ML22335A495
| ML22335A495 | |
| Person / Time | |
|---|---|
| Issue date: | 11/16/2022 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| NRC-2170 | |
| Download: ML22335A495 (1) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Fuels, Materials, and Structures Subcommittee Docket Number:
(n/a)
Location:
teleconference Date:
Wednesday, November 16, 2022 Work Order No.:
NRC-2170 Pages 1-91 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 FUELS, MATERIALS, AND STRUCTURES SUBCOMMITTEES 7
+ + + + +
8 WEDNESDAY 9
NOVEMBER 16, 2022 10
+ + + + +
11 The Subcommittee met via hybrid in-person 12 and Video Teleconference, at 1:00 p.m. EST, Ronald 13 Ballinger, Chairman, presiding.
14 COMMITTEE MEMBERS:
15 VICKI BIER, Chair 16 RONALD G. BALLINGER, Member 17 CHARLES H. BROWN, JR., Member 18 VESNA DIMITRIJEVIC, Member 19 GREGORY HALNON, Member 20 WALT KIRCHNER, Member 21 JOSE MARCH-LEUBA, Member 22 DAVID PETTI, Member 23 JOY L. REMPE, Member 24 MATTHEW SUNSERI, 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 ACRS CONSULTANT:
1 STEPHEN SCHULTZ 2
3 DESIGNATED FEDERAL OFFICIAL:
4 MIKE SNODDERLY 5
6 ALSO PRESENT:
7 RYAN HOSLER, Framatome 8
ANDREW MORLEY, Public Participant 9
CAROL MOYER, NRR 10 DAVID RUDLAND, NRR 11 CHRIS WAX, EPRI 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
3 T-A-B-L-E O-F C-O-N-T-E-N-T-S 1
Opening Remarks and Objectives 2
By Professor Ronald Ballinger, ACRS....
4 3
Staff Opening Remarks 4
By David Rudland, NRR...........
6 5
Issue and Staff Perspective 6
By Carol Moyer, NRC Staff.........
7 7
Industry Perspective EPRI/PWROG Focus Group 8
By Ryan Hosler, Chris Wax, EPRI...... 52 9
Public Comments 10 By Professor Ronald Ballinger....... 90 11 Adjourn 12 By Professor Ronald Ballinger....... 90 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
4 P-R-O-C-E-E-D-I-N-G-S 1
1:00 p.m.
2 CHAIR BALLINGER: Okay, the meeting will 3
now come to order. This is a meeting of the Fuels, 4
Materials, and Structures Subcommittee of the Advisory 5
Committee on Reactor Safeguards. I'm Ron Ballinger, 6
chairman of today's subcommittee meeting. ACRS 7
members present are Jose March-Leuba, Dave Petti, I 8
think Matt Sunseri will be here, is here.
9 MEMBER SUNSERI: Yeah, I'm here.
10 CHAIR BALLINGER: He's here. Joy Rempe, 11 Vicki Bier, Greg Halnon, Charlie Brown was here, I 12 think he'll be here, and let's see who else is here.
13 Vesna Dimitrijevic, and if I've missed anybody, I --
14 Walt Kirchner, and that probably should cover it, that 15 covers it. Mike Snodderly is the ACRS staff member 16 that's designated federal official for this meeting.
17 During today's meeting, the subcommittee will have an 18 information briefing with the NRC staff, and EPRI on 19 the French PWR safety injection system cracking.
20 I need to say ahead of time, this is an 21 open meeting. Nothing that we are going to have 22 presented here today is not publicly available.
23 Personally, I was on the committee in France for EDF, 24 along with another person who is here today, and we 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 will -- I will refrain from making any comments 1
related to the presentations, because for fear that I 2
might say something that's not open.
3 So, just to be careful with that, and we 4
would be happy to send these presentations to our 5
colleagues at EDF, just so that they know what was 6
said here. The rules of participation in all ACRS 7
meetings including today's were announced in the 8
Federal Register on June the 13th, 2019. The ACRS 9
section of the U.S. NRC public website provides our 10 charter, bylaws, agendas, letter reports, and full 11 transcripts of all full, and subcommittee meetings 12 including the slides presented here.
13 The meeting notice, and agenda for this 14 meeting were posted there. We have received no 15 written statements, or requests to make an oral 16 statement from the public. The subcommittee will 17 gather information, analyze relevant issues, and 18 facts, and formulate positions, and actions as 19 appropriate for deliberation by the full committee.
20 The rules for participation in today's meeting have 21 been announced as part of the notice of this meeting 22 previously published in the Federal Register.
23 Today's meeting is a hybrid meeting, is 24 being held in person, and over Teams. The bridge line 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 allowing participation of the public over the meeting 1
has been made available. A transcript of today's 2
meetings is being kept, therefore we request that 3
meeting participants on Teams, and on the Teams call 4
in line identify themselves when they speak, and to 5
speak with sufficient clarity, and volume so that they 6
can be readily heard.
7
- Likewise, we request that meeting 8
participants keep their computer, and, or telephone 9
lines on mute, otherwise we get feedback, and things.
10 The chat feature on Teams should not be used for any 11 technical exchanges. Okay, make sure everybody's 12 muted, okay. So, Dave? At this point I'll turn it 13 over to Dave Rudland of the NRC staff for initial 14 comments.
15 MEMBER SUNSERI: Thanks Ron. For those of 16 you who don't know me, my name is Dave Rudland, I'm a 17 senior technical advisor for materials in the division 18 of new and renewed licenses in NRR. And first off, I 19 want to thank you for the opportunity to come here, to 20 talk to you about this very important foreign 21 operational experience. And I wanted to let you know 22 that the NRC staff have been heavily involved in the 23 evolution of this issue.
24 We were first informed of it about a year 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 ago with the first cracking at Civaux Plant, and we 1
have been following it with the help of our 2
counterparts at ASN, the French regulator, as well as 3
at IRSN, ASN's technical service organization. So, 4
again, I just want to say thank you, the staff have 5
been following this, and we're happy to present some 6
of the things that we've learned over the last year.
7 And to hear from our EPRI counterparts 8
about their efforts, and how this relates to the U.S.
9 fleet. So, thank you very much.
10 CHAIR BALLINGER: Okay, I guess Carol, are 11 you on?
12 MS. MOYER: My name is Carol Moyer, I'm a 13 senior materials engineer in the Nuclear Regulatory 14 Commission Nuclear Reactor Regulation Office --
15 CHAIR BALLINGER: You're sounding like 16 you're a little bit in, and out, these microphones are 17 highly directional, so.
18 MS. MOYER: Is that better? Okay, I have 19 to slouch. Great, so thank you for this opportunity 20 to discuss this topic. As Dave already explained, 21 this is not our firsthand knowledge, this is a 22 compilation of what we've been able to glean from 23 various sources, and so we hope that we can add value 24 by pulling these facts together. But we're looking 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 forward to the discussion.
1 Okay, a quick overview of the operating 2
fleet. There are 56 reactors in operation, all of 3
them are pressurized water reactors built in the 70s, 4
through the 90s. They're in three main styles, 5
roughly classed as the 900 megawatt, 1300 megawatt, 6
and 1450, and up megawatt. The original 900 megawatt 7
version was based on a Westinghouse design, and the 8
others have been an evolutionary development from 9
there with some modifications to make them fit well in 10 the French grid.
11 Okay, so this is an overview of the time 12 line of these observations. As David mentioned 13 earlier, this was first discovered a little over a 14 year ago in last October. Flaw indications were 15 detected near welds in safety injection lines during 16 their scheduled ten year safety inspections. So, they 17 were first discovered at Civaux One and Civaux Two and 18 Poly One, and I'm going to try not to butcher the 19 French names too much.
20 Civaux One and Civaux Two, then at the end 21 of last year. The indications that were found in the 22 safety injection lines, and residual heat removal 23 lines on the IDF pipes circumferentially located 24 oriented -- jumping to the end. The laboratory 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 analysis confirmed that these were intra granular 1
stress corrosion cracking cracks. So, the regulator 2
ASN decided to expand the inspections, and reevaluate 3
prior NDE data to get an idea of the extended 4
condition.
5 MEMBER HALNON: This is Greg, is that near 6
welds, or in the welds, flaw indications? Is it a 7
heat affected zone?
8 MS. MOYER: Yes, adjacent to welds would 9
be a description.
10 MEMBER HALNON: Okay, so it is a part of 11 the weld essentially, issue.
12 MS. MOYER: Essentially, yeah. And you'll 13 see in the cross section that most of the cracks were 14 associated with welds, but not all. This was a nice 15 photo that I lifted from an EDF slide with permission 16 that shows that layout of the affected loops. This 17 diagram at the bottom, you'll see again, if I can 18 point at things. So, there are welds at elbows, and 19 various sections of pipe, and the cracking, as I said, 20 has been mostly associated with welds throughout the 21 injection lines.
22 So, with four separate loops, they're 23 stainless steel pipes, they are roughly an inch thick, 24 and eight inches diameter, give, or take.
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 MEMBER MARCH-LEUBA: One thing I'm reading 1
on the slide is that the long pipes with the stagnant 2
flow, or no flow at all.
3 MS. MOYER: Yes, we're going to get to 4
that too. Yes, there are pipelines of varying 5
lengths, and that we think is important.
6 MEMBER PETTI: Which alloy of stainless 7
steel is it, do you know?
8 MS. MOYER: 316LN. So, this is another 9
just blow up of -- well, a hand drawing, and a for 10 real photograph of the location of some of these 11 cracks. So, again, they are at the connections 12 between pipes, and elbows for example. Okay, so as I 13 said, they expanded the inspections, and tried to 14 understand the extent of condition. EDF was asked to 15 accelerate their plans to inspect safety injection 16 piping for similar degradation from the spring, and 17 through the summer.
18 The regulator also requested additional 19 information to assess the degradation in its extent to 20 determine whether it was a generic issue. Starting 21 with this summer, EDF deployed a new NDE method to 22 detect, and size flaws, and we'll talk about that a 23 little bit more also. So, they used a couple of 24 different methods to detect flaws, and to characterize 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 what was found.
1 The stress corrosion cracking indications 2
were mostly found in the 1300 megawatt, and 1450 3
megawatt type reactors, that I believe only one crack 4
was found in the 900 megawatt plants, and that we 5
think is also important. So, there is a list here of 6
plants that either had confirmed flaws, or had results 7
that suspected flaws that needed to be further 8
evaluated.
9 The list is not comprehensive, because it 10 kept changing as I looked at different reports, and 11 things, because no one was quite sure what was 12 confirmed, but let's say it's a lot. So, the pipes 13 were at -- the cracks were at pipe ID, located at all 14 those. When they did destructive examination of pipe 15 sections removed, they confirmed that these were 16 intragranular stress corrosion cracking in the base 17 metal 316L, or LN I later learned, and heat affected 18 zone adjacent to the welds.
19 Coming up in a couple slides I have a 20 photo, it seems in many cases the cracks started in 21 the heat affected zone, and grew to the weld, and then 22 stopped. But in some cases the extension could reach 23 the full circumference of the ID. Because of the 24 design of these lines, and the stagnant, 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
12 intermittent flow, there was a suspicion that they 1
could be susceptible to thermal fatigue cracking, and 2
that's really what they were looking for with the 3
ultrasonic test that they were doing on these pipes.
4 Although I'm told that they did not really 5
expect thermal fatigue cracking at this stage of life, 6
and the plant, for defense in depth reasons, they were 7
inspecting anyway, and were somewhat surprised to find 8
-- obviously were surprised to find the SCC cracking.
9 Most of the cracks were shallow, around a millimeter 10 deep. A few at heat affected zones were as much as 11 six millimeters deep. So, about an inch, or about a 12 quarter of the way through the wall of the pipe.
13 And yes, stagnant flow, and thermal 14 stratification are potential contributing conditions.
15 Okay, so it's our understanding that over 100 of these 16 faults have now been examined, either with a sensitive 17 penetrate test, or destructive examination, cutting 18 them apart, and actually opening the cracks, looking 19 at them. Cracked elbows were sent to the EDF's hot 20 lab.
21 Some were also examined by IRSN, the 22 technical support organization to the regulator, and 23 they were able to confirm then that there were IGSCC 24 cracks in the base metal, and heat affected zone, as 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 I said, max depth was about a quarter of the way 1
through wall. They also found elevated hardness in 2
the vicinity of the weld root pass. Not extreme, but 3
somewhat elevated, enough to have that noted I guess.
4 They also found unusual weld geometry in 5
some cases. There was the first pass, the root pass 6
of the weld in some of the cross sections was 7
unusually deep, or high. Without really a good 8
explanation for that, so it is possible that some off 9
normal welding conditions led to unexpected residual 10 stresses. However, there was no evidence of chemical 11 contamination.
12 You always look for chlorides, and things 13 like that that might exacerbate a CC, that was not 14 found. Okay, so here's a cross section. So, the 15 center picture shows one of these cracks that's five, 16 or six millimeters deep, and it's in the heat affected 17 zone adjacent to a weld, the crack tip goes up to the 18 fusion line, and then stops.
19 MEMBER MARCH-LEUBA: So, for us, the 20 people who don't know exactly what we're looking at, 21 the connect is the little hair line that we see at the 22 top of the cone, or where is it?
23 MS. MOYER: Yes, I'm sorry, if I can 24 figure out how to point on this. There we go, okay.
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 So, this is the pipe ID, the pipe inner wall, outer 1
wall, weld crown, weld root pass is here, and then 2
this small dark line beside the weld is the crack.
3 MEMBER MARCH-LEUBA: And the cone, the 4
thing that looks like a mountain is the weld?
5 MS. MOYER: That's the weld, yes. So, 6
this is the cast elbow on the left I believe. I don't 7
have an annotated picture in front of me, but I'm 8
pretty sure this is the cast elbow on the left, and 9
this would be a wrought stainless pipe on the right.
10 MEMBER HALNON: Was it always on the 11 vertical section that side of the elbow, on the 12 vertical?
13 MS. MOYER: I don't know the answer to 14 that, I don't think so, I think they were distributed 15 throughout these loops. And there were some that were 16 cracked on both sides of the same weld.
17 MEMBER KIRCHNER: This is Walt Kirchner, 18 I'd like to ask, you said the elbow is cast?
19 MS. MOYER: That is my understanding.
20 MEMBER KIRCHNER: That's -- when you lay 21 down a weld like this next to a cast, I don't know 22 Ron, that's not surprising to me.
23 CHAIR BALLINGER: I know nothing.
24 MEMBER KIRCHNER: Okay, you're our expert.
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 CHAIR BALLINGER: Can you say something?
1 MR. RUDLAND: I'm actually not sure if 2
they're cast, or not.
3 CHAIR BALLINGER: I don't think so, but --
4 MEMBER KIRCHNER: I would be surprised, 5
that's why I raised the question. Because the 6
difficulty of welding with castings is well known, but 7
8 MR. RUDLAND: But there's a sufficient 9
number of cast elbows in the French plants, because 10 they had an awful lot of thermal embrittlement issues 11 with their elbows, that they actually went through a 12 replacement.
13 MEMBER KIRCHNER: Yeah, of course.
14 MR. RUDLAND: But I'm not sure about these 15 to be honest.
16 MS. MOYER: It's second hand information, 17 but I think these are 316 elbows as well, not CFAs 18 like that, so perhaps they are -- nevertheless. So, 19 this is a cross section of a butt weld. So, you have 20
-- sorry for the folks online, it's a pipe, and an 21 elbow, and the weld, you start at the root pass, and 22 then grow out. So, the ends would have been dressed 23 at an angle, and then that gap is filled with weld 24 metal as you increase the circumference.
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 So, then these cracks occurred at the ID 1
surface, at the inner surface, and grew towards the 2
outer surface, but not very far.
3 MEMBER MARCH-LEUBA: It's easy to see how 4
you would leave a seal on the stresses there. Is 5
there any treatment afterwards, or? Because this is 6
7 MS. MOYER: It's my understanding that 8
these were plant welds, not shop welds. So, there is 9
likely not any post weld heat treatment applied, they 10 would be in the as welded conditions.
11 CHAIR BALLINGER: That I can say, you 12 don't post weld heat treat stainless steel welds, 13 otherwise you'd sensitize them. So, it's one of those 14 things where you just can't. It's good news, and bad 15 news with stainless steel, that's the good news, the 16 bad news, it's stainless steel.
17 MEMBER MARCH-LEUBA: I'm going back with 18 the CRE, with the professional, since the suspicion is 19 that this is stress corrosion cracking, it makes sense 20 that the internal, where you have the water. Can you 21 say yes for the record?
22 MR. RUDLAND: Yeah, the cracks are on the 23 wet surface.
24 MEMBER MARCH-LEUBA: On the wet surface, 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 where you will have the corrosion mixing.
1 MR. RUDLAND: That's correct.
2 MEMBER BROWN: That means the inside of 3
the pipe?
4 MR. RUDLAND: That's correct.
5 MEMBER BROWN: How in the world did you 6
find them in the first place then? You had to drain 7
everything, and then inspect it? I'm an electrical 8
guy, so I've got to ask.
9 MS.
MOYER:
They're inspected by 10 ultrasonic.
11 MEMBER BROWN: Okay, so exterior you can 12 pick it up?
13 MR. RUDLAND: They inspect on the outside.
14 MEMBER BROWN: I got that part. My memory 15 goes back 20 years --
16 MS. MOYER: They're running a probe down 17 the outside of the pipe.
18 MEMBER BROWN: So I'm starting to remember 19 some stuff.
20 MEMBER MARCH-LEUBA: This is actually an 21 ultrasonic picture, right?
22 MS. MOYER: No, this is a destructive 23 examination. They found the crack by ultrasonic, so 24 non-destructive examination, NDE, and then once 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
18 had located a crack, they cut that section of pipe 1
out, replaced it with a new one, and took the broken 2
one to the lab.
3 MEMBER MARCH-LEUBA:
So, that's a
4 microscope picture.
5 MS. MOYER: This is a cross section, 6
polished.
7 MEMBER SUNSERI: This is Matt, I have a 8
question, Matt Sunseri.
9 MS. MOYER: Yes?
10 MEMBER SUNSERI: So, maybe I don't 11 remember this correctly, but my recollection is for in 12 a granular stress corrosion cracking that occurs, you 13 need three things to happen. One is to have 14 susceptible material, two corrosive environment, and 15 three high temperature. So, it sounds like one, and 16 two are met, but the temperature seems a little low to 17 me, this is something south of 600 degrees Fahrenheit, 18 300 degrees C, so are these welds exposed to higher 19 temperature than that?
20 MS. MOYER: No, the essential requirements 21 are susceptible material, environment, and stress, not 22 necessarily temperature. You can get to stress by 23 temperature perhaps, and we may have here by 24 differential heating, but temperature alone doesn't 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 necessarily alone get you to SCC.
1 MR. RUDLAND: This is Dave Rudland, it's 2
more the environment, the chemistry of the water that 3
could cause that. So, there's certain chemicals in 4
the water that can cause susceptibilities to be worse.
5 So, if you have a high oxygen content in the water, it 6
makes that location more susceptible. So, when they 7
talk about the environment, they're talking in this 8
case more about the water chemistry than they are 9
about the temperature.
10 The temperature feeds into the growth rate 11 more so than the --
12 MEMBER MARCH-LEUBA: Asking questions 13 about the theory, 100 percent theory, is this an 14 accident event, I mean a transient, of having a crack 15 in a second, or is it a very long corrosion that goes 16 crack, crack, crack over months?
17 MR. RUDLAND: Yeah, it's a time dependent 18 cracking event.
19 MEMBER MARCH-LEUBA: It's a long time?
20 MR. RUDLAND: It took a while for the 21 crack to initiate, and a while for the crack to form.
22 MEMBER MARCH-LEUBA: So, it was not like 23 a surge of some very hot water.
24 MR. RUDLAND: That is 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
20 MEMBER HALNON: Matt, go ahead if you need 1
to finish your question.
2 CHAIR BALLINGER: I think it was done.
3 MEMBER DIMITRIJEVIC: I have a question, 4
this is Vesna Dimitrijevic. My question is we usually 5
don't see IGSCC in the pressurized water reactors in 6
the U.S., right? That's a more BWR problem. I mean 7
we might --
8 MS. MOYER: I think the answer to your 9
question is --
10 MEMBER DIMITRIJEVIC: You know, when we 11 analyze the degradation mechanism in class one piping 12 in the PWRs, that was never damage mechanisms. It was 13 the damage mechanism we saw in the BWRs.
14 MS. MOYER: Yeah. IGSCC has been reported 15 in BWRs primarily. When we see stress corrosion 16 cracking in pressurized water reactors, we tend to 17 call it primary water stress corrosion cracking, 18 PWSCC, but it's not really that different.
19 MR. RUDLAND: Yeah, and actually I think 20
-- this is Dave Rudland again -- the susceptibility of 21 the stainless steel in pressurized water is not as 22 high as it is in the water in BWRs, because of the 23 chemistry again. And so the cases in the U.S. where 24 we have seen stress corrosion cracking in stainless 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 steel have been kind of limited to locations where 1
there's something special going on.
2 There's been excessive grinding, or 3
something that really increases the stresses at the ID 4
surface, or in stagnant conditions, where you may have 5
water chemistry issues.
6 MEMBER DIMITRIJEVIC: I see, I mean I was 7
a little surprised to see this, but you know, I know 8
we have IS, integration stress corrosion cracking 9
program for BWRs, but -- all right, thanks.
10 MS. MOYER: Okay, as I was starting to 11 say, cracks were detected by manual ultrasonic 12 testing. So, in this case you're putting a sound wave 13 into the pipe from the outside, and listening for, 14 checking for a response, a reflection back to the 15 transducer. They use a 45 degree sure wave at 2.25 16 megahertz. This is pretty normal for looking for 17 thermal fatigue cracks, which is what they were doing.
18 It's manual scanning, which means there is 19 a human actually moving the transducer over the pipes.
20 Part of the story I think, the conjecture is that may 21 have been a contributing factor in the sense that 22 there's a dose implication to putting a human there to 23 do this examination. And one of the modifications 24 that was made in the later designs was to increase 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 certain piping run lengths.
1 Putting the person farther away from the 2
exposure, but also changing the configuration of the 3
pipe, so again, there's a trade-off. This ultrasonic 4
procedure, as I said, was designed for thermal fatigue 5
cracks, not really optimized to detect, nor to size, 6
or as we would say, characterize SCC cracks. But when 7
they did start to find some indications, they 8
confirmed the SCC cracks with destructive examination.
9 MEMBER HALNON: Carol, does that mean that 10 the SCC cracks had to get big enough beyond what 11 normally you would expect in order to see them? Or is 12 it just something that the inspector needs to be cued 13 into?
14 MS. MOYER: Yeah, SCC cracks tend to be 15 slow growing, and they tend to be very tight. And 16 sometimes they have adherent oxide in them that's 17 still somewhat conductive, and so it's not an easy 18 kind of flaw to find in the first place. But yes, 19 they do tend to be small, and tight, and a sure wave 20 that's just looking for a big sound bounce may look 21 past it.
22 MEMBER HALNON: They had to get bigger 23 than what we -- if we had a probe designed, or 24 optimized to detect SCC cracks, it would have caught 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 them a lot earlier?
1 MS. MOYER: Conceivably.
2 MEMBER HALNON: I guess that's the 3
inference, okay.
4 MS. MOYER: The other interesting thing 5
is, as I showed --
6 MEMBER DIMITRIJEVIC: I'm sorry, I have 7
another -- how does the extracting exams, you know, 8
which piping they have been performed on? I mean what 9
does --
10 MS. MOYER: I'm sorry, I don't understand.
11 MR. MORLEY: They've cut the pieces out of 12 the plant, and destructively examined those.
13 MEMBER DIMITRIJEVIC: Cut the pieces from 14 some retired plant?
15 MS. MOYER: No, the pipes that were found 16 to be cracked were cut out, and taken to a laboratory, 17 and replaced with new pipe sections.
18 MEMBER DIMITRIJEVIC: I see, so they have 19 to drain RCS?
20 MS. MOYER: Yes, this was done during 21 their ten year inspection.
22 MEMBER DIMITRIJEVIC: Yeah, but inspection 23 is done with the water in the piping, and the stress 24 exam, they don't have installation valves, so 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
24 have to drain the RCS to do this, right?
1 MS. MOYER: That's my understanding, I 2
don't know for sure how extensive that was, but yes, 3
that was part of the problem, and why it has taken a 4
long time for them to recover from this experience.
5 MEMBER DIMITRIJEVIC: I mean I -- so, 6
that's something I didn't see there, replacing the 7
class one piping. So, I mean okay.
8 MR. HOSLER: Yeah, this is Ryan Hosler, 9
they would have to drain down to the level of the cold 10 leg, and hot leg, clearly to get this done. But that 11 would remain above the majority of the reactor vessel.
12 MS. MOYER: Right, the injection point for 13 these lines is the cold leg.
14 MEMBER MARCH-LEUBA: Can you show us slide 15 four? Very clear.
16 MS. MOYER: Okay.
17 MEMBER MARCH-LEUBA: You can see the level 18 right there, right? It's the copper pipes, which are 19 not made of copper, but copper colored pipes in there.
20 MEMBER KIRCHNER: Yeah, you can't drain 21 the whole primary system with the core still in it, so 22 either they have to take the entire core out to drain 23 the primary system, before they drain the primary 24 system. Or, they take suitable precautions to -- if 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 the fuel is still in the core, to ensure that the 1
system doesn't drain out. Can you give us more 2
information on how they actually did the repairs?
3 MS. MOYER: I cannot, I have just been 4
looking at the materials implications of it, but I 5
think we'll get to that later in the meeting.
6 MEMBER KIRCHNER: Okay, thank you, yeah.
7 MEMBER HALNON: This is Greg, the industry 8
has cold leg plugs they stick in the cold legs, and 9
you go to mid loop operations, and that's how they can 10 drain the pipes.
11 MS. MOYER: So, all of the pipes that are 12 shown in this diagram, this cartoon, are here at the 13 hot leg, and cold leg level. So, all of the welds of 14 interest are in these loops, and in these pipes that 15 feed them. So, the vessel itself would not have to be 16 drained any lower than the hot, and cold leg level.
17 MEMBER HALNON: Correct.
18 MEMBER DIMITRIJEVIC: They go to mid loop 19 operation, my guess would be they had the fuel 20 outside, so.
21 MEMBER HALNON: Yeah, they had to defuel 22 the reactor, I mean you can't do this with fuel in it, 23 the dose rates would be too high anyway to do that.
24 The experience of this, cutting out the hot leg, was 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 paramount in figuring out how to do this probably.
1 MR. RUDLAND: Before Carol moves on, I'm 2
getting live emails from my French friends, and I just 3
wanted to point out that the elbows are forged, 4
they're not cast.
5 CHAIR BALLINGER: And now you said it, 6
okay.
7 MR. RUDLAND: Yeah, the elbows are forged.
8 And also there was a comment about the 900 megawatts, 9
there was a defect found in one of the reactors in the 10 RHR system, but it was due to lack of fusion, and not 11 due to SCC. So, there were no SCC cracks found in the 12 900 megawatts.
13 MR. MORLEY: I think it was emanating from 14 a weld defect in the 900.
15 MEMBER REMPE: We need to have whoever 16 just spoke say their name for the transcript please.
17 MR. MORLEY: I'm sorry, Andy Morley here, 18 Rolls Royce.
19 CHAIR BALLINGER: Say again?
20 MS. MOYER: Andy Morley from Rolls Royce.
21 CHAIR BALLINGER: Okay.
22 MS. MOYER: So, as Dave said, looking more 23 closely at all of the NDE data that they had in 24 history also might find other things along the way, 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 like this weld defect in one of the 900 megawatt 1
plants, and other non-relevant indications. So, that 2
was one of the things that was a concern too, was 3
whether it was possible that some other indications 4
from prior NDE examinations had been improperly 5
characterized, and were in fact cracked.
6 So, they looked again at some older data.
7 I don't have a breakdown of how those results came 8
out, I just know that these are complicated joints.
9 There's geometry, there's counterbore, there's the 10 weld tow, the welds were not crowned, so there's a 11 weld crown, which means it's difficult to inspect 12 right up close to the weld, because the transducer 13 just doesn't physically fit flat against the pipe 14 because of the crown on the weld.
15 So, some of these challenges may have 16 impacted the inspectability of some of the pipes. And 17 so, it was worth looking again just to make sure 18 nothing was overlooked. And then EDF developed a plan 19 for an accelerated schedule to inspect all of the 20 analogous piping. Okay, as I mentioned we were told 21 that there were inspections by remote penetrant test, 22 or ultrasonics, then some by destructive tests.
23 I've been informed that the new, and 24 improved, if you will, NDE procedure that was rolled 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 out is something called total focusing method full 1
matrix capture, and plain wave imaging. Total 2
focusing method plain wave imaging. So, this is still 3
a manual technique, but it's encoded, so you keep 4
track of the location, and there's a bracket that goes 5
around the pipe, and encodes the location information.
6 It's a very sensitive technique, it is a 7
developing technique. It is not fully qualified, at 8
least not in the U.S., and to my understanding, it's 9
not qualified in France either, by their standards 10 organizations. It's an emerging technique, and again, 11 as I said, very sensitive, and creates a great deal of 12 data. We among the NRC staff, when we were discussing 13 this, had some questions why a qualified phased array 14 ultrasonic technique was not selected.
15 Not to second guess anyone, but just in 16 the course of conversation, we said I wonder why they 17 went to such a sensitive technique, instead of a well 18 known phased array UT technique, but there we are.
19 Okay, so a little bit more about the root cause 20 analysis, as we've been saying, IGSCC was not 21 expected. It really didn't line up with international 22 operating experience, there was no SCC on the French 23 900 megawatt plant series after 30 years.
24 There was no contamination observed, like 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 chlorides, or something that didn't belong in the 1
water chemistry. However, there were some weld 2
repairs, and there were some unusual weld geometries 3
that were found. Again, I don't know a lot about 4
this, it is second hand somewhat, but EDF, I know did 5
a welding simulation to estimate hardening, and 6
residual stresses that were observed near the ID of 7
these pipes.
8 And found that there was an area of 9
limited depth on the ID that was subject to tensile 10 stress. But the stresses weren't compressive within 11 the weld, which does line up with the observation that 12 the cracks started in the heat affected zone, and then 13 grew to the weld, and stopped. That was confirmed by 14 their destructive examination.
15 MEMBER KIRCHNER: Carol, this is Walt 16 Kirchner again. Did the 900 series plants have those 17 cast elbows for those lines?
18 MS. MOYER: They apparently were forged 19 elbows, that was my bad information, sorry. Dave 20 Rudland has now confirmed those were forged stainless 21 elbows in both cases.
22 MEMBER KIRCHNER: Forged, not cast.
23 MS. MOYER: Right.
24 MEMBER KIRCHNER: Okay, thank you. But it 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 was the same material selection, same fabrication for 1
the 900s as the larger techniques?
2 MS. MOYER: Yes, as far as I know, it's 3
316LN, possibly with a nitrogen strengthening, so a 4
very common place material.
5 MEMBER KIRCHNER: Okay, thank you.
6 MS. MOYER: Okay. So, the regulator was 7
aware of some weld repairs. I think they knew which 8
welds had been repaired, or at least which pipes had 9
had weld repairs, but they weren't necessarily sure of 10 exactly where. Weld repairs may leave the pipes in 11 unusual, or unknown residual stress condition, that's 12 always a possibility. So, those welds were 13 especially, carefully reinspected.
14 Another thing that could be a contributor, 15 so it's worth checking out, is oxygenated water.
16 Chlorides certainly are very bad news for stress 17 corrosion cracking, but oxygen can be a contributor.
18 This is something that we learned from our colleagues 19 at IRSN, that some of the flows, like from core makeup 20 tanks, the boric acid tank, and then peroxides 21 injected at shutdown for various good reasons may also 22 bring along with them some oxygen that would 23 potentially throw off the chemistry in these lines.
24 Especially given that they are stagnant, 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 or mostly stagnant, so that's one other possible 1
concern. Thermal stratification is another concern 2
that could impose its own cyclic loading on the pipes.
3 As I mentioned, in the M4 type reactors, there were 4
some horizontal runs of pipe that were extended, had 5
a good justification for inspectability, but may have 6
contributed to having longer pipes that were stagnant, 7
and that had thermal stratification.
8 So, uneven temperature because of thermal 9
expansion means uneven stresses in those pipes. So, 10 I think IRSN has been doing some work to estimate just 11 how much of an effect that might have been. Okay, 12 mitigation plans include, as I mentioned, cracked 13 sections have been removed, and replaced with new.
14 From this fall, through 2025 EDF has a plan for a 15 complete examination program on all their operating 16 reactors on areas that might be affected by this 17 concern.
18 They're using this advanced UT procedure 19 that is optimized now for IGSCC detection. And I'm 20 told that they plan to seek qualification of that 21 technique, but anybody who's worked on standards knows 22 that's not a quick, or one step thing. That's usually 23 a protracted effort. The reactors that were 24 considered most sensitive, just based 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
32 population of flaws found so far are the N4 type, and 1
those have been stopped to carry out inspection, and 2
repairs where needed.
3 And looking forward, there is a plan for 4
a periodic reinspection program, with periodicity 5
based on the sensitivity of the NDE, the crack growth 6
rate of IGSCC, and mechanical, plastic fracture 7
mechanics analyses to try to stay ahead of it. So, at 8
present, and this at present is probably a week, or so 9
old, so it might not be quite accurate. There are 26 10 reactors down, 15 for stress corrosion problems, and 11 11 for other scheduled maintenance.
12 The repair work has been completed for six 13 reactors, and repairs on this problem are underway at 14 four.
15 MEMBER MARCH-LEUBA:
When you say 16 currently, you mean today, or you mean --
17 MS. MOYER: About a week, or so I think.
18 MEMBER MARCH-LEUBA: I mean it's not nine 19 months ago.
20 MS. MOYER: Right, this is -- my notes are 21 this is from World Nuclear News Daily on November the 22 4th.
23 MEMBER MARCH-LEUBA: At what speed are 24 they testing them, do they have more than one sensor, 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 one machine?
1 MS. MOYER: There are multiple machines, 2
but keep in mind, that you need operators, and you 3
need welders, and if there's an exposure component, 4
those folks have limits on how much they can do.
5 MEMBER MARCH-LEUBA: I'm kind of asking 6
what's your projection of when they'll be up, and 7
running?
8 MS. MOYER: I wouldn't even hazard a 9
guess.
10 MEMBER MARCH-LEUBA: Is it a matter of 11 weeks, a matter of years?
12 MS. MOYER: The whole inspection program 13 I was told was to be done by 2025. So, going all the 14 way through every plant ensuring that these lines have 15 been fully inspected, and repaired if needed may take 16 a couple of years. But when they can get to a point 17 where they can operate these plants, that's probably 18 a different answer, and I don't have enough 19 information to say.
20 MEMBER DIMITRIJEVIC: This is Vesna 21 Dimitrijevic, so, when you say the 15 for stress 22 corrosion, I know this is also outside information for 23 you too, so does that mean that they're doing testing, 24 or that means they've identified problems, and 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
34 have to do the pipe replacement? What does it mean 1
for stress corrosion problems, is it for testing, or 2
for replacement?
3 And what does -- for maintenance, is that 4
totally independent of IGSCC, or sorry, I have a 5
problem pronouncing this for some reason. Or is this 6
for the independent maintenance, or these are also 7
related to this? So, the maintenance, is that 8
independent of this problem, or it's for what?
9 MS. MOYER: It is my understanding that 15 10 of the plants had some indication of cracking in these 11 lines that was being further investigated, or 12 repaired. So, 15 of these were suspected, or 13 confirmed stress corrosion cracking, and 11 of the 14 plants were down for other maintenance, whether 15 scheduled, or unscheduled. Does that help?
16 MEMBER DIMITRIJEVIC: That gives me some 17 information. And my other question is also is your 18 understanding that the testing related to this will be 19 done in -- they will be scheduled on all these plants 20 in these couple years independent of the regular 21 service testing, it will be just for this, so they 22 will be shutting down all plants to do this testing?
23 MEMBER DIMITRIJEVIC: You were --
24 (Simultaneous speaking.)
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 MS. MOYER: I would have to ask --
1 MEMBER DIMITRIJEVIC: In order to perform 2
these tests, are they going to perform them on all the 3
plants, or are they just going to do them on the 4
plants where they have the most suspicion that this 5
corrosion cracking can be? And also I have a 6
question, do you have an idea what percentage of welds 7
they think is in the region which can be exposed to 8
that?
9 Because let's say that we have within 400, 10 600 welds in the cross run piping, what percentage of 11 those welds will need to be tested, do you have an 12 idea about that too?
13 MR. RUDLAND: This is Dave Rudland, if I 14 go back to your first question, I really need to 15 emphasize that this is still an ongoing, developing 16 issue in France. And discussions between EDF, and the 17 regulator are continuing to be ongoing. So, I hate 18 for us to speculate too much about their plans on what 19 they do, and when they plan to be up, and things like 20 that.
21 That really needs to be left to those 22 discussions between the EDF, and their regulators. In 23 terms of the number of welds, I know that them being 24 very conservative in the number of welds that they'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
36 looking at. I think Carol pointed out earlier that 1
they were doing a lot of penetrant testing on a lot of 2
the welds that had any types of suspicion at all, so 3
they're being conservative in how they're looking for 4
these.
5 And so, I think they're suspecting that 6
most of them that are in the same lines as the ones 7
where they've had cracks before are considered 8
susceptible enough that they want to do some kind of 9
non-destructive evaluation. In a lot of cases it is 10 this penetrant that they're planning to use.
11 MEMBER DIMITRIJEVIC: I'm really curious, 12 because that's a really terrible timing with this 13 energy crisis in Europe, they're shutting 50 percent, 14 or more of their plants, it's just really bad timing.
15 MS. MOYER: Certainly.
16 MEMBER MARCH-LEUBA: Yeah, I'm reluctant 17 to put unverified data on the record, but Google says 18 that France intends to restart 27 of the reactors by 19 December of this year, and they're still working on 20 five more, they expect to be restarting in February, 21 but of course that's Google.
22 MS. MOYER: Okay, thank you.
23 MR. MORLEY: Andy Morley, Rolls Royce 24 here, do we now, have they only looked at these branch 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 lines, the SIS, and the RHR system, or have they 1
looked at the main loops as well? And if they haven't 2
looked at the main loops, is there a reason why?
3 MEMBER REMPE: I'm sorry to interrupt, and 4
I'm chairman of the ACRS, and this is not a time when 5
the public can ask questions.
6 MR. MORLEY: I'm sorry, I didn't realize 7
that.
8 MEMBER REMPE: I realize that it's a 9
different country, and all that, and I don't mean to 10 be rude. But at the end of this meeting, I assume the 11 supplementary chairman will ask for public comment, 12 and if you have a comment, you can state it for the 13 record at this time, but that's all we can do.
14 MR. MORLEY: My apologies.
15 CHAIR BALLINGER: My apologies, I thought 16 that you were one of us in some respect, so that's the 17 reason why I said okay. And nobody corrected me until 18 now. Now we can keep going.
19 MS. MOYER: Okay. As a regulator, I found 20 it interesting that the utility, that EDF in this case 21 performed a stress analysis, and proposed a flaw 22 evaluation criteria for continued operation. That is 23 they essentially requested permission to restart a 24 reactor with known cracks saying well, they're very 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 small, and we know where they are, and we can show by 1
a stress analysis, and crack growth data, that we can 2
operate to the next outage, or something like that.
3 The regulator, with support from their 4
PSO, evaluated that proposal, and the level of 5
uncertainty associated with the data that went into 6
it, and decided that that was not prudent, and as a 7
result, that reactor was not given permission to come 8
back online with flaws in place. I found that 9
interesting.
10 CHAIR BALLINGER: I can say, I might add 11 that the rules in France are different than the rules 12 in the U.S. In the U.S. my guess is that it's allowed 13 by the code, they would have done a risk analysis, and 14 probably be able to operate, but in France, the rules 15 are much more final if you want to use that word.
16 MS. MOYER: Right, that's true.
17 MR. RUDLAND: To add to what Ron said, I 18 agree, I would suspect that we would probably prefer 19 it to be mitigated, and not an -- an SCC left in 20 service. By mitigated, a weld overlay, or something 21 like that.
22 MS. MOYER: Okay, so we talked about 23 extent of condition, we talked about root cause. One 24 of the other things that we would normally look at is 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 similar operating experience. What have we seen that 1
looked anything like this in any other reactors? A 2
similar appearing crack was found in the Japanese 3
plant, OE unit three pressurizer spray weld in August 4
of 2020.
5 That flaw was attributed to hardening from 6
high weld stress on the inside diameter surface.
7 Again, as Dave mentioned earlier, we've typically seen 8
this kind of cracking when you have a lot of grinding, 9
or a really unusual weld upset, a weld procedure that 10 was off normal in some way. And you would have 11 unusually high residual stresses. So, that was 12 essentially the conclusion in that case.
13 In the U.S., PWRs have observed stress 14 corrosion cracking in 316 only when there was cold 15 work, grinding contamination, very oxygenated water 16 conditions, something like that, that you can really 17 point to. Okay. So, in the U.S., we have the ASME 18 code, American Society of Mechanical Engineers Boiler 19 and Pressure Vessel Code. Section 11 covers in 20 service inspection.
21 That code is mandated by U.S. regulations, 22 by 10 CFR 5055A. Most plants use an NRC approved risk 23 informed in service inspection plan as an alternative 24 to section 11. So, they have to meet at least one 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
40 those. U.S. plants examined 10 to 15 percent of the 1
class one safety injection, and RHR piping under their 2
programs. So, they do inspect these class one pipes 3
looking for this kind of flaw.
4 They do apply a sampling, they don't 5
necessarily look at 100 percent of the welds every 6
outage, they cycle through them --
7 MEMBER MARCH-LEUBA: So, you have a 8
population, this number, 10 to 15 percent is every 9
year, or ever?
10 MS. MOYER: Every inspection interval, 11 which is ten years.
12 MEMBER HALNON: There is an escalation 13 aspect to it.
14 MS. MOYER: Certainly, if anything is 15 found, then you have to expand --
16 MEMBER HALNON: This 10 to 15 percent has 17 got to be clean.
18 MEMBER DIMITRIJEVIC: It is not certain, 19 they always inspect the same 10 percent, or 15 20 percent. There is no -- I mean they don't cover every 21 ten years, the same ten percent, more than that is 22 inspected.
23 MS. MOYER: I'm sorry, was there a 24 question? No, okay, I missed the beginning of 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
41 MEMBER DIMITRIJEVIC: My point was if 1
we're inspecting the same ten percent, and we have 2
seen here that higher power, the plants, we can be 3
missing this in United States, because we are not 4
looking for that, we are looking for thermal fatigue, 5
and inspecting ten percent of the valves. Or this is 6
something which will be lessons learned from this 7
French situation.
8 MEMBER HALNON:
But we
- also, the 9
experience here shows that a small leak will occur.
10 We won't get a catastrophic failure, so it's you will 11 leak before break.
12 MR. RUDLAND: And this is Dave Rudland, 13 let me also point out that the risk informed programs 14 are designed such that the ones that they are looking 15 at are the most susceptible locations, be it stress, 16 or geometry, or whatever it happens to be. So, they 17 are looking at the most susceptible.
18 MEMBER HALNON: One thing we do miss in 19 this is the weld records. I was at the D.C. summer 20 plant when the hot leg was found, and until you went 21
- back, and looked at the actual records of 22 construction, you didn't realize that you had 23 excessive grinding, and excessive rewelds on that root 24 pass. And this 10 to 15 percent, it may be most 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 susceptible from a system condition perspective, but 1
not necessarily the most susceptible from a rework 2
perspective.
3 So, I don't know how you factor that in, 4
but again, it's like the previous slide said, the most 5
cases in the U.S. it's excessive grinding, and 6
excessive rework, but that's not factored into the 7
section 11, is it?
8 CHAIR BALLINGER: I feel compelled.
9 MEMBER HALNON: Go for it Ron.
10 CHAIR BALLINGER: My personal opinion is 11 that stainless steel of this type is fundamentally 12 unstable in 300 degrees C water. So, the material has 13 to maintain that protective film. There are so many 14 variables involved, everybody says there's too much 15 grinding, there's too high stress, but this material 16 will crack if you violate that film. And so, anything 17 that violates that film will probably result in 18 cracking.
19 The reason the difference is between PWRs, 20 and BWRs is one operates as an oxygen over pressure, 21 the other one operates as a hydrogen over pressure, so 22 the potential is much lower, but they're still 23 susceptible. So, that's one of the reasons why we're 24 having this meeting, to kind of warn people that we 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 need -- 80 years is a long time. Thanks.
1 MEMBER MARCH-LEUBA: You came in on that, 2
what film are you talking about, is this put in on the 3
inside?
4 CHAIR BALLINGER: It's a Faustian bargain 5
that we play. You've got a chromium oxide based film 6
that's on the surface that isolates the material from 7
the environment. But the material is fundamentally 8
unstable, it wants to convert to an oxide. So, if you 9
breach that film, then the underlying material is 10 exposed. And if you do it in the right way, you 11 crack.
12 So, in that respect, plain carbon steel 13 would be better than 316 stainless steel because it 14 doesn't stress corrosion crack.
15 MEMBER MARCH-LEUBA: I don't know anything 16 about this, why I'm asking the question. Chromium 17 oxide is generated when you put water inside --
18 CHAIR BALLINGER: Yeah, it's a film.
19 MEMBER MARCH-LEUBA: It's in itself 20 something --
21 CHAIR BALLINGER: I'm being a little fast, 22 and loose with the chromium oxide. It's an oxide 23 based film that's tough, and adherent, and it is a 24 strong function of the chromium content, which is 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
44 reason why the higher nickel chrome, and higher chrome 1
alloys, 690 for example, nickel based, that has 30 2
percent chrome, that's much more stable. But this 3
material is right at the limit where you need to be 4
very careful about stress, and grinding, and all that 5
kind of stuff.
6 MEMBER MARCH-LEUBA: But isn't that film 7
self-generated, so grinding won't get rid of it, you 8
will generate a new one?
9 CHAIR BALLINGER: Yeah, it reforms very 10 quickly.
11 MEMBER MARCH-LEUBA: Anyway, you guys are 12 the experts.
13 CHAIR BALLINGER: But you can create 14 conditions in a crack which make the environment more 15 aggressive. And if the stress is five millimeters 16 deep, you're driving this crack through. So, again, 17 we just need to be careful.
18 MEMBER MARCH-LEUBA: And would the stress 19 be dependent, I mean if we have much more stress 20 inside the pipe than outside?
21 CHAIR BALLINGER: For a single V weld, 22 typically what happens is you have a tensile residual 23 stress on the ID, but it goes compressive right away, 24 which is what they see here.
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 MEMBER MARCH-LEUBA: So, the probability 1
of the crack growing more than a quarter of an inch is 2
small?
3 CHAIR BALLINGER: Low.
4 MEMBER MARCH-LEUBA: Low.
5 CHAIR BALLINGER: But again, the weld, 6
you've heard the expression, is it a Friday weld, or 7
is it a Monday weld? There are all kinds of things, 8
parameters that affect welding. And these are -- look 9
at the multi, these are like 20 pass welds. And the 10 code allows regrinding. In other words you do an 11 inspection part way through, and if you see a defect, 12 the code requires you to grind it out.
13 So, once you grind it out, you change the 14 whole residual stress pattern, and everything. So, 15 there are a lot of variables that can affect this, and 16 you just have to be mindful of that.
17 MEMBER MARCH-LEUBA: Would it be a good 18 application for artificial intelligence to review all 19 those records that we're talking about, and identify 20 which of the records?
21 MS. MOYER: Especially with a new NDE 22 technique that's going to generate boatloads of data, 23 an AI technique would be very --
24 MEMBER MARCH-LEUBA: We have a separate 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 subcommittee on artificial intelligence, and --
1 MS. MOYER: Machine learning is being 2
looked at it for it.
3 MR. RUDLAND: The repairs haven't been 4
documented, a lot of them haven't been well 5
documented. And again, like Ron mentioned, the 6
stresses are a function of the number of passes in the 7
thickness, but also if there's any repairs, and the 8
depth of those repairs on when this thing may go into 9
compression. You have a large inter diameter repair, 10 you can have even very high stresses through much of 11 the wall because of that repair, so.
12 CHAIR BALLINGER: But the N4 reactors were 13 the newer ones.
14 MR. RUDLAND: In reality the smaller the 15 pipe, the harder it is to do an internal repair, 16 right? So, for this size pipe, it's a lot less likely 17 than it is for some of the hot legs, where they've had 18 issues where they had to grind, and do welds.
19 MR. HOSLER: This is Ryan Hosler, just 20 wanted to make a quick comment. Inspection coverage 21 for these branch lines, the safety injection RHR 22 lines, so as was mentioned, the risk informed 23 inspection program looks at susceptibility, but also 24 consequence. And obviously the non-iceable portion, 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 which is what we're interested in, is going to have a 1
higher consequence, and is going to get a focus of the 2
inspections.
3 On top of that, as discussed previously, 4
thermal fatigue is a big part of a potential issue in 5
these non-iceable portions of the piping, and there's 6
an MRP-146 inspection program that also looks there.
7 So, there is focus at this particular branch 8
connection, so you don't have to necessarily -- the 9
specific condition isn't necessarily the entire class 10 one, it's just the non-iceable portion of the piping.
11 End of comment.
12 CHAIR BALLINGER: Just one, or two more.
13 MS. MOYER: Just one, or two slides more, 14 yes. I know we are on time. Okay, as Ryan just 15 mentioned, there is MRP-146 guidance document, I 16 guess, what do you call it, white paper? That 17 enhances owner's voluntary programs of inspection to 18 look specifically for cracking in these lines. That's 19 enough of that. Okay, class one pipes are examined 20 using multiple ultrasonic
- angles, or scanning 21 directions.
22 So, these are well inspected in U.S.
23 plants, in any of the plants there are class one 24 pipes. Personnel procedures, and equipment used on 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 piping welds must pass rigorous performance 1
demonstration testing under ASME code section 11 2
appendix 8. The examinations in the U.S. are 3
optimized for thermal fatigue flaws also, but are 4
capable of detecting stress corrosion cracking.
5 And the UT examinations have a current 6
ability to detect cracks 5 to 15 percent through wall 7
with a very good probability of detecting larger 8
cracks. Challenges, there are the same metallurgical 9
challenges of metal grain structure, geometric 10 features like counterbore, and weld crowns, and the 11 well geometry itself that complicate the inspection 12 procedure, and the interpretation of the inspection 13 data.
14 But NRC continues also to do research on 15 both SCC initiation, and growth, and NDE capabilities.
16 Okay, so ASM has concluded that stresses caused by 17 thermal stratification were one of the major 18 contributors to the root cause of this IGSCC. The 19 older French reactors appear to be less affected, 20 possibly due to design. So, we talked earlier about 21 this is a developing kind of degradation.
22 This is a time dependent degradation, so 23 one might deduce then that the older plants would be 24 more susceptible than newer plants, that's why this 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 was a bit puzzling at first. And so we think that 1
design had a bigger effect than age in this case.
2 After the inspections have been completed on all the 3
reactors, there's a periodic inspection program that 4
will be defined, and implemented across the French 5
fleet.
6 The U.S. fleet configured similarly to the 7
older French
- plants, continues its regular 8
inspections, no similar degradation has been observed 9
in the U.S. plants. So, due to the past U.S.
10 operational experience, the continued robust ASME 11 section 11 exams in these locations, and the follow up 12 by the U.S. industry, the staff concludes that there 13 is no immediate concern for a similar issue in the 14 U.S. fleet.
15 MEMBER HALNON: Is that taking into 16 consideration comparable position, and places on the 17 pipe? I know we did 10 to 15 percent for the NDE, but 18 was there a concerted effort to go look at this 19 specific line? I guess this is the high pressure 20 injection, or low pressure injection line, what line 21 is this comparable to the U.S. line, do you know Ron?
22 CHAIR BALLINGER: High pressure injection.
23 MEMBER HALNON: So, we had the high 24 pressure injection.
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 (Simultaneous speaking) 1 MEMBER HALNON: So, we have the high 2
pressure injection nozzle issues, and some valve 3
issues, but have we specifically checked to make sure 4
our records are giving us what you said, that there's 5
been nothing found at the U.S. plants? Is that 6
because we haven't seen anything in our general 7
inspections, or is that because we've actually gone, 8
and looked at these pipes?
9 MS. MOYER: We have not changed our 10 inspection plans for U.S. plants in response to this 11 observation. We are taking it into account as we do 12 all operating experience, but we have not changed 13 anything in the near term. I think we have another 14 presentation that I would be stepping on if I injected 15 about --
16 MEMBER HALNON: Okay, we can wait, thanks.
17 MEMBER DIMITRIJEVIC: This is Vesna 18 Dimitrijevic, I work on the development of EPRI for my 19 side, and we apply this to over 30 plants in the 20 United States, so I have some experience with that, 21 and I can really give you some previous ASME section 22 11, which is using France's inspect 25 percent every 23 10 years in class one, and that 25 percent is the same 24 25 percent every 10 years.
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 In United States, with the risk informed 1
ISI, which is now applied almost in every plant, I 2
think only maybe a couple don't have a risk informed 3
ISI, actually that was changed at this location, 25 4
percent were selected based on stress, temperatures, 5
changes, and things like that. So, in the risk 6
informed ISI what happens is the valves were spreading 7
risk, high risk, medium risk, and low risk.
8 High risk corresponds to high 9
consequences, and high degradation mechanism, and only 10 one high probability of the pipe fail to degradation 11 mechanism, and only flow oscillated corrosion is one 12 of those mechanisms.
All other degradation 13 mechanisms, including IGSCC, and the thermal fatigue 14 are considered medium failure probability degradation 15 mechanism.
16 But when combined with high consequences, 17 like once when somebody just said, the pipe cannot be 18 insulated, they will come in the medium category, 19 where they inspect ten percent. Those ten percent are 20 selected based on presence of degradation mechanisms.
21 Based on all this ISI data up to that moment, only 22 degradation mechanism identified in about 20000 23 inspections in the class one was thermal fatigue.
24 And therefore those ten percent are put in 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 those thermal fatigue locations, and that's always the 1
same ten percent. However, the risk informed ISI 2
program has, if ever a different degradation mechanism 3
has been discovered, then the program has to be 4
reevaluated. So, the question is can the current 5
inspection (audio interference) thermal fatigue, also 6
identify IGSCC, I don't know.
7 But I bet if this information comes from 8
some outside programs, we'll also require plants to 9
look in their risk informed ISI programs, that would 10 be my understanding of that. There is also ASME court 11 cases covering, I think one just covering class one is 12 N760, but I'm not sure, I forgot those numbers. So, 13 covering risk informed inspections.
14 CHAIR BALLINGER: Okay, well thank you.
15 Other questions from members? Okay, we have -- we're 16 way ahead of scheduled, we're scheduled for a break at 17 2:40, but I think we can probably just pick it up, and 18 keep going. And so if there aren't any other 19 questions, can we just pick it up? Your name tag has 20 been changed twice already. You were an EPRI guy, and 21 now you're a Framatome guy.
22 MR. HOSLER: Yeah, somebody is moving me 23 around.
24 CHAIR BALLINGER: I don't know what'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
53 going on here.
1 MR. HOSLER: I'm Framatome. So, yeah, 2
someone is going to have to drive the presentation for 3
me, I wasn't able to get my computer hooked up to 4
Teams. So, all right, while he's getting it up, I'll 5
introduce myself. I'm Ryan Hosler, I'm the materials, 6
and fracture mechanics supervisor at Framatome, that's 7
the U.S. part of Framatome, but still part of the 8
mother company.
9 I'm also the technical lead of the focus 10 group that is addressing this issue, or addressing the 11 potential impact of this issue in the United States.
12 That focus group is developed by the PWR Owners Group 13 in collaboration with EPRI. This focus group is the 14 auxiliary piping stress corrosion cracking operating 15 experience, and -- there we go, coming up here.
16 So, this focus group includes experts from 17 the industry, utility members, also the vendors, and 18 also EPRI as well, including materials experts, 19 welding experts, NDE experts. A quorum to come 20 together to try to see how this may impact the United 21 States. Actually the focus group is also looking at 22 Owners Group members outside of the United States, but 23 for this presentation we'll focus on the U.S.
24 One thing I want to briefly cover 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
54 not in my slides, I probably should have added, is 1
just why is this stress corrosion cracking atypical, 2
why are we sitting here? So, historically, stress 3
corrosion cracking in PWRs has been found in isolated 4
locations beyond the first isolation valve, or clearly 5
stagnant conditions, or up in the control audit drive 6
mechanisms.
7 Where especially for certain designs, with 8
certain plants have a very large volume. So, when you 9
start up you get a lot of oxygen up there during start 10 up. So, you have those off chemistry conditions, 11 you'd see stress corrosion cracking there. Also for 12 cases of stress corrosion cases in PWRs, and well 13 controlled water chemistry, it required heavy cold 14 work. So, imagine pressurizer heaters that were 15 assuaged, and not stress relieved afterwards.
16 We saw stress corrosion cracking in those, 17 and some chemical volume control system heat exchanger 18 tubes, thin wall tubes that were bent as part of the 19 design. As high cold work, and cracking in those.
20 Now, and so for the non-iceable portions of the branch 21 connections, we haven't really seen stress corrosion 22 cracking prior to this EDF OE. So, that's -- so, 23 looking at this EDF OE, the first factor is it seems 24 like that the weld stress is a primary contributor 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
55 initiation.
1 Meaning there's no evidence cold work, the 2
destructive exams haven't shown any slip planes, or 3
you wouldn't be able to grind the inside of these 4
pipes anyway, it would not be normal practice to grind 5
the inside, and the destructive exams, as mentioned 6
earlier, shows that the weld crown is there, there is 7
no -- at the root, I'm sorry, there is no grinding 8
afterwards.
9 So, and also this is 316L, so you wouldn't 10 expect sensitization, we haven't really seen any 11 evidence of sensitization, and in destructive exams 12 we've seen indications close to the flowing line, and 13 also far from the flowing line. So, water chemistry 14 does not seem to be a big factor in my opinion, from 15 what I've seen. It seems like exceptionally high 16 stress at the root is what's the primary driver.
17 And that's what makes this unique, is we 18 have not seen that previously, typically it requires 19 high cold work for this to happen. So, that's one 20 piece, the other piece is the flaws have a large flaw 21 aspect ratio. Meaning they're very shallow, and can 22 be very long, which is again not typical of what we 23 see with stress corrosion cracking. And that's, in my 24 view, primarily driven by, as you discussed earlier, 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 the stress profile you typically see for weld residual 1
stress.
2 A high ID weld residual stress, and then 3
quickly becoming compressive. So, I just wanted to 4
start off, the Japanese OE that was mentioned, one 5
flaw has been found there, that appears very similar, 6
has all those same characteristics. So, those being 7
the cases that we're aware of, that have this atypical 8
stress corrosion cracking. I just want to kind of 9
give that as a lead in.
10 So, that comes to the agenda here, which 11 is what are the industry actions to consider this 12 operating experience, and how it may impact the U.S.
13 fleet? So, first, as I said, it's a collaboration 14 between the Owners Group, and EPRI. So, EPRI 15 completed a white paper, which I'll talk a little bit 16 about. We're in the process of revising MRP-236, 17 revision one, which is among other things, a database 18 of the PWR stress corrosion cracking operating 19 experience.
20 So, that was last revised in 2017, so 21 we're adding new operating experience to it to 22 understand the new trends, and what they may mean.
23 And then lastly I'll talk about the effort in the 24 focus group itself, and the two main efforts are 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
57 safety assessment, and applicability assessment. Next 1
slide please. Okay, so for the white paper, the MRP 2
letter, and number is listed there if you'd like to 3
take a look at it.
4 CHAIR BALLINGER: We have it.
5 MR. HOSLER: So, it was developed by 6
experts in the field of stress corrosion cracking, 7
Peter, and Jason, Peter Scott I believe. So, here are 8
a few of the inclusions from the paper. The first is 9
that the most important factors that accelerate stress 10 corrosion cracking to higher levels than expected in 11 PWR primary water are residual deformation from cold 12 work, or welding stress.
13 And then stress from
- welding, and 14 pressurization fit up, etcetera, and environment, 15 oxidants, and contaminates in creviced areas. Also 16 there's a good deal of discussion of a newly developed 17 crack growth rate for stress corrosion cracking that's 18 discussed, MRP-458, which is also available for 19 review. And also last point being stress corrosion 20 cracking, and stand still components exposed to 21 flowing PWR primary water will continue to occur.
22 But there is no evidence of aging that 23 accelerates stress corrosion cracking in raw stainless 24 steels, and no sudden increase in stress corrosion 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 cracking initiation growth is expected after decades 1
of operation. I'm happy to take questions on this 2
paper, but I didn't write it, so I can only say so 3
much.
4 CHAIR BALLINGER: That statement is almost 5
self-contradictory. I mean you take up before the 6
comma, and keep everything after that, everything is 7
fine. If you take everything out after the comma, the 8
meaning changes.
9 MR. HOSLER: So, you know, I know Peter, 10 so I'll try to interpret what I believe it's saying.
11 What I believe it's saying is stress corrosion 12 cracking is going to happen. As you said earlier, 13 it's hot water, and stainless steel, if you give 14 enough time, it'll happen. If the three components, 15 if they're minor enough, then it might not happen in 16 the life of the plant.
17 But if one, or two of the components are 18 severe enough, then it'll happen at some point. I 19 think he's trying to hit that point, but he's saying 20 there isn't a late blooming phase, or something, 21 there's no sudden increase expected, I think is what 22 he's saying.
23 CHAIR BALLINGER: Got it.
24 MR. WAX: Yeah Ryan, that point you made 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 at the very end is really what that statement's trying 1
to imply there. And I'm Chris Wax from EPRI. That 2
statement is saying there's no significant increase in 3
propensity for stress corrosion cracking as you age 4
these plants, it's there from day one, it'll be there 5
in day infinity.
6 CHAIR BALLINGER: I rest my case.
7 MR. HOSLER: All right, next slide please.
8 Okay, so this is going to discuss a little bit in the 9
next couple slides, the preliminary results of the 10 provisioned MRP-236. So, as I said earlier, this 11 report contains information concerning stress 12 corrosion cracking of primary circuit pressure 13 boundary stainless steel, including an operating 14 experience database.
15 The last revision was completed in 2017, 16 which found no cases of stress corrosion cracking in 17 the non-iceable portions of the branch piping. The 18 new revision, which is in progress, is reviewing the 19 operating experience since 2017. The only confirmed 20 cases of stress corrosion cracking in the non-iceable 21 portions of the branch piping have occurred in the EDF 22 fleet, and the one case at the Japanese unit.
23 Also, we've also been reviewing the 24 thermal fatigue OE database, and that's MRP-85, 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
60 MRP-468. And the reason for this is that in some 1
cases you'll be looking for thermal fatigue, you'll 2
inspect, you'll find the flaw, and then perform an 3
overlay, and you'll never actually do a destructive 4
exam, so you can't know with 100 percent certainty, 5
it's just because you were looking for thermal fatigue 6
that it was in fact stress corrosion cracking.
7 Like with EDF, they were looking for 8
thermal fatigue, and then they ended up finding stress 9
corrosion cracking. So, we reviewed that database for 10 the purpose of seeing if we could find any potential 11 cases of this atypical stress corrosion cracking that 12 was believed to be thermal fatigue. And the way we 13 went about doing that was first looking at the OE 14 database, there's about 34 cases over the last several 15 decades, and look for cases where they did not perform 16 a destructive exam.
17 And in cases where there was a flaw with 18 a large aspect ratio, meaning shallow, and quite long, 19 similar to what we (audio interference) events in this 20 category. And so, we went through our database, and 21 identified 2 of the 34 cases where it met this 22 criteria. Does that mean stress corrosion cracking?
23 Can't say, because weld overlays were performed, but 24 it does meet the criteria.
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 So, those will be summarized in the 1
revision to this report. Next slide please. So, this 2
is again, preliminary results, I'll try to walk 3
through this figure here. So, this figure is showing 4
at the bottom, X axis is number of events, stress 5
corrosion cracking events for PWR stress corrosion 6
cracking of stainless steel. And on the left side is 7
the stress driver. Whether it be weld residual 8
stress, cold work, cold work, and residual stress.
9 Just operating stresses, or depending on 10 operating stresses, and so the blue represents 11 stagnant conditions, and the red represents well 12 controlled conditions. And I'll explain that in a 13 little bit. And saying it might be better to look at 14 it as the blue represents cases where we believe 15 aggressive environment was a significant factor in the 16 stress corrosion cracking.
17 And the red cases represent events where 18 we believe the environment was not a primary 19 contributing factor in the stress corrosion cracking.
20 Okay, so that being said, as I mentioned earlier, so 21 the top right there, you see the weld residual stress, 22 the big blue block there that is being driven by 23 residual stress. Again, CRD, and CED housings, and 24 fuels, and valve drain lines in iceable piping is 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 where we see the most issues.
1 And for cold work, in the cold work case, 2
we do see there's some stagnant conditions, but the 3
well controlled conditions, we also would see driven 4
by cold work, I mentioned the pressurizer heaters, and 5
also the vent heat exchanger tubes. What's new is 6
that now there's well controlled water chemistry 7
events that are driven by residual stress, that top 8
right.
9 And those are the new events, that's the 10 EDF OE, and the safety injection RHR piping, and also 11 the one case in pressurizer spray piping at the 12 Japanese plant. So, that's kind of a visual look at 13 the outlier here, and why we're focusing on this, and 14 how it may impact the U.S. fleet. All right, I think 15 I hit that there. Next slide. Okay, so now I'm going 16 to go over where we are currently with the PWR EPRI 17 focus group addressing this issue.
18 So, first, I'll discuss the safety 19 assessment, the purpose of the assessment is to assess 20 the potential safety impact of this operating 21 experience in the industry, and then the applicability 22 assessment is to assess the applicability of this OE 23 to the industry. Next slide. The safety assessment's 24 in progress, so this is all preliminary, but 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
63 approach essentially, to determine safety impact is 1
based on risk.
2 And risk is a function of likelihood, and 3
consequence, so what is the likelihood of this type of 4
stress corrosion cracking being present in the United 5
States? And we're going to determine that by 6
reviewing inspection data to determine if applicable 7
locations are being inspected. Review the UT method 8
to determine if these type of flaws would be 9
identified.
10 And then also review the operating 11 experience database, which I have been discussing, to 12 determine if this type of stress corrosion cracking is 13 occurring outside of EDF. Then we'll also look at the 14 consequences of this type of stress corrosion 15 cracking. So, we'll review the available flaw 16 evaluations to determine if a flaw could reach a 17 critical flaw size, and then also compare design basis 18 analysis breaks to branch line breaks.
19 And then after all that's done, if 20 appropriate, we'll issue recommendations. Next slide.
21 All right, so this picture on the right here is an 22 example of a branch line within EDF stress corrosion 23 cracking -- I'm sorry, EDF safety injection piping 24 from an N4 design plant. And the reason I'm showing 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 this is in order to determine whether we're inspecting 1
the right locations, we have to define what is the 2
area of interest?
3 And so, looking at the EDF OE, all of it 4
is in the safety injection RHR branch lines, non-5 iceable portions, and elbow welds, or heat affected 6
zone adjacent to those elbow welds. And there was 7
discussion of this earlier, but all these are large 8
diameter pipes, 8 to 14 inches. The safety injection 9
is passive, and the RHR is the big RHR suction, 10 sometimes it's 14 inch diameter pipe.
11 And so, that's the main area of focus for 12 the OE review, or I should say the inspection review 13 I will cover in a moment. Next slide. Okay, so the 14 preliminary results for the likelihood of stress 15 corrosion cracking being present in the fleet. So, 16 I've reviewed the inspection results for 56 units, and 17 focusing on each unit's last inspection that occurred 18 in the last ten year period.
19 And so going through each of these, the 20 passive safety injection piping, large diameter, it 21 was about 130 welds inspected with no reportable 22 indications. Have the other SI piping, which is high 23 pressure, smaller diameter pipe, could be one, and a 24 half to maybe six inches in diameter. About 250 welds 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 have been inspected with no reportable indications.
1 We'll go to the double asterisk there.
2 And I'll note that historically some 3
indications have been identified, and attributed to 4
thermal fatigue. But again, these results are the 5
last time they inspected, and the thermal fatigue they 6
observed was prior to that, and some repair 7
replacement was performed.
Then
- lastly, the 8
pressurizer spray -- I'm sorry, the RHR piping, large 9
diameter, we looked at 180 welds that were inspected 10 with no reportable indications.
11 And also the pressurizer spray piping with 12 60 welds inspected with no reportable indications.
13 So, there's still 17 units in the U.S. I have not 14 reviewed their data yet, we're still gathering. It's 15 a big effort to gather all that information, to go 16 through it. So, we're looking, and we're not seeing 17 it, at least not in the last ten year period.
18 And we looked at the OE database, and 19 we're not seeing -- before the last ten year period, 20 the OE database goes up to 2017, we're not seeing 21 stress corrosion cracking in these locations. We look 22 to the thermal fatigue database, there might be two 23 units that meets the criteria, but we're not sure, 24 because there's an overlay. So, then the question 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 becomes well, are we using the appropriate UT method 1
to see it?
2 So, talking with ND experts, UT experts at 3
EPRI, and the vendors in the industry, the conclusion 4
was that the UT methods employed were appropriate.
5 Meaning that while the techniques used may not be 6
specifically designed for IGSCC protection, they do 7
provide reasonable assurance that if a significant 8
cracking was present, it would be detected. So, there 9
are new UT methods, and qualifications for personnel, 10 and methodologies specific for identifying IGSCC.
11 They've been used for years in BWRs, and 12 so we are considering those methods here potentially 13 for the future. At least to some, maybe a one time, 14 or maybe something else. It's part of what we're 15 considering as far as the safety assessment.
16 MEMBER HALNON: So, what if we missed it, 17 what happens?
18 MR. HOSLER: That gets to the consequence, 19 which is the next slide here.
20 MEMBER HALNON: Okay, I knew that.
21 MR. HOSLER: All right, good lead in. All 22 right, so for the consequence evaluation, this is a 23 preliminary result, it's in process. A branch break 24 line is bounded by a break, considered by design basis 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 analysis, that's no surprise. Design basis analysis 1
consider large break LOCA, and those lines are clearly 2
more limiting than a break in a branch line. So, 3
considering could simultaneous breaks happen in 4
multiple branch lines.
5 Given the paucity of this type of OE, and 6
the extent of inspection coverage, simultaneous breaks 7
in multiple branch lines is highly unlikely. We don't 8
have the statistics to put a number on highly 9
unlikely, but that's where we are right now.
10 MEMBER HALNON: Is a break in any line 11 likely? Or, I mean less unlikely? In other words if 12 this is left to crack, are we going to expect a 13 catastrophic break?
14 MR. HOSLER: What makes this -- I talked 15 about why this stress corrosion cracking is atypical, 16 and part of the reason is the large flaw aspect ratio.
17 Typically you would expect a flaw to be driven through 18 a wall, and then leak, and so worse case it would 19 leak, and you would identify it, because stainless 20 steel is highly ductile, and flaw tolerant.
21 In this case it's less clear, because as 22 EDF has observed, you can have a 360 degree flaw, that 23 so far they've only seen them go the duct to the root 24 pass wall, maybe 20 percent through a wall. EDF has 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 performed flaw evaluations under those criteria, and 1
concluded that faulted events would not cause that to 2
rupture. And as part of this focus group, we have not 3
yet done an analysis like that.
4 MEMBER HALNON: Okay. Are the cracks 5
arrested, or are they just growing slower, and slower 6
because of less stress?
7 MR. HOSLER: The residual stress profile 8
that is driving the flaw is highly tensile at the ID, 9
and then drops off very quickly to become compressive.
10 The stress intensity factor, the K driving the crack 11 won't go to zero, it will flatten out to --
12 MEMBER HALNON: It'll continue to crack, 13 but just maybe at a slower rate.
14 MR. HOSLER: Correct.
15 MEMBER HALNON: If it's going all the way 16 around, that slow rate doesn't give me any comfort.
17 MR. HOSLER: Right. And ideally, that is 18 correct.
19 CHAIR BALLINGER: I mean the good news is 20 that the maximum amount of unidentified leakage would 21 easily be detected before you got to a case before you 22 had a real problem. The bad news is that at no time 23 during the Davis-Besse event, did the leak rate ever 24 exceed the unidentified leakage limit.
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 MEMBER HALNON: And that's the same in 1
D.C. Summer, we never, it was a third of a gallon a 2
minute, which is less than the one -- we found it was 3
400 pounds of boric acid crystals hanging form the hot 4
leg. So, the boric acid is the key, keep boric acid 5
in your system so you can find the leaks.
6 CHAIR BALLINGER: That's part of my 7
argument about the tactile business of walking through 8
the plant.
9 MEMBER HALNON: That's right.
10 MR. RUDLAND: This is Dave Rudland, I have 11 one comment about the stability of these types of 12 flaws. And realize that we're talking about mainly 13 just the membrane type of stresses. If the type was 14 under just totally, membrane stress, you might get to 15 a condition where you might get a crack that continues 16 to grow evenly around the circumference even. But 17 under bending, which a lot of these pipes have bending 18 stresses on them.
19 Of course it's going to favor one side, or 20 the other to wherever the bending tensile stress is.
21 Which again would lead to a leakage behavior probably 22 before it ruptured.
23 MR. HOSLER: All right, next slide. Okay, 24 so applicability assessment. This is also in 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 progress, this is preliminary. But the approach being 1
applied is identify, or consider the root causes to 2
find for the EDF stress corrosion cracking operating 3
experience. So, the primary is elevated stresses in 4
the elbow weld region, and through a thermal 5
stratification, I should say that it's identified as 6
the primary by EDF.
7 And EDF has identified the secondary as 8
weld residual stress. The IRSN has suggested that 9
elevated dissolved oxygen due to dissolved oxygen in 10 makeup water may be a contributing factor. EDF does 11 not consider this to be one of the root causes, but it 12 has been brought up by the IRSN. I can briefly give 13 my thoughts on that. I agree with EDF on this.
14 Dissolved oxygen in the makeup water 15 would, if that was a contributing factor, you would 16 see stress corrosion cracking in the cold leg. Not 17 only that, we're seeing stress corrosion cracking at 18 the EDF branch lines near the flowing line, and down 19 the branch line far from the flowing line. So, there 20 doesn't seem to be -- oxygen doesn't seem to be a big 21 driver in my view.
22 CHAIR BALLINGER: The question has been 23 asked in another setting, with respect to the U.S.
24 fleet, on makeup water tanks, some of them are 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 inerted, some of them are not, is that true?
1 MR. HOSLER: That's my understanding.
2 CHAIR BALLINGER: And has somebody checked 3
to see which ones are, and which ones aren't? If 4
somebody claims that oxygen is an issue, would that be 5
a discriminating?
6 MR.
HOSLER:
If the applicability 7
assessment concludes that oxygen is one of the main 8
drivers, then that could be a way to focus inspections 9
potentially. But we're not going that direction 10 currently for the reasons we discussed. Oxygen can 11 certainly aggravate stress corrosion cracking 12 absolutely.
Especially when the material is 13 sensitized, and there's no evidence of sensitization.
14 And again, we would expect to see cracking 15 elsewhere if oxygen was really an issue. And beyond 16 that, there's also been IGSCC found in the RHR branch 17 line, which comes out of the hot leg, so clearly 18 whatever makeup water you had after it's through 19 before, it's gone. So, there's no oxygen you would 20 expect in the RHR line.
21 CHAIR BALLINGER: Got it.
22 MR. HOSLER: All right. So, considering 23 those root causes, assessed whether these conditions 24 are present in the industry so we can consider 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
72 potential for thermal stratification in industry 1
branch lines. This has been considered for quite a 2
few years for concerns with thermal fatigue. And so 3
MRP-146, which has NEI0308 guidelines for performing 4
inspections specifically in these elbow welds, looking 5
for thermal fatigue.
6 So, stratification is definitely something 7
that's been a focus historically in the U.S. For the 8
residual stress aspect, which as I said earlier is, in 9
my view, the main driver, partially because of how the 10 flaw is growing. I think as mentioned earlier, if the 11 thermal stresses, or stratification is going to be 12 preferential on the bottom of the pipe, so you would 13 imagine the flaw would preferentially grow, and go 14 through on that side.
15 But looking at the EDF OE I haven't seen 16 that. So, that being said, for the applicability 17 assessment, considering weld residual stress, it would 18 be very beneficial to be able to review the weld 19 procedures used, and weld records at the EDF units, 20 and compare that to the U.S. fleet. That is easier 21 said than done. You don't have access to the EDF 22 records.
23 And also in the U.S. it's highly varied in 24 how it was done, because it's done differently 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
73 over the industry. The branch lines weren't welded by 1
the OEM of the reactor, or the primary loop, or the 2
steam generators, they were all done by other 3
organizations. So, I can't go look at the Framatome, 4
or the BMW fabricated plants, and pull those records, 5
because the branch lines were done by other companies.
6 So, gathering those records is very 7
difficult, so making that comparison is still being 8
considered how feasible it is. Lastly, so we can --
9 as we discussed, review common practices for makeup 10 water control, and monitoring programs. If we do end 11 up feeling the oxygen is a primary contributor, that's 12 a path we're going to have to go down, but right now 13 we're not going that direction.
14 Next slide please. All right, so been 15 working on this for a little while, it's going a 16 little slower than I like for two reasons. One is 17 gathering all this inspection data for the fleet has 18 taken some time. And also as mentioned earlier, it's 19 very much an ongoing investigation in France, 20 identifying where there are flaws, determining the 21 extent of the condition.
22 And so, it's taken some time to gather up 23 that information, and only so much is available to the 24 U.S. And so at a point now where we've discussed in 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 the focus group, said okay, this is the information we 1
have now, or do we take what we have, and write our 2
safety assessment, and then revise it later if we get 3
more information later? So, that was what we've 4
decided to do.
5 So, now we're moving forward, going to put 6
out the safety assessment in January, and if 7
additional information comes out later, we'll just 8
revise it, and update it. And then applicability, 9
assessment sometime in the first quarter of next year.
10 I'll leave that as my last slide there.
11 CHAIR BALLINGER: Questions from the 12 members, or consultants? I apologize.
13 MEMBER PETTI: I just had a question given 14 these French units, are there any designs similar to 15 the French units in terms of the way the piping runs?
16 MR. HOSLER: So, the N4 design that seems 17 to be most affected, the EDF type design, there are 18 passive safety injection lines, large diameter safety 19 injection lines, which for the vast majority of their 20 flaws, are in those lines are in kind of a downward 21 horizontal configuration, which is not true with the 22 U.S. fleet. The U.S. fleet, pretty much all the 23 passive safety injections are either up horizontal, or 24 horizontal.
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 CHAIR BALLINGER: Thanks.
1 MEMBER BROWN: Is that good, or bad?
2 MR. HOSLER: If thermal stratification, as 3
identified by OE, is a primary driver, then that would 4
mean that the safety injection lines in the U.S. fleet 5
are not affected by the issue.
6 MEMBER BROWN: Less susceptible.
7 MR. HOSLER: Correct.
8 CHAIR BALLINGER: So, then you tumbled to 9
the idea that subtracting all that out, all that's 10 left is residual stress, weld residual stress, and if 11 that's the case, what's the difference between welding 12 procedures in the U.S. versus France?
13 MR. HOSLER: Besides the operating 14 experience -- besides seeing the types of flaws being 15 observed at EDF, and not seeing those flaws in the 16 U.S., that is the indirect comparison. The direct 17 comparison would be actually reviewing the weld 18 procedures, and weld records. And if those were made 19 available, then that would be useful.
20 CHAIR BALLINGER:
Because there's 21 obviously a difference, one cracked, and one didn't.
22 MR. HOSLER: I believe, and maybe this is 23 me talking, Ryan Hosler, and not representative of a 24 consensus, but I definitely see a correlation between 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 the branch line configuration, and the amount, the 1
extent of stress corrosion cracking in the EDF fleet.
2 Certainly the N4 design has the most indications, and 3
it has the most stratification. So, there's a clear 4
correlation there.
5 I'm not seeing the cause effect, but 6
everything I've seen, much of what I've seen is 7
similar to others, which is press reports, some 8
PowerPoint presentations, not a full report. And 9
maybe if I saw a full report, I'd say clearly, I see 10 how they got from A to B. But right now I don't see 11 how they got from A to B, I see a correlation, but not 12 a cause effect right now.
13 MEMBER BROWN: Is there -- I was looking 14 at the age of the plants. Our plants are older than 15 theirs by not an insubstantial decade, or so.
16 MR. HOSLER: Yeah, the U.S. fleet is 17 certainly older than the N4 design branch point.
18 MEMBER BROWN: Yeah, based on the 19 statistics, you're talking 20 years, and these aren't 20 cracking, and those are.
21 MR. HOSLER: Correct.
22 MEMBER BROWN: And their materials are 23 still the same? Was it 316LN, or whatever the name 24 was?
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 MR. HOSLER: Yeah, the materials vary in 1
the U.S., but much of it is 316, sometimes 316L, for 2
the larger diameter pipes, 316 seems more common. For 3
the smaller diameter, 304 seems more common. But 4
they're all stainless steels that are generally 5
similar.
6 MEMBER BROWN: So, the U.S. plants have 7
had more stress applied to them over a period of time 8
you would think, just from normal operations.
9 MR. HOSLER: I wouldn't say -- the stress 10 is imparted during fabrication, the weld residual 11 stress. The thermal stratification --
12 MEMBER BROWN: I wasn't thinking about 13 thermal stress, but just due to operations, and flow 14 usage.
15 MR. HOSLER: Right. Yeah, I'd say the 16 time is a factor, and temperature is a factor, but all 17 other conditions being equal, higher temperature will 18 make it occur more quickly, but temperatures are 19 similar in both designs, U.S., and EDF. All other 20 components being equal, then yeah, it's just a matter 21 of time. So, clearly we're not seeing -- in the EDF 22 fleet, their older plants, they're only seeing it in 23 their newer plants.
24 MEMBER BROWN: The 900s.
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 MR. HOSLER: In the 900s they're not 1
seeing, they had one that was really associated with 2
a weld defect, doesn't seem to be part of this extent 3
of condition. The extended condition at EDF, they 4
defined it as the N4 plant safety injection RHR. And 5
then for the P Prime4, which is new, but not their 6
newest plants, they've seen stress corrosion cracking 7
at one unit.
8 So, they've included the safety injection 9
lines for those 12 units as well as part of their 10 inspection destructive examination program. But yeah, 11 generally the EDF, they've only seen it in their newer 12 plants, and not their older plants.
13 MEMBER DIMITRIJEVIC: Yeah, but those are 14 also coming up --
15 MEMBER BROWN: There's a benefit to being 16 old.
17 MEMBER DIMITRIJEVIC: Those are higher 18 power plants, over 1300 megawatts. So, I mean in 19 United States, if we look at this, it would have to be 20 in higher power plants. I don't think that aging is 21 maybe such a big part. It's your opinion, the aging 22 is an effect, or the power of the plant?
23 MR. HOSLER: I don't see a cause effect 24 between the power of the plant, and susceptibility 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
79 stress corrosion cracking. It's certainly an aging 1
effect question --
2 MEMBER DIMITRIJEVIC:
No, not the 3
progression, but the temperatures could affect thermal 4
transfer, so.
5 CHAIR BALLINGER: Nobody really knows when 6
initiation occurred, and these plants are 40, 30 7
years.
8 MR. HOSLER: For the EDF fleet, they 9
performed inspections ten years ago, and they didn't 10 see anything, they performed inspections this time, 11 and they did see something. Was it there ten years 12 ago? It's possible it may have been to a small 13 degree, and they just weren't able to detect it. But 14 to them, it appears as though -- I mean it's their 20 15 year life is when all these started popping up at the 16 N4 design.
17 Which, now U.S. plants are hitting their 18 50 years around this time, some of them are. And 19 they're not seeing it, which indicates that clearly 20 there is, in my view, a significant difference between 21 the weld practices that were used at those two 22 designs.
23 MR. RUDLAND: This is Dave Rudland, if I 24 can make an observation, the NRC, and EPRI a few years 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 back did a very extensive weld residual stress 1
validation program, where we made welds in plates in 2
small diameter pipes, and very large diameter pipes, 3
and looked at how these residual stresses are formed.
4 And something that struck me when I first saw this, 5
and looked at it was the micrographs that I saw 6
definitely showed the root pass of the weld was very 7
large compared to the fill passes.
8 And again, typically in these multi pass 9
welds, the secondary, the second, and third passes 10 kind of anneal the first pass a little bit, and reduce 11 the stresses slightly. In the cases that we've seen 12 the root passes are very large, indicating to me it's 13 going to have very high residual stress because of how 14 big that root pass is. And that evidence is further 15 brought forward by how deep the crack went.
16 Which, the crack went to about the depth 17 of the root pass, suggesting again that it's some weld 18 procedure, or some issue with the residual stresses 19 that come from that particular weld. And those were 20
-- granted, we've only seen a very small sampling of 21 the weld sections, but that was an observation that we 22 had from those weld sections.
23 MR. HOSLER: And that was true for the EDF 24 destructive exam results I've seen, and also 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
81 Japanese one as well.
1 MEMBER HALNON: I wanted to ask Carol, 2
given all this information, what are the plans for the 3
NRC? Is it leveled effort with research to the SCC, 4
and NDE research, or are you guys finding a generic 5
communication, and you're working up towards that?
6 What's the NRC engagement as they finish their safety 7
assessment?
8 MS.
MOYER:
This is an emerging 9
experience, and response. None of that has been 10 decided for sure yet. I will restate that we had 11 already research in place on SCC initiation, crack 12 growth rate, NDE techniques, and their sensitivity, 13 and their reliability, including some of these newer 14 techniques that EDF is employing now. So, we will 15 continue those works to better understand what tools 16 we have, and what susceptibility we may be looking 17 for.
18 I'm not aware of any plans for a generic 19 communication. I'm looking at Dave because I don't 20 know the answer to that one.
21 MR. RUDLAND: Yeah, so this is Dave 22 Rudland. Just coincidentally this morning, the NRC 23 management, and the industry management had a 24 materials meeting, and this topic came up of course.
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 And the status of the industry's focus, effort stuff 1
came up. We asked that when they finish their effort, 2
that we have another public meeting to kind of go over 3
their final results.
4 And then I think any action we take will 5
have to wait until after we hear their final results.
6 CHAIR BALLINGER: Yeah, that's good, 7
thanks.
8 MR. RUDLAND: We suspect, and correct me 9
if I'm wrong, but we suspect sometime this spring, 10 early summer, we would have that meeting.
11 MEMBER KIRCHNER: Ron, this is Walt, may 12 I ask a question?
13 CHAIR BALLINGER: Of course.
14 MEMBER KIRCHNER: Going back to the 15 earlier presentation with that nice picture of the 16 crack, so we have Carol, and David, and Ryan who are 17 metallurgists, I'm not. I look at that kind of crack, 18 and that doesn't look like stress corrosion cracking.
19 I shouldn't have said that. How would you describe 20 that crack from the picture, from the inner elbow 21 picture from EDF.
22 MR. HOSLER: So, this is Ryan Hosler. The 23 main indicators that it's stress corrosion cracking, 24 and it's hard to see in that picture I agree, but 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 there's some better pictures where it's closer up, and 1
it's etched, so you can see the grain boundaries, and 2
you can clearly see that the flaw is following the 3
grain boundaries. Also some other destructive exam 4
results did in fact break open some of the flaws, and 5
look at the fracture surface with an SEM.
6 And you can clearly see that the faceted 7
type intra granular stress corrosion cracking 8
indicative surface.
9 MR. RUDLAND: And this is Dave, one of the 10 first questions I asked was were these things 11 initiated by thermal fatigue, and then grown by SCC?
12 Transitioned, and grown by SCC, and from what I've 13 heard, and read through my counterparts in France, 14 that the SCC is across the entire crack face. So, 15 it's not something that was transitioned, it was 16 purely stress corrosion cracking.
17 MEMBER KIRCHNER: So, how do you reconcile 18 that with looking at root causes, and saying it's 19 residual weld stress?
20 MR.
HOSLER:
Residual stress is 21 historically primary driver for stress corrosion 22 cracking. For the BWRs, all the issues they had in 23 the 80s, they were all weld residual stress driven.
24 MEMBER KIRCHNER: Okay, 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
84 CHAIR BALLINGER: I might add, there's 1
kind of a rule of thumb, Vicker's hardness number for 2
initiation, the Vicker's hardness number, if I recall, 3
exceeded these numbers.
4 MR. HOSLER: They're very close to the ID, 5
the hardness is quite high.
6 MR. RUDLAND: And just to be clear, stress 7
corrosion cracking is a constant load mechanism crack.
8 Fatigue is an alternating load, on, off. For stress 9
corrosion cracking you need to have constant load, and 10 so that's why residual stress is so detrimental to 11 those that are susceptible to stress corrosion 12 cracking.
13 MEMBER KIRCHNER: Great, thank you, thank 14 you.
15 MEMBER BROWN: Earlier I asked a question, 16 not being a metallurgist, where we're going now, it 17 seems to me, the way I take away this is our ASME 18 standards, and everything else in the weld 19 requirements for Part 50, and Part 52, whatever we do 20 to build plants seems to have put us in good stead.
21 Is that a reasonable conclusion?
22 MR. HOSLER: I think what we've been doing 23 has worked very well. Stainless steel generally has 24 been -- stainless steel welds have been extremely well 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 performing.
1 MEMBER BROWN: Is anything changing with 2
the Part 53 with risk informed stuff, does that get 3
applied to welding? I know a wide variety of plants, 4
and light water reactors could use Part 53 if they 5
wanted to, I presume. Do we still have special --
6 MEMBER PETTI: It still pushes the section 7
level.
8 MEMBER BROWN: Okay, I didn't remember 9
that from our previous --
10 MEMBER PETTI: Yes, but again, in some 11 cases they're not going to use stainless steel, 12 they're going to use high end nickel alloy, it depends 13 on the reactor.
14 MEMBER BROWN: Yeah, stuff of that nature.
15 I just needed to make sure it was clear in my mind.
16 MEMBER PETTI: Yes, and the code, and such 17 is actively evolving for some of these more advanced 18 reactors, and the newer materials, and higher 19 temperatures, so they're in the process of evolving to 20 try to meet those needs.
21 MEMBER BROWN: Just worry about you don't 22 want to see the standards not meet the tests, we've 23 been building these things since the late 50s, and 24 that's a long time to have some of the very, very 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 early plants, which have been decommissioned already, 1
but it just seems to have put us in good stead, that's 2
why I asked the questions.
3 CHAIR BALLINGER: Stainless steel has 4
produced a number of bad hair days, and expenses, but 5
it hasn't resulted in anything really bad.
6 MS. MOYER: As we've been saying, it is a 7
resilient material, corrosion resistant, stainless, 8
but corrosion resistant anyway. And between the 9
design, and the inspection of section 11, leak before 10 break has been maintained, and yes. So, if there are 11 problems, typically they can be detected before they 12 become big problems.
13 MEMBER BROWN: My point is inspection 14 requirements, and things like that are costly, but 15 they have at least been able to ensure that we're 16 okay, which is the good news to me.
17 MS. MOYER: Fair enough.
18 MEMBER BROWN: I'm not criticizing anybody 19 else, it's not the point, it's just a matter of 20 whatever standards people have, and people complain 21 about, they seem to have worked. Thank you.
22 MR. SCHULTZ: Carol, this is Steve 23 Schultz. You mentioned that the inspection techniques 24 that have been newly developed in France had 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
87 advantages of more rapid testing, and also other 1
opportunities for testing, and you said that we were 2
going to be looking further with that. I'm not sure 3
if this is a question for you, or for the Owners 4
Group. Is there --
5 MS. MOYER: That's not quite an accurate 6
portrayal. I would say the techniques they're 7
employing are very sensitive, and they find they're 8
capable of detecting flaws, they're also capable of 9
detecting geometry, they're capable of detecting grain 10 structure, and weld micro structure. And the 11 difficulty with a sensitive technique that is capable 12 of detecting many things is it can be challenging in 13 the interpretation of those results.
14 So, while EDF has selected an inspection 15 technique that will give them a lot of information, 16 turning that into actionable information will be a 17 challenge for them. We are continuing to look at 18 those inspection techniques, as I'm sure a lot of 19 other people are. I mean there are systems available 20 that you can buy, then take to your power plant of 21 whatever sort, and employ that kind of an inspection.
22 But there are not ASME section 11 appendix 23 8 qualified procedures that say if you do this, you 24 will get that predictable result. You see what I 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 mean? We're in that somewhat sort of grey area with 1
these new techniques.
2 MR. SCHULTZ: I thought you mentioned that 3
there would be a
dose reduction off (audio 4
interference) technique, did I miss that?
5 MS. MOYER: That's not true.
6 MR. SCHULTZ: Okay, thank you.
7 MR. WAX: This is Chris Wax from EPRI.
8 From a inspection perspective, we do have the 9
performance demonstration institute within our EPRI 10 NDE Center of Excellence. So, we do require folks to 11 come, and actually qualify on these procedures to 12 prove that they can actually go find an indication in 13 a plant. They get their qualifications there, they 14 get their approval, they get the -- essentially the 15 thumbs up that they can actually go perform these.
16 And following the IGSCC experience from 17 the BWR plants, we have a pretty extensively well put 18 together IGSCC procedure that has been utilized in the 19 past to look for these types of indications. And like 20 Ryan mentioned in our presentation, that's one of the 21 things we're looking at for future inspection needs.
22 Do we need to go to that route, to that level, and 23 have this IGSCC procedure deployed on these 24 inspections?
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 that's something that will come out of 1
our findings, and our discussion from our focus group.
2 CHAIR BALLINGER: Thank you.
3 MEMBER DIMITRIJEVIC: Do you mean, okay, 4
are you talking outside of the current ISI program, 5
outside of these ten years, ten percent? You mean 6
additional inspection for IGSCC, is that your meaning?
7 MR. WAX: I mean we have the thermal 8
fatigue programs that we have required the thermal 9
fatigue procedures to be utilized on in the past, and 10 we're looking at some of those inspections, and the 11 periodicity that they are utilized on. And 12 potentially providing input that we propose the use of 13 the IGSCC procedures. Again, that's in conversation, 14 it's not formal yet, but that's just a thought we're 15 having.
16 MS. MOYER: And I would want to reiterate 17 that the newer techniques that I mentioned earlier, 18 I'm not going to say the letters, because I'll say it 19 wrong, but you can look back at the slide that 20 mentions these matrix type things. They are very 21 challenging to deploy in a real power plant at this 22 time. They require a transducer that is very large.
23 And so, a large transducer on a small diameter pipe 24 means you don't always get very good sound connection 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 between the transducer, and the pipe.
1 So, unless they can shrink the size of the 2
transducer with additional technology development, I 3
mean trying to get something that's three inches 4
square to meet against a pipe, and give you a 5
predictable sonification of that volume is just hard.
6 It also produces, as I said earlier, vast quantities 7
of data that will be an interpretation challenge. So, 8
this is why we've chosen not to go this direction just 9
yet.
10 CHAIR BALLINGER: Are there questions from 11 members, or consultants? Well, that's the end of your 12 presentation, right?
13 MR. HOSLER: Correct.
14 CHAIR BALLINGER: So, I'm sure the other 15 members would thank you very much for the 16 presentation, it's been very informative. We are 17 hopefully going to get more presentations as this 18 evolves. This was for information only, and I think 19 you provided us with as much as you can by way of 20 information that we can -- to give us a perspective on 21 the issue, which is really important for us.
22 So, if there aren't any -- I have to go 23 out -- excuse me. Now we can ask, are there any 24 members of the public that would like to make 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
91 comment? If there are members of the public out there 1
that would like to make a comment, please state your 2
name, and then make your comment. Hearing none, 3
again, I'd like to thank you guys very much for 4
coming, and talking with us. And this meeting is 5
adjourned.
6 (Whereupon, the above-entitled matter went 7
off the record at 3:05 p.m.)
8 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
Stress Corrosion Cracking in French PWRs:
Operating Experience Discussion Carol Moyer, Sr. Materials Engineer, NRR/DNRL/NVIB David Rudland, Sr. Technical Advisor, NRR/DNRL November 16, 2022
French Operating Fleet
- 56 reactors in operation, all Pressurized Water Reactors (PWRs)
- Built during 1970s - 1990s
- 3 main styles
- 32 are 900 MW (CP0 and CPY types)
- 20 are 1300 MW (P4 and P4 types)
- 4 are 1450+ MW (N4 type)
- Based on Westinghouse design, modified for French grid 2
Figure courtesy of ASN
Timeline of Cracking Detected in French NPPs in Fall 2021
- Flaw indications were detected near welds in safety injection (ECCS) lines during scheduled decennial safety inspections
- 10/2021 - Civaux-1 (N4)
- 11/2021 - Civaux-2 (N4) and Penly-1 (P4)
- 12/2021 - Chooz-B1 and Chooz-B2 (N4)
- Indications found in safety injection (SI) lines and residual heat removal (RHR) lines
- Laboratory analyses of Civaux-1 cracks identified intergranular stress corrosion cracking (IGSCC)
- Regulator (ASN) decided to expand inspections and reevaluate prior NDE data 3
Affected Piping Loops 4
- 4 separate loops, connected to the cold leg of the primary circuit
- Stainless steel piping Diameter: 25 to 30 cm Thickness: 2.85 cm
- Design modifications increased the length of piping runs in the safety injection circuits to reduce inspector dose.
Cracking Locations in Safety Injection Piping 5
Figures courtesy of ASN. https://french-nuclear-safety.fr/asn-informs/news-releases/stress-corrosion-phenomenon-detected-on-reactors Nominal pipe sizes are 8 - 14 diameter, 1 wall thickness
Additional Indications of SCC in Spring 2022
- EDF accelerated plans to inspect safety injection piping for similar degradation in Spring 2022
- The regulator (ASN) requested additional information from EDF to assess the degradation and its extent - whether a generic issue.
- Summer 2022 - EDF deployed new NDE method to detect & size flaws
- SCC indications were mostly in 1300 MW and 1450 MW type (newer) reactors, not in 900 MW (older) plants
- SCC indications have been reported at these reactors:
- Civaux-1 & -2 (1561 MW)
- Chooz-B1 & -B2 (1560 MW)
- Penly-1 (1382 MW)
- Cattenom-3 (1362 MW)
- Flamanville-2 (1382 MW) (unconfirmed)
- Golfech-1 (1363 MW) (unconfirmed) 6
Characterizing the Crack-like Indications
- Indications were at pipe inner diameter, located at elbows.
- Through destructive examination, the cracks were confirmed and attributed to intergranular stress corrosion cracking (IGSCC).
- In the base metal (AISI 316L) and heat affected zone (HAZ) adjacent to welds
- Crack depth varies
- Crack extension can reach 360° (the full circumference)
- Thermal fatigue cracks had been anticipated.
- Stagnant or intermittent flow is expected in affected lines.
- Thermal stratification in affected lines is postulated.
7
Destructive Examination
- Over 100 welds have been examined with either penetrant test or destructive examination.
- Cracked elbows were removed and sent to EDF hot laboratory for assessment. Cracks were confirmed, attributed to intergranular stress corrosion cracking (IGSCC).
- In the base metal (AISI 316L) and heat affected zone (HAZ)
- Max. depth confirmed ~1/4 through-wall depth
- High hardness detected in the vicinity of the weld root pass
- Unusual weld geometry (height of the weld root pass) or weld repairs detected in some cross-sections
- No evidence of chemical contamination 8
Destructive Examination of Civaux-1 Crack 9
- Non-sensitized 316LN
- Non-polluted hydrogenated primary water at
~300°C
- 20 years in service
- Elevated hardness
(~270 HV) at ID surface
- Fully IGSCC Susceptibility to IGSCC of cold work austenitic stainless steels in non-polluted primary PWR environment, T. Couvant et al., EDF, Fontevraud 10, Sept. 2022
Non-Destructive Examination (NDE) (1/2)
- Cracks were detected by manual ultrasonic testing (UT)
- Procedure used 45° shear wave at 2.25 MHz
- Manual scanning has worker dose implications
- The UT procedure was designed to detect thermal fatigue (TF) cracks. It was not optimized to detect or size SCC cracks.
- Destructive exams have confirmed intergranular SCC cracks, with depths up to 6 mm (0.25), up to 360° circumference.
- EDF has reanalyzed prior NDE data to look for missed calls, data characterized as non-relevant indications.
10
Non-Destructive Examination (NDE) (2/2)
- Safety injection lines are being inspected by remote penetrant test or by ultrasonic test (UT)
- Total Focusing Method/Full Matrix Capture (TFM/FMC ) Plane Wave Imaging (TFM/PWI)
- Scanning is still manual but encoded for location
- EDF claims flaw depth sensitivity is +/- 1mm
- TFM/FMC is a research technique, not qualified - will pursue RSE-M
- Why a qualified phased array UT (PA-UT) technique was not selected is unclear.
11
Root Cause Analysis (1/2)
- IGSCC degradation was not expected, not in accordance with the international operating experience. No SCC on the French 900 MW plant series after 30 years.
- No evidence of contamination was observed
- EDF has performed a welding simulation to estimate hardening and residual stresses in the areas where IGSCC is observed.
- An area of limited depth on the inner side of the weld is subject to tensile stress.
- A compression zone exists within the bulk of the weld. This compression zone could significantly slow down the propagation of the cracks.
- Destructive exams confirmed hardening near pipe inner diameter 12
Root Cause Analysis (2/2)
- Residual stress changes and localized hardening may occur
- Regulator asked for all repaired welds to be re-inspected
- Oxygenated water may contribute to SCC susceptibility
- Oxygen ingress in reactor coolant system might come from non-deaerated fluids (core make-up tanks, boric acid tank) or H2O2 injected at shutdown.
- Thermal stratification is expected to impose cyclic loading on the pipes
- Increased in N4-type reactors, associated with longer horizonal runs for inspectability
- Modeling by IRSN shows more stratification than previously expected 13 Main RCS Auxiliary system lines
Mitigation Plans - France
- Cracked piping sections have been removed and replaced.
- From September 2022 through 2025, EDF intends to carry out a complete examination program on all operating reactors, on the areas that might be affected by IGSCC.
- Advanced UT procedure optimized for IGSCC detection
- Technique is based on Phased Array Ultrasonic Testing (PAUT) combined with the Total Focusing Method (TFM)
- EDF plans to seek qualification of the technique in 2023
- All reactors considered as most sensitive (N4 type, reactors with indications that may have been considered as non relevant) have been stopped to carry out inspection and repairs where needed.
- A periodic inspection program will be defined, with a periodicity based on the sensitivity of the NDE, the growth rate of IGSCC, and mechanical elastic-plastic fracture mechanics analyses.
14
Status of the French Fleet
- Currently 26 reactors are shut down: 15 for stress corrosion problems and 11 for maintenance.
- Repair work was completed at 6 reactors
- Repairs are underway at 4 reactors
- EDF performed FEA stress analysis and proposed flaw evaluation criteria for continued operation.
- For Cattenom Unit 1, EDF requested continued operation for 8 months without repair of cracks (4mm and 6mm).
- On advice of IRSN, ASN denied a re-start authorization due to uncertainties in crack sizing and pipe stresses.
15
Similar Operating Experience
- A similar crack was found in Japanese Ohi Nuclear Power Station Unit 3 pressurizer spray weld in August 2020, which was attributed to hardening from high weld stress on the inside diameter (ID) surface.
- Unusual heat input at the weld
- Restriction of weld deformation (constraint)
- Residual stress would be expected
- Operating experience in U.S. PWRs has shown that stress corrosion cracking of 316 stainless steel is unlikely without significant abnormal conditions, e.g., cold working, grinding, contamination, oxygenated water conditions.
16
USA Operating Experience
- The use of ASME Code Section XI is mandated by 10 CFR 50.55a with most U.S. plants using an NRC-approved risk-informed inservice inspection (RI-ISI) plan as an alternative to Section XI.
- U.S. plants examine 10-15% of the ASME Class 1 SI and RHR piping welds under their RI-ISI programs.
- Welds are susceptible to thermal fatigue cracking.
- There have been ten incidents of thermal fatigue cracking since 2013, with seven found through UT examination and three by leakage.
- EPRI updated MRP-146 in 2018 to enhance the owners voluntary programs of inspections to detect thermal fatigue cracking.
17
More on U.S. NDE Examinations
- Class 1 stainless steel pipes are examined using multiple ultrasonic angles from four directions.
- The personnel, procedures, and equipment used on piping welds must pass rigorous performance demonstration testing under ASME Code Section XI, Appendix VIII.
- The examinations in the U.S. are optimized for thermal fatigue flaws but are capable of detecting stress corrosion cracking.
- The UT examinations have a current ability to detect cracks of 5-15%
through-wall depth and a good probability of detecting larger cracks.
- Challenges include the metal grain structure and geometric features of the pipes and welds.
Conclusions
- ASN concludes that the stresses caused by thermal stratification are likely the most significant factor in the root cause of the IGSCC.
- Older French reactors appear less affected, possibly due to design.
- French fleet inspections will be carried out with advanced UT.
- After inspections have been completed on all reactors, a periodic inspection program will be defined, with a periodicity based on the sensitivity of the NDT, the expected growth rate of IGSCC, and elastic-plastic fracture mechanics analyses.
- The US fleet, configured similarly to the older French plants, continues regular inspections. No similar degradation has been observed.
19
Global Expertise
Agenda Industry actions to consider EDF OE
- EPRI white paper (complete)
- Revision to MRP-236R1 (in progress)
- PWROG Focus Group (in progress)
- Safety Assessment
EPRI White Paper (MRP Letter 2022-018)
Conclusions:
- The most important factors that accelerate SCC to higher levels than expected in PWR primary water are residual deformation (from cold work or welding),
stress (from welding, pressurization, fit up, etc.) and environment (oxidants and contaminants in creviced areas)
- SCC in stainless steel components exposed to flowing PWR primary water will continue to occur, but there is no evidence of aging that accelerates SCC in wrought stainless steels, and no sudden increase in SCC initiation and growth is expected after decades of operation Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22
Revision to MRP-236R1 (1/2)
- MRP-236 contains information concerning SCC of primary circuit pressure boundary stainless steel, including an OE database
- Last revision was completed in 2017, which found no cases of SCC in the non-isolable portions of branch piping
- New revision (in progress) is reviewing OE since 2017
- Only confirmed cases of SCC in non-isolable portions of branch piping have occurred in the EDF fleet and one case at a Japanese unit
- Revision will also include a review of the thermal fatigue OE database (MRP-85R2/MRP-468) for cases that have the potential to be EDF-like SCC
- Focused on flaws with large aspect ratios and where a destructive exam to confirm mechanism was not performed
- Two units have presumed thermal fatigue in RHR piping with these characteristics Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22
Revision to MRP-236R1 (2/2)
Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22 EDF OE (SI and RHR piping)
Japan OE (one case in PZR spray piping)
CRDM/CEDM housings/seals Valve drain lines Isolable piping Bent CVCS HX tubes PZR heaters (non-stress relieved)
EDF/Japan OE is atypical because
- given well-controlled water chemistry, WRS alone is sufficient for SCC
- Located in non-isolable portion of branch piping
- Previous OE indicated that only thermal fatigue occurred in this piping (MRP-85R2/MRP-468), which is addressed by an inspection program (MRP-146R2)
PWROG/EPRI Focus Group
- Safety Assessment
Purpose:
Assess potential safety impact of EDF OE on the industry
- Applicability Assessment
Purpose:
Assess applicability of EDF OE to the industry Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22
Safety Assessment (1/4)
Approach
- Safety is based on risk, which is a function of likelihood and consequence
- Likelihood of SCC
- Review inspection data to determine if applicable locations are being inspected
- Review UT method to determine if EDF-type flaws would be identified
- Consequence of SCC
- Review available flaw evaluations to determine if a flaw could reach critical flaw size
- Compare design basis analysis breaks to branch line breaks Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22 Issue Recommendations (as appropriate)
Safety Assessment (2/4)
Areas of interest per EDF/Japan SCC OE Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22
Example branch line (EDF SI piping)
Safety Assessment (3/4)
Preliminary Results for Likelihood of SCC
- Inspection results reviewed for 56 units*
- PZR Spray Piping: ~60 elbow welds inspected with no reportable indications
- Gathering of inspection results continues (17 units remain for US fleet)
- UT methods employed were appropriate
- While the techniques used may not be specifically designed for IGSCC detection, they do provide reasonable assurance that if significant cracking was present, it would be detected
- UT methods specific to IGSCC are being considered Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22
- Results from last inspections performed during most recent 10-year interval
- Historically, some indications have been identified and attributed to thermal fatigue
Safety Assessment (4/4)
Preliminary Results for Consequence of SCC
- A break in a branch line is bounded by breaks considered by design basis analyses
- Given the paucity of this type of OE and the extent of inspection coverage, simultaneous breaks in multiple branch lines is highly unlikely Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22
Applicability Assessment Approach
- Primary: Elevated stresses in the elbow weld regions due to thermal stratification
- Secondary: Weld residual stress
- Assess whether these conditions are present in the industry
- Consider potential for thermal stratification in industry branch piping
- This is a primary focus for existing thermal fatigue inspection requirements (MRP-146R2)
- Compare weld procedures
- Review common practices for makeup water control and monitoring programs Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22
- EDF does not consider O2 to be a root cause
Schedule Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22 Report Projected Completion Safety Assessment January 2023*
Applicability Assessment March 2023
- Safety Assessment may need to be updated in the future as more information becomes available
Questions?
Auxiliary Piping SCC OE Focus Group Update - ACRS Meeting 11/16/22
Full Name Timestamp Court Reporter1 11/16/22, 12:45:20 PM Russell Cipolla Intertek 11/16/22, 12:45:20 PM Larry Burkhart 11/16/22, 12:45:20 PM Thomas Dashiell 11/16/22, 12:45:20 PM Michael Snodderly (DFO, ACRS) 11/16/22, 12:45:20 PM Christopher Brown 11/16/22, 12:45:20 PM Carol Moyer (NRR) 11/16/22, 12:50:09 PM DUBOIS Olivier 11/16/22, 12:51:21 PM Carol Nove 11/16/22, 12:52:31 PM Beth Kehler Haluska (Services - 6) 11/16/22, 12:52:52 PM Vesna Dimitrijevic (Guest) 11/16/22, 12:54:12 PM PETIT Marc 11/16/22, 12:54:18 PM Timothy Wyant (Guest) 11/16/22, 12:55:05 PM Dan Widrevitz 11/16/22, 12:55:50 PM On Yee 11/16/22, 12:56:05 PM Sandra Walker 11/16/22, 12:56:16 PM Matt Sunseri 11/16/22, 12:57:12 PM ALEXANDER STANISLAV CHAPMAN 11/16/22, 12:58:12 PM Bengt (Guest) 11/16/22, 12:58:14 PM LOUET Guillaume 11/16/22, 12:58:16 PM Bushmire, Anthony J.
11/16/22, 12:58:17 PM Udyawar, Anees 11/16/22, 1:00:57 PM Emma Haywood 11/16/22, 12:59:16 PM Seppnen Tommi 11/16/22, 12:59:42 PM DE CURIERES Ian 11/16/22, 12:59:48 PM Edwin Lyman 11/16/22, 12:59:51 PM Tammy Skov 11/16/22, 12:59:57 PM Thierry SOLLIER IRSN (Guest) 11/16/22, 1:00:44 PM Ness (Guest) 11/16/22, 1:00:45 PM Dariana LLANES VEGA 11/16/22, 1:01:11 PM Walt Kirchner 11/16/22, 1:02:13 PM Marlette, Steve E 11/16/22, 1:02:23 PM Enes ARACI 11/16/22, 1:02:42 PM Gregory Halnon 11/16/22, 1:02:55 PM Zhai, Ziqing 11/16/22, 1:04:26 PM Steven Wessels 11/16/22, 1:04:36 PM Mycle Schneider 11/16/22, 1:05:03 PM Andy Morley (Guest) 11/16/22, 1:05:15 PM Wilson, Bryan M.
11/16/22, 1:06:51 PM Adrien THIBAULT (ASN) (Guest) 11/16/22, 1:07:21 PM Dennis Bley (Guest) 11/16/22, 1:09:16 PM Matthew Mitchell 11/16/22, 1:10:12 PM FRANKLIN BARNABAS NOMBO 11/16/22, 1:11:26 PM Shim, DJ 11/16/22, 1:11:32 PM Fajar KURNIA (Guest) 11/16/22, 1:15:45 PM R (Guest) 11/16/22, 1:16:40 PM Christina Antonescu 11/16/22, 1:21:43 PM Zena Abdullahi 11/16/22, 1:23:01 PM
CATTIAUX Gerard (Guest) 11/16/22, 1:33:46 PM Cory Parker 11/16/22, 1:42:39 PM GARCIA HERAS MARIA LUISA 11/16/22, 1:44:05 PM Tim Watkins (Guest) 11/16/22, 1:59:20 PM Ryan Hosler (Framatome)
Chris Wax (EPRI)