ML033300218
| ML033300218 | |
| Person / Time | |
|---|---|
| Site: | Byron  |
| Issue date: | 09/18/2003 |
| From: | NRC/OCIO |
| To: | |
| References | |
| FOIA/PA-2003-0444 | |
| Download: ML033300218 (95) | |
Text
1 1 U.S. NUCLEAR REGULATORY COMMISSION 2 + . . .
3 PUBLIC MEETING 4 ON 5 FOIA/PA-2003-0444 6
7 THURSDAY, 8 SEPTEMBER 18, 2003 9 + + - +
10 11 12 The meeting commenced at 1:30 p.m.-
13 14 15 16 17 18 19 20 21 22 23 24 Transcript produced from audiotapes provided by the 25 Nuclear Regulatory Commission.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
-1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
2 1 P-R-O-C-E-E-D-I-N-G-S 2 1:30 p.m.
3 MS. PETERSON: We've got quite a few 4 people here today. It's a little bit different than 5 we had anticipated with offices being shut down, we 6 don't have the video conference we had expected. But 7 we will begin with what we have.
8 I'm Cindy Peterson; I'm the Director of 9 the Division of Reactor Safety here in NRC Region 3.
10 Before we get into introductions, I'd like to talk a 11 little bit about how we're organized here.
12 We do have people on the phone from 13 Headquarters, which we will get to, as well as members 14 of the public on the phone.
15 Because of the unique circumstances, we 16 are going through out Headquarters Operation Center, 17 which means that the bridge we're on is a recorded 18 bridge.
19 That's atypical of a meeting of this 20 nature, but I wanted to make sure everybody was aware 21 of that before we began.
22 Today's meeting is to talk about. the 23 recently reported issues at Byron and Braidwood.
24 There Exelon determined they may have exceeded their 25 licensed thermal power limits.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
3
- l. So with that said, I would like to go to 2 introductions. I'd like to first have the NRC 3 introduce themselves here in the room, and then have 4 Exelon and then we'll go to the phone.
5 MR. CALDWELL: My name is Jim Caldwell.
6 I am today the Deputy Regional Administrator, as of 7 Sunday I'll be the Regional Administrator.
8 MR. REYNOLDS: Steve Reynolds. I am the 9 Acting Director, Division of Reactor Projects.
10 MR. HILLS: I'm David Hills. I'm a 11 Mechanical Engineering Branch Chief here in the 12 Regional Office.
13 MS. PETERSON: Thank you. (Inaudible).
14 SPEAKER: (inaudible).
15 MR. AINGER: Ben Ainger, Licensing Manager 16 for environmental safety.
17 MR. BENN: I am Carl Benn (phonetic), site 18 site director at Braidwood.
19 MS. PETERSON: Did we miss someone, I'm 20 sorry. Thank you. I know we have at least one NRC 21 person on the telephone. Bill, are you there?
22 MR. ROWAN: Yeah, this is Bill Rowan 23 (phonetic), I'm a Project Director for Project 24 Directorate 3 in the Office of Licensing and Project 25 Management in NRR. I think, George Dick, are you on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. -
(202) 234.4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
4 1 the line?
2 MR. DICK: Yes, I am, Bill. This is 3 George Dick, I am the Licensing Project Manager for 4 Byron.
5 MS. PETERSON: Any other NRC people on the 6 telephone? Since we don't have the benefit of 7 visuals, are there any other persons on the phone?
8 We had anticipated somebody from inside 9 NRC to be dialing in, so that person may join us.
10 Thank you, everyone.
11 MR. DICK: Cindy?
12 MS. PETERSON: Yes, go ahead, George.
13 MR. DICK: If I just interrupt for a 14 second, we're having a little bit of difficulty 15 hearing folks unless they are close to the microphone.
16 We can hear you fine, but some of the 17 introductions, as they went on, we weren't able to 18 catch.
19 MS. PETERSON: Thank you. We have some 20 people who aren't sitting at the table near the 21 microphone. That's a good reminder. For the 22 microphones, if the green light is on, it is live.
23 And if we have people elsewhere speaking, 24 it would be helpful if they could come closer to a 25 microphone.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
5 1 MR. DICK: Great, thank you.
2 MS. PETERSON: You're welcome. So we have 3 Cecil from IEMA (phonetic). Is there anyone else who 4 would like to introduce themselves for the public?
5 (Introductions are inaudible.)
6 MR. ROWAN: Cindy, this is Bill Rowan 7 (phonetic), just so you're aware, you know, we're 8 here, and one of the things we have to do because of 9 the hurricane is deal with contingencies, Licensees, 10 relative to security orders or something along those 11 lines.
12 So I might have to drop off the line. And 13 I'm not going to interrupt, if I have to do to that, 14 I'm just going to get off the line.
15 MS. PETERSON: Okay, we understand, Bill.
16 MR. OWEN: Okay.
17 MS. PETERSON: The other thing, because we 18 are going through the Operations Center, should they 19 need to recover this bridge for ongoing events, we'd 20 lose the connection also.
21 MR. OWEN: Okay, okay, thank you.
22 MS. PETERSON: Mr. Tuckman (phonetic),
23 have you joined us? Let's begin. On August 28th, 24 Exelon reported to us that three of their units may 25 have exceeded their licensed thermal power limits.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.neaIrgross.com
6 1 Byron Unit 1, potentially up to one and a 2 half percent. Byron Unit 2, 0.6 percent, and 3 Braidwood Unit 2 up to .08 percent. Now today we want 4 to talk about those issues.
5 We want to talk about the conditions and 6 circumstances that led up to the discovery of those 7 that you reported in August, and why it took so long 8 to get to that discovery point.
9 We're also interested in the current 10 conditions of those units. Also, as we understand, 11 you've got an investigation ongoing, we'd be 12 interested in your information, to date, on what 13 you've found and what your future actions would be.
14 And also we're interested in what 15 assurances you can give to us that suspicions that 16 you're making at this point in time, those that you 17 plan in the future and those that will be made before 18 your full investigation is complete.
19 What assurances do we have that similar 20 long-standing issues wouldn't have the same decision 21 making process that led to this protracted period of 22 time for the thermal power limit.
23 Again, since this is a public meeting, I 24 remind you if there is a need to discuss proprietary 25 information, we would need to hold that to a closed NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
7 1 session. Probably the easiest thing to do, if we need 2 to have a proprietary discussion, would be to finish 3 the public portion of the meeting and then close the 4 meeting.
5 Now regarding the public observation role, 6 we would ask any members of the public that are 7 present or on the phone to hold comments or questions 8 until we conclude the business portion of the meeting 9 between NRC and Exelon.
10 And at that time we would open it up for 11 any comments or questions that members of the public 12 may have. At this point, if anybody has any other 13 openings, we turn it to Exelon and your presentation.
14 SPEAKER: Just as a matter of protocol, we 15 do have feedback forms. I think they're on the 16 (inaudible). For any members of the public, we always 17 try to get any feedback from our public meetings and 18 how to improve them.
19 So, please feel free to pick up a form and 20 fill it out. You can either leave it with us or mail 21 it in.
22 SPEAKER: (Speaker is inaudible.)
23 SPEAKER ON PHONE: Hello, if you can hear 24 me out there, there's a problem with the audio, we 25 can't hear the speaker, or I can't, anyway.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
8 1 MS. PETERSON: Okay, we'll try to rectify 2 that.
3 SPEAKER ON PHONE: Okay, turning the 4 microphone on I think should help.
5 MS. PETERSON: Oh, a bright green light, 6 versus a dim green light would indicate the microphone 7 is live.
8 SPEAKER: Okay, thank you very much.
9 PARTICIPANT: Before you get started, on 10 the, you mentioned that Jack Skoals (phonetic) has 11 commissioned a group. Is that, what level is that?
12 PARTICIPANT: It's going to be led by one 13 of our site vice-presidents who was not involved with 14 this, Bob Bement from the Clinton Power Station.
15 And there will be other technical 16 individuals on the team, individuals from our 17 oversight board.
18 PARTICIPANT: That is internal, right?
19 PARTICIPANT: That is internal. We have 20 had members of our Nuclear Safety Review Board look at 21 the information that we're going to go through on the 22 slides today, looking for opportunities for statements 23 for, you know, other questions such as that.
24 But in direct response to your question, 25 the decision making evaluation effort. Okay, James.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
1 SPEAKER: Okay, that being said, thank 2 you, Chip. Can I be heard all right on the bridge 3 now.
4 PARTICIPANT: Yes, you're coming in and 5 out, but I can hear you.
6 MR. MICER: Okay, thank you very much. As 7 I said earlier, I'm Jim Micer (phonetic), Vice 8 President of Engineering for Exelon.
9 What I'd like to do first is, quickly, the 10 agenda is on Page 3 of the packet, but I'd like to 11 quickly take us to Page 4.
12 As you'll see shortly, Bill Cuba (phonetic) 13 and Brad Adams (phonetic), will describe the details on 14 the items that are bulletted on Page 4.
15 Their discussions focus on the background, 16 investigative efforts over time, the root cause and 17 safety implications and our actions going forward.
18 I would like to open the meeting with a 19 perspective on the overpower issue and to lay out for 20 you what our future actions will be.
21 As you are aware, both of our 22 organizations have expended a great deal of energy and 23 resources on figuring this issue out, over the past 24 several years.
25 Many communications have occurred between NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 wwwnealrgross.com
10 1 our respective staffs, and we have literally spent 2 thousands of hours and hundreds of thousands of 3 dollars in attempting to resolve this issue. While 4 we're satisfied that we have finally determined the 5 cause and a pathway for resolution of the issue, we 6 are dissatisfied with the fact that it took us so long 7 to identify that cause.
8 One action that we're taking to ensure 9 that the cause has been fully explored and challenged, 10 is an independent, technical evaluation of the root 11 cause conclusion.
12 This evaluation will be in addition to our 13 normal senior management and plant operations review 14 committee reviews, and will be completed prior to 15 further implementation of the ultrasonic flow meters 16 at the station.
17 Another aspect of this issue, and perhaps 18 the most critical, is the lessons learned with respect 19 to our decision-making process.
20 We preliminarily evaluated portions of 21 that process over part of the time period. However, 22 as Chip mentioned, as CNO Jack Skoals has commissioned 23 a separate team to thoroughly critique the decision-24 making and management actions that were applied over 25 the large span of this issue.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
1 As Chip said, the team is led by Bob 2 Burnett (phonetic), the Site VP from the Clinton 3 Station and will report their results to Jack Skoals.
4 We will systematically apply these lessons 5 learned to our decision-making processes and, as Chip 6 said, we will share the lessons learned with you in 7 the evaluations.
8 I'd like to now introduce Bill Cuba, the 9 Corporate Engineering Director, who will start with 10 the background on the ultrasonic flow meter.
11 MR. CUBA: Thank you, Jim. For this 12 portion of the presentation, what I want to do is I'm 13 going to provide some background on the ultrasonic 14 flow meter technology, implement at both Byron and 15 Braidwood and then go into the chronological events of 16 the issues and investigations that we performed.
17 So, if you go to Page 6, the purpose of 18 the ultrasonic flow meters, which shows the UFM 19 technology, because it does more accurately measure 20 feed flow.
21 And it was based on this accuracy that we 22 could potentially recover megawatts lost due to the 23 feed water Venturi of flow and accuracy.
24 This installation was not part of a 25 measurement uncertainty recapture of a K uprate. So NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
12 1 there was no intent to raise power beyond the existing 2 license limits.
3 The intent was to measure feed water flow 4 more accurately. Based on the design of the flow 5 meters, the ultrasonic flow meters, they, I believe 6 they have gotten accuracy down to about .5 percent, 7 .25 percent.
8 The feed water flow Venturis can be as 9 accurate as, I believe, around one and a half percent.
10 Okay, so you can have feed water Venturi falling.
11 There's inaccuracies with the instruments associated 12 with the flow Venturis.
13 Okay, so to move on, on Page 7, there's a 14 simple diagram of the ultrasonic flow meter. And just 15 to walk you through it very briefly, if you take a 16 look at the sensors alpha and sensors bravo, a signal 17 is sent from alpha and it strikes an eddy current 18 within the flow stream, and these are embedded in the 19 flow stream. Based on that characteristic of that 20 eddy, it modifies the signal and then that signal is 21 collected. (Inaudible) flow sensor, a similar signal 22 is set, and again it hits the same eddy.
23 It hits the eddy current and modifies that 24 signal. And those two signals are collected and then 25 compared. And if the compared signals match each NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
13 1 other within a statistical accuracy, in a time delay 2 calculated, and that's, that time delay is what we 3 call the cross-correlation.
4 The comparing of the two signatures. And 5 then based on that time delay that's calculated, at a 6 known distance between the probes, you can go ahead 7 and calculate a flow velocity.
8 PARTICIPANT: (inaudible) 9 MR. CUBA: It does to the calibration of 10 the equipment.
11 PARTICIPANT: How far are they apart?
12 MR. CUBA: About a foot apart, twelve 13 inches.
14 PARTICIPANT: (inaudible) 15 MR. CUBA: Okay, so a time delay is 16 calculated. With a known distance you calculate a 17 velocity based on the properties of water you then get 18 to a mass flow rate.
19 That's when we get to the ultrasonic flow 20 measurement, and that's flow rate. And then to come 21 up with the correction factor, which is what we use, 22 you take the ultrasonic flow and you divide it by the 23 Venturi flow, and that becomes your correction factor.
24 If you move to Page 8, that's basically what I've 25 discussed associated with the drawing.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
14 1 If you move to Page 9, Venturis are 2 located on the individual branch line. Just centering 3 the steam tunnel goes down the steam tunnel all the 4 way.
5 The UFMs are fairly close. They're 6 upstream of Venturis. I don't know, do we have a 7 distance? Ten diameters? Ten pipe diameters? Which 8 is roughly 130 inches?
9 You have a main header, d-reg valve, okay.
10 Then you have the ultrasonic flow device, the 11 Venturis, then you go on to (inaudible). And there's 12 a drawing.
13 Okay. The installations were supported by 14 our ultrasonic flow meter vendor. They were installed 15 in accordance with vendor procedures.
16 The NRC subsequently approved the 17 ultrasonic flow meter technology in March, 2000, for 18 use in Appendix K (inaudible). And the vendor 19 procedures that we use are consistent with those 20 approved in the NRC topical report.
21 The ultrasonic flow meters are installed 22 on an H-feed water branch line supplying the steam 23 generators and they're installed in the same manner as 24 both Byron and Braidwood.
25 They are installed on individual lines.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
\_ _, 234-4433 (202) __. ..__ WASHINGTON, D.C. 20005-3701 www.nealrgross.com
15 1 They are also installed on common headers.
2 PARTICIPANT: (Inaudible).
3 MR. CUBA: Correction factors available 4 are used as a multiplier in our calemetric(phonetic) 5 calculation and correct feed water flow, and then feed 6 water flow is a major input to the calemetric to 7 determine the reactor power.
8 We determine correction factors as 9 snapshots. They are determined on a standard 10 periodicity(phonetic). Also, after a defined change 11 in power, which has the potential to (inaudible) a 12 Venturi and/or during plant parameter trending,' and 13 that's done by our thermal performance engineer.
14 And all that is controlled by 'site 15 (inaudible). The periodic is every nine months, okay.
16 And then after a potential for a plan scram we would 17 have to go ahead and do a recalculation of the 18 correction factor.
19 And then also under a step change in load 20 or a major power (inaudible). At that point the 21 thermal performance engineer would go ahead, look at 22 the secondary site parameters and then go ahead and 23 determine a correction factor.
24 PARTICIPANT: (Inaudible).
25 MR. CUBA: We do calculate thermal power NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234.4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
16 1 daily. This correction factor is a multiplier on the 2 feed water input to the calemetric. The multiplier 3 correction factor is not changed, okay. So it is 4 calculated periodically, it's just (inaudible).
5 Okay. Moving on. Next, I want to go over 6 the chronology of issues and investigations associated 7 with the Byron and Braidwood megawatt differences 8 issues.
9 Going to Page 11, the ultrasonic flow 10 meters were implemented at Braidwood in June of 1999, 11 and they were implemented at Byron in May of 2000.
12 And I will go over why the gap between the two time 13 frames.
14 Installed and used to determine thermal 15 power. They were installed in May of (inaudible). We 16 noticed these electrical output differences upon 17 initial installation and also during the five percent 18 power uprate in 2001.
19 And it was really as a result of these 20 differences in megawatt output that we performed 21 multiple evaluations from '99, through 2002, to 22 determine a reason for this difference.
23 Installation was in '99. Turn on at 24 Braidwood was in '99. Turn on at Byron (inaudible).
25 Moving to Page 12, like I said, for Byron, because of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
A__ _,2344433 (202) __ . ..__ WASHINGTON, D.C. 20005-3701 www.nealrgross.com
17 1 the differences, evaluations were performed between 2 June, 1999, to May, 2000.
3 Items that were re-evaluated include the 4 installation was re-checked, the software was re-5 checked, the computer. We tried to understand the 6 differences.
7 We did dual-instrument testing with the 8 ultrasonic flow instrument and UFM bender. And what 9 we did there was we, on one of the feed water lines, 10 we actually installed a second ultrasonic flow meter 11 just upstream, and compared the two results.
12 And we got a very good correlation. We 13 looked at the Venturis, both at Braidwood and Byron to 14 look for clues, and we found a difference between 15 Byron and Braidwood in the cleaning methodologies in 16 the Venturis.
17 And why that is potentially significant is 18 Braidwood used a power wash, where Byron wiped their 19 Venturies and their piping down. A power .washing 20 could roughen the pipe which could actually make the 21 ultrasonic flow meter read more conservatively.
22 So what we did at Byron was we took a flow 23 meter and we put it down stream of anywhere that would 24 be affected by this power washing.
25 And, once again, we came up with a very NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(2021 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
18 1 good correlation. So there was no effect with respect 2 to whether we wiped down the Venturies or we power 3 washed them, to see if that was causing the difference 4 associated with the feed water flow.
5 We also did design reviews, especially 6 focused on secondary site plant parameters, heat rates 7 implementing procedures and OPECS.
8 And a lot of those reviews did identify 9 differences, especially in the secondary site but no 10 root cause could be established for that difference.
11 Now based on the multiple checks we did, 12 following vendors instructions on the installations, 13 our evaluations concluded that the Byron ultrasonic 14 flow meters were implemented correctly, they were 15 installed correctly and operating in the criteria 16 (inaudible).
17 Three times. Then we're going to talk 18 about (inaudible). Okay. (Inaudible). That's 19 correct.
20 Okay, moving on to Page 13, following the 21 upgrades in 2001, at both Byron and Braidwood and with 22 the differences in megawatt electric output still 23 present, we put together an independent Exelon review 24 team, which performed that review in February, 2002.
25 And those were Engineers from the Mid-NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 www.nealrgross.com
1 Atlantic Region, Thermal Hydraulic Engineers. We did 2 not include our ultrasonic flow measurement vendor in 3 those particular evaluations.
4 And really what we asked them to do is 5 take a look at what we've investigated so far, see 6 what else you can come up, see if there's something we 7 missed.
8 PARTICIPANT: Hey, Bill, you might want to 9 cover what drove that investigation, just to provide 10 linkage from --
11 MR. CUBA: Well, really what drove that, 12 again, tied to the uprates, was a, a CR was written.
13 Once we had Byron and Braidwood Unit l's uprated and 14 with all the turbine changes and equipment changes we 15 were able again to do a kind of a one-on-one 16 comparison between the units and (inaudible).
17 So that was an actual CR written. And 18 that was in response to that CR condition report. The 19 team did look and they did identify- Do you have a 20 question?
21 PARTICIPANT: I wanted to just go back to 22 the '99-2000 evaluation, because I was looking at the 23 Condition Report.
24 One of the reasons why you were checking 25 this out, besides the fact that you were getting more NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 www.nealrgross.com
20 1 megawatts out of Byron, was that all, almost all of 2 the parameters were higher at Byron.
3 MR. CUBA: That's correct.
4 PARTICIPANT: Which would indicate unless 5 you just got lucky enough that everything was on the 6 high side of the tolerance, all of them were.
7 So that would, you know, good engineering 8 judgement would tell you that that's a problem. And 9 then you did your assessment.
10 And I was reading in the conclusion of the 11 document, after doing that assessment, and it said 12 that the conclusion was that there was not enough 13 evidence to refute the primary flow measurement 14 standard, which is a different conclusion than what 15 you just said.
16 It's saying that, you've come at it from 17 a different angle. You have the question and you are 18 not, you're, I guess, basically saying that the, even 19 though the question exists, you can't say that it's 20 not accurate, instead of saying it is accurate.
21 MR. CUBA: And maybe I misled you. In the 22 June, 1999, through May, 2000, we concluded that the 23 ultrasonic flow meter was installed, per design, per 24 requirement.
25 It was measuring flow correctly. And it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005.3701 www.nealrgross.com
21 1 was on that'basis that we dedided to go ahead and 2 implement them. Okay. But that was based on that 3 technology, verifying the installation and verifying 4 'it met all the criteria for installation.
5 PARTICIPANT: So this is a Byron document.
6 So this might be different than what you're talking 7 about. It was a Byron position document.
8 MR. CUBA: That was the apparent cause 9 evaluation that was performed. That 'is a roll up 10 document from all the -activities to that date I regarding this issue.
12 So, what Bill is communicating is 13 essentially the conclusions that were reached through 14 that evaluation.
15 PARTICIPANT: Okay, but I was just going 16 on the' way it was stated was that you were basically 17 saying, since we can't prove it's not wrong, 'we're 18 going to say it's right; Is that the way you did it?
19 MR. CUBA: I think what you're getting to, 20 Jim, is, is a question that we're asking ourselves in 21 the decision.
22 We worked really-hard to look at all of 23 the various installation, start up tests and various 24 computer checks on the device, continually looking at 25 do we meet the standard in each one of these
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
22 11 1 .I parameters in terms of setting up the machine, and so 2 on and so forth.
3 And upon reaching the conclusion that we 4 did, we were confident because we met all of the 5 acceptance criteria we had set up properly that it was 6 therefore reading accurately.
7 And what we see as we look at this is, 8 that there were opportunities where there was other 9 data that was telling us we should look in a different 10 direction and we didn't (inaudible).
11 PARTICIPANT: Well, I have the advantage 12 of 20/20 hindsight, but I was just looking at the 13 information you had, which would tell you generally 14 that there was something wrong in the direction which 15 you had in question.
16 And then you went back and looked at the, 17 how it was installed, how it was operated. You 18 checked all the things out you could check out.
19 But the way it was stated was that since 20 we can't prove that it's inaccurate, we're going to 21 submit it's accurate. And that's what I was asking, 22 was that the way the conclusion was reached?
23 MR. CUBA: I think based on the conclusion 24 of the three reviews that were performed, the ones 25 from '99 and 2000, the independent review and then the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005.3701 www.nealrgross.com
23 1 one that followed after that, in conjunction with 2 that, combining all three of those reviews, we did 3 conclude, from a technical standpoint, that the 4 technology was working correctly and we accepted it.
5 So, again, as part of the independent 6 review, an issue was brought up identifying two burn-7 up anomalies and in particular the let down curves 8 (inaudible).
9 As a result of that question on burn up 10 and while it could be investigated, Byron went and 11 removed the correction factors and went through the 12 Venturis while that investigation was going on.
13 PARTICIPANT: This was Rich Lapiori 14 (phonetic), who was the (inaudible), at that point and 15 myself, we did work with this independent team when 16 they brought their conclusions. It was a Friday 17 evening, and brought the question of the fuel burn up.
18 That was a different question than had 19 been asked previously. In response to that, until we 20 further reviewed that question, we did remove the 21 constant and put the plant back in the Venturis for 22 that period of time.
23 And subsequent to that, as Bill will go 24 into, we did analyze in more detail that aspect, and 25 concluded that we were within expected ranges and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
11 24 1 subsequently thereafter we reinserted the count.
I1 2 MR. CUBA: They looked at the core models, 3 completion prediction, industry, other industry 4 variances and Boron letdown curve and nuclear fuels to 5 determine that they had the variation seen could not 6 conclude or quantify that type of difference.
7 PARTICIPANT: I'm not a Nuclear Engineer, 8 especially on the fuels, but how accurate can you get 9 on fuel burn up?
10 MR. CUBA: I can't answer that question.
11 But when they looked at that, the sensitivity looking 12 at your model, I don't think (inaudible).
13 PARTICIPANT: Because I was just visiting 14 Dresden(phonetic), and I understand they were told 15 they were about to, they were using more fuel than 16 they thought.
17 So they thought they were going to have to 18 go into coast down a couple months ago, and they're 19 not yet into coast down, which is to say it's not as 20 accurate as you might think. So that's why I was 21 asking.
22 MR. CUBA: That is correct.
23 PARTICIPANT: This would be a weekly 24 correlating comparatory. So our prediction pulls our 25 ability to estimate cycle length burn up in the order NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005.3701 www.nealrgross.com
25 1 of week from beginning of cycle to end of cycle.
2 So, by no means, can we look at burn up 3 over a month period and then extrapolate what thermal 4 power is. It is, however, a, there are opportunities, 5 especially on the (inaudible) looking (inaudible) to 6 either corroborate or refute information (inaudible).
7 Your question, Jim, I would say there is 8 it is not without value. It is certainly not 9 something they can immediately draw a conclusion.
10 PARTICIPANT: No, I thought it was a good 11 thing to look at, I just didn't know how accurate it 12 was and how, whether you could use that an ability to 13 make a decision from.
14 MS. PETERSON: During the time that you 15 went back to the Venturis, what megawatt change did 16 that cause? What was the result, I should say?
17 MR. CUBA: It would be roughly just 18 removing the constant. I think at that time we were 19 roughly around two percent on either one. Maybe 1.4 20 on (inaudible) two. It may not be exact, but 50 or 55 21 megawatts total, approximately.
22 So it was, again, based on the nuclear 23 fuels review and the fact that based on the 24 independent review, they really could not find any 25 other new insights, other than those previously review NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
26 1 at the site.
2 Turned the, or went back to the correction 3 factors. However, they did recommend that the Exelon 4 continue and put together an additional team together 5 to do, again, to keep looking at this issue and 6 perform a detailed evaluation.
7 Which, takes us to Page 14. So as a 8 result of that independent team's recommendation, we 9 did put together a team, again with the ultrasonic 10 flow meter vendor, corporate engineering and flight 11 engineering, again to review the UFM implementation.
12 But this time, a much more detailed, hand 13 over hand type of review of the installation where we 14 hand the calemetrics. We actually looked at the 15 piping, the piping design, the material in the piping, 16 transducers, cables, software.
17 We actually took a machine from 18 AMEG(phonetic), used it on the system and got 19 correlating results. We took the Braidwood machine, 20 put it on Braidwood, got correlating result.
21 We took calemetric data from Byron and 22 plugged it into the Braidwood calemetric to see if 23 that would make a difference and we got the same 24 result, and vice versa with Braidwood input into 25 Byron.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
27 1 PARTICIPANT: When you say you're getting 2 correlating results,-what do you mean by that?
3 MR. CUBA: Meaning very close. Meaning if 4 we calculated through the calemetric, 98.2 percent 5 power, we get something very close.
6 If we'took all that same data and put it 7 in the Braidwood computer, put all that data in there, 8 we would get a very close (inaudible), to see if 9 somehow we could find where there might be a 10 difference causing the different output.
11 PARTICIPANT: And you may not be able to 12 answer this, because I know you're still working at 13 the root cause and the time line and how you made your 14 decisions, but you concluded that, based on the burn 15 up, that the correlation factor or that you could use 16 the correlation factor and you reinstated it.
17 But the recommendation was to continue to 18 further review it. Why?
19 MR. CUBA: Again, there were differences 20 noted, but they could not come up with a root cause or 21 an explanation for the difference. So they felt it 22 needed to be pursued further.
23 PARTICIPANT: So, if not, did you not 24 (inaudible) do it here,'maybe it's a question you can 25 take back with you. Is that, would that be your, if
,-NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
28 1 you still had a question about whether you were 2 operating at your thermal power limit, would it be 3 normal to reinstate the correlation factor which would 4 put you back above it, if in fact you were?
5 I recognize you did not believe you were, 6 but there was still enough question in your mind that 7 you wanted to further review it.
8 Is that going to be the normal, I guess 9 the question is, is that what you would, your 10 expectation, and maybe, like I said, maybe you can't 11 answer that now and maybe you can look and see what 12 the thought process was.
13 PARTICIPANT: Yes, we're clearly looking 14 at that aspect, I think, rather than speculate. And 15 two, what this will boil down to is when do we arrive 16 at enough information that appears to be contradictory 17 or nonsupportive (inaudible).
18 It's not infrequent for us to have to 19 resolve (inaudible) or disposition question. It's 20 likely not a day that goes by at a station where in a 21 condition report we capture some kind of question.
22 However, I see your point is, that is 23 essentially the genesis for the (inaudible) how do we 24 process it? What points are, you know, triggers 25 (inaudible) and revaluate our current (inaudible).
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
29 1 So, I see your point. And I think that 2 will be the central theme to follow on.
3 MR. CUBA: Okay,- good. *We will be 4 interested in the answer to that.
5 Moving on, we also did a different type of 6 independent type testing and that was we, this was the 7 first time we put an ultrasonic flow meter transducer 8 on the main header that that-then feeds each of the 9 individual branch lines.
10 Now we were able to do a comparison 11 between a line that feeds all the individual feed 12 water lines.
13 And when we did that, we did get a, what 14 I'll say is a correlation within the accuracy of the 15 ultrasonic flow meters, but it was at the high end of 16 the accuracy (inaudible).
17 This was also during a period of time when 18 we had it scheduled that we were starting our coast 19 down, so this was done at 92 percent power. And these 20 are two key items, I'think here, that we need to go 21 back and look at.
22 Because at the time we did this, and with 23 the understanding of how the-ultrasonic flowmeters 24 work, they should be power independent with respect to 25 accuracy.
NEAL R.GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
30 1 This would later turn out to be something 2 that we found out as part of our root cause was not 3 the case, but the problem that we found. And Brad 4 will go into that.
5 So the study, based on the correlation 6 within the accuracy of the ultrasonic flow meter 7 design, concluded that the ultrasonic flow meters did 8 indeed measure flow per design and were implemented 9 properly.
10 And that was based on that correlation 11 test. And then all the, all the reviews we did on the 12 installation.
13 PARTICIPANT: (Inaudible). Are you going 14 to talk about (inaudible) the March, 2002, comparison 15 testing and what I understand was (inaudible) some 16 sort of test you did this summer (inaudible)?
17 MR. CUBA: Yes, I will. I'll talk about 18 that.
19 PARTICIPANT: The same test but there were 20 differences, I guess.
21 MR. CUBA: Well, yeah, because when we 22 repeated this test on Unit 1, it turned out that we 23 repeated it at full 100 percent power. Okay.
24 And when you get into the root cause of 25 what the problem was, a design that was power NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
31 1 independent was actually not working correctly, which 2 made the correction factor power dependent.
3 PARTICIPANT: It was more accurate at full 4 power?
5 MR. CUBA: The'accuracy for the correction 6 factor changed in a non-linear fashion as power 7 changed. So it's not-supposed to --
-8 PARTICIPANT: It's the fact that we took 9 the data at 92 percent power (inaudible) masked the 10 fact that there wasn't good correlation. It wasn't 11 until we took the data at 100 percent power this year, 12 that that really beca'me'evident.
13 PARTICIPANT: Is that something, is that, 14 that's new information from the vendor? I mean, is 15 the vendor aware?
16 PARTICIPANT: That's the information based 17 on our test that we completed. That (inaudible) Byron 18 on Unit 1 in late August.
19 PARTICIPANT- Is that something that needs 20 to be put out to the rest of the industry?
21 MR. CUBA: That is part of the root cause, 22 and how that occurred. And we will get into that --
23 PARTICIPANT: That's something that needs 24 to get out?
25 PARTICIPANT: AMEG guys issued a tech NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
32 1 bulletin on this specific issue, and I know that they 2 do have extended a condition review ongoing for their 3 other applications.
4 MS. PETERSON: That's correct, we issued 5 a tech bulletin and we will also be discussing all of 6 this in detail with NRR next Friday.
7 PARTICIPANT: I'm just concerned about any 8 other plants out there that might want to do their 9 calibration at lower than 100 percent power and they 10 would not be aware that that's not a linear --
11 MS. PETERSON: All of our customers are 12 aware and (inaudible).
13 PARTICIPANT: There's also one additional 14 point here is one of the things that, one of the 15 opportunities is when we did the comparison testing, 16 we found that our data fell on, we had an acceptance 17 ban, but we were close to the edge on the acceptance 18 ban.
19 And that's something else in the root 20 cause in the evaluation we're looking at was, there's 21 not enough information and you should have had enough 22 information to trigger additional follow through.
23 So I wanted to make sure that that point 24 was covered.
25 MR. CUBA: And then, moving on, one of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
. . 234-4433 (202) WASHINGTON. D.C. 20005-3701 www.neaIrgross.com
33 1 recommendations of the 2002 study was to take data 2 over a long period of time, continuously, to see if we 3 could uncover some kind of response or anomaly that we 4 could tie to plant conditions or something only not in 5 the plant to help us determine what the cause or 6 potential cause of the mismatch between the two units 7 was.
8 And this was going to be a lessons 9 learned. This turned out to be a missed opportunity 10 as well. We collected the data, we did not fllow 11 through and give this long term data to the vendor to 12 help them analyze it.
13 We were really weak in developing a plan 14 to get this data collected, collected under certain 15 conditions, and give it to the vendor. And when I go 16 ahead and talk about the final plan.
17 When we got to the root cause, we took 18 that and shored that up to make sure that they got all 19 the data they needed to help us with this 20 investigation.
21 MS. PETERSON: What did you do with the 22 data you were collecting?
23 MR. CUBA: Essentially, it was collected 24 at the site, but it was not forwarded to the vendor.
25 MS. PETERSON: To try to evaluate it in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 www.nealrgross.com
34 1 any fashion, or it was just collected and they were 2 not evaluated?
3 MR. CUBA: They were evaluations that were 4 ongoing as part of the, I think the document Mr.
5 Caldwell(phonetic) was reading there.
6 It says, you know, 2000 document. So 7 there was evaluations that were ongoing, but the 8 specific data that Bill is referring to was taken, but 9 it wasn't forwarded on to AMEG.
10 MS. PETERSON: (Inaudible) was just Unit 11 1 or was it Unit 1 and Unit 2 of Byron?
12 PARTICIPANT: Continuous data on Unit 2.
13 MR. CUBA: The other question that was 14 asked was what did we do with that data? Were we just 15 storing it? Was anybody looking at it?
16 PARTICIPANT: That specific data that 17 they'd asked about, Jeff, was not reviewed by 18 (inaudible).
19 PARTICIPANT: Continuous data from the UFM 20 instrumentation itself.
21 PARTICIPANT: The four? Or is the four 22 compared to the before?
23 MS. PETERSON: Again, just Unit 2.
24 MR. CUBA: Unit 2 (inaudible).
25 PARTICIPANT: After the completion of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005.3701 vmw.nealrgross.com
35 1 tests (inaudible) being a temporary installation was 2 removed from the (inaudible) procedures for temporary 3 modifications to the plant (inaudible) other test 4 equipment, it was removed after that test.
5 MS. PETERSON: And I just want to ask one 6 question you may have answered already. After you did 7 the comparison test on Unit 2, back in March of 2002, 8 did you see the signal noise that you saw just 9 recently.
10 MR.' CUBA: We did not see the signal 11 noise. But we did not, let me clarify that. We did 12 not look for the signal noise in the fashion we did in 13 2003.
14 MS. PETERSON:- Have you been able to go 15 back and look at the data and see if the signal noise 16 was there?
17 MR. CUBA: You'll have to help me. I"m 18 not sure if you can go look at that data.
19 PARTICIPANT: I can answer that question.
20 The data was in the context of the modulation. It's 21 not (inaudible). So'therefore we cannot go back and 22 extract that data.
23 But we have data (inaudible).
24 MS. PETERSON: Okay, so that information 25 is different than what was being collected for NEAL R. GROSS
. . COURT REPORTERS AND TRANSCRIBERS ,
1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
36 1 continuous data collection.
2 PARTICIPANT: What you're collecting 3 (inaudible) in the continuous data as the actual 4 (inaudible).
5 MS. PETERSON: Okay.
6 PARTICIPANT: There's no data used to 7 calculate the (inaudible).
8 MS. PETERSON: And it's not the raw data, 9 it's the, what comes out after the calculation.
10 MR. CUBA: (Inaudible).
11 PARTICIPANT: A couple days worth of work 12 when you did remove the insulation, prep the piping, 13 assure the dimensions, hook up a bracket to calibrate 14 the probes. You know, two days, half a week's worth 15 of work.
16 PARTICIPANT: (Inaudible).
17 MR. CUBA: A combination of technicians 18 from the vendor with some site support.
19 MS. PETERSON: I have another question.
20 I'm sorry. I guess this was not clear to me. If the 21 testing was done on Unit 2, we had a concern with Unit 22 1. What testing was done on Unit 1?
23 MR. CUBA: The testing, the comparative 24 testing was done on Unit 1, okay.
25 MS. PETERSON: Unit 1, okay.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
37 1 MR. CUBA: Okay. The data collection was 2 being collected on Unit 2, because they had looked and 3 they were measuring for a correction factor and had 4 one anomalous reading on the two alpha.
5 So we were collecting data-to see what 6 potentially could be causing that. So he was 7 collecting data for that.
8 PARTICIPANT: So the way I would 9 characterize that on a little higher level ;is the 10 expectation was over time to collect it on both units, 11 and Unit 2 was done first out of convenience, because 12 there were existing questions on that unit?
13 That was the unit on which the (inaudible) 14 at the time. However, as we said earlier, the'data 15 was (inaudible).
16 So the important point on 14 is, that 17 there were two opp6itunities that we believe in the 18 decision making (inaudible).
19 One was the collection of data and the 20 second was the fact that we were on the outer band of 21 (inaudible), the main header in the branch line.
22 And neither we nor our vendor said find 23 the outer band, we really need to call the 24 (inaudible). We met an acceptance criteria, but not 25 by (inaudible). I think those are both important.
NEAL R.GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234.4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
38 1 MS. PETERSON: Thank you.
2 MR. CUBA: Moving on, the last quarter of 3 2002. Resident Inspector did issue an unresolved item 4 in this area. We did receive the letter to Exelon 5 concerned about Byron's Unit 1 thermal power level.
6 We did conclude with their response based 7 on the testing we had done reviously that the Unit 1 8 ultrasonic flow meters were indeed installed 9 consistent with the vendors procedures, and that the 10 Unit 1 was operating within its licensed thermal power 11 limit.
12 Now to put things into perspective, it's 13 four years, we've got three major investigations we 14 still have the major or we still have this difference 15 in megawatts electric between Byron and Braidwood.
16 There's regulatory interest. So, we sat 17 down and we developed a plan. We knew we had to 18 organize this plan.
19 We had to get the right data. We felt we 20 had to get data, not only at Byron, but also at 21 Braidwood, so we could get the right data, the data 22 possible to help us evaluate the issue and kind of 23 broaden our scope and get data trends.
24 And from the lessons we've learned on 25 collecting data on Byron, on Unit 2, that became also NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.neaIrgross.com
39 1 part of the plan.
2 MS. PETERSON: Back here, in your remarks, 3 that NRC's interest in it prompted you to step back 4 and look at it and-make this plan for attack.
5 MR. CUBA: I think it was a piece. I 6 think it definitely had an influence.
7 MS. PETERSON: Can you maybe tell me a 8 little more about that?
9 MR. CUBA: Well, as Jim said in his 10 opening remarks, we have been working with this 11 investigation for several years, okay.
12 And we've put a lot of time in it, a lot 13 of money, and we still were not yielding results. And 14 now we're getting questions from you about, hey,
- 15 what's going on?
16 PARTICIPANT: In addition, Cindy, we did 17 have a meeting of NRR in January, following the 18 issuance of the letter and (inaudible) we did get some 19 good input with respect to some ideas going forward, 20 in terms of things that they perceived that we should 21 look at, based on their experiences not only with this 22 technology, but' with some of the competitors 23 (inaudible) in'the industry.
24 So that was factored into the thought 25 process and market this year also.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
40 1 MR. CUBA: Yeah, Cindy, just to walk 2 everybody through the sequence here. In January, mid 3 January we got a letter from you all asking the 4 question on the over power at Byron.
5 And we said, you know, look, we'd like to 6 come out and meet, the meeting was at NRR. The 7 Technical Coach, Bill and Brad, being two of the 8 primary folks, went out to NRR.
9 And as Brad stated, they went through a 10 number of the issues. Here's what we know. Here's 11 what we don't know. Everything that we've walked 12 through.
13 And there appeared to be a level of 14 increased understanding on both sides. Both from our 15 standpoint and what your questions were, and from the 16 NRC standpoint, principally NRR, what we had done and 17 why were drawing the conclusions that we did.
18 Subsequently, I believe it was about two 19 months later, we had some additional questions to go 20 through and at that point we started working with NRR 21 and said here's the ones that we can answer now.
22 Here's the ones that we don't know if we 23 will ever be able to answer. And then lastly, here's, 24 there's a portion of these that we think we could 25 answer with going through what we call our test plan NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005.3701 www.nealrgross.com
41 1 here.
2 And we're willing'to-implement this test 3 plan to get through to answer the rest of these 4 questions. It was something that was set out and that 5 was being walked through when we ultimately discovered 6 that while we were looking at this testing process to 7 go from data collection from, in essence, July through 8 early 2004.
9 So there 'was a concerted effort here to 10 get to that, but if you look at the NRC's involvement, 11 you know, I think'everybody would have to say, yes, 12 obviously when you get a letter from the NRC it 13 raises, you know, it gives you another opportunity to 14 pause and reflect.
15 And when the RAI questions came out, some 16 of those we weren't in a position to answer. We did 17 decide to go forward'-and implement the test plan.
18 MS. PETERSON:' Had the- NRC-not made those 19 inquiries, not had those questions, do you think you 20 would have gotten to the point of identifying the 21 issues that you identified in August?
22 MR. CUBA: That's a, you know, Steve and 23 I talked about that yesterday. And that's a good 24 question. Not in August, is probably the safest way 25 to answer that.
- NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234.4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
42 1 You know, ultimately, I think that as 2 additional operating experience became available and 3 this continued to get looked at, would we have come 4 across it and when we would come across it, I don't 5 think we can answer that.
6 MS. PETERSON: But you didn't have a plan 7 in place at that time that would have driven you to 8 collect this data, is that correct?
9 MR. CUBA: That's correct.
10 MS. PETERSON: And, if I understand 11 correctly, once that condition report was issued, you 12 were considering the issue was (inaudible). The 13 condition report you had mentioned earlier. The one 14 that was identified in January.
15 PARTICIPANT: (Inaudible) evaluation, 16 there were a couple of follow up action, out of that 17 (inaudible) to continue to look. But not with the 18 rigor of the plan that was generated.
19 MS. PETERSON: Okay.
20 PARTICIPANT: But there were, you know, 21 continuing a couple of actions to review again fuel 22 performance over a cycle, continue to trend. There 23 were actions attached to that (inaudible).
24 But, like I say, not with the same 25 complexity of the plan that was generated in March.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.neaIrgross.com
43 1 MR. CUBA: And one additional point that 2 was, in having been involved in probably all the 3 discussions, one of the drivers, and, Cindy, I think 4 the NRC definitely played a part.
5 But one of the drivers in all the 6 discussions were what possibly can there be that we're 7 missing? I mean from our evaluation, from our 8 vendor's evaluation, from everybody that looked at 9 this.
10 Outside technical experts, everybody said 11 it's installed, it's operating in accordance with its 12 design you guys have implemented. Your, the 13 calibrations are good.
14 And the test plan was developed for just 15 that purpose, to say what else could there possibly be 16 that we're missing. And a number, a significant 17 portion of the' test plan was going to go back over 18 ground that we had already gone over, just to make 19 sure that there wasn't something we missed.
20 So, it wasn't, you know, it was something, 21 the test plan itself was something that we developed.
22 It wasn't, the NRC didn't tell us you need this-test 23 plan.
24 You guys asked us questions and we thought 25 that that was the best path to resolution, to finally NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(2021 2.4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
44 1 and completely address this issue.
2 MS. PETERSON: If I understood your 3 earlier answer, had we not asked those questions, that 4 test plan would not have been developed?
5 MR. CUBA: We wouldn't have found it by 6 August. To go back to my earlier answer, I mean, you 7 know, I don't know if ultimately if, as Steve said, 8 there were actions to follow up.
9 Would we have gotten there, when we got 10 there? That answer, I believe, is clearly no. Would 11 we have gotten there eventually? You know, that's a 12 tough thing to answer.
13 PARTICIPANT: For example, one of the 14 longer term actions coming out of that was to do just 15 what one of these steps is and collect this long term 16 data, okay.
17 And that was one of the steps in our plan.
18 Would it have been part of this overall schedule, I 19 don't know. But we definitely had to go and take this 20 long term data and get it to the vendor to try to 21 understand, try to verify that the data was being or 22 the correction factor was not changing over time due 23 to conditions in the plant.
24 You know, to go find out and try to 25 determine that. And that became part of this plan.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 www.nealrgross.com
45 1 There were other parts'of the plan in response to the 2 NRC letters.
3 I mean that was the same group involved 4 with testing. We assigned owners to help develop the 5 responses to the question area.
6 MR. CUBA: And I don't know if Bill and 7 George are still on the phone, but you know, in 8 understanding that you guys donlt typically put 9 yourselves in consultant mode, but based upon the 10 meeting in Washington, we were staying open to ending 11 all inputs.
12 Because from -our internal and external 13 evaluation, we couldn't find it. So the meetings and 14 discussions that-were occurring with NRR, were in that 15 context.
16 They weren't, hey, we have this question 17 and we answered'and, you know, as shortly as possible.
18 It was more in the'exchange of information, here's 19 what we do know, here's what we can figure out and 20 why.
21 And even as we worked 'through the 22 development of the RAI questions, that was the spirit 23 in which the questions ultimately came out, was in 24 that cooperation.
25 And from'a technical approach, what could NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234.4433 WASHINGTON, D.C. 20005-3701 www.neaIrgross.com
46 1 we be missing because, and we appreciate, as those 2 guys said, there's a couple of folks in NRR that had 3 pretty good insights that we immediately took back and 4 applied at the plant.
5 To continue on, I'm on Page 17. We 6 discussed a little bit about investigating long term 7 trends. We had this issue where we had collected data 8 and essentially done very little with it.
9 So, what we did was we installed a direct 10 data link from the data being collected by the 11 ultrasonic flow meters straight to the vendor.
12 So he could study it daily, and the other 13 reason was, is we could watch the correction factor 14 trends during steady stay power, reduced power, during 15 power changes, and really pre and post refuel outage.
16 Potentially Venturi affects. If the 17 correction factor shifted, shifts due to power, which 18 it's not supposed to do.
19 So, we got that data directly to the 20 vendor, no middle man. They could collect all the 21 data they want to help us to try to evaluate the 22 problem.
23 The other item is we did gain some 24 insights when we put on the common feed water header 25 and compared it to the individual branch lines.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
. . 234-4433 (202) WASHINGTON, D.C. 20005-3701 www.nealrgross.com
47 1 Now we only had one point, and that was 2 Byron. So we wanted to collect Byron Unit 2, 3 Braidwood Unit 1, Braidwood Unit 2, to get additional 4 data to check out to see if there was correlation, to S see if Byron 1 is-an (inaudible).
6 So we moved ahead with the plans. As a 7 result we did our first test at Braidwood Unit 1. I'm 8 on Page 18. We did a comparison that showed almost 9 identical correlation between the common header and 10 the individual line, ultrasonic flow meter.
11 We had a very good correlation with 12 respect to that testing. We then moved on to Byron 13 Unit 1. Again, this time the difference was we were 14 at full power.
15 We did the comparison on the feed water 16 header to the sum of the individual lines, and at this 17 time we were outside the statistical allowance.
18 And what we did, signal noise was 19 observed. So an additional test using spectrum 20 analysis was done. And we noted that there was, under 21 the spectrum analysis, and this gets into part of the 22 root cause, there was no noise on the main feed water 23 header, but we actually saw noise on each of the 24 individual four lines.
25 And we'll get into that in the root cause.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005.3701 www.nealrgross.com
48 1 PARTICIPANT: (Inaudible).
2 PARTICIPANT: Oh, if you bear with us, 3 I'll talk a little bit about that root cause and talk 4 about the (inaudible).
5 MR. CUBA: So we had a definitive problem 6 now identified where we were outside the tolerance 7 between the two measurement locations, and we now had 8 noise indicators.
9 So, with the problem identified, we made 10 the decision at that point to go ahead and reduce 11 power. And I'll turn it over to Steve.
12 PARTICIPANT: You know, at the site we're 13 obviously watching and monitoring the site very 14 closely. Brad came to me as soon as there was some 15 verbal information regarding his variance and outside 16 of what was expected between the common header and the 17 individual loops and the potential noise issue.
18 So, at that point, we took action similar 19 to February of the year before when there was a 20 question of new information.
21 We removed (inaudible) and went back on 22 the Venturi flow. We did that on both units that day.
23 Particularly Unit 1 was the variance between the 24 common header and the individual loops.
25 And on Unit 2 we saw noise on one of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
49 1 individual lines. So, that kicked off a condition 2 report that kicked off an extended condition review 3 for both Byron units and to both Braidwood units.
4 Okay. So based on that extent of 5 condition and the test that was developed for noise, 6 we went to Braidwood Unit 1, I believe, two days 7 later. And we took measurements. We did see some 8 noise on two lines.
9 However, we had this very close 10 correlation, between a header that was measuring flow 11 without noise, and the individual line. So the 12 correction factors remain in effect at Braidwood Unit 13 1.
14 Braidwood Unit 2, we did not have the 15 correlating main header feed water flow measurement 16 bracket on, and based on seeing noise on two of those 17 loops, we went ahead and turned those constants, the 18 one, we removed the correction factors from two of the 19 four and reduced power appropriately.
20 PARTICIPANT: And I assume that's the same 21 thing on Byron Unit 2. Even though you saw noise on 22 one, you also didn't have a correlation.
23 MR. CUBA: That's correct, we did not have 24 a correlating bracket on Byron Unit 2. Okay. Now, we 25 turned off all the correction factors on Byron Unit 2.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
50 1 We turned off the affected loops on 2 Braidwood Unit 2. Braidwood Unit 2 had the other 3 advantages. They did not have all these questions on 4 their secondary.
5 Their thermal performance was right in 6 line with where (inaudible). So, at Braidwood the 7 unit that was running as predicted in that secondary 8 site indication.
9 PARTICIPANT: So you reset the factors, 10 all the correlation factors on, put back to one on 11 Unit One, Byron Unit 1, Byron Unit 2, and on, you left 12 the ultrasonic (inaudible) or the correlation factors 13 on Braidwood Unit 1 and you set two of the loops back 14 to one on Braidwood Unit 2.
15 MR. CUBA: That's correct. And then the 16 units notifications were made in accordance with the 17 license commission.
18 So, in summary, I tried to give you a 19 little perspective on the investigations we performed.
20 obviously we've evaluated the issues several times, 21 but unfortunately it was four times before we got it 22 right.
23 We, and the vendor missed opportunities in 24 our early evaluations and really in the developmental 25 or failure mode to be focused on equipment outside the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
1 pipe. Not the water and the noise that could be 2 generated from water inside the pipe.
3 As Brad will discuss, I think we're 4 confident we got this root cause and he'll talk about 5 that in his slides and we are going to continue to go 6 forward on those portions of the test plan.
7 And as Jim mentioned, we're going to do an 8 - independent technical review as part of this root 9 cause analysis. If there's no questions, I'll go 10 ahead --
11 MS. PETERSON: I do just have one question 12 and I'm not sure, Brad, if you're going to get into 13 (inaudible) about it.
14 In one of the, one of the evaluations that l5 was done, the only thing I have here is Westinghouse 16 April of 2002,(inaudible) of your condition report.
17 It mentions that there was a loose nut 18 that caused a high noise level in the transducer 19 response characteristics. -
20 Is that the same type of noise or is it a 21 different type?
22 PARTICIPANT: It's a different type of 23 noise. Because when 'the thing is slow, the coupling 24 to the transducer (inaudible). So because of that you 25 have physically amuch lower-connector noise ratio.
'NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
-1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 www.neaIrgross.com
52 1 That is not the noise that we will be 2 talking about.
3 MS. PETERSON: Okay.
4 PARTICIPANT: That particular noise is 5 (inaudible) and it will be (inaudible) transducer, and 6 there will be no correlation (inaudible). It wouldn't 7 even come into the picture.
8 MS. PETERSON: Okay, so that's why the 9 data was rejected in that case, but not rejected with 10 this noise.
11 PARTICIPANT: Would you expect this root 12 cause to affect units in different ways or would you 13 expect it to impact them in the same way in every 14 case?
15 PARTICIPANT: I'll talk about that. When 16 you go through the technical aspects of what we saw, 17 in reality depending on certain conditions you can end 18 up with a conservative, non-conservative or no impact 19 result. And I'll talk about that.
20 Good afternoon. What I want to talk to 21 you about is the root cause that we've found to date 22 and our corrective actions going forward.
23 Then I'm going to spend sometime on, at 24 least, our preliminary evaluation and safety 25 implications of our historical condition.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
53 1 If you'd turn to Page 22, immediately 2 after the event of August 28th, there, we formed a 3 root cause team at the site. And the purpose of the 4 root cause team was obviously to determine the cause 5 of inaccurate feed water flow measurement.
6 The root cause team was formed with 7 personnel from Byron and Braidwood, along with 8 Canteer (phonetic) engineering and technical expertise 9 from AMEG and Westinghouse.
10 And we have both root cause expertise on 11 this team and technical expertise to make sure that we 12 have the proper mix of personnel to go at that.
13 The team has been working very diligently 14 for the past three weeks on this activity. What we 15 have at this point is what I characterize as a 16 preliminary root cause.
17 But I have a high degree of confidence 18 that we have identified the cause. The reason it's 19 listed as preliminary because root cause still has to 20 go through our formal management review process at the 21 site.
22 And that is still in progress. We, our 23 schedule is such to support the upcoming licensing 24 event reports associated with that and, but I can tell 25 you that we have a high degree of confidence that we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
,^<\~flln-7%
'"A WA;HfNr.T0N flC. 20005-3701 www.neafrgross.com
54 1 have found the issue.
2 Essentially, and what we found is, that 3 there is an ultrasonic flow correction factor error, 4 as you've heard. And the correction factor is caused 5 by noise.
6 And this noise is impacting the time 7 delay. Bill talked about how the AMEG bracket 8 basically measures the time delay. It's really fairly 9 simple, conceptually. But it integrates those 10 individual eddy currents and comes up with an average 11 time delay, which is then correlated to, eventually to 12 feed water flow.
13 Basically, what this error was, maybe if 14 I can say it in different words. It's a physical 15 shift in the time delay that's determined from the 16 AMEG.
17 So, in effect, the error or the bias 18 caused a shift in that time delay, either right or 19 left. And I'll talk about, a little bit why and why 20 that's important.
21 And basically what we've seen is that the 22 bias varies as a function of noise structure and 23 intensity. And what I mean by that is that depending 24 on the frequency of the noise and the magnitude of the 25 noise, as it's factored into the cross-correlation, it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
55 1 can impact the total time delay calculation, either 2 positively or negatively or theoretically stay the 3 same. If that makes sense.
4 The frequencies that we're talking about 5 are on the order of 30 hertz or less. So we're seeing 6 noise frequencies in the flow stream at specific 7 points.
8 And I'll talk about how we came up with 9 that determination. The bottom line to remember here 10 is, though, with respect to this specific noise, there 11 are no acceptance criteria or were no acceptance 12 criteria provided originally as part of the 13 installation to evaluate for this noise.
14 And we didn't really ask for it because 15 of, I guess, a lack of knowledge on our-part. And at 16 this point, up until sort of the revelation here at 17 the end of August, we really collectively, between us 18 and AMEG had a lack of recognition (inaudible).
19 If you turn to Page 23, at-the top of the 20 page there, the presence of noise in the individual 21 loop' flow signal cause the non-linearity in the 22 calculated (inaudible) correction factor as a function 23 of power level.
24 And as I stated before, depending on the 25 frequency and the magnitude of the noise signal, when NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234.4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
56 1 you go through the actual physics of the calculation, 2 the bias can be conservative, non-conservative, or it 3 can have no impact.
4 And it really depends on your individual 5 configuration. If you had a constant driver, however, 6 which is what we see associated with this, then it is 7 going to impact your correction factor in a same 8 direction with the magnitude.
9 And it's also integral to why we see a 10 variance at power level. As we have gone back and 11 looked at it, we see a difference in correction factor 12 from power level.
13 In other words, from the 95, from the five 14 percent power upgrade we start changing the correction 15 factor. When you look at that, because of the 16 magnitude the noise that we're seeing, it has an 17 impact in terms of shifting that time delay from one 18 point to another on the curve, and the feedback that
.9 we get from the system.
20 We determined that the noise is caused by 21 pressure pulses in the feed water piping and the 22 pressure pulses in the feed water piping are caused by 23 acoustic resonance.
24 And we've confirmed that both analytically 25 and by taking physical measurements in the plant, at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 ww.nealrgross.com
57 1 least on Byron Unit 2. So we have matched up, with 2 our analytical mind, by our expert in that area in the 3 corporation, along with the actual physical data with 4 very sensitive pressure instrumentation that we hooked 5 up as test.
6 We've validated that we do have these 7 pressure pulses in the feed water line.
8 PARTICIPANT: What's the driver?
9 SPEAKER: We don't know what the driver 10 is. That's part of our continued investigation. So 11 we can confirm that the, yeah, I've got that as my 12 next point here.
13 So I understand that's a question that I 14 get all the time. I wish I could tell you that we 15 knew what the driver was.
16 We have some theories. We believe that 17 maybe pressure dropped across the (inaudible) valve.
18 It could be check valves that we have downstream.
19 If you look on the schematic, the check 20 valves that are the feed water (inaudible) check 21 valves are right downstream of the Venturis and UFMs.
22 It could be associated with the feed water 23 pumps. There are a number of factors. At this point, 24 though, we have not been able to identify the driver.
25 But, we have identified the noise, and we have NEAL-R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
~,4% n-3A AA-772 waczwIlrTAIJ n r nnn57f1 www.nealraross.xom
58 1 confirmed that the noise impacts the correction 2 factor.
3 So I'll talk about corrective action going 4 forward. The corrective action going forward is to 5 install a bracket that is noise-free from an impact on 6 the correction factor standpoint and to verify that 7 status as you go forward and recalculate the 8 (inaudible).
9 PARTICIPANT: Can you talk more about the 10 (inaudible).
11 SPEAKER: We're talking about the AMEG 12 bracket itself. The difference between the common 13 header installation and the individual line.
14 We had an installation on the individual 15 line as part of this project plan we've installed on 16 all four of the Byron and Braidwood units, common 17 header brackets.
18 We say brackets, that's a term that we use 19 for the AMEG (inaudible).
20 PARTICIPANT: (Inaudible).
21 SPEAKER: If you turn to the top of Page 22 24, we're talking about the corrective action. So 23 we've identified the source of the bias on the 24 correction factor.
25 So what I've got listed here are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 www.nealrgross.com
1 corrective actions associated with it. Obviously we 2 removed the correction factors (inaudible), that was 3 done on the 28th at Byron.
4 And what we've done, obviously, is pursue 5 installation of the common feed water header UFM to 6 provide for an alternate calculation, a second 7 correction factor. And that's important for several 8 things.
9 And I'll talk about its importance in our 10 historical evaluation of power condition here in a few 11 pages. Based on criteria provided by Westinghouse in 12 their evaluation, we're revising our appropriate site 13 procedures to check the UFM for noise contamination 14 that could potentially impact the signal.
15 - So we have this criteria that's been 16 provided by Westinghouse. In addition, they will be 17 evaluating, in the foreseeable future, our power 18 installation and our calculations to ensure that we 19 don't have this noise present whenever we would do an 20 ultrasonic flow measurement.
21 And we also have validated that all four 22 of the installations on the common headers, for the 23 four Byron and Braidwood units, are, I say noise free.
24 Obviously all pipes have noise, that's 25 something you can't get rid of. But they're noise NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234-4433 WASHINGTON, D.C. 20005.3701 www.nealrgross.com
60 1 free to the extent of not adversely impacting the 2 calculation (inaudible).
3 PARTICIPANT: They have a common header 4 and you put a UFM on it that measures total flow.
5 SPEAKER: Right.
6 PARTICIPANT: You have UFMs on each 7 individual line and you total that to get, add all to 8 get total flow.
9 SPEAKER: Correct.
10 PARTICIPANT: So to get total flow you've 11 got to add four UFMs, you've got to do that work. And 12 you said that's several days. Now you're comparing, 13 you say well I'm not going to check that with the one 14 common header.
15 Why didn't just the industry start with 16 this, these common header if that gets you total flow?
17 SPEAKER: Well, I can answer part of that.
18 I'm just, based on my review, some like to see the 19 view of this technology, as you are aware, 20 (inaudible).
21 And they're using it in a constant 22 monitoring mode, where there Venturis are basically 23 out of the system. They're feedback to their 24 precision calemetric is directly from AMEG, this type 25 of system.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
61 1 And so they have done that and hooked it 2 up on the four individual feed water lines. In 3 theory, you get better uncertainty now when you have 4 four, four independent lines hooked up as opposed to 5 one and you do the statistical analysis and you divide 6 by the square root of the total number of the loops 7 anid you end up with an overall better uncertainty.
8 So that's the thought process. However, 9 you can qualify the system, the installation, on a 10 common header. And as long as the instrument, and 11 certain analysis that you do, fits within the overall 12 bounds of your safety analysis, which is part of the 13 validation process for installing these and it's 14 acceptable to utilize.
15 PARTICIPANT: So the way you did it, do 16 you factor in (inaudible) to flow line?
17 SPEAKER: That's correct.
18 PARTICIPANT: I'd just like to go back for 19 a moment. Again, on Unit 1, Byron Unit 1, you saw 20 noise on all four lines?
21 SPEAKER: That's correct.
22 PARTICIPANT: But you did not see noise on 23 thecommon header?
24 SPEAKER: That's correct. Noise that 25 impacts the, as I stated,- and I want to be clear, all
'NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
-. -1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
62 1 pipes that have flow through have noise.
2 The noise that we're talking about is the 3 specific contamination that's impacting this signal or 4 a low frequency of a magnitude of such that it's 5 impacting the instrumentation.
6 PARTICIPANT: So what you're saying, then, 7 that the pressure pulses occurred, then obviously, 8 sometime or someplace between the common header and 9 the -- or the individual by itself?
10 SPEAKER: The pressure pulses are present 11 in the individual feed water lines. We've done 12 acoustic monitoring on the, we've done acoustic 13 modeling on the main eed water header and we see 14 resonances from that particular modeling similar to 15 what was on the individual feed water line.
16 At this point, we've not taken the 17 pressure data on the main feed water header. What we 18 do have is an evaluation of the traces from that 19 common header by AMEG using their tool.
20 They have a tool, they call it diagnose, 21 that evaluates and has determined that the signals 22 that they were getting from those common header 23 installations are clear o noise that would impact 24 that correction factor.
25 In addition, we did perform a test this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
63 past weekend on Byron Unit 2. We had our periodic 2 turbine valve/governor valve surveillance which causes 3 us to down power from 100 percent down to 4 approximately 80 percent.
5 And throughout that process we took 6 pressure data and we also monitored the AMEG 7 correction on the, on the common header.
8 And from that we would determine very 9 close correlation between the Venturis and the common 10 header installation, which are the behavior that we 11 did not see previously with respect to the individual 12 feed water line installation.
13 So we've done some additional things to 14 help us validate and have a high degree of confidence 15 in the installation on the common header. Did I 16 answer your question, Marie?
17 PARTICIPANT: I think so.
18 PARTICIPANT: You talked about the vendor 19 having a tool that detects the noise.
20 SPEAKER: I'm sorry?
21 PARTICIPANT: You talked about the vendor 22 having a tool to detect the noise?
23 SPEAKER: Correct.
24 PARTICIPANT: How long has that tool been 25 around?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
,n-) A AA3 /1ACIAWIArTANi n . nnnrV7nl www.nealrarossmm
64 1 PARTICIPANT: We have the tool to use, but 2 we don't use (inaudible). And I believe what happened 3 is (inaudible).
4 (Inaudible).
5 PARTICIPANT: And how long have you had 6 the tool?
7 PARTICIPANT: (Inaudible).
8 PARTICIPANT: So, back in 1999. It's easy 9 to sit here and say, why didn't you use it back then?
10 But it was available if somebody would have thought --
11 PARTICIPANT: (Inaudible).
12 SPEAKER: Based on the information that 13 AMEG has provided, obviously AMEG supports the 14 installation. Their technicians are here working with 15 our personnel.
16 But they really, really oversee and 17 perform the installation of these brackets and do the 18 calibration and testing.
19 The information they provide is they 20 believe that the test may have been run on one loop 21 and it appears that that loop may have been one of the 22 ones that is noise free.
23 And given no indication of a concern going 24 forward. Now, as I stated before, we don't have an 25 acceptance criteria for that, and so from our NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234-4433 WASHINGTON, D.C. 20005.3701 www.nealrgross.com
65 1 standpoint, I'm not defending it.
.2 We were really ignorant of this piece of 3 it.
4 PARTICIPANT: Another way to look at it 5 though, there is a tool out there, somebody though 6 enough to use it at least once.
7 And if (inaudible).
8 SPEAKER: That's a great point, see, but 9 all electronic installations of a similar nature like 10 this have signal noise as an issue.
11 PARTICIPANT: Right. Well, that's another 12 point.
13 PARTICIPANT: Let me just understand that.
14 When you're revising, in the process of revising your 15 procedures to check for the noise?
16 SPEAKER: That's correct. So, in other 17 words --
18 PARTICIPANT: Are you sending the 19 information to --
.20 SPEAKER: That's correct.
21 PARTICIPANT: And so every nine months or 22 when the power decreases at your Venturis (inaudible),
23 during that time is when -you're sending the 24 information?
25 SPEAKER: Correct. The criteria, the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
.1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.neaIrgross.com
66 1 procedures have specific criteria as you stated before 2 where we are required to recalculate friction factor.
3 Whether its a deeo down power a reactor 4 trip, other changes. In the particular case going 5 forward, until AMEG comes up with a tool that we could 6 utilize ourselves, we will have AMEG evaluate the 7 affectibility of our traces for calculation and 8 determination (inaudible).
9 And that will be included in the 10 procedures.
11 PARTICIPANT: Shortly after this was 12 discovered, at the end of August, as the Site Vice 13 President, I was very interested in the question of 14 why in the Spring of 2002, when we did a testing 15 between the common header and the four loops, did we 16 not discover that?
17 And now in August of 2003, we did? I 18 chartered a group to go figure out that question. So, 19 I would like the group to just walk through, and there 20 is some basis of how we got the use this diagnose tool 21 this time around.
22 So, Bill or Ormondo (phonetic), how did we 23 get to use the diagnose this time when we didn't use 24 it last time? I think that's an important point.
25 PARTICIPANT: (Inaudible). First of all, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 ww.nealrgross.com
67 1 we redo our measurement. We want to make sure that 2 they are not affected by (inaudible).
3 (Inaudible). So what happened in the case 4 of the (inaudible). We look at the individual photos 5 and we find that they are (inaudible) significantly 6 compared to our expectation (inaudible).
7 So during the time we (inaudible), and we 8 have one of our engineers actually constantly look at 9 (inaudible) to see (inaudible) factor, we look at the 10 (inaudible).
11 And one of the things they noticed is that 12 the close correlation (inaudible). So combining the 13 information of the (inaudible) because that's the only 14 component that we can look at (inaudible).
15 We look at the (inaudible) and everything 16 else. But we (inaudible). So that (inaudible).
17 PARTICIPANT: I know you are going to go 18 into NRR next week and talk in detail about 19 (inaudible). I can't help myself but ask the 20 question. Is this the first time that you've 21 determined that noise affects the (inaudible)?
22 PARTICIPANT: (Inaudible), this is the 23 first time that we have. We have to check the noise 24 before, but usually there are (inaudible) from the 25 frequency that we are looking.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
68 1 But looking at a (inaudible).
2 (Inaudible). And sometimes we saw a noise, but it was 3 a very sharp (inaudible).
4 PARTICIPANT: What are the possible 5 differences in the units that would cause it to affect 6 one unit more than it would affect another unit.
7 SPEAKER: e do have some differences, and 8 we're still evaluating those. The root cause team has 9 evaluated. We looked at pipe lengths. We've looked 10 at installation.
11 We do have different check valve designs.
12 The plant and the units are more similar than they are 13 different, I guess. We are continuing to evaluate 14 that, but at this point, as I stated previously, we 15 have not been able to firmly identify what the driver 16 is. We just have several theories.
17 PARTICIPANT: Is there something unique 18 about Byron and Braidwood that would cause it to 19 affect them versus any other plants in the industry?
20 SPEAKER: That's a question that think 21 AMEG and others will have to answer. I feel, as I 22 stated before, an obligation to continue to pursue the 23 driver as to what's causing the frequency problem that 24 we're seeing.
25 And with respect to whether this impacts NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
69 1 other plants or not, theoretically, from my 2 standpoint, I believe it's possible.
3 Whether it's fair or not, that's something 4 that AMEG is going to have to answer.
5 PARTICIPANT: We, at AMEG, (inaudible) are 6 reviewing all our installations just to make sure that 7 we didn't have similar problems.
8 We find that (inaudible). (Inaudible).
9 PARTICIPANT: Saying that you found this 10 at Braidwood and Byron, are there any other Exelon 11 sites that use the UFM. So they are installed but 12 they are not currently being used.
13 I guess what my bottom line question is, 14 are you sending other, from other sites, are you 15 sending the information to Westinghouse to check for 16 this noise also?
17 PARTICIPANT: Ann Marie, we have because 18 of the unique configurations, particularly at Dresden 19 and Clinton, because of the status there and their 20 power upgrade, at this point in time, although we have 21 the UFMs installed they have, they're not specifically 22 critical because we're generator-limited on those 23 plants.
24 And at LaSalle(phonetic), when we used 25 them, we found that there was no recovery, so we have
'NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(2021 234-4433 WASHINGTON, D.C. 20005-3701 www.neargross.corn
70 1 continued to use the flow Venturi at that station.
2 So, we'd need to, as we complete the root 3 cause, go back and assess the impact for going forward 4 on those units.
5 However, we don't have an immediate 6 concern on any of those units today, because of the 7 way they're operating.
8 But at the end of the day, we will have to 9 close the loop (inaudible), but they're not in the 10 immediate part of the root cause.
11 PARTICIPANT: Okay.
12 SPEAKER: Let me go back to 24 there, in 13 the middle of the page. Let me kind of summarize here 14 with respect to this.
15 Even though we don't know the driver, we 16 have identified the cause of the bias in the 17 correction factor. And we can reliably detect that 18 noise when it exists.
19 And therefore, we believe that we, I 20 believe we can put in corrective actions to ensure 21 that when we calculate correction factors in the 22 future, that they're accurate and consistent based on 23 that particular piece.
24 But, going forward, prior to us going back 25 to this technology. In other words, reinstituting the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
71 1 system, as Jim stated before, we're going to 2 commission an independent technical review, and this 3 is'separate from the decision making piece now.
4 But the technical review is going to 5 review the root'cause and our corrective action, along 6 'with all of the other physical, aspects of the 7 installation.
8 So the plan -there is to really give a
. hard,' independent technical evaluation of our 10 installation along with assuring that the corrective 11 action, the root cause and- corrective action 12 (inaudible).
13 And then Jim, Jim also talked about the 14 additional team that's been chartered to 15 comprehensively evaluate our decision making on a 16 broader level and over the entire life span of the 17 issue.
18 So, this team will provide the results of 19 their work to the (inaudible)-Officer, and obviously 20 we'll share those findings with you at the appropriate 21 date.
22 PARTICIPANT: But also part of a decision 23 going forward to do that, there will be significant 24 discussion at the corporate office up to an including 25 (inaudible).
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 - www.nealrgross.com
72 1 We've put that hard stuff in place.
2 Again, I don't mean to do a root cause evaluation 3 (inaudible). We talked a lot about the noise 4 (inaudible).
5 What other sorts of root causes or 6 (inaudible) causes have you looked at and haven't 7 dispositioned yet? And, I guess I'm particularly 8 interested I guess in the mechanical side of it.
9 By that is more installation than these 10 accuracies. I heard you've got to install it within 11 a thousandth of an inch.
12 That can go lots of different ways when 13 you're installing the (inaudible) transducers on a 14 piece of pipe. What other root causes of the 15 (inaudible) are you looking at?
16 SPEAKER: That's a very good question, 17 Steve. Obviously over the past three years, we've 18 evaluated these installations intensively. And a lot 19 of the things that you're referring to have been 20 evaluated.
21 As part of this root cause we actually 22 went back and revisited all of those things along with 23 all of the other potentials.
24 We looked at cable affects, software, 25 hardware, obviously the pipe vibrations. You know, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 wyw.nealrgross.com
73 1 inside diameters of pipe, the pipe layout.
2 Other types of issues. So those things 3 were exhaustively evaluated as part of this root 4 cause. So we just didn't stumble on noise and say, 5 hey, we've got it and that's it.
6 Those things are all part of the standard 7 process that we applied in doing the root cause 8 evaluation for this or any other issue.
9 And that's why, as part of a team, when we 10 set that up, we had individuals with specific root 11 cause training and expertise as part of that team.
12 PARTICIPANT: Have you concluded that pipe 13 thickness has no particular affect?
14 SPEAKER: In this particular case, we have 15 concluded that pipe thickness has no particular affect 16 on this particular issue.
17 So, if we could, I'd like to move on to 18 Page 26, and talk a little bit about safety 19 implications.
20 The installation of the common header is 21 important to us, and I'll explain that in a minute.
22 From a historical perspective, we determined that 23 Byron Unit l's over power was limited to 101.6 24 percent.
25 And the basis for that is the difference NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.: -.
rqn> v.U4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
74 1 between our historical maximum correction factor and 2 the current correction factor that we get from the 3 common header.
4 So, in other words, there's about a 1.6 5 percent difference in feed water mass flow when you 6 finalize that calculation. And we have a high degree 7 of confidence that that (inaudible) historical over 8 power condition for Byron Unit 1.
9 PARTICIPANT: Are you saying that's the 10 highest you went to?
11 SPEAKER: That's correct.
12 PARTICIPANT: Okay.
13 SPEAKER: Using the same --
14 PARTICIPANT: If I remember correctly, 15 there have been times when there were no moderations, 16 you've exceeded 100 percent power.
17 SPEAKER: That's correct. And if --
18 PARTICIPANT: So if you add those 19 instances, those excursions, on to this. Have you 20 done that and see what you get?
21 SPEAKER: As part of the, as part of the 22 final review to generate the Licensee Event Report we 23 will evaluate that.
24 I believe, based on our historical review, 25 that there's nothing of significance in there that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 wmm.nealrgross.com
75 1 would fall within anything too far outside of that.
2 But we'll make sure that that gets 3 incorporated in our Licensee Report.
4 PARTICIPANT: (inaudible).
5 SPEAKER: No. Not to my knowledge.
6 PARTICIPANT: Is the 1.6 percent the delta 7 between the Venturi flow and what you got in megawatts 8 thermal? Between where you are now and where you were 9 when you (inaudible) ultrasonic?
10 Or is that, is the 1.6 what you believe 11 would be inaccuracy (inaudiblej?
12 SPEAKER: The 1.6 is the, what we believe 13 to be the inaccuracy or the bias in the correction 14 factor, due to the signal noise.
15 PARTICIPANT: Okay, so you, but you, so 16 you think that if it was working correctly, you would 17 have gained some megawatt (inaudible)?
18 SPEAKER: Correct. We would have gained 19 megawatts, but not many megawatts obviously, where we 20 were at. We had a correction factor on the order of, 21 I'm giving an approximate value now, on the order of 22 two and a half percent.
23 So the, on Unit 1, now I'm talking about.
24 And if you take the data today, it's on the order of 25 one percent. So there's about a one and a half COURTNEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
76 1 percent delta between the individual feed water line 2 contaminated signal calculation and the common header 3 installation with (inaudible).
4 When we used the same analysis for Byron 5 Unit 2, we come up with a determination of the over 6 power limit at 100.4 percent. And for 7 Braidwood Unit 2 on the same basis is limited to 100.3 8 percent.
9 Now as part of this investigation, we have 10 commissioned our corporate nuclear fuel organization 11 along with Westinghouse to re-evaluate our 12 (inaudible). And that review is currently in 13 progress.
14 And those evaluations are being performed 15 with conservative assumptions that (inaudible). And 16 at least at this oint the preliminary evaluation, the 17 results indicate that the applicable safety analysis 18 accepted criteria were met.
19 We expect that work to be completed here 20 in the near term and will be included as part of the 21 (inaudible) renort and (inaudible).
22 PARTICIPANT: Jim, just to illustrate the 23 evaluations that were being performed, two and a half 24 percent or approximate two and a half percent that 25 Brad is talking about from the individual lines back NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 wwv.nealrgross.com
77 1 to the Venturis, we're including that in the analysis 2 plus additional conservatisms for inaccuracies, 3 etcetera.
4 So we wanted to make sure that from a 5 safety analysis standpoint that we were completely 6 bound in limited, even though there's the confidence 7 that the over power was actually limited to 101.6.
8 Follow me there.
9 PARTICIPANT: You're saying you're going 10 to take the worst case, two and a half percent, and 11 say that you were still within your action analysis?
12 PARTICIPANT: Yes. Yes. Actually, it's 13 the 102.5 plus three uncertainty associated with 14 Venturis --
15 PARTICIPANT: Right.
16 PARTICIPANT: -- associated with the 17 (inaudible).
18 PARTICIPANT: (inaudible).
19 PARTICIPANT: Five percent.
20 PARTICIPANT: When you permitted that, did 21 you have to redo any of these analyses or was your 22 original action analysis up to 105 percent 23 (inaudible).
24 SPEAKER: The entire gamut of safety 25 analysis were re-evaluated for a power upgrade.
COUR. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
%A1ACUIdPor-fI n'o nA1fnl77n vwn.a;Irnrn~qqmm
78 1 PARTICIPANT: Up to 105 percent.
2 SPEAKER: One hundred and five percent of 3 the previous condition, that's correct.
4 PARTICIPANT: Right.
5 SPEAKER: Well, uncertainty, what I'm 6 referring to here is Westinghouse is doing a check and 7 evaluating our maximum historical condition, plugging 8 that back into the code and re-evaluating that 9 particular impact to determine what results we get 10 (inaudible) to determine whether we would need the 11 applicable safety analysis (inaudible).
12 So we've talked a little bit about future 13 plans and actions but basically a lot of this is 14 reiteration of previous stuff, but I'll say it again 15 just so it's clear.
16 On Page 28, we have this project plan in 17 progress, installed common feed water header 18 (inaudible) on all our Byron and Braidwood units.
19 We discussed this in our August, or, yeah, 20 August transmittal to the NRC that's laid out to our 21 project plan.
22 We had done testing on those common header 23 UFMs and they are free of noise contamination that 24 could adversely affect the signal.
25 I want to make sure that, again I say, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 vAwl.nealrgross.com
79 1 they all five had noise. But with respect to this 2 specific noise that's 'causing the issue, we have 3 confirmed that the, that currently those common 4 headers are free of that particular noise.
5 And then we're also revising our 6 procedures to ensure that' we evaluate for this 7 particular phenomenon as we go forward to ensure that 8 we recalculate correction factors, that they are 9 accurate and consistent with what we expect.
10 This overall project is scheduled for 11 completion by the Byron and"Braidwood units in the 12 September/October time frame.
'13 We will continue that project plan.
14 Obviously, what we communicated in our previous 15 transmittal, had us going out into 2004, with specific 16 action.
17 Given the events of August, we have 18 obviously accelerated some actions and we are 19 revisiting the types of actions and additional things 20 that we're going to do.
21 But our intent is to continue that project 22 plan to evaluate this in addition to continue to 23 investigate sort of the source of the noise of the 24 driver, because I feel, you know, very strongly about 25 trying to continue to attempt to identify that piece
-- NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D.C. 20005.3701 www.nealrgross.com
80 1 of it.
2 As I stated previously, we're going to 3 perform an independent technical review of the root 4 cause and corrective actions and then we've got this 5 broad review of (inaudible) initiated, and those 6 results will be shared with the NRC.
7 And before we make any decisions to 8 restore the AMEG system, we'll make sure that we have 9 all these pieces in place.
10 PARTICIPANT: Do I understand, when you 11 said the comparison test between the common headers 12 for (inaudible), you did not see the noise. The 13 noise, you don't know if the noise was there?
14 Did you see it? It's interesting to know 15 if the noise, if it's only at the higher power and, or 16 does it start only after the (inaudible)?
17 SPEAKER: We did, I talked about the test 18 that we ran as a test, the data that we took during 19 our scheduled down time (inaudible) this past weekend 20 for the governor valve, turbine valve/governor valve 21 surveillance.
22 We did also take pressure data at that 23 lower condition, and we still saw the noise 24 contamination by the pressure pulses, is what I'd say, 25 in those individual feed water lines.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 www.nealrgross.com
1 The magnitude was different and the 2 frequency shifted somewhat in certain circumstances of 3 some of the resonancy.
4 So we still saw the noise out of the 5 pressure pulses in those lines, but at a different 6 level and magnitude of-frequency.
7 PARTICIPANT: Any time you do a review of 8 what happens (inaudible) the root cause (inaudible).
9 (Inaudible) low frequency pressure pulses at another 10 sit'e.
11 It had EPUs, it had (inaudible) low 12 pressure, low frequency pressure pulses. I 'just 13 wondered, in the big scheme of things, (inaudible) low 14 frequency pressure pulses (inaudible).
15 PARTICIPANT: Regarding the last bullet in 16 the decision making process review. When do you 17 anticipate getting to the point where you'll be ready 18 to share'that with the NRC?
19 PARTICIPANT: 'It's probably premature for 20 me to speculate, but I think that we-will be done, 21 certainly with the data gathering and'the first path 22 analysis I expect'in about the order of a month.
23 But obviously the point at which we say 24 we're satisfied with the results is -going to be a 25 function of what we're'finding. So (inaudible).
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
mnv, of 4AlAA WASHINGTON. D.C. 20005-3701 www.nealrgross.com
82 1 PARTICIPANT: As part of your analysis, 2 technical review, do you plan on any additional 3 testing to be done?
4 SPEAKER: We are D01anning on taking 5 additional data, going forward, on, for Byron Unit 2.
6 And we're also pursuing that for Braidwood, to again, 7 as I have stated several times now, with the intent to 8 figure out what the driver is and also to identify 9 what the differences between the four units may be.
10 PARTICIPANT: Getting back to your 11 questions, within the next two weeks we'll be prepared 12 to establish a date with you when we can come back 13 here and talk about.
14 I think we'll, we should be able to get 15 enough clarity on the endpoint, I think, we owe you 16 that. So within the next two weeks we'll get that 17 date set with you all.
18 PARTICIPANT: Thank you.
19 SPEAKER: If there are no other questions, 20 for Brad, I'll take us to Page 30. As you've heard 21 Brad explain, we believe that we've identified the 22 root cause to be an error in the ultrasonic flow 23 measurement correction factor that was caused by 24 signal noise.
25 And that the signal noise was caused by NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 mmw.nealrgross.com
83 1 pressure pulses in the'feed water pipe. It was not an 2 easily identified, it took our multi-disciplined root 3 cause team, with thermal hydraulics, system 4 performance, INC and AMEG experts to pull that-level 5 of detail together.
6 And in order to solidify our understanding 7 of this root cause we will, as we've said, conduct an 8 independent technical evaluation of the root cause 9 conclusions using outside resources.
10 Our preliminary evaluation determined the 11 safety significance of the over'power to be.'low and 12 bounded by the relevant safety analyses. However, 13 that conclusion is not intended to minimize the 14 significance that weire assigning to this issue.
"'15 It continues to have the attention of our 16 entire organization, up to and including Mr.
17 Scolds(phonetic) 'and Mr. Kingsley(phonetic).
18 Our plans for future implementation of the 19 ultrasonic flow meter will be deter-mined after the 20 performance of that independent technical review of 21 the root cause.
22 The results will be discussed with our 23 senior management to determine the next steps. And we 24 will communicate those plans to you when they are 25 established.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS a 1323 RHODE ISLAND AVE.. N.W.
('=)l 91Ai4mA WASHINGTON. D.C. 20005-3701 www.nealrgross.com
84 2 As I stated earlier, although we put a 2 significant effort into trying to determine the cause 3 of this problem over the last four year period, we are 4 disappointed that we didn't get to the answer much 5 sooner.
6 We've all discussed the team that was 7 chartered by Jack Scolds to conduct the broader 8 evaluation and decision making, and from that team we 9 expect to better the missed opportunities and the 10 lessons learned, and to share those with you as we 11 just agreed.
12 Pending oher questions or comments, this 13 ends our presentation.
14 PARTICIPANT: I have a question. It's 15 going to take some time, and I understand it takes 16 some time to thoroughly examine the decision making.
17 Are there any lessons that you've learned, 18 anything that you've internalized at this point, 19 compensatory measures, if you will, however you want 20 to call or consider them.
21 Anything that's impacting current decision 22 making that you've taken from this so far?
23 SPEAKER: We've gone through other 24 situations that we have at our operating power plants 25 and have not yet tumbled on any that we, based on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
85 1 lessons learned out of this, we feel like we to 2 immediately go backwards.
3 I will, where it does come up more 4 prevalently is pending decisions we have. I'll give 5 you a simple example. One of the projects that we're 6 evaluating to reduce (inaudible) at (inaudible) is 7 ultrasonic fuel cleaning.
8 So we're taking some of the lessons 9 learned that, although we haven't stepped our way all 10 the way through the process, it is obvious how we are 11 treating technology that perhaps we are not routinely 12 conversant in.
13 Flow measurement using ultrasonics is one.
14 Effective ultrasonics on a fuel pin would be another.
15 So we are taking these lessons learned and applying 16 them toward actions that are pending in the short term 17 to the best expense that we're able to.
18 It's going to be an ongoing process.
19 Because everyday that we learn more about this, causes 20 us to go back and revisit either past decisions or 21 decisions that we'll may be go through in the near 22 term, prior to completion of the other initiative.
23 PARTICIPANT: That fuel cleaning process, 24 is that different than the one that was used 25 Hungary?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
PnM 'aA.AATA WARHINTN D.C. 20005-3701 www.nealrpross.com
86 1 SPEAKER: It's, it is similar in concept 2 to that, and used at Calloway(phonetic) in south Texas 3 and such. But its application on the type of fuel, 4 specifically the boiler fuel, will be a first of a 5 kind at (inaudible).
6 PARTICIP.NT: So you're also taking the 7 lessons learned from other places as well?
8 SPEAKER: Oh yeah, yes.
9 PARTICIPANT: Are there other questions?
10 Let me ask Bill or George, on the phone, any questions 11 from you?
12 PARTICIPANT: This is George, I have one 13 question. When Exelon reported that they thought that 14 Byron Unit 1 was over Dower, and it was a combination 15 of the contamination of the signals from the 16 individual feed water instruments as well as the fact 17 that the, some of the parts did not equal the whole 18 when they compared it to the common feed water header.
19 The, my understanding is that there had 20 been some continuous monitoring of the data of the 21 individual feed water legs for a while, prior to 22 adding the UFM on the common header.
23 And my question is were there indications 24 in those signals, at that time, that may have given 25 the clue that there was something irregular in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
87 1 feed water measurements with the UFM?
2 SPEAKER: They have taken several months 3 of data, one significant note, it's (inaudible) power 4 is we have not seen any change in the correction 5 factor.
6 So one thing we were looking for was this 7 change, and we had not seen it. Now an important 8 piece of this plan was to get this correction factor 9 data on the individual loop as we begin coast down, 10 which we now have.
11 Now I have not heard whether or not we've 12 seen changes due to the coast down we're currently 13 experiencing. But correction factors have been steady 14 as a result of this data being taken for a long period 15 of time.
16 PARTICIPANT: Okay, thank you. I don't 17 have any other questions.
18 PARTICIPANT: Okay, Bill? Okay.. Any 19 other questions from the NRC? A couple of logistic 20 things before I turn it over to Jim for the closing.
21 That presumption that there is no 22 proprietary information in what you've given us and 23 you've prepared it for a public meeting, so:do you 24 want to verify that? Thank you.
25 And for members of the public, we will be NEAL R. GROSS COURT. REPORTERS AND TRANSCRIBERS
- 1323 RHODE ISLAND AVE., N.W.
Innfl "A AAV: %A1AJUIKlrT't n l ru )nnnr.'7ni
- un.A. n r nre* enm
88 1 available after Jim closes this portion of the 2 meeting, for questions and answers, and we do have the 3 public feedback forms available.
4 PARTICIPANT: The NRC will be available.
5 PARTICIPANT: And with that, Jim.
6 SPEAKER: Well, we appreciate you coming 7 over. I certainly appreciate you being open and 8 honest because we asked a pointed and hard questions 9 and you guys didn't get, you answered the question the 10 way it occurred and I appreciate that.
11 I hope the next session we have, which is 12 a little more difficult discussion, can be the same 13 way. Because that's what we're looking for. Just to 14 get to the bottom of this so we can move forward.
15 So I do appreciate the information we 16 gained today, and I know you'll be going to 17 Headquarters and they'll probably get into a lot more 18 technical details (inaudible).
19 You indicated that you were disappointed 20 in the time it took to figure this out, obviously.
21 Echo that. There was a lot of discussions that have 22 been going on over the past year and a half within 23 this region and with Headquarters, it's gotten all the 24 way to the highest levels in the Agency.
25 And questions on what to do, how to get NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
89 1 where we need to get to and concern about, you know, 2 we typically want to make sure that we understand what 3 we're doing and that we have a good foundation for 4 decisions that we make.
5 And as you know, it took you a while to 6 get to this root cause. But we were, saying we don't 7 have some of the same expertise you have,' it was 8 difficult for us-to figure out'how to react.
9 And in fact, we rely on you as the
- 10. Licensees to, because you do have that expertise to 11 make the correct decisions.
12 And in this'case, I think, I won't preempt 13 your review, but we're not conservative. There's a
'14 couple of things, I asked the question about the 15 statement, only to try to bring out a point.
16 And that is something that we've talked 17 about on several other issues and, with Exelon plants
- 18. and that is that we'look to see, for the approach to 19 be you, it's your plant, it's your resource.
20 We look to see that 'you are looking for, 21 is this right, and then justifying'why, as opposed to 22 justifying why you're right before determining whether 23 you are or not.
24 And I'm not, and I'll wait and see-what 25 you have to say, but that's the way it appears-to us NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W..
{on?% aA. WAqiNrTnN n . Vnnn.37ni vww n mrn.cs.r m
90 1 now that you wanted this to be correctly and that's 2 what you ended up doing is justifying it.
3 And, in fact, I asked the other question 4 about your practice of taking a non-conservative 5 decision, even though you weren't exactly sure it was 6 right.
7 You had good, you know, it was, you felt 8 it was reasonably okay because everything you tried 9 did not indicate it was a problem.
10 But maybe to ask the question more 11 pointedly, going forward would your expectation be 12 that you would take the normal practice review, take 13 the conservative approach and then decide whether it's 14 a problem or not before you move forward.
15 That's a fair question. Is that --
16 PARTICIPANT: Yeah, I think certainly our 17 expectation of ourselves is, given all the information 18 (inaudible) analyzing and a responsible job of 19 gathering information that we won't proceed, whatever 20 form perceived may take, in a direction such as 21 raising power or extending runs, those types of 22 things, without a thorough understanding of what the 23 issues are and a fundamental belief that we are 24 correct.
25 With that said, do understand the nature NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 wwvi.nealrgross.com
91 1 of your questions and certainly; as we've gone through 2 this, some of the documentation'appears that we don't 3 (inaudible), like the impression it leaves is we don't
.4 feel good about this,' but we can't find anything 5 wrong.*
6 And at what point should-that turn to a "7 -positive action and drop to a position where we can 8 distinctly say, we may not be where we're going to end 9 up, but we're at a position today that we're very 10 comfortable.
11 PARTICIPANT- -As'you know, we just do a 12 sampling and we g-et, we do a smart sample, but we do
-13 'a sampling of what's'going on.
14 And we do have gut feel, on things, but in
15 the end,' we rely on you guys'to make the right 16 decisions. And to'be honest with you, Exelon has done 17 a very good job in the past-few years of making the 18 right decisions.- -
19 Just a few examples we've had an
- 20 opportunity to talk about here, and'that's what the 21 -iindustry obviously is looking'for and certainly what 22 the Agency is looking for.
23 We're struggling-through a situation with
. .2 24 another facility in this region that went completely 25 the other direction and we're looking at our processes
- NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
- 1323 RHODE ISLAND AVE., N.W.
runm% 71w
' A %WAINtTt Ntr .sn fnnnFl7n1 : www nPnIRnrnsrtnm
92 1 to see what we can do to analyze things that are 2 difficult to understand.
3 And we won't need to do that if we can 4 trust you to always do the right thing. People make 5 mistakes, but if you're always thinking to do the 6 right thing, then it's not necessary for us to be 7 sitting back saying, all right, are we being, did we 8 get accurate information or are they trying to pull 9 our leg or what?
10 Now, we're all skeptics on this side of 11 the table, that's how we got to doing what we're 12 doing, but, so we'll continue to ask those questions.
13 But, in the end, it doesn't do any good 14 for any of us to take the approach where, let's see 15 what we can get by with. We may not always get it, 16 but we get there at some point.
17 And by that time, sometimes it's too late, 18 as evidenced by the plant out east. In any case, I 19 did appreciate this. I thought you guys were very 20 open and honest with the information that you've 21 provided, and I look forward to the next meeting so we 22 can discuss that portion of it.
23 Because that's probably the part that 24 concerns us the most. One last thing. I know this 25 wasn't safety significant, but you did get EPU, which NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 www.nealrgross.com
93 1 you got an extended power uprate because you had 2 margin in your plant to be able to do that.
3 And then on top of that we were concerned, 4 you know, you were doing something that was beyond 5 that which should have given it a little bit more of 6 a significance.
7 Even though I understand it's not really 8 safety significant, it was at the bad time to be doing 9 something that got you above your 100 percent. But 10 thank you very much for coming and we look forward to 11 the next meeting.
12 PARTICIPANT: We'll take just a couple of 13 minutes to let folks gather their things, if they're 14 choosing to leave at this point, and then we'll open
,15 it up for the NRC to take questions.
16 We'll start first here in the room, if 17 there's anyone that would like comments or questions 18 and then we'll go to the phone. So just give us a 19 couple of moments, please.
20 (Asides away from microphone.)
21 PARTICIPANT: Okay, for those that are 22 still with us, we'll open it up for any questions for 23 the NRC? Anyone on the phone?
24 PARTICIPANT: Yes, Greg (inaudible) from 25 NRC, but I have no questions.
.-NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
94 1 PARTICIPANT: Okay, thank you very much.
2 Anyone else? Okay. Sir, anything? Okay, thank you.
3 That concludes it then, we have no public questions.
4 Thank you.
5 (Whereupon, the foregoing matter was 6 concluded.)
7 8
9 10 11 12 13 14 1s 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 wmv.nealrgross.com
CERTIFICATE This is to certify that the attached proceedings before the United States Nuclear Regulatory Commission in the matter of:
Name of Proceeding: Public Meeting on FOIA/PA-2003-0444 Docket Number: n/a Location: n/a were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and, thereafter reduced to typewriting by me or under the direction of the court reporting company, and that the transcript is a true and accurate record of the foregoing proceedings as recorded on tape(s) provided by the NRC.
Jamie BtiffU Official Transcriber Neal R. Gross & Co., Inc.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 wvw.nealrgross.com