2007 Transcript[191]

Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION Title: Advisory Committee on Reactor Safeguards 539th Meeting (not applicable) Docket Number: Location: Rockville, Maryland Date: Thursday, February 2, 2007 Work Order No.: NRC-1422 Pages 1-193 NEAL R. GROSS AND CO., INC. Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W. Washington, D.C. 20005 (202) 234-4433 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Two White UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR SAFEGUARDS (ACRS) 539TH MEETING + + + + + THURSDAY, FEBRUARY 2, 2007 VOLUME II + + + + + The meeting was convened in Room T-2B3 of Flint North, 11545 Rockville Pike, Rockville, Maryland, at 8:30 a.m., DR. WILLIAM J. SHACK, Chairman, presiding. MEMBERS PRESENT: WILLIAM J. SHACK, Chairman JOHN D. SIEBER, Vice Chairman SAID ABDEL-KHALIK, Member GEORGE E. APOSTOLAKIS, Member J. SAM ARMIJO, Member SANJOY BANERJEE, Member MARIO V. BONACA, Member MICHAEL L. CORRADINI, Member THOMAS S. KRESS, Member OTTO L. MAYNARD, Member DANA A. POWERS, Member GRAHAM B. WALLIS, Member NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 STAFF PRESENT: ZENA ABDUALLY WILLIAM H. BATEMAN GARY HAMMER CORNELIUS HOLDEN MICHAEL JUNGE RALPH LANDRY TIMOTHY R. LUPOLD RALPH MEYER BOB RADLINSKI TANEY SANTOS TED SULLIVAN JENNIFER L. UHLE SUNIL WEERAKKODY ALSO PRESENT: JOHN ALVIS MICHAEL C. BILLONE BERTRAND DUNNE NAYEM JAHINGIR CHRISTINE KING ALEX MARION ODELLI OZER JIM RILEY MIKE ROBINSON GLENN WHITE NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS 9) 10) 8) AGENDA ITEM 6) 7) Opening Remarks by the ACRS Chairman Proposed Revision to 10 CFR 50.46 LOCA Criteria for Fuel Cladding Materials 7.1) 7.2) Remarks by the Subcommittee Chairman Briefing by and discussions with representatives of the NRC staff Draft Final Revision 1 to Reg Guide 1.189 (DG-1170), "Fire Protection for Nuclear Power Plants," and SRP Section 9.5.1,"Fire Protection Program" 8.1) 8.2) Remarks by the Subcommittee Chairman Briefing by and discussions with representatives of the NRC staff Subcommittee Report Wolf Creek Pressurizer Weld Flaws 10.1) 10.2) 135 136 90 91 89 6 7 I-N-D-E-X PAGE 4 5 Remarks by the Subcommittee Chairman 136 Briefing by and discussions with representatives of the NRC staff 137 (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 to order. of the 6) P-R-O-C-E-E-D-I-N-G-S (8:33 a.m.) OPENING REMARKS BY THE ACRS CHAIRMAN CHAIRMAN SHACK: The meeting will now come This is the second day of the 539th meeting Committee on Reactor Safeguards. Advisory During today's meeting, the Committee will consider the following: Proposed revision to 10 CFR 50.46 LOCA criteria for fuel cladding materials; draft final revision 1 to regulatory guide 1.189 (DG-1170), "Fire Protection for Nuclear Power Plants," and SRP section 9.5.1, "Fire Protection Program"; subcommittee report on ESBWR PRA; Wolf Creek pressurizer weld flaws; proposed revisions to regulatory guides and SRP sections in support of new reactor licensing; future ACRS activities and report of the Planning and reconciliation of ACRS Procedures Subcommittee; comments and recommendations; and preparation of ACRS reports. This meeting is being conducted in accordance with the provisions of the Federal Advisory Committee Act. federal meeting. A transcript of portions of the meeting is NEAL R. GROSS Mr. Taney Santos is the designated for the initial portion of the official (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 interview being kept. And it is requested that speakers use one of the microphones, identify themselves, and speak with sufficient clarity and volume so they can be readily heard. I remind members that we are scheduled to two candidates during lunchtime today. Hopefully we'll stay on schedule and actually be able to eat lunch also. Our initial item this morning is the work on the 50.46 fuel clad criteria. And since I have a conflict of interest on that, Jack Sieber will be running this portion of the meeting. VICE CHAIRMAN SIEBER: Mr. Chairman. 7) PROPOSED REVISION TO 10 CFR 50.46 LOCA CRITERIA FOR FUEL CLADDING MATERIALS VICE CHAIRMAN SIEBER: And, without Okay. Thank you, further ado, I would like to introduce Jennifer Uhle to provide the staff's introduction to the presentation on 50.46 this morning. Jennifer? MS. UHLE: Thank you. Good morning. MEMBER ARMIJO: Mr Chairman, we did have And maybe I could a subcommittee meeting earlier. give you a little bit of a briefing. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 7.1) VICE CHAIRMAN SIEBER: charge of this session? MEMBER ARMIJO: (Laughter.) VICE CHAIRMAN SIEBER: Okay. It's okay with me. Why don't you take REMARKS BY THE SUBCOMMITTEE CHAIRMAN MEMBER ARMIJO: I just wanted to say that we did have a full day of subcommittee meeting on the 19th. Several members of the Committee were present. And we covered this topic in some depth. We had presentations, of course, from the staff and from Argonne National Laboratory as well as presentations from Westinghouse, AREVA, and G&F on the issue of the phenomenon. As we have learned at the It's a complex Committee meeting, it's complicated. phenomenon going on. The staff has done and research people done an admirable job in the research to try and understand these various components. There has been generally very good support from industry to this program, but the industry people have been reluctant to support use of the embrittlement criteria at this point because they believe the research is not yet complete. And the way to incorporate those research results into a rule is still not settled. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 7 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 7.2) staff and So we will be hearing today from both the industry. And I think the time was allocated roughly about 50/50 to give everybody a chance to make their points. With that -CHAIRMAN SHACK: MS. UHLE: Go ahead. Good morning. Thank you. BRIEFING BY AND DISCUSSIONS WITH REPRESENTATIVES OF THE NRC STAFF MS. UHLE: My name is Jennifer Uhle. I am the Deputy Division Director for Materials Engineering in the Office of Nuclear Regulatory Research. I would like to thank the Committee for taking the time to meet with us today to talk about our research program dedicated to the development of revised fuel clad acceptance criteria for postulated loss-of-coolant accidents. Of course, these famous criteria of 2,200 degrees Fahrenheit and 70 percent local clad oxidation are contained in 10 CFR 50.46. Today we will try to describe to you our understanding of these complex phenomena that contribute to the embrittlement of fuel clad under these conditions. This understanding has been developed over a period of ten years. And we will do our best to (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 8 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 summarize it in the time allowed. To facilitate our communication, we will be providing a set of proposed acceptance criteria. But I want to stress that today And that we are not presenting to you rule language. will be developed at a later date in NRR along with research support as well as stakeholder involvement. We feel there is a great need for a revision to the present rule for a variety of reasons. First, the current criteria are non-conservative. The NRC has managed this issue of ensuring plants are taking voluntary measures to ensure safety in the event of a LOCA. Second, we have shown that the criteria are affected strongly by burnup as well as a choice of alloy and even fabrication process. Third, the current rule is written to be clad-specific. And licensees are required to get exemptions from 50.46 to be able to use the new and better-performing clads. We find this to be unnecessarily burdensome to the licensees and, more importantly, to the staff because we're spending our time reviewing these submittals. need for exemptions may also And, of course, the be hampering the introduction of superior clad materials. So research believes this program has NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 generated a sufficient amount of data to proceed with the rulemaking in one presentation. In Ralph Meyer's presentation, you will see that there is one prominent area deemed the F factor, some of it citing, where we have data but we have also used some judgment to provide the basis for our proposal. Our research believes that proposed criteria will ensure safety. And it's important to go forward with the rulemaking, one of the concerns I previously mentioned, although you will hear from the industry. I think other stakeholders desire to postpone the rulemaking to provide more of a database. Our goal today is to try to convince you to support our decision and our goal to move forward with the rulemaking. views. We look forward to hearing your If there are no other questions about what we're trying to accomplish -MEMBER ARMIJO: Real quick one. If you went ahead with this, what is your time frame in which you would actually have wording that would go into the rule? MS. UHLE: Well, we have a NUREG. And we're writing them. And Research has the NUREG. it's hoping to finish it and transfer it over to NRR the end of March time frame. Then the NRR has, of (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 to that? requested rulemaking course, a rulemaking schedule developed that involves certainly the legalities of rulemaking, which is stakeholder involvement. The current rulemaking, at one point the plan that was developed a year ago indicated that the final rule would be out on the street January 2009, so early January 2009, so a few years from now. Right the now the to Commission, the of course, staff prioritize rulemaking activities. And with this realization of the non-conservatism of the current rule, the staff is questioning whether or not we need to prioritize this higher and perhaps expedite. Ralph Landry, do you want to add anything Ralph Landry is NRR. He would be in charge of the rulemaking activities. MR. LANDRY: Ralph Landry, NRR. I'm not in charge of rulemaking activities. MS. UHLE: You're in charge of the technical aspects of rulemaking activities. MR. LANDRY: said is very accurate. The point of what Jennifer We have not initiated the We are following very rulemaking at this point. closely. We have been very involved in this work with NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 different the Office of Research. a very orderly fashion We would like to proceed in to a new rulemaking, to changing the acceptance criteria. We have had acceptance criteria in 50.46 that have withstood 30-plus years of use. And as we move forward, I want to make sure that we proceed to criteria that would withstand another extended period of time that we would not need to go back and change in a very short time. And ways. we're looking at it a couple of This was brought up at the Could we put performance-based subcommittee meeting. words into the rule and details in a regulatory guide or do we have to put some details into the rule? haven't approach pursued to take exactly at the legalities but it of is We which very this point, appealing to have performance-based words in the rule itself and the details left to a regulatory guide. MEMBER ARMIJO: MS. UHLE: Okay. Thank you. So if that is all, then, I would like to introduce Dr. Ralph Meyer from the Office of Research, who is the lead technical staff member in charge of the research program. In addition, Dr. Billone, who is the principal investigator from Argonne. NEAL R. GROSS He is also here (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 that. DR. MEYER: from the industry. Okay. We've had cooperation mute now. minutes. PARTICIPANT: the bridge right now? PARTICIPANT: Thank you. PARTICIPANT: Thank you. Sorry about Westinghouse. I'm going on Can we ask who else is on working on if you have any particular questions you'd like to ask. DR. MEYER: cladding Good morning. and fuel We've been response to loss-of-coolant accident conditions for almost ten years now and have had a fair amount of cooperation that I want to mention. this program. (Pause.) MEMBER ARMIJO: Ralph, our apologies. MEMBER APOSTOLAKIS: the other side? DR. MEYER: That will cost you five Who is the person on Okay. We're ready to go. The industry has had us in I want to mention quickly that EPRI has been involved with us from the beginning. Global Nuclear Fuel, AREVA, its preceding companies, NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 13 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 and Westinghouse have all provided fuel rods and fuel cladding materials for testing in the program. And they have been very free to give us their opinions as well. In addition to that, I want to mention another program that I sometimes forget to mention in doing this work. And that's a program that we have had with the Kurchatov Institute in Moscow. The French IRSN and the NRC for almost the same number of years had been providing some support to Kurchatov to do related work. And they have done almost a parallel study to what we have done up at Argonne National Laboratory and documented that in a NUREG IA report that we issued almost two years ago. This is very extensive and unraveled some of the pieces of the puzzle that we will talk about today. So I want to mention the Kurchatov work and And I also want to mention the IRSN support work. Russian fuel manufacturer, Tivel, is also a sponsor of this work and, in fact, probably paid the lion's share of the cost, although we ran the content of the program from this little international arrangement that we had. Now, the work at Argonne has been documented in a draft NUREG CR report, which I think NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 the Committee has. We sent it to the Committee. It's a fairly lengthy report. And we spent a lot of time talking about that at the subcommittee. So there are a lot of things that happen to the fuel during a loss-of-coolant accident. And our research has looked into a number of them but has focused primarily on the loss of ductility that takes place in a process that we just generally refer to as embrittlement. During a loss-of-coolant accident, the cladding temperature goes up. vicinity softens. of 800 degrees And somewhere in the the cladding It Centigrade, It balloons. It pops. It ruptures. relieves the pressure. It also goes through a phase They're change just about at the same temperature. not totally related to each other, but they do happen at about the same time. Now, only above that temperature, starting at around 900 degrees Centigrade does the oxidation rate on the surface because it's in steam, the surface oxidation rate, picks up enough that you will accumulate a lot of oxidation during the period of the transient. And at the same time, the oxygen that is laid on the surface begins to diffuse into the metal. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Then eventually the cooling water from the emergency cooling systems comes in, cools, and quenches the material. Then it goes back through the phase change. The low-temperature phase change we refer to as the alpha. phase. The high-temperature one is the beta And I'll come back to that in just a second. Now, the current embrittlement criteria you're probably all familiar with this. CFR 50.46, part B. It's in 10 In paragraph 1, there's a That's temperature limit of 2,200 degrees Fahrenheit. 1,204 degrees Centigrade. And we will just glibly speak of 1,200 degrees Centigrade in the presentation. There is an oxidation limit of 17 percent. This is really a time limit because it was understood at the beginning and we know it now that the embrittling process does not take place on the surface where the oxide is accumulating. diffusion of oxygen in the metal. The diffusion process and the oxidation process run at about the same speed. oxidation limit was used. And so an I It is related to the It's very convenient. won't go into the details, but it turns out to be a very convenient thing to do. It gives you a nearly constant number that you can use as a limit. In running a LOCA calculation, you (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 calculate -well, your basic LOCA transient And then calculation is just time and temperature. you run along with that some equation for oxidation and get a calculated oxidation amount during the transient. And you keep that less than 17 percent, less than or equal to 17 percent. One-sided oxygen pickup is assumed everywhere along the cladding except in the balloon. And in the balloon, you recognize that you have hit a rupture. And the steam can get into the inside of the And then oxygen balloon and lay oxide on the inside. will diffuse in from the inside simultaneously with the diffusion in from the outside. So you use a two-sided assumption within the balloon. In 1998, after we became concerned about the effects of burnup on these criteria, NRC issued an information number. notice that clarified the 17 percent And we said at that time the 17 percent was total oxidation, meaning the transient oxidation plus any corrosion that accumulated on the fuel rod during normal power operation. Now, in the next ten slides, I want to just give you a brief overview of the type of work that's been done to support the criteria that we're going to describe to you later on. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 the This is work that Mike Billone spent three hours describing to the subcommittee. I'm going to I just spend about three minutes on it literally. want to give you a feeling for the magnitude of the experimental program that has been undertaken. So, first of all, here is a list of all of cladding materials that we have tested, And in Zircaloy-2, 4, ZIRLO, M5, and a Russian E110. some cases, we have had multiple subsets of these. Zircaloy-4, for example, we have three distinct varieties of Zircaloy-4. We have some older vintage 15 by 15 Zircaloy-4, some modern 15 by 15 Zircaloy-4, and some modern 17 by 17 Zircaloy-4, in addition to having the high burnup Zircaloy-4 of the older variety. MEMBER BANERJEE: by 15," "17 by 17"? DR. MEYER: The fuel geometry, the -Oh, the bundles, yes. And the What do you mean by "15 MEMBER BANERJEE: DR. MEYER: -- bundle size. geometry turns out to be important because the more rods in the array, the thinner the cladding. And you're going to see that cladding thickness shows up in one of the equations. effect on embrittlement. NEAL R. GROSS And so it has a direct (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 materials. So we have looked at all of those We have a furnace that is radiant-heated. It has reflectors and a central tube going down through there with a specimen. specimens. We can use short We can pass We can use long specimens. steam over the outside only. We can pass it up All of those through the middle and the outside. kinds of tests are done in this apparatus. MEMBER POWERS: Ralph, you indicated in your introductory comments that most of the period of time you're interested in, rapid oxidation is not taking place. Did you have to get up to above some rapid steam critical temperature before you get oxidation in the cladding? DR. MEYER: Yes. MEMBER POWERS: That means in the real reactor accident, the heat is coming from the inside out to the clad. But in your experiments, you're Does that make going from the outside in on the clad. a difference? DR. MEYER: Actually, most of the testing And so there that we have done has been two-sided. was a time when we were concerned that by doing so much of the work with two-sided oxidation, that we were not setting the test up right. NEAL R. GROSS And we did then (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 staff. see, we're do some one-sided oxidation tests. In the end, as you saw and the rest will going to suggest that the two-sided analysis be done everywhere on the run so the tests are exactly the right ones for that. MEMBER POWERS: MEMBER Thank you. So typically in a BANERJEE: bundle, at these temperatures, some portion of the heat is coming from radiation onto the surface in some form inside. What is that fraction? DR. MEYER: MEMBER The heat source is -Inside, but it's BANERJEE: radiating, right, as well? DR. MEYER: Well, but, I mean, you just So they're all -- have similar rods all around. MS. UHLE: This is Jennifer Uhle from the It depends on the I mean, that's hard to say. transient. It depends on exactly the view factors, the peaking factors because obviously you need the strong delta-T to provide the driving force. I think being from NRR, when I was in NRR, review maybe at most 20 percent, I think is from radiation at the real high temperatures. when you're up at the -The two main heat sources are But that's DR. MEYER: (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 handout. PARTICIPANT: DR. MEYER: The handout is good. -- the ring of the cladding We have a couple Anyway, the heat from the inside coming from decay heat and stored heat from the beginning of the transient and then the heat from the metal-water reaction. those are all accounted for in the analysis. So temperature is a very important. This And metal-water heat affects the temperature rise during the transient. So in setting up the experimental apparatus, a lot of effort is put into calibrating the furnace and the temperatures on the rods to be tested. That picture looks so good on the file. the main test that we do is a ring-compression test. You can hardly see it here, but there is an Instron machine that's squeezing a -MEMBER POWERS: It is much better in the that's about eight millimeters long. of Instron machines doing this. One is in a glove box And one is where we can squeeze irradiated pieces. just sitting out in a laboratory where it's easier to get to. The ring-compression test results have to be interpreted. Our techniques for doing this are much more sophisticated than they were back in 1972 NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 the time? and 1973, where the same ring compression, general ring compression, technique was used. how to do this quite well now. The furnaces are generally programmed in a way that more or less represents the temperature rise during a postulated accident. MEMBER APOSTOLAKIS: DR. MEYER: Yes? It may be obvious to Ralph? And so we know MEMBER APOSTOLAKIS: a lot of people here, but where are you going with this? What are you trying to get out of these experiments? DR. MEYER: All I want to do at -- what we're trying to get at are criteria that can be used to identify when the cladding loses ductility during this transient so you can use that as a limit and then with that limit show that the emergency core cooling systems have been adequate to protect the ductility of the material. MEMBER APOSTOLAKIS: So when do you mean How long it will take to lose ductility? DR. MEYER: Well, that's basically what we determine experimentally. MEMBER APOSTOLAKIS: DR. MEYER: Right. And then that information is (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 underlying geometry. contained in the temperature limit and the oxidation limit and applied in the analysis, the safety analysis, when you analyze the thing. VICE CHAIRMAN SIEBER: Those limits are surrogates for the loss-of-coolable geometry, which is the endpoint. You want to maintain coolable geometry. Endpoint is loss-of-coolable DR. MEYER: There were big discussions about this It came down to during the hearing in 1972 and 1973. a position of maintaining ductility in the cladding as the way to ensure a coolable geometry. And we have not tried to change any of the philosophy or the basic experimental approach to it but just do it in such a way that we can see the effects of burnup and manufacturing variables and update the criteria. We were able to do four what we call integral tests on high burnup rods before we lost access to the alpha-gamma hotcell at Argonne. these are pictures of those four. And All four of these And fuel rods were BWR fuel rods with low corrosion. you can see the single balloon and ruptured area in each of those. We analyzed those in detail. MEMBER ARMIJO: Ralph, I'm sorry. You (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 didn't do the H. B. Robinson? Robinson? DR. MEYER: Robinson before the -MEMBER ARMIJO: DR. MEYER: we have the specimens. Before they shut down? So But We didn't get to the H. B. Didn't you do an H. B. -- hotcell was shut down. And we want to test them. we have had no ability to do that since July 26, 2005. MEMBER ARMIJO: DR. MEYER: Okay. We remember the day. And you also have the M5 MEMBER ARMIJO: fueled rods and the ZIRLO? DR. MEYER: No, no. It's a very painful process to get fuel rods from a power plant for testing. And over the years, we have been able to get a set of BWR rods from the Limerick plant and a set of PWR rods from the Robinson plant. These are relatively older fuel types. We have plans to get ZIRLO-clad rods and M5-clad rods with high burnup for this program. rods have not been provided yet. the current test program. What we were able to get were some small pieces of M5 and ZIRLO cladding from high burnup rods, getting those pieces from the Skuzda Laboratory, where NEAL R. GROSS Those So those are not in (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 EPRI. they had such fuel rods for testing and we made arrangements to get those pieces. not been tested yet either but Those pieces have will be tested hopefully in the next two months. DR. BILLONE: Billone from Argonne. Excuse me. This is Mike Just for clarification, the high burnup M5 rods that we and EPRI have agreed to put into the program are in transit to Argonne. They have been in transit for six months, but they're in transit. PARTICIPANT: DR. BILLONE: Slow truck. Slow truck. But you physically have MEMBER ARMIJO: the H. B. Robinson rods, -DR. BILLONE: MEMBER ARMIJO: old vintage -DR. BILLONE: MEMBER ARMIJO: DR. OZER: Yes, yes. -- even though that's an Correct. Okay. This is Odelli Ozer, They're at the Excuse. The M5 rods have been shipped. Idaho National Laboratory. They're just awaiting shipment from Idaho hotcell over to wherever Argonne wants them. And we are in discussions with (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 effects Westinghouse for ZIRLO rods. DR. MEYER: Well, since we've gotten onto this subject, let me say that we have had a program plan for this program since 1998. 2003. It was updated in It has been reviewed by the subcommittee and by the full Committee several times. In that program plan, we always knew that we would not have the high burnup ZIRLO and M5 rods in time in the time that we wanted to try and revise the embrittlement criteria. And so the plan for the beginning was to examine unirradiated rods of Zircaloy-2, Zircaloy-4, M5, and ZIRLO and irradiated Zircaloy rods. With this cut of the variables to make an assumption that the burnup effects that you saw in the Zircaloy would apply to M5 and ZIRLO because we realize that we wouldn't have those rods in any timely way to make the test. And that turned out to be the case. So what we're going on here are burnup measured on Zircaloy and, by assumption, carried over to M5 and ZIRLO with the alloy and manufacturing properties measured on the unirradiated material. I think we understand enough of what is going on that this is a reasonable approach. NEAL R. GROSS And I (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 hope that I can convince you of that. Sometimes if we want to do a mechanical test in the balloon, instead of cutting a ring and compressing it because of the deformation, we do a bending test. And we do a lot of microscopy to look at the details of metallurgical phases in the oxide layers that build up on the rod. This happens to be We do a lot a scanning electron microscope picture. of optical microscopy also. Okay. So that was my three-minute sweep Now what I want to through the experimental program. do is to slow down and talk about what is really happening and what we have learned from the results and then how we propose to use those results. So imagine that a fuel rod has been through a temperature transient such as the one that I showed and has now been cooled back down to near room temperature and you look to see if it's brittle or ductile. So what you see when you look at the sample is that there is O 2 on the surface, oxide on the surface, and then you see material that when it went up in temperature had all transformed to the beta phase. oxide But as oxygen diffused into the metal from the that's lying on the surface, the oxygen (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 concentration in the beta phase got above the solubility limit and caused it to go back to the alpha phase, which had a lot of oxygen in it. And so when you take it all back down to room temperature, what you see is a region that was in the beta phase at high temperature. this oxygen-stabilized alpha layer. you see the oxide layer. Among these phases, the only one that has any ductility is a portion of the prior beta phase. It's the portion of that phase that has a low oxygen content, a content lower than about six-tenths of a percent of oxygen. VICE CHAIRMAN SIEBER: Could you tell us You clearly see And, of course, which phase is body-centered cubic and which is -DR. MEYER: Yes. -- phase-centered? The VICE CHAIRMAN SIEBER: DR. MEYER: Yes, I can. low-temperature alpha phase is a hexagonal close-pack structure. And the high-temperature beta phase is a body-centered cubic. When the original work was done in the late '60s and early '70s and the rule was first written, there was this Appendix K that you are probably all familiar with. Appendix K required that (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 don't. they don't. MEMBER ARMIJO: Okay. Confirmed that they you use the Baker-Just oxidation equation. And that was because the data that Hobson had taken, which were used as the basis for the 17 percent number, had been reduced with the Baker-Just equation. oxidation. Hobson did not measure the amount of He calculated it with Baker-Just. So he used Baker-Just going in. out. And it worked. We're He used Baker-Just coming switching from the Baker-Just correlation to the Cathcart-Pawel correlation because it's a much more accurate correlation. And I just wanted to put in your handout the equations that we're using so that they would be for reference. I don't think I need to talk about those in any detail. MEMBER ARMIJO: Ralph, I just want to ask Have you confirmed one question and just to be sure. or is it well-known that the oxidation kinetics for the, let's say, various types of zirconium alloys, Zircaloy-2, 4, M5, and ZIRLO, have the same activation energies and pre-exponentials so that this one equation represents the whole family? DR. MEYER: Yes. We have confirmed that So would you use a different equation for each NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 if you're -DR. MEYER: explain this. Okay. I really need to The And I know I ought to do this. Cathcart-Pawel equation works very well for all of the alloys we have tested, the ones you have mentioned: Zirc-2, Zirc-4, M5, and ZIRLO, at the high temperature end of the range of interests. At 1,200 degrees Centigrade, a Cathcart-Pawel works quite well for all of them. As you go down in temperature, particularly the M5 alloy, which has no tin in it, it's just zirconium-1 niobium, it has slower oxidation kinetics, say, around 1,000 degrees Centigrade. much slower. Now, by using the Cathcart-Pawel equation, even for M5, we're not introducing any error into the situation because it's just the parameter that we correlate against. It's our surrogate for time. So It's it does not represent the true oxidation rate for M5 at lower temperatures, but it is still a good time yardstick. MS. UHLE: This is Jennifer Uhle. Couldn't you just stick MEMBER POWERS: with Baker-Just, then? DR. MEYER: We could have used Baker-Just. (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 staff. MS. UHLE: This is Jennifer Uhle from the We just want to point out that in the regulatory guide or perhaps in the ruling, which when it gets worked out will be the guidance to make sure that whatever correlation or whatever equation, oxidation equation, you're using to reduce your data to show when you lost ductility, you have to use that in your system analysis code that will tell you what your fuel rods would be, how brittle they would be during a loss-of-coolant accident. So right now there is a disconnect because in the 17 percent limit currently, that was derived using Baker-Just. However, in best estimate methods that the licensees have and vendors have been using for they have NRR approval to use, they're free to use whatever correlation is acceptable for the oxidation equation. So there is currently a disconnect. Now, thankfully it's not that much in error, but in the future, we need to make sure that those two are consistent. MEMBER CORRADINI: you tried to explain it. I don't have it. Just to clarify because I thought I got it, but now So let's just stick with Cathcart-Pawel. And you were to take a set of data. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 just ask, time. So you said this is just a surrogate for So what you're using this for is to compute a percent reaction given the protocol, which is if it's not ballooned, it's one-sided. If it's ballooned, it's two-sided and then with that percentage, then come back to a time. I'm still not clear about that because what you said about M5, I remember being the case. I'm not exactly sure how it still sounds to me like using Cathcart-Pawel. you With a range of temperatures as cook the fuel, you're going to overestimate oxidation. DR. MEYER: You will overestimate oxidation for M5, for example, because it spent some time at a lower temperature. the oxidation process But, as it turns out, doesn't control the embrittlement process. It's diffusion into the metal So we're that controls the embrittlement process. just using oxidation rate as sort of a surrogate for diffusion rate because we can measure it. MEMBER CORRADINI: then, the obvious and I understand. question? a May I So if I for overestimate diffusion, oxidation it's surrogate why am I not also overestimating the diffusion time and, therefore, overestimating the NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 will form embrittlement? DR. BILLONE: DR. MEYER: Ralph, can I -Help me out, Mike. No. Let's look at it a DR. BILLONE: different way. If I test M5 at 1,200 degrees C. and Zirc-4 at 1,200 degrees C. or 1,000 degrees C. -let's go to the 1,000, where they're very different -they pick up weight, oxygen, at different rates, but they embrittle at about the same rate because what's controlling is a diffusion process of oxygen into the metal and through the metal. So M5 forms a thin oxide layer. a thick oxide layer, which Zirc-4 doesn't contribute at all as long as you have an oxygen source there to drive your diffusion. The simple fact is when you plot M5 ductility goes down like that with time versus Zirc-4 ductility, which goes down. level. MEMBER CORRADINI: So one last question, They go down at the same and then I'll be quiet, which is then the oxidation kinetics is nothing. You are using the A and the Q and the R essentially as a solid diffusivity model, which is approximately right, regardless of the oxidation. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 MEMBER BONACA: Right. Okay. Fine. MEMBER CORRADINI: DR. MEYER: Now, I want to point out that you would not want to do this in calculating the metal-water heat, the separate matter. The metal-water heat you would want to use a best estimate oxidation correlation. But for us it turned out to be convenient just to use this same calculation, plot all of our data not as a function of time but as a function of what we call CPECR, Cathcart-Pawel Equivalent Clad and Reactive. MEMBER ABDEL-KHALIK: Would you expect this to work for any and all yet-to-be-developed alloys? DR. MEYER: I expect this to work for any and all zirconium-based alloys that are in the tin niobium family at the concentrations of around one percent; in other words, the range of things we -MEMBER ABDEL-KHALIK: DR. MEYER: And this is -- We tested all the way from Anything in that zirconium-tin to zirconium-niobium. range I believe these results will be applicable. MEMBER ABDEL-KHALIK: is based on what? Intuition? It's based on testing that And this expectation DR. MEYER: (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 wide variety of materials that are in this range which have not only differences in composition but also differences in fabrication and understanding which differences cause some change in the ductility behavior and arranging the criteria in such a way that it would catch all of them. MEMBER BANERJEE: What is that 87.8 there? DR. MEYER: It's just a geometric factor. There are four A couple of questions. Let me define equivalent cladding. hours of details involved in this subject at least. Equivalent cladding reacted is where you can do a calculation and you assume that all of the oxygen that is consumed goes into ZrO 2 at the surface. is lost by diffusion into the metal. That's what ECR is. was used 35 years. concept. It's a concept that And none There's nothing wrong with the And we stick with it, with the concept. MEMBER BANERJEE: What is it? It says it's 20 percent oxidized or something, 17 percent? What does that sort of pertain to? DR. BILLONE: It pertains to the fraction of the wall thickness that you consume. DR. MEYER: The temperature of the time and the wall thickness, yes. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 T as well? DR. MEYER: Yes, yes. Okay. MEMBER BANERJEE: MEMBER ARMIJO: MEMBER Thank you. Okay. Now, this APOSTOLAKIS: temperature, Ralph, which temperature is this peak? Is that time-dependent as well? DR. MEYER: MEMBER time-dependent? DR. MEYER: Yes. So time is varied in Yes. Big T. That's APOSTOLAKIS: MEMBER APOSTOLAKIS: MEMBER APOSTOLAKIS: DR. MEYER: So you do the calculation. And you, just like that first slide that I showed you, have temperature running along with time and changing. And you can in the models integrate the amount of oxidation that takes place, which is a good surrogate for integrating the amount of diffusion that takes place because they have the same kinetics and roughly the same coefficients. Okay. The first and main result that we see is that, sure enough, the high burnup material embrittles in less time; that is, at a lower calculated oxidation level, than the fresh material. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 knows, your Robinson vintage. So what you have here is irradiated H. B. Zircaloy-4. It's 15 by 15. It's old It has a rough surface. And it is embrittling at around eight percent ECR, which is well below the 17 percent number that we have talked about. Now, if you take very similar unirradiated material -- I'm not quite sure it deserves to be called archive material, but it's as close as we could get to archive material. So here we have this same And we test vintage unirradiated 15 by 15 Zircaloy-4. that. And it tests out at about 14 percent. Now, just as a little matter of interest, this is with the Cathcart-Pawel model. If we had been using the Baker-Just report, Baker-Just equation, there's a 3 percent difference. percent. It would be 17 This is exactly what was tested, the result that was obtained in the early 1970s, on which the original rule was based. MEMBER ARMIJO: ductility Ralph, just so everybody is two percent reference ductility. achieve. That's your target that you want to DR. MEYER: Yes. Okay. Sorry about that. I MEMBER APOSTOLAKIS: DR. MEYER: Sorry. (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 burnup of -DR. BILLONE: DR. MEYER: Sixty-seven. -- 67 gigawatt days per ton. You just have glibly used the word "ductility" here. actually have two techniques that we use. We One of them comes directly from the Instron machine, where we look at the displacement versus time and can get something we call an offset strain. And the other method is actually simpler. measure the diameter of the ring with micrometers before you squeeze it and after you squeeze it, right at the point where you develop the first through-wall crack. And in one case because of bending and other things that I don't understand but I hope Mike understands, in one case the zero is at one percent when you use micrometers and it's two percent when you're using this offset strained value that we measure. MEMBER APOSTOLAKIS: "high burnup" roughly -DR. MEYER: High burnup. This had a So when you say, You can see that the specimen that was tested here had a corrosion thickness of about 80 microns. If you run the numbers and take 14 percent, convert 80 microns to NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 and let -MEMBER BANERJEE: You keep saying "F yet. DR. MEYER: Wait for a couple of slides comes in. equivalent cladding reacted as a percentage, multiply that by 1.2, which is the F factor, and subtract it from 14, you get 8. So this is where the so-called F factor The reason that we didn't just say 1.2 right off the bat was before we made the measurement, we didn't know what the number was going to be. so we just put a factor in the equation. After we measured it, we found some And sensitivity to heat-up rates and cool-down rates, which could cause this F factor to have several values. So we have, in fact, explored the possible range of those values and, as a matter of judgment, selected 1.2 as the most appropriate value to use. This is the point where judgment has entered into the final result and where there can be some difference of opinion on what the F factor should be. MEMBER BANERJEE: what the F factor is? MEMBER ARMIJO: He hasn't gotten there Could you just repeat (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 factor. factor." And I don't know where it is. DR. MEYER: F is for factor. All right. It's just a And Look, this is an empirical correlation. what we're doing is we know that the main effect -and I forgot to say it -- here is a result of hydrogen that gets absorbed into the cladding during normal operation as a consequence of the corrosion process. We know that about 15 percent of the released hydrogen gets absorbed into the cladding. But I said before that oxygen was the embrittling agent in the material. And so what we believe is going on here is that the hydrogen is controlling altering rates. So it's not necessarily doing any the both the solubility limits limits the or it's solubility and diffusion embrittling on its own because it's all in solution at the high temperature, but it is affecting the oxygen diffusion into the metal. And on this slide, I simply show that we have confirmed that hydrogen is having this effect by taking unirradiated Zircaloy-4 and other materials, pre-hydriding them in the laboratory, and then testing them in the same way. effect by doing that. NEAL R. GROSS And you can reproduce the (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 MEMBER ARMIJO: Ralph, there is a contention that I want to give you a shot of answering before the industry folks talk that by virtue of quenching these materials from high temperature, you introduce a hydrogen embrittlement, in addition to the oxidation embrittlement, because that's an issue that is going to come up we'll have to wrestle with. Have you confirmed that the hydrogen effect is strictly oxygen or is it oxygen embrittlement plus hydrogen embrittlement? DR. MEYER: Well, now, I think that there is a component of direct hydrogen embrittlement in the samples that have been -- is it the quenched ones or the slow-cooled ones? I get confused on this. But all of this is wrapped up in the cooling rate -MEMBER ARMIJO: DR. MEYER: Right. -- effect, which we have Mike, do you looked at and made some judgments about. want to -DR. BILLONE: Yes. I would say most of that loss of ductility that you see is due to increase in oxygen. There's a small but significant -- in other words, if you're setting two percent as the limit, if you slow-cool the sample, you might get three percent ductility where you expect less than NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 41 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 hydrogen in three. And so essentially quenching freezes in solution in places where it causes embrittlement. So if you quench at 800 degrees C., your sample is going to be more brittle than if you just cool to room temperature with no quench. MEMBER ARMIJO: DR. MEYER: Okay. Now, the next big Okay. effect that we found in this study was actually noted first in some Eastern European tests that were done in the '90s. And we learned from what we did that this breakaway oxidation process had it been seen earlier, in fact, affects the embrittlement process. So what happens with the zirconium alloys? And it can happen to all of them. It turns out that the old E110 Russian cladding was the most susceptible to this and provided the most dramatic pictures of it. But what happens is that as you enter this high temperature region and you start laying down the oxide on the surface, that the type of oxide that we normally see is black and shiny. form. It's a tetragonal And it's rather protective and doesn't allow the hydrogen to enter in any significant amount during the period of the high temperature transient. Under some conditions, this oxide can NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 42 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Chairman? oxide? DR. MEYER: Yes, those were flakes. That switch from a tetragonal to a mono-clinic form. mono-clinic form is not black and shiny. looking. It's full of micro cracks. A It's dull And it lets hydrogen in. And so as soon as you get into this break-away process, hydrogen starts getting sucked into the cladding and has the same effect as it had before. So you have to be careful with all of these alloys to make sure that you don't have the conditions that promote the bad oxide to grow. VICE CHAIRMAN SIEBER: Is that flakes of was a very advanced case of stuff. picture because of its dramatic effect. MEMBER ARMIJO: I like that Ralph, to be sure that everyone has some time, it might be a good idea to get to your proposed. DR. MEYER: Okay. It's 9:27. And we're MEMBER ARMIJO: supposed to wrap up at 10:00. DR. MEYER: MEMBER Okay. Is that right, Mr. ARMIJO: So I think it's important that people understand your proposal. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 one. MEMBER ARMIJO: DR. MEYER: Yes. Okay. one there? DR. MEYER: I do need to talk about this then. MS. UHLE: Why don't you talk about that DR. MEYER: Okay. Let me skip over these, Okay. So the concept is that diffusion of oxygen into the metal is the embrittling factor, not laying down the oxide on the surface. It turns out that you have a big source of oxygen on the inside of all cladding materials, UO2 fuel full of oxygen. And we know from our present work and from some historic work that we looked up that as soon as the cladding and the fuel stick together, that source of oxygen then becomes available for diffusion into the cladding. I think we have incontrovertible -- is that the right word? -- evidence that this effect is real and it is at least when you have a bonded fuel layer, which you generally would have at high burnups, there is ample oxygen on the ID. So that you get diffusion from both directions, whether you're in a balloon or not in a balloon. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 44 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 hydrogen. readily. VICE CHAIRMAN SIEBER: comes from the UO2? DR. MEYER: comes from -VICE CHAIRMAN SIEBER: pretty tightly bound. DR. MEYER: It comes from the UO 2 That ample oxygen It comes from the UO 2 . It I thought that was . One other thing I need to point out -- and then I'll get right to the criteria -- is that within about an inch of the center of the rupture, you also have hydrogen absorption on the ID. You had steam getting in, oxidizing the inner surface of the cladding, where it can get in the balloon. And, again, the oxidation process frees up And the hydrogen isn't swept away very It's trapped inside. And so you get high hydrogen absorption in the vicinity of the balloon. MEMBER BANERJEE: So does the oxygen diffuse through the oxide layers, oxide layer crack, and get through the reaction zone? DR. BILLONE: No, no. What happens is you are getting oxidation in the opening, the balloon opening region. MEMBER BANERJEE: I'm saying imagine you've got this bonded fuel or whatever. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 45 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 not cracked. DR. MEYER: No. It diffuses in. Aren't there kinetics the ID. bonded. DR. BILLONE: have expanded 50 percent. MEMBER BANERJEE: understand that. DR. BILLONE: I'm trying to answer about Right, right. Yes, we Well, in the balloon, you DR. BILLONE: Right. Oxygen is now diffusing MEMBER BANERJEE: through the oxide or are there cracks in the oxide and allows oxygen in? DR. BILLONE: an oxide layer. that layer. VICE CHAIRMAN SIEBER: MEMBER BANERJEE: DR. BILLONE: MEMBER Right. The steam oxygen is creating And oxygen is also diffusing through From the inside? From the inside. But when it's just BANERJEE: I'm missing the point. DR. MEYER: It is present on the surface, and it just diffuses in. MEMBER BANERJEE: It diffuses in. It's MEMBER BANERJEE: associated with that diffusion? NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 46 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 DR. BILLONE: That's the same kinetics. It goes to teach us how? MEMBER BANERJEE: DR. MEYER: I need to go through this. So we get And then I think I'm where you want to be. these high hydrogen concentrations, very high hydrogen concentrations, balloon. The balloon does not stay ductile. It has 3,000 ppm, the vicinity of the some strength left, but in spite of the fact that the current regulation has detailed prescription on how to analyze the balloon, it really doesn't work because the balloon has hydrogen in it that wasn't realized when the rule was put together that causes the balloon to be -- let me go right to here. if I have to. So here is what we are proposing to do. We're proposing to keep the temperature limit right where it is with no change. with this. There's a lot of history And I'll come back And there's also an effect that we see in the present work. Once you get above about 1,200 degrees Centigrade, the oxygen diffusion rate picks up. the oxidation limits would then be lower. And And so you basically have more parameters here than you need. And you can just fix this temperature NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 right where it has always been at 2,200 degrees Fahrenheit and then work the rest of the problem from there. So that is what we have done. And now what we are looking for here is a replacement for 17 percent, which accounts for the effect of burnup. And so we will start with a measurement on unirradiated cladding at 1,200 degrees. And there's a reason for choosing the 1,200 degrees. This is the analogue of 17 percent. show you some values. I'll And we subtract from that the corrosion thickness multiplied by a scaling factor, just an empirical factor, to fit the data. MEMBER ARMIJO: is one, right? DR. MEYER: Yes. If you were to use the Now, currently that factor information notice recommendation, that factor would be one. I have to tell you that at the time the information notice was written, we did not understand this process. It was a guess. We expected that there would be an effect, and it was a logical guess to make. MEMBER ARMIJO: You have incorporated all burnup effects into that 1.2 times -DR. MEYER: Well, not quite all because there is the matter of break-away -NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 data and factor. MEMBER ARMIJO: DR. MEYER: separate limit. two-sided Right. Okay. -- that is accommodated by a And then there is the matter of the All of these are oxygen penetration. accommodated by everything that is on this page, but the first line takes account of the basic burnup effect that is a consequence of corrosion and hydrogen absorption during normal operations. MEMBER APOSTOLAKIS: formerly one, ECR corrosion -DR. MEYER: Yes. -- was that in the This factor of one, MEMBER APOSTOLAKIS: information notice? DR. MEYER: Yes. In the information notice, we simply said, "Interpret the limit to be the sum of the transient and the corrosion thickness." So, in effect, you're subtracting the corrosion thickness from 17 percent. And we didn't say multiply it by an F We just said -MEMBER APOSTOLAKIS: DR. MEYER: some So F is 1.2? F is 1.2 based on our current about the appropriate judgment adjustments to make to account for these cooling rate effects. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 49 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 650, Okay. Now what we're suggesting here is that we assume two-sided oxygen pickup everywhere on the run and simply not do a calculation in the balloon. And if you'll let me, I'll show you how I get to that point on the next slide. And then, finally, we measure the minimum break-away time and use that time as a time limit for the period in the transient above 650 degrees Centigrade. numbers -- The reasons for all of these choices of MEMBER ARMIJO: That time is the same as the time allowable for the entire transient, that you can't get break-away during that transient? DR. MEYER: you're not The period above 650. to creating Below this susceptible break-away oxide, but above 650, you can get the break-away oxide. may persist, even And once it starts developing, it if you change and move to a So we look different temperature in the transient. for the minimum. examples. MEMBER CORRADINI: And I'll show you some numerical Just for clarification, Ralph, I just wanted to -- so the ECR is using the Cathcart-Pawel model at 1,200 C.? DR. MEYER: The ECR unirradiated is the (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 You do test. experimentally observed. MEMBER CORRADINI: understand what you just said. computation, what am I I'm just trying to So if I were to do a So the ECR computing? unirradiated is using the 1,200 C.? DR. MEYER: This is a measured result. And you find the transition from ductile to brittle behavior, just like we showed on those slides. And you do that with Cathcart-Pawel ECR on the x-axis, instead of time. MEMBER CORRADINI: DR. MEYER: Right. I understand. And And you take that number. that's what you have right here. MEMBER CORRADINI: Okay. But I am going You don't So to say it back to you so I get it right. have a stylized time history for the temperature. you're using that ECRunirradiated at a constant 1,200 C.? DR. MEYER: That's correct. And then the second ECR MEMBER CORRADINI: corrosion is what again? MS. UHLE: What is that ECRcorrosion? That's the preexisting corrosion that occurs when the rods are just burned at normal temperatures. MEMBER CORRADINI: DR. MEYER: Calculated how? The H. Again, it's measured. (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 51 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 corrosion B. Robinson rods had 80 microns of oxide on the surface of them. MS. UHLE: DR. MEYER: The licensees -The vendors know what their rates are so they can tell you approximately. MEMBER CORRADINI: So this is allowable by the way you're doing this that this is allowable relative to some predetermined corrosion rate buildup as a function of burnup? DR. MEYER: MEMBER Correct. So for a given APOSTOLAKIS: burnup, given kind of fuel, the right-hand side of the inequality is a number? DR. MEYER: Yes. On the left, you go MEMBER APOSTOLAKIS: to the equation, right? DR. MEYER: On the left is your -Calculation. MEMBER APOSTOLAKIS: DR. MEYER: MEMBER -- calculation, your -But you used BANERJEE: Cathcart-Pawel first, too. MEMBER APOSTOLAKIS: result of that is? Time. So my last question is Yes. And then the MEMBER CORRADINI: (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 only? DR. MEYER: Yes. In this case now, the temperature. result. to get to the -MEMBER BANERJEE: It's whatever ECR is the It has to be less than the right-hand side. MEMBER APOSTOLAKIS: Right. But that determines what? MEMBER BANERJEE: MEMBER KRESS: Time and temperature. But in making that calculation, you use the area of the clad on both sides? MEMBER CORRADINI: It's the thickness. DR. MEYER: DR. BILLONE: DR. MEYER: We're using -Two-sided oxidation. -- two-sided equations. And it's a function of It doesn't matter. MEMBER KRESS: So you have to do it along the whole wad at different temperatures? DR. MEYER: Well, you do it just at the peak, like you do now at the peak temperature node. MEMBER KRESS: You're looking at the peak peak maximum oxidation would always occur at the peak node, peak temperature. MEMBER KRESS: Okay. So you do it at the (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 53 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 -DR. MEYER: MS. UHLE: -- for both of them. We have to subtract off the So you that's peak. But the temperature is a function of time? DR. MEYER: Yes. One last clarification. So that MEMBER CORRADINI: So Sam asked you about t in the last arrow. t always has to be less than the actual transient time because you're going to be much below 650 for a lot of the transient. I know you don't want to do this, but just to ask it theoretically, so you have gone through all of this effort in the first arrow to take time and wrap it into an ECR. time measure. measures? DR. MEYER: way I look at it -MS. UHLE: Well, because we also have to Well, that is basically the But, yet, you come back to a So why not just simply have two time preexisting corrosion from the ECR calculated. need it to be in some sort of format consistent. MEMBER CORRADINI: crazy academic. So I'll -No. I understand. I'll Okay. But I'm just a DR. BILLONE: (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 -MEMBER CORRADINI: But you know it's the answer your questions. MEMBER CORRADINI: fuel is inside the core. I know how much the So It's in so much time. I'm got cooking time at one temperature. cooking time during a transient. I've got I have time. So if you're going through all of the effort to get an ECR and have a stylized thing to be a surrogate for time, then you come back to a second requirement that's time. time? DR. MEYER: Well, in most cases, hopefully Why not just simply use real time of how you have the fuel in the core. DR. BILLONE: DR. MEYER: the transient. DR. BILLONE: presenting this to you. Time is a simplistic way of In his first viewgraph he May I try something? The time is the time during showed you of temperature versus time, you're going to be integrating ECR over that and you -MEMBER CORRADINI: DR. BILLONE: not pure time. measure of On the left-hand side? It's On the left-hand side. It's time and temperature, which is a oxidation. And it relates to (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 55 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 That's true. issue there embrittlement. So if you only go up to 1,000 degrees You're C., you're going to have the same time period. going to get a low ECR calculation. MEMBER CORRADINI: I'll stop now, but I'm still getting that you have a correction and the correction factor is time of operation time. So I can rearrange the thinking process and take the right-hand side negative, put it over there, and operation -MS. UHLE: because But there's still a temperature the different rods are at different temperatures. Okay? So we don't know what the -- we can't just say this rod is going to be the limiting rod and we know it's operating temperature throughout the entire life span of that rod. MEMBER POWERS: It would be a difference between a small break and a large break LOCA. MS. UHLE: Yes. So you still have a time-temperature type couple there that you need to factor in. MEMBER CORRADINI: I understand. That's a good point. Thank you. So, then, last question about the arrow on the little t. So the history of how any individual rod is sitting inside the core is not going to affect that? That is, I can have a hot rod -NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 56 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 -MEMBER ARMIJO: Ralph give his thing. I think we should let example -MEMBER ABDEL-KHALIK: Before you take this correct. MS. UHLE: It's just the time in the MS. UHLE: a burnup effect. DR. MEYER: So far as we know, that's That's right. It does not have transient that exists above the 650 has to be -- and the calculated transient that the licensees provide would have to make sure that the time above that was less than the minimum time. MEMBER CORRADINI: DR. MEYER: Thank you. If you let me do one numerical I think it's a little bit complicated, but that second criteria is just to prevent really crummy alloys from getting into your reactor. And that's a real simple thing. The real meat of the issue is the ECR during the LOCA transient. And so there are really two things that they are trying to protect. And I think belaboring that break-away thing isn't worth much, but Ralph should give an example of how he would apply this to NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 57 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 I guess. DR. MEYER: It depends on the alloy. It problem. between. MEMBER ARMIJO: DR. MEYER: Okay. Thank you, Ralph. And if a bad one. MEMBER ARMIJO: DR. MEYER: Okay. to -DR. MEYER: Let me show you a good one and a real material. So I think with that, I am going to have And I'll skip over the five in Okay. So here is M5. we take a fresh piece of M5 tubing and find the point at which it loses its ductility, it's about 20 percent in this ECR definition. This is a typical value. At end of life, M5 might have 40 microns of corrosion. And you not make a geometric conversion And it happens to be of 40 microns to the ECR unit. four percent. MEMBER APOSTOLAKIS: unirradiated was 17 percent. DR. MEYER: I thought the ECR It's not. That's part of the It's not. One size doesn't fit all. MEMBER APOSTOLAKIS: MEMBER BANERJEE: Okay. Okay. It depends on the alloy, (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 58 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 number? DR. MEYER: Measured. The second one is probably depends more on a couple of fabrication steps than on the alloy composition. So now if you take 1.2 times 4 and subtract it from 20, you get 15.2 percent. The current limit would be 17 percent minus 4 percent or 13 percent. So, actually, you have a higher limit with this material. And the measured time at which break-away occurs at the worst temperature is on the order of 5,000 seconds. So you have A typical LOCA is what, 1,800 seconds. no problem with break-away on this material. And you would use in your calculation 15.2, Everything else would run the same way instead of 17. that the current analysis is done. This is going to ensure you that you have covered the effects of manufacturing variables, alloy, burnup, everything that we have found. MEMBER BANERJEE: measured number -DR. MEYER: MEMBER Yes. -- or a calculated The first number is a BANERJEE: also measured, but it will come from the vendor's correlation from measurements in the plant. MEMBER CORRADINI: We know the bad one (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 59 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Dr. Ozer. corrosion. already. DR. MEYER: You know the bad one already is the old style Russian E110, where we measure with the fresh cladding of transition of about 12 percent, not 17 percent. Now, this cladding is very resistant to It has low oxygen content and it's also And so the like the M5 is Zirconium-1 niobium. material, we had some 50-gigawatt day per ton cladding in the Russian program. percent. So you get a limit of 11.4 percent for this material. If you were using the current rule, it And that converts to only .5 would be 17 percent minus the .5 or 16 and a half. But look at this. starts in about 500 seconds. these limits no longer apply. embrittle. The break-away process So after 500 seconds, Very quickly, it will And so if you had a LOCA transient with this fuel that spent more than 500 seconds above 650 degrees Centigrade, it probably would not retain ductility after that transient. Do you want me to quit now? MEMBER ARMIJO: Yes. We are going to have Dr. Odelli Ozer from EPRI is going to speak for the industry people, although there are some here NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 60 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 available to answer questions. Dr. Ozer? MEMBER ARMIJO: your presentation? DR. OZER: Ralph? The blue folder? Thank you very much. Do we have handouts of MEMBER ARMIJO: DR. OZER: Yes. I would like to thank the Committee for giving me the opportunity to express the industry's position on this. time. I know we are kind of short on So I am going to try to be rather concise. First of all, let me state that we are fully in support of the overall objective of NRC in trying to develop performance-based criteria because such criteria will allow the introduction of new materials without the concern about getting exemptions so the licensing process will be much smoother, will go much faster. We are also very much in support of the excellent work that is being done at Argonne, the work that Ralph covered in three minutes. very much in support of that. Our concern is primarily with the You know, we're interpretation of that work and with the proposed changes, the changes that are being proposed to the NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 61 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 current criteria. First of all, I think we all agree that the data has not shown the presence of any public safety issues. MEMBER POWERS: I guess I just don't Do you mean to tell understand that statement at all. me that it's perfectly okay to embrittle a clad during a transient so that when it shatters during cooling -DR. OZER: If you let me go -No. I want to understand MEMBER POWERS: this sentence. DR. OZER: -- I will address it. I would As I like to address that to some greater extent. said, we do have concerns about the interpretation of the rules and, in particular, the use of the F factor, which has been discussed at length, and the fact that we may be getting oxygen ingress from the ID, how to address that, whether to address it by assuming double-sided oxidation. as well. And, you know, the main concern is that we feel that a rather bounding approach will have a rather significant negative impact on the industry. Again -MEMBER POWERS: I am still coming back to We are concerned about that (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 62 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 the your first sentence. better to me. You've got to explain that It seems to me that it is really a bad idea to embrittle a clad. DR. OZER: current One more slide. are We think that The criteria conservative. embrittlement issue was set up some 33 years ago because at that time the concern was that we really didn't know what kind of forces would be exerted on fuel during a LOCA event. Since then, a lot of experiments have been done, both in Japan and in the U.S., that show that even zero ductility fuel has enough strength to withstand the stresses and strains that result from the quench operation as well as a wide range of impact loads that may be expected following the LOCA. feel that there is conservatism in there. We also feel that there is conservatism in trying to determine when you will lose ductility from ring-compression tests done on de-fueled cladding. We So we think that those are tests that are very localized; whereas, the response of fuel in the reactor will be more of an integral nature and will be affected by the fuel column that should be present there. So, you know, we feel that those are conservatisms that are present right now. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 63 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 We also feel that we have margin, considerable margin, today, particularly with regards to high burnup fuel. And this is from a presentation that was made by Westinghouse to the subcommittee a couple of weeks ago which shows the power levels of different fuel as a function of burnup. And we can see that the higher-burnup fuel is way down compared to fresh fuel or even once-burned fuel. What we have over on the right-hand side is the calculated response of either high-power fuel or the lower-power high burnup fuel, the temperatures that fuel will experience during a LOCA event. What we can see is that the high burnup fuel is in the 200-degree range. the limit. It's nowhere near And the only way you can get this high burnup fuel to reach the limit of temperatures is by exceeding the limit everywhere else. So, you know, if we're putting a cap on the fresh fuel, we're also de facto putting a cap on the high burnup fuel. MEMBER ARMIJO: So you are saying that the high burnup effects that are the primary issue related to the 1.2 factor occur only in fuel that cannot reach these temperatures if that's what I heard you say. DR. OZER: What I am saying is that -Reach the 1,200, can't MEMBER ARMIJO: (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 64 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 operates at that. reach 1,100. DR. OZER: That's right. There is not enough -- MEMBER ARMIJO: DR. OZER: There is not enough power in And we are the fuel to reach those temperatures. arguing about F factors that will apply for that kind of a fuel. MEMBER CORRADINI: I apologize. DR. OZER: much Why is that? I missed Why is that? Because the higher burnup fuel powers. This is in the lower reactor, the power distribution in the reactor, fresh fuel, once-burned fuel and second-burned fuel. MEMBER CORRADINI: stored energy effect? So it's strictly a It's not a decay heat effect? It's just the opposite Heat is not going to matter. then. If it's a decay heat effect, that's irrelevant. If it's a stored energy effect, that's relevant. I mean, if you're telling me it's power at the moment I have the event I essentially redistribute the stored energy, I accept that, but if it's a decay heat effect, that's not the case. DR. OZER: Well, again, the decay heat and This is the power, stored power, produce these lines. the response during a LOCA. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 65 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 the first? DR. OZER: The first one is burnup. Burnup? time. MEMBER APOSTOLAKIS: What are the axes on VICE CHAIRMAN SIEBER: heat is much shorter. Near term decay Long-term is higher. But it's all over by that MEMBER ARMIJO: MEMBER APOSTOLAKIS: DR. OZER: MR. DUNNE: Yes. This is Bert Dunne from Areva. And What you are looking at is the peaking factor. the cladding temperature transient is determined by a normalized decay heat rate times the peaking factor. So your peaking factor carries through into your decay heat as well with time, at least for the time period of a LOCA. So we find that the stuff out here in the third cycle is operating about half of the decay heat that the fresh fuel would be. MEMBER CORRADINI: energy effect? MR. DUNNE: No. I think it's mainly decay So it's mainly stored heat and partly stored energy. VICE CHAIRMAN SIEBER: But those are pretty, on that graph, that shows to me pretty, wide power deviations, which I don't recall power (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 66 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 deviations as severe as that. MEMBER ARMIJO: cycle to cycle? VICE CHAIRMAN SIEBER: No. Differences Well, it runs You mean differences from from fresh fuel to twice-burned fuel. between about 70 percent and 130 percent, as opposed to I see assemblies there running less than 50 percent. MEMBER ARMIJO: Yes. They're pretty dead. They're pretty low. VICE CHAIRMAN SIEBER: MEMBER ARMIJO: They're pretty dead, but, you know, you get a lot of burnup in one cycle nowadays, so two cycles of burnup. VICE CHAIRMAN SIEBER: Well, modern fuel designs try to flatten the core as best we can to -DR. OZER: Well, we tried to reduce leakage as well so that the high burnup assemblies will be on the periphery. MEMBER ARMIJO: But, you know, I want to make sure that everybody understands that that is what they're saying, that the temperatures achievable as a function of burnup are defined by curves like this. It may be different for BWRs and some kind of PWRs. So the real risk is limited by the achievable temperature during the LOCA. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 67 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Anatech. about it. DR. OZER: So, conservative. Exactly. we do Thank you. feel that it is again, And now we feel that there are some And additional conservatives that are being added on. that is the use of a single F factor, the requirement of assuming double-sided oxidation, not only in the balloon but everywhere throughout the rod, and, of course, the assumption that the high burnup furl will also oxidize at the limit temperature. MEMBER POWERS: Let me ask you a question In bright red, you have "Experimental evidence supports the view that embrittled material." That experimental evidence on the forces or is it on the material? And if it's on the forces, gee, I'd like to know where that information comes from because I have searched in vain for some idea of what kinds of impulses and forces you get during an ECCS recovery. DR. OZER: This is based on experiments that were done in Japan where fuel was passed through a LOCA heat-up coolant scenario and then quenched. John, would you? MR. ALVIS: Yes. This is John Alvis from The Japanese run their integral samples They hold an through a large-break LOCA heat-up. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 68 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 up. what was it? oxidation period. And then they cool down and quench. But what the Japanese do with their tests, their requirements are that they hold their tests with axial constraint. So they apply a load to their And they integral samples during the quench process. have discovered that, even with high burnup interval rodlets, that they can reach ECRs out to 20 percent without losing the coolable geometry. MEMBER ARMIJO: these minuscule? What kind of loads? Are Are they significant loads? I think they hold their -- MR. ALVIS: Five newtons? DR. BILLONE: PARTICIPANT: MR. ALVIS: Five hundred seventy. The quench assembly. Right. Their hypothesis is that the grids would lock up or the rods would lock up at the grid spans MEMBER ARMIJO: So they put these things in bending or some way that would -DR. BILLONE: DR. OZER: They hold it. Intention, intention. What they do is they heat it You know, they clamp it. And then they quench. And MEMBER ARMIJO: DR. OZER: And then they quench it. they see whether it will break or not. NEAL R. GROSS And what they (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 69 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 zero ductility? DR. OZER: MR. ALVIS: DR. OZER: ductility That's correct, yes. Correct. So that's why I'm saying even has enough strength to see is that even the 17 percent ECR fuel will not break, that you need much higher ECRs to break it. MEMBER ARMIJO: Based on strength, not on material withstand stresses resulting from quench. MEMBER ARMIJO: But you are not arguing You accept against a ductility limit, though, right? -DR. OZER: Not at this point, no. No. But I'm trying to say that there is conservatism in using ductility as a surrogate for what the fuel -you know, what we are concerned about is coolable geometry. And we're trying to make sure that the fuel will survive a LOCA event. And ductility was used as a surrogate for anything that may be happening in the reactor during a LOCA event. MEMBER POWERS: That's exactly what happened, is that nobody knew what kind of forces were going to be placed in the fuel. that you do know. This seemed to say And I'm asking, how do you know NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 70 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 get-go. NEAL R. GROSS that? DR. OZER: Yes. The only thing that we can say here is that it will withstand the quench, the stresses resulting from quench. MEMBER POWERS: It's the stresses resulting from quench in a particular experimental apparatus -DR. OZER: Yes, correct. -- with a particular kind MEMBER POWERS: of configuration. DR. OZER: Yes, with a particular load. What I'm asking about is MEMBER POWERS: now how do I take that and then imply that it's conservative in the reactor? It may be, for all I know, but I just don't know how to do that because I don't know what the forces are. DR. OZER: Again, I am only using this as an indication that there is some reason to feel that the sky is not falling exactly. MEMBER POWERS: Well, I mean, when the The original rule was developed, people said, "Yes. ductility criteria will be conservative criteria." DR. OZER: Yes. And they knew it from the MEMBER POWERS: (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 71 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 this point? DR. OZER: Yes. I'm trying to -The question we have now MEMBER POWERS: is, what do we do about all of these new fuels that are coming along? And how do we keep bad fuels that look compositionally the same as good fuels out of the system? MEMBER ARMIJO: high burnup effects? And how do we account for I think that's a fundamental issue, how much emphasis is on the high burnup effect, because that's where the F factor is and the 1.2. And that's where I think the focus of the industry issue is. And we've got to understand that. That's right. May I ask a question at So, DR. OZER: MEMBER CORRADINI: Sam, I think, characterized it. really, if I understand your original slide, you have done some calculations. And going from the notice effect, which is in the '98 notice, essentially correcting for it at a factor of one, correcting for it as a factor of 1.2 is going to cause, your point is, undue conservatisms, because already you are correcting for the high burnup using the factor of one if I understood what we were told? DR. OZER: Yes. Our concern is that a single F factor to account for all of these heat-up, NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 72 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 here. cool-down, material quench temperature, effects is material not property, fabrication going to be defensible in a licensing environment. MEMBER ARMIJO: So to repeat it differently, you would rather go on a case-by-case basis with separate fuel to the staff? DR. OZER: No. I think our argument is that, really, we are not ready to, we don't have sufficient data to defend the 1.2. MS. UHLE: Can I interrupt at this point because I think the conversation is getting a little off base in the sense that we're not talking about rule language. It may be the option that NRR decides that a licensee or a vendor getting a fuel design certified would come up with the F factor. we're getting a little off base. DR. OZER: DR. MEYER: Yes. I -So I think Could I also make a comment I'm trying to restrain myself, but for these modern alloys, the 1.2 factor has very little effect because, as you saw in the numerical example, the corrosion thickness is low. And the only time that this really is going to have a big effect is when you're dealing with one of the older claddings. There is still some in the plants, like Zircaloy, where the NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 73 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 corrosion is high. DR. OZER: And Ralph gave some nice examples, where we indeed seem to be gaining some margin. So why are we complaining? MEMBER CORRADINI: was thinking. DR. OZER: MR. DUNNE: Right. Well -That's kind of what I This is Bert Dunne from AREVA. One of the things that AREVA wants is for the criteria to be on well-established scientific grounds because we think that is the location at which we can have a long-living criteria. And what I look at is a learning curve to tell me whether or not I am on well-established scientific grounds. I think we are still learning. Two years ago we had two new effects that we needed to consider. This time we're back up here. We again have two the relatively newly discovered or realized effects: potential for quench temperature cooling rate to have an effect and the ID oxidation. So we're just kind of saying go slow if you go or we would rather have a period of time when we didn't discover a new effect tomorrow. DR. OZER: Let me mention the concerns The F factor is that we have with the F factor. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 74 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 trying to cover a lot of territory. It's going to be a function, a very complicated function, of hydrogen content, cladding design actually on time, and temperature history. We have information that the cooling rate and the temperature at which quench is introduced does have a significant effect on the F factor. Would low quench temperatures, temperatures below 600 degrees, give us a much better F factor, even an F factor less than one? You know, the impact of these variables cannot be addressed to a single factor. Plus, the F factor is really not appropriate for BWRs because F factor is a multiplier on oxide thickness. And for BWRs, really, the parameter that should be used is the hydrogen content in the cladding. There is a larger variety or uncertainty about the oxide thickness that would have to be accounted for. And this was penalized at better And this fact was performing BWR cladding alloys. recognized, in fact, by NRR in preparing the proposing interim RIA criteria, which for BWRs are based on hydrogen content, rather than oxide thickness. There are other problems as well. You know, how do you determine the F factor a priori from NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 75 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 prehydrided data? We tried to do an exercise where we took the experiments that were conducted by Argonne on unirradiated matching pairs of experiments, on unirradiated cladding material and hydrided cladding material, and tried to derive an F factor from that. And we see that the F factor is all over the map, going from almost two down to, again, less than one, .7, .8. What is interesting to note here is that when you go to slower-cooled cases, what we have here is cases that were quenched at 800 degrees because all the quenches, most of the quenches that Argonne has done are done at 800 degrees. And when you either don't cool it or cool it at lower or quench it at lower -- I'm sorry. If you quench it at lower temperatures or slow-cool without quench, you get much better F factors. We are concerned that the use of 800 degrees for quench temperature is inappropriate or it's overly conservative. Again -- and I'm basing this on this time a calculation or evaluation provided by AREVA for different scenarios. large-break LOCA scenarios. These are two And they estimate the This one is a quench to occur below 600 degrees. small-break LOCA. And the quench here is around 250 NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 76 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 that doing that? DR. OZER: I'm sorry? Currently they are doing notice. They're degrees. Now, we have similar results reported from our BWR colleagues that indicate that most of the quench they expect to be occurring around 600 degrees or less. And, again, when you quench at 600, you get So that's another uncertainty. Dr. Ozer, before you leave You a better response. MEMBER ARMIJO: the BWR situation, what is your argument on that? say the oxidation is not the right parameter to use. Why do you say that? DR. environment, you OZER: have Because in a licensing to account for all the uncertainties, the uncertainty that you will expect in predicting the oxide thickness. And the BWR people can predict the hydrogen content with less uncertainty than they can predict the oxide thickness. So if it's based on oxide thickness, they would have to take a higher penalty. MEMBER ARMIJO: Currently aren't they MEMBER ARMIJO: through the information including the oxidation, external oxidation. DR. OZER: Yes. And now we are applying (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 77 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 a question. G&F. a factor on top of that. And we are questioning the adequacy of that factor for BWRs. DR. MEYER: If you will look at my slide 27, you will see that there wouldn't be any penalty for the BWRs. MR. JAHINGIR: This is Nayem Jahingir from Just to clarify Sam's point, we have the And ductility loss is more related to And there is some indication ductility loss. hydrogen than oxidation. that at higher exposure, hydrogen uptake is much higher for like same oxidation for BWR cladding, too. That's why for RIA, we are kind of weighing to the hydrogen space, rather than an oxidation space, because that's actually more related to the ductility. MEMBER BANERJEE: I just want to ask you If you go back to the previous slide, the pre-cooling phase, before quench, is a fairly rapid cool-down anyway you can see. DR. OZER: It says here. Yes. So quench is only There's MEMBER BANERJEE: a calculation for when the surface rewets. extensive heat transfer, which brings the surface down. So why do we put so much emphasis on the quench per se, compared to a process which might be dropping the temperature fairly rapidly? NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 78 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 analyze our DR. OZER: The easy answer is that And the the experiments showed that there is an effect. effect tends to give better results when experiments are quenched at 600, as compared to 800. MEMBER BANERJEE: you cool them down? DR. BILLONE: DR. OZER: Yes, it does. It doesn't matter how I'm sure it does. May I clarify one point? In the Argonne DR. BILLONE: You're talking about CEA experiments. experiments, we found no difference between quenching at 800, 700, and 600 degrees C. And getting back to the F factor, we can data and say conservatively we want If you conservative numbers, 1.6 for the F factor. want to take into account that our experimental cooling rates are faster than what you see there and our quench temperatures are higher, then we can justify moving the F factor down. But 1.2 really applies to quench temperatures below 600 degrees C. and cooling rates 5 degrees C. per second or less on the cooling part before you get to the quench. MS. UHLE: Again, this is Jennifer Uhle. This is rule language we're talking about -NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 79 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 could be DR. BILLONE: MS. UHLE: Right. -- because it may be decided that the licensee is -- it's up to the licensee to determine the appropriate F factor. MEMBER ARMIJO: be in a NUREG somewhere. MS. UHLE: Right. So the guidance would Well, the guidance would say that this is the type of test that you need to run and here is the value you need to come up with. But then it could be such that the vendor would then be responsible for coming up with the F factor. That could be a possible approach if we're talking about -- the concern I think here is that the 1.2 doesn't apply to all different clads. MEMBER BANERJEE: I guess the question was that the cool-down rate affects this F in terms of whether it's 5 degrees per second, 10 degrees per second, or 15 degrees per second, correct? MS. UHLE: Well, Argonne has indicated that the temperatures of -- what was it? -- 800, 700, 600 didn't make -- we didn't see that much of a difference. But, again, if this is something that incorporated into the testing program associated with coming up with this F factor, if NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 80 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 M5. that's, in fact, the way NRR wants to go, then that's up for debate. And that would be discussed in the stakeholder involvement period. I think the question is whether or not the phenomena is applicable. MEMBER ARMIJO: you're right. Right. Right. I think I think how much emphasis you put on ID oxidation due to bonded fuel, you know, how much of an effect that is, the effect of hydrogen and the -those are the fundamental issues. And you're still arguing how important those things are. DR. MEYER: This is Ralph Meyer. With regard to the F factor, keep in mind that there is only one set of data in the world. The industry doesn't have another set of data with high burnup fuel rods than those one. So, you know, you can speculate about how many variables are involved, but it's very tough to go out and measure it for another cladding type when you don't have the data. MEMBER ARMIJO: But we will have the data in a year or so, won't we, if you get your program going? You know, you get your new hotcell access. DR. BILLONE: Yes. You will have data for And you will have data for high burnup ZIRLO. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 81 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 again. MEMBER ARMIJO: I mean, you've got -DR. BILLONE: hotcells for that. now. MR. DUNNE: Mike, this is Bert Dunne Yes. We don't need new And that's committed R&D? That's what we're working on right You're talking about the Skuzda examples now? DR. BILLONE: MR. DUNNE: Right. What we really want to do is wait for the Oak Ridge program, where we're talking about fuel that -- cladding that has fuel inferior to it so we can learn something about the ID oxygen source and the relative merits of testing irradiated fuel with simulated cladding that's been preloaded with hydrogen. DR. BILLONE: That's correct. That will be F.Y. 2008 for the fuel tests, but for the cladding tests with the modern alloys -MR. DUNNE: Still within the time frame that was just mentioned of a couple of years, I hope, if we could stay on schedule. MEMBER ARMIJO: you finished? DR. OZER: Let me just say a few words Okay. Well, Dr. Ozer, are about our concern about the oxygen pickup on the ID. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 82 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 on PWR fuel. BWR fuel. We are not disputing that that may occur. It's just the assumption, again, the recommendation that we assume. We account for it by assuming double-sided oxidation. We don't -- you know, for this to occur, you have to have strong, either very strong, contact or bonding. And we think that the results, the This can only experimental results, are inconclusive. be or can best be demonstrated from integral tests. So far there have been no integral tests The only integral tests we have are on And those are -- you know, I'm taking this And this is cladding that graph from the draft NUREG. has been irradiated. 52. The burnup of the fuel rod was We estimate that at this elevation, the burnup here is 57, where bonding should have been rather significant. We see a clear alpha layer on the outside. On the inside, there are some regions where it is said there is no alpha and other regions where it is said there is alpha. I think one has to be really quite a metallurgy expert to differentiate any kind of an alpha layer here, much less differences between A and B. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 83 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 mean, isn't So we feel that this is a question that has to be resolved with additional experiments. And there have been some statements made that additional experiments, integral experiments, will not be available for years. MEMBER ARMIJO: But you know that in high burnup BWR fuel, there is fuel clad bonding. DR. OZER: That's right. You've seen it. It's not MEMBER ARMIJO: 100 percent uniform. and some clad designs. It has higher And it's a function of burnup Is it the same in PWR fuel? pressure, maybe tighter external contact. I don't know. DR. OZER: MEMBER We don't know. I think that can be Argonne has ARMIJO: explained by doing integral experiments. -MEMBER POWERS: this just a Can we do this with -- I matter of looking at irradiated fuel? I mean, I'm trying to think about I don't think you how you would do it experimentally. can do a persuasive experiment here. DR. OZER: You would have to run it through a local scenario, the heat-up scenario, to see how. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 84 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 know that. proposed. MEMBER POWERS: DR. OZER: fact that it occurs. I thought that -- Again, we're not disputing the It's just how to account for it And also we are concerned and how important is it. that if we assume double-sided oxidation, this may be interpreted, assuming you're calculated double-sided oxidation. And we may have to take, may be required to take, into consideration the energy of oxidation, which at high temperatures could be quite significant and would result in a penalty in the -DR. BILLONE: No, no. That was never You're not forming any oxide in this event on the idea of the -DR. OZER: about that because -DR. BILLONE: No. We were very clear I think you have to be clear about what you use for the -MEMBER POWERS: You're heat of dissolution is going to be so close to the heat of oxidation that I don't think you have gained anything here. DR. BILLONE: Very clever. MEMBER POWERS: I mean, it's hard to Very clever, actually. You imagine how you would keep up here on the inner surface uniformly. I think you get mass transport NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 85 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 program. do we. year. MEMBER ARMIJO: confirmatory research? DR. OZER: NRC participates in the Halden And so That is part of the NRC limited on the oxidation. DR. OZER: As far as experimental data is concerned, I would like to say that we don't think that Argonne is an island of information in itself. I think Argonne and has has worked closely a with lot other from laboratories benefitted interactions with other labs in trying to resolve discrepancies. is ongoing. And I think probably the most relevant work is being done at the Halden Lab, where, indeed, high burnup fuel rods are being subjected to LOCA-like scenarios in reactors. So, you know, these questions about And I think work at these other labs heat-up, heating up from the inside, as opposed to heating up from the outside, you know, the Halden results will not be as sensitive a results being done in laboratories like ANL. These results are expected later this I mean, they send representatives. But, you know, it's a Halden program. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 86 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 message. DR. OZER: -- with respect to that it's rupture. DR. OZER: Yes. And so if there was project. DR. MEYER: Yes. We're in the Halden They're not doing any embrittlement They're looking at ballooning and axial measurements. fuel relocation. DR. OZER: And they should be able to provide us with metallurgy information about away from the balloon, extent of oxidation. MEMBER ARMIJO: But they will take it to MEMBER ARMIJO: oxidation from the ID when they do their metallography, they should confirm or correct -DR. MEYER: DR. OZER: go into this. Yes, this is true. I don't know whether I should I think we're out of time, but -ARMIJO: I think we got your MEMBER going to be quite costly for the industry to implement this, in conclusion, again, we don't feel that there is a public safety, urgent public safety, issue at this point. And we feel that the bonding approach that is being proposed is premature. MEMBER ARMIJO: Well, you know I have a (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 87 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 dilemma. And I'm just going to make my little speech. industry and the staff. And I And that's to both parties: I hear that there is no urgency. hear that there is urgency. I hear that there is Yet, But if going to be a big impact to the industry. Ralph's calculations show there is no impact. there is no impact, why is there urgency? So I can't get around all of these claims. MS. UHLE: The urgency primarily stems from what is required in the regulation and what is voluntarily done by the licensees. break-away regulation. If a new cladding were to be submitted for approval, there is nothing in the regs that would require any concern about the break-away oxidation. Yet, you can see with the fabrication process of the E110 that that was a strong effect. MEMBER ARMIJO: Okay. So -oxidation metric, that For instance, the is not in the Right now you have no guidance or no regulations that require the -MS. UHLE: Break-away. -- suppliers to even think MEMBER ARMIJO: about break-away. MS. UHLE: MEMBER That's right. Okay. So that's a ARMIJO: (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 88 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 that. MS. UHLE: Well, again, that would come in That could be deficiency. MS. UHLE: And so when we say there is no urgency, that is because we have talked to NRR about this. And NRR has gone out voluntarily and done a Is there a safety issue spot check to see "Okay. looking at how the licensees are currently operating, voluntarily operating that way?" That doesn't preclude them from changing the way they operate. So with that, we can say -MEMBER ARMIJO: They're not likely to do from introduction of a new clad design. a change in the way they operate within the regs as written. And they are free to do so. They don't have to tell us exactly what they're doing on a day-to-day basis. DR. OZER: through a reg guide? MS. UHLE: There is no regulatory But couldn't that be addressed requirement that would force anybody to take this into consideration. MEMBER ARMIJO: Unless there are some other questions, I think I am probably way out of time, Mr. Chairman. And I would like to end this part NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 89 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 8) into break. of the session. Okay. Thanks, everybody. Okay. I think we got the issues on the table. yours. Dr. Shack, it's all CHAIRMAN SHACK: Yes. It's time for a We would like to make it a short break since If we could come back in we are a little bit behind. ten minutes? (Whereupon, the foregoing matter went off the record at 10:29 a.m. and went back on the record at 10:43 a.m.) CHAIRMAN SHACK: session, everybody. I would like to come back Yesterday I read the qualifications and experience of our new senior staff engineer, Ms. Zena Abdually. in the Committee's And she will be helpful of power uprate review applications, thermal hydraulic issues, and TWR sump performance issues. What I neglected to do yesterday was to welcome her aboard. today. MS. ABDUALLY: (Applause.) DRAFT FINAL REVISION 1 TO REG GUIDE 1.189 Thank you. And I would like to do that (DG-1170), "FIRE PROTECTION FOR NUCLEAR POWER NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 90 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 PLANTS," AND SRP SECTION 9.5.1,"FIRE PROTECTION PROGRAM" CHAIRMAN SHACK: Next up on our agenda is a presentation on reg guide for fire protection and the SRP 9.5.1. And we'll be lead through that by Jack Sieber, our Fire Protection Subcommittee Chairman. VICE CHAIRMAN SIEBER: much, Mr. Chairman. 8.1) REMARKS BY THE SUBCOMMITTEE CHAIRMAN VICE CHAIRMAN SIEBER: this is a major Thank you very effort by the staff, the revision of reg guide 1.189, which was draft guide 1170 in its earlier days. It is sort of a companion to the 805 risk-informed fire protection effort. And the purpose of reissuing this, among others, is to consolidate all the references, of which there are over 100, to preexisting documents and consolidate those into a document that is easier to read and easier to follow. not introduce or break The latest document does new ground in the fire protection area, but it is more a consolidation. We mentioned SRP section 9.5.1. That has now been incorporated into the draft reg guide, which we're reviewing. And so because of that consolidation, we need not conduct a review of a NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 91 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 slide. VICE CHAIRMAN SIEBER: Yes. Just so the pretty well. VICE CHAIRMAN SIEBER: MR. RADLINSKI: Okay. 8.2) separate document. What I would like to do now is introduce Cornelius Holden, who is responsible for the overall effort in this guide, to introduce to us the staff personnel who worked on this and are responsible for it. MR. HOLDEN: Thank you very much. BRIEFING BY AND DISCUSSIONS WITH REPRESENTATIVES OF THE NRC STAFF MR. HOLDEN: I am Cornelius Holden, With me today is Division Director, Risk Assessment. Sunil Weerakkody, who is our Branch Chief for Fire Protection; and Bob Radlinski, who is our senior person on this effort, will be conducting the briefing today. With that, Bob? MR. RADLINSKI: Good morning. everybody. Dr. Sieber, you covered my introduction So move on to the next Committee recognizes it, a couple of months ago, we wrote a letter on this draft guide for public comment, suggesting that the staff issue it. NEAL R. GROSS And now the (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 92 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 comment. MEMBER APOSTOLAKIS: VICE CHAIRMAN SIEBER: lot of comments that came back. comments. MEMBER APOSTOLAKIS: make comments today? VICE CHAIRMAN SIEBER: them even better two months ago. (Laughter.) VICE CHAIRMAN SIEBER: MR. RADLINSKI: Okay. Okay. The objective, as You could have made So we are allowed to Comment on -Well, there were a We did not have any comments are back and we're revisiting the subject again. MEMBER APOSTOLAKIS: said to issue it? VICE CHAIRMAN SIEBER: Issue it for public So our letter just we have mentioned, is that we are going to describe how the NRC staff addressed the public comments that were received on the reg guide and also, of course, to obtain ACRS permission to issue the reg guide. Just to summarize the comments and the responses, the NRC received 95 what are called new comments on the draft guide. were from NEI. All of those comments The reason I say "new comments" is NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 93 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 because they also included 16 additional selected comments that were made on the previous draft guide of the original version of draft reg guide 1.189 when it was issued the first time. There constructive. were excellent comments, very We incorporated or agreed with 67 of It's over 70 percent. And the final the 95 comments. draft will reflect those comments. Also, earlier this week we had a public meeting to summarize what our resolution was of those comments, an opportunity for additional discussion. And that went very well. Also, in the interest of time, my presentation today is only going to talk about the comments that we did not agree with and/or significant issues. MEMBER APOSTOLAKIS: Can you give us some I idea of what kinds of comments you agreed with? mean, were they editorial or substantive or -MR. RADLINSKI: Combination, nothing that It just added would change positions or anything. clarifications. They were very helpful in identifying areas where we may have assumed or we had thought that the regulatory requirements were clear. because of the comment, they were not. NEAL R. GROSS But obviously So we added (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 94 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 by the way? MR. RADLINSKI: of discussion about this. There has been quite a bit With all of the actions and in generic some additional clarification. Okay. There were basically seven categories of comments that we did not agree with. The first was that some of the guidance in the revised reg guide is a backfit. Now, the second is that we should not issue the reg guide at this time because of the comments that the Commission had with respect to the generic letter that we recently submitted for publication on spurious actuations. The third is that we should endorse industry standards in lieu of issuing the reg guide. Next is that the guidance that is provided letter 81-12 should be applicable to III.G.2 areas, the appendix R III.G.2, as well as III.G.3 areas. Of course, I'll be getting into more detail in each of these issues. The next one is that detection and suppression are not necessarily required with operator manual actions when they are accredited for a III.G.2 area. MEMBER APOSTOLAKIS: What does that mean, (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 95 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 discussions that have gone on over accrediting of operator manual actions for III.G.2 areas, we issued a RIS, 2006-10, that talked about this. The industry contends that if they are able to credit an operator manual action in lieu of the protection requirements of III.G.2, then detection/suppression, which is generally required by III.G.2 or portions of III.G.2 and III.G.3, are not necessarily part of that design. necessarily be required. VICE CHAIRMAN SIEBER: It has been the They would not staff's position that they are required. MR. RADLINSKI: Yes. We have been Item 6, some steadfast in that position. Let's see. of the new reactor guidance that we have added to the reg guide. Actually, the reg guide did not have any new reactor guidance in it before. A lot of it was -- there had been some in the previous version of the SRP. And we rolled that over into the reg guide and The comment is that is not a specific also added some new guidance. some of that new guidance requirement of the regulation. And, finally, I think we mentioned this in our last meeting before we sent the reg guide out, NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 96 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 that we would like to revert to 50.59 as a basis for evaluating plant changes for fire protection. Okay. The details. On the backfit, again, the gist of the comment was that some of the new and revised guidance in the draft guide would be a backfit for existing plants. We went back to the process with the original issuance of reg guide 1.189. the CRGR meeting minutes. reviewed that document. We looked at And the full Committee And they reach a conclusion that it was not a backfit, that a backfit analysis was not required. That was essentially based on the fact that compliance with the reg guide is not required, it's not imposed compliance, and that compliance should be assessed against a plant-specific licensing basis, not against the reg guide. And licensees performing their own self-assessments should also do those assessments against their licensing basis and not the reg guide. Although we added some guidance and changed some of the existing guidance in the original version, the same basis for a no-backfit conclusion would also apply to the current revisions of the reg guide. And, in addition to that, we did review NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 97 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 the latest version with the CRGR chairman. It was agreed that the update is likewise not a backfit and does not require a backfit analysis. MEMBER MAYNARD: Just from a practical standpoint, I understand your bullets, your points there relative to the backfit. But that would almost be saying that the reg guide basically is not going to be used in any assessment or evaluation or anything. It's saying that compliance is going to be based against the licensing basis, not the reg guide. So what's the purpose of the reg guide if it's not going to be used in any assessment? MR. RADLINSKI: Okay. Of course, any reg To guide is one acceptable approach to regulations. my mind, it will be used as a baseline for a licensee who has a configuration that isn't addressed in his plant licensing basis, isn't addressed even in the regulations. This would be the baseline for an inspector to say, "Okay. will work. This is one approach that This is one approach that would be acceptable to the staff for meeting the regulations in general. If you are not doing it this way, then you can propose something else and explain to us why that's acceptable and why it meets the regulations." NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 98 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 So this is kind of a starting point. It gives the inspectors, gives the licensees a baseline for what would be considered by the staff to be an acceptable approach. used. MEMBER MAYNARD: So if they propose an So that's kind of how it will be alternative approach, you would be assessing that against the licensing basis, not starting with the reg guide 1.189 as a minimum level of effort? VICE CHAIRMAN SIEBER: MR. RADLINSKI: VICE CHAIRMAN Correct. Correct. SIEBER: In fact, the starting point, reg guides are issued to licensees as well as internal use by the agency. And my experience in licensing is that's where you go first because it's the easiest amount of work. If you do the things in the reg guide, then you don't have to come up with an alternative solution. If you can't do them because of configuration in your plant or you have a better idea, that becomes an exception which you identified to an inspector when he comes to inspect you for compliance. MEMBER MAYNARD: But it's also been my experience that these tend to become more or less minimum acceptable requirements. NEAL R. GROSS You may propose (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 99 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 topic: alternatives, but a reg guide does set kind of a threshold level there. I understand your point. It still can't be waived MR. RADLINSKI: by the inspector saying, "You're not complying with this." Okay? It's not a basis for compliance. CHAIRMAN SIEBER: There is a VICE statement right in the preamble to the reg guide that explains what its legal purpose is. CHAIRMAN SHACK: legal purpose. VICE CHAIRMAN SIEBER: Yes. Well, I mean, You all sort of know the if you want to interpret it in your own way, that's up to you, but I read what is written down. MEMBER MAYNARD: There's also a practical So I side of how this actually gets implemented. think we need to move on. VICE CHAIRMAN SIEBER: MR. RADLINSKI: Okay. Right. Another favorite You are multiple spurious actuations. probably all familiar with the generic letter that was prepared on this issue, particularly with the respect to the approach of one at a time is an assumption that would provide a basis for post-fire safe shutdown circuit analyses. The comment was that you shouldn't be NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 issuing this reg guide because of Commission comments on that generic letter. Our response is that we did revise the wording in the draft guide, which was from the public comment version, that we did basically water down the language. We are now no providing any specific guidance on what approach to use for circuit analyses with respect to one at a time. However, we do include a note and continue to include that that based on the industry cable fire tests, a one-at-a-time assumption for spurious actuations may not adequately address the potential risks due to fire, so just kind of a flag to licensees that there may be a problem if you use that assumption as a basis for your circuit analysis. We also note or the Commission comments on the generic letter based on our changes that we made to the design guide that really don't warrant not issuing the reg guide. It's one issue. And we have kind of watered it down or softened it quite a bit. VICE CHAIRMAN SIEBER: So the generic letter now still rests as a draft and the issue is still out there. If the Commission changes its mind about the staff's approach to the generic letter, would that warrant the change to this reg guide? MR. RADLINSKI: A future revision to the (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 101 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 that? (No response.) MR. RADLINSKI: Okay. One of the comments tuned? MR. RADLINSKI: Yes, yes. Okay. Thank you. reg guide probably will incorporate the guidance that we plan to put into the ultimate generic letter that is issued. One of the main comments the Commission had was that we basically said, "Hey, industry, you have a problem," but we didn't tell them how to fix it. VICE CHAIRMAN SIEBER: MR. RADLINSKI: Okay. And they want us to work with the industry to come up with the methodology and the acceptance criteria to address the potential problem. VICE CHAIRMAN SIEBER: So we should stay VICE CHAIRMAN SIEBER: MR. RADLINSKI: Any more questions on referred to some public law that basically said that the government agencies should use industry consensus standards if they were available as a replacement for things like reg guides. They specifically mentioned NFPA 804. NEAL R. GROSS For (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 102 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 the coming out those of you not familiar with 804, it is strictly for new reactors. And it is a deterministic approach to Okay? another for new version, reactors, 806, which that's is a fire protection. There's that's risk-informed, performance-based approach. And I might note that AP-1000, also the SBWR have referred to 804. Now, 804, like any other NFPA standard, is, you know, an appropriate standard to be referred to and provide guidance for the design of the fire protection program, however porous. in accordance with the regulations. Also, 804 was just reissued, revised and reissued, in 2006. I think the first version was And there were It must be done 2001, but that was a previous version. a lot of changes. And we're reviewing it, but we haven't completed our review yet. And by issuing reg guide 1.189 now that does not preclude 804, you're -CHAIRMAN SHACK: question of what would I was sort of left with a new plant use for a possible future endorsement of guidance. MR. RADLINSKI: For performance-based? (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 103 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEI-0001, reg guide? CHAIRMAN SHACK: MR. RADLINSKI: The reg guide, yes. The reg guide is fine. CHAIRMAN SHACK: and even deterministic. MR. RADLINSKI: say, they haven't -CHAIRMAN SHACK: This is complete enough? Yes. Well, deterministic, as I Well, performance-based VICE CHAIRMAN SIEBER: MR. RADLINSKI: When you say, "this," the And, like I say, they are also referring to 804, just like they would refer to NFP 13 for sprinkler systems and 15 for water spray, it provides some additional guidance. But I would also like to point out that there are some things in 804 that we don't agree with that we don't consider them to be meeting the regulatory requirements. The comments in this regard also mentioned which is the industry guidance for performing post-RSA shutdown analyses. really a consensus standard. That's not And also we have already not endorsed necessarily, but we have provided statements of staff acceptance of NEI-001 in a RIS and also in reg guide NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 104 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Weerakkody. have -MR. WEERAKKODY: Yes. This is Sunil this? good point. 1.205 for the -CHAIRMAN SHACK: On the performance base, they still have the problem, right, because that guidance is not out yet for new reactors? MR. RADLINSKI: For new, yes, that's a There is an 806 coming out, NFPA 806. That will be the industry consensus standard for new reactors using a performance-based environment. CHAIRMAN SHACK: Now, will you have a very high priority on reviewing that when you -MR. RADLINSKI: We are. We have already submitted two sets of comments. We reviewed it in great detail, submitted a lot of comments the first time around. Most of those were incorporated. Maybe 80 percent were incorporated. the final review by the staff. It's back now again for MEMBER APOSTOLAKIS: When will we see Is the ACRS going to see that, 806? MR. RADLINSKI: Sunil, I don't know if you We have no plans to bring 806 to SRS on It's still in the works. It's still what? this unless you request. MEMBER APOSTOLAKIS: MR. WEERAKKODY: It's still being (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 105 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 delivered by the Code Committee. MEMBER APOSTOLAKIS: I understand. But at some point, you will have to issue a regulatory guide, whether you agree or not, and accept 806 with exemptions and so on. stage? And we get involved at that MR. WEERAKKODY: We don't plan to because we believe what's in the updated reg guide that you see today, which incorporates the high-level guidance on new reactors is sufficient. Now, in our review process, what we are trying to do is make sure that 806 or 805 is in that plan. at -CHAIRMAN SHACK: MR. WEERAKKODY: From performance-based? Right now we don't have So we have no initiative to endorse 804 or 806 a plan to go in and endorse 806 for new reactors. CHAIRMAN SHACK: So you're saying you don't plan to have guidance for a performance-based approach for new reactors? MR. WEERAKKODY: think that's necessary. At this point we don't I think if That's correct. you look at the advanced rectors, if you look at the advanced rectors, what we have really Okay? done is risk-informed the design itself. NEAL R. GROSS You don't (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 106 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 a need. Now, if advanced need a deterministic indicator. every area fully supported. MEMBER APOSTOLAKIS: reactors on slide 11. We'll come to the And I had some You basically have problems with the appendix. Whatever it is. Is it an appendix? But Yes, I think it's an appendix. I'm a bit surprised. B. I mean, this is, yes, appendix Don't we typically, I mean, following this public law, look at these industry standards and then express a view as to how much of those standards is applicable? You will do that sometime in the -MR. WEERAKKODY: Yes, if there should be That's what I'm saying, Dr. Shack, saying. there is a need, if why somebody don't you said, any stakeholder said, "Look, consider endorsing 806 in the rule," we definitely would look at it at that state. Okay? But if nobody wants it, why would we want to spend the time? But in the meantime, though, like But Bob said, we are very closely wording the review. I have two people in my staff in that code committee. MEMBER APOSTOLAKIS: of timing and need. MR. WEERAKKODY: Yes, sir. It is just a matter (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 107 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 MR. RADLINSKI: We will look at it in the I mean, they future, but it's just not time yet. haven't even issued 806 yet. MEMBER APOSTOLAKIS: I understand. The reason for VICE CHAIRMAN SIEBER: issuing this reg guide as a draft at this time is because of the potential for new reactors. And this goes along with a whole suite that the staff has been working on the last few months. MR. RADLINSKI: have fire PRAs. VICE CHAIRMAN SIEBER: MEMBER APOSTOLAKIS: Right. Well, I don't know And the new reactors will about that, but I am waiting until your slide 11. MR. RADLINSKI: MEMBER All right. See how well I APOSTOLAKIS: control myself, Bob. VICE CHAIRMAN SIEBER: MEMBER APOSTOLAKIS: MR. RADLINSKI: Okay. Moving right along. Discipline. We are on slide 8 The now, generic letter 81-12 and appendix RIII.G.2. comment was that the guidance in generic letter 81-12, which has a very general title of "Fire Protection Rule," should apply to appendix R, section III.G.2 areas as well as III.G.3 areas. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 108 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Our response to this is that this was covered extensively in a RIS that we published in 2005. It's RIS 2005-30. It addresses the issue. That RIS was issued for public comment because of the controversial nature of it. We received public comments, numerous public comments. meeting to We even had a follow-up public each and every one of those address comments. backfit. That RIS was also reviewed by CRGR for And it was issued final in December, on December 20th in 2005. And essentially what it says is that generic letter 81-12, the guidance provided by the generic letter. And there was a follow-up memorandum that provided additional guidance. All of that is clearly applicable to alternative dedicated shutdown capability and not to III.G.2. I mean, it's related to the III.G.2 indirectly in the sense that some of these associated circuits of concern could cause damage or prevent a redundant train from shutting down the plant. But other than that, I think the industry -- they haven't said it specifically, but I think they're focused on the fact that one of the mitigating components or one of the options for mitigation of a NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 109 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 spurious actuation of an associated circuit of concern is an operator manual action. apply that to III.G.2. case. Okay. Operator manual actions and As I mentioned before, the And they would like to Of course, that's not the detection/suppression. comment is that we have inappropriately implied that if you credit an operator manual action in the III.G.2 area, then you don't necessarily have to provide detection and suppression. a baseline, it to should the be That may be true, but as assumed protection that that that is fundamental fire you're providing in that area. As we all know, there are three components to fire protection defense-in-depth. fire. You prevent the And If you do have a fire, then you detect it. you suppress it. And then, finally, you assure safe shutdown in the event of that fire. Operator manual actions typically support the third component. the electrical They serve as a substitute for fire barrier system or raceway separation. They do not eliminate the need for the Okay? They also support it, other components. MEMBER APOSTOLAKIS: though, don't they? NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 110 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 take that -MEMBER APOSTOLAKIS: MR. RADLINSKI: approach, then Okay. I The industry -If I do that, if I have to have sub-bullet? MR. RADLINSKI: They support? The operators detect MEMBER APOSTOLAKIS: the fire and alert the fire brigade. MR. RADLINSKI: MEMBER Right. Is it just the APOSTOLAKIS: When we say, "Operator manual actions," I guess we mean a specific set of manual actions." MR. RADLINSKI: Well, we're talking about We have a situation where you have a III.G.2 area. redundant trains in the same fire area. And you have removed your electrical raceway fire barrier system, thermal lag, or whatever and you have replaced it with an operator manual action to mitigate the failure of that circuit that's no longer protected. MEMBER APOSTOLAKIS: So it's that specific set of manual actions that OMA refers to? MR. RADLINSKI: Right. And the industry don't detection suppression in the area of consideration. MEMBER MAYNARD: If you have detection and suppression, why do you have to rely on operator actions? NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 111 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 operations. MR. RADLINSKI: -- spurious actuations separation barrier. VICE CHAIRMAN SIEBER: can There's a be achieved by MR. RADLINSKI: VICE CHAIRMAN Defense-in-depth. SIEBER: Detection and suppression are not a substitute for the fire barrier. MEMBER APOSTOLAKIS: It's for the fire requirement, which barriers or distance. MEMBER APOSTOLAKIS: Because I guess some of the plants could not meet the appendix R separation criteria or is that the idea? MR. RADLINSKI: In a number of cases, it was because of the thermal lag issue, where they just took the thermal lag off or just didn't credit it any longer and said, "Okay. We'll assume that that cable Okay? Since that cable tray is going to burn up." trap is going to burn up, I'm going to have to take some operator manual action to mitigate the -VICE CHAIRMAN SIEBER: Spurious that could prevent safe shutdown. MEMBER BANERJEE: But the new plants will be able to meet the separational requirements. MEMBER APOSTOLAKIS: NEAL R. GROSS They should. (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 112 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 them, right. reactor. MEMBER APOSTOLAKIS: VICE CHAIRMAN SIEBER: MEMBER APOSTOLAKIS: us that the -VICE CHAIRMAN SIEBER: Moving on. NEAL R. GROSS You have to add -- diverse vessels. Moving on. Dr. Kress will tell reactors. MEMBER APOSTOLAKIS: With different possible. MR. RADLINSKI: enhanced fire protection. later. Yes. That's part of the We'll talk about that But they won't be able to do that 100 percent It's not -Well, why not? It's just not physically on the cases. MEMBER BANERJEE: MR. RADLINSKI: I mean, you just have areas of the plant where things come together. VICE CHAIRMAN SIEBER: room cable spreading. MR. RADLINSKI: Right, obviously the Like the control control room but under the reactor vessel and areas like that. VICE CHAIRMAN SIEBER: You have two manufacturers for the -VICE CHAIRMAN SIEBER: A reactor and B (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 113 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 MR. RADLINSKI: Okay. The main control room complex fire protection. The comment was that "The following should be deleted from the guidance that's in the reg guide." One is to provide suppression for peripheral rooms that are adjacent to the main control room. The other is that the industry does not believe that smoke detection in the individual cabinets within the main control room is necessary. First of all, the auto suppression in the peripheral rooms may be required by appendix R, section III.G.3. a III.G.3 area. the -VICE CHAIRMAN SIEBER: strength of the barrier. MR. RADLINSKI: Right. That's a natural It depends on the Okay? Obviously the control room is You have alternative shutdown and in VICE CHAIRMAN SIEBER: place for a fire, computer rooms, offices. MR. RADLINSKI: Right, offices with paper. Are these suppression MEMBER BANERJEE: systems primarily sort of rapid system? MR. RADLINSKI: MEMBER BANERJEE: MR. RADLINSKI: No, no. Water. Water sprays? And Just like an office. (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 114 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 lot. NEAL R. GROSS with respect to cabinet detectors, they made the argument that control rooms continually man. operators are there. The But products of combustion detectors, inside cabinets may detect the fire more quickly than an operator's eyes or nose since they're detecting visible products of combustion. But, more importantly, the detectors in the cabinets tell you exactly where that fire is. Okay? If you're an operator and you smell smoke or the ceiling detectors set off the alarm, you may not know where that fire is. You may have to go around opening cabinet doors to try to find it. MEMBER ARMIJO: What's the logic for saying, "Don't do that"? -- I mean, why would they say VICE CHAIRMAN SIEBER: MR. RADLINSKI: MEMBER ARMIJO: MR. RADLINSKI: It costs money. It costs money and -It can't cost that much. To be honest, the NRC has allowed them to not do that in a number of cases. They've submitted exemption requests. approved them. MEMBER ARMIJO: It can't cost that much. Everything costs a And we have VICE CHAIRMAN SIEBER: (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 115 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 on page 10? MR. RADLINSKI: But, interestingly enough, And and the 804 actually required cabinet detection. that's the industry standard that they would like to adopt. MEMBER APOSTOLAKIS: Is that the standard that we have approved, the agency has approved? MR. RADLINSKI: No. No. No? Now we're on -Is there an error VICE CHAIRMAN SIEBER: MEMBER APOSTOLAKIS: MR. RADLINSKI: Okay. VICE CHAIRMAN SIEBER: Is there an error? On page 10, is there an error on that slide? MR. RADLINSKI: Did I mention that? Oh, yes. On the handout? For some reason, the handout It's correct in mine. It didn't get the correction. should be III.G.3, not III.G.2. VICE CHAIRMAN SIEBER: members who would make that -MR. RADLINSKI: that to my attention. All right. The comment with respect to Thank you for bringing Okay. So for the new reactors, one of the comments, was that the guidance that we've added is not specifically required by the regulations. And specifically we made comments NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 116 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 guidance. goals for to the effect that new reactors should have minimal reliance on operator manual actions and alternative shutdown and also that operator manual actions should be avoided. Furthermore, we said that reliance on electrical raceway fire barrier systems should be minimized. They objected to the use of these terms. And the comment was that those terms and that guidance is not in the regulations anywhere. This is guidance. The reg guide provides And these are considered to be appropriate new plants, where the fire protection protection program can be integrated into the planning and design phase of the plant. Furthermore, it supports the Commission's concept of enhanced fire protection for new reactors, although, again, it's not in the words or the description of the enhanced fire protection. also consistent with GD-C3. MEMBER APOSTOLAKIS: Yes. But it's The issue of new reactors in fire protection, all the risk-informed initiatives we have undertaken the last eight, nine years have been voluntary. And the argument has been, you know, we have already licensed the existing reactors using NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 117 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 here to separate criteria. So we can't really go back and So they're impose now that they become risk-informed. voluntary. And that has led to situations where we are really dancing around an issue. you do that. new reactors, If you do this, But for they be If you do this, you do that. why don't we demand that risk-informed? In other words, it seems to me that there is a general consensus that NFPA 805 is a good thing to have. And we like plants to follow NFPA 805, assess the risk. And then if they want to change later, you know, they can do a risk evaluation and go to the regulatory guide and so on and so on because it gives an integrated view of the plant. Why can't we say that new reactors should follow the NFPA 805? MR. RADLINSKI: hear you say I wish Ray Galucci were I'm sure he would that. appreciate it. MEMBER APOSTOLAKIS: Is there anything in the regulations that forbids that? VICE CHAIRMAN SIEBER: do it by rulemaking -MR. RADLINSKI: Right. I think you have to (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 118 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 VICE CHAIRMAN SIEBER: -- if you want to impose it as an absolute requirement. MR. RADLINSKI: argument, I mean. Okay. That's a legalistic Then you go to appendix B, which And you refers to fire probablistic risk assessments. see things like a detailed fire PRA is not necessarily required for a new reactor fire protection program. And then later on it says, however, if an applicant for a combined operating licenses references a certified design and if that certified design developed a fire PRA, then we impose additional requirements that the PRA has to be reviewed, right, and all that stuff, which I don't see here right now. But, I mean, we put all these "ifs." And we rely on the good will of the applicant to do the PRA. So if somebody doesn't do a fire PRA, then they don't have to do all these things and they go back to being deterministic and all of that. In other words, we are perpetuating this situation of having two parallel regulatory systems, I mean. And at the same time, we see major utilities right now switching to NFPA 805 because they believe it's to their advantage. MR. RADLINSKI: Right. Why have all of these MEMBER APOSTOLAKIS: NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 119 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 ifs that a detailed fire PRA is not necessarily required but if their certified design developed a fire PRA -VICE CHAIRMAN SIEBER: Where are you reading from, George? MEMBER APOSTOLAKIS: MR. RADLINSKI: Appendix B of this. I believe the "ifs" are there because we don't have the regulatory rule in place for that. But it's very important to note that AP1000 and ESBWR, both DCDs, both have fire PRAs, -MEMBER APOSTOLAKIS: MR. RADLINSKI: Yes, but where -- -- which means that the COL applicants must adopt that fire PRA and maintain it. MEMBER APOSTOLAKIS: Yes. It says, "Then the COL applicant is to use that PRA and update it to reflect site and plant-specific information that may not have been available at the design stage. addition, the licensee that has a In risk-informed performance-based FPP similar to NFPA 805 or that plans to evaluate plant changes using a risk-informed approach must have a detailed fire PRA." And you look at all of this and say, "Well, gee, they're asking me to do all of these things if there is a fire PRA in the certified design. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 120 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 reactors. regulate? MEMBER BANERJEE: Well, you can not PRA? MEMBER BANERJEE: yet, but I presumed it would. MEMBER APOSTOLAKIS: Is that a good way to Well, it hasn't come right? MEMBER APOSTOLAKIS: Does EPR have a fire And if there isn't, then what do I do? appendix R?" MR. RADLINSKI: The reality of the I go back to situation is that you are going to get what you want. They do have fire PRAs. MEMBER APOSTOLAKIS: But how do you know that in the future they will also have fire PRAs? MEMBER BANERJEE: Well, it will be EPR, reference the design if you don't want to reference. MEMBER APOSTOLAKIS: It seems to me, though, that the NFPA 805 appears to be the way to go. MR. RADLINSKI: But it's not for new It's specifically for -MEMBER APOSTOLAKIS: No. It's doesn't say anything, right? MR. I mean, there is -RADLINSKI: No, no. It says specifically for existing operating reactors. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 121 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 upset. MEMBER APOSTOLAKIS: upset anybody. business. VICE CHAIRMAN SIEBER: MR. WEERAKKODY: Yes. Pretty easygoing. I can't give you a Yes. It wouldn't wait. rulemaking. MEMBER APOSTOLAKIS: major problem. MR. RADLINSKI: Then we can do that. VICE CHAIRMAN SIEBER: need an SMR to start one. MEMBER APOSTOLAKIS: Well, but then, Two years. You It takes two years to And that's such a MEMBER APOSTOLAKIS: Well, but since we like it for existing reactors, why don't we like it for future reactors? MR. RADLINSKI: We do. Yes. All you need is a MEMBER APOSTOLAKIS: VICE CHAIRMAN SIEBER: again, it seems that two designs we have certified already have a fire PRA that wouldn't upset anybody because -VICE CHAIRMAN SIEBER: They wouldn't be And it would be the good way of doing complete full answer on this issue, but I know I NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 122 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 what we reactors. issue? MR. WEERAKKODY: Yes. MEMBER APOSTOLAKIS: MR. WEERAKKODY: CHAIRMAN SHACK: MR. WEERAKKODY: are doing, Dr. We are raising it? This is just on new suspect we are kind of talking about is this a policy issue that's under consideration in the new reactor space? I know the ACRS has its -- what I'm saying is like Bob is up there. And we are kind of parroting what the current policy is as we know it from the new reactor folks. So I don't know whether we can solve it in fire protection. For example, even if we agree with you that we should require fire PRAs for all new reactors, it's not under the purview of the Fire Protection Branch. But I have heard from the grapevine that you are interested in this issue in other forums. MEMBER APOSTOLAKIS: Who is raising the That's what I -In PRA in general. In PRA in general. Apostolakis, is we So are following, as opposed to leading, that policy in the fire protection area. But in the meantime I think what Bob is NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 123 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 vendor had the cable saying is he is also the lead for his ESBWR. knows 805. He knows the new reactors. And he From a safety standpoint, the new designs are taking care of the safety business by keeping things in separate rooms. The only place they bring things together, is using the control room and in the containment. So we are looking at core damage frequencies like 100 times lower than our current operating plants. MEMBER APOSTOLAKIS: a fire But even if the PRA in the design the included certification utility that application, will have a this new implies reactor that doesn't necessarily have to go to NFPA 805. says. That's what it It can if they want, but they don't have to. CHAIRMAN SHACK: It's performance-based. My problem with this MEMBER APOSTOLAKIS: is that -- and maybe you're right, Sunil, that it's not your business to do these things, but we have lived with a very strange situation so far since 1998 for existing reactors because of the license issue. But to perpetuate this for new reactors and have these parallel systems forever doesn't sound to me like it's a rational way to proceed. And maybe it's not your job to do that but certainly I think the NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 124 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 issue. safety Committee's job. VICE CHAIRMAN SIEBER: policy issue. believe it to But that becomes a And I think that it's fair for us if we recommend to the staff that they consider developing a policy issue. way a rulemaking would start. MR. WEERAKKODY: But that's the If I may, one thing with respect to 805, we specifically excluded new reactors from 805 because, even though concept-wise, you know, risk-informed, performance-based is okay for new plants as well, it's kind of like the get-by rule, so to speak. We build a plant. And we want to fix the plant using risk-informed because if you think of the reg guide and the thresholds we applied in the core damage frequency changes that allows self-approval, for the new reactor, it's way too liberal in a sense because they start with a much advanced, much lower core damage frequencies. Now then you run into another policy Should we be holding new reactors to higher standards? So if there is a need to risk-inform new reactors, we should be looking at 806, not 805. MEMBER APOSTOLAKIS: NEAL R. GROSS Well, yes. That's a (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 125 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 detail as far as I'm concerned. And I suspect the reason why 805 did not refer to new reactors is this fear of not putting something there that you don't have to when you approve a document. It's always, you know, focus on the immediate problem and don't say anything about 20 years from now. As a philosophical issue, though, it seems to me that this is a good opportunity to go with a new system, which a lot of the utilities with existing reactors acknowledge is a good system, right? MR. WEERAKKODY: Yes. In fact, how many MEMBER APOSTOLAKIS: plants now, units? MR. WEERAKKODY: Forty-two. Forty-two out of -- MEMBER APOSTOLAKIS: MR. WEERAKKODY: when Browns Ferry starts. MEMBER APOSTOLAKIS: comment on this. VICE CHAIRMAN A hundred and three, 104 So that is really my SIEBER: Why don't we continue on with your remaining slides? MR. RADLINSKI: Okay. But, again, a fire MEMBER APOSTOLAKIS: PRA should receive a peer review to the extent that adequate industry guidance is available. NEAL R. GROSS So if I (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 126 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 fire PRA. that. don't have a fire PRA, what do I do? review or something else? VICE CHAIRMAN SIEBER: review if you don't have it. MEMBER APOSTOLAKIS: VICE CHAIRMAN SIEBER: review if you don't have one. MEMBER APOSTOLAKIS: You don't review the What? There is nothing to There is nothing to Do I get a peer But then I'm doing something in lieu of And I would like to know, would there be a peer In other words, this review for that alternative? sends a message that if you dare go into a fire PRA, we're going to hit you with 100 requirements to try to discourage you from doing it. VICE CHAIRMAN SIEBER: I think it is far easier to do a PRA of any set and get it peer-reviewed than it is to build architectural features into your plant. And that's really the choice you have. You know, you have to do all of your thinking up front in the design stage if you want to avoid having to take the route of risk-based fire protection. It's still a policy issue. MEMBER APOSTOLAKIS: It is. It is. But we are sending the wrong message, it seems to me, here. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 127 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 George. review. CHAIRMAN SHACK: There's no other to send, You wouldn't have a PRA without a peer A peer review of a deterministic program I mean, it's perfectly makes a whole lot less sense. sensible. VICE CHAIRMAN SIEBER: probably for all PRAs. -CHAIRMAN SHACK: And that's true I still would like to move on We had better move on. -- and be no later VICE CHAIRMAN SIEBER: than the fuel folks left us. CHAIRMAN SHACK: subcommittee report. MEMBER APOSTOLAKIS: CHAIRMAN SHACK: MR. RADLINSKI: We're taking up George's What? Onward. Right. The next comment had to do with new reactors and the guidance that we have provided that they should be maintained safe for all modes of operation. This entire slide is a summary of their comment, basically to say to delete the guidance that addresses fire protection for non-power operation. Their basis is that the staff has already approved new designs without disposition, that passive shutdown plants would have to evaluate fire effects on NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 128 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 me here. VICE CHAIRMAN SIEBER: Okay. And if there anything, active systems that are used when the plant is too cold for passive cooling. VICE CHAIRMAN SIEBER: If you don't do the plant will become warm enough for passive cooling. MR. RADLINSKI: You are getting ahead of is a requirement or guidance by the NRC, the comment is that the NRC should provide the specific method of analysis that the industry should use to address this. And, finally, they made the comment that the staff was directed to cease activity on the shutdown rule in 1997. I still haven't figured out what that has to do with this, but -- so our response is basically plants have to have a fire protection program that maintains plant safety in the event of fire in all modes of operation. Okay? If you want to find bases in the That's fundamental. regulations, 50.48(a)(2)(iii) requires that the means to limit fire damage to structures, systems, and components is important to safety so that the capability to shut down the plant safely is ensured. That means keeping a safe shutdown. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 129 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 a Last, but not least, 50.59. about this in the last meeting. We talked Again, we believe that it would be appropriate to put fire protection back under 50.59. The Commission has said they do not like the idea of a separate license condition for fire protection, no adverse effect approach to evaluating changes. 50.59 is good for the rest of the planet. So we It should be good enough for fire protection. are proposing to do that. Okay. VICE CHAIRMAN SIEBER: MR. RADLINSKI: Yes. I have one. George? Is that it? MEMBER APOSTOLAKIS: VICE CHAIRMAN SIEBER: MEMBER APOSTOLAKIS: statement. Yes. Again, there is There is a discussion of the self-imposed station blackout somewhere there on page 19. And there is speculation. The risk of self-imposed station blackout may greatly exceed the actual risk posed by the fire. And the licensee should consider the risk carefully when evaluating the plant safe shutdown design and procedures. How are they going to do this if they don't have an estimate of the risk? NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 130 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 problem. bit. again. VICE CHAIRMAN SIEBER: That solved that And that, in fact, we go on and say, "However, acceptable operator manual actions that are implemented in accordance with" such and such and such and such may present a lower risk than the self-imposed station blackout approach. And I'm trying to understand how in a deterministic world a utility may decide that one or the other represents a lower risk. MR. RADLINSKI: that down a bit. First of all, we did water I We took out the word "greatly." don't imagine that answers your question. (Laughter.) MR. WEERAKKODY: This is Sunil Weerakkody I think maybe I can address this a little You know, some of these things in the absence of a PRA, which probably aren't going to do as you discover a fire in certain areas made by engineering judgment or operator judgment as to "Do I want to cope with a self-induced station blackout or do I want to go and put out a fire the size of a wastebasket?" And so it becomes a judgment call in those clear-cut cases. Beyond that, I think that you are right, NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 131 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 George. You would have to do some kind of analysis for the big events where the risk is not well-defined. But just undergoing a station blackout is an operator's challenge. MEMBER APOSTOLAKIS: another statement. And then there is New reactor design should not rely on self-imposed station blackout to mitigate potential fire damage to safe shutdown systems. Is that a policy issue or is it a technical issue or -MR. WEERAKKODY: Even though you don't have numerical calculations to show that inducing a station blackout is not a good thing, there is overwhelming knowledge that that is not a good thing to do. sense. Why would you want to take out your I mean, it is kind of almost like common operating equipment intentionally because you want to be in the licensing basis. because the regulation We have had to limit that does not, the current regulation does not, prohibit that. In some of the cases, such as this, what we have done is we have basically told the new plants, "Please don't design your plants to rely on that kind of mitigation. It just doesn't make sense." You are blacking VICE CHAIRMAN SIEBER: NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 132 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 situation? MEMBER ARMIJO: MR. WEERAKKODY: In real. I don't know the answer, scenarios. MEMBER ARMIJO: Has anybody ever done it? No. that -VICE CHAIRMAN SIEBER: -- in some fire out the plant to avoid some spurious operation, which is pretty drastic. MR. WEERAKKODY: I mean, you could confirm with risk verification that that is, in fact, the case and say how big it is, but just to say that I'm going to kill these or I'm going to turn all of these off so that they don't get damaged by a spurious actuation -VICE CHAIRMAN SIEBER: It is my understanding that few plants have that as a provision -MR. WEERAKKODY: That's correct because VICE CHAIRMAN SIEBER: MR. WEERAKKODY: Do you mean in actual but we do know that in some plant procedures, they rely on it. Whether they actually have had a fire to do it I do not know. MEMBER POWERS: been done, Jack. NEAL R. GROSS I think, in fact, it has (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 133 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 doesn't VICE CHAIRMAN SIEBER: United States? MEMBER POWERS: Yes, in some U.S. plant. Where? In the For some reason, Pilgrim comes to mind, but I don't know that for a fact. VICE CHAIRMAN SIEBER: I don't know. I think it would be a good thing to find out. MEMBER APOSTOLAKIS: So for a clarification question, for new reactors, if they don't go the risk-informed approach, appendix R applies? MR. apply RADLINSKI: to plants No, no. Appendix R after '79 licensed technically. But the guidance is very -- I mean, it's It's -I think that we like appendix R. VICE CHAIRMAN SIEBER: have pretty much come to a conclusion of the formal presentation part of the meeting. is I read through all of My personal opinion documents and these particularly the questions and answers. I think both the industry, including NEI and other licensees, did a pretty good job of supplying comments. And the staff did a pretty good job of responding to those. I understand there is an NEI member here. And if anyone would want to make a statement, they can NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 134 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 decision. NEAL R. GROSS for you. CHAIRMAN SHACK: That's a Committee know I can. MR. RADLINSKI: Is there a take-away that do that now. you. CHAIRMAN SHACK: The next item in our If not, Mr. Chairman, I turn it back to agenda is a subcommittee report from George on our ESBWR Subcommittee. words? VICE CHAIRMAN SIEBER: CHAIRMAN SHACK: Does he know that? If you would like to say a few He does. Are we writing a MEMBER APOSTOLAKIS: letter on this, by the way? CHAIRMAN SHACK: No. VICE CHAIRMAN SIEBER: comments if you'd like. MEMBER APOSTOLAKIS: You can have added Thank you, Jack. I we assume you're going to approve the -MEMBER APOSTOLAKIS: MR. RADLINSKI: What? Is there a take-away that we assume you're going to approve the issuance of the reg guide or -MEMBER APOSTOLAKIS: There is a question (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 135 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS for this. MEMBER APOSTOLAKIS: CHAIRMAN SHACK: record for this. (Whereupon, a luncheon recess was taken at 11:38 p.m.) Yes. Pardon me? We can go off the MR. RADLINSKI: May I conclude from your comments that the Committee will approve the issuance of the reg guide? VICE CHAIRMAN SIEBER: (Laughter.) MEMBER APOSTOLAKIS: sort of a letter. VICE CHAIRMAN SIEBER: what I think right now. 9) SUBCOMMITTEE REPORT Okay. We had a I can only tell you You will get some Watch your mail. MEMBER APOSTOLAKIS: meeting on December 14 and 15. CHAIRMAN SHACK: We can go off the record (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 136 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 10) A-F-T-E-R-N-O-O-N S-E-S-S-I-O-N (1:06 p.m.) WOLF CREEK PRESSURIZER WELD FLAWS CHAIRMAN SHACK: Our first presentation this afternoon is on the Wolf Creek pressurizer weld flaws. Armijo. And our cognizant member for that is Sam Sam, I'll turn it over to you. REMARKS BY THE SUBCOMMITTEE CHAIRMAN MEMBER ARMIJO: Okay. Mr. Chairman, we're 10.1) going to have an informational briefing this afternoon related to the October 2006 indications of potential cracking at Wolf Creek. We will hear from representatives of the staff as well as from Duke Energy and NEI. We're not expected to write a letter or make any decisions, but we are free to ask as many questions as we think we need to understand this. With that, I would like to turn it over to -- I think it's Mr. Sullivan who will start out for NRR. MR. SULLIVAN: MR. BATEMAN: Thank you very much. Excuse me. Ted, before you get started, I would just like to add one more thing. This is Bill Bateman from the staff. We do have a subcommittee meeting scheduled for February 21st, at NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 137 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 10.2) the NRC. MEMBER APOSTOLAKIS: You have to speak to which point we will have a lot more time to talk about details here -- I know you have only got an hour for us now -- and then a full Committee meeting subsequent to that in March. MEMBER ARMIJO: And also I think we have Is there? someone on the phone, but I'm not positive. CHAIRMAN SHACK: MR. LUPOLD: I don't know. Our understanding is that our contractor, Dave Rudlin called. MR. RUDLIN: I'm here. Who are these people? MEMBER APOSTOLAKIS: MR. LUPOLD: Dave Rudlin is a contractor that we have utilized to evaluate some of the flaws that we discovered at Wolf Creek. MEMBER APOSTOLAKIS: MR. LUPOLD: Are you NRC yourself? I'm with I am Tim Lupold. the microphone, though, because -MEMBER ARMIJO: Okay. Well, just as long as the folks on the phone just please put their phones on mute so we don't hear any kind of background. With that, Ted, it's all yours. BRIEFING BY AND DISCUSSIONS WITH REPRESENTATIVES OF THE NRC STAFF NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 138 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 background. MR. SULLIVAN: My name is Ted Sullivan. And I work in the Division of Component Integrity. And I've been working on this Wolf Creek law issue since about November time frame. I wanted to set out some very brief I know this is kind of industry stuff, but I thought it would be appropriate to help put the Wolf Creek information in a little bit of context. And at the subsequent meeting, I expect that either industry or ourselves will talk about this more. The context for these inspections is an industry "mandatory program" under some guidelines that were issued by NEI. This particular program is And it deals very customarily referred to as MRP-139. with inspection and mitigation of dissimilar metal butt welds and reactor coolant system of PWRs. It provides, among other things, guidance for volumetric and visual inspection of alloy-82/182 butt welds. It is over and above what is required by the ASME code in that it requires -- in the industry context, I'm using the word "require" -- inspections that are more frequent than those required by the ASME code. around And the whole program is somewhat oriented temperature in that, for example, the pressurizer weld locations need to be inspected first NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 139 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 six nozzles. and most frequently. So it was in that context that these indications or flaws at Wolf Creek were found. This licensee was performing inspections of the dissimilar metal butt welds in the nozzles of the pressurizer. And these indications were found as part of inspections that were done prior to applying weld overlays, which was their plan all along. And I'm going to talk about that more in subsequent slides. We were notified of it in mid October by an event notification. So flaws were found in three of I guess And I'll get into them one by one. In the surge line, there were three flaws found. were circumferential in orientation. They are of varying sizes. They One, the first one, has an arc of about 38 degrees; the second about 21-degree arc; and the third one is a much smaller, about 7 and a half-degree, arc. This weld was last examined in 1993 using techniques that predated the performance demonstration initiative qualification program. I want to say a little bit about the qualification of the procedure and the examiner. manual procedure. The procedure that was used was a It was qualified for flaw detection NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 140 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 last and length sizing. The examiner was qualified for detection. He had apparently not gone through or passed -- I don't know which -- the qualification for length sizing. Notwithstanding, readings were taken for informational purposes on length and depth and all the readings were confirmed by a person from EPRI. that note, which will appear on some subsequent viewgraphs, is true for all of the welds examined. MEMBER ARMIJO: expert or did you -MR. SULLIVAN: The EPRI person was a Was the EPRI person an And person who administers the PDI qualification exams. MEMBER ARMIJO: MR. SULLIVAN: But he's experienced? I would say he was very I just can't call experienced, and he was an expert. him qualified because EPRI doesn't qualify its own people. They administer the exams. MEMBER BONACA: volumetric The 13 years between the is normal, the long examination, period of time? MR. SULLIVAN: such a long period of time. time. I'm not sure why there was It does seem like a long It's more than an interval. MEMBER BONACA: Yes. (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 141 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 MR. SULLIVAN: That's a good question. Do you know? Well, if I could That's not something we asked. MEMBER CORRADINI: address that, the requirement would be to inspect it once every interval. And the ASME section 11 gives latitude to defer some exams from one period to a next. So it's ten years plus or minus is what the So it's not unheard of to have 13 exams would be. years between subsequent exams. MR. SULLIVAN: Okay. On the relief It was a nozzle, there was a very large flaw. 170-degree arc. one flaw also. And on the safety nozzle, there was It had about a 55-degree arc. ARMIJO: I've seen prior MEMBER presentation material that the staff has issued, maybe a month or so ago. And I've seen numbers that are higher, like 11-inch cracks or indications, as opposed to 7.7. What is going on? MR. LUPOLD: The numbers that you're referring to would be the lengths of the flaws, as projected on the OD of the pipe. This is these numbers that you're seeing right here -MEMBER ARMIJO: MR. LUPOLD: flaws on the ID. NEAL R. GROSS ID. -- would be the length of the (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 142 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 the MR. SULLIVAN: When we initially got the And we weren't clear data, it was over the telephone. where these links were. ID. We thought they were on the They subsequently clarified it was on the ODs. So we had to do a little conversion. Okay. Our concerns with these inspection results were that they were the first large multiple circumferential flaws identified. Previous circumferential flaws have been identified, but these were large. We found a very large flaw. And we found multiple indications in one of the nozzles. The expectation was to see smaller flaws and see axial flaws. data shows more Predominantly the inspection you get axial flaws than often circumferential. And, of course, the concern with circumferential flaws is it can lead to rupture, as opposed to the concern you have with the axial is that it's much more likely to just lead to leakage. And our concern with the large flaws and multiple flaws was that it seemed to us to increase the need to complete the baseline inspections on a timely basis. So we did fracture mechanics evaluations of this data. We took it as though it was axial, We didn't actual, even though we couldn't confirm it. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 143 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 nozzles. change the sizes. to do analysis. The analyses were done in such a way as to basically work the problem backwards to try to get an estimate of when the cracks might have initiated. And We just used the information data then we worked the problem forward to get an estimate of when the flaws could lead to leakage if they were left in service, if they had been left in service. And we estimated times to reach critical flaw size, again, if they had been left in service. We analyzed the flaws in all three We didn't assume that the flaws in the surge We just picked the largest of those line interacted. three flaws. We calculated time ranges based on three different residual stress profiles, two different fracture mechanics models, and two different through-wall flaw models. And I think we can talk about that a lot more in the meeting on the 21st of February, but the reason I'm bringing it up now is that 2 times 3 times 2 turns out to be 12 different cases that were analyzed. slide. And that will come up on a subsequent These were not best estimate calculations. And they're not considered bounding. NEAL R. GROSS They were just (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 144 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 calculations we did to try to scope the problem. VICE CHAIRMAN SIEBER: Did you get any clue as to the validity of the leak before a break assumption? MR. SULLIVAN: going with this. VICE CHAIRMAN SIEBER: first and then leaks. MR. SULLIVAN: the next slide. results for we the saw Well, this will come up on Okay. It breaks Yes. That's where we are On this slide, which talks about the surge some line, time in all 12 cases we analyzed, rupture. between leakage and And you can see that in the rows of this particular table on this viewgraph. So that is the salient point, I think, of this viewgraph other than the fact that the times could be fairly short, less than two refueling cycles. MEMBER MAYNARD: One thing to be pointed out, Wolf Creek did not take credit for leak before break. This was analyzed without taking credit for So they leak before break for this particular line. were not outside their design basis. important to note. MR. SULLIVAN: That's true. I think that's On this plant, the surge line was not a leak before break, NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 145 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 assumptions? wait. pipe that had not been requested of the staff nor reviewed. And, as with other plants, the smaller nozzles, the safety and the relief lines were never submitted to the staff as candidates for leak before break. CHAIRMAN SHACK: I mean, those ranges of I mean, times don't pass my sanity check, actually. you know, I would say measured size to leak could be one year to infinity. Initiation to measured size could be -- I would be astounded if it were .3 years. It could well be 16 years. MEMBER CORRADINI: astounds you? MR. SULLIVAN: CHAIRMAN The first one. What were the Which one did you say SHACK: Well, maybe that's something we can just We'll wait until I'll just make that comment. we get to the subcommittee meeting. MEMBER ARMIJO: Even though there was no claim on leak before break, those are pretty big pipes, 15-inch, 16-inch pipes. piece of metal there. MEMBER MAYNARD: 15-inch line. lines. NEAL R. GROSS The surge line is a That's a pretty hefty And then those nozzles are 8-inch (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 146 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 of the day. NEAL R. GROSS MR. SULLIVAN: In our evaluation of this information, we didn't really give any particular credence to the time. analysis as I'll And I think the results of this into it really are not get surprising. It's not surprising that on the surge line, you would see leak before break behavior. On the smaller lines, which are not as flaw-tolerant, it's not surprising that you would see rupture turnout in the calculations before leakage. And that really is pretty much how we used the information. MEMBER BONACA: It still troubles me when I think about what we're saying in license renewal, that a 10-year inspection was good when the plant was 10 years old. years old. Then it's good when the plant is 50 And this is confirming otherwise. MR. SULLIVAN: Well, I guess the reason we are pretty comfortable with this is that industry has put together a reasonably aggressive program to mitigate these welds. space, And so in license renewal we think that that's really what license renewal is relying on, is the program to mitigate these welds and address PWSCC. MEMBER BONACA: Yes. This is the problem (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 147 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 have to MR. SULLIVAN: MEMBER BONACA: That's right. Then tomorrow there is I mean, there has going to be some other component. to be a recognition that aging is going to create new flaws. It just is inevitable. MR. SULLIVAN: MEMBER BONACA: Right. And I'm just saying that we'll have to reflect on the inspection intervals. CHAIRMAN SHACK: Yes. Let me ask another question about the inspections. I mean, every section XI inspection now of a welded pipe is going to be done with a PDI-qualified inspector? MR. SULLIVAN: CHAIRMAN SHACK: That's correct. Okay. So there will be no more inspections that will be done by anybody that's not through the qualification process? MR. SULLIVAN: recognize, That's true. that I mean, you are PDI though, there supplements to address, at least the cast stainless steel. That problem is still being worked. CHAIRMAN SHACK: MR. SULLIVAN: Right. And I think one of the points of this -- and industry will probably make this point on the 21st, but there are a lot of these welds that can't be inspected because you don't have access NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 148 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 further on overlay? or you've got materials that are not inspectable. But one of the thrusts of the MRP-139 program is to make the configuration inspectable, even if you have to put a weld overlay, a full-structure weld overlay, on the weld to accomplish that. CHAIRMAN SHACK: So you can inspect the You still can't inspect the pipe? MR. SULLIVAN: Depending on the material, underneath it, you can inspect into the original weld, at least some distance, again, depending on what the adjacent materials are. Okay. We have kind of covered the point In here already, but I'll just get into it briefly. the leak to rupture row, the fourth row on this table, the important information is in the note. And what it shows is that in 8 of the 12 cases we analyzed, there wasn't any time between leak and rupture. And, contrasting that with the safety nozzle, we found something similar, although not quite as dramatic, which is that in 4 cases, 4 out of the 12 cases, there was no time between leak and rupture. And I think that we can discuss this the 21st. We're trying to make arrangements to send over to the ACRS the report that our contractor put together that will discuss this in NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 149 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 it? MR. SULLIVAN: -- Dr. Shack was commenting what -MEMBER ARMIJO: That's Bill's issue, isn't treated the a lot more detail. You can see exactly which assumptions led to which results. MEMBER ARMIJO: Is your primary assumption that this was PWSCC and that the crack growth rates, you had crack growth rate data that you could use in the analysis? MR. SULLIVAN: flaws as Yes. In this analysis, we which was the most PWSCC, probable causae that was identified by the licensee. And we used the MRP-115 crack growth rates, which were generated by the industry using a lot of data, both industry data, probably some NRC data, and some Navy data. MEMBER ARMIJO: Okay. So then you worked back from the time to -- you worked backwards from those. So that left a long period of time for initiation, right? MR. SULLIVAN: Well, basically this is on, that it shows the possibility that these flaws generated in a non-credibly short period of time. CHAIRMAN SHACK: I mean, even to do these (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 150 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 things, you have to make all sorts of assumptions about how many cracks initiated, you know, whether these cracks are 11 inches long because you initiated 100 short cracks that linked up or there is this one crack that grew that arrested itself going through the wall and then grew around the thing. number. So you pick a I can come through here and give you an analysis that can be just about any number you want. MEMBER ARMIJO: Or the state of stress. What's the stress where these things are growing? MR. SULLIVAN: Well, we had to make We used the design assumptions about part of that. loads that came from the licensee and maybe ultimately from Westinghouse. And we used three different So that's where the stress residual stress models. assumptions came from. Okay. Moving on into some less numerical material, some general observations are that long circ flaws decreased time between leak and rupture. Your flaw tolerance goes down if you start out assuming that you've got long circ flaws to begin with. And the second observation is basically that smaller diameter welds are less well-tolerant than large diameter welds. And then specifically I think the rest of this slide just kind of reiterates NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 151 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 distribution conservatism. flaw. didn't what we just talked about, that the relief line had the least margin based on our analysis, with 8 of 12 cases showing no time between leak and rupture. The safety line analysis had a shorter It showed that 4 out of the 12 cases analyzed produce any evidence of leakage prior to rupture. And the surge line, I think in part because of the way we analyzed it, not linking up any of the flaws, we sold it in all cases with some time between leak and rupture. And the shortest time on all of these analyses or most of them, not every single one, or most of them was less than two operating cycles, I think between initiation and failure. I've got a little treatment here of conservatisms, non-conservatisms, and uncertainties. And it's kind of difficult in this case to try to figure out which box to put some of these aspects in. Residual stress relaxation is a problem that was worked by industry prior to our last meeting. That's a potential conservatism. The only reason I say "potential" is I think it could vary depending on what residual stress models are used. The is axisymmetric generally residual thought of stress as a That's something I haven't mentioned up NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 152 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 until now, but the way the residual stresses are modeled for practicality and possibly also because of lack of better information, they're modeled as axisymmetric. conservatism, necessarily be. And that's generally viewed as a although I don't think it would There are some potential non-conservatisms in the analysis. Not to overwork this, but we have The first one talked about some of these already. certainly I have talked about. The pipe loads that we used were not necessarily bounding. numbers. industry. We got Wolf Creek-specific And we're aware they aren't bounding for the The indication sizes may not be bounding. The We really don't know what is out in the fleet. indications we use may be bounding, but they may not be. The industry recommends and uses the 75th percentile crack growth rate. this analysis. That's what we used in That's not necessarily bounding. And in terms of uncertainties, I think we have hit on some of these. distribution intended. MEMBER APOSTOLAKIS: NEAL R. GROSS I don't understand is certainly The residual stress an uncertainty, no pun (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 153 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 developing bounding. MR. SULLIVAN: more conservative. As I think we may get into later, there are 37 units that have not been addressed under MRP-139. That's a little bit just slightly bigger That's true. One would be the statement of the 75th percentile is not necessarily bounding. What does that mean? Well, a 95th percentile I'm MR. SULLIVAN: crack growth rate would be more conservative. just pointing out that what was used in the analysis was the 75th percentile. MEMBER CORRADINI: MR. SULLIVAN: Growth rate? Growth rate, yes. But neither one is MEMBER APOSTOLAKIS: than half the units. And flaw depth is another uncertainty. As I pointed out before, the flaw depths were measured, but they weren't measured with qualified techniques. The is position based on that the the staff has that been the thinking inspections or mitigations need to be accelerated from the current industry schedule for some plants. I know that statement is a little bit in a vacuum, but if we have time, I'll talk more about what that means. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 154 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 to now right now. CHAIRMAN SHACK: You aren't going to give us a hint as to what some plants are? MR. SULLIVAN: Okay. I'll get into that I said that 37 plants haven't completed There were 19 plants that their MRP-139 evaluations. don't even have dissimilar metal welds. There are something like 13 plants that up have already implemented the MRP-139 inspections or mitigations. mitigated. Most of them have Some have just inspected with an augmented inspection frequency requirement in MRP-139 over that in the code. There are 26 or 27 plants that are Two thousand scheduled to do the inspections in 2007. and seven is the schedule that was originally in MRP-139 for completing the baseline program. leaves 9, 10, 11 plants somewhere in there. The reason I'm being a little bit vague is that it hasn't happened yet. on what is planned. We just have information That But somewhere around ten plants are slatted to do the examination after the original schedule in MRP-139, namely in 2008. really the target of this first bullet. MEMBER ARMIJO: The plants that don't have And they're dissimilar metal welds, are they exempt from this NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 155 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 materials? that? MR. SULLIVAN: address PWSCC. The program is designed to inspection? MR. SULLIVAN: MEMBER ARMIJO: Yes. Is there a reason for And PWSCC has only been found to date in alloy 82, 182 welds and alloy 600 products. MEMBER ARMIJO: And what are these Are those -MR. SULLIVAN: This program and the Wolf Creek welds only applies to 82, 182 -MEMBER ARMIJO: about the 11 that -CHAIRMAN SHACK: with 308 in all likelihood. MR. SULLIVAN: No. The 11 plants are 11 They would be stainless Right. No. I'm talking plants who have planned to do the inspections in 2008 that all have alloy 82 or 182 welds. MEMBER ARMIJO: Yes. I got that. I'm going back to the ones that are exempt from this issue. MR. LUPOLD: Okay. We are referring to the plants that we said don't have materials that are susceptible. And those materials typically are Some of those materials stainless steel materials. (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 156 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 152. MR. LUPOLD: Alloy 152 is a nickel-based could be alloy 52 or alloy 152 also. Now, MRP-139 actually talked about those type materials. materials. They're considered to be resistant And all MRP-139 would do is have you go back and inspect in accordance with the ASME section 11 program. MEMBER ARMIJO: Okay. So there is some basis for those materials to be viewed as lower risk or no risk? MR. LUPOLD: That's correct. And at the subcommittee MEMBER ARMIJO: meeting, I would like to get more information on why that is true. MEMBER POWERS: I am not familiar with alloy which has a much higher chromium content in it than alloy 82 or alloy 182. And having the higher chromium content has demonstrated it is more resistant to primary water stress corrosion cracking and testing that is being conducted on the material. CHAIRMAN SHACK: equivalent of 690. MEMBER CORRADINI: good statement. NEAL R. GROSS Yes. That's a very It's sort of the weld (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 157 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 shortchange already. viewgraph. Now, the I don't want to in any way industry, but we put together a We have They're MR. SULLIVAN: Okay. Returning to this viewgraph, the second part of our developing position is that we view that enhanced RCS leakage monitoring with action levels to shut down and visually inspect welds would be a very desirable thing to do until inspections or mitigations are completed. And in developing this position, we considered a number of factors. I think we talked about most of these So I think I will just move on to the next listing of bullets of the industry position. lifted these strictly out of their documents. going to have time to explain their position more, but I just wanted to lay out a couple of things. Industry has stated they believe the inspection findings are an anomaly. We don't think And have we're in the position to treat it as such. anomalies have been -inspection findings occurred in the past that have been ascribed to anomalous behavior. And most of the time they don't turn out to be anomalous. Industry agrees with an enhanced leakage NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 158 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 now. detection program. I think our differences at this point have to do with action levels and specific time lines for completing action levels and shutting down the plant if that's what it comes to. They have a very good program, but it's not as prescriptive as we would like to see. Industry is undertaking some non-linear finite element analyses to try to address some differences between industry results and what they think is a more realistic outcome. that in the next slide. And I think that's probably enough for Industry is going to have time to talk about it I'll comment on some more. MEMBER BONACA: Sorry. The issue, you had some bullets about bounded by plant design basis accident analysis, existing safety analysis conclusions remaining valid. Of course, frequently of And so the breaks is an element of those analyses. somebody will explain why these would be acceptable. MR. SULLIVAN: to be up in a few minutes. MEMBER BANERJEE: the finite element analysis. I think industry is going So maybe they can -I have a question about This has to assume some sort of a residual stress distribution, right, when NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 159 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 correct. MEMBER ARMIJO: time, right? something -MR. SULLIVAN: I think it was. But do these actually That was your longest you do this? made there? MR. SULLIVAN: In the analyses that have So what sort of assumptions would be been done so far, we used three different residual stress assumptions. One of them was an ASME model. It was pegged to a It appears in the ASME code. higher yield stress than the one in the code because the materials have a different yield stress. The second model is one that was developed by our contractor based on finite element analyses of weld deposition. MEMBER BANERJEE: MR. understanding. SULLIVAN: When the weld was done? Right. That's my And the third assumption was no residual stress at all. MEMBER ARMIJO: MR. SULLIVAN: Just applied loads? Just applied loads, And it should have been if it wasn't MEMBER BANERJEE: bound the situation? MR. SULLIVAN: No, we don't think they (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 160 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 bound it. That's why I made the statements earlier that these analyses are just scoping analyses in our view. They're not bounding or best estimate. We just tried to do some analyses to show what could happen. MEMBER BANERJEE: What will industry do to improve this situation or are they going to tell us? MEMBER ARMIJO: MR. SULLIVAN: They are going to tell us. Well, I think they are available to answer in more detail, but I think the main thing is that these analyses will remove the constraint that the flaws remain elliptical. MR. LUPOLD: the next slide. MR. SULLIVAN: We have some skepticism. We certainly think it We should just go right to This isn't about the analyses. will be interesting. We think it's important work. We're interested in understanding what's going to happen from these analyses. And the NRC is interested in doing some similar work itself. But in terms of using this for regulatory decision-making, that's kind of another matter. We think that these analyses will basically turn out to just be another scoping study. with different results. leak before a break. NEAL R. GROSS And they may come up They may show that you get (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 161 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 right. NEAL R. GROSS these leaks? MEMBER CORRADINI: Typically RCS leakage bullet. about But I think the end result would be what we already know. a break. You may or may not get leak before I think that unless these analyses could rule out rupture prior to leakage, I don't think they're going to help us in regulatory decision-making. So that's kind of the point of the first I already made the second bullet. that. We talked We don't consider these results anomalous. experience results. ever do. We don't think that's a position that proves out with previous inspection And, you know, that's not something we would I previously kind of alluded to our It's concern with industry's leak-monitoring program. an excellent program, but it doesn't have time constraints for implementing actions. And it doesn't require shutdown depending on what could be found. MEMBER BANERJEE: How do they monitor is measured just through a mass balance for the reactor coolant system. MEMBER BANERJEE: In the system itself, (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 162 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 time period. answer this MEMBER CORRADINI: Yes, in the system. It's measured at every plant at least every 72 hours. Some plants will do it 48 hours. it every 24 hours. MEMBER BANERJEE: Well, what are the Some plants will do thresholds of detection here? MEMBER CORRADINI: question a Industry may be able to bit better, but little typically you could measure into the hundredths of a gallon per minute leakage. MEMBER BANERJEE: MEMBER CORRADINI: MEMBER BANERJEE: MEMBER CORRADINI: MEMBER BANERJEE: Hundreds of gallons. Hundredths, .01. Hundredths? .01 galloon per minute. So you can actually monitor all the inflows and outflows and everything down to .01 of a gallon? MEMBER CORRADINI: It's monitored over a So you collect how much leakage you have And then you do the mass over like a 24-hour period. balance. And you can come up with changes of a couple of hundredths of a gpm, you know, from one day to the next. You can see that in the calculations. And typically, though -MEMBER BANERJEE: It depends on the (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 163 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 your -MEMBER BANERJEE: So just to go back to accuracy with which you can measure various -MEMBER CORRADINI: Yes, it does. It depends on the accuracy of your measuring instruments and, you know -MEMBER MAYNARD: This has tech specs associated with it, not only the instrumentation but the requirements to do it. In addition to being able to do the mass balance and leakage that way, if you get a leak in this part of the system, you also have radiation monitors and you have containment temperature, containment pressure. You have a number of other things that are going to alert you to a leak from an area like this. MEMBER CORRADINI: Right. You also have this mass balance thing, when we had these leaks in alloy 600 and alloy 600 welds, were such leak-monitoring programs underway to do a mass balance and detect the leaks? MEMBER CORRADINI: Yes. Utilities have A very, used the mass balance for some time period. very small leak from an alloy 600 weld or an alloy 82 weld will probably not be detected in a mass balance. MEMBER BANERJEE: So with Davis-Besse, (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 164 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 of leakage. MR. RUDLIN: Right. Do you have some idea of that would this have been detected or not? MEMBER CORRADINI: because I don't I hate to speculate on really have the background to really answer that information question. on Davis-Besse MR. SULLIVAN: We've done some calculations of situations where a flaw goes from just a pinhole. A circ flaw, for example, goes from a pinhole to a longer flaw assuming that the overall length is short enough to remain stable. And we believe you get enough flow out of a long, stable -- well, not a long -- a short -- can anybody help me here? Dave? Yes? You did some calculations MR. RUDLIN: MR. SULLIVAN: MR. SULLIVAN: how long the flaw might have to be before you would see something on the order of, say, .1 gpm? MR. RUDLIN: monitoring factors. MR. SULLIVAN: Did we do these It depends on the load and calculations assuming the Wolf Creek loads? MR. RUDLIN: Yes, yes, but we didn't do (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 165 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 ask you. any calculations from obtainable type loads. them for ideas like through-wall crack types. MEMBER CORRADINI: Dave, I was going to We did Is it safe to say that a leak rate would be a high enough volume to detect before we encroached rupture of a pipe? MR. RUDLIN: The problem is that when you have just the flaw just breaking through, the time between the first pinhole to the time it becomes an idea like a through-wall crack, it's probably going to be very small. The growth in that little ligament area is going to happen very, very quickly. In the relief line type of calculation, a surface crack was actually unstable. And so before even leakage, the surface crack would have failed, creating a large opening that would have been longer than the critical through-wall crack size. That was a specific unique case, I think, with the relief line. cases, where the I think in most of the other surface crack was stable until leakage, there probably would be enough time for detection before you can get the critical through-wall cracks stopped. MEMBER ARMIJO: Yes. Well, those kind of details I think we have to address in the subcommittee NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 166 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 meeting. going to But I just want to ask one question. ask industry the same thing. I'm Are you convinced that these are cracks -- it's as simple as that -- these are cracks and not just some other anomaly, bad NDT signals or -MR. SULLIVAN: I don't think we can say that we are convinced because there is no destructive examination data. MEMBER ARMIJO: MR. SULLIVAN: Right. But the analysts called it as a multi-faceted indication, which this is the sort of indication you can get from stress corrosion cracking, although you don't necessarily only get it from stress corrosion cracking. I think the position of the regulatory agency is we have to treat it as stress corrosion cracking. take. It's the only sensible position for us to We cannot be in a position of saying, "Well, we So we're going to treat it as don't know for sure. though it's not cracking." MEMBER ARMIJO: I know that Wolf Creek did not take a sample for metallographic examination. Does anyone in the industry intend to do that if they find something so you can put it to bed that this is really PWSCC and not something else? NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 167 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Alex Marion. someone -MEMBER ARMIJO: MR. SULLIVAN: in the next segment. MEMBER ARMIJO: (No response.) MEMBER ARMIJO: Okay. Well, I think the Any other questions? Okay. Fine. MR. SULLIVAN: I think that maybe Alex or -- can answer that question next speaker is -- where did we have our little -- who is the next speaker? Is it Alex Marion? Yes. NEI. MR. HAMMER: Sam, I understand that Duke, the Duke representative is not here but that Alex Marion is going to make the presentation. MEMBER ARMIJO: MR. HAMMER: MR. MARION: Okay. NEI. Good afternoon. My name is I'm the Executive Director for Nuclear Mike Robinson was Operations and Engineering at NEI. scheduled to give this presentation, but he was unable to attend because of weather conditions in the south. I have with me Glenn White from Dominion Engineering, one of the technical consultants that the industry is using; and also Jim Riley, who is the Director of Engineering of NEI. Hopefully Mike Robinson is on the telephone. Mike, are you there? (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 168 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 everybody. MR. MARION: Yes. The EPRI program, as charts. MR. MARION: Yes. But you are presenting for MR. ROBINSON: MR. MARION: I am here, Alex. Good. And we also have Christine King from EPRI on the phone as well. MS. KING: That's right. I am here. MR. MARION: So I have a team of four people to keep me out of trouble. MEMBER ARMIJO: Well, these are EPRI logo MEMBER ARMIJO: Ted Sullivan indicated, comes under the auspices of an industry-wide initiative that was undertaken by the Nuclear Energy Institute. And the EPRI materials reliability project is one of the issued programs that come within that program or within that initiative. And their primary focus is on pressurized water reactor piping systems and components relative to degradation. What I would like to do is offer the industry perspective relative to this question of the generic implications of the Wolf Creek inspection findings. Let me just say that our position is that the industry has put forth a very proactive management NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 169 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 program to assess the condition of alloy 82, 182 butt welds and PWR primary systems. And we have developed that with a focus on the more susceptible areas. Basically the first phase involves the welds located in the vicinity of the pressurizer. And we have reevaluated the schedule and the focus of our program, which is documented in MRP-139. And we do not believe that we need to accelerate the schedule. So fundamentally our first principle is that we feel that the bases for MRP-139 inspection program as well as the safety analysis that was developed to support that inspection program remain valid in light of the findings at Wolf Creek. MEMBER ARMIJO: When that program was set up, were you basing that on the existence of axial cracking or did you have circumferential cracking also in mind when you came up with these? MR. MARION: I believe predominantly axial cracking based upon the available information from laboratory data as well as field experience on the kind of cracking phenomena we have been experiencing on an international basis. And all of that was factored into the program that we have developed thus far. I don't know if Mike or Christine want to NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 170 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 question elaborate on that at all. MR. ROBINSON: Just a quick comment, Alex. When we put 139 together, we did assume axial cracks, but we also went back and accounted for the fact that certain cracks were very much a possibility. So 139 considers the possibility of both and circ cracks. MR. MARION: MEMBER about Okay. Just to ask a leakage monitoring ABDEL-KHALIK: enhanced this program, what is being proposed here? Tightening tech spec limits on unidentified leaks or -MR. MARION: I will speak to that in a little more detail later in the presentation if I can defer that question. MEMBER ABDEL-KHALIK: MR. MARION: All right. Basically, as we indicated, the pressurizer locations were the more susceptible locations based upon the knowledge that was available at the time that we put the program together. clearly they have our highest priority. Fundamentally with regard to the Wolf Creek findings, we think they're anomalous because they're not validated or confirmed by any of the previous findings in basically the worldwide And experience to date. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 171 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 leakage I will elaborate a little bit more on the monitoring but we programs, think as I indicated previously, this is very important because we as an industry believe safety needs to be maintained and it is being maintained. And one of the key aspects of doing that is to have an effective responsive leakage monitoring program. MEMBER ARMIJO: Just to make sure I understand, you said you don't think these are valid. Does that mean you still have doubts whether these are cracks, that there may be just some NDT anomaly? MR. MARION: Yes. Hindsight being 20/20, we wish we had taken a boat sample at the time, but we didn't. MEMBER ARMIJO: MR. MARION: Me, too. And so, as the staff indicated, they feel that they're in the position where they have no choice but to take a very stance Wolf relative to the inspection conservative findings of Creek. And because of their uniqueness, we don't feel that we have to take the same position. There are discussions going on -CHAIRMAN SHACK: That doesn't inspire confidence in your inspection program, though, if you NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 172 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 are that skeptical about the results. MR. MARION: Well, the reason we're skeptical about the results is because there wasn't a sufficiently comprehensive NDE conducted to really determine the depth size, et cetera. question that remains. And the uniqueness of the indications on -- was it five indications? -- basically averaged anywhere from 22 to 33, 35 percent through all going circumferentially around the pipe. been seen before at all. CHAIRMAN SHACK: But, I mean, that is You know, we And that has never And that's a big fairly typical of a crack in a weld. have core shrouds cracked partway through by the foot. You know, there must be -- well, make it the kilometer. MEMBER ARMIJO: Yes. In BWR pipe cracks, we have had multiple indications and -MR. RILEY: This is Jim Riley, NEI. A couple of the reasons that we felt this was unusual is that there was no axial component to these. And where we have been predicting axial all along kind of being inspected degradation pipe, there was no axial component here. And, in addition, we found all of these NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 173 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 have indications at basically the same place, through-wall, which is a little curious also because if they are cracks and if they are growing rapidly, to find this many at basically 20-some percent through-wall, all at a snapshot in time, is unusual. MEMBER ARMIJO: certainly seen that Not for BWR piping. kind of We circumferential cracking, that depth, also hard sizes on BWR pipe cracking and -MR. RILEY: Did you find them all about the same depth at the same time? MEMBER ARMIJO: CHAIRMAN SHACK: Sure, sure. That is exactly what I would expect, actually, from stress corrosion cracks in a pipe weld. MR. RILEY: They would all be growing on a basis we started at the same time growing at the same -CHAIRMAN SHACK: as they go through the weld. No. That they slow down And now the guess is, have they stopped or have they just slowed down? PARTICIPANT: CHAIRMAN They're growing laterally. So they're going to So you're SHACK: around and spread and initiate around. going to get long cracks growing slowly through the NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 174 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 wall. But now the question is, how slow is slow and how long? MR. MARION: Well, if I am not mistaken, we're talking about different materials and different forms of degradation. MEMBER ARMIJO: I don't think so. They're definitely different materials. CHAIRMAN SHACK: It is a residual stress, and it is a stress corrosion crack. MEMBER ARMIJO: Unless you're sure it's not a stress corrosion crack by virtue of that you don't have confidence in your NDT methods, then you've got to assume that it is, I guess. Go ahead. MS. KING: This is Christine King. I would like to offer one other point relative to this being an anomalous indication. We have recently taken samples out of the North Anna Unit 2 reactor vessel head and cut into them. And those were indications that were called large circumferential flaws as well. When we actually cut into those flaws, what we found was a repair that had intruded into the nozzle. And that's what was actually found and called by the NDE. We had similar -- it had facets and things NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 175 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 comment. was like that. So it's not that we don't have confidence in our NDE, but sometimes you do get a repair that by a UT method looks as if it is a PWSCC flaw. MEMBER ARMIJO: involved in BWR pipe Well, let me tell you I cracking at the very beginning of that problem. And I can't tell you how many times people said we had an anomalous finding, one of a kind, and it turned out to be a major problem for the industry. So I think the prudent thing to do is assume they're real until you prove that they're not real cracks. And you're going to save yourself a lot of money in the long run. MR. ROBINSON: Alex, just one other You know, the cracks at the indication at Wolf Creek aren't the first indications of cracking in these pressurizer nozzle locations. There are, I think, if memory serves me correctly, about 20 worldwide other occurrences where cracking has been found in these locations. And when you go back and look at the indications that were reported from the other 20 or 17 locations, you find that most of those were axial in orientation. You also find that where there were other circ cracks, they were much smaller in scale. But they also had an accompanying axial component, which, NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 176 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Thank you. again, there were other reasons why we believe that part of what we're seeing here at Wolf Creek really doesn't fit the model of what we have seen elsewhere. MEMBER ARMIJO: Yes, but nature doesn't It does what it wants. Anyway, go feel it has to fit your model. And then your model has to fit the data. on. MR. MARION: That's a point well-taken. I would like to move on with the presentation material because I only have 40 more slides to go in the next 5 minutes. All pressurized water reactors will have inspected or mitigated pressurizer locations by their next normally scheduled refueling outage, which is less than 16 months away. perspective. Let me offer another And we'll get into details on this when we have the subcommittee meeting. If you look at the timeline of activity and when MRP-139 was issued where plants were in their outage cycles, regardless of whether 18-month or 24-month cycle, and you look at the timeline and you could see clearly that not everybody was going to complete the inspections by the end of 2007, we recognize that on the front end. And the December 31, 2007 was from the industry perspective a reasonable NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 177 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 date to basically put on the table as a goal to be achieved, but we recognize everyone couldn't meet that gaol. CHAIRMAN SHACK: that goal under your plan? MR. MARION: The utilities that have When will everybody meet planned to do inspections in 2008 have evaluated their justification and rationale for not meeting the goal. And that evaluation has been reviewed independently by the utilities. MR. ROBINSON: Alex, a more direct answer, right now there are nine plants that are planning to do either inspection or litigation in the spring, in the Spring 2008. There is one plant that has an outage scheduled the first week of February 2008, three plants that have outages scheduled for the first week of March of 2008, a fourth plant that has outages scheduled in April of 2008. And the last plant that has an outage to do with this particular material and this issue occurs in early June 2008. CHAIRMAN SHACK: MR. MARION: Thank you. One other thing All right. that had come up is the NRC was concerned about not having specific information on what utilities have completed relative to this inspection program, nor do NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 178 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 disciplined, they have complete information relative to what utilities were planning to do under this program. And so all the utilities have agreed to submit letters to the NRC. And those letters were to be in by the 31st of January articulating the status of their inspection results or mitigation results to date as well as their plans going forward. And to date all plants have completed bare metal visual examinations. And a number of them have already completed volumetric examinations. This graphic represents the inspection mitigation plans by plant. the utilities that do We already talked about have the susceptible not material. There are four plants that have replaced And the material that they're their pressurizer. using in the weld is nonsusceptible material. Inspections have been completed at two plants thus far. Mitigation has been completed at 11. And I'm not going to go through all the statistics because of lack of time. But I think You have that information. this represents to a very this balanced approach executing inspection program. MEMBER ARMIJO: If any of these people find circumferential cracking of a reasonable size, is NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 179 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 there any new requirement to get a bolt sample so you can confirm what the mechanism is? MR. MARION: That's an excellent question. One of the things that we're doing with this program is conducting lessons learned after each of the outage campaigns. And we just completed evaluating potential lessons learned from the Fall 2006 outages. We're going to do the same thing in the spring of this year as well as possibly in the fall of this year. And we clearly recognize that we needed to improve on the communication, the communication from the individual utility at the time that they find an inspection indication or inspection result that calls into question some of the fundamental assumptions we have already made. And we have positioned the industry resources to be responsive to that particular utility so they can do an evaluation and provide some recommendations on what the utility should do going forward. timely. There were communications that were And we're trying to set that up so it's very conducted as a result of the Wolf Creek inspection findings. But, quite frankly, we feel that we can NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 180 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 think one ready to -MEMBER BONACA: important One question I have, I in the timing of improve on that process. place. I will never say never, but I can tell you that we're putting whatever checks and balances we need going forward so that we can identify these findings right away, communicate them to the right technical resources within industry, and then provide some guidance to utility in the middle of an outage so they can make an informed decision. CHAIRMAN SHACK: How about a standby team And so we have that in element inspection would be in my judgment how long has it been since a utility has done volumetric inspection of its own pressurizer flaws? I mean, are you considering that? MR. MARION: Yes. Okay. We've asked the MEMBER BONACA: MR. MARION: Yes. utilities to look at the documentation they may have relative to the fabrication of the original welds as well as the results of inspections that were conducted previously. And we talked about a little bit during Ted Sullivan's presentation on the ten-year ISI. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 181 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 information. As I mentioned before -- and I am going to go through these quickly because I want to try to -MEMBER ABDEL-KHALIK: Excuse me. Would you, then, have a modified graph like the one you have on this previous slide that shows time between the planned inspection and the last inspection? MR. MARION: We can provide that We'll make that a slide for the subcommittee meeting later this month if that's okay. MR. ROBINSON: that line also. Alex, just a point along Part of the reason most are going straight to mitigation, as opposed to trying to do inspection, is simply because many configurations that currently exist in the plants are not inspectable. The current PDI, you know, your protocol, we may have I'm sure also have done inspections, but the question remains how many have done? You have the PDI-qualified inspections, which is the rules we're playing by. MEMBER MAYNARD: CHAIRMAN SHACK: We didn't hang up on him. You are right. The conference lasts an hour, right. MR. MARION: Okay. So I'll move on. Enhanced leakage monitoring. things in place. There are a couple of What the utilities had communicated NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 182 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 developed to the NRC was their current enhanced leakage monitoring program. Now, that program goes well beyond what's in the tech specs primarily because of lessons learned from the Davis-Besse experience. And I have a graphic that will speak to that in a little more detail. Additionally, INPO was conducting review visits of the utility programs relative to managing degradation of primary system components. And one key aspect of that is an effective leakage monitoring program. The data we have collected thus far for the 2007 and 2008 plants indicating that the utilities are taking action up to around .3 gpm, that's .3 gallons per minute unidentified leakage. CHAIRMAN SHACK: .1 to .2, right? MR. MARION: Wasn't that the average? No. I think it was like .6. But Davis-Besse was like I'm sure the NRC can speak to that at the meeting, the next meeting of the subcommittee. The some Westinghouse guidance on Owners' an Group has enhanced leakage monitoring program. being evaluated by And that guidance is currently the Pressurized Water Reactor (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 183 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Owners' Group. And they understand quite well what the NRC staff expectations are relative to the action levels, specifically taking shutdown action at certain thresholds. The way our program is set up, we allow the issue programs the opportunity to determine what positions they want to take that become mandatory for all of the utilities that are affected by that particular program. That's something that's in play. And we expect that to be resolved within the next month or so. But that group is taking a serious look at these programs. This represents the results of a quick and dirty survey we took based upon responses from 44 of the 69 plants. It gives you a range of the thresholds that they have in their programs to date. When we refer to the baseline, each -well, not each one, but there are different baselines that people are using based upon the current conditions or leak rates from the last inspection, et cetera. So it is a little bit of a variable. But And we these are the action thresholds, if you will. will hopefully have more data on this as we prepare for the subcommittee meeting on the 21st. The real big issue between the industry NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 184 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 and the NRC is the fact that we think there is adequate time between leakage and failure of the pipe such that appropriate corrective action can be taken by the utility. We did duplicate, if you will, for lack of a better term, -- and if I'm saying an incorrect term from an analyst's point of view, I expect to be corrected. We did duplicate the NRC analysis and came up with relatively similar conclusions. But we feel that a more detailed analysis would be warranted. And we, quite frankly, believe that it may indicate that there is additional margin between leakage and rupture. Now, the industry is prepared to deal with the results of this analysis. And if the results show there is additional margin, then that information will be provided to the NRC, but if the results show that nothing has changed from what we have already concluded, then the utilities will take appropriate action. The point of doing this analysis is to make sure that we have the best information available to the utilities so they can make the best decision they possibly can as to whether or not they should continue with their current plans to do inspections in NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 185 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 analysis 2008 or possibly expedite those inspections by doing some in 2007. I can tell you right now my own personal opinion, for what it's worth, I don't think all nine plans can do inspections in 2007. have the infrastructure. I don't believe we I don't believe the good conditions will allow it above and beyond what's currently planned for 2007, but that's a personal opinion at this point. I don't know if Glenn wants to add anything relative to this non-linear finite element analysis. We just started the work. We had already indicated to the staff that as we go through this technical work, we will be engaging them and keeping them apprised of what assumptions we're making, what load conditions we're considering, et cetera. And our objective is to try to get this completed midsummer so that we can communicate the results to the utilities again so they can make an informed decision on what their actions ought to be going forward. MEMBER ABDEL-KHALIK: understanding is that the time I guess that my period that the unidentified leak remains unidentified in tech spec is relatively short before the operator if the leak NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 186 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 remains unidentified for a relatively short period of time, meaning a day or so, then the operator has to take some action. So significantly that shorter time than period the seems to of be any accuracy modeling that you come up with. So I am not sure what are you gaining by sort of sharpening your pencils as far as the models are concerned? MR. WHITE: The main question at issue is whether you're going to have a through-wall flaw that can leak at all before there is a rupture of the weld. If one has a large enough crack that does not penetrate through the entire thickness, that could still cause a rupture directly with no opportunity at all for detection of leakage. MR. ROBINSON: This again is Mike. But I think it's important to point out if you look back up on slide 8, there's a reference to a Palisades and a Tsaruga 2 event. And both of those are in these small bore lines that we're talking about. And what the experience there showed us is we had small leaks that were identified on plant instrumentation and plant walk-downs. And these are There's the same lines that we're talking about. essentially being a very small increment of time NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 187 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 between leakage and pipe failure. So here are two clear examples where that is not the case. MR. WHITE: follow up on Alex. I would add a few comments to The program that we are in now It's a five-month Alex mentioned we just started. program, but within the first month, we will have results. The whole five-month period is to allow time for reaching consensus on assumptions to look at sensitivity cases, to look closely at the conditions for the nine plants that are most at issue that are planning to do mitigation in Spring of '08. So it's a program that is intended to bring in experts within the industry on the NRC side, outside the industry together to look towards bounding calculations and towards consensus. to be another scoping calculation. MEMBER ARMIJO: Exactly what is this It's not intended analysis is expected to change, for example, the geometry of the growing track? MR. WHITE: we are looking at. crack. There are two main things that The first item is the shape of the Previous calculations have assumed it stays as a semi-ellipse and driven by crack growth at the deepest point and the surface point, which were NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 188 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 to change. assumed to have relatively high stresses in comparison to the stresses at other points along the crack front. So, in reality, the crack shape is going And preliminary work indicates it can be a significantly smaller cross-sectional area of that crack when it reaches through-wall penetration versus this semi-ellipse assumption. assumption. CHAIRMAN SHACK: But how are you going to So it's a technical handle the range of residual stresses that you -MR. WHITE: we're looking at. That's the second part that That is to a multi-prong approach. We have done many calculations simulating welding residual stress in the past. We're going to build on that to look specifically at these nine plants at issue. On top of that, we're going to look at more sensitivity cases and then use that as the basis for sensitivity cases, different magnitude, residual stresses, different profiles through the wall, different profiles around their circumference, and to look at enough cases to build consensus that we have sufficient assurance about how these cracks should grow. CHAIRMAN SHACK: I can understand you (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 189 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 getting probablistic results. I have a hard time believing you'll get a bounding result that you can live with. MR. WHITE: Well, we want to have sufficient confidence in our result in order to -CHAIRMAN SHACK: MR. MARION: It will be interesting. In conclusion, we Okay. fully understand NRC concerns with regard to recent inspection results and their basis for extending those concerns to the remainder of the fleet. fundamentally think that the NRC's But we is position extremely conservative. I talked about the letters that utilities have submitted to the NRC. So the NRC now has docketed commitments, if you will, of what the plans are for those utilities to conduct inspection mitigation in 2007 and 2008. As I mentioned before, the program we have laid out in MRP-139 we continue to believe is valid, reasonable, and is responsive to our understanding of this important degradation mechanism. And, lastly, we believe that the plants are still in a position where they can continue to operate safely until the next refueling outage when the inspection and mitigation activity is completed. NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 190 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 needs to And that concludes my presentation. would like to speak to the one question that was raised about opportunities to conduct destructive examination of the Wolf Creek. We have had some discussions with the management of Wolf Creek along those lines. discussions are still in play. suggest any conclusion. I think Wolf Creek's next rescheduled outage is the Fall of 2008 if my memory serves me right. And, as we progress, once a decision is made And the I I am not at liberty to relative to what Wolf Creek may do or may not do, we will be more than happy to communicate that with this Committee and the NRC staff. The decision at this point rests with Wolf Creek management. Okay. That completes my presentation. I will be more than happy to answer any -MEMBER MAYNARD: be factored into One other thing I think this if we look at accelerating schedules is there are limited resources that can do a quality job in both the inspection and especially in the mitigation of these. And I think we need to be careful we don't overstretch the resources. I think it's important to get these things mitigated correctly, rather than just NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 191 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 toss a whole bunch of money or something. MR. MARION: Mr. Maynard's comments. I would just add, speak to We have spoken with key One can always vendors who support these inspections. conclude that you could squeeze another inspection or mitigation activity in in the middle of summer, but the question is whether you can implement that outage in the middle of the summer, when you need the electricity. And so that's one of the -MR. ROBINSON: This is Mike. What we're talking about, to do a typical overlay of these nozzles on a pressurizer, you're talking about a minimum of roughly 30 days from the time you shut the plant down, get it into a condition where you can do the overlay, perform the overlay, perform the work, demode the area, and then put the unit backbone, you're talking about roughly a good 30-day period. And that assumes you don't have any rework or other issues that you encounter as you're going through the project itself. MR. RILEY: There's a myriad of other The dose considerations that come to play here. considerations are one. You can fit so many of these in based on the resources of being able to do the overlay. But these overlays actually hold quite a bit NEAL R. GROSS (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433 192 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 perspective. of dose. And the people who are the folks who do the overlay are -MEMBER ARMIJO: MR. RILEY: Plant to plant. -- limited from that And it can be pretty significant. Another thing that -MEMBER ARMIJO: I just want to make clear basically everybody who is doing the inspection is going to be prepared or plan to do an overlay anyway. MR. MARION: The majority of utilities are There are only two that planning to do overlays. we're aware of who are planning to do inspections. And those are going to be conducted this year, in 2007. MEMBER ARMIJO: So they go in. They do an inspection hoping or anticipating there would be no findings of concern. MR. ROBINSON: I think what you would find is that the smart way to plan these if you just plan to do the inspection is you do have a contingency to bring in a vendor should your inspection results find something. So I don't think anybody would plan to do an inspection without having a pretty well-thought-out and planned overlay as a backup. MEMBER ARMIJO: That's what I expected. (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 NEAL R. GROSS (202) 234-4433 193 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS I'm sorry. original I just want to make sure. MEMBER MAYNARD: plan had been I know at Wolf Creek, the to inspect and have a contingency plan. Actually, the cost of having resources standing by turned out to be about as much as going ahead and planning the mitigation. So I believe they made the decision to go straight to mitigation because it didn't cost that much more. MEMBER ARMIJO: Okay. If there aren't any other questions, Mr. Chairman, it's all yours. CHAIRMAN SHACK: a good presentation. Thank you very much for It's Let's see where we're at. back to you, Otto, for our work on the reg guides and SRP sections, our favorite topic. MEMBER MAYNARD: Our favorite topic here. Do we need the recorder? CHAIRMAN SHACK: We don't need the recorder any more this afternoon. (Whereupon, the foregoing matter was concluded at 2:19 p.m.) (202) 234-4433 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. WASHINGTON, D.C. 20005-3701 (202) 234-4433