ML26012A015
| ML26012A015 | |
| Person / Time | |
|---|---|
| Issue date: | 12/04/2025 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| NRC-0499 | |
| Download: ML26012A015 (0) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Docket Number:
(n/a)
Location:
Rockville, Maryland Date:
Thursday, December 4, 2025 Work Order No.:
NRC-0499 Pages 1-90 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1716 14th Street, N.W.
Washington, D.C. 20009 (202) 234-4433
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
1 2
3 DISCLAIMER 4
5 6
UNITED STATES NUCLEAR REGULATORY COMMISSIONS 7
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 8
9 10 The contents of this transcript of the 11 proceeding of the United States Nuclear Regulatory 12 Commission Advisory Committee on Reactor Safeguards, 13 as reported herein, is a record of the discussions 14 recorded at the meeting.
15 16 This transcript has not been reviewed, 17 corrected, and edited, and it may contain 18 inaccuracies.
19 20 21 22 23
1 UNITED STATES OF AMERICA 1
NUCLEAR REGULATORY COMMISSION 2
+ + + + +
3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
(ACRS) 5
+ + + + +
6 731ST MEETING 7
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 8
(ACRS) 9
+ + + + +
10 OPEN SESSION 11
+ + + + +
12 THURSDAY 13 DECEMBER 4, 2025 14
+ + + + +
15 The Subcommittee met via Video-16 Teleconference, at 8:32 a.m. EST, Walter Kirchner, 17 Chair, presiding.
18 MEMBERS PRESENT:
19 WALTER KIRCHNER, Chair 20 GREGORY H. HALNON, Vice Chair 21 VICKI M. BIER, Member 22 CRAIG D. HARRINGTON, Member 23 ANNIE KAMMERER, Member 24 ROBERT P. MARTIN, Member 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
2 SCOTT P. PALMTAG, Member 1
DAVID A. PETTI, Member 2
THOMAS E. ROBERTS, Member 3
MATTHEW SUNSERI, Member 4
5 DESIGNATED FEDERAL OFFICIAL:
8 ACRS CONSULTANT 9
RONALD BALLINGER 10 ALSO PRESENT:
11 KATHY BARNES, Public Participant 12 ALAN BLIND, Public Participant 13 JESSE DEER, Public Participant 14 JACQUELYN DRECHSLER, Public Participant 15 ARNIE GUNDERSON, Public Participant 16 ANDREW JOHNSON, NRR 17 KEVIN KAMPS, Public Participant 18 MICHAEL KEEGAN, Public Participant 19 PAUL KLEIN, NRR 20 APRIL NGUYEN, Region III 21 KARL RABENHORST, Public Participant 22 KRAIG SCHULTZ, Public Participant 23 ANN SCOTT, Public Participant 24 SPENCER TOOHILL, Public Participant 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
3 P-R-O-C-E-E-D-I-N-G-S 1
8:32 a.m.
2 CHAIR KIRCHNER: This morning's topic is 3
the Palisades Steam Generators Operational Assessment.
4 We have received quite a few written comments from 5
interested members of the public and at least one 6
request to make an oral statement from Mr. Arnie 7
Gunderson. There will be opportunity for these 8
statements and other public input after the committee 9
has heard from the staff on the subject.
10 And with those introductory remarks, I 11 will now turn to my colleague, Greg Halnon, who is 12 vice chairman of the ACRS and our lead member on this 13 topic. Greg.
14 VICE CHAIR HALNON: Thank you, Chair 15 Kirchner. During our 728th full committee meeting on 16 September 3rd through the 5th, we discussed the 17 Palisades restart activities. Since the documentation 18 was not available at the time, the operational 19 assessment of the steam generators will be discussed 20 today as one topic, one sole topic, of this morning's 21 meeting session.
22 As part of the interaction with the 23 committee, the NRC staff will present and discuss 24 their thoughts on the subject and the document. Once 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
4 this is complete, we will have approximately 15 1
minutes for public comments. Unless previously 2
allocated more time, comments will be limited to two 3
minutes to ensure that as many individuals as possible 4
have the opportunity to speak about the operational 5
assessment of the Palisades steam generators. As you 6
make your comments, I do ask that you please focus 7
your comments accordingly.
8 The committee will then discuss what we've 9
heard today, and I anticipate that we will provide a 10 short write-up through our deliberations of the 11 meeting's summaries.
12 So with that, I'll turn the presentation 13 over to Paul Klein of the staff. Paul.
14 MR. KLEIN: Thank you. Good morning, 15 ACRS. Paul Klein from NRR staff. My colleague seated 16 off to my right here is Andrew Johnson, and we're both 17 in the Division of New and Renewed Licenses in the 18 Corrosion and Steam Generator Branch, and we're happy 19 to be here this morning presenting the results of the 20 Palisades steam generator OA.
21 CHAIR KIRCHNER: Paul, just pull your 22 microphone a little closer.
23 MR. KLEIN: Is that better?
24 CHAIR KIRCHNER: Yes.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
5 MR. KLEIN: Thank you. We just want to 1
put a list of the acronyms upfront to help members of 2
the public that might be joining and those that aren't 3
as familiar with general nuclear and more steam 4
generator-specific acronyms to better follow the 5
discussion this morning.
6 Some of the ones on the list that we'll 7
probably touch on the most during the course of this 8
morning's presentation on the left side would be CM, 9
which is condition monitoring; eggcrate, which is the 10 term for the horizontal lattice tube supports in the 11 steam generators at Palisades. On the right side, 12 we'll be speaking quite a bit about SCC, or stress 13 corrosion
- cracking, and OA, the operational 14 assessment, which is the forward-looking projection of 15 tube integrity to the next inspection.
16 Next slide. So we recognize the focus for 17 today is the Palisades Steam Generator Operational 18 Assessment for Cycle 29. We did want to start, 19 however, with a summary slide to provide a high-level 20 perspective upfront of the Palisades work as a whole.
21 And since it's been a few months since September, 22 since we last presented at ACRS, we thought it would 23 also be beneficial if the OA discussion was preceded 24 by some slides that relate to the steam generator 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
6 design to some of the results from the tube 1
inspections, the subsequent tube sleeving; and then 2
we'll follow that up with some of the condition 3
monitoring and operational assessment work that was 4
done. And then we'll wrap-up with some information 5
related to recent steam-generator chemical cleaning at 6
Palisades and provide concluding remarks.
7 So this slide is really intended to be a 8
high-level summary, and we'll get into more details 9
later on in the presentation. We previously have 10 briefed the committee about degradation was detected 11 during 1D28 at Palisades, the D designating a 12 defueling instead of a refueling outage. The big 13 surprise during the steam generator tube inspection 14 was detection of over 1,200 axial ODSCC indications at 15 the horizontal eggcrate supports. And because in 16 steam generator A, if all of the tubes with ODSCC 17 indications at the supports would have been plugged, 18 it would exceed the 15-percent plugging design limit 19 in the tech specs, so the licensee instead submitted 20 a license amendment request to install sleeves as a 21 repair method in lieu of plugging; so, at this point, 22 there's been a total of almost 300 sleeves placed that 23 will be -- oh, thank you. Sorry. Almost 3,000 24 sleeves will be placed in service to preserve the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
7 plugging margin. And it's worth noting that, of that 1
almost 3,000, about 900 of them are what they term 2
corrective sleeves, which means they were placed over 3
locations where ODSCC was present. The other, 4
roughly, 2,000 were preventative sleeves that tended 5
to be placed in locations where cracking could occur 6
in the future. And we can discuss that strategy in a 7
couple of slides when we see the steam generator 8
schematic.
9 In terms of condition monitoring that 10 showed that tube integrity was maintained, as part of 11 that effort, 23 indications passed in-situ pressure 12 testing as part of the condition monitoring. There 13 was no burst or leakage detected.
14 And we did want to correct one mistake 15 that was on preliminary slides. In that third bullet, 16 I think the preliminary slides had 23 tubes. It was 17 actually 23 indications and 22 tubes. So I just 18 wanted to note that.
19 VICE CHAIR HALNON: This is Greg. Before 20 you go on, can you describe what the no burst or 21 leakage criteria is? What is that test and how is it 22 done?
23 MR. KLEIN: Well, as part of the in-situ 24 pressure test, there's usually two points of interest.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
8 One would be a pressure representative of main 1
steamline break, and so you would be wanting to check 2
for any potential leakage at that point.
3 And then the burst would be to test 4
against the structural integrity performance criteria, 5
which, in this case, was three times the normal 6
operating pressure differential. And then they add 7
some margin for related to testing at ambient 8
temperature and also for the account from the AGR and 9
other type things.
10 VICE CHAIR HALNON: And these indications 11 that they tested, they were worst case or at least 12 indications that would be expected to --
13 MR. KLEIN: Yes, they are the worst-case 14 indications. In terms of the eggcrate supports, they 15 would be the ones that, analytically, would be most 16 likely to burst, if that were to happen during one of 17 the tests.
18 And then for the circumferential 19 indications, they were tested for leakage. At the top 20 of the tube
- sheet, there were a
number of 21 circumferential indications that we talk about on a 22 later slide that were tested for leakage. And then if 23 you detect leakage, then you would test those for 24 burst at that point; but there was no leakage during 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
9 any of those tests or with the other tests.
1 VICE CHAIR HALNON: So just in summary, 2
the licensee took the worst-case known cracks, 3
subjected them to pressure three times the design 4
pressure, and there was no leakage or burst; is that 5
correct?
6 MR. KLEIN: That is correct for the 7
eggcrate indications for the circumferential ones at 8
the top of the tube sheet that were tested to the main 9
steamline break conditions. And since there was no 10 leakage at that point, there was no need to test them 11 to burst.
12 VICE CHAIR HALNON: But just to clarify, 13 you said three times the design pressure. Is it three 14 times the normal operating normal pressure? Okay.
15 Slightly less than the design. Well, not slightly.
16 Less than.
17 MR. KLEIN: That's correct.
18 VICE CHAIR HALNON: Yes. Okay. Thank 19 you.
20 MR. KLEIN: It's a good clarification.
21 MEMBER HARRINGTON: This is Craig. It's 22 a differential pressure.
23 VICE CHAIR HALNON: Differential.
24 MR. KLEIN: All right. Should I continue?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
10 For the last bullet here we're on, Cycle 29 1
Operational Assessment, we have Revision 2 that we 2
took a look at. That Revision 2 was for 1.5 effective 3
full-power years.
4 The bottom line, and we'll discuss this in 5
more detail later, the analysis results did meet all 6
the performance criteria. There was a little margin 7
for eggcrate structural integrity performance 8
criterion. But there are some very key conservative 9
assumptions as part of that analysis. One is that the 10 crack growth rates at the eggcrates are based on 11 assuming all cracking at 1D28 was just below the 12 detection threshold level at 1R27. And we'll discuss 13 that later, but that is clearly a conservative 14 assumption. And it also assumes no improvements from 15 the chemical cleaning that was just performed on both 16 steam generators.
17 We did have a verification call with the 18 licensee. And based on that call and some of the 19 discussions, they indicated that a OA revision is in 20 progress, and we're expecting to receive a Revision 3 21 at some point here in the near future.
22 This just shows the Palisade's Model 2530 23 steam generators. I think, in one of our earlier 24 presentations, we were asked if we could show some 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
11 schematics to give the committee a
better 1
understanding of the generators. And as you can see 2
on the left side, the horizontal supports are labeled, 3
once you get above the tube sheet level, 1H up to 8H.
4 The actual sleeves were inserted from 01H up to 05H.
5 And in terms of the preventative strategy that they 6
used, because, once you sleeve a tube, you can't get 7
in in the future to a higher elevation to sleeve at a 8
higher support, they tended to sleeve from 01H to 05H 9
and a lot of the tubes where they had indications.
10 And that's how they ended up with approximately twice 11 as many preventative sleeves as corrective type 12 sleeves.
13 Overall, the steam generator has 8,219 14 tubes, 0.75-inch diameter, 0.042-inch wall thickness, 15 combination of lower row U bends and then higher row 16 square bends, as shown on the right here. And I think 17 that is most of the highlights we want to cover.
18 This next one shows schematically on the 19 left side a more blown-up schematic of the upper 20 support structure, the vertical supports, or vertical 21 straps are sometimes also called. And then you can 22 also see the diagonal bar hot and diagonal bar cold in 23 that schematic, and that's referred to oftentimes as 24 a batwing support. And then the lattice type support 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
12 on the bottom right of the left side, which then on 1
the right side schematically lays out the eggcrate-2 type support arrangement. I think we can move on.
3 VICE CHAIR HALNON: I think we have time 4
to set for clarification on this chart you just put 5
up. There's the vast majority of the axial educations 6
are actually two supports. Go back to the previous 7
picture. Where in this picture were these cracks 8
within the -- I'm sorry -- the length of this? You 9
know, on the right you got the eggcrate, and where are 10 the cracks?
11 MR. KLEIN: So they would be within that 12 width of that a eggcrate support on the tube in an 13 axial direction. You can imagine there would be a 14 number of deposits around the tube. And this 15 schematically also doesn't show the tube density, so 16 the actual tube bundle would be a 0.75-inch tube 17 arranged on a 1-inch triangular pitch so that there's 18 a much denser packing of tubes for heat-transfer 19 purposes.
20 VICE CHAIR HALNON: Okay. Thank you. So 21 the cracks tend to occur where the deposits that are 22 trapped by those various structural injuries?
23 MR. KLEIN: That's correct. The deposits 24 help to trap impurities, which then also promote 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
13 cracking at those locations.
1 So this slide's a summary of some of the 2
more important SCC indications and volumetric wear 3
indications at support structures that exceeded the 4
40-percent wear plug-in criteria during the 1D28 5
inspection. In terms of the total number of SCC 6
cracks, you can see that the tube supports, by far, 7
have the highest numbers. In terms of the actual 8
wear, we're just showing the indications that exceeded 9
40 percent, but, of course, there's actually thousands 10 of wear indications, which is not unusual for steam 11 generators to see some type of wear at support 12 structures.
13 MEMBER HARRINGTON: This is Craig. The 14 high number of wear locations in the condition 15 monitoring table like this, that's not just newly 16 observed ones, that's cumulative over time?
17 MR. KLEIN: Yes, that's correct. So in 18 the condition monitoring, you would assess all the 19 indications, all the historical ones, plus any new 20 ones. And you, typically, for a generator that's been 21 in service for a number of years where it tends to 22 attenuate over time and so the wear growth rates drop 23 over time and it becomes a stable situation. And I 24 think that's reflected by the relatively few number of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
14 indications that you see that were greater than 40 1
percent here.
2 MEMBER MARTIN: Question here. Bob 3
Martin. Previously, when you visited us, I think you 4
talked about some cleaning activities. I mean, I 5
can't remember whether they had to clean at the time 6
or plan or, I mean, obviously you've already, you 7
know, if you haven't said it, you've implied debris 8
matters and contributes to the stress corrosion 9
cracking at these eggcrate locations. They've gone in 10 and cleaned them all up so that at least they're ready 11 to go.
12 MR. KLEIN: Yes. So last time we were 13 here, that was a planned activity at the plant. And, 14 at this point, it has been done and we have a slide 15 that will speak to that later on in the presentation.
16 But, yes, they did a chemical cleaning and there was 17 thousands of pounds of deposits removed from the 18 generators.
19 MEMBER MARTIN: Okay.
20 MR. KLEIN: I think the final thing I'll 21 highlight here is shown in red, and that would be the 22 effective plugging percent. So that that would 23 represent the tubes that were plugged, plus accounting 24 for the sleeves that were installed, to come up with 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
15 an effective plugging percent. And you can see where 1
the steam generator A and B are relative to the 15-2 percent criteria.
3 CHAIR KIRCHNER:
Can we ask the 4
participants to mute their phones and computers.
5 Thank you.
6 PARTICIPANT: But could we ask the 7
speakers to say who they are? Thank you.
8 CHAIR KIRCHNER: Okay. We're just, again, 9
we're having a presentation from the staff. It's Paul 10 Klein and Andrew Johnson. Go ahead.
11 MR. KLEIN: All right. Thank you. So 12 this slide talks about the condition monitoring. I 13 just want to remind everyone that that is a backward-14 looking process where you determine if you have 15 maintained tube integrity during operation up until 16 the inspection point. So it uses the inspection 17 results, and you assess whether your tube integrity 18 performance criteria were met.
19 So, typically, you pre-establish CM limits 20 for each degradation mechanism and location so that 21 you can rapidly evaluate that during the inspection 22 process. Once you begin to receive any EE results, 23 the flaws that are detected can either clearly meet 24 the CM limit with initial ND sizing, some of the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
16 larger flaws might require enhanced A-current flaw 1
sizing, which is also termed profiling, to determine 2
if condition monitoring is met. And then if you're 3
analytically unable to demonstrate that condition 4
monitoring was met, the next step would be the in-situ 5
pressure testing.
6 So all tubes met condition monitoring 7
during 1D28. We previously mentioned there were 23 8
indications that did require in-situ pressure testing, 9
and there's a breakdown below the third bullet, that 10 the locations of those 23 were at eggcrate supports, 11 there were two axial indications at the top of the 12 tube sheet, 11 circumferential indications at the top 13 of the tube sheet, one axial ODSCC at a diagonal bar 14 hot, and there was one obstructed tube that was tested 15 since they couldn't pass the eddy current probe 16 through the tube. And as we previously mentioned, no 17 leakage or tube burst or an in-situ pressure test.
18 MEMBER PALMTAG: This is Scott Palmtag.
19 So I have a question on the timeline. So the plant 20 was laid up, and I assume that condition monitoring 21 was done after they decided to restart in terms of the 22 timelines.
23 MR. KLEIN: Yes. So the condition 24 monitoring would have been done following the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
17 inspection that was performed.
1 MEMBER PALMTAG: Okay. And then, in a 2
condition monitoring report, it says degradation 3
exceeded what was expected, and there were some 4
questions about how the plant was laid up after these 5
latest inspections. Is there any indication of, you 6
know, why there was -- degradation was exceeded or 7
expectations were exceeded?
8 MR. KLEIN: Well, I think they clearly 9
under-predicted the amount of cracking that was 10 occurring at the eggcrate supports. If the question 11 is whether it would occur at temperature in the last 12 cycle or occur during shutdown; is that what you're 13 asking?
14 MEMBER PALMTAG: Or why, if they could 15 have some indication of why this occurred.
16 MR. KLEIN: I mean, I think it occurred 17 because of the axial ODSCC, right? So the question is 18 when it occurred. And I think, you know, the last 19 time we were before the committee, we made a statement 20 that we couldn't for a certain rollout. There was 21 some contribution during shutdown, but the more that 22 we thought about it and the more we've looked into 23 some of the precursor eddy current signals, it seems 24 more likely to us that degradation occurred at 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
18 temperature. And we'll talk about that in a couple of 1
slides here why we think that.
2 MEMBER PALMTAG: Okay. That's important 3
because, you know, now you have a better explanation 4
of why this occurred over shutdown. All right. Thank 5
you.
6 MR. KLEIN: Okay. This schematic shows an 7
example of how the more detailed eddy current or flaw 8
profiling is used as part of the condition monitoring 9
process. So this is a snapshot of condition 10 monitoring of steam generator B for axial ODSCC at 11 eggcrate supports. And if you look on the left chart, 12 each of those data points would indicate an indication 13 that was detected in size by eddy current. On the 14 right side, this would show that same data but now 15 with the benefit of the flaw profiling. So you would 16 essentially take each slice of data and try to develop 17 a flaw profile so that you can better map the actual 18 crack profile. And when you do that, you'll 19 oftentimes find that initial sizing oversized the flaw 20 and you can see that there's relatively few flaws at 21 this point at or above that condition monitoring line, 22 which is shown by the red dashed line. And those 23 indications and the similar ones that would occur in 24 steam generator A then would be candidates for in-situ 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
19 pressure testing.
1 VICE CHAIR HALNON: This is Greg Halnon.
2 Real quick. The flaw characterization and the 3
profiling, is that a new process that Palisades came 4
up with because of these generators, or is it part of 5
a well-established industry process that every PWR 6
goes through?
7 MR.
KLEIN:
It's part of a
8 well-established process that every PWR would use as 9
part of any inspection of their steam generators.
10 VICE CHAIR HALNON: So this is not a new 11 methodology or unproven methodology?
12 MR. KLEIN: Not at all.
13 VICE CHAIR HALNON: Thank you.
14 MEMBER ROBERTS: Paul, this is Tom 15 Roberts. Can you speak to the condition monitoring 16 limit? Does that assume one cycle of operation after 17 the inspection, or is that based on the time at the 18 inspection, or is it something else?
19 MR. KLEIN: The condition monitoring limit 20 is backward looking, so it's not assuming any forward 21 projection of one cycle or two cycles or anything. So 22 it's really to say, at this point in time, based on 23 your previous operation, did you maintain tube 24 integrity?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
20 And so for a given mechanism at a given 1
location, they'll develop a combination of flaw depth 2
and length that would meet that condition monitoring.
3 And so for a different top of tube sheet indication 4
that was axial, you'd see a different condition 5
monitoring line.
6 And I think it's important to point out 7
that the condition monitoring line is also a 8
conservative line. So it's not intended to indicate 9
that flaws above that line would necessarily lead to 10 burst. And I think that's confirmed by the fact that 11 they had 23 indications that were above the condition 12 monitoring line tested, and there was no leakage or 13 burst from any of those flaws.
14 So it's intended to represent a
15 conservative flaw above which you need to pay more 16 attention to and below which you're confident that it 17 met tube integrity.
18 MEMBER ROBERTS: Right. But it's some 19 other analysis that projects forward in time and 20 acceptability. This just says that, as of today, they 21 do expect to operate, but you have to do more analysis 22 to conclude where they are in a cycle or two cycles.
23 MR. KLEIN: Yes. So the second half of 24 the process, the operational assessment that we're 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
21 going to get to in a few slides, that that is the 1
forward projection of tube integrity until the next 2
inspection point.
3 MEMBER ROBERTS: Okay. Thank you.
4 MR. KLEIN: The next slide here shows a 5
profile of outside diameter deposits from steam 6
generator A. And in this case, this schematic shows 7
deposit loading of 20 mils and thicker, as shown in 8
the yellow highlights on the left chart here. And so 9
deposit mapping is done in most plants to understand 10 the condition of the tubes and to determine when it 11 might be appropriate to do chemical cleaning. And I 12 think what you see here on the left side shows that 13 the highest deposit loading tended to be from above 14 the third support up to about the seventh there.
15 The right part of this slide, which was 16 taken from the apparent cause analysis from the plant, 17 shows the support number with the bottom support, 01H 18 up to 05H, in terms of the number or percentage of 19 total eggcrate cracks that occurred at that location.
20 So you can see from the chart to the right compared to 21 the highest deposit loading on the left, the chart on 22 the right suggests that it's a temperature-driven 23 mechanism, which is consistent with cracking at 24 temperature at the eggcrates. The 01H and 02H have 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
22 combined about six out of every ten eggcrate cracks in 1
the steam generator.
2 So that leads us into the operational 3
assessment part, which we just had a question on a 4
minute ago. So that is the forward-looking 5
projection, and that provides a technical basis for 6
tube integrity until the next steam generator 7
inspection. And that's done for all tube degradation 8
mechanisms, so it includes not only the SCC at the 9
eggcrates, but anywhere in the steam generator plus 10 wear structures, such as vertical straps or diagonal 11 bars.
12 This process is a pretty mature process, 13 and it is addressed in the EPRI guidelines. And 14 degradation can be projected using a number of 15 different techniques, from fully deterministic to 16 fully probabilistic, and we'll cover these methods in 17 a little more detail in the Palisades-specific results 18 coming up in the next few slides.
19 CHAIR KIRCHNER: Paul, this is Walt 20 Kirchner. Just going back one slide, on that -- I 21 know you have a slide coming up on the chemical 22 cleaning. So going back to those deposits, those are 23 mainly on what I'll call the hot side of the U-tube.
24 MR. KLEIN: Yes, that's correct.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
23 CHAIR KIRCHNER: Yes, at the lower support 1
levels for the eggcrates, so to reinforce your 2
temperature-driven mechanism.
3 MR. KLEIN: Yes. I think in terms of this 4
particular cut, which was for 20 mils and thicker, I 5
think it's shown that the highest deposits are 6
actually a little bit higher than where you're seeing 7
the most cracking. So they don't necessarily line up.
8 CHAIR KIRCHNER: They don't line up on 9
through one.
10 MR. KLEIN: So that would suggest that 11 temperature has an important part, which is typical 12 for SCC and steam generators.
13 CHAIR KIRCHNER: I know you're coming up 14 to this, but, post-cleaning, was another scan done to 15 look at deposit, depositions?
16 MR. KLEIN: I don't know if they've gone 17 through that step or not.
18 CHAIR KIRCHNER: Thank you.
19 MR. KLEIN: So in terms of OA options, 20 there's different approaches that are available. One 21 would be a full bundle probabilistic approach. I 22 won't read all the details here. Some of the more 23 simplified techniques include arithmetic, simplified 24 statistical mixed techniques, which are a combination 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
24 of the above. And, typically, two integrity engineers 1
start with the most conservative and simplest and then 2
work their way towards a probabilistic, if needed.
3 So this schematic is taken from the EPRI 4
Integrity Assessment Guidelines and, since that is a 5
proprietary document, I want to note that the staff 6
did obtain written permission from EPRI to present the 7
slide in this ACRS meeting since it is open to members 8
of the public. Having said that, the idea shown here 9
is that you go through and perform a number of 10 iterations on your full bundle. And for each run, you 11 have an outcome that is shown as a plot here and then, 12 in red, it shows the worst possible tube. And after 13 you go through a number of these Monte Carlo 14 iterations, you can construct a plot on the lower 15 right of all the worst cases from each of your runs.
16 And then they take the lower fifth worst pressure and 17 compare that to the Structural Integrity Performance 18 Criteria in order to determine if the analysis was 19 successful or not, or met the criteria I should say.
20 Next slide. We are aware that ACRS did 21 receive some public comments challenging the use of 22 the 95/50 acceptance criteria for steam generators, 23 instead of a 95/95, so we thought we would provide 24 some comments on that. And first is that the industry 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
25 consensus for steam generator OAs is the 95/50, and 1
it's been that way for quite some time. There's an 2
EPRI report on the Technical Basis for Tube Integrity 3
Performance Acceptance Standards from 2006 that 4
discusses that, and it's incorporated into the EPRI 5
Integrity Assessment Guidelines.
6 The public commenter did provide some 7
instances of where 95/95 was used as an acceptance 8
criteria and that typically was related to either 9
leakage or dose, not the Structural Integrity 10 Performance Criteria. There are some cases like the 11 H amendments that relate to tube cracking deep within 12 the tube sheet where a 95/95 criteria was used for the 13 alloy 600 TT fleet, and there are individual cases 14 before the consensus approach was developed.
15 The 95/50 criterion has been in use now 16 for quite some time, 20 years, and it has been shown 17 to be very effective. And we do want to note that 18 that criterion is used in conjunction with the 19 deterministic safety factor, so either 3 times the 20 normal operating pressure differential or 1.4 times 21 the design basis accident level, depending on what's 22 more conservative for a given unit. But using that 23 95/50 in conjunction with the deterministic safety 24 factor does provide multiple layers of protection.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
26 So we'll turn it now to --
1 MEMBER ROBERTS: This is Tom Roberts. I 2
was wondering if you could give some perspective on 3
the 95/50. One interpretation is that there's a 50/50 4
chance that 5 percent of your tubes will rupture. Is 5
that the right interpretation? Obviously, there's 6
conservatism in the conditions under which that would 7
happen, but is that the right way to interpret it or 8
should you interpret it some other way?
9 MR. KLEIN: No, you should not interpret 10 it that way. It's related to whether you're going to 11 meet the three times the normal operating pressure 12 differential structural integrity performance 13 criterion, so it's a very conservative value that 14 you're comparing against.
15 MEMBER ROBERTS: Right. Yes. That's what 16 I think I'm trying to say is that if you had 17 conditions of 3 times the normal operating pressure 18 differential or 1.4 times the DBA pressure, then 19 there's a 50/50 shot that 5 percent of your tubes will 20 rupture. Is that the right interpretation, or is that 21 not right? I'm expanding the conservatism of the 22 pressure margin.
23 MR. JOHNSON: This is Andrew Johnson. I 24 would say that you still have a probability of 0.95 of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
27 meeting 3 delta P or 1.4 delta P.
1 MEMBER ROBERTS: Sure. If there's 1,000 2
affected tubes, that would lead to, roughly, 50 would 3
be the mean value of the number of ruptures at those 4
conditions, if I'm understanding this right, put a 5
50/50 chance. So, in reality, it's the 3 times, you 6
know, operating differential pressure and the 1.4 7
times actual pressure that's really the margin; is 8
that right? But you don't expect to meet those 9
conditions in the real world.
10 VICE CHAIR HALNON: This is Greg. I 11 thought that was like a confidence level, not a 12 deterministic number of tubes. You don't take 1,000 13 tubes and put 0.95 on it and multiply them together 14 and get the answer. It's 95-percent confidence level 15 or 50-percent confidence. It's 50-percent confidence 16 that there's a 95-percent chance the tube won't 17 rupture. That's how I interpret it. That's not a 18 chance. It's a probability.
19 MEMBER ROBERTS: Right. But if you apply 20 it 1,000 times, like flipping a coin with a 0.5 21 probability, you get heads half the time.
22 MEMBER MARTIN: It depends on how the 23 analysis is done. If it's a simple analysis, it might 24 just be one tube. That has a 95 percent with a 50-25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
28 percent confidence. But, anyway, obviously, it 1
doesn't appear you know the details of how the 2
statistics was applied. But 5 percent of tubes, one 3
tube either way, it's a pretty conservative type 4
perspective. Three times differential pressure.
5 MEMBER ROBERTS: Well, conservative is in 6
the 3 times --
7 MEMBER MARTIN: That's a margin kind of 8
argument there.
9 MEMBER ROBERTS: Do you have any sense of 10 what the probability would be at a more reasonable 11 level of pressure? Is there some sort of extrapolated 12 curve that you've seen or could judge?
13 MR. JOHNSON: This is Andrew Johnson. So 14 you say a more reasonable pressure. I guess are you 15 saying a pressure less than 3 delta P?
16 MEMBER ROBERTS: Yes.
17 MR. JOHNSON: Well, then, I mean, clearly, 18 the probability would be much lower if you're going to 19 have any problems.
20 MEMBER MARTIN: You've got to think of a 21 linear no-threshold theory. That might be a 22 conservative model.
23 MEMBER ROBERTS: Yes, exactly. I was 24 thinking about it a lot from the EPRI document a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
29 couple of slides ago. That might give some, you know, 1
perspective on how quickly that probability falls off 2
as the pressure goes down. Is that, like, a legal way 3
to look at it? Because that curve fell off pretty 4
quickly. Going from 3,500 to 3,000 psi, at least from 5
the readability of the curve, it comes to basically no 6
probability, no chance of -- so that would -- again, 7
just try to get a perspective. If you had 1.4 margin 8
of design basis acting on differential pressure, that 9
1.4 would be quite a bit to the right of that bottom 10 red point with a 5-percent degree of the anchor.
11
- Anyway, I
was just trying to get 12 perspective on this. My sense is that you're not 13 right on the cliff edge. I'm just trying to get an 14 understanding, you know, whether the pressure margin 15 really does support that view, you know, given the 16 margin and the pressure that you use for things.
17 Again, this plot is a representative plot, I assume, 18 not specific to palisades, but it would seem to 19 support that the probability would be extremely low at 20 design basis pressure differential or something closer 21 to the operational differential pressure. Again, I 22 was hoping to see if I understood that right.
23 MR. KLEIN: It seems to me you have the 24 right understanding.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
30 MEMBER ROBERTS: Even though, if you were 1
to get the pressures at a 95/50, you would, you know, 2
presumably be in, I guess, a hard-to-justify condition 3
because there would be a lot of tubes potentially 4
rupturing with a large population of available tubes.
5 MR. KLEIN: I don't think we can provide 6
exact probabilities, but I think you can take comfort 7
in the fact that they tested the worst flaws from the 8
steam generators and they all met the 3 delta p 9
criteria. So that's one measure of the conservative 10 nature of the approach. Another is that since this 11 approach has been implemented industry-wide there's 12 been no tube ruptures.
13 MR. JOHNSON: And this is Andrew Johnson 14 again. I guess another point that might give you 15 confidence is that historical testing performed many 16 years ago that we have records of showed that, you 17 know, brand-new virgin tubes, when tested for rupture, 18 they were, the vast majority, rupturing around 10,000 19 PSI. So that's significantly more than what we're 20 looking at here.
21 MEMBER ROBERTS: Okay. Thank you. The 22 conservatism and the strong capability in addition to 23 conservatism in the assumed pressures. Okay. Thank 24 you.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
31 MEMBER MARTIN: Another point here, like, 1
how many would go? Once you have one go, the pressure 2
drop or the differential pressure would, of course, 3
drop, as well. So you probably wouldn't have a 4
cascading type condition, you know, 5 percent or 5
whatever. It would probably be one, maybe two, 6
depending on how rapid these things were. And, of 7
course, that's still within, I think, operating 8
history fleet, having one.
9 VICE CHAIR HALNON: That would be 10 undesirable.
11 MEMBER MARTIN: Undesirable.
12 VICE CHAIR HALNON: Let's establish that.
13 This is Greg. We're not advocating that it's okay to 14 have one rupture.
15 MEMBER MARTIN: No, but it has happened.
16 VICE CHAIR HALNON: True. It has 17 happened.
18 MEMBER HARRINGTON: And this is Craig.
19 It's also, just to reiterate the point, this is in 20 condition monitoring space. The next step after, not 21 strict timeline but this doesn't mean that the tube 22 passed and, therefore, it necessarily stays in 23 service. Some of those that were evaluated would 24 still be maybe taken out of service or sleeved.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
32 So condition monitoring doesn't say the 1
tube is fine to continue operating necessarily. It 2
depends on the degradation mechanism and other people.
3 MR. KLEIN: Well, it's probably good, at 4
this point, to mention that, in terms of the approach 5
that's taken, it's plug-on detection for cracks. So 6
due to the challenge of sizing stress corrosion 7
cracks, you know, the conservative approach is taken 8
in industry, and that's to plug-on detection. So any 9
of those indications that you saw in that CM plot 10 would either be sleeved or the tube plugged.
11 MR. JOHNSON: And this is Andrew Johnson.
12 As a point of clarification, each plant has to have 13 specific approved procedures to sleeve, right, or 14 perform an alternate repair criteria. If they don't 15 have that specifically approved, then they have to 16 plug.
17 MR. KLEIN: Slide 15. In terms of the 18 Palisade's specific operational assessment shown here 19 in the two bullets or the degradation mechanisms that 20 were addressed by a fully probabilistic OA approach 21 and those that use some type of mixed arithmetic Monte 22 Carlo approach.
23 So this slide kind of gives a summary of 24 Revision 2 Cycle 29 OA for Palisades, which was for a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
33 1.5 EFPY duration. Probabilistic full tube bundle 1
method was used for most of the degradation 2
mechanisms, including ODSCC, at the eggcrate supports.
3 And then what we tried to show was the lower 95th 4
percentile burst pressure versus the three times NOPD 5
acceptance criteria for some of the mechanisms here.
6 For the top three, I'll address probabilistically.
7 And for the U-bend, it was a mixed arithmetic Monte 8
Carlo approach.
9 It's worth probably noting for the one of 10 most interest, the ODSCC at eggcrates, they simulated 11 ten outages in the modeling that was performed, so ten 12 outages 50,000 times, in order to address that 13 particular mechanism. And we included the U-bend 14 stuff just because there was some discussion the last 15 time we were here with respect to Indian Point tube 16 rupture, but I don't think we need to spend too much 17 more time on that unless you have questions.
18 VICE CHAIR HALNON: So, Paul, this is 19 Greg. Earlier, you said that the probabilistic method 20 was used if, for lack of a better term, simpler 21 methods could not be utilized. Is that implying here 22 that this was a very complicated set of indications 23 such that probabilistic had to be relied on to 24 characterize these?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
34 MR. KLEIN: I mean, it's common for plants 1
to use probabilistic assessment for a number of 2
mechanisms. So I would say the complicating factor 3
here that they had to deal with in modeling was just 4
the unanticipated jump in the number of indications at 5
the support plates, and the next few slides will get 6
into how they try to account for that. And so that 7
was probably one of the more difficult parts of the OA 8
model. But the use of probabilistic approach is not 9
unusual at all in OAs.
10 VICE CHAIR HALNON: So the type of 11 indications and the actual characterization of 12 indications was not unusual. It was just the amount, 13 the volume, of the step change and the numbers; is 14 that correct?
15 MR. KLEIN: That is correct. I think the 16 OA process is a mature process. I think there are 17 options that can be used for each mechanism, and, 18 because of the large number, they did make some 19 assumptions that we'll discuss here in the next few 20 slides to try and fit the Weibull function to the 21 results when they wanted to benchmark the results of 22 1D28.
23 VICE CHAIR HALNON: So would you say 24 there's nothing unusual about the indications other 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
35 than the increased numbers? I mean, the numbers, we 1
didn't find any mechanisms or things we didn't 2
understand with the eddy currents?
3 MR. KLEIN: No. I would say axial ODSCC 4
at the eggcrate supports and combustion engineering 5
plants is a well-known degradation mechanism.
6 So the next slide kind of speaks a little 7
bit to what we just discussed in terms of what was 8
different for Palisades. And so what we tried to do 9
on this slide was just show some of the degradation 10 mechanisms on that left column, and then we compared 11 the projections from the previous OA. So how many of 12 each of these degradation mechanisms were expected at 13 1D28, if you will, and then the projection from 14 Revision 2 of the current OA for 1R29.
15 And so you can see for most of these 16 mechanisms, there's some increment increase in the 17 number that's projected, which is not unusual as the 18 plant ages, particularly a 600-millimole plant. But 19 for the first row here, the axial ODSCC, you can see 20 a significant underprediction on the number of 21 indications. And so that really was probably the one 22 thing that created the greatest challenge in terms of 23 updating your model so that you reflected the 1D28 24 special results.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
36 MEMBER HARRINGTON: This is Craig again.
1 In the projections from the previous outage, why was 2
the axial PWSCC in the tubesheet not performed? Is it 3
because they hadn't seen that condition yet or --
4 MR. KLEIN: They had seen it, but it had 5
been intermittent and not a great number. And so I 6
think, in this particular outage, they found more 7
PWSCC indications within the tubesheet compared to 8
historical. And so, because of that, you see that 9
reflected in the projection then for 1R29. And that 10 may be -- this is just speculation, so I should make 11 that clear. But some of the detections may have been 12 because of the cleaning that was done in the RCS at 13 the plant and maybe that allowed for improved 14 detectability. I guess the next inspection or two 15 will determine whether that was a one-time jump or 16 whether it's part of a trend.
17 MEMBER PALMTAG: This is Scott Palmtag.
18 I just want to clarify, 1292 indications is a high 19 number; but that's what you use to decide whether you 20 sleeve it, or is that the number of indications after 21 you sleeve it?
22 MR. KLEIN: That's the total number of NDE 23 indications. So in both steam generator A and steam 24 generator B, that's the total number of ODSCC 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
37 indications that the eggcrate supports.
1 MEMBER PALMTAG: Before sleeving or after 2
sleeving?
3 MR. KLEIN: Well, it's the same number of 4
indications, so that number caused them to pursue 5
sleeving instead of just plugging each of those tubes.
6 MEMBER PALMTAG: Right. Okay. So now you 7
don't expect to get 1292 because you sleeved, because 8
you did sleeving?
9 MR. KLEIN: No. The sleeving should not 10 affect that because locations that were sleeved as a 11 preventative method, you already have a crack there.
12 In terms of sleeve locations that were preventative, 13 you could develop a crack in those locations in the 14 future, but, since they're already sleeved, there's no 15 safety concern with that.
16 MR. JOHNSON: Paul, you said preventative 17 the first time. You meant correct.
18 MR. KLEIN: Oh, thank you, Andy. Yes. I 19 don't know if I addressed your question. You look 20 like I didn't address your question properly.
21 MEMBER HARRINGTON: Scott, this is Craig 22 again. That is a projected number. That's a big 23 number from condition monitoring, as well, but that's 24 not necessarily a real number. It's a projected 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
38 number.
1 MEMBER PALMTAG: It was projected, and 2
then they sleeved, or is this projected after?
3 MEMBER HARRINGTON: The sleeving was based 4
on the condition monitoring and other things.
5 MEMBER PALMTAG: Okay. So some of the 6
projections even with this is for next outage.
7 MEMBER HARRINGTON: What they might see 8
next outage as part of their operational assessment to 9
make sure that they can go, you know, to determine 10 when they have to inspect again, in part, and that 11 they can make it at least to the next inspection 12 without having the failure.
13 MEMBER PALMTAG: Okay. Thanks. That 14 answers the question.
15 VICE CHAIR HALNON: So this is Greg. In 16 a perfect world then, in 1R29, the condition 17 monitoring would say you have 1292 indications that 18 you have to assess going forward for the next 19 operational assessment.
20 MR. JOHNSON: This is Andrew Johnson. You 21 could, you know, if your model was exactly right and 22 matched reality, you would find 1292 in the next 23 inspection. You could very easily find 292 in the 24 next inspection or you could find more, right? You 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
39 just -- this is a model.
1 MR. BALLINGER: This is Ron Ballinger, a 2
consultant. Were that to actually happen, and they 3
were to sleeve every one of them, what would the 4
margin be?
5 MR. JOHNSON: The margin to what, sir?
6 MR. BALLINGER: The 15 percent. I haven't 7
done the math.
8 CHAIR KIRCHNER: You're at 11 and 6.4. So 9
you're currently -- effective plugging for steam 10 generator A is 11.1 and B is 6.4. This is, roughly, 11 1,000 tubes out of 16,000 total, so you're probably 12 still under your 15 percent, but it depends which 13 steam generator.
14 MR. JOHNSON: Right. Well, and you also 15 have to consider that there's a variable number that 16 affects sleeving, right?
17 CHAIR KIRCHNER: Right, yes.
18 MR. JOHNSON: When you first start 19 installing sleeves, every 10 or so sleeves is 20 equivalent to plugging one tube. As you keep 21 installing more sleeves, that number goes up.
22 Eventually, it goes to 11, then a little bit more than 23 11, right? As you keep restricting flow, right, each 24 additional sleeve has a greater effect. But that's 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
40 kind of the range, between 10 to 12 sleeves, per plug.
1 MR. BALLINGER: You also have an imbalance 2
in flow now between the steam generators.
3 MR. JOHNSON: Well, that's true. And 4
there's a certain limit between what you can have 5
between the generators. That's right.
6 MEMBER PETTI: This is Dave. You can have 7
more than one indication in a given tube.
8 MR. JOHNSON: That's true.
9 MEMBER PETTI: So, again, that's not 10 necessarily directly translatable.
11 MR. JOHNSON: Right.
12 MEMBER HARRINGTON: And, of course, the 13 other -- this is Craig. The other key point is that 14 the operational assessment helps assure that you can 15 operate safely until the next inspection. If you 16 arrived in the condition that Ron's describing where 17 you have to respond to all these and sleeve them and 18 maybe you approach your plugging limit, that's an 19 economic problem. It's a life of plant problem, but 20 it's still not a safety problem because you haven't 21 tripped the safety wire.
22 MEMBER ROBERTS: Tom Roberts. What's the 23 role of the proactive sleeving in this projection?
24 Presumably, all the tubes they thought would show 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
41 cracking or most likely should crack in the next 1
cycle, they'd proactively sleeve. Is this 1292 in 2
addition to this, or does that count all tubes, 3
including those that are sleeved?
4 MR. KLEIN: So I think there's a little 5
confusion between the sleeving and the projections for 6
the number of crack indications at the next outage.
7 They probably don't necessarily overlay with each 8
other. So in terms of the approach for the leaving, 9
the idea was if you -- let's say most of your cracks 10 were at O1H and O2H, well, you can put sleeves in one 11 of those locations, like O1H. And then if you did an 12 inspection at the next outage and you found a crack at 13 O2H, for example, you would need to plug the tube at 14 that point. So a lot of the sleeving that was 15 performed would quickly become null and void at the 16 next inspection, so they tried to go up to the next 17 few elevations and put a sleeve in place so that, if 18 cracking were to occur at that location, it would 19 already be sleeved and the tube could remain in 20 service.
21 MEMBER ROBERTS: So that number is not 22 indicative of cases that are unsleeved? That's all 23 tubes?
24 MR. KLEIN: That number is all tubes, yes.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
42 CHAIR KIRCHNER: So that's just kind of a 1
statistic --
2 MR. KLEIN: I think if we get to two more 3
slides, we'll be able to see the projection on a plot 4
and maybe it will make more sense.
5 CHAIR KIRCHNER: Before you go on, Paul, 6
could you just loop back and address this? So, 7
roughly, there were 1,000 tubes that were corrected 8
sleeves and 2,000 preventative sleeves. So would 9
those preventative sleeving of tubes address these 10 projected indications? Is that where they did the 11 preventative sleeve installation?
12 MR. KLEIN: They did the preventative 13 installation in tubes that already had an eggcrate 14 crack indication. If you look at the tube bundle as 15 a whole, you don't really know where your next cracks 16 may show up, so it's really difficult to try to look 17 at the projections on total number indications and 18 relate that to sleeve or unsleeved location.
19 CHAIR KIRCHNER: This is Walt again. The 20 preventative sleeving was if you find a crack now at 21 01H, the preventative sleeve above that because you 22 can't do it later, and so it's not that they said, 23 well, that's a tube we're going to preventatively 24 sleeve. It was a tube with a problem and there's --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
43 yes, understood.
1 MEMBER MARTIN: This is Bob Martin. A 2
number of these questions kind of get to the heart of 3
the model you have. I assume it's relatively simple.
4 Numbers like 1292 is more of a statistical derived.
5 Actually, it's not a 3D, you know, solid model. You 6
don't have all of that. So, I mean, you're nodding 7
your head.
8 MR. JOHNSON: This is Andrew Johnson.
9 You're absolutely correct. That number 1292 is a 10 purely probabilistic number. It's derived. It has no 11 relevance to any location in the generator.
12 MEMBER MARTIN: And as you described, it's 13 a conservative model overall. You know, whether it's 14 1292 or 2, I mean, it really could be anywhere. But, 15 you know, you take a conservative approach to kind of 16 assess where you're at from there and whether actions 17 taken at the site are appropriate.
18 MR. KLEIN: I think it's probably worth 19 following up that comment to note, earlier, it was 20 asked if the model was exactly correct to be 1292 21 indications at the next inspection. 0Ideally, the 22 tube integrity engineer wants to model conservatively, 23 so you want a model such that your number of actual 24 indications is less than your projections. And if you 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
44 look at the non-eggcrate locations where there's 1
cracking in the generator, you can actually see the 2
plots where they've consistently been able to 3
overestimate the number of indications and the model 4
continues to be conservative over time.
5 And so, in this case, in the eggcrates, 6
they needed to make an adjustment in modeling. And 7
the expectation is those adjustments would create 8
conservative predictions moving forward.
9 MEMBER HARRINGTON: Right. This is Craig 10 again, too. It's noteworthy steam generator B in the 11 CM and what half the number of indications for the 12 projection, they took higher numbers for steam 13 generator A and applied them to B. It's conservative.
14 MEMBER MARTIN: And this would drive maybe 15 a unique inspection or early inspection, or what 16 decision would be coming out of this analysis?
17 MR. KLEIN: So this particular analysis 18 was for 1.5 EFPY. Later on, we're going to discuss --
19 and we've mentioned several times that they are in the 20 process of revising, so it's possible that that 1.5 21 EFPY could become shorter.
22 Okay. So moving on, a little more 23 specifics about the modeling of ODSCC at the 24 eggcrates. So the way that they treated it, they 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
45 modeled it with the Weibull function, starting with 1
1R27, and then the model parameters were adjusted.
2 You typically iterate these models until you benchmark 3
them or you're able to reproduce the current 4
inspection in your model, and then it's considered 5
benchmarked.
6 They made a very conservative assumption 7
that, to fit the large increase in detected 8
indications in 1D28, they treated outages prior to 9
1R27 as having perfect ODSCC detection. In other 10 words, all ODSCC cracks at eggcrates in those prior 11 outages were perfectly detected by NDE and taken out 12 of service.
13 MEMBER BIER: Excuse me. A question on 14 that. I just want to make sure I'm understanding 15 properly. My mic seems to have gotten much more 16 sensitive for some reason. Sorry. Vicki Bier.
17 So you're saying this is conservative 18 because the new indications are treated as 19 representing brand new problems and, hence, a large 20 increase, whereas, in reality, some of them probably 21 pre-existed and the increase was less dramatic; is 22 that correct?
23 MR.
KLEIN:
That is correct.
24 Historically, with stress corrosion cracks in steam 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
46 generators, it takes some time before they reach an 1
level where the eddy current will pick them up.
2 And then the other conservative assumption 3
that they made with respect to growth rate was that 4
all flaws detected on 1D28 were at the threshold of 5
detection in 1R27. And the result of that assumption 6
is that it results in a very large average growth rate 7
of 9 percent through wall per EFPY. And if you 8
compare that, the EPRI integrity assessment guidelines 9
have two sets of growth rates for alloy 600 mill and 10 yield tubing. And those growth rates and integrity 11 assessment guidelines are based on a total of about 35 12 plants over a period of decades, so it's a pretty 13 robust database.
14 And so if you took the EPRI default 15 average growth rate and adjusted that for the 16 Palisades operating temperature, which is on the lower 17 side, it would suggest a 2.15 percent through wall per 18 EFPY. And I think in the last OA, I think they 19 modeled the average growth rate at this location as 3 20 percent through wall per EFPY. That 9 percent is also 21 greater than the growth rate that EPRI uses to 22 conservatively bound any alloy 600 unit or 23 temperature, and I'm not going to provide that number 24 since it's proprietary.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
47 MEMBER HARRINGTON: This is Craig. I 1
assume, though, that that assumes that cracks were 2
right at the threshold of detection. But if they had 3
not even initiated yet, the growth rate would be 4
remarkably higher, right?
5 MR. KLEIN: If the cracks detected in 1D28 6
had not initiated at 1R27, it would be extremely high 7
growth rates. And I don't think that's very 8
plausible.
9 MEMBER HARRINGTON: Yes, unrealistic.
10 MR. KLEIN: It's unrealistic for a number 11 of reasons, including the fact that, when they went 12 into the 1D28 locations and then performed any current 13 look back, they could see precursor signals in prior 14 outages. And so as they went further back in time in 15 those outages, they could see that the percentages of 16 those detections dropped over time. But that would 17 suggest a growing population of flaws that eventually 18 reach a detection level during 1D28 were detected. So 19 this is not as conservative as it could be, but it's 20 judged reasonably conservative under the circumstances 21 of --
22 MR. KLEIN: I think it's conservative 23 because, if you look at the Palisades T-hot 24 temperature from the last operating cycle, it's about 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
48 583. And that conservative EPRI upper bound, I mean, 1
you can apply that to plants that are operating at any 2
temperature, so, say, 650 degrees, for example. So to 3
exceed that upper bound rate with your average, to me, 4
is a conservative approach.
5 MEMBER HARRINGTON: I don't know about 6
ODSCC or PWSCC. That's a very susceptible temperature 7
range, but, okay, sufficiently conservative under the 8
circumstances. Thanks.
9 MEMBER ROBERTS: Hey, Paul, Tom Roberts.
10 What is the role of the sleeving to this analysis? Is 11 the sleeving accredited, or is this assuming that 12 these are, you know, cracks that grow on unsleeved 13 tubes?
14 MR. KLEIN: Well, I think cracks can grow 15 on both sleeved or unsleeved tubes. I mean, really, 16 the sleeving is on the ID, and it's put there to 17 maintain tube integrity. The cracks are propagated 18 from the outside based on the chemistry associated 19 with deposits, et cetera.
20 MEMBER ROBERTS: I'm trying to understand 21 why it matters if the tube is sleeved. If the tube is 22 sleeved, doesn't the sleeve take the structural --
23 MR. KLEIN: Well, if you have a tube that 24 is sleeved and then a crack develops at that location 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
49 behind it in the future, yes, you've already accounted 1
for that, and the sleeve is the pressure boundary. So 2
you're right.
3 MEMBER ROBERTS: So this analysis then 4
must assume correct crack growth in unsleeved tubes?
5 What you said, if the tube is sleeved, it doesn't 6
matter. That's what I'm trying to understand.
7 MR. KLEIN: I don't think the modeling, 8
again, I don't think the modeling is trying -- it's 9
taking a population, which is the entire tube bundle, 10 and it's taking the number of detections from the 11 current inspection, and then it's projecting that 12 forward, and sleeving is not part of that modeling 13 assessment.
14 MEMBER ROBERTS: So that would seem to be 15 a significant conservatism; is that right? It's 16 essentially assuming that you haven't sleeved 17 anything, and so you're susceptible to all of the 18 crack growth in your confidence model, whereas, in 19 reality, well, the sleeved tubes have been protected.
20 MR. KLEIN: Okay. I would consider some 21 of these other items listed here as more conservatisms 22 than that, but that could be considered as some of 23 those locations where they might develop are already 24 pre-sleeved.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
50 MEMBER HARRINGTON: Yes. This is Craig.
1 It would be a statistical conservatism, but not a 2
mechanistic conservatism basically.
3 MR. KLEIN: Correct. So this slide is 4
figure 610 from the operational assessment, and this 5
shows the probabilistic modeling for steam generator 6
A, which, of course, bounds steam generator B. And so 7
what you see on the plot here are a number of 8
different inspections shown by the red dots. And if 9
you look at prior to 1R27, which is shown here on the 10 vertical line, it's close to about 24 EFPY. So prior 11 to that it assumed perfect detection, and then at 1R27 12 is when it first initiated. You can see in the green 13 triangle here shown at the top that the model now 14 initiates a large number of new indications, cycle 15 between 27 and 28. And because of that, you then have 16 this large increase in number of indications that is 17 benchmarked by the model here shown around EFPY 26.
18 You can see that the predicted detections and actual 19 detections match at that point, and that's part of the 20 benchmarking process.
21 And then so the third vertical line to the 22 right then would be the projection out to the next 23 inspection point, and you can see what the model is 24 predicting in terms of number of detections at that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
51 point, which is a significant increase from the 1
current inspection.
2 So in terms of staff comments for the 3
Section 2908, the OA did meet all the acceptance 4
criteria for the 1.5 EFPY. We didn't know that there 5
was a small margin for meeting the SIPC at the 6
eggcrate supports. We noted some of the conservatism 7
is already about the average SCC growth rate, assuming 8
that all cracks are present at the detection threshold 9
and only grew during the last operating cycle.
10 One of the questions that was outstanding 11 was what if the cracking occurred also during the 12 extended shutdown when the chemistry was being 13 controlled. So if you included that as part of your 14 model that cracking occurred both over that last 15 operating cycle and/or shutdown, that would give you 16 a longer time from crack growth, which would result in 17 a lower number and also assumes no benefits from 18 chemical cleaning. And we'll talk about chemical 19 cleaning, I think, in the next slide or two slides 20 maybe.
21 We did have a clarification call with the 22 licensee and their vendor on the 14th of October. The 23 purpose of that was to obtain clarification on 24 portions of the Cycle 29 modeling. So, as you know, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
52 these are not simple documents to generate or to 1
review. And so we did have a number of questions just 2
to have a better understanding of their modeling 3
approach and the OA model that was used in this case, 4
the assumptions that were made. And, in particular, 5
we asked about the upper tail growth and how well 6
their model accounted for the upper tail growth that 7
was seen during the 1D28 inspection with the 8
assumptions that they made. And we also talked to him 9
about, you know, whether they had done sensitivity 10 analysis to understand how much small changes and 11 different assumptions would affect the overall OA 12 results.
13 Based on those discussions, the licensee 14 vendor is performing additional sensitivity analysis.
15 A new OA will be issued and our understanding is that, 16 at this point, it has been received at the site. The 17 NRC staff has not yet had a chance to take a look at 18 it.
19 MEMBER PETTI: Paul, this is Dave. Just 20 a question. The issue of the chemical cleaning and 21 any benefits, the models don't take that into account, 22 or is it that, since there's sensitivity study, they 23 could look at where they kind of normalize things?
24 MR. KLEIN: Yes, I don't want to speculate 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
53 about what the Rev 3 is actually going to do because 1
they may make a number of changes. But the Rev 2 that 2
we reviewed, there's no credit for chemical cleaning 3
and that's definitely --
4 MEMBER PETTI: I thought you, in your 5
discussions with them, you wanted them to pursue some 6
additional sensitivity analyses. I mean, like the 7
chemical cleaning, is the model that sophisticated 8
that it could take that into account?
9 MR. KLEIN: Well, the way you could try to 10 take that into account, and I'm not suggesting that 11 they are, is there might be a change in the number of 12 initiations or the crack growth rates. Later on, in 13 the chemical cleaning sides, we're going to show that 14 they took a significant amount of copper out of the 15 generators, and so that should have a beneficial 16 effect in terms of initiations and growth rates. But 17 they're modeling based on what was detected in 1D28 18 without any effect from chemical cleaning.
19 CHAIR KIRCHNER: Paul, before you go on, 20 could you elaborate on bullet two on this slide?
21 Small margin for structural integrity performance 22 criterion.
23 MR. KLEIN: Yes. I think that's 24 reflective of the 95th lower value for burst from the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
54 probabilistic modeling is above the SIPC value, but 1
it's not above that value by much. And so that's 2
something that we pay attention to. But having said 3
that, you could make less conservative assumptions and 4
provide for more margin also.
5 So it really is a combination of how they 6
model and what the results are, not necessarily just 7
a given number. That's one of the reasons, you know, 8
that kind of led to the discussion about modeling the 9
upper tail growth and making sure that that was well 10 represented in your model and also understanding maybe 11 in a more formal sensitivity type analysis of how some 12 of the parameters affected. When they develop the 13 model, they get a sense of that, you know, by changing 14 parameters and seeing the effects, but we were hoping 15 to see something a little more formal.
16 MR. BALLINGER: This is Ron Ballinger, 17 consultant, again. You may not know the answer to 18 this, but a lot of plants spend a lot of effort to get 19 rid of copper. Have these guys done that?
20 MR. KLEIN: They have. And I'm going to 21 address that in the next slide.
22 MEMBER ROBERTS: This is Tom Roberts.
23 Before you go on to the next slide, I want to follow 24 up on the previous question. The credit for chemical 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
55 cleaning, I certainly acknowledge that they didn't 1
take any credit for it, but your earlier slide showing 2
where the deposits were compared to where the cracks 3
were would seem to indicate that a chemical cleaning 4
wouldn't have as big an effect as you might think 5
because there were deposits where a lot of the cracks 6
were. Is that right?
7 MR. KLEIN: I wouldn't say there's no 8
deposits there. Keep in mind that that slice of 9
deposits was for 20 mils or greater based on eddy 10 current, and so I would think there are deposits at 11 all elevations, maybe just not to that same thickness.
12 MEMBER ROBERTS: Okay. Thank you.
13 MEMBER KAMMERER: This is Annie Kammerer.
14 I have a question. So what I'm understanding from 15 this presentation is that you do the inspections, you 16 see the condition, you do the modeling that we saw 17 first, and then you do a forward modeling to try to 18 assess what, I guess, the timeline or what you would 19 see in the future, and that that sort of is stepped 20 forward in time in a periodic basis. And so, first of 21 all, I hope that's the correct situation. Please 22 correct me if I'm wrong. And I guess my question is 23 has this cleaning happened in the past? And if so, is 24 there some data or information that could be used or 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
56 some sort of assessment done on how big of an impact 1
it made in the past to understand how much of a 2
difference it could make at this time? Thank you.
3 MR. KLEIN: It's a very good question. I 4
think your understanding is correct. So there's a 5
backward-looking process called condition monitoring 6
that makes sure you maintain tube integrity and then 7
the forward-looking process or operational assessment 8
that models the future. Palisades has not performed 9
a hard chemical cleaning prior to this past October.
10 It was recommended that the plant do that, I think, 11 back around 2008 - 2009 timeframe. Had it done it at 12 that point, we probably wouldn't be sitting here is 13 just my guess.
14 But there is data from other plants that 15 show the effects of chemical cleaning, and I think we 16 tried to hit on that on our third bullet on the 17 current slide, which is it does show a clear 18 improvement, but it's hard to quantify. Some plants 19 have seen greater benefit than other plants, but 20 there, no doubt, will be a benefit from removing all 21 that copper from the steam generators. The degree, I 22 think, will be determined moving forward by looking at 23 the rate of new indications and crack growth rates.
24 MEMBER KAMMERER: Thank you.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
57 MR. KLEIN: Some of the plants with data 1
also that complicates it. When you get to a large 2
number of cracks and you do chemical cleaning, some 3
plants also would drop the temperature in order to get 4
a benefit from that, as well. And so that makes just 5
assessing the chemical cleaning by itself a little 6
more difficult.
7 MEMBER BIER: I have a follow up question, 8
and this is a hypothetical. It's not something I'm 9
raising because I think it's happening but just, 10 conceivably, you could imagine that chemical cleaning 11 could actually increase the risk either by damaging 12 the tubes themselves or by removing debris that is 13 covering up a crack and, you know, preventing 14 problems. I assume that, you know, we have enough 15 experience from past cleaning at other plants that 16 this is not a concern; is that correct?
17 MR. KLEIN: That is correct. And the 18 chemical cleaning process has been around for quite 19 some time. They're very careful in the steps that 20 they take, and they include coupons in the generator 21 at the time of cleaning to make sure that, if they 22 were to have a transient or something they didn't 23 expect, that it would be represented in the coupons 24 that would then be removed and assessed. So we have 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
58 no expectation that chemical cleaning would be 1
detrimental.
2 MEMBER BIER: Okay. Thank you.
3 MR. JOHNSON: This is Andrew Johnson. I 4
would add the comment about removing deposits and 5
revealing more cracks, you know, I wouldn't call that 6
a risk increase, right, because that's greater 7
detectability. And I think that's actually likely as 8
a result of this cleaning.
9 MEMBER BIER: Okay. Thank you for the 10 explanation.
11 MR. KLEIN: So continuing on with the 12 chemical cleaning slide, it was performed in October.
13 Second bullet here shows the amount of material 14 removed, which was substantial, and one of the things 15 to point out is the amount of copper that was removed, 16 which is between 80 and 90 pounds per generator and 17 that's a lot of copper. So that will be beneficial 18 moving forward in terms of crack growth and crack 19 initiation. And I'd also note that these are 20 preliminary amounts, so they are subject to change.
21 MR. BALLINGER: This is Ron Ballinger 22 again, but you say they have removed -- when I say 23 removed copper, I mean feedwater heaters and stuff 24 like that to eliminate copper that's put in the steam 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
59 generator.
1 MR. KLEIN: I didn't understand that at 2
first. So the copper heat exchangers, I think, have 3
been long removed from the plant. They had, I think, 4
90/10 copper-nickel condensers, and they had some 5
brass feedwater heaters, and I think those type of 6
things have long been removed because people 7
understood a long time ago the detrimental effects of 8
9 MR. BALLINGER: So this copper had been in 10 there for a long time?
11 MR. KLEIN: Yes, that's correct. Next 12 slide. So in terms of concluding remarks, we did want 13 to note that the sleeve amendment we talked about the 14 last time we were here, the safety evaluation was 15 completed and the amendment was issued on October 16 30th. That session number is shown in that first 17 bullet.
18 In terms of safety issue related to 19 Palisades startup, we thought it was important to say 20 that we don't have a safety issue related to tube 21 integrity and startup. We do have some questions 22 about the operational assessment and modeling, but we 23 would characterize those more as a longer-term 24 question, not an immediate Palisades startup issue.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
60 They are performing additional OA analysis; and 1
Revision 3, we believe, is at the site and in the 2
process of being reviewed there and, at some point 3
after the site accepts it, we expect it will be sent 4
to the NRC for review.
5 That, I think, concludes our remarks.
6 VICE CHAIR HALNON: Thank you, Paul. This 7
is Greg. Any other discussion at this point? We'll 8
have another chance to deliberate after public 9
comments, but any other discussion or questions?
10 MEMBER PALMTAG: I just had a quick 11 question. Do you know when the startup is planned for 12 Cycle 29?
13 MR. KLEIN: I'm going to defer that to 14 April if she's listening.
15 MEMBER PALMTAG: And the question is 16 related to the timing on Rev 3.
17 MS. NGUYEN: Yes. So this is April Nguyen 18 with Region III. I'm the team lead for the restart 19 oversight activities. So, currently, the plant is 20 still undergoing activities to prepare for the startup 21 sequence. So there's still a good amount of 22 maintenance and testing activities that need to be 23 completed, including loading fuel into the core, which 24 has not been completed yet. So, you know, as the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
61 schedule moves along, we continue to follow the 1
schedule.
2 MEMBER PALMTAG: So there's plenty of time 3
for Rev 3 to get finished and evaluated before the 4
startup is the conclusion there?
5 MS. NGUYEN: That is correct. And as 6
mentioned previously, we do anticipate to see that in 7
short term.
8 MEMBER PALMTAG: Just to go back to 9
Craig's point, you made the point in there, but I 10 think it's important to stress, especially for the 11 public, that we did pass the OA, that there is a large 12 crack indication, but it is acceptable under the OA.
13 I just wanted to stress that. Thank you.
14 VICE CHAIR HALNON: Thank you, Scott.
15 Yes, I wanted to, you know -- they've been doing a lot 16 of conservatisms, taking the present condition of the 17 generators, the tubes.
Probabilistically, 18 statistically, deterministically, added a lot of 19 conservatisms and projected that there would be a 20 number of tube indications, and the next inspection 21 will tell us where we are based on that sliding scale 22 of how conservative that is.
23 And then, from a structural integrity 24 perspective, we passed what I would call the go-no-go, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
62 which is all the worst-case case tubes passed very 1
conservative structural integrity. So, given that, I 2
can see, you know, that can align with your concluding 3
remarks, of course, adequate justification.
4 When Rev 3 does come out, I will ask that 5
if there's anything that takes any concern, other 6
than, you know, outside of your concluding remarks, 7
that you would come back and inform us from that 8
perspective. We don't plan on reviewing Rev 3 unless 9
we get a flag from you all that says something unusual 10 or something different at this point. It may change 11 by the end of the meeting, given the comments from the 12 rest of the members.
13 Again, questions from any of the members 14 or consultant?
15 MEMBER HARRINGTON: This is Craig. Do 16 you, relating this to the rest of the plant, do you 17 see any connection between what's been observed here 18 and, in particular, the fact that the plant is 19 implementing weld overlays in PWSCC susceptible 20 locations elsewhere in the piping of the plant? Do 21 you see any connection there at all?
22 MR. KLEIN: We did speak to members of our 23 piping branch about the weld overlays. I think one of 24 the public comments that was received suggested that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
63 that was due to active crack at that location. And 1
based on the information that was provided on the 2
docket that those are preventative overlays that are 3
being performed and the licensee has not detected 4
cracks at those locations, so that's worth noting. I 5
think, as part of their analysis, they do postulate 6
cracks, though, as part of the case that allows the 7
structural weld overlay, so it's a postulated flaw.
8 In terms of how it relates to steam 9
generators, we do stay abreast of other locations, but 10 I think the steam generator kind of stands on its own 11 because the degradation assessment that's performed 12 prior to each outage, those look at all potential 13 degradation mechanisms in the steam generator, 14 including primary water, stress corrosion cracking, 15 and outside diameter stress corrosion cracking. So in 16 terms of that impacting the steam generator processes, 17 I think the degradation mechanisms are already 18 considered and it doesn't have a large impact.
19 MEMBER HARRINGTON: Okay. Thank you.
20 VICE CHAIR HALNON: So Dr. Ballinger, as 21 part of our consulting staff and expert in materials, 22 what's your impression of the OA and the work that was 23 done by the contractor to establish it?
24 MR. BALLINGER: Yes. I mean, I think that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
64 they did a very good job, very thorough, given the 1
uncertainties that were doing this. Anybody that 2
knows anything about the definition of mill and yield 3
will understand the difference between A and B.
4 That's not surprising at all.
5 It's also not surprising that, once you 6
start detecting these things, you start detecting a 7
lot because these tubes are what, 50 mils, round 8
numbers, thick? The cracks start responding to stress 9
at 50 microns. Personal experience. And so the 10 initiation process is multiple sources of initiation 11 that eventually link up. But if the initiation is at 12 50 microns, it starts responding to the stress, it's 13 not surprising that you see the kind of behavior, and 14 that's why they use a Weibull distribution in the 15 first place.
16 So it's well accounted for in the EPRI 17 guidelines, in the EPRI procedures, they know what's 18 going on. To my knowledge, I think there was a test 19 program, maybe funded by the NRC, a long time ago 20 where they looked at the effect of a tube rupture on 21 adjacent tubes and the possibility of additional 22 rupture, and they found an L-set. So there was no 23 issue related to if they had a rupture impacting 24 another tube.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
65 So, again, overall, I'm sure we'll see 1
what happens at the end of the next cycle.
2 VICE CHAIR HALNON: Thank you, Ron. Last 3
chance before I go open for public comments.
4 MR. KLEIN: I think one thing that I 5
probably didn't note but it's worth mentioning, during 6
the clarification call with the licensee and their 7
vendor, at that point, they expected the Rev 3 OA to 8
be more conservative.
9 VICE CHAIR HALNON: Certainly, as I 10 mentioned, if that's not the case, then at least 11 inform us so that we can make a decision if we need to 12 look at it.
13 Okay. I'm going to, at this time, open it 14 up for public comments, and we'll start with the 15 person who pre-applied, if you will. Mr. Gunderson, 16 your two minutes is starting. Are you there?
17 MR. GUNDERSON: Yes. Hi, this is 18 Gunderson here. Can you hear me?
19 VICE CHAIR HALNON: Yes, sir. Go ahead.
20 MR. GUNDERSON: Okay. I've witnessed the 21 Palisade's resurrection by Holtec for several years.
22 Despite steam generator tube hideout damage caused by 23 Holtec's neglected chemical treatment during wet 24 layup, it's unlike any other reactor in history. It's 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
66 clear to me that the NRC is going to do nothing to 1
stop the Palisades from restarting. I no longer 2
consider the NRC to be protecting Michigan's health 3
and safety. The Trump administration's executive 4
order requiring the NRC to speed up licensing has 5
turned the NRC into a sock puppet for Holtec.
6 I'm sure your committee is aware that I 7
was retained as an expert and have on numerous 8
occasions expressed well-documented concerns about the 9
deterioration of the steam generators. The evidence 10 suggests the steam generators will fail before the 11 next refueling outage, releasing radiation into the 12 state of Michigan. And what's been the response to my 13 concern? Holtec has said that my concern should be 14 discounted because, quote, Gunderson left the nuclear 15 industry decades ago, yet continues to hold himself 16 out as an expert in nuclear power plant operations, 17 closed quote. This from a company with no operating 18 experience. And the NRC has ignored my concerns, like 19 the Moai statues on Easter Island.
20 The question remains how much radiation 21 will spew from Palisades when a generator fails?
22 Today's slide 17 suggests that the NRC expects 1,000 23 new indications if Palisades runs for 18 additional 24 months. Rather than proving that the reactor coolant 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
67 pressure boundary is safe and strong, the NRC seems 1
content to place its faith in mitigating systems after 2
the leak.
3 Is it too late to change course on 4
Palisades? The reality is that the NRC's Palisade 5
train has not left the station. Events in the last 6
year make it clear that the NRC's train barreled right 7
through the station without ever intending to stop.
8 Thank you.
9 VICE CHAIR HALNON: Next on the list, Mr.
10 Kraig Schultz.
11 MR. SCHULTZ: Can you hear me?
12 VICE CHAIR HALNON: Yes, sir.
13 MR. SCHULTZ: Yes. Good morning, and 14 thank you for the opportunity to speak. My name is 15 Kraig Schultz with Michigan Safe Energy Future.
16 What is happening at Palisades is novel 17 and noteworthy. Restarting a plant from formal 18 decommissioning has never been done in this country.
19 This is completely new ground, and new ground calls 20 for higher standards, not lower. A new plant, a new 21 plant, would never be licensed with margins this 22 narrow or with this level of unresolved uncertainty.
23 Yet, somehow, the first restart from decommissioning 24 is being done with a steam generator that has passed 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
68 its age-degraded cliff, and we are expecting a 1
heightened crack indication after the next operating 2
cycle. That's your slide number 17. That mismatch is 3
the core issue.
4 The question before the ACRS is not 5
whether the calculated probabilities can be made to 6
pass. The real question is, what should the standard 7
be for the very first restart of a decommissioned 8
plant. Whatever is decided here becomes the precedent 9
for every future attempt. This decision will shape 10 the reputation of nuclear energy, the trust of the 11 public, and the credibility of the regulator.
12 The good news is that there's no financial 13 or political pressure to take unnecessary risk.
14 Federal funding is available. Replacing the steam 15 generators would bring more jobs and more economic 16 activity, and it would set a clean foundation for the 17 proposed SMR project next door. There is simply no 18 reason to accept a just barely adequate performance as 19 a -- I mean, we're starting with the steam generators 20 as a huge problem here. So the repair campaign was an 21 honest effort, but noble attempts do not change 22 physical limits. The results show that the repaired 23 steam generators do not meet the appropriate standard 24 for a first-of-its-kind restart.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
69 This is the moment for the ACRS to define 1
that standard and, if the committee takes the 2
conservative prudent path, you will not be remembered 3
as the group that slowed a restart. You will be 4
remembered as the group that protected the long-term 5
future of nuclear energy by ensuring the first restart 6
was one worth emulating.
7 Thank you for your time.
8 VICE CHAIR HALNON: Thank you, Mr.
9 Schultz. Mr. Kamps, Kevin Kamps.
10 MR. KAMPS: Hello. Can you hear me?
11 VICE CHAIR HALNON: Yes.
12 MR. KAMPS: This is Kevin Kamps with 13 Beyond Nuclear, and I'm speaking to you from 14 Kalamazoo, Michigan, which is 35 miles downwind of 15 Palisades. I've heard the term cliff edge and cliff 16 mentioned by the previous speaker but, also, I 17 believe, by an ACRS member. We live on that cliff 18 edge here in Kalamazoo.
19 Full disclosure: Beyond Nuclear is a 20 legal intervener against the relicensing, the restart, 21 unprecedented, of Palisades. We represent our members 22 and supporters, some of whom live less than a mile 23 from the atomic reactor. And I wonder where Entergy 24 is today because I've heard a lot of blame about the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
70 year 2009. How have these problems been unknown for 1
what is it? Sixteen years? There are some 2
disconnects. There are some elephants in the room 3
that are being ignored.
4 If a single tube bursts, that will result 5
in a release of hazardous ionizing radioactivity into 6
the local environment. Alan Blind had to correct an 7
NRC staff person last January on that.
8 Another important part of that January 9
2025 meeting was an NRC staff admission that Holtec 10 had neglected to implement proper wet layup on the 11 steam generators for two solid years, and I just hear 12 denial that this damage was a result of that. And I'd 13 like to point out that Consumers Energy, the original 14 owner and operator of Palisades, testified to the 15 State of Michigan in the spring of 2006 that the steam 16 generators needed replacement. NRC didn't require it, 17 so consumers didn't do it, Entergy didn't do it, 18 Holtec is not doing it.
19 There is a requirement in the Atomic 20 Energy Act for reasonable assurance of adequate 21 protection of public
- health, safety, and the 22 environment. I'm not feeling it. So if I'm right, 23 NRC is violating the law and our organization, our 24 environmental coalition, has taken NRC to court so far 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
71 on the exemption request, but we intervened on the 1
steam generators, and we were shut down quickly --
2 VICE CHAIR HALNON: Mr. Kamps, I need you 3
to wrap up your comments. You're well past two 4
minutes.
5 MR. KAMPS: Yes. Quit experimenting at a 6
grand scale with this nuclear monster on the 7
lakeshore. Us guinea pigs downwind do not appreciate 8
it. Thank you.
9 VICE CHAIR HALNON: Thank you for wrapping 10 your comments up. Any other public comments? I have 11 no one online. If you're on the phone --
12 MR. RABENHORST: Yes.
13 VICE CHAIR HALNON: State your name, 14 please.
15 MR.
RABENHORST:
My name is Karl 16 Rabenhorst.
17 VICE CHAIR HALNON: Go ahead, Karl.
18 Please keep your comments relevant to the meeting 19 today.
20 MR. RABENHORST: Yes. I served as a FEMA 21 Region 5 rep program site specialist for the Palisades 22 Nuclear Plant, and I provided written comments 23 documenting both the NRC and FEMA have failed to 24 enforce 44 CFR 350 off-site rep program regulatory 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
72 mandates for over a decade. In September 2014, I 1
reported Michigan emergency management failure to 2
produce required radiological emergency response staff 3
rosters and was overruled by Sean O'Leary, the Region 4
5 regional assistance committee chair. During an 5
annual letter of certification staff assistance 6
meeting in February of 2016, O'Leary verbally granted 7
blanket waiver to 44 CFR 350 requirements in favor of 8
the Michigan Emergency Management Act and routinely 9
waives 44 CFR 350 requirements.
10 At 3 p.m. on October 10, 2018, I provided 11 documentation of the above to NRC OIG Special Agent 12 Joseph Lee at the Alexandria, Virginia Public Library.
13 I was also FEMA Region 5's rep to the 44 CFR 351 Part 14 B mandated FRPCC --
15 VICE CHAIR HALNON: Excuse me, Mr.
16 Rabenhorst. Could you get more specific to the steam 17 generator operator's operational assessment, please?
18 MR. RABENHORST: Well, okay. The concern 19 becomes any discussion of restarting the plant needs 20 to consider the ability to implement an off-site 21 response. If any response is similar to the chemical 22 spill at East Palestine, Ohio, which is a Beaver 23 Valley-supporting jurisdiction, or a Hurricane Helene 24 response, it would undermine public confidence in the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
73 nuclear power enterprise. Collapse of public 1
confidence after Fukushima resulted in the German 2
government prematurely decommissioning the entire 3
German fleet.
4 Any public support of the nuclear power 5
industry is an essential national security concern, 6
and a decision to restart the plant needs to consider 7
off-site radiological emergency preparedness, 8
regulatory compliance, and the ability to implement an 9
off-site response. Thanks.
10 VICE CHAIR HALNON: Thank you. Next on 11 the list is Kathy Barnes. Kathy Barnes, unmute 12 yourself. Looks like you're having a little trouble.
13 We'll give you a second. She's got her hand up.
14 Kathy, I'm going to give you a couple more seconds to 15 get your mic taken care of. If not, then you can 16 supply written comments to Quynh Nguyen of the ACRS 17 staff, and he will enter those into the record if you 18 can't make them verbally.
19 I see you now. Yes, finally got you. Go 20 ahead, Kathy.
21 MS. BARNES: Yes, I'm sorry. I had to go 22 into my settings and turn my microphone on. For some 23 reason, the meeting app turned it off.
24 Anyway, I wanted to say thank you for all 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
74 your research, but it is based on risk and 1
probability. And probability statistics and all that, 2
that's a lot of mathematics, but you're not taking 3
into account that you just don't know everything.
4 And if some of these tubes get clogged, et 5
- cetera, it's going to increase pressure and 6
temperature on the other ones. You just can't stop 7
these tubes from functioning because that would be a 8
meltdown, correct? If all the tubes, if they did not 9
function right, if they burst, if they were clogged, 10 it would be a meltdown because you couldn't pump the 11 water in, and I don't think you should take that risk.
12 And even if you totally replaced all of 13 those tubes and put in new tubes, you still have an 14 aged, embrittled nuclear reactor there that you've run 15 out of -- you know, you talked about core samples.
16 You ran out of those. You don't have those. That's 17 also based on risk and probability.
18 And talk about downwinders, that goes a 19 long range. At one former meeting with the NRC, I 20 asked about evacuation zones and about what would 21 happen. Basically, if there was a meltdown at 22 Palisades, it would affect agriculture. It would 23 affect people's residences. It would be probable 24 deaths. Cancer rates would go up. It would be 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
75 horrendous. Is that something that you're willing to 1
be responsible for? Is that something that you're 2
willing to take a risk that it might happen?
3 Another thing is that that plant needs to 4
be cleaned up. It's got tritium underneath the plant 5
that needs to be removed, and that hasn't been done.
6 And, eventually, it's going to contaminate the water 7
of Lake Michigan as it tritiates the water and the 8
groundwater. I believe it's already contaminated the 9
groundwater, and eventually it will go out in the 10 lake.
11 VICE CHAIR HALNON: Kathy, you're past two 12 minutes. Can you wrap up your comments, please?
13 MS. BARNES: Pardon?
14 VICE CHAIR HALNON: You're past your two 15 minutes.
16 MS. BARNES: Oh, keep it to the tubes?
17 Oh, okay.
18 VICE CHAIR HALNON: I need you to wrap 19 them up, so I have other people waiting to make 20 comments.
21 MS. BARNES: Okay. Okay. One more thing.
22 Fukushima is still melting down. That's still 23 contaminating the ocean. They're pumping that 24 contaminated water to cool it down into the ocean.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
76 They haven't been able to stop that meltdown. Is that 1
something that we want for Lake Michigan?
2 So those are my comments. Please, please, 3
take them seriously. I do not think it should be 4
restarted. I think it should be shut down and cleaned 5
up. Thank you.
6 VICE CHAIR HALNON: Michael Keegan, you're 7
next up. Two minutes, please.
8 MR. KEEGAN: Hello. Thank you. So 9
Michael Keegan with Don't Waste Michigan. We're 10 interveners to prevent the restart of Palisades, and 11 we've had multiple contentions and we're on appeal 12 with the NRC Commission themselves. And we're also in 13 federal court.
14 I'm a sociologist by training, and a 15 colleague of mine, Charles Perrow, has written a book 16 called Normal Accidents, and what he does is he goes 17 through complex systems and shows where there are 18 failures that are baked into the cake. Now, the steam 19 generators just happen to end up in your lap, and 20 that's in the limelight. But there are a multitude of 21 problems at Palisades.
22 To previous ACRS commissions, the utility 23 had promised to change out the strainers on the 24 reactor, on the steam generators, on the whole system.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
77 There are five metric tons of calcium silicate as an 1
insulation, and it's a GSI-191. It's a generic issue.
2 It's been known about since 2000. There have been 3
promises to remedy it. It's not been remedied.
4 There's just layer upon layer. The 5
control rod drive mechanisms, in August of '24, the 6
public was promised that those systems would be 7
changed out. Well, then they're in a hardship, and 8
now they have to do overlay welds. You've got 9
band-aid upon band-aid. Every system you look at is 10 compromised.
11 What's special about Palisades? There's 12 no regulator ever been regulating at Palisades.
13 Everything goes.
So you're way out there 14 probabilistically, risk assessment, all these systems 15 are compromised. Normal access. What's the 16 probability something's going to happen? Catastrophic 17 failure. One over one. It is going to happen. You 18 baked it into the cake.
19 So do not allow the restart. Step 20 forward. Previously, you were promised that that 21 strainer issue would be resolved. It is not resolved.
22 Thank you.
23 VICE CHAIR HALNON: Okay. I still have a 24 few minutes. Is there any other comments?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
78 MS. DRECHSLER: My name is Jackie 1
Drechsler, and I would like to make a comment.
2 VICE CHAIR HALNON: Go ahead, Jackie.
3 MS. DRECHSLER: Thank you so much. I 4
appreciate this meeting. I've been to just about 5
every single meeting regarding the restart of the 6
Palisades Michigan plant. I've been to the big 7
meetings, the small meetings.
8 This is what I would like to say: There 9
have been so many years of lack of oversight, and 10 there is documentation that shows the real 11 possibilities of the steam generator failure due to 12 cracking of the tubes. All I have heard today is 13 supposition, probabilistic, and risk modeling. It's 14 all speculation.
15 The NRC fellow said earlier, we need to 16 make it to the next inspection without a fail. And 17 then another person said -- or it might have been the 18 same because I can't necessarily know who's speaking.
19 The other person said large crack indications, but 20 they're acceptable. Well, this plant, first of all, 21 needs to be cleaned up. And the lady who talked about 22 the tritium in the waters, that, you know, that 23 changes DNA. There is the possibility of a real 24 meltdown, should there be, you know, a failure.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
79 This is high risk, pollution of waters, 1
land, and health damages to people and wildlife. Why 2
would the NRC approve a restart based on speculation?
3 Thank you very much.
4 VICE CHAIR HALNON: Thank you. It looks 5
like we have Alan Blind for the next question.
6 MR. BLIND: Yes. Hi. Thank you. Do you 7
hear me okay?
8 VICE CHAIR HALNON: Yes, Alan. You have 9
two minutes. Go ahead.
10 MR. BLIND: My comment, actually, comes 11 from the Nuclear Regulatory Commission in their 12 October 23, 2000 report, 25 years ago, on their 13 summary of the Indian Point steam generator tube 14 failure. Let me read it: The task group concluded 15 that communicating the safety significance of the 16 Indian Point experience is difficult. During the NRC 17 significance determination process, the staff found 18 that the tube condition during the last cycle was risk 19 significant due to the degradation of safety margin.
20 And I want to emphasize that, due to the degradation 21 of safety margin, and there's been so much discussion 22 today on margin.
23 Back to the quote: Notwithstanding the 24 loss of safety margin, Indian Point 2 was designed to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
80 mitigate the effects of a steam generator tube 1
rupture. I digress. Palisades is not designed, as 2
Kevin Kamps quoted me earlier as correcting the NRC.
3 And I think the committee needs to consider that's the 4
consequence that stands Palisades apart.
5 Let me come back now to the NRC quotation:
6 This distinction may not be understood by all 7
stakeholders. Now, this is the most important point, 8
so let me finish. NRC will probably face this 9
communications challenge again because steam generator 10 tube failures and ruptures -- and notice they say 11 failures and ruptures -- have occurred before and will 12 occur again. Let me repeat: will occur again.
13 Therefore, the task group recommends that the NRC 14 should incorporate experience gained from Indian Point 15 in the significance determination process into planned 16 initiatives on risk communication and outreach to the 17 public.
18 VICE CHAIR HALNON: Alan, you're over your 19 two minutes. Can you wrap up and conclude?
20 MR. BLIND: Yes. I'll just conclude that 21 we need to put these margins back into the allowed 22 operating period of time. Why allow Holtec to operate 23 all the way up to 1.5 effective full power years with 24 little margin, rather than a reduced period of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
81 operation and regain operating margin? And then let's 1
take a look at what the inspection results tell us.
2 Thank you.
3 VICE CHAIR HALNON: Okay. I don't see any 4
other people in the queue. No public in the room here 5
that I can see. Oh, okay. So Jesse Deer. Oh, is 6
that the second comment? If you have another comment, 7
please put it in writing and send it to Quynh Nguyen 8
of the ACRS staff. His email is on the agenda.
9 So I think this is it. Jesse Deer, have 10 you made your comment already or is this your first 11 time? You keep on muting and unmuting. Try again.
12 Yes. Go ahead, Jesse. Go ahead.
13 MR. DEER: Good day, you all. Pardon me, 14 I'm a little under the weather. Got a little flu, but 15 I still am tuning in. My name is Jesse Deer In Water.
16 I live in Redford, Michigan, basically on the other 17 side of the state, but work on different nuclear 18 issues. You know, I'm no technical expert like some 19 of you all are, but I just wanted to come here and, 20 basically, after the presentation here and kind of 21 following what's going on, I just wanted to reaffirm 22 some of the things I heard, you know, that 1,200 tubes 23 are at risk at Palisades. There's tons of tracking 24 and degradation at critical support components, and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
82 the operational assessment, the OA, the degradation is 1
projected using different determining factors.
2 And I just wanted to, you know, just 3
reaffirm what everyone's kind of said here. You know, 4
I know some of it was, you know, this and that. But 5
I've known about Arnie Gunderson since I was a child 6
and the different analysis and the work that he's done 7
and followed. And so I would just like to just uplift 8
Arnie's messages, his comments and everything that he 9
had said. I mean, this is a big issue. You know, I 10 mean, as an indigenous person with relatives who are 11 Potawatomi and live over there in the area, I can say 12 that there is tons of concern for safety and the 13 future and that every possible thing needs to be 14 looked at, every possible thing that needs to be fixed 15 needs to be fixed. I mean, and if these steam 16 generators are compromised, they should just be 17 replaced and fixed.
18 I mean, I'm against the reactor reopening 19 in general because, you know, I come from a community 20 impacted by nuclear issues in Oklahoma. So I have, 21 you know, this lived experience with it. And so, 22 basically, I just want to uplift Arnie's comments and 23 work and then also, at the same time, just give my 24 honest thing that a nuclear meltdown could impact the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
83 lakes terribly, you know. I mean, bad. And the land 1
I love over there in the western part of the state 2
that my relatives fish, forage wild rice, manoomin, 3
all that's at stake.
4 VICE CHAIR HALNON: Jesse, you're past 5
your two minutes, so go ahead and please wrap up.
6 MR. DEER: All right. Sorry. I'll wrap 7
up. But the United States owes a moral debt to the 8
indigenous people here on Turtle Island and in 9
Michigan to do the right thing. And you all are part 10 of that, you know what I mean? You have this 11 opportunity to do the right thing. If there's the 12 money there, if there's all this other stuff to fix 13 it, man, freaking fix it. I mean, you know, make them 14 fix it. Don't run tests that, you know, they could --
15 yes, yes, just fix it. Just fix it. That's all.
16 Just fix it.
17 VICE CHAIR HALNON: It's 10:30. I'm going 18 to close public comments at this time, and we will --
19 okay. I'm going to allow two minutes just because you 20 got in just before I said I was done. So go ahead.
21 MS. TOOHILL: Hi. Can you all hear me?
22 VICE CHAIR HALNON: Yes.
23 MS. TOOHILL: Okay. Great. Sorry. My 24 name is Spencer Toohill, and I'm with the Breakthrough 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
84 Institute. I just wanted to give a very brief 1
comment.
2 I am in support of the Palisades restart.
3 And we are as the Breakthrough Institute. I just 4
wanted to point out we are very big believers in 5
public engagement and public comments. And, you know, 6
these public meetings are very important forms of 7
public engagement with the NRC and with the ACRS. So 8
I think that this meeting is scheduled until noon, and 9
I just wanted to encourage to allow this important 10 forum of public engagement. And if people had more 11 comments, I would just encourage to allow people to 12 provide their public comments, their verbal public 13 comments, in this forum for as long as they can.
14 So that's all I had. Thanks.
15 VICE CHAIR HALNON: Okay. No other hands 16 are raised. I'm going to declare an end to public 17 comments. It seems like we've received quite a few.
18 I appreciate it.
19 If anyone beyond this has a comment, 20 please enter it into the record, email Quynh Nguyen 21 the ACRS staff, and we will try to get that into the 22 public record. So I'm going to ask the members is 23 there any additional thoughts or questions before we 24 recess out of this period? Don't see any.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
85 So with that, Chairman Kirchner, I will 1
pass the meeting back to you. I want to thank Paul 2
Klein and Andrew Johnson for your time and efforts.
3 And, again, I want to reiterate that we will probably 4
include a summary and probably, in that summary, I'll 5
just ask that you, as part of the meeting minutes, 6
provide us a notification if there's anything 7
non-conservative or of concern in Rev 3 of the 8
operational assessment that comes out.
9 So with that, Chairman Kirchner.
10 CHAIR KIRCHNER: Thank you, Greg. So we 11 will capture our activities here in our meeting 12 summary. Are you looking for further input from 13 members for that?
14 VICE CHAIR HALNON: You know, I just need 15 to take some time to draft some thoughts, and then I 16 can probably, if you want, after taking a break -- if 17 we take a break now, I can draft some thoughts up, and 18 then we can iterate on that.
19 CHAIR KIRCHNER: Reconvene and iterate on 20 that. So with that, we'll take a break until 10 21 minutes of 11. That's 10:50 Eastern time.
22 VICE CHAIR HALNON: I wasn't watching 23 carefully, and it appears that we've got one more 24 person that would like to make a comment, given the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
86 fact that we do have a few minutes prior. We need to 1
take a break, granted.
2 CHAIR KIRCHNER: Let's take the comment.
3 VICE CHAIR HALNON: Yes, let's take the 4
comment. Ann Scott, I'm going to give you two 5
minutes. I apologize for missing your hand before I 6
stopped. I assumed you had it up before I stopped; 7
otherwise, I probably would not have had it up. But 8
I want to give you two minutes. Go ahead, Ann. Ann, 9
can you unmute yourself and take your two-minute 10 comment? Ann Scott, we still don't see you unmuted.
11 MS. SCOTT: Hello. My name is Ann. Can 12 you hear me?
13 VICE CHAIR HALNON: Yes, we can. Go 14 ahead.
15 MS. SCOTT: My name is Ann Scott. I live 16 in Palisades Park. I consider my home Palisades Park.
17 I consider the plant Palisades Park. I live walking 18 distance from the plant. I've grown up there. I 19 watch it built. I watched it with all of its failures 20 and the fines against it and the attempts to protect 21 all of us who live near the plant, who live in 22 Michigan, who live on Lake Michigan, who live in the 23 United States, who live in the world. I've watched 24 the NRC panel. I know their jackets. I know their 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
87 faces. I've heard them tell me over and over again 1
that they're here to protect us, they're here to 2
protect our country, our state. I don't believe that.
3 And I just want to acknowledge that you're not being 4
honest with us and you're not taking the proper 5
precautions. And I feel so badly for you and for all 6
of us.
7 And the idea that you're giving two 8
minutes to these important people. I'm not. I'm just 9
an emotional human being. But there have been 10 important people trying to get a message across, and 11 the suggestion that they're allowed two minutes.
12 That's shameful.
13 That's all. I'm done. My husband will 14 have to turn off the microphone but really let down, 15 really let down.
16 VICE CHAIR HALNON: Okay. Back to you, 17 Karen.
18 CHAIR KIRCHNER: Okay. Thank you. So 19 let's take a break now to the top of the hour, 11 20 o'clock, and we'll come back and discuss next steps 21 forward at that period. So we're recessed until 11 22 a.m.
23 (Whereupon, the above-referred to matter 24 went off the record at 10:38 a.m. and went back on the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
88 record at 11:01 a.m.)
1 CHAIR KIRCHNER: Okay. It's 11 o'clock, 2
and we are back in session. And I'll turn it back to 3
my colleague, Greg Halnon.
4 VICE CHAIR HALNON: Thank you, Chair 5
Kirchner. So I wanted to thank everybody for the 6
process that we went through today, and I wanted to 7
also thank the public for the comments that they made.
8 I know it's been a long process, and we clearly see 9
some frustration in the results of how the process is 10 coming out, and we certainly understand many of us 11 have either worked or lived near nuclear plants 12 ourselves and certainly understand that frustration.
13 We, as the
- ACRS, are an advisory 14 committee;
- and, accordingly, we provided the 15 commission our thoughts on the restart process in the 16 September 22nd, 2025 letter, which is available on the 17 public website of the ACRS. And we continue to stand 18 by our conclusions that the steam generators' 19 integrity is of concern in that the condition 20 monitoring and operational assessment reports are 21 essential to ensure integrity and that the NRC staff 22 should continue to scrutinize the performance of the 23 steam generators, both during the operational period 24 and during the next inspection period, very closely.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
89 And we did ask the staff to come back and tell us 1
about Rev 3 if there is a non-conservative or some 2
other concern that approaches that.
3 So with that, that's what our summary will 4
reflect. And we're not going to continue to -- we're 5
not going to let loose of this, but we're certainly at 6
a point where, unless the staff comes back and tell us 7
otherwise, we'll stand by our September 22nd one.
8 So, with that, Walter, I will send it back 9
to you.
10 CHAIR KIRCHNER: Any comments from the 11 members? Obviously, we'll monitor closely the next 12 phase of the startup, which will include hydro testing 13 once the system is buttoned up. That will be another 14 check on the integrity of the steam generators and the 15 tubes.
16 Anyone else at this point? Any new 17 business for this meeting? Hearing none. Then we can 18 let the court reporter go, and we will adjourn the 19 731st meeting of the ACRS.
20 MEMBER BIER: Were we going to finalize 21 written statements, or that's going to be done 22 offline? Written summaries for today and yesterday.
23 VICE CHAIR HALNON: We're going to 24 discuss, Bob, the write up --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
90 CHAIR KIRCHNER: No, we already finalized 1
that yesterday. Thank you for the reminder, Vicki.
2 For anyone listening in, that was on the full-spectrum 3
LOCA topical report from Westinghouse, so we finalized 4
our summary of that yesterday after yesterday's 5
discussions. No new business.
6 Okay. Then we are adjourned.
7 (Whereupon, the above-referred to matter 8
was concluded at 11:05 a.m.)
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
Advisory Committee on Reactor Safeguards Full Committee Meeting December 4, 2025 Palisades Steam Generator Update - Operational Assessment Paul Klein and Andrew Johnson NRR/DNRL/NCSG
Acronyms 1RXX - refueling outage number AILPC - accident induced leakage performance criterion BOC - beginning of cycle CIRC - circumferential CM - condition monitoring DBH - diagonal bar hot Eggcrate - horizontal lattice tube support EOC - end of cycle EPRI - Electric Power Research Institute FSAR - final safety analysis report ID - inside diameter NDE - nondestructive evaluation NOPD - normal operating pressure differential OA - operational assessment OD - outside diameter PW - primary water SCC - stress corrosion cracking SG - steam generator SGOG - Steam Generator Owners Group SIPC - structural integrity performance criterion TEH - tube end hot TSH - top of tubesheet hot TTS - top of the tubesheet
Agenda
- Summary Up Front
- Palisades SG Design
- SG Inspection (1D28) Results
- CM OA Process
- 1D28 CM Summary
- Cycle 29 OA Review
- SG Chemical Cleaning
- Concluding Remarks
Summary - Palisades SG Tube Degradation 1D28 SG inspection: Over 1200 axial ODSCC indications at eggcrate supports
- Plugging would exceed 15% SG A design limit; licensee submitted LAR to install sleeves A total of 2971 SG sleeves will be placed in service to preserve plugging margin
- 906 corrective sleeves (SCC present), 2065 preventative sleeves (no SCC present)
Condition Monitoring (CM) showed tube integrity was maintained, 23 tubes passed insitu pressure tested as part of CM (no burst or leakage)
Cycle 29 Operational Assessment, Rev. 2 - 1.5 Effective Full Power Years
- Analysis results meet all performance criteria, little margin for eggcrate SIPC
- key conservative assumptions: (1) crack growth rates for ODSCC at eggcrates based on assuming all cracking at 1D28 was just below detection threshold at 1R27 (2) assuming no improvements from EPRI SGOG chemical cleaning of both SGs
- Based on NRC-licensee clarification call, OA revision is in progress
Palisades Combustion Engineering Model 2530 SGs Installed in 1990, each SG has 8,219 tubes Alloy 600 mill annealed tubing Nominal 0.75 inch diameter, 0.042 inch wall thickness Rows 1-18 U-bend, Rows19-138 square bend design Horizontal stainless steel lattice type eggcrate support plates Sleeves installed 01H to 05H
Vertical and Horizontal (Eggcrate) Supports EGGCRATE
1D28 SG Inspection - Final Results SG Location Type Indications Tubes A
TSH Axial ODSCC 17 16 A
TSH Circ PWSCC 29 10 A
TSH Circ ODSCC 73 60 A
Tube Supports Axial ODSCC 900 608 A
TEH to TEC Wear > 40%
5 5
A EFFECTIVE PLUGGING PERCENT 11.1%
B TSH Axial ODSCC 10 9
B TSH Circ PWSCC 10 10 B
TSH Circ ODSCC 1
1 B
Tube Supports Axial ODSCC 313 220 B
TEH to TEC Wear > 40%
3 3
B EFFECTIVE PLUGGING PERCENT 6.4%
1D28 Condition Monitoring
- CM uses inspection results to assess whether tube integrity (SIPC, AILPC) was maintained until the inspection
- CM limits pre-determined for each degradation mechanism/location for rapid evaluation:
- Flaw clearly meets CM limit with initial NDE sizing
- Flaw needs enhanced eddy current flaw sizing (profiling) to determine if CM is met
- Insitu pressure testing (ISPT), if NDE alone cannot confirm CM is met
- All tubes met CM during 1D28, ISPT needed for 23 indications
- 8 eggcrates, 2 axial TTS, 11 circumferential TTS, 1 DBH, 1 obstructed tube
- No leakage or tube burst during ISPT
CM - Axial ODSCC at Eggcrates SG B
Eggcrate Support Cracking SG A Support Number Eggcrate Cracks 05H 7%
04H 10%
03H 8%
02H 24%
01H 37%
ODSCC Distribution at Eggcrates is Consistent With a Temperature Driven Mechanism
~ 70%
Operational Assessment (OA)
- Forward-looking projection of SG tube degradation, OA provides the technical basis for meeting tube integrity until the next SG tube inspection
- Addresses all tube degradation mechanisms such as wear at support structures, SCC at the lattice supports, SCC at the top of tubesheet
- Degradation can be projected using deterministic (worst case tube),
mixed, or fully probabilistic methods
OA Options
- Two general approaches Full bundle probabilistic
- Monte Carlo simulation methods are used to predict the future distribution of the structural integrity parameters at the end of the next inspection cycle (EOC), so that the lower 95th percentile of the worst case burst pressure values may be obtained. Similarly, the upper 95/50 total bundle leakage is also calculated.
-Simplified techniques, including:
- Arithmetic
- Simplified Statistical
- Monte Carlo
- Mixed techniques - a combination of the above.
OA - Full Bundle Probabilistic Process
95/50 Criterion vs 95/95
- 95/50 Probability/Confidence is the industry consensus for OA:
- EPRI Technical Basis for SG Tube Integrity Performance Acceptance Standards, 2006
- EPRI SG Program Guidelines - Integrity Assessment Guidelines
- Some uses of 95/95 acceptance criteria, typically related to leakage and dose, not the structural integrity performance criterion
- H* amendments for tube cracking deep within tubesheet, Alloy 600 TT fleet
- Individual unit examples before consensus approach developed
- Note the 95/50 criterion is used in conjunction with deterministic safety factors, (3 x NOPD burst normal operation, 1.4 design basis accidents) providing multiple layers of protection
Palisades OA for Operating Cycle 29
- Full tube bundle probabilistic OA model used for:
- Axial Outside Diameter Stress Corrosion Cracking (ODSCC) at Eggcrates, Vertical Straps, and in the Freespan
- Axial ODSCC and PWSCC at Top of Tubesheet (TTS)
- Freespan Axial Intergranular Attack (IGA)
- Circumferential (Circ) ODSCC at TTS
- Tube Support Wear
- Mixed Arithmetic/Monte Carlo method used for:
- Circ PWSCC at Expansion Transitions and Expanded Tubesheet Locations
- Axial ODSCC at Dents/Dings
- Axial PWSCC at Dents/Dings and Small Radius U-bends
- Tube-to-Tube Wear
Palisades Cycle 29 OA, Rev. 2 (1.5 EFPY)
- Probabilistic full tube bundle method used for most degradation mechanisms, including ODSCC at eggcrate supports
- A lower 95th worst case tube calculated burst pressure that is greater than 3 times NOPD meets structural tube integrity Mechanism Lower 95th Burst (psi) 3 times NOPD (psi)
Comments ODSCC at Eggcrates 3978 3960 Simulated outages 1R20 to 1R29 50,000 times Axial ODSCC at TTS 4340 3960 50,000 trials Circ ODSCC at TTS 7499 3960 50, 000 trials Axial PWSCC U-bends BOC structural depth from 95th POD, upper 95th growth EOC Lower 95th Flaw Burst pressure is 4180 psi (3960 psi 3DP)
Mixed Arithmetic/Monte Carlo approach
OA Indication Prediction Comparison Damage Mechanism Projections for 1D28 Projections for 1R29 Axial ODSCC at Supports (Evaluated Probabilistically) 22 indications/SG 1292 indications (SG A bounding; used for SG B)
Axial ODSCC at Top of Tubesheet (Evaluated Probabilistically) 16 indications/SG 21 indications/SG Axial PWSCC within Tubesheet (Evaluated Probabilistically)
NOT PERFORMED 68 indications/SG Circumferential ODSCC at Top of Tubesheet (Evaluated Probabilistically) 23 indications/SG 35 indications/SG Wear at Supports (Evaluated Probabilistically) 3372 indications (both SGs) 3408 indications (both SGs)
Modeling of ODSCC at Eggcrate Supports
- Crack initiation modeled with a Weibull function starting with 1R27, model parameters adjusted until 1D28 results were benchmarked
- To fit the large increase in detected indications in 1D28, outages prior to 1R27 were treated as having perfect ODSCC detection
- OA input for SCC growth rates assumed all flaws detected during 1D28 were at the threshold of detection at 1R27
- Results in average growth 9% TW/EFPY
- Conservative to both the EPRI Guidelines default average growth adjusted to Palisades operating temperature (2.15% TW/EFPY), and the EPRI average growth rate meant to conservatively bound any Alloy 600 unit/temperature
- Cycle 29 upper limit growth of 19.23% TW/EFPY
OA Model Results
- Wiebull distribution with parameters adjusted to benchmark 1D28 (~26 EFPY)
- Assumes perfect detection prior to 1R27 (~24 EFPY)
NRC Staff Comments, Cycle 29 OA Cycle 29 OA meets all acceptance criteria Small margin for meeting SIPC - ODSCC at eggcrate supports OA conservatively calculated average SCC growth rates by assuming all cracks were present at detection threshold and only grew during the last operating cycle (assuming cracking also occurred during the extended shutdown period decreases crack growth rate)
OA assumes no benefits from chemical cleaning Clarification call with licensee on October 14, 2025
- Obtain clarification on portions of Cycle 29 OA modeling
- Upper tail growth and sensitivity analysis discussion Licensees vendor is performing additional Cycle 29 OA sensitivity analyses; a new OA (Revision 3) will be issued
Palisades SGs Chemical Cleaning
- EPRI SGOG (Hard) chemical cleaning performed in October 2025
- Separate cleaning steps for copper and magnetite deposits
- Preliminary amount of material removed:
- SG A: approximately 3000 pounds total deposits
- SG B: approximately 3100 pounds total deposits
- Alloy 600 MA tubing fleet operating experience shows a variable but clear improvement in tube cracking from hard chemical cleaning
Concluding Remarks - Path Forward
- NRC SG Sleeve Amendment issued October 30, 2025 (ML25303A280)
- NRC staff have no safety issues related to Palisades start-up and initial SG tube integrity, however, staff has questions related to OA modeling of cracking at the eggcrate supports
- Palisades is performing additional OA analysis based on NRC staff questions about the OA
- NRC staff evaluation of Palisades Cycle 29 OA continues, OA revision (Rev. 3) will be submitted to NRC after licensee review
Back-up slides
Bobbin POD Curve: Eggcrate, Vertical Strap, and Freespan Axial ODSCC
Eggcrate Support Bobbin Noise Comparison
Burst pressure Evaluation - Eggcrate
Palisades SG Tube Sleeving Amendment
- Issued October 30, 2025, ML25303A280
- Sleeve life is limited to 10 years
- Sleeve installation limited to hot leg only
- LC establishes a minimum sleeve inspection scope, the actual scope will be determined by the degradation assessment and OA
- 100 percent of in-service sleeve/tube assemblies by the end of 1R29
- 50 percent sleeve/tube assembly inspection by the end of each RFO thereafter, with scope expansion per EPRI Integrity Guidelines if flaw detected in pressure boundary portion of sleeve or parent tube
- LC intended to maintain high probability of detecting structurally significant parent tube flaws should they initiate
Gundersen ACRS Comments 12/4/25 I have witnessed Palisades resurrection by Holtec, an inexperienced decommissioning contractor. Despite Palisades steam generator damage caused by Holtecs neglect and long history of exemptions to other licensing issues, it is clear to me that the NRC will do nothing to stop Palisades from restarting.
I no longer consider the NRC to be protecting the publics health and safety. The Trump Administration Executive Order requiring the NRC to speed up licensing has turned the NRC into a sock puppet for the nuclear industry I am sure that your committee is aware that I was retained as an expert and have, on numerous occasions, expressed well documented concerns about the deterioration of the Steam Generators. The evidence suggests that the Steam Generators will fail before the next refueling outage, releasing radiation on the State of Michigan.
And what is the response to my concerns? Holtec said that my concerns should be discounted because Gundersen left the nuclear industry decades ago yet continues to hold himself out as expert in nuclear power plant operations1. This by a company with no operating experience! And the NRC has ignored my concerns, acting like the Moai statues on Easter Island.
The question remains How much radiation will spew from Palisades when the steam generators fail? Rather than proving that the reactor 1 ASLBP No. 25-988-01-LA-BD01, APPLICANTS ANSWER OPPOSING BEYOND NUCLEAR ET AL.S PETITION TO INTERVENE AND REQUEST FOR HEARING, July 11, 2025. Page 19
coolant pressure boundary is safe and strong, the NRC seems content to place its faith in mitigating systems after the leak.
Is it too late to change course at Palisades? I have given up hope that the NRC will even read, let alone acknowledge, my steam generator safety concerns concerns. The reality is that the NRCs Palisades train has not left the station. Events of the last year make it clear that the train barreled right through without ever intending to stop!
ACRS October 10 Meeting Palisades Nuclear Power Plant 2 Minute Verbal Presentation by Arnie Gundersen You are meeting today behind closed doors to review the supposedly non-public proprietary operational assessment of the safety of Palisades. I am not confident that the publics health and safety are well served by your top secret process.
My technical position on the integrity of the Steam Generators is already on the record. After I presented my concerns to the ACRS, new information became available that is deeply concerning. Holtec asked the NRC for eight relief requests because it has now detected stress corrosion cracks in eight dissimilar metal welds in the reactor coolant system! Two in the hot legs, four in the cold legs and two in the pressurizer.
EIGHT reactor coolant welds have experienced SCC in addition to the 3000 flaws that were sleeved in the steam generators. The entire reactor coolant system is degraded because Holtec chose not to maintain EPRI reactor coolant standards for two years.
Those of us who have appeared before the ACRS to criticize NRC Staff decisions have a joke about the ACRS: The ACRS asks the Staff great questions. The problem is that the ACRS accepts crappy answers! Since I only have two minutes, here is just one good question the ACRS might want to ask the Staff:
Good Question: What is the probability that the Steam Generators will spring a leak before the next refueling?
Poor Answer: We dont know but there are systems in place to mitigate the leak when it occurs.
That answer is an admission that the Staff is relying on mitigation when a Reactor Coolant breach is almost inevitable.
Will you accept that crappy answer? After this meeting is finished, will you be able to say that you put safety before expediency on the Palisades resurrection??
1 Additional Written Information to The Advisory Committee on Reactor Safeguards Concerning the Safety of the Palisades Nuclear Plant October 10, 2025 Arnold Gundersen In January 1986, two NASA contract engineers identified that the Challenger Space Shuttle was endangered if it were to be launched in cold weather. Those engineers used all the professional channels available to prevent the launch. But the bureaucratic inertia within NASA to maintain the launch schedule caused those NASA engineers to be overruled. We all know the outcome of that safety lapse. I write to you today in the spirit of those two NASA engineers as I continue to express my safety concerns to the members of the ACRS. You provide the last possible public safety oversight before resurrecting the Palisades nuclear plant.
First, I wanted to thank you for allowing me to share my concerns about the condition of the diminished integrity of the Reactor Coolant System at Palisades for five minutes during the Palisades subcommittee hearing on August 21, 2025. And I also want to thank you for your thoughtful Steam Generator questions to the NRC staff during the full committee meeting of September 3, 2025. I appreciate that the ACRS appears to be taking its oversight of the Palisades resurrection precedent seriously.
That said, new information just placed on the Palisades docket has amplified my previously expressed concerns. I know the NRC staff has not been forthcoming with information for me to analyze as an expert. I fear that the NRC staff has not been forthcoming to the ACRS either.
Never in my 54 year professional career have I been more concerned about the integrity of the reactor coolant pressure boundary than I am about the condition of Palisades. Please let me explain.
All operating nuclear reactors are required to provide detailed Steam Generator (SG) Tube Inspection Reports to the NRC identifying flaws discovered during eddy current inspections. Six months after the inspections are completed, these detailed tube inspection reports become available to experts like me in the Public Document Room (PDR). Based on my prior industry experience, I knew that prolonged corrosive chemical exposure from extended shutdowns is deleterious to the metal components in both the Reactor Coolant and Secondary systems. I suspected that degradation was occurring at Palisades after it was permanently closed by Entergy in May 2022 and acquired by Holtec in June of 2022. But I had no hard data from the PDR to support my concerns. The last detailed Palisades SG tube Inspection Report in the PDR is from the 2020 SG inspections performed by Entergy. Five years of tube inspection data on both the primary and secondary systems is lacking from the PDR.
Since Holtec acquired Palisades, it appears to have used regulatory loopholes to avoid filing years of detailed Steam Generator Tube Inspection Reports indicating the extent of the damage.
The NRC Staff has even acknowledged that Holtec has failed to provide some Steam Generator inspection details, which is why the NRC staff delayed issuance of the SG sleeving LAR. Here is the NRCs statement about the cause of that schedule delay:
2 NRC staff has estimated that this licensing request will take approximately 940 hours0.0109 days <br />0.261 hours <br />0.00155 weeks <br />3.5767e-4 months <br /> to complete. The NRC staff expects to complete this review by September 30, 2025. Due to the eddy current qualification data not being provided by the licensee, the review date is beyond their originally requested date of August 15, 2025. (March 20, 2025, https://adamswebsearch2.nrc.gov/webSearch2/main.jsp?AccessionNumber=ML25076A177)
There are only two publicly available documents that discuss the condition of Palisades SG tubes. The first is the September 18, 2024 Preliminary Notification of Occurrence (PNO)
(ML24262A092) issued by the NRC staff based on their concerns after the shocking August 2024 Holtec SG inspection results. The second is a letter containing meeting notes from October 1, 2024 (ML24262A092) between Holtec and the NRC that summarize the August inspection and make vague promises about follow-up analyses. Thats it. If additional information is in the possession of the NRC staff, it should also be in in the PDR, and there is no such information.
That leads me to the conclusion that the NRC staff is not in possession of some critical Steam Generator tube inspection data from 2024 and 2025 or that the staff does not want the public to analyze the condition of the SGs.
In your September 3, 2025 meeting, the NRC staff told the ACRS that approximately 3,000 sleeves were inserted into about 700 tubes since May of 2025. Each sleeve is 18 inches long, which means that 4,500 feet of sleeves (0.85 miles!) were installed. That is an astounding length of sleeving and is not supported by the publicly available flaw data from the September 18 and October 1, 2024 PDR documents. For an expert like me, it would be a simple matter to compare the existing 2020 Entergy Inspection with both the 2024 and 2025 Holtec Inspections to search for trends and their root cause of the increased cracking indications, but none of the 2024 and 2025 inspection data is available. However, it appears likely that the tube damage that was identified and sleeved in 2025 exceeded the tube damage that was identified in 2024.
The general rule for plugging is that tubes are sleeved or plugged when an indication has reached or exceeded 40% through wall. So a 20% indication will not be plugged but will be reexamined during the next refueling outage based on Electric Power Research Institute (EPRI) water chemistry guidelines. But the chemical hideout at Palisades is anything but normal. When Holtec did examine the tubes in 2024, it found some previously unaffected tubes had Stress Corrosion Crack indications exceeding 80% through wall cracks after remaining in cold unpressurized water for two years. Slow, anticipated crack growth that EPRI assumes is not realistic for Palisades. Hence 3,000 sleeves, already a huge number, may be inadequate to prevent additional tube failures because of hideout before the next Palisades Steam Generator inspections.
Traditionally, eddy current testing begins several inches above the tube sheet. The tube sheet is part of the reactor coolant pressure boundary which is where chemical hideout would be expected to be most prevalent. Because of this hideout, it is not clear that either the SG tubes or the SG tube sheet will survive for even half a year after Palisades resurrection is complete.
Now, new information of degradation has become available. In addition to all the steam generator tube and tube sheet indications indicating both SCC and PWSCC in the steam generator, on August 20, 2025 Holtec filed a series of relief requests (ML25232A195 ) indicating
3 that it has discovered Primary Water Stress Corrosion Cracking (PWSCC) in at least eight dissimilar metal welds within Palisades Primary Coolant System. The affected welds include indications in two hot leg welds, four cold leg welds and two pressurizer welds.
The record indicates that Holtec did not take samples of either primary or secondary water chemistry at Palisades for two years and also that it is aware that Palisades was not in compliance with EPRI water quality guidelines. Clearly the absence of adequate water chemistry control at Palisades and its effect on the primary coolant system boundary are issues that deserve the thorough attention of the ACRS before allowing Palisades to set a new licensing precedent. This is a generic issue, as there are other decommissioned reactors now in the queue to be resurrected that have also not maintained adequate water chemistry during closure.
The existing evidence suggests that the reactor coolant pressure boundary degradation detected was caused by inadequate water chemistry control at Palisades, which places the facility in violation of two General Design Criteria:
Criterion 14Reactor coolant pressure boundary. The reactor coolant pressure boundary shall be designed, fabricated, erected, and tested so as to have an extremely low probability of abnormal leakage, of rapidly propagating failure, and of gross rupture.
Criterion 15Reactor coolant system design. The reactor coolant system and associated auxiliary, control, and protection systems shall be designed with sufficient margin to assure that the design conditions of the reactor coolant pressure boundary are not exceeded during any condition of normal operation, including anticipated operational occurrences.
The last time a steam generator tube completely ruptured was at Indian Point more than two decades ago. The condition of both the Primary Coolant System and the Steam Generators is even worse at Palisades with extensive SCC and PWSCC already identified. Luckily Indian Points design allowed it to dump the radioactive steam into the condenser where it was contained. Palisades does not have this feature and would use Atmospheric Dumps to discharge radioactivity directly into the atmosphere.
Previously, I have seen the ACRS advise the NRC staff and vendor (General Electric) of its concerns that regulatory expediency was placed before public safety. About two decades ago, I was one of a few experts who petitioned the ACRS to evaluate Net Positive Suction Head concerns relating to the request for regulatory relief on Containment Overpressure during Boiling Water Reactor Power Uprates. The ACRS did the right thing then by refusing to allow for the containment overpressure relief which was championed by the NRC staff and GE. I have previously applauded the ACRS personally for making that decision.
My concern initially started with SCC and PWSCC discovered in Palisades SGs but new Holtec relief requests have identified significant PWSCC corrosion at eight other locations within the reactor coolant system. The loss of the reactor coolant pressure boundary can lead to previously unimaginable impacts to the general public. The ACRS must be keenly aware of what could
4 happen in the event of primary coolant system failure or a Steam Generator tube failure due to years of neglect from improper wet layup by Holtec at Palisades.
I pray that you will thoroughly question the integrity of the reactor coolant pressure boundary and steam generator tubes caused by Holtecs failure to meet EPRI primary and secondary water chemistry standards before allowing Palisades to set a new licensing precedent.
Thank you, Arnie Gundersen Expert Witness for Beyond Nuclear, Dont Waste Michigan, et al.
Written Comment for the 729th ACRS Full Committee Meeting October 10, 2025 ACRS Written Comment - Palisades Steam Generator Life Cycle Considerations and Chemical Cleaning ConsiderationsAlan Blind
Dear Members of the Advisory Committee on Reactor Safeguards:
In your September 3, 2025 session, NRC staff, in response to your detailed questions, placed significant weight on the planned steam generator cleaning and emphasized the two-year period during decommissioningwhen chemistry was not programmatically maintainedas the primary path forward to ensure public health and safety. Staff also appeared to rely heavily on Framatomes technical evaluations, and that reliancewithout clear evidence of independent expert reviewshould itself be of concern to the ACRS.
While the decommissioning lay-up period deserves consideration, I respectfully urge the Committee to maintain a broader view that accounts for the full thirty-plus years of operating experience with the current Palisades steam generators. Too much emphasis was placed on assuming that two years without chemistry controls were the sole cause of the unexpected rate and severity observed in the 2024 inspection results. The longer history clearly shows that degradation mechanisms were active well before the recent lay-up period, and it is essential that the operational assessment reflect those cumulative effects. NEI 97-06 requires that all degradation mechanisms be evaluated and that uncertainties be summed in determining the appropriate period of operationmeaning the full prior life cycle of the steam generators must be considered.
Also, There must also be a questioning attitude about the specific chemical cleaning technique selected, its practical contribution to the operational assessment inputs, and whether it can realistically address the full life-cycle challenges of these steam generatorsincluding the embedded copper and lead deposits that have historically, up until 2010, and maybe beyond, proven difficult to remove.
My perspective comes from direct responsibility: in 2010, while serving as Engineering Director at Palisades, I was asked to lead an evaluation of whether secondary-side chemical cleaning could meaningfully extend the life of the replacement steam generators. That studycompleted more than a decade before Entergy elected to shut the plant down, and before many additional years of chemical deposit accumulationreached conclusions that remain directly relevant today.
In 2010, we concluded that without a proven chemical cleaning technique capable of removing the embedded and historically difficult copper and lead deposits, outside diameter stress corrosion cracking from these mechanisms would continue.
Even if cleaning could reduce the rate of degradation, it could not, by itself, assure operation through the 2031 license renewal period. We decided to not chemically clean the Steam Generators.
That finding remains directly applicable today. While chemical cleaning may contribute to life extension, it cannot be viewed as a stand-alone solution. The NRC and the Committee must weigh the broader record of operating experience, the persistence of copper and lead contamination, and the historical difficulty of fully removing hardened deposits when judging the adequacy of the operational assessment.
Operating History The Palisades replacement steam generators, Combustion Engineering Model 2530, were installed in late 1990 and entered service in 1991. The tubing is mill-annealed Alloy 600, a material widely recognized in industry for its susceptibility to corrosion. From the outset, over 300 tubes in each generator were preventively plugged, with additional plugging accumulating in every refueling cycle. By 2009, effective plugging rates reached ~5% in both units. Predictive models in the 2010 Entergy study showed that without chemical cleaning, Palisades would likely exceed its 15% plugging limit by the mid-2020s, short of the 2031 license renewal period.
Unique Issues Identified in the 2010 Study The 2010 chemical cleaning evaluation identified several unique challenges at Palisades:
Persistent sludge pile on the tubesheets containing hardened deposits of copper and lead, historically at Palisades, resistant to mechanical lancing.
Copper transport legacy from admiralty brass and copper-nickel condenser and heater tubing, which was replaced only after the steam generators were already in service.
Lead contamination originated from low-pressure turbine rupture discs, which were not replaced until 2012more than 20 years after the steam generators entered operation.
Top of tubesheet collars and tube scale deposits, containing copper and lead, nearly impossible to remove mechanically.
The report concluded that unless copper and lead deposits were removed, long-term service to 2031 was unlikely.
Chemical Cleaning Techniques Four principal methods were reviewed in 2010:
EPRI-SGOG process: most thorough and documented, capable of targeting copper/lead/iron; but costly and operationally intrusive.
ASCA (Advanced Scale Conditioning Agent): less expensive, widely used, but originally intended for maintenance rather than full sludge removal.
High-temperature chemical cleaning (HTCC): aggressive but risky; prior use at Waterford 3 created iron redeposition issues.
Deposit Minimization Treatment (DMT, AREVA, now Framatome):
newer, lower cost, but limited operating experience and uncertain effectiveness against copper/lead.
Unique Considerations at Palisades Because Palisades replacement generators entered service with pre-existing copper/lead contamination and hardened deposits, not all cleaning methods may be fully effective. The ACRS needs to question NRC staff on the particular cleaning method Holtec plans to use. The ASCA and DMT methods in particular may not penetrate hardened sludge piles or collars, raising questions about their adequacy
for the Palisades application. Even the most effective process (EPRI-SGOG) was predicted to provide only partial mitigation, not a guarantee of service through 2031.
Conclusion and Recommendation As you meet with NRC staff on October 10, I urge the Committee to:
1.
Acknowledge, as you have already done, the role of poor lay-up chemistry during decommissioning, but also weigh the prior three decades of degradation mechanisms that are equally important to todays operational assessment.
2.
Question the NRC staff closely on the basis for confidence in the specific cleaning method selected, particularly whether it can address Palisades hardened sludge piles and copper/lead contamination.
3.
Probe whether staff has critically evaluated vendor claims and benchmarked outcomes from other Alloy 600 plants.
4.
Ask the NRC staff how it expects the wide range of degradation mechanisms and they uncertainties to be quantified in the operational assessment.
Closing Remarks In closing, I urge the Committee to keep in mind the unique circumstances at Palisades. Steam Generator Life-cycle management must be considered in full, not just the two-year period of unmaintained secondary chemistry during decommissioning, but the entire thirty-plus years of operation that preceded it. The upcoming operational assessment will necessarily be far more complex than those at other plants, with a wide range of uncertainties for each degradation mechanism.
As required by NEI 97-06, these uncertainties must be treated quantitatively summed as the square root of the sum of the squaresnot minimized or considered in isolation. Finally, Palisades licensing basis for a steam generator tube rupture accident relies solely on the atmospheric dump valves for rapid depressurization when offsite power is available, and on primary system feed-and-bleedan operator-induced LOCAwhen it is not. While this approach is licensed and permissible, it must weigh heavily on any judgment about the acceptability of additional operating periods given the condition and age of these steam generators.
Respectfully submitted, Alan Blind, Retired Baroda, Michigan
Written Comment for the 729th ACRS Full Committee Meeting October 10, 2025 Re-Examining NRC Staffs Tube Integrity Assessment Response to ACRS: A Thought Experiment for ACRS Consideration Alan Blind At the September 3 meeting, in response to the ACRS request to NRC Staff to address the most severe cracks found in the October 2024 inspection, NRC staff stated that Palisades met tube integrity requirements during its last cycle because all 22 suspect tubes passed in-situ pressure testing, showing no rupture or leakage at three times normal operating differential pressure.
While NRC staffs statement is an assessment of tube integrity on the inspection date, I present this thought experimenta back-of-the-envelope review of the most degraded tube, R73C94not as a definitive analysis, but to highlight that the staffs response to the ACRS appeared to lack in-depth intellectual curiosity and
focus on conservative decision making. I respectfully suggest that the Committee probe this issue further in your October 10 meeting.
During Holtecs 2024 inspections, Tube R73C94 was identified as the most degraded and stood out with its long axial and deep penetration crack. It exhibited a 73% through-wall axial crack, 1.7 inches long, located at a tube support platea known site for corrosion and flow stagnation.
Crack Origin and Timing of Degradation Outer Diameter Stress Corrosion Cracking (ODSCC) requires three elements: a susceptible material, an aggressive environment, and sustained stress. The stress component is particularly important, arising from operating pressure differentials across the tube wall as well as thermal stresses from repeated heat-up and cool-down cycles during normal plant operation. During the shutdown and decommissioning phase, when Palisades steam generators were depressurized and at ambient conditions, these stress drivers were absenteven though water chemistry controls had also lapsed. This means that the severe crack found in Tube R73C94 could not have grown significantly during the decommissioning interval; instead, it must be conservatively assumed that the defect was already present at
the time of plant shutdown and entry into decommissioning. This point is important within my thought experiment because it indicates Palisades ended power operations already carrying a tube in near-failure condition, making the margin to rupture even smaller than NRC staff acknowledged.
Basic Concepts: Failure Mechanics and Stress Environment Stress Orientation and Crack Growth: Hoop (circumferential) stress is the dominant stress in thin-wall, pressurized tubes and acts perpendicular to axial cracks. Axial ODSCC forms along the tube axis due to stress concentration, stagnant chemistry, and thermal conditions near the TSP. In R73C94, the crack approached through-wall depth. As the wall thins, remaining ligament stresses increase non-linearly, leading to imminent structural failure.
Likely Failure Characteristics:
- Burst-Type Rupture - Failure would occur rapidly, resulting in a high-pressure rupture. Though the crack is axial, the final rupture would be circumferential due to internal pressure acting radially.
- Leak-Before-Break Unlikely - The high crack length and depth offered
This 4.5-month estimate represents a bounding upper limit. It does not account for the non-linear intensification of hoop stresses as the ligament thins, which could plausibly accelerate failure and lead to rupture sooner under actual operating conditions.
Key Implication Tube R73C94 was approaching structural failure with insufficient margin for detection or operator intervention. The most probable failure mode was a sudden, circumferential rupture at high pressurea scenario that would have posed a serious challenge for operator response and triggered a General Emergency declaration with potential offsite evacuation under Palisades licensing basis.
Thus, while tube integrity could be demonstrated at the time of the October 2024 inspection, the underlying data show Palisades was operating on the edge of a steam generator tube rupture. Continued operation for only a few additional months could have resulted in a markedly different outcome.
I respectfully submit this perspective for the Committees consideration: tube integrity assessments should not be viewed only as point-in-time
demonstrations, but also in the broader context of time-to-failure and the limited margins shown in the October 2024 results.
Alan Blind
Written Comment for the 729th ACRS Full Committee Meeting October 10, 2025 Reassessing NRC Benchmarks for Palisades Steam Generator Tube Integrity: CE Fleet Experience, 2024 Inspection Results, and the Role of Copper/Lead Oxides and Metal Deposits For Future At Power Operations Alan Blind Introduction In this comment, I will first explain why Beaver Valley is not a valid benchmark for Palisades steam generator performance, given the major design differences between Westinghouse and Combustion Engineering (CE) units. I will then compare Palisades operating history to the broader CE fleet, showing that Palisades has already exceeded the end-of-life service years of all other CE steam generators. Next, I will discuss why the extensive tube degradation found during Palisades 2024 inspection should not have been unexpected, based on known Alloy 600 behavior and the well-documented exponential growth of cracks.
Finally, I will highlight the unresolved issue of copper and lead deposits in Palisades steam generators, including the risk that these deposits may have oxidized during the uncontrolled chemistry period in decommissioning, and propose specific questions for the ACRS to raise with NRC staff at the November 10 meeting.
Note: This paper refers to copper and lead deposits in Palisades steam generators. For a more complete explanation of this issueincluding chemical cleaning considerationsI respectfully direct the Committee to my separate comment titled Palisades Steam Generator Life Cycle Considerations and Chemical Cleaning Considerations Benchmarking At the September 3 meeting, an ACRS member asked NRC staff what other plants inform the Palisades design and SG tube evaluations. Staff replied that only one other plant with Alloy 600 tubing remains for benchmarking: Beaver Valley. The staff appeared to place some weight on this benchmark as a validation of their evaluation of Palisades vendors analysis of steam generator tube condition.
This comparison is deeply problematic. Beaver Valley is a Westinghouse three-loop PWR, not a CE two-loop plant like Palisades. That difference matters technically when thinking about ODSCC mechanisms:
Loop & system layout: Westinghouse SGs at Beaver Vally are designed for three-loop primary systems; CE plants like Palisades operate with two large steam generators and four reactor coolant pumps, a different hydraulic and thermal environment.
Support structure architecture: Westinghouse SGs use broached or quatrefoil tube support plates; CE SGs (including Palisades Model 2530s) use egg-crate lattice supports with vertical/diagonal straps and a stay-cylinder region. This geometry dictates how and where ODSCC initiates, as well as how inspections and repairs must be performed.
Tube bundle geometry: Westinghouse SGs use all U-bend tube bundles, while CE SGs combine both: at Palisades, rows 1-18 are U-bends and rows19-165 are square bends, supported by egg-crate lattices, vertical straps, and diagonal straps. This difference in tube geometry creates very different stress and wear environments, as well as distinct inspection challenges.
Tube scale and collars unique to CE: At Palisades, tube scale and top-of-tubesheet (TTS) scale collars remain in place on the outside of tubes. These collars, along with copper/lead scale, are extremely hard deposits nearly impossible to remove mechanically because of the CE designed tri-pitch tube geometry. They directly promote ODSCC in dents, freespan, and egg-crate regions. Westinghouse SGs do not face this same combination of tri-pitch geometry and persistent areas of scaling, including for Palisades, copper/lead collars.
Materials: Both plants may have Alloy 600 tubing, but Palisades specific mill-annealed Alloy 600 (0.75 OD x 0.042 wall), coupled with CE support geometry, has shown a much higher susceptibility to ODSCC than most Westinghouse designs.
Because material, geometry, and scale/collar buildup drive ODSCC progression, Beaver Valley cannot be considered a technically valid benchmark for Palisades.
A More Meaningful Benchmark: The CE Fleet Experience
Rather than relying on Beaver Valley, the more meaningful benchmark is the actual fleet history of CE plants with steam generators of similar mechanical design.
Following are the nominal steam generator effective full-power years (EFPY), as of the year 2010, at which other CE units reached end-of-life and replaced their steam generators:
Palo Verde 1 - 15 EFPY Palo Verde 2 - 14.8 EFPY Palo Verde 3 - 20 EFPY Arkansas Nuclear One 2 - 15.5 EFPY Calvert Cliffs 1 - 19 EFPY Calvert Cliffs 2 - 18 EFPY St. Lucie 1 - 16 EFPY St. Lucie 2 - 20 EFPY Fort Calhoun - 25 EFPY San Onofre 2 - 21 EFPY
San Onofre 3 - 21 EFPY Waterford 3 - 23 EFPY In every case, CE units retired their steam generators well before reaching Palisades present replacement steam generator operating age. Palisades today is estimated at ~28 EFPY of operation on its replacement steam generators, already exceeding the highest steam generator service life of any other CE unit.
Note: I do not indicate the tube alloy used at each of these CE plants, as that information is not always reliably available. It is well understood that allow 600 tubes, like Palisades, are the most susceptible alloy. However, regardless of alloy differences, the fact remains that the entire CE fleet reached end of SG life well before Palisades current estimated EFPY age. This makes the CE fleet record the most meaningful benchmark for ACRS consideration of Palisades tube condition.
Why the 2024 Palisades Inspection Results Were Predictable Viewed against this fleet record and the well-established behavior of Alloy 600 tubing, Palisades 2024 inspection resultsshowing widespread and exponential
increase in cracking and tube degradation at roughly 28 EFPYshould not have come as a surprise. Let me explain:
Established damage modes: Alloy 600 steam generator tubing is prone to both mechanical wear (caused by interaction with support structures such as egg-crates, straps, and tube-to-tube contact) and stress corrosion cracking (both primary water SCC and outside-diameter SCC). These modes are well documented across the CE fleet.
Exponential growth of cracks: Once initiated, cracks do not grow in a linear fashion; instead, they accelerate, often increasing dramatically after a relatively quiet incubation period. This behavior aligns with industry findingsfor example, A Review on the ODSCC of Steam Generator Tubes by H. Chung et al. (2013) describes ODSCC as a thermally activated process that accelerates once initiated. See Summary of Literature Search.
Role of deposits: At Palisades, persistent copper and lead deposits in sludge piles, tube scale, and TTS collars have created a corrosive environment that promotes ODSCC. These deposits act as stress concentrators and chemical initiators, ensuring that once cracking began, its progression would be rapid.
Summary of Literature Search: End-of-Life and Rapid Exponential Crack Growth Increasing Stress Intensity Factor The growth of a crack is primarily driven by the stress intensity factor (K) at its tip.
Because K depends directly on crack size, as a crack becomes longer or deeper the stress at its tip increases. This accelerates the growth rate and creates a positive feedback loop: a larger crack produces a higher stress intensity, which in turn drives the crack to grow even faster.
Crack Coalescence During the incubation period, numerous small, isolated microcracks can form in high-stress regions, such as tube support plate crevices. As these microcracks extend, they begin to link and merge into larger, more significant cracks. This coalescence dramatically increases the effective crack size, producing a sudden surge in growth rate and a rapid increase in the number of detectable defects during inspections.
Worsening Chemical Environment The aggressive chemical environment within crevicesnecessary for ODSCC initiationintensifies over time. Localized boiling, driven by heat transfer from the primary coolant, concentrates impurities inside the narrow crevice region, sometimes by factors as high as 10. Continued tube operation allows sludge and deposits to accumulate, restricting water flow and further concentrating chlorides, sulfates, and caustics. The result is a progressively more aggressive microenvironment that accelerates crack initiation and growth.
Sources Chung, H., Hong-Deok Kim, Seungjin Oh, et al. A Review on the ODSCC of Steam Generator Tubes in Korean NPPs. Nuclear Engineering and Technology, Vol. 45, No. 4, August 2013.
Cizelj, L. Trends of Degradation in Steam Generator Tubes, 1998.
U.S. NRC, NUREG/CR-5752, ANL-99/4, Assessment of Current Inspection and Monitoring of Steam Generator Tubes.
Palisades Copper and Lead Deposits: The Unresolved Question of Oxidation Beyond these fleet-wide degradation patterns, Palisades faces an additional, unresolved risk: the extent and persistence of copper and lead accumulations within its steam generators. These deposits are concentrated in precisely the locations most vulnerable to ODSCCdents, freespan regions, egg-crate supports, and the tubesheet transition. Historically these deposits were present in metallic formhence hard to remove.
Basic chemistry shows why this distinction matters. Metallic copper and lead are relatively stable, but when exposed to oxygenated water or uncontrolled secondary-side chemistry, they can oxidize to form compounds such as copper oxides or lead oxides. In oxidized form, these compounds are well known to promote electrochemical reactions that degrade Alloy 600 tubing and accelerate ODSCC initiation and growth.
During Palisades decommissioning period, for nearly two years, secondary-side chemistry was not controlled or monitored. This raises a critical unresolved issue:
whether some portion of the long-standing copper and lead metallic deposits in Palisades steam generators may have shifted from metallic to oxidized form under
these conditions. If so, the potential for localized aggressive environments around the tube surfaces will be a factor for future power operations, when stresses are again introduced.
This is an important consideration that requires further explanation. Specifically, NRC staff should clarify to the ACRS:
Whether copper and lead deposits at Palisades have oxidized following the extended period of uncontrolled secondary chemistry; What impact any such oxidation would have on ODSCC progression during the approved operating interval; and How these risks will be addressed in the upcoming operational assessments before the next full inspection of the steam generator tubes.
In short, the chemistry of these deposits is no longer an academic question. It is central to whether Palisades can reasonably be assumed to maintain tube integrity until the next mandated inspection.
Conclusion
Taken together, these factors meant that Palisadesoperating beyond the demonstrated steam generator service lives of all other CE plantswas virtually certain to experience extensive tube degradation by the time of its 2024 steam generator tube inspection. Perhaps, in 2020 Entergy understood this very well, and it was a factor in their decision to discontinue future operations?
The October 2024 inspection results validated long-standing industry knowledge about Alloy 600 tubing rather than representing an unforeseen development. They confirm that the aging process for these steam generators follows the well-documented trajectory of slow initiation followed by rapid, exponential crack growth.
When viewed through the broader lens of CE plant steam generator experience clearly showing the industry end of CE steam generator life, and the extensive research on rapid, exponential crack growth after prolonged years of operation, the ACRS must press the NRC to explain how continued operation in this high-risk region of accelerated crack growth can be considered acceptable in the operational assessment when defining the permissible operating period before the next shutdown and inspection.
Moreover, the current 2024 Palisades data available today indicates that these accelerated growth rates are continuing unabated, as the research indicated. The literature shows that little can be done to halt degradation once it has been set in motion years earlier, given the inherent metallurgy, tube geometry, and crevice chemistry conditions driving the process. This reality underscores the urgency of adopting truly conservative decision-making goin forward.
Suggestions for ACRS Questions at the November 10 Meeting Given the CE fleet experience, the unique design differences of Palisades, and the predictable progression of ODSCC, I respectfully suggest that the Committee ask NRC staff at the November 10 meeting to address the following questions:
Benchmarking Basis: Why is Beaver Valleya Westinghouse three-loop plant with fundamentally different steam generator design features considered a meaningful benchmark for Palisades, when the CE fleet record provides a more relevant basis for comparison?
Fleet Experience Envelope: Does NRC staff agree that Palisades, at ~28 EFPY, is already operating beyond the demonstrated service lives of all
other CE steam generators, and if so, what technical justification supports allowing further operation?
Deposit-related Degradation: How is NRC staff accounting for the persistent copper/lead scale, tube scale, and top-of-tubesheet collars unique to CE SGsfeatures that are absent in Westinghouse designs but known to accelerate ODSCC at Palisades?
Predictability of ODSCC: Given the well-established, non-linear progression of ODSCC once initiated, does NRC staff agree that Palisades 2024 inspection results were predictable, and if so, how is this being factored into the operational assessment and inspection interval planning?
Alan Blind
Five-Minute Verbal Comment - ACRS Full Committee, October 10, 2025 Alan Blind Good afternoon, Members of the Committee.
I appreciate the opportunity to speak. I will summarize my written comments today. Let me emphasize up front: the October 2024 Palisades inspection results confirm that accelerated crack growth is continuing unabated. There is nothing Holtec can do now to reverse this. The literature shows this is exactly what happens once the degradation process has been set in motion years earlier. Given the inherent metallurgy, tube geometry, and crevice chemistry of Alloy 600 steam generator tubing, very little can be done to halt it once underway.
By way of background, I served as the Engineering Director at Palisades.
In that role, I managed the benchmarking of Palisades steam generator remaining life against the full population of Combustion Engineering plants. That benchmarking effort, completed in 2010, showed clearly that Palisades was at the front edge of the CE fleet experience curve, where the rate of outside-diameter stress corrosion cracking accelerates dramatically what we described as a third-derivative growth curve.
We also accounted for plant-specific history. Decisions made in the 1990s introduced persistent lead and copper deposits into Palisades steam generators. These scale collars and sludge piles are extremely difficult to remove and created localized conditions that exacerbated ODSCC.
Combined with the known morphology of Alloy 600 cracking and the CE fleet benchmarking, we projected that Palisades would reach the 15%
tube-plugging limit around 2025. The Holtec October 2024 inspection results confirmed that projection almost exactly.
Those results show rapid expansion of both the rate and severity of ODSCC cracks. Several tubes approached or exceeded 90 percent through-wall depth, with one tube R73C94 within months of rupture under operating conditions. This is not random chance, and not simply the result of the recent period without secondary water chemistry controls. It is the textbook exponential growth phase of Alloy 600 cracking a progression that began decades ago, and one we already understood and anticipated in 2009.
I applaud the Committee for shining a light on the Operational Assessment, and I recognize that in your September 22 letter to the Commission you highlighted the uncertainties introduced by the extended layup without chemistry control. That is indeed a concern, and perhaps
one that chemical cleaning may help address. But the broader and more fundamental issue is this: the accelerated crack growth rates documented in 2024 show that Palisades has entered the high-risk, end-of-life phase where failures can occur suddenly and no credible mitigation strategy exists. The lack of proper wet layup during decommissioning did not cause this the process was already well underway.
Fleet experience we reviewed in 2009 confirms this. Every other CE plant replaced its steam generators well before reaching this point. None attempted to continue operation in the exponential growth region of Alloy 600 degradation. Yet here we are in 2025, and Palisades is the outlier.
So the central question before you, as advisors to the Commission, is this:
how can continued operation in this region of accelerated crack growth be considered acceptable in an Operational Assessment that defines the permissible interval before the next inspection? The slope of the exponential curve is approaching infinity.
In my view, conservative decision-making is the only defensible path forward. The data, the published research, the CE fleet benchmarking, and now the Palisades 2024 inspection all point to the same conclusion: once initiated, this degradation cannot be arrested. Palisades has crossed into a
region of unacceptable risk, and that fact must shape both NRC staffs review and the Committees recommendations.
Thank you.