 l l
'9250 20 06 300 )
jlhDAVbw 1 to go in the direction of putting safety system reliance on 2 the engineered safety feature systems, that is, the ECCS 3 systems. 4 What we did was, we eliminated in System 80, the . 5 high concentration boric acid makeup tanks that has 6 previously existed in the CVCS and replaced in the ECCS 7 systems with a higher concentrt: ion of boron in the 8 refueling water tank. 9 We attempted to consolidate all the safeguards 10 functions in the safety ejection system or the ECCS system. 11 The 34 10 megawatt plants do initiate charging flow on 12 safety injection. I) 13 okay. So now we had another system besides the 14 safety injection that was involved in this safety system 15 actuation. And System 80 consolidated that. The other 16 functional difference that came into play is, System 80 has 17 different reactor coolant pumps, as you're probably aware. 18 That's the KSV pumps, which require seal injection. That is 19 now supplied by the CVCS system. 20 So those functional differences, again, jumping 21 ahead a few slides, but it's probably the correct time to do 22 it -- 23 (Slide.) 24 -- result in the fact that those plants wind up (o) 25 with the volume control tank isolated on a safety injection ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-3*6 6646
l I 19250 20 07' 301 1 1 } DAVbw 1 . signal,.which means that the valves required to do that are 2 lE valves. They are powered from the' control room. 3 The boric acid makeup pumps also are started on a 4 safety injection actuation signal, as well 'as the gravity 5 feed lines from the boric acid makeup tanks being aligned on 6' safety injection. , 7 So again, those valves are off of emergency 8- power. They are controllable from the control room, and the , 9 34 10 megawatt plants can take the actions that Palo Verde 10 took and do them from the control room. 11 MR. REED: I'd like to just explore some other 12 differences, pluses and minuses. f) s- - 13 I would guess that Palo Verde has a slight plus 14 because it's out in the desert, and it can tolerate lots of 15 radioactivity discharge on the steam generator tube 16 ruptures. 17 Also Palo Verde has a high pressure injection 18 system which is on the order of 1800 pounds shutoff. I was 19 wondering about other sites like Waterford. Are they closer 20 in? Can they tolerate the same kind of radioactivity 21 discharge? Same primary to secondary leakage before it's 22 terminated or equalized, and do they perhaps have higher 23 safety injection pump shutoff heads? 24 MR. TURK: Let's take them in reverse order. () 25 The safety injection shutoff heads on local ACE. FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 3366646
Ib L= ll c
~,;9 250 ? 20 08 - 302 1
j )DAVbw 1 plants'are actually slightly lower. As opposed'to 18 [ -2 pounds, I think they're down around 1500 or 1600 pounds. I l
.3. don't recall exactly.
4 MR. REED: So there are no CE System 80 plants 5 with this auxiliary spray slow depressurization system that 6, have.like 2300 pound HPSI's? ! 7 MR.-TURK: That's correct. 8 MR. REED: I'd say that's a plus. 9 MR. TURK: Certainly, with regard to pressurized 10 thermal shock and other considerations, we feel that it is. 11 With regard to the tube rupture, I think the correct way to 12 answer that is, they're all subject to the same () 13 requirements. They certainly have to'make plant-specific, 14 as we were discussing earlier, meteorological calculations 15 and other aspects, but in the end, they're all subject to 16 the same basic requirements. f 17 MR. EBERSOLE: There's no site population density 18 difference. It's all the same standards. The fact that 19 you're out in the desert doesn't do you any good, except in 20 a philosophical sense. 21 MR. TURK: Your zones are probably spaced out 22 further, but the method of calculating what has to happen in 23 those zones would change. 24 MR. EBERSOLE: Are there any CE plants that have l( ) 25 in-built to them the problems discovered at Palo Verde which ace FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 4646
..9250'20-09 ~303
[]DAVbw . I have no PORVs?
.v i
~2 MR. TURK: The four that we're discussing here 3 are the four. ,
4 MR. EBERSOLE: That's the entirety of them? , 5 MR. TURK: Current. So I kind of got out of
- 6. . order here a little bit. I think that's essentially the.
7 essence of what we looked at from the deterministic 8 standpoint. That is, the ability to meet RSBS-1. In your
~9 handout there are some-comparative flow diagrams between the 10- systems, if you want to examine them.
11 I think I'd like to move on to the second aspect, t 12 which was the probabilistic analysis which was done. I 13 think that the bottom line here is that the.probabilistic (( f
- 14 analysis that we did ---
15 (Slide.) 16 -- which is not unlike what Andy Marchese's 17 people are doing for A-45. We looked at essentially the
- 18 difference in core melt probability or core damage 19 probability with and without PORVs to try and deternine if 20 there was a benefit to backfitting PORVs to these plants.
. 21 I think it should come as no surprise that 22 when we're looking at core damage situations and core damage 23 transients, the auxiliary spray does not come into play to a 24 large degree. It is the situation we talked about this () 25 morning, or as long as you Save heat -emoval through the 4 ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 3364646
l 9250 20 10 304 I~'DAVbw 1 steam generators that you have the aux feedwater working s_/ 2 that the time frame that you have for depressurizing and 3 getting down into the other mode, that is the shutdown 4 cooling mode, are quite lengthy, and it is consistent to 5 rely on systems that might require manual actions. 6 That analysis in 239, which was the report given 7 to the Staff, essentially addressed three potential PORV use 8 events, that is, loss of heat synch, where the PORV 9 conceivably could have been used to back up not the aux 10 spray, but the aux feed system and perform the heat rule 11 function. 12 Okay. Conceivably, PORVs could be used in the () 13 tube rupture, as we talked about earlier today; however, 14 with regard to the probabilistic analysis here, on core 15 damage, the only requirement is, as we were discussing, I 16 thirix that eventually, you have to get down onto shutdown 17 cooling to stop that leakage path, if, for some reason, the 18 leakage exceeds the makeup available, that would eventually 19 lead to a core melt type scenario. 20 And then the last scenario that was evaluated was 21 the PORV initiated LOCA, which was intended to be evaluated 22 to show the down side of potentially adding these valves. 23 MR. EBERSOLE: Rick, let me make an observation 24 on this point. () 25 You now have a one-inch vent system, which, in a ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33Mi646
- 4. .. - . ... -. .
3
\ )
9250-20 11 305
't, ' T^';DAVbw 1 way, is a substitute for the PORY-initiated LOCA, and you've i/
2 designed it. soothe potential for even a small LOCA through 3 it is virtually impossible or negligible. It's got three 4 valves in series. 5 I am confident that when you put Item 3 up there, 6 that you optimize the PORV configuration, to enhance the 7 possibility of accidental blowdown'through it, rather?chan 8- go the,way that you did, through the one-inch vent. That's 9 always the. ways the business is done. 10 MR.'TORK: No, that's not true. In the 239 11 analysis, we did not take, for instance, the existing PORV
\
12 syadoms that are out there, where the PORV is normally 13- aligned and set to go on high pressure. We model the syntam s 14 that would have normally been unisolated, would not be
'[- 15 actuated, except by perator-action, and he would have to ld j open two valves. He would have' to open both the block valve
<t 17 --
18 oha. EBERSOLE: You did do e system somewhat like - 19 che current vent system. 20 MR. TURK: Only larger. It was essentially a i. 21 four-valve system. Two trains, two valves in each train. 22 MR. EBERSOLE: So it was a matrix of four valves 23 to guarantee'both closure and (opening. l
, 24 ,
MR. TURK: Correct. i . l (). (25 MR. EBERSOLE: That, in fact, did have a spurious ACE-FEDERAL REPORTERS, INC. 202-347-37'00 Nationwide Coverage 800 336-6646
1 I 9250 20 12- 306 - I'3DAVbw- LOCA potential'with the good valves. 1 'O 2 MR. TURK: Yes. And basically, the LOCA 3 potential was one of -- after, it was initiated for cause. 4 In other words, there would be certain cases -- 5 MR. EBERSOLE: -- where you' thought it was doing 6 the right thing. 7 MR.~ TURK: -- where you would use it, and there 8 was some potential; that's right. 9 MR. EBERSOLE: So these are unnecessary 9 10 deliberate initiations; right? 11 MR. TURK: Right. 12 MR.-EBERSOLE: What was that frequency? 13 MR. TURK: I don't recall offhand. I really {} 14 don't have the information from the reports. It was low, as 15 were the other frequencies. If you remember, one of the 16 results of that overall effort was that the frequencies that 17 we're dealing with, both for the non-PORV and the PORV, 18 including case, were down. 19 MR. EBERSOLE: Was it as low as the B-1 event, 20 the coupling of primary to secondary excess pressure? 21 MR. TURK: I'm not sure. 22 MR. EBERSOLE: That's a standing problem right 23 now, and that has the unique potential, of course, of 24 blowing up the low pressure coolant complex, direct damage () 25 effect, rather than an indirect one. And I'd be interested ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-6646
9250 20 13 307 T'~T DAVbw 1 in the augmented hazard of having these valves against the L) 2 background of that which is already there. 3 That is also compounded now by the dual tracks, 4 the low pressure coolant, you know, because of the 3 5 requirements of duality to go to low pressure. You now have 6 two opportunities in existing valves, not only to discharge 7 primary,. coolant , coolant fluid into open space containment, 8 but right directly into the low pressure cooling complex 9 which destroys the opportunity to cool the core in an 10 ' eventual sense. 11, , , The incremental presence of these other valves, 12 I think, would be negligible, in the presence of these other 13 two that are already in place. (G~'\ - 14 Did you look at that? 15 MR. TURK: No. This was not a full PRA. It was 16 limited, very closely, to the events we discussed with the 17 Staf f, where there was a potential for using the PORV, in 18 -lieu of concurrently existing functions. 19 MR. EBERSOLE: Let me ask the Applicant, Palo 20 Verde or Arizona Power, to have a look at the relevant 21 hazard afforded by the presence of PORVs in your 22 configurations, the 4 by 4, versus that which they already 23 have, with respect to going into primary. 24 MR. DAVIS: Mr. Ebersole, this is George Davis 25 from Combusion Engineering. To answer your earlier question (]) ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(Hi646
9250 20.14 308 T^'DAVbw 1 about the PORV-initiated LOCA probability, the numbers we
~U 2 have come up with before were on the order of 10 to the 3 minus 7.
4 MR. EBERSOLE: So it's way out in the mist 5 anyway. L 6 MR. DAVIS: A very low number. The thing you may 7 remember about our previous discussion 'is that when you 8 looked at the pros and cons, the potential for a 9 PORV-initiated LOCA was a very small number, but the benefit 10 to a secondary heat sink, et cetera, was also a very small 11 number, so you were looking at the comparison of two very 12 small numbers, positive and negative. 19 13 v 14 : 15 ; 16 17 18 19 20 21
. .: 2 23 24 25 ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 4646
i 9250 21 01 309
'DAVbur 1 MR. EBERSOLE: What is the current event B J'
2 probability? 3 MR. TURK: Inadvertent unisolation of the 4 shutdown. 5 MR. DAVIS: That we didn't have in the numbers. 6 MR. TURK: I don't know that I would have a 7 comparable number of that. We would have to go back. 8 MR. EBERSOLE: That is really a chaotic event 9 because it puts primary coolant in the precise place where 10 you can't tolerate it. 11 Thank you. 12 (Slide.) () v 13 MR. TURK: We were on the subject of event 14 systems. I would point out this is the way the event system 15 is configured on San Onofre, and Waterford has a similar 16 configuration. 17 Instead of having the double valve arrangement 18 that Palo Verde has around on the pressurizer side, both 19 valve paths are subject to the small orifice. 20 (Slide.) 21 I think that essentially really covers the 22 comparison we wanted to make with those other plants. 23 The conclusions from the previous effort from 24 CEN-239 that remain valid in light of relooking at these 25 systems vis-a-vis Palo Verde's experience are, first of all, (v~~') ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6686
9250 21 02 310 JggDAVbur 1 that the aux spray system provides the safety related method 2 for depressurizing with heat removal maintained by the 3 generators, and'the system designs are consistent with the 4 requirements of the branch technical position for this class 5 of plant. 6 The performance of the aux spray system at Palo 7 Verde was consistent with the assumptions we made in the 8 analysis; that is, that manual actions were taken and taken 9 in a reasonablo amount of time to reestablish suction flow, 10 and finally that the aux spray reliability really had a 11 negligible effect on the probabilistic analysis of 239, 12 basically because the system was modeled in a realistic [ }j 13 way. It is not figured in as an engineered safeguard 14 system, and there is a relatively long time for operator 15l- action. 16 MR. EBERSOLE: Any questions? 17 (No response.) 18 MR. EBERSOLE: If not, that leaves us with just 19 the residual discussion topic. 20 MR. REED: Jesse, I have been listening to the 21 discussions today and seeing these plots of slow 22 depressurization, and I have heard the arguments that Palo 23 Verde is not unique because of its site and that the 24 radioactivity that might go from primary to secondary over a
^
25 long period of time -- I have heard the argument that there (x. s') ACE-FEDERAL REPORTERS, INC. ; I 202-347-3700 Nationwide Coverage 800-336-6646
9250 21 03 311 DAVbur 1 is not -- I have some concern that the size of tube rupture 2 has been selected as too small. 3 We talk all the time about single tube rupture, 4 quantities of the order of 4- to 600 gallons per minute, I 5 guess on Westinghouse type plants at max pressure, and we 6 talk about multiple tube ruptures, and I always can be 7 concerned. 8 We used the term in the ACRS' letters a number of 9 times, this absence of rapid depressurization, and I guess, 10 -as we see again, certainly there is not rapid 11' depressurization. 12 I am sort of standing on the fence. I don't know 13 ' whether rapid depressurization is necessary or not. It all .- (~) us 14- goes back to an assumption of what is the. size of the tube 15 rupture and a lot of other things and their spillover or 16 overfill and all these other things that haven't been quite 17 fully reviewed. 18 I do, however, feel that the issue of the 19 alternato paths to decay heat removal are certainly 20 diminished by this auxiliary spray technique versus PORVs, 21 but on the issue of rapid versus slow depressurization, I 22 don't think I have enough information to say that rapid is a 23 requirement. 24 MR. EBERSOLE: Glenn, we are still right where we 25 always were. f} ACE-FEDERAL REPORTERS, INC. 202-347-3700 Natiortwide Coverage 800 336-6646
9250 21 04 - 312 1 The mode of heat removal is still what it always JggDAVbur 2 was. It is still through the low pressure RHR system. It 3 hasn't changed. We have enhanced the reliability and, to 4 some extent, the rapidity with which we can get there, but 5 we haven't changed its nature at all. 6 So all I have really heard is that we appear to 7 have a somewhat more reliable track to get to low pressure 8 cooling, but it is no better when we get there than it 9 always was. 10 To this extent, I do not see that the Advisory 11 Committee's position, for which Commissioner Asselstine 12 asked us to review this, that there can be any change in it () 13 -as a result of all this flap about the method by which we 14 depressurize. 15 Let me read you the question that we are standing 16 , with here. 17 Commissioner Asselstine requested the Advisory 18 Committee on Reactor Safeguards to provide an updated 19 committee position on the Palo Verde Unit 2 prior to any 20 anticipated low power or full power operating license 21 decision for this plant by the staff of the Commission if 22 the committee has a different view from that contained in 23 their December 15, 1981 report on Palo Verde. 24 I would say it is rather evident that we can't 25 have any different position because nothing has really (v~) ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 33M486
9250 21 05 313 1 changed as a result of all this. We still have to ride on {' }DAVbur 2 the.RHR system as it always was. 3 We appear to be better able to get there, sooner 4 than we first thought, by virtue of the improvements that 5 have been made, and I really see no particular need for very 6 much of an elaborate presentation to the full committee. A 7 very abbreviated statement would seem to me to be all 8 right. 9 But I would open this topic to the subcommittee 10 to reflect the individual opinions. 11 Dave. 12 MR. WARD: I think I am in pretty good agreement () 13 . with your bottom line. But something you said bothered me. 14 I think what I have learned over the past few 15 weeks is -- and today -- that the auxiliary spray system is 16 not as reliable as I thought it was three years ago. 17 MR. EBERSOLE: Right, but they are going to fix 18 it. 19 MR. WARD: There seems to be some things that are 20 being done to improve its reliability, and with those I am
- 21. not persuaded that our position need be any different than 22 it was --
23 MR. EBERSOLE: That is exactly what I am saying. 24 MR. WARD: -- three years ago, now almost four (~)
\_/
25 years ago. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide coverage 800-336-6646 ^~
~9250.21 06 314 jlgDAVbur 1- MR. REED: To address Commissioner Asselstine's 2 ' request, I think I would have to go back and read all'three 3 letters, not the December 15th, 1981 alone, because I think i
4 the '83 letter and maybe the '85 letter have modified a 5 little bit that '81 position. 6 So I think I would have to read tonight all three 7 letters and reflect on what I have heard here today to 8 decide whether we have a different view, and then we would 9 havo to qualify the view an he has related it through the 10 '81 letter. 11 MR. EBERSOLE: What I have heard is we have 12 discussed the meaas by which we translate t,o the same system
'() 13 . we always knew about, the low pressure RHR system, and we 14 found out - there are ways that we need to improve it and 15 will.
. 16 But it is the same' system in the long run when we 17 get there, anyhow. 18 Carl? 19 MR. MICHELSON: A little different tack on the ] 20 question. 21 I am trying so sort out in my mind the importance 22 'of seismic qualification of this function or lack of seismic
.23 qualification of this function. Maybe we can have a little 24 discussion about why one need not worry about the fact that
() 25 it isn't seismically qualified. I am still sorting it out. l l ace-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646 l
9250 21 07 315 gggDAVbur 1 I would like the benefit of anybody else's thoughts on it. 2 I am surprised, though I haven't had very much 3 time to think about it yet, but I thought it was seismically 4 qualifiad. 5 MR. EBERSOLE: Is there any real concern about a 6 particular vulnerability in the seismic context? 7 Rick, we are supposed to have a qualified system 8 here, right? 9 MR. TURK: A system that meets the requirements 10 of RSB 5-1. That means that it is safety grade, and there 11 were three exceptions. I think they were noted in one of 12 the early presentations with regard to manual actions. 13 MR. EBERSOLE: What is the impediment to making (v) 14 this fully safety grade in the seismic context -- a few pipe
~ 15 supports?
16 MR. TURK: I am not sure that I heard. 17 Mike? 18 I am not sure that I heard it was not seismic. 19 As far as I know, the system is seismic on all plants. 20 MR. MICHELSON: For your plants, for Palo Verde, 21 the answer I think is "no." It is a mixed bag. 22 MR. EBERSOLE: What is the explicit vulnerability 23 to the seismic influence? 24 Can someone tell us from Palo Verde? 1 1 (~^) 25 It may be just something you could fix v ACE-FEDERAL REPORTERS, INC. , 202-347-3700 Nationwide Coverage 800-336-6646 f I
9250-21 08 316 jlgDAVbur 1 straightforward 1y. 2 MR. KOROSKI: Mike Koroski, from Combustion l 3 Engineering. , 4 I am not too'sure of the scope of the question. 5 MR. EBERSOLE: What is the seismic vulnerability 6 in'this depressurization system, using the auxiliary spray? 7 MR. KOROSKI: If you follow the lines and sources 8 of water from the charging pump suction all the way to the 9 pressurizer, that entire section is seismically qualified 10 and seismically designed. 11 MR MICHELSON: We are talking both the power 12 side as well as the pressure boundary side, of course. () 13 Is that a statement for both?- 14- MR. KOROSKI: I think there are a couple of 15 valves for which we indicated that there was not. 16 MR. MICHELSON: Well, you remember now we had 17 discussions earlier -- I assume you were here -- concerning 18 the automatic transfer from the volume control tank to the 19 refueling water storage tank. This particular system I 20 pursued to some extent, and we found out it was not 21 seismically qualified. 22 MR. KOROSKI: That is true. 23 MR. MICHELSON: It is a surprise to me, and I was 24 trying to sort out, and I haven't had time yet to think () 25 about whether that is important or not. I was just asking l 1 i ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800-336 4646 L. l
9250 21'09 317 '{];DAVbur_ 1 .for a little help. 2 So there are parts of this system, at least for 3~ Palo Verde, that are not seismically qualified unless I 4 misunderstood. 5 MR. EBERSOLE: Is that just the electric power 6 part? 7 MR. KOROSKI: Yes. 8 MR. MT.CHELSON: There was the transfer system. 9 MR. KOROSKI: I think generally you are talking 10 about the electrical components themselves and you are not 11 talking about the pumps, the valves, the ~ piping, and that 12 type of thing. 13 MR. MICHELSON: ( )~ That is the area we pursued, at 14 least. 15 MR. EBERSOLE: Can that be overcome by the thesis 16 that you can manually operate anyway? 17 MR. TURK: Yes. 18 MR. MICHELSON: That raises the question, of 19 course, whether or not we had anything, and you also have to 20 now speculate concerning inadvertent actuation, since it is 21 not qualified for the earthquake, for instance, and I don't 22 know which position it might refer back to on an after shock 23 to the earthquake as an example. 24 So I don't know whether to worry about it or I l
-( ) 25 not. I don't know if you have thought about it or not.
l ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(H5646
9250 :21 10 318 TO DAVbur 1 But I am just saying that'it is something that I kJ 2 haven't sorted out, and I don't have any other comments. 3 MR. EBERSOLE: What about you, Charlie? 4 MR. WYLIE: I agree with what has been said. I 5 don't have anything further. 6 MR. EBERSOLE: I think it would probably be just 7 as well that we have an absolute minimum presentation from 8 the staff and from the applicant and possibly CE to take up 9 any unexpected questions from the full committee, but I 10 think three men is enough and we won't have this convention 11 to take up what I think is no change in the issue. 12 Okay. Does anybody else have any comments? Do () 13 you care to make any comments or questions? 14 MR. WYLIE: I would like to ask a: question just 15 to clarify in my own mind regarding the 13.8 KV buses which 16 lie somewhere between the plant and the switchyard. 17 Mr. Barrow, I believe you may be able to answer 18 that question. 19 The buses are what, metal-clad buses that cross l 20 over to tie all this stuff together, the 3000 13.8 KV buses, 21 and they are what, aluminum, aluminum housing, aluminum 22 conductors? 23 MR. BARROW: I am not sure. I believe they are 24 steel housing. () 25 MR. WYLIE: Are they rectangular or round? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-66 4
9250 21 11 319 T'NDAVbur 1 MR. BARROW: Rectangular. .V 2 MR. WYLIE: Okay. 3 , With relation to these transformers themselves, 4 where do these buses lie across the cross-buses, 5 physically? 6 MR. BARROW: The buses you are talking about -- 7 MR. WYLIE: You have three 500 KV 13.8 8 transformers on the double load secondaries, and then you go 9 to three sets of switchgear basically. But then'you have 10 got cross-ties between the switchgear, Unit 1 to Unit 3, and 11 between the various units there. 12 Physically, where are those buses? f() 13 MR. BARROW: Can I put a slide back up? 14 MR. EBERSOLE: Sure. 15 MR. BARROW: Let's go over the handout. 16- MR. WYLIE: That is in II-D, I guess. 17 You have got enclosed there a startup 18 transformer, and you have got startup transformers 1, 2, and 19 3. The secondaries to those transformers connect across the 20 buses. 21 MR. BARROW: That is right. 22 MR. WYLIE: Physically, where is that located 23 _ with relationship to startup transformer 2? 24 MR. BARROW: By that, are you talking about the () 25 non-seg bus that we had trouble with? ace-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(HWi46
>9250.21 12 320 IN 1 MR. WYLIE: No, I am talking about these Q;DAVbur 2 cross-buses up here.
3 MR. BARROW: These cross-buses between S0-5 and 4 S0-6 are done -- the connections from the normally open and 5 normally closed breakers are all made with cable -- 6 MR. WYLIE: The crosses are? 7 MR. BARROW: The cross-connections are all made 8 with cable and duct below ground. 9 MR. WYLIE: Okay, that makes a big difference. ,= 10 MR. BARROW: Like I said, the 13.8 lines run from 11 S0-5 to S0-3 and from S0-6 to S0-4. Those are overhead 12 conductors. Il 13 MR. WYLIE: Okay. S0-5 to S0-6. v 14 MR. BARROW: These connections -- 15 (Slide.) 16 -- are all underground and made by cable in 17 underground duct. These connections from S0-5 -- I will 18 just use Unit 2 -- from S0-5 to S0-3 and from S0-6 to S0-4 19 are 13.8.KV overhead lines on regular transmission towers, 20 and then these connections from the aux transformer to S0-1 21 and 2 and from the S0-1 to S0-3 are nonsegregated phase 22 duct. 23 Those are the conductors that I am talking about 24 ~being rectangular, the aluminum rectangular bus. f] 25 MR. WYLIE: Okay. You would examine ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-66 4
9250 21'13 321 1 transformer-2 for explosions, and it wouldn't take out all X-}/T~ DAVbur 2 that? i 3 MR. BARROW: These transformers are 4 well-separated and the cables which come out of these 5 transformers immediately go down from the bushings down into 6 the duct. We feel they are quite well protected. 7 The cabl'e that run by these transformers run 8 through other ducts, which are in the encasement, which 9 wouldn't be affected. 10' MR. MICHELSON: I assume they are also protected 11- from the subsequent fire following the transformer 12 explosion? () 13 MR. BARROW: Yes, they are because they are in
- 14 separate ducts. I think we use ABS plastic duct encased in 15 concrete, and they are in absolute separate duct runs. Not 16 all of the ducts in the encasement come back together.
17 MR. REED: Jesse? 18 MR. WYLIE: Thank you. 19 MR. REED: Jesse, I would like to point out that 20 since 1981 -- and that is the letter. referred to by 21 Asselstine -- there was an October 18th, 1983 letter. Now, 22 it is not Palo Verde specifically. The letter relates to CE 23 System 80 and the need for rapid depressurization. 24 I think it is unfortunate that the 1981 letter () 25 and the 1983 letter and even the 1985 letter have focused ACE-FEDERAL REPORTERS, INC. I 202-347-3700 Nationwide Coverage 800-336-6646
9250 21 14 322 1 more on the need for rapid depressurization, more than the ()DAVbur. i 2 issue of decay heat removal. 3 I would'like to read the last paragraph of the I 4 October 18th letter for our reflection because wu have to 5_ decide what we are going to do here. 6 The committee agrees with the NRC staff's 7 recommendation to integrate any new requirements for rapid 8 depressurization into the more comprehensive new 9 requirements for improvements to decay heat removal systems. , 10 expected to be forthcoming from Task Action Plan A-45 within 11 one year. 12 That would be 1984. n ) 13- We see no need for earlier resolution of the PORV 14 issue. 15 I have listened to the depressurization issue. 16 I know we have focused on it and used it as a title. We 17 have perhaps overfocused on it versus the decay heat issue, 18 the option, the alternative here, the bleed and feed 19 option. 20 But I really think the two issues exist, and I
-21 think I could probably say, okay, if I want to solve this 22 thing piecemeal I can be convinced that maybe they don't 23 need, let's say, more rapid depressurization.
24 But I come back again to an open A-45 PORV thing, () 25 you know. Don't they go along together? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(HWi46
!= 9250 22 01 323 jDAV/bc 1 MR. EBERSOLE: That last statement that you read
-v 2 is appropriate there. We must integrate the method of 3 getting to low pressure with the adequacy of low pressure 4 when we get there.
5 MR. REED: I would assume that if A-45 ever comes 6 out with that and comes out with bleed and feed, you're 7 -going to get-the rapid depressurization capabilities there. 8 MR. EBERSOLE: Oh, sure, we'll get it. What I 9 thought I heard you say was though that, in getting through -1 10 low pressure, we must consider the means by which we get' 11 there, which is what we've been talking about today, and 12 integrate the problems in getting there as well as the ; 13 . problems after we get there. (V"T ll.4 MR. REED: What I'm trying to decide in answer to 15 your dilemma is whether we need to bring it up. before the
, 16 full. committee or whether we have to reply to Asselstine. I 17 guess we should, and what do we then say?
18 I think one thing we can say is perhaps the need 19 for rapid depressurization is not the chicken or the egg, 20 whichever one comes first, but the one that comes second. 21 And A-45 still rules wi,th respect to decay heat removal and 22 reaching an acceptable plateau on the core melt risk. 23 MR. EBERSOLE: Any further comments? 24 MR. WARD: Jessie, I'd like to address a comment 25 which will eventually be a question to the staff. If we go (])- ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-33&M46
' 9250 22102 324
^ ~- T' DAV/bc- 1 back to 1982 or whatever the issue of the PORV's in the M 2 System-80 plants-wasLfirst focused on'as an issue, as I 3 recall the position the staff took, it was that they were~
- 4. taking tha position that PORV's should be required as a I
5 backfit for these plants. But the need was not, well, the 6 pros and cons were fairly close. And they were willing to
'7 wait in some interim for the resolution and a better 4'
8 understanding of the whole situation from A-45 before-9 insisting on an implementation.
- 10 That's the position of the ACRS. I guess, 11' unfortunately, A-45 has stretched out. The resolution is
- 12. -maybe 18 months late. But has the staff changed its 13 position from that? Can anybody answer that? Did you
(])' 14 understand my question? 15 DR. YANG: That is not our position, that we will e
-16 wait for A-45 for the resolution.
17 MR. WARD: I realize that, but the original staff 18 position was that it appeared that the requirement was going . 19 to be that PORV's would be' required as a backfit. 20 DR. YANG: But the staff study on PORV issue was 21 limited in scope and the staff considered A-45 would have 22 more consideration in the study. So we'll have a better 23 decision. 24 MR. WARD: Maybe it's an acade.uic distinction. () 25 MR. EBERSOLE: Maybe another system, more ACE-FEDERAL REPORTERS, INC.
- X)2-347-3700 Nationwide Coverage 800 336 6646
9250 22 03 325 i lsDAV/bc 1 competent than PORV's, might be proposed. You know, a 'd 2 dedicated shutdown system. 3 MR. WARD: Okay. 4 MR. EBERSOLE: I want to thank everybody for 5 coming and making such a clear presentation and taking up 6 all these complicated matters. 7 MR. LICITRA: Mr. Chairman, I'd like to have a
. 8 question I'd like to ask. I mentioned that the staff and 9 our applicant presentation on Thursday should be limited.
10 The question I'd like to ask is with regard to the operating 11 experience. The staff representative who made that 12 presentation, Mr. Jessie Crews, of Region V.
/') 13 MR. EBERSOLE: I want him. I want him to be A.J 14 present. I thought that was a considerable addition to the 15 entire meeting and will add'the proper flavor to what's
~
16 going on out there.. 17 MR. HERNON: The other part of it, having 18 essentially brought the committee up to date as of 80 19 percent and holding and getting ready to rerun the load 20 rejection test, the ACRS letter said that you wanted to 21 review the total startup experience for Unit I. 22 The question I have is: Do you want another 23 installment of the same thing after this meeting? 24 MR. EBERSOLE: My impression now is that things 25 are going very well. I don't see any need for that. [} ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
9250.22 04 326 {,DAV/bc 1 Any comment by the subcommittee members on that? 2 (No response.) 3 MR. EBERSOLE: Having gone through the rigors of 4 the first phase of s,tartup, everything seems to be coming 5 .out very well. 6 MR. WARD: . I don't disagree with what you say. 7 MR. EBERSOLE: I again thank everybody for coming 8 and giving us such fine presentations. 9- With that, we'll adjourn the meeting. 10 (Whereupon, at 5:20 p.m., the meeting was 11 concluded.-) 12 h 13 14 15 16 17 18 19 20 21 22 23 24 25 ACE-FEDERAL REPORTERS, INC. 2 -347-3700 Nationwide Coverage 800-336-6646
2 CERTIFICATE OF OFFICIAL REPORTER is I (); i This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the matter of: 1 NAME OF PROCEEDING:- ADVISORY JMMITTEE ON REACTOR SAFEGUARDS SUBCOMMIPTEE ON CESSAR/PALO' VERDE NUCLEAR ENERATING STATION i DOCKET NO.: PLACE: WASHINGTON, D. C. DATE: TUESDAY, NOVEMBER.5, 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) y 6 (TYPED) DAVID L. HOFFMAN Official Reporter ACE FEDERAL REPORTERS, INC. ReportTar's Affiliation O
,, _ , , _ _, - , . , - , r -
. - . . ~ .. . - - . . .- -.
k O . REVIEW 0F CEOG EVALUATION OF RAPID DEPRESSURIZATION IMPACT OF CURRENT AUXILIARY SPRAY ISSUES ON: 4 o PREVIOUS DEPRESSURIZATION EVALUATION (CEN 239) o 3410 MW PLANT AUXILIARY SPRAY DESIGN 4 O - i 1 5 1 4 O js 9 O
~
f 4
-e,--- .--.,.--_-_-,-----.,,,__.,,,m.,n -
=. . . . . ._ - -
i RESPONSE T0'14 OUESTIONS ([ REGARDING POTENTIAL BENEFIT OF PORV's TOPIC AUX SPRAY IMPACT i
- 1. AUX SPRAY CAPABILITY X !
- 2. MINIMIZED CHALLENGES TO RPS 3.- EFFECT OF PORV's ON ATWS
- 4. EFFECT OF PORV's ON PTS
- 5. MULTIPLE FAILURE SCENARIOS
- 6. ALTERNATE SG HEAT REMOVAL
- 7. CHEMISTRY CONSIDERATIONS
- 8. RISK DUE TO LOSS OF HEAT SINK X
- 9. RISK DUE TO SGTR X
- 10. RISK DUE TO PORY INITIATED LOCA X
- 11. EFFECT OF PORV's ON RISK
- 12. COST OF PORY
- 13. SG TUBE PLUGGING CRITERIA
- 14. SG VIBRATION ANALYSIS
h D DEPRESSURIZATION ANALYSES HAVE BEEN REVIEWED' RELATIVE TO PVNGS l AUX SPRAY ISSUES l l
- 1. DETERMINISTIC REQUIREMENTS l o RSB 5 ' )
o PLANT CuMPARISON i
- 2.
- PROBABILISTIC ANALYSIS o LOSS OF HEAT SINK o SG TUBE RUPTURE o SMALL BREAK LOCA O .=
') 4 .
\
4 e O 8 e
~
~
O
DETERMINISTIC REQUIREMENTS 1 AUX SPRAY SYSTEMS MEET THE REQUIREMENTS OF RSB 5-1 FOR CLASS II PLANTS (ACHIEVE COLD SHUTDOWN WITH SAFETY GRADE SYSTEMS) CONSIDER FUNCTIONAL DIFFERENCES BETWEEN SYS 80 AND 3tl10 MW PLANTS
.(1) l-LIMINATED HIGH B0RON CONCENTRATION BAMT'S (2) CONSOLIDATED CVCS SAFEGUARDS FUNCTIONS IN SAFETY INJECTION SYSTEM O (3) INCLUSION OF RCP SEAL INJECTION M
11/05/85 1
~
O
._. .f
~
O O O . PLANT COMPARISON
- AUX SPAY SYSTEM
- COMP 0NENTS SONGS WATERFORD. PVNGS CHARGING PUMPS QUANTITY 3 3 3 TYPE POS. DIS. POS. DIS. POS. DIS.
, CLASS (ASME III) 2 2 2 SEISMIC I I I POWER IE 1E lE i
- AUX SPRAY VALVES j QUANTITY 2 2 2 TYPE MOTOR / MANUAL SOLEN 0ID SOLEN 0ID CLASS (ASME III) 2 2 2 SEISMIC I I I POWER 1E/NA IE lE RCS ISOLATION VALVES QUANTITY 1/ LOOP 1/ LOOP 2/ LOOP-l TYPE MOTOR SOLEN 0ID PNEUMATIC CLASS (ASME III) 2 2 2 SEISMIC I I I POWER 1E 1E 1E l
~
c O . MRILIARY SPRAY O * [ F0Ril0N OF CVCS AUXILIARY SPRAY r-------- 7 '- 1 in . i 1 .T. g y i 8 I LITD(He l CW203 g ,W,K,s,,, FW 5 l l NY NOR 91. O WIG 1tet TOCES
! l , I WILVES I-7t? 10 LIX)P 1A ,,
LITDubde FLOW 9 m ,,9 CH-524 T-729 I
; .N g Rt6D(RATIVE l IKM8GU ELAT p.y cxowd1R 9 a g:5 g n
a fdGHuL CINIGING 10 LDuP 2A I n griposeg Flow tig J. (*5 8
,'L",c, " at5 PALO VERDE -
?' s &
r- -~~-~~---l yy gs I I I p
,, N i
i l W) l T-0129 gcgp g i r, sce,- W g EtN-l 9203 (_ _El 8 [ LE1DMa FLDI - m atLIN<Y KI9AY ppz 10 CVCS 70R110N OF DE CVCS m litrtitRAfiVi L_in LOOP 2A I WATERFORD s Exc,% " E2J I~ g
*
- rRan"' at$emin
.....,,, , .m ,
SONGS bI e 2 4 i
"5 g l 9 x :9:
- =3--D C '
N kb I l i'z% i l r-- ------- 1"22'- ----------J l l 0 N l G I I 13ut. g P2R l _ __ _ _ ' ' " _ _ _ _ ___ _ _.._._ _ _ _ ___ _ _l (=miumV Snv am:0N C OF DE LMS
O 3410 MW CVCS FEATURES o VCT ISOLATION ON SIAS o BAM PUMP DELIVER ON .SIAS o VCT SWITCH 0VER WITH 1E POWER
. .s l
O
. Innsad ,. y RWSP - Rocctor W: tar Ste Pool ,
i BAMP - 5:ric Acid M:kcup Pump
%QrnQ 1%.(4Q BAMT - Boric Acid Makeup Tank
~ "" " *"
, CHARGING PUMP ALIGNMENT NO - Valve Normally Open TO VCT/BAMT/RWSP NC - Valve Normally Closed ' (A) - Emergency Power from Diesel Train A (B) - Emergency Power fron BAMT - - Diesel Train B BAHT VCT
- A B
- This valve does not have (6) SIAS actuation i
I. RWSP Volume 475.000 gal. N HAM () (
\
BAM (/ II. Batching Tank Volume 600 gal. III. Primary Water Storage Tank A g g h IV. Volume 260,000 gal. Batching Tank and Primary Water i Storage Tank can provide water N .N . j inventory to BANT (A) and (B)
- V. Primary Water Storage Tank can be (3) (3) directly aligned to charging pumps 8 AMP (A) BAMP (A) VI. Min. T.S. inventory per RANT
- 1) BAM A B BAM .. CVC 113B 183 Tank 3800 gal. (varies upward 113A with decreasing baron cone.ncracios NC (B) (8) (1) NC . (8) F2)NO f g (1) - Valve opens on SIAS
~
(2) - Valve closes on SIAS (3) - BAMP start on SEAS
,, (4) - Charging Pump start on SIAS
- (5) - Valve opena on low-low VCT BAM (6). * - VCT level ind'ication safety -
133' grade (A) (1)NC _\ . (4) a ? To RCS ONC * (3) . . From RWSP hCVC SQ7
CEN 239 - PROBABILISTIC ANALYSES
) *- -THREE EVENTS:
- 1. LOSS OF SECONDARY HEAT SINK
- 2. SGTR-
- 3. PORV INITIATED LOCA LOSS OF HEAT SINK DOMINATED BY SG HEAT REMOVAL NOT DEPRESSURIZATION l* CONTRIBUTION'0F SGTR TO CORE MELT WAS SHOWN TO BE NEGLIGIBLE PORV LOCA PROBABILITY ESSENTIALLY OFFSET ANY FEED AND BLEED BENEFIT Q . . .
o RESULTS ARE EXTREMELY INSENSITIVE TO AUX SPRAY SYSTEM R IABI!ITY AUX
- SPRAY FAILURE SCENARIOS RESULTED IN EXTREMELY LOW CORE kELT FREQU$NCIES E.G.: SGTR LOOP AUX SPRAY FAILURE STUCK OPEN ADV UNCORRECTED FOR 16 HOURS FREQUENCY LESS THAN ANALYSIS CUT OFF 0F 1 X 10-8 YEAR 11/05/85
($)
- - - - . - - _ _ - _ _ - _ _ _ _ - _ _ _ _ - _ _ _ _ . _ _ _. _ t.
w- " 4 - h CONCLUSIONS o THE AUXILIARY SPRAY SYSTEM PROVIDES THE SAFETY RELATED METHOD FOR DEPRESSURIZATION. THI SYSTEM DESIGN IS CONSISTENT WITH NRC BRANCH TECHN: CAL POSITION RSB 5-1 o PERFORMANCE OF THE AUXILIARY SYSTEM AT PVNGS IS CONSISTENT WITH THE ASSUMPTIONS MADE IN DEPRESSURIZATION ANALYSIS o AUXILIARY SPRAY RELIABILITY HAS A NEGLIGIBLE EFFECT-ON THE~PROBABILISTIC ANALYSIS OF CEN 239 NOT ASSUMED TO BE A HIGHLY RELIABLE SYSTEM LONG TIME AVAILABLE FOR OPERATION
l' i l S CEN 239 l
!s s AUX S' RAY MODEL !
SYSTEM FUNCTION: DEPRESSURIZE T0 SHUTDOWN COOLING SYSTEM PRESSURE i BASED ON SYSTEM SPECIFIC DESIGNS BEFORE ANY RESULTING UPGRADE MODELS EXPLICITLY INCLUDED: EQUIPMENT INSIDE CONTAINMENT CHARGING PUMPS RELATED POWER SOURCES CHARGING PUMP SUCTION LINE-UPS WERE ASSUMED AVAILABLE SEVERAL SOURCES TIME AVAILABLE ( 16 HOURS) 0THER FAILURE MODES (EXPLICITLY MODELED) WERE MORE LIMITING ([)* PZR VENT WAS NOT CREDITED AS BACKUP -- 11/05/85
r .!' PRESSURIZER (19@0 FT 3) ! <
. l 4" SPRAY NOZZLE 6" SAFETY VALVE- m m 3.EEVE NOZZLE I
[ *. 7, SPRAY HEAD ' t ASSEMY 1
.M INSTRUISENT ERE1778 F SHEAR LUG l
96.25" REF. ID l s' 107.48 REF DIA TYP 8' 10 1/2" ? 5 ELECTRIC HEATER i HEATER SUPPORT 5 (35 REQ'D) I , PLATES h TEMPERATURE N g NOZZLE ' ' THERMAL SLEEVE I
= -
1 l "
! SURGE SCREEN L
SUPPORT SKIRT s ) I I Q ::.; k . / .M INSTRUMENT LE i l SURGE IJOZZLE , l y figure 66
, , ,.. --,.,,--.---,.-..,;,,., , c . . , , . . - .
ggy - _g a I -- el p; , ,
=
A C
.mv, Petssua1ZER VAPOS et
~
. WACE SAMPLE W n 34-I V3
'( PREssen ZEs ,
I gr FC i LC v4 m 2 FC J4- -+ C GB I ;,; e s i
;o -
Jo-a 4C h % 2 a c ^ n-- - FC 1 P
- 4 L r
. Q J+- -+ C Qs.
y( SUENCHTANK
, , et J
" te n-e1=
eM6ONOFRE 95LfCLEAR GENERATING STATION Useis 2 Si 3 O - - .t
- Sestet SKE"IM
- 1 ..
%giBTS II.I.1-1
NRR STAFF PRESENTATION TO THE O ACRS l
SUBJECT:
STATUS OF FEED & BLEED UORK IN USI A-45 PROGRAM, INCLUDING FOREIGN POSITIONS. DATE: NOVEMBER 5,1985
~
PRESENTER: ANDREW R. MARCHESE O PRESENTER'S TITLE / BRANCH /DIV: TASK NANAGER, GEi1ERIC ISSUES BRANCH DIVISION OF SAFETY TECHiiOLOGY PRESENTER'S NRC TEL. NO.: (30D 492-4712 SUBCOMMITTEE: ACRS C0f1BINED CO.!1BUSTION EGINEERING/PALO VERDE SUBCOMf1ITTEE fiEETlilG, O
? :
^
~- y p g.+- 1 AP'PR0ACH
- ASSESS VALUE OF. FEED AND BLEED VIA PRA. .
COMPARE WITH.0THER ALTERNATIVES ASSESS PHENOMEN0 LOGICAL.' VIABILITY.'VIA TRAC SIMULATIONS-(LANL).
- LOOK-FOR KEY PARAMETERS (VALVE SIZE, PUMP. HEAD, TIMING, ETC.)
_a v .- - . - - . 2.m.-
1 . Of . C . Q i-
. 1 i-i' ,
i c - STATUS-
- a I
l' - TWO-PLANTS EXAMINED T0.DATE WITH PRA (P0 INT BEACH AND TURKEY' POINT). c . THREE PLANTS YET TO BE EXAMINED WITH PRA (ANO 1, TROJAN, ST. LUCIE).. l t
- FOUR PLANTS-STUDIED.WITH-TRAC (ZION, OCONEE, CALVERT CLIFFS,.
H.B. ROBINSON). - i - EXTENSIONS TO OTHER PLANTS PROVIDED. i . i i i J i
- _ - 1' lOL _O iO; _
I w PRELIMINARY.'RESULTS FOR POINT BEACH PLANT. LIMITED PRA-(TRANSIENTS /SMALL LOCA) INTERNAL EVENTS ONLY POINT BEACH - BASE CASE ASSUMES FEED AND BLEED AVAILABLE (PLANT HAS PROCEDURES IN. PLACE)
- P(CM) = 1.76E-4/R-YR
-IF: FEED AND BLEED IS' NOT.. AVAILABLE
- P(CM) = 4.05E-4/R-YR THEREFORE FEED AND: BLEED REDUCES P(CM) BY-8 =.2.29E-4/R-YR
- - - - - - - g ,, __ . 4 _ s
- , c PRELIMINARY RESULTS FOR TURKEY POINT PLANT TURKEY POINT - BASE CASE DOES NOT INCLUDE A FEED AND BLEED CAPABILITY (PLANT DOES NOT HAVE PROCEDURES)
- P(CM) = 4.98E-4/R-YR IF FEED AND BLEED WERE AVAILABLE
- P(CM) = 8 46E-5/R-YR FEED AND BLEED REDUCES P(CM) BY li = 4.13E-4/R-YR
'g"2,
=
r A , 4, 1
- RESULTS OF~PHENOMEN0 LOGICAL STUDIES 4
1 i
- FEED AND BLEED' GENERALLY SUCCESSFUL IF INITIATED PRIOR-T0 STEAM GENERATOR'
- DRY OUT.
i
- TIME TO DRY OUT IS PLANT SPECIFIC AND. SENSITIVE TO TIME OF-REACTOR SCRAM (FOLLOWING. TIMES RESULT
- FOR SCRAMS L< 20 SEC INTO ACCIDENT): ,
- OCONEE rse 8! . MIN, i --
CALVERT CLIFFS ->20-MIN. j - ZION A'40 MIN, H B ROBINSON ^340 MIN. ! FEED 'AND BLEED' SUCCESSFUL IN TW0. PLANTS.(0CONEE AND ZION) IF INITIATED NO LATER THAN TIME 0F PRIMARY SYSTEM SATURATION, l l
, ~ , .
O - dt .d .
] -
i ' I .. i .
. CONCLUSIONS TO DATE I
j .a 1- . i
'A FEED'AND-BLIED CAPABILITY DOES PROVIDE SOME RISK REDUCTION,
~
i - ! TIMING 0F FEED AND BLEED INITIATION-.IS CRITICAL.(FOR SOME PLANTS DECISIONS MUST BE MADE LESS THAN 20 MINUTES INTO AN INCIDENT).- . l i i f l . I 1 t ). ] I f' 1
r ,
,,-- ,sn C) CI Q ,
FOREIGN POSITIONS ON FEED AND BLEED SOURCES OF INFORMATION
-s -1. " REGULATORY POSITION. RELATIVE TO PRIMARY SYSTEM. FEED AND BLEED IN OTHER COUNTRIES,"' ~
MEMORANDUM FOR ROGER J. MATTSON FROM-BRI AN W. SHERON, MAY 22, 1983. +
- 2. MEETING WITH ITALIANS ON THEIR NEW GDCs ON JANUARY 30, 1984.
- 3. DECAY HEAT REMOVAL SYSTEMS, PROCEEDINGS OF.THE CSNI SPECIALIST MEETING, APRIL 25-29, 1983,.WURENLINGEN, SWITZERLAND.
- 4. TRIP REPORT - CSNI SPECIALIST MEETING ON DECAY HEAT REMOVAL SYSTEMS, WURENLINGEN, SWITZERLAND, APRIL 25-29, 1983, MEMORANDUM TO KARL KNIEL FROM ANDREW'MARCHESE, JULY 14, 1983.
- 5. VISIT TO SITES IN WESTERN EUROPE WITH NRC TECHNICAL TEAM FOR USI PROGRAMS A-45 AND A-29, TRIP REPORT, D. WARD, DECEMBER 4, 1984.
- 6. TRIP REPORT - FOREIGN TRAVEL IN SUPPORT OF USI A-45 AND GENERIC ISSUE A-29 PROGRAMS, MEMORANDUM TO KARL KNIEL FROM ANDREW MARCHESE, JANUARY 15, 1985.
P
-O; - 10 -
&L
-/
SilMMARY OF FOREIGN APPROACH TO FEED AND BLEED GENERALLY, NOT IN FAVOR OF USE OF FEED AND BLEED. SOME VARIATION FROM COUNTRY T0' COUNTRY. LAST DITCH MEANS FOR EMERGENCY-DHR. PRIMARY CONCERNS: PORV AND RELIEF VALVE OPERATION. PROMPT AND ACCURATE OPERATOR ACTION REQUIRED. ADVERSE ENVIRONMENT CREATED INSIDE CONTAINMENT.
, .n t
I . t WEST GERMANY NOT DEALT WITH AS A LICENSING ISSUE. BASED ON RELIABLE AFW SYSTEM AND SPECIAL EMERGENCY SYSTEMS,-FEED-AND-BLEED IS NOT ONLY UNNECESSARY BUT UNDESIRABLE. FOR RAPID PLANT DEPRESSURIZATION, PREFER TO USE RAPID l SECONDARY-SIDE C00LDOWN THOUGH ADVs COUPLED WITH USE OF SAFETY GRADE SPRAY SYSTEM. CONVOY DESIGN HAS SMALL RELIEF VALVES, INSUFFICIENT FOR FEED AND BLEED. RECENT CONSIDERATION GIVEN T01 FEED AND. BLEED FOR SEVERE ACCIDENT POLICY. l l l
. ~ . . ,.. .. - .- - - --. . .- . - .- . - ..- . .
L i l' t SWITZERLAND ', NO PLAN TO USE FEED AND BLEED MODE OF DHR. i
- PREFER RELIANCE ON STEAM GENERATORS-AND BUNKERED SPECIAL. EMERGENCY
- SYSTEMS, .
-FOR RAPID PLANT DEPRESSURIZATION, PREFER TO USE RAPID SECONDARY SIDE .
- C00LDOWN THROUGH ADVs COUPLED WITH USE.0F SAFETY GRADE SPRAY SYSTEM.
1 l PRIMARY CONCERNS: l ADVERSE' ENVIRONMENT.
- AFFECT ON: ESSENTIAL-OPERATING EQUIPMENT, .
i l ! 1 1
}
q
) , .-. - -
.-sw a
O O O. . BELGIUM EVALUATING FEED AND BLEED FOR OLDER PLANTS, FOR NEWER PLANTS, PORVs, STEAM DUMP VALVES AND AUXILIARY SPRAY SYSTEM ARE SAFETY GRADE. PREFER RELIANCE ON STEAM GENERATOR AND BUNKERED FEEDWATER SYSTEMS. I ALSO SYSTEM REDUNDANCY AND DIVERSITY IS GREATER THAN IN U.S. PLANTS,
O O f1 THE NETHERLANDS FINAL POSITION ON FEED AND BLEED HAS NOT BEEN MADE; HOWEVER, THEY DO NOT THINK FEED AND BLEED IS A GOOD APPROACH TO EMERGENCY DHR, PRIMARY CONCERNS: ADVERSE ENVIRONMENT.
- INSTRUMENTATION AND CONTROL,
- PORV OPERABILITY, m_ _ _ _ _
?O; O !O' .
' ITALY ENEA EVALUATING EFFECTIVENESS OF FEED'A'ND BLEED-FOR~ ADVANCED
-PWR DESIGN.(PUN) AS AN EMERGENCY MEANS OF DHR.
PREFER USE OF STEAM GENERATORS. PRIMARY. CONCERNS WITH FEED'AND BLEED: TIMELY AND< ACCURATE OPERATOR ACTION REQUIRED. OPERATION OF INSTRUMENTATION IN ADVERSE ENVIRONMENT.
+
t f
O O 10 ? . s FRANCE FEED AND BLEED HAS.BEEN ANALYZED AS A DIVERSE MEANS FOR DHR. OPERATING PROCEDURE IS AVAILABLE FOR USING FEED AND BLEED IN PWRs.
~
IMPLEMENTATION 0F PROCEDURE UNPROVEN:
. PROCEDURE DOES-NOT ASSURE SUCCESS.
OPERATOR TRAINING-ESSENTIAL. 9
O O O ENGLAND SIZEWELL B DESIGN - FEED AND BLEED IS A DIVERSE METHOD FOR DHR. NO CREDIT GIVEN IN SIZEWELL B SAR FOR FEED & BLEED. SIZEWELL B OWNERS MUST ADOPT OPERATING PROCEDURE FOR FEED AND BLEED.
, .x : .;
Ol a;
'O
~
1 'OL 1 ( w JAPAN FEEDANDBLEEDNOTREQUIRED, BASED'.0N;RELIABIOTYOFAFWSYSTEMS. CONFIDENT OF STEAM GENERATOR TUBE INTEGRITY. ANNUAL INSPECTION OF-STEAM GENERATOR TUBES.
a. STAFF POSITION e REANALYSIS REQUIRED TO SUPPORT LICENSING BASIS e SGTR EVENT SHOULD BE REANALAYZED ASSUMING WORST SINGLE ACTIVE FAILURE.AND USING ONLY SAFETY GRADE SYSTEMS AND COMPONENTS FOR RCS DEPRESSURIZATION FOLLOWING THE EVENT e . SYSTEMS ~ ENHANCEMENTS AND VERIFICATION REQUIRED e PROVIDE POWER TO CH-501 AND 536 FROM AN IE MOTOR CONTROL CENTER FOLLOWING A LOSS-0F-0FFSITE POWER AND/0R A SAFETY INJECTION ACTUATION SIGNAL e ENHANCE THE RELIABILITY OF THE VOLUME CONTROL TANK (VCT) LEVEL INDICATION e PROVIDE AUTOMATIC REALIGNMENT OF CH-501 AND 536 ON LO-LO VCT LEVEL AND LOSS-0F-0FFSITE POWER, TO ALIGN CHARGING PUMP TO TAKE SUCTION FROM THE REFUELING WATER TANK s O
~'t'_
A v STAFF-POSITION (CONT'D) .
+
e- MAINTAIN THE' VALVE HV 532 AT -A LOCKED OPEN POSITION t e . VERIFY THAT-THE WATER VOLUME IN THE-RWT AB0VE THE UPPER CONNECTION OF WATER. SUPPLY LINE HAS SUFFICIENT INVENTORY TO SATISFY THE AMOUNT OF WATER REQUIRED FOR THE APSS~ e- REMOVE POWER AND LOCK OPEN VALVE 524 TO PREVENT A SPURIOUS VALVE MOTION'OR OPERATOR ERROR FROM DEFEATING THE FUNCTION OF THE AUXILIARY PRESSURIZER SPRAY SYSTEM-e VERIFY.THE PALO VERDE EQUIPMENT QUALIFICATION PROGRAM TO ENSURE-THE COMPONENTS ASSOCIATED IN THE FLOW PATH OF APSS ARE PROPERLY ENVIRONMENTALLY QUALIFIED e PROPOSE i.PPROPRIATE TECHNICAL SPECIFICATIONS TO ENSURE OPERABILITY OF ALL SYSTEM COMP 0NENTS OF THE APSS INCLUDING ITS WATER SOURCES O
~ . , . i
. 'A-AUXILIARY PRESSURIZER SPRAY SYSTEM b
.e - APSS FUNCTIONS IMPORTANT TO SAFETY-e 'APSS'IS REQUIRED TO FUNCTION FOR PLANT SHUTDOWN PER THE 4 ui REQUIREMENTS OF BTP RSB 5-1 e APSS IS REQUIRED TO FUNCTION FOR MITIGATION OF A POSTULATED SGTR EVENT e APSS FUNCTION WAS FACTORED IN.THE STAFF'S PORV STUDY
'p~' ' e SAFETY CLASSIFICATION OF APSS e THE CURRENT APSS DESIGN MEETS BTP RSB 5-1 REQUIREMENTS FOR CLASS 2 PLANT IMPLEMENTATION (D0ES NOT REQUIRE . SYSTEM TO BE COMPLETELY SAFETY GRADE) e THE DESIGN OF THE APSS WATER SUPPLY SYSTEM DOES NOT MEET THE SAFETY GRADE REQUIREMENTS o
g . e APSS WATER SUPPLY SYSTEMS VULNERABILITY o VOLUME CONTROLLTANK (VCT) LEVEL INSTRUMENTS ARE VULNERABLE TO SINGLE FAILURE
- s. VCT ISOLATION VALVE UV-501 IS VULNERABLE TO SINGLE FAILURE e WATER SUPPLY LINE FROM THE REFUELING WATER TANK (RWT)
VIA VALVE HV 536 IS VULNERABLE TO SINGLE FAILURE e GRAVITY WATER SUPPLY-LINE VALVE NO. 521, AND CONTAINMENT (']) ISOLATION VALVE HV 532 ARE VULNERABLE TO SINGLE FAILURE I e POWER SUPPLIES TO VCT LEVEL INSTRUMENTS AND VALVES UV-501 AND HV 536 ARE NOT RELIABLE e CHARGING PUMPS ARE VULNERABLE TO GAS B0UNDING DUE TO A FAILURE OF VCT ISOLATION
N
- VALVE POWER SUPPLY CHANGE 1
EXISTING MODIFIED lE BUS IE BUS
) SIAS TRIP ON
)
NON lE MCC IE MCC I I) I) i) i)
++
501 co 536 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 e O O
([) LICENSING BASIS FOR PALO VERDE UNIT 2 RESOLUTION OF THREE REMAINING ISSUES
~
POST ACCIDENT SAMPLING SYSTEM ECCS REANALYSIS PRESSURIZER AUXILIARY SPRAY SYSTEM , CERTIFICATION BY APPLICANT THAT DESIGN, CONSTRUCTION AND TESTING 0F UNIT 2 HAVE BEEN COMPLETED IN CONFORMANCE WITH FSAR AND OTHER DOCKETED COMMITMENTS, O COMPLETION OF POWER ASCENSION TEST PROGRAM FOR UNIT 1 PRIOR TO INITIAL CRITICALITY OF UNIT 2, i l I i
. i e b COMPENSATORY MEASURES FOR RESTART OF PALO VERDE UNIT 1 FOLLOWING 9/17/85 CONFIRMATORY ACTION LETTER i* DAILY MON'ITORING OF REFERENCE LEG,FOR VCT LEVEL INDICATOR , REVISED PROCEDURES TO ASSURE ALIGNMENT OF RWT TO CHARGING PUMP SUCTION UPON LOSS OF 0FFSITE POWER PROCEDURAL CAUTIONS ON RESTART OF CHARGING PUMPS ASSURE POST-TRIP REVIEW PROCESS EVALUATES EVENTS WITH RESPECT TO POTENTIAL SAFETY SIGNIFICANCE ASSURE APPROPRIATE VENDOR REVIEWS HAVE BEEN MADE FOR REMAINING POWER ASCENSION TESTS REVIEW SHIFT COMPLIMENT FOR REMAINING POWER ASCENSION TESTS , O
O H.B. thrt i POWER ASCENSION TESTING EXPERIENCE
- 1. GENERAL OPERATING PERFORMANCE
- 2. EXPERIENCE WITH POWER RUNBACK SYSTEM LOAD REJECTION TEST EXPERIENCE - ROOT CAUSE OF PROBLEMS EXPERIENCED O -
l l l O
+ l
1
.O
- DEVELORENT OF PAT PR0(RAM SCEDULE ACTUAL TEST PROGRAM DLRATION AT OTWR T PUWTS l
SCDPE OF TESTING FOR PVNGS vs. OTER CE PLAN 1B l ACTUAL TEST TDE TEST PREDICTIONS vs. TEST RESULTS VERY 0000 AGREEENT BETWEEN PREDICTION /RESlLTS O
- N0 LEXPLAIED TEST RESULTS OR PLANT TRANSIENT RESPONSE
~^
PLANT TRIPS: 7 TRIPS (TWO WiILE TESTING) TWO MAIN FEED PitP TRIPS OE CEAC CIRCUIT BOARD FAIllRE l TIREE LOSS OF PGER TRIPS (ONE MfILE lESTING) OE S/G L(W LEVELS TRIP (WHILE TESTING) LO JGH 11/05/85
1
-- -. - -~ >
u$' (D TY eg em I M >
~
e O, ,
. 1 4
= . O l _
.o4
-o "a#
- e- ,
. % E '
. 4 '
W i
> ~~, s-
, e . o F-5 e i M_
. ph
- g. . -
ul O #'s v-O
- o -
l f
- +-e P
g g Y
. . . n h.
6
. < $ = - /g
Unit 1/ Unit 2 Schedules 100% 80% 50% M 30% ~
~~
LPPT UNIT 2 PCHPT
.RX ASSBLY FUEL i LOAD i
' 10 R.
80%
, PWy gygy y ! s a s , , , , , , ,
totes 11/as 12/as itse n/se 3/se 4/se s/ss else rise g g g TG H
\\/o5lS5
t O REACTOR POWER CUTBACK SYSTEM- +
- v. :
A CONTROL GRADE SYSTEM TO EfiHANCE THE UNITS AVAILABILITY. [
- j. -
THE SYSTEM IS DESIGNED TO OPERATE Ill C0flJUNCTION WITH THE STEAM { BYPASS CONTROL SYSTEM TO DROP SELECTED CEA'S FOR: i A LOSS OF FEED PUMP i A LOAD REJECTION - TO MATCH STEAf1 BYPASS CAPABILITY WITH REACTOR POWER. 1 GWW 11/05/85 t o s O
,w- - - - . , - - ,,-.<->,,.L4 , ,.,.. - - - - . ,,,--a,-, ,_,, ,-.,e..,--._,-.--.---w--w--,,. - - - . . . - . . .w.-,...- - . - , , . . , . ~ . . - - , - - . _ , -
^
;/>s - EXPERIENCE WITH REACTOR 2's 30WEF, CUTBACK SYSTEM
*- TEST PERFORMED
-LOSS OF FEED PUFP BELOW 50% - SUCCESSFUL N0 REACTOR POWER CUTBACK SECOND PUMP. ASSUMED LOAD LOSS OF FEED PUMP AT 70% ' SUCCESSFUL
. TURBINE SETBACK
-N0 REACTOR POWER CUTBACK LOSS OF LOAD FROM 50% - INITIAL TEST : UNSUCCESSFUL
- SECOND TEST : SUCCESSFUL ...,,
NO REACTOR POWER' CUTBACK { ): , LOSS OF LOAD' FROM 80% - INITIAL TEST : UNSUCCESSFUL REACTOR POWER CUTBACK ACTUATION
~
TURBINE SETBACK TESTS TO BE PERFORMED RE-PERFORM LOSS OF LOAD FROM 80% LOSS OF FEED PUMP FROM 100% LOSS OF LOAD FROM 100% i GWW 11/05/85 ; 1
SUMMARY
OF OCTOBER 24, 1985 EVENT j ]) PLANT WAS PREPARING FOR THE LOSS OF LOAD TEST FROM 80% POWER PLANT INITIAL C0fiDITI0flS 83% REACTOR POWER [ TURBINE /GEliERATOR C0flNECTED TO GRID i ALL STATI0fi LOADS BEING FED FROM 0FFSITE POWER SOURCES ANTICIPATED RESULTS 0F TEST NO REACTOR TRIP RPCS ACTUATES TO REDUCE POWER ACTUAL TEST RESULTS ^ 1(() - REACTOR TRIP CONCURRENT WITH TURBINE TRIP ' REACTOR POWER CUTBACK DID NOT OCCUR DUE TO REACTOR TRIP ALL SAFETY SYSTEMS PERFORM PER DESIGN 4 GWW 11/05/85 - j 9 9
-, e . , - - - , - , , - - - - , , - - , - - - - -,- +-,,-e-e ,, w + -~ em - , - , - ~ - , - - - , - -
-O MORW FUM 10-24-85 M
i 1010.00--l-TWBIE/ GEN. TRIPPED REACTOR TRIP
-l-STEAM BYPASS QJICK OPEN 1010.30 - -SIAS/CIAS/MSIS 1011.00 - -2 RCP's TRIPPED 1012.00 - --tPSI INJECTION FLO4 STOPPED O --
1013.00 - --ATMOSPERIC VALVES OPEED 1 1015.00 - --NON-SEI&iIC AFWS P&P STARTED 1046.00 - --RESET SIASICIAS 1
- 5. -
--RESET MSIS O
I I 6M 11/05/85
.- i 6-l ROOT CAUSE EVALUATION f)))
OCTOBER 24, 1985 EVENT REACTOR TRIP CAUSE - SG LOW LEVEL SIGNAL T00 FAST TO ACCOMM0DATE PRESSURE WAVE CORRECTIVE MEASURE - INCREASE REACTOR TRIP BISTABLE RESPONSE TIME FOR SG LOW LEVEL C00LD0WN'0F PRIMARY SYSTEM CAUSE - INCORRECT SETP0 INT IN STEAM BYPASS CONTROL SYSTEM ($) CORRECTIVE MEASQRE - ADJUSTED SETP0 INT IN THE STEAM BYPASS CONTROL -
- REVERIFICATION OF ALL SETP0INTS IN SBCS, FWCS, RRS GWW 11/05/85
-($)
,e--
4 ,,w. --, , , , , , , , - , , , , - - ,-,,--r, ,m--,-,- - m--, ,_,-,,---n-- -
.-O II.C. AUXILIARY PRESSURIZER SPRAY SYSTEM (APSS)
- 1. APSS DESIGN i
- 2. APSS OPERATIONS, SEPTEMBER 12. 1985 EVENT Root CAUSE EVALUATION
- 3. APSS DESIGN ENHANCEMENT
- 4. SGB ANALYSES O .
3 e-I l O l
O U/T1TARY FRESSLRT7FR SMY PLRPOSE: PROVIDE A WANS FOR REACTOR (DOUNT SYSTEM DEIRESSLRIZATION, MEN l'AIN PRESSLRIZER SPRAYS ARE NOT AVAll)BLE ESTABLISKD DESIGN GITERIA: BR!Eri TEGNICAL POSITION RSB 5-1 FOR A OJSS 2 PLAT
- OERATOR ACTIONS OUTSIDE CONTROL R001
- OERATOR ACTIONS INSIDE CONTAIMiENT AFTER SSE
- R:.%IKING AT HOT STAKBY LNTIL MAMJAL ACTIONS OR REPAIRS ARE CORETED O
TF0 11/05/85'
O O O . 9 SIffLIFIED SOD % TIC 0F PALO VERDE CVCS , Sl[NING AljXILIARY SPRAY PORTION NftlLleAf Sf*AT
,_[f(Alle
_ _ _ _ _t of CWS l _. - gros d f');, tyg. . 2a m .l _$g_m_y wu l
- t. .w _ _ _ _ _ _ s l c,.-m ,.,,,
l 'aida" : .d ; *[ @ N"'"Nq m j ><,.m. ja A
} us. b re% he" i I l O_ASSIFICATIONOFi - _l l ASME SEI ENV1H0ffENTAL !
@P0ENTS 000E CATEGORY IE PWER WALIFICATION CH-529 III, 0. ASS 2 I l A YES
.CH-239 III, CLASS 2 I NON-IE YES I l CH-240 III CLASS 1 I NON.-JE YES l CH-203 IIII,CLAssi I B YES ,
CH-205 III, CLASS 1 jI A YES
. .1 O .
. O O .
SIW LIFIED SCE MATIC 0F PALO VERDE CVCS *- SHOWING SOLECES OF BORATED WATER FOR AUXILIARY SFRAY - q Tiest le. tlc ACIO mMesseet Imat
~
EnstlC K lO w.: k MIO4WG Titift 11stl81CAllt8t l
\ Et4Eleft " "'## g -)
g_ g ' Itt:(LiseG 08W'F*1883 Mir 3 tie c,. or.I, O lE'#5'E ner i b-N-- k-N-- -
- g. F h CH-530 ,
6
, , , , , 3 oi-se, ca.sm u nr.us ner V "l%"'" .
f nec I >< - c,,. N iss "o r-rir r-rfr g ,' a,. n uc una O-3rH414 >+ %
' S** E El IN ()
, - po- N a b -N-b4 (-]) :}: stt usat u, ,,c,,,. :
r e.t g omnr. ner 1 80 IICP SEAf. lHECflGI asAl [ NOWS 44 I I l O_ASSIFICATION OF M10R OW0ENTS - ASE SEISMIC I ENVIR0lMNTAL GPF0ENTS nmE CATEG(RY .IE PGER QJAl TFICATION I CHARGING PUMPS (3) III. CLASS 2 I A/B YES ICH-501 l'III CLASS 21 I I A are.s-1 . CH-536 . III, CLASS 3 I A nTe.s-1 , 1 i i i l
( s, SlM%RYOF SEPTEl1ER 12,1985 EVENT PREPARING TO PERFORM A LOAD REJECTI0fl TEST
' INITIAL PLANT C0flDITI0flS REACTOR AT 53% P0tER TURBItE/GEfERATOR C0flIECTED T0. GRID N0fi-CLASS STATION LOADS Oil GEfERATOR TEST STARTED BY IllITIATING LARGE LOAD REJECTI0ft I
l
- EXPECTED RESLLTS OF TEST NO REACTOR P0tER CUTBACK SYSTEM ACTUATI0ll O
- NO REACTOR TRIP - . ,
ACTUAL RESlLTS OF TEST fl0 REACTOR POWER CtRBACK SYSTEM ACTUATION REACTOR TRIP OCCURRED L , N
. 11/05/85 O
.t A REACTOR TRIP #4
- v 09-12-85 L
r IlE. 2208.29 OPEN GENERATOR BREAKER 918 2208.53 -- REACTOR TRIP (RCP SPEED 90%) 2210.00 OPERATOR CLOSES MAIN STEAM LINE DRAINS 2210.53 -- LOST STATION LOADS 2213.00 NATURAL CIRCULATION ESTABLISHED 2215.00 -- STARTS NON-SEISMIC AFWS PUMP ! 2218.00 -- RESTORED STATION LOADS , l -- MAIN STEAM LINE DRAIN VALUES RE0 PEN 2222.00 SIAS/CIAS '-O .2224.00 neS1 eLOW inaEC110n ST0eS l 2230.00 -- OPERATOR SECURES ALL 3 CHARGING PUMPS i 2232.00 LINED UP GRAVITY FEED 0000.00 -- STARTED BAM PUMPS 0002.00 STARTED CHARGING PUMPS B AND C RESULT SIAS 8, CIAS 0102.00 -- RESTART 2 RCP'S l GWW 11/05/85 O
i (~T ROOT CAUSE EVALUATION s.> - . SEPTEMBER-12,1985 C00LDOWN 0F PRIMARY SYSTEM
.CAUSE
*' RE0PEtlING 0F T1AIN-STEAM LINE DRAIN VALVES CORRECTIVE MEASURES .
REMOVE AUTO 0PEllING 0F VALVES ON TURBINE TRIP LOSS ~0F. POWER T0 fl0ft-ESSENTIAL STATION LOADS CAUSE: INABILITY OF THE TURBINE / GENERATOR TO MAINTAIN HOUSE LOADS CORRECTIVE MEASURES l[]
- MODIFY THE TURBINE SUPERVISORY SYSTEM
[10DIFY TURBINE ELECTR0-HYDRAULIC CONTROL SYSTEf1 LOSS 0F CHARGING FLOW CAUSE INACCURATE VCT LEVEL INDICATION CORRECTIVE MEASURES DAILY SURVEILANCE OF REFERENCE LEG REVISE PROCEDURES TO LINEUP CHARGIllG PUMP SUCTI0ft MODIFY PROCEDURES TO WARN OPERATORS OF POTENTIAL LOSS OF SUCTION TO THE CHARGING PUMPS VARIOUS ADMItlISTRATIVE ENHANCEMENTS DESIGN EllHANCEllENTS O GWW 11/05/85
MODIFICATION
SUMMARY
A,. j .
- 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 S,IAS i
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 llI*~ LOCK OPEN TWO NORMALLY OPEN VALVES IN AUXILIARY SPRAY FLOW PATH
" ELIMINATES P0TENTIAL FOR ISOLATION OF GRAVITY FEED OR CHARGING LINE DUE TO SPURIOUS ACTUATION OR OPERATOR ERROR MAR 11/05/85 4
0
,,r _,. ._ .. _,
- - - - , ..n- - . , , . v-a - ., .,- - - - , . - - , , , , _ , , , , , . . . , , _ . _ _ _ . _ , , , , , _ . , . . , , ,- ,
O STEAM GEERA10R 11BE RUPTIRE ANALYSIS EXPLAIN THE USE OF AUXILIARY PRESSURIZER SPRAY IN THE PALO VERDE STEAM GENERATOR TUBE RUPTURE ANALYSIS DEMONSTRATE THAT THE DOSE CONSEQUENCES FOR THE PALO VERDE SGTR ANALYSIS ARE RELATIVELY INSENSITIVE TO THE TIME AUXILIARY PRESSURIZER SPRAY IS INITIATED SHOW THAT AN ALTERNATIVE SYSTEM EXISTS TO FACILITATE DEPRESSURIZATION FOLLOWING A STEAM GENERATOR TUBE RUPTURE '-C s 4 9 O as 11/05/85
.-7} w ~ ANALYSIS
SUMMARY
.(1) -
PVNGS. FSAR' APPENDIX.15A
' -STEAM GENERATOR.TUEE RUPTURE
-* LOSS OF 0FFSITE POWER RECOGNIZED EMERGENCY. OPERATING GUIDELINES FULLY STUCK 0 PEN ATMOSPHERIC DUMP VALVE-AUXILIARY PRESSURIZER SPRAY OPERATION AT I.0 f(2) . OCTOBER 15, 1985 SUBMITTAL SAME AS -(1) EXCEPT:
AUXILIARY PRESSURIZER SPRAY 0PERATION AT TW
.(3) NOVEMBER 4, 1985 SUBMITTAL
^
-SAME AS (2) EXCEPT:
NO AUXILIARY PRESSURIZER SPRAY PRESSURIZER VENT OPERATION AT TL'O HOURS. GWS
/Q .
11/05/85 S 4, f I- _O
,---....,~.,,m., .,,r.,r.._- ,,,,., ..,,,_-t ,_ 4. _ . . , __y,.. ._mm.,.,,,m_..-a_,.,, , , , , ..-,my,. y.-.m._. , , , . _ ~ ,,_.,._y._ , r -- . . . , .
l COMPARISON OF THE RCS PRESSURE RESPONSE FOLLOWING A SGTR
)
O WITH AUX SPRAY INITIATED AT 1015 SECONDS vs 2 HOURS i 2500 i i i i i OPERATOR TAKES CONTROL OF PLANT-0 PENS ONE ADV IN l l EACH SG . 2000 - l PERATOR INITIATES AUXILIARY SPRAY
<c l m -. /
- m. 1500 f \ ~
OPERATOR BEGINS TO THROTTLE HPSI FLOW i W i i 8 1 ' 1000 g O ,0 \
\
\ RCS REACHES SHUTDOWN
\
l 500 - COOLING ENTRY CONDITIONS 0 I I I I I 0 5.000 10000 15000 2000 25000 30000 TIME, SECONDS SGTR WITH AUX SPRAY AT 1015 SECONDS
---SGTR WITH AUX SPRAY DELAYED UNTIL 2 H00RS i
O STEAM G2NERATon war. RUPTURE WITH I4SS OF OTTSITE POWER AND A FULI,Y SWCK CFIN ATHOSPHERIC' DUMP VALVE
t PRE-ACCIDENT 10 DINE SPIKE 7 THYR 0ID DOSE-REM . 2 HOUR 10SES 3 HOUR DOSE , EXCLUS101 A RA LOW FOPULA110N B0UNDARY ZONE :
- , (REM) (REM) l, c .
l i PVilGS FSAR 200 41 ! APPEllDIX 15A
- AUX-SPRAY AT '
1015 SECONDS AUX-SPRAY AT 208 44 TWO HOURS . i 10CFR100 DOSE LIMIT IS 300 REM ! O' i i i GWS l 11/05/85 i L f O i 4
, -..-...,,.n --- . - - .- - . . -.-..-, . -.-,.,.,,-- -,-n - - - . _ - - . - , - - . - - - , - - . - ~ _ , , , , - - - -~ -- -, .. -. -, -
s O mssmme am A som DEPRESSURIZATION RESULTS PRIMARILY BY THROTTLING IPSI FLOW i BEFORE f9SI FLOW IS THROTTLED LEVEL IN THE PRESSURIZER MUST BE RECOVERED f l. i AUXILIARY PRESSURIZER SPRAY IS USED TO RESTORE LEVEL LO i as 11/05/85
PRESSLRIZER VEN1 PATHS o
'd C,o4 TAl d MENT ATM os PME.
Ak ei gc-Ioci K'c - 108 D 6 W M o 1" o FJKClob '
@ (x j" f' D' y W F4\05 g.s Y2 ys [
rm i FC lo3, if KDT N O ME%d8lZE R I I CLASSIFICATION OF MAJOR CUf0NEtlTS ASME SEISMIC l Et#JRCeliENTAL 0]MPONENTS WDE CATEGORY IE POWER CUALIFICATION RC-103 , III. CLASS 2 I A l YES I l l RC-105 I III. CLASS 21 I B , YES I RC-106 IIII. CLASS 21 I I A . YES < l l l 1 ' IRC-108 III. CLASS 11 I 8 YES l l l 1 1 RC-109 III. CLASS i I B YES l i 11/05/85 _ _
O COMPARISON OF THE RCS PRESSURE RESPONSE FOLLOWING A SGTR WITH
. AUX SPRAY INITIATED AT 2 HOURS vs PRESSURE VENT OPENING AT 2 HOURS 2500 i i i i i OPERATOR TAKES CONTROL OF PLANT-0 PENS ONE EACH SG 2000 -
\ OPERATOR I ITIATES DEPRESSURIZATION PERATOR BEGINS TO THROTTLE HPSI FLOW.
1500
-f _
$ h
$ 1000 -
', s -
m . E
's./'s RCS REACHES ~
% SHUTDOWN COOLING 500 ENTRY CONDITIONS 14, _
\
O I I I I I 0 5,000 10000 15000 20000 25000 30C00 TI?E, SECONDS SGTR WITH AUX SPRAY AT 2 HOURS O ---SGTR WITH PRESSURIZER VENT AT 2 HOURS STEAM GENERATOR TURE RUPTURE WITH IDSS OF OFFSITE POWER AND A FUflY STUCK OPEN ATMOSPHERIC DUMP VALVE
's PRE-ACCIDENT I0 DINE SPIKE
-(G THYROID DOSE-REM 2 HOUR JOSES 8 HOUR DOSE EXCLUSIO'4 A<EA LOW POPUi_A110fl B0UtlDARY Z0fiE (REM) (REM)
PVilGS FSAR 200 41 APPEf1 DIX-15A AUX-SPRAY AT 1015 SEC0flDS AUX-SPRAY AT 208 44
- TWO HOURS O
PRESSURIZER 208 44 VENT AT TWO HOURS 10CFR100 DOSE LIMIT IS 300 REM GWS 11/05/85 O
O SlftMRY ; DOSE RESULTS ARE RELATIVELY INSENSITIVE TO THE TIME OF AUXILIARY PRESSURIZER SPRAY ACTUATION DURING THE FIRST TWO HOURS 10CFRiOO DOSE LIMITS ARE NOT EXCEEDED WHEN APSS USE IS DELAYED UNTIL TWO HOURS AFTER THE EVENT INITIATION AN ALTERNATE SYSTEM EXISTS THAT CAN BE USED TO FACILITATE DEPRESSURIZATION DURING A STEAM GENERATOR TUBE RUPTURE, WITH NO IMPACT ON THE DOSE CONSEQUENCES.
']
i O se 11/05/85
.- I (1 AUXILIARY PRESSURIZER SPRAY
SUMMARY
v
~
S THE AUXILIARY PRESSURIZER SPRAY SYSTEM DESIGN MEETS THE REQUIREMENTS OF BTP'RSB 5-1. THE OPERATORS SUCCESSFULLY DEALT WITH THE LOSS OF CHARGING IN THE SEPTEMBER 12 EVENT. COMPENSATORY MEASURES HAVE BEEN TAKEN TO REDUCE THE PROBABILITY OF A TEMPORARY LOSS OF CHARGING. DESIGN ENHANCEMENTS ARE BEING MADE TO
. REPLACE SOME OF THE COMPENSATORY MEASURES.
ADDITIONAL STEAM GENERATOR TUBE RUPTURE ANALYSES CLARIFY THE REQUIREMENT FOR USE OF THE AUXILIARY PRESSURIZER SPRAY, OR THE BACKUP VENT, TO MINIMIZE OFFSITE DOSE. O GWS 11/05/85 ()
,, -, ~ ,c. ,
O ILD. OFFsITE POWER RELIABILITY
- 1. OCTOBER 3 AND OCTOBER 7, 1985 EVENTS Roor CAUSE EVALUATION
- 2. SWITCHYARD To POWER Block INTERFACE O
i i l l O .
,, , - , - , - - . . - - - - - , , , - , , - , , - - , - - , - - - - - - - - - - - - , , . - . - - , - - , - - - ---- - , - - , - - n-,-- a , - ,--- -
O REACT (R 1 RIP #5 (LOP) 10-03-85 i IRE ll 16Ll3.00-- --LOSS OF P(MER 1641.00-- --REACTOR TRIP 1641.10 - --DG STARTS #D LOADS 1641.15 - --MAIN STEAM SAtt.uts OPEN 16tl7.00--l-ADV's OPEN 16Ll7.10- --MSSV RESEAT !' Q .. 1710.00 - --0FF-SITE P(MER RE-ESTABLISKD i 1 2018.00- --RCP's RESTARTED l l l O su 11/05/85 i-
0CTOBER 3, 1985 EVENT
*(])
? ROOT CAUSE SPURIOUS SIGNAL FROM THE :LANT MULTIPLEXER CAUSED AN If! ADVERTENT 0PEllING 0F THE MAlfi FEEDER BREAKERS SUPPLING 0FF-SITE POWER MULTIPLEXER SYSTEM REMOTE OPERATI0ft 0F 'i0:1-SAFETY 13.8 KV SWITCHYARD BREAKERS COMPUTER BASED SYSTE!i UTILIZING FIBER OPTICS
- OCTOBER 7, 1985 EVEllT ROOT CAUSE SPURIOUS SIGilAL FROM THE ?LAliT MULTIPLEXER CAUSIflG THE MAIN r s FEEDER BREAKERS T0 OPEN WilLE TROUBLESHOOTING WITH THE
.V REACTOR SHUTD0Wil CORRECTIVE MEASURES C0f9LETE HARDWARE MODIFICATI0ft FOR CONTROL OF MAIN FEEDER 3REAKERS
?ERFORM SURVEILLANCE TEST TO EliSURE PROPER BREAKER CONTROL GWW 11/05/85
Q
.O -
- ,we w,w- w s : - w n ...,
a m o <, i i i s s i i 3j = -- - _a_}b_}p
) }L
}i
,, ,, ,",, "! J __ _ __1_.,,,,,,,1__ .____ _ _ _ 3 .- 1 I____ U STARTUP !"a""""
wu wu I Tw eI, @E""*" I TRANSFORMER gl T - T YARD ~ i Z -- I 1 I I . _I NC,) NO) NO) NC) NC) Ng) N g) NC) NC) NO) NC) Ng)fiai=j ==iwas
"""' _3R sj!! _ ,jil .sj,s__f21 s3s ju sg,jp sss _jp s6s_ _,f
\
7"" j
- k. \
an an f3 an f1 f% I i l l ) l l} -
)
soi so2 ' I soa i) soi i soa i) so2 soi ' I I)so4 I I)so4 soa i) soa l l) soc THREE UNIT ONSITE/ 0FFSITE POWER
O II.E. AUXILIARY FEEDWATER SYSTEM RELIABILITY
- 1. AFWS DESIGN AND OPERATION
- 2. COMPARISON WITH DAVIS-BESSE DESIGN
- 3. EXPERIENCE WITH MAIN FEEDWATER AND Al1XILIARY FEEDWATER SYSTEMS f
O l O l
O MS AUXILIARY FEEDWATER SYSTEM-(AFWS) PVNGS AFWS DEFENSE IN DEPTH i
. i COMPARIS0N OF PVNGS AND DAVIS-BESSE AFW DESIGN COMPARIS0N OF PVNGS AND DAVIS-BESSE AFWS ACTUATION LOGIC DAVIS-BESSE EVENT-ANALYSIS
-Q ..
I i e 9 AKK 11/05/85 O i
O
", r+- o O- ~
& N--" PALO VERDE itL. .
_ m. ..... g. R .m.1 9 X / LO u,,, ,,.. , m ,w Ehe < :8:-!: )- O
=. 7 ,i 7 a. A--
A - -J.. g. 1 a v.: - g a
, i e- '
. G 3"- x M :::
. . . . . .*g: :
- }- $$ O N 2,'.~,.i.
a.. m4 -- m
- __s ..-
a ==. .. . , - n
... R
-N-+< N Rc:::04Mw ' N V Qso gg..g = = = ,... . . ...
AUXILIARY FEEDWATER SYSTEM t
9 SIGNIFICANT DESIGN DIFFERENCES - PALO VERDE VS. DAVIS-BESSE PALO VERDE JAVIS-BESSE DIVERSELY POWERED EMERGENCY IDENTICALLY POWERED EMERGENCY PUMPS PUMPS 3 SEPARATE CST SUCTION LINES COMMON CST SUCTION LINE FOR 3 PUMPS 3RD PUMP HAS DG BACKUP POWER 3RD PUMP HAS N0 BACKUP 3RD PUMP IS 100% CAPACITY 3RD PUMP IS 50% CAPACITY RD PUMP OPERABLE FROM THE 3RDPUMPREQUIRESLOCALMANl[AL C0:lTROL ROOM ALIGNMENT DISCHARGE VALVES ON 3RD PUMP NO BACKUP POWER ON 3RD TRAIN HAVE BACKUP AIR DISCHARGE VALVES COLD PIPING TO AFPT MINIMIZED LONG RUNS OF COLD PIPING TO AFPT'S TO PREVENT STEAM CONDENSATION THAT ARE SUSCEPTIBLE TO STEAM CONDENSATION CE U-TUBE STEAM GENERATOR DESIGN B&W OTSG DESIGN MA!!UAL INITIATION OF AFAS ONLY MANUAL INITIATION OF SFRCS REQUIRES REQUIRES AFAS SWITCH OPERATION PARAMETER INPUT BY THE OPERATOR b AKK 11/05/85
O , COMyAR IGoN OF ~#Vd(56 Ado Wl6 65001E A, F d Ac Td ATIOd Lor = tc. 7Vd45 D A vis . B E. e E. m s ca Sc= bet 7xes,dge . L.sveL ~ Fit.Esd ne. L svec 4Wda L , 147UT 4 O TdFI4ED KdPTJKap
*sa gdf7 dKE ETKdfTd g.E.
L.ocic I' p o
' isola}E $# AiG l
i S4 sq kT Kd[s7d gep M.v4JA.L h'oT KJgTJgep tydT Y Y aras Arwe l _v_.:rd AT to 4 A c TJ ATioW tiGWAL Stq4AL l O
DAVIS-BESSE EVENT ANALYSIS EVENT POSSIBLE EVENT AT PVNGS? PVf1GS COMMENTS MFWP TRIP YES PVNGS SUCCESSFULLY TESTED TRIPS OF MFWP WITHOUT PLANT TRIP REACTOR, TURBINE TRIP YES CONTROL SYSTEMS ARE DESIGNED TO PREVENT RX TRIP AT PVNGS SFRCS INITIATES MSIV CLOSURE N0 A TURBINE TRIP DID RESULT IN AN AflD ISOLATES STEAM TO THE MSIS. MODIFICATIONS HAVE BEEN MFPT IMPLEMEf1TED TO PREVEllT RE0CCURENCE OP ATOR MISTAKENLY ISOLATES N0 OPERATORCAllNOTISOLATEMFWAh1DAF fhUTh L PbSRE hERRIkbRk IREb RkbdN COMP 0NENTS OVERSPEED TRIP OF BOTH STEAM N0 1 TURBINE AND 1 MOTOR DRIVEN 100% TURBINE AFW PUMPS RESULTS IN CAPACITY SAFETY AFW PUMPS. MOTOR TOTAL LOSS OF AFW PUMP CAN FEED BOTH SG'S RESET OF SFRCS DID fl0T OPEN NO PVNGS HAS NO COMPARABLE VALVES. N0 VALVES AF-599 AND 608. VALVES ONE VALVE ISOLATES ONE SG FROM BOTH WERE NOT DESIGf1ED TO RE0 PEN ESSEf1TIAL AFW PUMPS. ALL AFW AGAINST HIGH PRESSURE ISOLATION AND REGULATING VALVES ARE DESIGNED AND SUCCESSFULLY TESTED TO OPERATE AGAlfiST DESIGN PUMP HEAD
" AKK 11/05/85
~
{) DAVIS-BESSE EVENT ANALYSIS (CONTINUED) EVENT POSSIBLE EVENT AT PVNGS? PVNGS COMMENTS ATMOSPHERIC VENT VALVES NO PVNGS ATMOSPHERIC DUMP VALVES ARE AUT0f1ATICALLY OPEN AND NOT AUTOMATIC AND CONTROLLED BY DEPRESSURIZE STEAM OPERATOR ACTION ONLY GENERATOR USE OF 3RD PUMP REQUIRES N0 ALL VALVES TO USE THE 3RD PUMP HAVE LOCAL MANUAL ALIGNMENT ACTUATORS OPERABLE FROM THE CONTROL ROOM. POWER FROM DIESEL "A" AVAIL-ABLE. PHYSICALLY SEPARATED FROM THE SAFETY RELATED PUMPS. LOCAL MANUAL RESET OF YES MANUAL RESET OF THE TRIP LINKAGE IS T INE DRIVEN AFW REQUIRED. A MOTOR OPERATOR (1E) P IS PROVIDED TO OPERATE TRIP VALVE. AKK 11/05/85
/
^
0PERATION OF FEEDWATER SYSTEMS y(~y
- NORMAL PLANT OPERATIONS-
- STARTUF FROM 0 TO 3% POWER USING 3RD AFWS PUMP SWITCH FROM AFWS PUMP TO ONE MAIN FEEDWATER PUMP AT 3% POWER START SECOND MAIN FEEDWATER PUMP AT 50% POWER
~*
EMERGENCY OPERATIONS AUXILIARY FEEDWATER ACTUATION SIGNAL (AFAS): ACTUATION AT 25.8% WR SG LEVEL RESET AT 40% WR-SG LEVEL AFAS ACTUATES:
" -^
.O TURBINE DRIVEN AFWS PUMP MOTOR: DRIVEN AFWS PUMP VALVES TO SG f
GWW
-11/05/85 1
0 0
e . c .. . p EXPERIENCE WITH FEEDVATER SYSTEMS V l'AIN FEEDWATER SYSTEM ADJUSTMENTS DURING POER ASCENSION SYSTEM OPERATING EXPERIENCE NORMAL OPERATIONS REACTOR TRIPS OF JU:iE If4, AliD JULY 1,1985 0 .=s e 0 0 t GWW 11/05/85 e O :
OL EXPERIENCE WITH FEEDVATER SYSTEMS
. . AUXILIARY FEEDWATER SYSTEM USE DURING PLANT STARTUPS USE DURING ~ ABNORMAL OPERATIONS AFAS INITIATION AFWS PUMP OPERATION 9 -
sa W 4 6 e
- GWW 11/05/85 g, , ,m , . . . . . _ y , . _ . , _ , c , -,
, _ _ , , . - _}}