ML20205F612
| ML20205F612 | |
| Person / Time | |
|---|---|
| Issue date: | 03/26/1987 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1578, NUDOCS 8703310217 | |
| Download: ML20205F612 (290) | |
Text
_ _-_
OgIGINAL .
UlN H E.D STATES
- o'"
1 NUCLEAR REGULATORY COlW4SSION P
k,
- <- . u-- -=,t === ',
IN THE MATTER OF: . DOCKET NO:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS t
u.
j SUBCOMMITTEE ON METAL COMPONENTS ',
v < \
4 O
v .
LOCATION: WASIIINGTON, D. C. PAGES: 1 210 DATE: T!!URSDAY, MARCH 26, 1937 f
_ . - - - ._ _c
- =. _. -
lih i
iI .
30 V0:: 30m0Ve vom Acas6ivice. ;
ACE-FEDERAL REPORTERS, INC.
n Official Reporters
444. North Ca itol Street '
1 I
D\
Washington, .C. 20001 l (202) 347 3700 0((3 M ,
r l ', n NA11C uwIDF. COV ERACE m ., nk - -- - ~
I;/
,- PUBLIC NOTICE BY THE
' ' k UNITED STATES NUCLEAR REGULATORY COMMISSIONERS'
, ADVISORY COMMITTEE ON R.?. ACTOR SAFEGUARDS
, ,/
THURSDAY, MARCH 26, 1987 The contents of this stenographic transcript of the proceedings of the United States Nuclear Regulatory Coinmission's Advisory Committee on Reactor Safeguards (ACP.3), ac reported herein, is an uncorrected record of the discussions recorded at the meeting held on the above
,.. date.
No member of the ACRS Staff and no participant at P- ,
this meeting accepts any responsibility for. errors or q,3 inace'praciesofstatementorfatacontainedinthis a
tranncript.
4 {'
f i
k Ii I.
j
(.?
.l
, ,. (
r i t i
O 4 - ,
[
CR30335.0 i BRT/Ojg 1
,, 4,. 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION l
3 ADVISORY CO'MMITTEE ON REACTOR SAFEGUARDS 4 SUBCOMMITTEE ON METAL COMPONENTS I
Nuclear Regulatory Commission l Room 1046 y 6 1717 H Street, N.W.
~*
Washington, D. C.
7 s l l 8 Thursday, March 26',' 1987-f i 9 l The meeting convened at 8:30 a.m. '\
10 11 ACRS MEMBERS PRESENT:
DR. PAUL G. SHEWMON 12 g . MR. CARLYLE MICHELSON g . . .
MR. DAVID A. WARD 14 '
15 ,
ACRS CONSULTANTS: ,
16 M. BENDER H. ETHERINGTON 17 T. KASSNER
' E. RODA'BAUGH -
W. SHACK 18 s
19 l' 20
(
21 22 )
23 ,
24 O 25 ACE-FEDERAL' REPORTERS, INC.
? -> L --- ~ ~ ~c - - ~~
130335.0' 2 BRT E
r- ,
l 1 P,R O C E,E D_ I_.N__G S, 2 DR. SHEWMON: The meeting will come to order.
-3 .This is a meeting of the ACRS Subcommittee on Metal 4 Components. I'am Paul Shewmon, Chairman of the .j 5 Subcommittee. The~other Subcommittee.-- or ACRS~ members in 6 attendance are Carl Michelson. The ACRS consultants j .7- present are Harold Etherington, Ev Rodabaugh, and Tom-l 8 Kassner.
9 The purpose for-this meeting is to discuss the L
10 following matters: Beaver Valley 2 WHIPJET Program, public 11 comments-and their resolution on NUREG-0313,.Rev. 2, and 12 the status of hydrogen water chemistry ani$ its effect on 13' material behavior.
- o 14 Al Igne is the cognizant ACRS Staff member for
~
15 today's meeting. The rules for participation in today's 16 meeting have been announcled as part of the notice of this !
17 ~ meeting and was published in the Federal Register on March .
~
I 18 11,'1987. .
19 It is requested -- we are recording the session, 20 so it's requested that each speaker first identify himself 21 or herself and r. peak with sufficient clarity and volume so 22 they can be readily heard.
23 As of this morning we have received no written 24 comments or requests for time to make oral statements from 25 members of the public.
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
J30335.0 3 BRT.
(~h 1 Any other announcements? If not, we'll proceed G
2 with it, and I'll call upon Peter Tam, licensing project 3 manager, to start things.
4 MR. TAM: Good morning, I am Peter Tam, the 5 licensing project manager for Beaver Valley Units 1 and 2.
6 My introductory talk will just consist of a project 7 management perspective of this program that we called 8 WHIPJET.
9 First of all, I would like to point cut that 10 Beaver Valley is the first plant to try to apply the 11 leak-before-break, LBB technology, to balance-of-plant 12 piping. ,
13 Balance-of-plant has been kind of misleading.
O 14 It really does not pertain to the secondary side. It 15 really pertains to certain piping that was not supplied by 16 Westinghouse, but it's kind of cumbersome to describe it so 17 we typically have been calling it balance-of-plant; but 18 it's really not balance-of-plant in its usual sense. Our 19 staff reviewers later on will probably give more details 20 about what kind of piping was involved.
21 Just a very quick overview of the chronology of 22 events.
23 Prior to September 6, there has been a number of 24 meetings between the Staff and various people talking about 25 application of leak-before-break to piping beyond primary
)
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
r 30335.0 4 BRT.
[')
ss 1 coolant loop. However, on September 6, 1985, the Applicant 2 formally submitted a proposal to propose to use LBB 3 technology.
4 On November 13, 1985, the ACRS, in its letter 5 addressing the application of Beaver Valley Unit 2, said 6 this obviously is a departure from practice but it may have 7 benefits.
8 Finally, on March 3, 1986, the Staff wrote a 9 letter formally agreeing to start reviewing the use of LBB 10 to Beaver Valley Unit 2, and the Staff also indicated the 11 conditions under which exemption to GDC-4 will be granted.
12 The Staff also suggested that the Applicant ,
- 13 would work.very closely with the Staff, and informing the 14 Staff at regular intervals, it was suggested at quarterly 15 intervals, but it turned out we had meetings much more 16 often than that. It turned out during 1986 and early 1987, 17 we have had seven technical meetings, one on-site review by 18 the Staff, one audit carried out at Boston at the Stone &
19 Webster office. The Staff also paid a visit to an EPI test 20 site to observe leak detection tests.
21 After all of that Beaver Valley, th9 Applicant, 22 finally submitted its WHIPJET report in early 1987, and the 23 Staff completed its review and it is documented in this 24 report, supplement number 4 to the Beaver Valley safety 25 evaluation report. That concludes my introduction.
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
[
-30335.0- 5
.BRT
(} l Our lead reviewer is Ted Sullivan, and I believe he will introduce the other. people as we move along.
2 3 DR. SHEWMON: Fine. Thank you.
4 MR. SULLIVAN: My name is Ted Sullivan.
5 (Slide.)
6 I'm a section leader of the materials engineers 7 section in the PWR licensing division A.
8 I would like to give, here, a summary of some of 9 the major steps in the process of introducing 10 leak-before-break into the regulatory process, and I would 11 like to acknowledge on the slide some of the -- not all, 12 but some of the past ACRS participation in this process.
13 In May and June of 1983, there were Subcommittee 14 and ful-1 Committee presentations on the leak-before-break
. 15 resolution of the assymetric LOCA loads issue.
16 In the winter of 1984 and '85, the ACRS heard a 17 presentation on NUREG 1061, volume 3, which was the r 18 culmination of the work of the piping review committee that 19 developed criteria on leak-before-break.
20 I bring this up mainly to build on the fact that 21 this program essentially followed that criteria.
22 In October of 1985, although -- in October of 23 1985 there was an exemption granted on the dynamic effects 24 of postulated pipe breaks in the primary coolant loops. I 25 want to point out that's not part of the WHIPJET program.
O ACE FEDERAL REPORTERS, INC.
202 3d-3700 Nationwide Coverage 800-336-6646
30335.0 6 BRT 1 WHIPJET, as Peter Tam pointed out, had to do with branch
)
2 line piping, but with respect to leak-before-break, the 3 primary coolant loop was treated by a exemption process, 4 and that exemption was granted, as I said, in October of 5 1985.
6 In February of 1986, approximately a year ago, a 7 year and a month ago, there was an ACRS presentation on 8 some of the early status of the Beaver Valley 2 WHIPJET 9 program. That was given by Duquesne Light and their 10 consultants.
11 In April of 1986, what we call the limited scope 12 rule was published and became effective in May. This 7 13 limited scope rule pertains to- PUR primary coolant loop
()
14 piping. So any leak-before-break reviews that we have done 15 or completed since April of 1986 on primary coolant loop 16 piping has no longer required an exemption.
17 Then I wanted to also point out that beyond 18 today's meeting, there will be, we envision, two more 19 meetings on this subject. None of them are scheduled, but 20 we are hoping that they can be next month and the month 21 after. They will cover subjects that pertain to the heavy 22 component support redesign, considerations in 23 leak-before-break reviews, and then, once the NRC has 24 completed its own internal review of the final broad scope 25 rulemaking process, there will be another presentation on n
v ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335'.0 7 CRT:
.( ) 1 all of'the generic aspects of this program related to the 2 so-called broad-scope. rule. So'we are hoping that that 3 will take place in May of this year.
4 DR. SHEWMON: Before we leave that, the April 5 86, then, is then the limited GDC, limited scope?
6 MR. SULLIVAN: Right.
7 DR.'SHEWMON: And the broad scope is yet to be i, 8 issued, although we have seen it in various incarnations, 9 several times.
10 MR. SULLIVAN: That's correct.
11 DR. SHEWMON: Can you tell me where the Staff's 12 interpretation of the change in the arbitrary intermediate .,
1,3 breaks came in-that chronology?
( Bob Bosnak, the.SRO.
14 MR. BOSNAK: The fi'rst 15 plant we approved for arbitrary intermediate break approval 16 was Catawba 2 and that was in the time frame of April-May, 17 '84. All the plants have been licensed since that time, 18 have applied for and have received the arbitrary i 19 intermediate break change; and there is a standard review i-l 20 plan which I think you have seen, it's in final form, it's 21 due to be published within a.few weeks, and that will 22 modify SRB.3.6.2 to eliminate the arbitrary intermediate
. 23 breaks.
24 MR. MICHELSON: Relative to the -- this question 25 of the broad scope rule, I gather you are already allowing 1
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6M6
1 30335.0 8 BRT 1 in essence certain aspects of the broad scope rule to be 2 implemented? Or have there been cases where you have been 3 allowed leak-before-break outside of containment?
4 MR. BOSNAK: We have not. The breaks apply 5 across the board, inside and outside.
6 MR. MICHELSON: I understand that, but on the 7 broad scope rule --
8 MR. BOSNAK: The broad scope rule, this is the 9 first plant we are talking about and this is all still 10 limited to inside containment.
11 MR. MICHELSON: So you are saying there have not 12 yet been any cases of people applying for leak-before-break
, . 13 philosophy for outside of containment?
(' ')
14 MR. BOSNAK: That's right. .
15 MR. MICHELSON: Thank you. That will be a part 16 of the meeting I guess we are going to have pretty soon, 17 though. Is that coming up pretty soon, Paul? The broad 18 scope rule?
19 MR. SULLIVAN: That meeting will be scheduled as 20 soon as the Staff has --
21 DR. SHEWMON: April-May, '87, it says there. I 22 hadn't heard of it before but --
23 MR. MICHELSON: I knew it was coming up. I 24 thought it kept being delayed and I thought it was delayed 25 again.
C'i V
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-66*6
- 30335'.0 9
- BRT DR. SHEWMON: Bob, take this once more, if you-
{J 1 2 .would, and tell me briefly what the difference -- what the 3 philosophy was on the change in the arbitrary intermediate 4 break ruling? You said before we got started you-thought 5 of that as fundamentally different from the fracture 6 ' mechanics involved?
7 liR. BOSNAK: It is fundamentally'different from 8 the fracture mechanics, in that fracture mechanics are not 9 applied.to the arbitrary intermediate break elimination.
10 What we have there is, currently breaks are postulated 11 based on stress and usage factor and, in addition we have.
12 said there has to be two additional breaks. Even though 13 your stress le. vel may be below the specified limit and your O 14 usage factor may be below that threshold, you had to have N 15 two additional ones. These are so-called arbitrary 16 intermediate breaks.
17 We have, by making this change, this was 18 reviewed by the piping review committee, it was recommended 19 that the benefits to be gained far exceed what you would 20 have by keeping those present -- to have the massive 21 restraints and all the rest of those that go along with the 22 postulated arbitrary intermediate breaks was not in the 23 best keeping of safety of the plant. That's-why the change 24 was made.
25 What we are talking about, now under the limited O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 T
e - -, - , , ,
30335.0 10 BRT
'} :1 2
scope rule and under.the broad scope rule, involves the use of fracture mechanics. And leak-before-break.
3 DR. SHEWMON: Thank you.
4 (Slide.)
5 Yes?
6 MR. SULLIVAN: The criteria that was followed in 7 performing the WHIPJET review is treated in its fullest 8 context in NUREG-1061, volume 3, which I mentioned 9 previously was presented to the ACRS shortly after it was 10 issued. The criteria that's in that report is also the 11 same criteria that are followed in the narrow scope, 12 so-called limited scope rule which has already become p 13 effective, and that was published in the Federal Register-14 notice when that rule was issued in April of 1986. This is 15 the same criteria that pertains to WHIPJET, which is also 16 summarized in the proposed broad scope rule, which was 17 issued in July of 1986, and we hope that's going to come to 18 a culmination this calendar year.
19 (Slide.)
20 In performing the WHIPJET review, one of the 21 first things that was done was to assess each line for its 22 potential for significant degradation during service. This 23 is the so-called screening criteria step, to make sure, as 24 I said, that none of the proposed lines could be 25 susceptible to some form of significant degradation.
O ,
i ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
30335.0 11 BRT In that assessment the lines -- the candidate
-( ). 1 2 lines were evaluated against history of service experience, 3 service cracking, service failures, for PWR -- similar PWR 4 systems. These are basically the criteria that were looked 5 at in the screening. considerations.
6 MR. MICHELSON: Question on water hammer. We 7 have under preliminary consideration a generic issue 135, 8 dealing with steam generator overfill and so forth.
9 If, as a result of working that generic issue, 10 you find at a later date that we have a potential water 11 hammer problem, say in the horizontal run off the top of 12 the steam generator or a typical Westinghouse-type PWR, how 13 will that determination be treated if, in a particular case, b3 14 the utility no longer has restraints to withstand a break .
15 in the line at that point?
16 MR. SULLIVAN: You are talking about in steam 17 lines?
18 MR. MICHELSON: Yes. Off the top of the steam 19 generator, inside of containment.
20 MR. SULLIVAN: Well, so far we haven't had any 21 . applications, so it's not a concern in the sense that --
22 MR. MICHELSON: It isn't necessarily a 23 hypothetical question, by any means. There's a good 24 possibility that if we have a water hammer problem at all 25 it could very well be in that horizontal run. Also,
()
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M6
30335.0 12 BRT 1 generally those horizontal runs are in very close proximity 2 to the containment, a few feet, 2 or 3, or less.- There's a 3 very good chance if you would rupture there, you'd lose not 4 only the steam line but you'd lose primary containment. If 5 this is in conjunction with a steam tube rupture that 6 caused the overfill that led to the break, you'd have a 7 steam tube rupture without confinement.
8 MR. SULLIVAN: I guess the only thing I can 9 offer is when and if we receive --
10 MR. MICHELSON: The thing that puzzles me is 11 that we say, all right, today there's no water hammer.
12 Based on our knowledge today. That's simply because we
~
j_ 13 haven't developed the understanding o'f how these water
~
14 hammers might go.
15 If at a later date we develop the understanding, 16 is that a backfit, then, to go back and ask them to put 17 appropriate restraints on that line? How will it be 18 treated?
19 MR. SULLIVAN: I guess we'd try and treat it up 20 front in the particular area you cited.
21 MR. MICHELSON: Clearly you can't yet, because 22 you don't have the information with which to treat it. We 23 can't treat it today.
24 MR. BOSNAK: One thing you could do -- I mean 25 this is an area that people know, have a potential for
(~h G
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
30335.0 13 BRT 1 water hammer. If someone were applying, even now, to do (V~'1 2 that, these kinds of questions would be asked.
3 MR. MICHELSON: But on today's knowledge we 4 don't know how to answer. You know we don't know whether 5 there's a water hammer problem or not.
6 MR. BOSNAK: We know there's a potential for 7 water hammer there.
8 MR. MICHELSON: Is that good enough?
9 MR. BOSNAK: In the screening, if there's a
' 10 potential and the potential is fairly high, that would be 11 enough to rule it out.
12 MR. MICHELSON: If we knew how to decide, if it
, 13 was f airly high, I'd agree with you.
14 MR. BOSNAK: That's the thing that we haven't 15 settled on.
16 DR. SHEWMON: Are there any lines in a 17 pressurized water reactor system where creep is a design 18 consideration?
19 MR. SULLIVAN: I don't think there is.
20 DR. SHEWMON: I don't think so either. I just 21 wondered why you had it on your list.
22 MR. SULLIVAN: Well, I put it on the list 23 because it is a criteria that we would follow in these 24 programs. Highest temperature on any of the lines I am 25 aware of is around 650.
O i
ACEJFEDERAL REPORTERS INC.
202-347 3700 Nationwide Coserage 800-336-6646 1
30335.0 14 BRT 4
1 MR. ETHERINGTON: Does the code presume any m-2 necessary allowance will be made for aging of centrifuga11y 3 cast stainless steel?
4 MR. SULLIVAN: I don't know how the code treats 5 it. Keith, can you address that?
6 MR. WICHMAN: I don't think so. First, there 7 are no cast stainless steel within the lines of WHIPJET.
8 MR. ETHERINGTON: We are talking Beaver Valley.
9 I thought we were talking more generally than that.
10 MR. WICHMAN: I thought this meeting was Beaver 11 Valley. But the code does not explicitly treat, to my 12 knowledge, aging of stainless steel.
,, 13 MR. ETHERINGTON: But it does require plant
( ) -
14 specification mechanical properties?
15 MR. WICHMAN: Yes. We are talking about class 1 16 -- code class 1 systems here. We have to meet the code 17 allowables; specifications and allowable stresses. Yes.
18 And, obviously, you have to meet the material 19 specifications.
20 MR. MICHELSON: Maybe I misunderstood the 21 purposes of our meeting on the Beaver Valley WHIPJET 22 program. I thought it was being touted as an example, you 23 know, a lead proposition, a lead program and, therefore, we 24 view it as a generic possibility, not just Beaver Valley.
25 MR. WICHMAN: Well, it is a lead-in effect, o
V ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6646
4 30335.0 15 BRT
~
/^' 1 because-this is the first time we have gone outside-the A
2 main coolant loops.
3 DR. SHEWMON: He's being a little. legalistic, I 4 -think. Because the same plant has already applied the 5 limited scope GDC-4, and that certainly does include the 6 primary system, and there are these problems.
7 MR. WICHMAN: All of these systems are within 8 -- are code class 1 within the primary system, except 9 portions of lines in the accumulators, which I believe go 10 into code class 2. But they are all inside of containment.
i 11 DR. SHEWMON: But your answer is in the primary 12 system there's no explicit allowance for aging of the cast i
13 components; is that right?
i
(-) - - -
14 MR. WICHMAN: Not to my knowledge. -
15 DR. SHEWMON: Okay. Please go ahead.
16 MR. SULLIVAN: Okay.
17 (Slide.)
- 18 "Some of this information you probably would have 4
19 heard on your trip on Tuesday to the site. This is a list
- 20 of the piping that was originally in the scope of the 21 program. What I've tried to show here with the division 22 there is that the first five systems above the line are the 23 pipes that remain in the program. The other four are 4
24 systems that were eliminated from the program for various 25 reasons. Some of them had to do with screening criteria.
()
ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646 l
. -- -__ _ _ . - , . . _ . - _ . _ , _ . . _ . . , _ . _ . _ . _ _ - . _ _ _ _ _ . ~ , _ _ . _ _ _ _ _ _ _ _
30335.0 16 BRT
( ). 1 Some of them had to do with economics. One of them had to 2 do with leak detection difficulties, and one of them was 3 too small to pass the leak-before-break criteria.
4 I also wanted to point out that originally the 5 program included ferritic and austenitic pipe, that was 6 originally inside and outside containment, and in the 7 original scope the range of pipe sizes was from about 1.5 8 to 16 inches.
9 DR. SHEWMON: Which are in and which are out?
10 MR. SULLIVAN: These are still in the program.
11 DR. SHEWMON: Of the bottom ones,.could you 12 tell me, for example, why main steam was out?
g3 13 -
MR. SULLIVAN: ,
Main steam was eliminated for V
14 economic ' reasons. As I understand it, the whip'restrainca 15 were already designed, the construction was almost 16 completed at the initiation of the program. Duquesne Light 17 made an economic decision that it was more cost-effective 18 to just complete the construction rather than pursue the 19 analytical approach in the program.
20 DR. SHEWMON: RCS stands for reactor coolant 21 system, doesn't it?
22 MR. SULLIVAN: That's right.
23 DR. SHEWMON: What are the small diameter piping 24 lines in that?
25 MR. SULLIVAN: I believe they were vent and i
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 17 BRT f~') drain lines.
v .
1 2 DR. SHEWMON: Okay.
3 MR. MICHELSON: I'm a little confused, the last 4 statement talks about outside of containment. I thought 5 this was not going to apply outside of containment.
6 MR. SULLIVAN: Let me go over that point again.
7 In the original scope of the program, there was piping that 8 was inside as well as outside containment. But after the 9 various considerations, the screening criteria, economic 10 and so forth, the scope of the program narrowed down such 11 that the only remaining lines were inside containment --
12 or are inside. .
13 MR. MICHELSON: What does that last sent'ence
(~) 14 mean, " included"? The exclusion includes?
15 MR. SULLIVAN: Initially.
16 DR. SHEWMON: Initially those were included and 17 then they decided not. So they can, under the general GDC-4, 18 include out-of-containment lines, but the utility decided 19 they didn't want to try.
20 MR. MICHELSON: I was just trying to figure out, 21 make sure I understood the earlier reply.
22 What's the smallest line inside containment this 23 is being applied to?
24 MR. SULLIVAN: It's a six-inch SIS pipe.
25 (Slide.)
O ACE FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-6646
30335.0 18 BRT 1 I would like to summarize now what hardware was m
2 eliminated. The objective of this program was to eliminate 3 the pipe rupture restraints and the jet impingement shields.
4 This also shows what the line sizes were by system for the 5 remaining pipes in this program.
6 At the bottom of the slide, what I'm indicating 7 is what type of materials are in the typing systems that 8 remain in this program. They are all wrought materials, 9 there are no cast piping materials in this material, they
~
10 are all 304, 304N, or 316 pipes with either shielded metal 11 arc welds or sub arc welds.
12 DR. SHEWMON: It's the sub arc welds that have
,~
13 the poorer toughness; is that right?
14 MR. SULLIVAN: That's right.
15 DR. SHEWMON: And the applicant in much of these 16 has used the properties of these poorer welds as the 17 limiting, since they would probably often be in the high 18 stressed areas?
19 MR. SULLIVAN: My understanding is that the sub 20 arc welds are only in these three piping systems.
21 DR. SHEWMON: Which three?
22 MR. SULLIVAN: The bottom three, the surge line, 23 the 12-inch RHS, residual heat removal line, and the 24 10-inch residual heat removal line.
25 DR. SHEWMON: The larger systems.
n U
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 8004346M6
30335.0 19
-BRT (v ) 1 Tell me what the star.means under PRR?
2 MR. SULLIVAN: Right. This is a little 3 confusing. There were no pipe whip restraints on the 4 10-inch residual heat removal line, but there was a break 5 in here that they wanted to get rid of the dynamic effects 6 for, because that break would have required a pipe whip 7 restraint to be placed on one of these lines. They are 8 connecting. This line feeds into the accumulator line.
9 DR. SHEWMON: Sometimes if you get crass about 10 this, or unknowledgeable, you want to talk about tons or 11 cubic feet of steel instead of numbers. Could you --
12 MR. SULLIVAN: I never asked that question.
13 DR. SHEWMON: You are too much the specialist. ~
(,s) 14 I would like to get sometime some idea as to whether --
15 other ways to compare this to what came out with the' 16 limited scope GDC-4, but maybe I can get some of that from 17 the Applicant later. Thank you. Go ahead.
18 The last column on that was what remains? What 19 was JIS? Jet impingement -- okay.
20 MR. SULLIVAN: The JIS stands for jet 21 impingement shield. And the last two columns are a summary 22 of the amount of hardware that did not have to be installed 23 because of this program.
24 DR. SHEWMON: Thank you.
25 MR. SULLIVAN: This is a summary of the actual l (
14CE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-66.M
30335.0 20 BRT
~
) 1 analytical procedures that are used.
2 (Slide.)
3 And the margins that the Staff has in their 4 criteria that appear in the various criteria documents.
5 This criteria is the same criteria that was 6 discussed in the context of limited scope rulemaking. If 7 you have any questions about it, I'll address them, or if 8 you would like me to go through it, I'll address it.
9 Otherwise I'll just go on.
10 MR. ETHERINGTON: I have a question, the matter 11 of general understanding. Would a valve ever be considered 12 the terminal end of a pipeline?
13 MR. SULLIVAN: No. We do not consider valves to 14 be terminal ends.
15 MR. ETHERINGTON: It has to be something more or 16 less anchored?
17 MR. SULLIVAN: That's correct. The terminal end 18 is something that restrains the rotational displacements 19 and translational displacements.
20 DR. SHEWMON: When you say "as-built loads,"
21 they had to go out and establish that the as-built -- what 22 it was, as distinct from the as-designed or on the drawings?
23 Is that it?
24 MR. SULLIVAN: That's correct. What they would 25 have to do in that process is verify that the final design
(^)b
(_
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
1 30335.0 21 l
-BRT j l
1 condition or the final design loads are consistent with the
(~J')
2 as-built configuration.' i 3 DR. SHEWMON: Where does the square root of 2 4 come from?
5 MR. SULLIVAN: I'm not sure I answer that ]
6 question. Keith, can you address that?
7 DR. SHEWMON: Is it just a good number between 1 8 and infinity?
9 MR. WICHMAN: Something close to that, it came 10 out of 1051 volume 3, in the NUREG. It's genesis was 11 probably from Section 11, originally.
12 DR. SHEWMON: Fine.
13 MR. ETHERINGTON: I think it's a factor of two 14 and then K-1 equals stress times the square root of pi A, 15 the square root of 2 comes in there, I believe.
16 DR. SHEWMON: So it's the fracture mechanics --
17 MR. ETHERINGTON: Yes.
18 DR. SHEUMON: Okay.
19 MR. SULLIVAN: One of the considerations, 20 obviously, in leak-before-break, is establishing, as was 21 outlined on the previous slide, what the minimum detectable 22 leak is.
23 (Slide.)
24 What the utility proposed and what we eventually 25 accepted was that the minimum detectable leak was half of a O
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8C4336-6646
.30335.0 22 BRT
()
tm I gpm.-
2 The primary system that would be employed for 3 detecting such leaks would be the containment sump. The 4 sump flow rate is monitored once a shift. That's backed up 5 by_ inventory balancing which is done every 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. The 6 average unidentified leak at Beaver Valley 1 is 7- approximately .3 gpm, and this is consistent with 1 8 experience at other PWR plants. There are operating 9 procedures that require containment entry, if unidentified 10 leakage is equal to or exceeds a half a gpm.
11 These types of procedures are also followed at 12 Beaver Valley 1, and there have been occurrences where the ,
13 operators have gone in to track down sources of leakage at 14 that level.
15 MR. RODABAUGH: Ted, I thought I heard Peter Tam 16 say that you did some sort of test on the leakage rate 17 determination.
18 MR. SULLIVAN: Yes, he did. I'll elaborate on 19 that a little bit.
20 There were some tests that Duquesne Light :
21 sponsored at EPRI. What they had to do with was visual 22 detectability of very small leaks. What they were trying 23 to establish was some sort of a threshold for visual 24 detectability of steam and water dripping out of the line, 25 and I believe that the threshold, the comfortable threshold ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 23 BRT
(~'N 1 was established at about .1 gpm.
'w) 2 MR. RODABAUGH: I was thinking maybe you had 3 somehow tried to test, say put 1 gallon of water on the 4 floor up here and see how soon you pick it up in your sump 5 pump?
6 MR. SULLIVAN: No. They didn't do that. The 7 relationship of that test to the program really had to do 8 with the time during which there was still piping systems
~
9 in the program that were outside containment. But it does 10 feed into this aspect in that at the point where an 11 operator would go in, we believe they would be able to 12 determine small leakages visually.
,- 13 MR. MICHELSON: If this is a hot water line and
\-) 14 so forth, you are probably not going to see any water on 15 the floor, are you?
16 MR. SULLIVAN: What happened in the EPRI program 17 was that because there's so much insulation around the pipe, 18 the insulation essentially captures most of the steam.
19 Some steam does seep out the seams.
20 MR. MICHELSON: But that doesn't necessarily end 21 up on the floor either for a long time. It probably ends 22 up in coolers and places like that, and not on the floor.
23 MR. SULLIVAN: The steam would end up in coolers 24 and then the condensate from the coolers funnels directly 25 to the sump.
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80 4 336-6646
.30335.0 24 BRT
(~T 1 MR. MICHELSON: You have to decide, when you see U
2 condensate from the cooler, which you do all the time to 3 some extent, you have to decide whether this is detectable 4 or nondetectable leakage.
5 If you had a certain amount, you know, of 6 leakage beforehand, how do you know that the leak that was 7 there then isn't just getting a little bigger, and it's 8 still a nonproblem as opposed to a leak in the primary 9 pressure boundary, which may be a serious problem. How do 10 you detect the nondetectable and detectable leakage?
11 MR. SULLIVAN: What came out of the EPRI test 12 was that even at the very small flow rates, on the order 13 of .l.gpm, there was still water that condensed out of the 7_
14 insulation and landed up on the floor. 'So I think that 15 would be one clue.
16 Eventually the water that would be evaporating 17 and being picked up by the coolers would tend to push up 18 that approximately .3 gpm background, and at the point it 19 becomes .5, someone has to go in and start looking.
20 MR. MICHELSON: You might just speculate that 21 the leak that you had before in some nonsafety piece of 22 equipment is just getting a little bigger.
23 MR. SULLIVAN: I think one of the procedures 24 involved in this is that if there is some identified 25 leakage, the utility has to measure it. They have to come i
ACE-FEDERAL REPORTERS, INC.
L 202 347-3700 Nationwide Coverage 800436-6646
)
30335.0 25 BRT 1 up with some means of measuring it when'they go inside
("))
u 2 containment to try and sort it out from the unidentified 3 leakage.
4 MR. MICHELSON: It's not real easy to sort out 5 condensate from an air cooler to decide where it came from.
- 6 MR. SULLIVAN
- Right.
7 DR. SHEWMON: To what extent -- it would seem to 8 me that we are primarily concerned about leaks in the 9 primary system. These leaks will bring with them 10 radioactivity. .
11 How does that tie into this and can one make the 12 argument that the more dangerous leaks are usually found by 13'
()' 14 rises in radioactivity, or isn't that an effective help?
MR. SULLIhAN: Well, there a're two other means 15 of leak detection in the containment. That's essentially 16 what the first line up there was leading to, even though I 17 didn't talk about it.
18 One of them has to do with radiation monitors.
19 Keith, can you elaborate on what the other --
, 20 MR. WICHMAN: There are three systems that are 21 mandated. Among those three systems is a particulate 22 monitor and a gaseous rad monitor, okay; you have your
, 23 choice between the gaseous rad monitor and I understand the 24 containment cooling condensate. Okay?
25 So you've got the sump, a particulate monitor, Q
ACE-FEDERAL REPORTERS, INC.
4 202 347 3700 Nationwide Coverage 800-336-6646 l I
30335.0' 26 BRT and your choice between the gaseous rad waste monitor and
,{ 1 2 collecting the condensate from the coolers.
3 DR. SHEWMON: When you say a choice, does that 4 mean when they design the plant they don't have to put one
.5 in? Or on that morning they decide --
6 MR. WICHMAN: You have to have two systems, 7 three systems, two are mandated by Reg Guide 145, and'you 8 have your choice between the gaseous rad monitor and the 9 cooling condensate from the coolers -- monitoring the 10 condensate from the coolers.
11 You could have, for example, two monitors that 12 would detect radioactivity.
13 MR. MICHELSON: This plant has a very large air
.o ,
14 cleanup system inside of containment which is trying to 15 strip the stuff out as fast as it is leaking in. Have you 16 determined what threshold before you would even be able to 17 detect the activity?
18 MR. WICHMAN: I think I'll turn this over to 19 Mr. Lefave.
20 MR. LEFAVE: Bill Lefave from NRC Staff.
21 There is a threshold for these. They are very 22 sensitive, but they also respond to any kind of change in 23 radioactivity in the primary coolant system, background 24 radiation and all. I don't have the numbers. I recall --
25 I don't know what the numbers are offhand, but I know they O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
es
-= -
33 q y a .q n
,f@ '
s;;;
-30335.0' '
- 27 Mc 1
-BRT- e --
y6 0,- ?,- O; r .
, A f-E il are'wellilessthanhalf'agpm.$omeleakageIsgoingN6 l set d
2 off thb~ alarm over'what-the background setting is. /' y l
3 >
MR. MICHELSON: Where.are they located?' . ; J ,
k
^
-4 N~.zLEFNVE:
R Well, at various pointa in the
{p N w /p pe:
5 containment. ,
s
,, y, 6 MR. MICHELSON: But in the air, not near he u
R .,
7 primary. piping. *
.( .f 8 . M R .' p$F' AVE : I'm not sure er.:actIy whera in the ,'
1 :
9 system they are'. r' -
y ;.
10 MR. ELLIOT: . Gayy.Elliot, . PWRA. '
/, ,
11 In our evaluation there is' radiation detection'J P W 12 and sump monitoring, but we rely very heavily on the s'uing
\
13 monitoring. s n
s ft
-Q. 14 Badically, what we've observed, and this is ..
1 9,
- . 'Y (i 15 basically from the EPRI tests, is'that a .5_ gallons per 1 t
\ s 6 16 minute leak'is 56 enormous leak and wduld not go undetected ,y.
. ,- a 17 . visually. What we are relying on la that the sump,wG1 -- 7 i i 18 water will eve ~ntually r'each the sump, be detected ~2 rom sump 19 measurements and inventory balance,.and then the operators >
20 will _go into the plant and look for the loak. ,
21 MR..MICHELSON: If you have a large identified
.i 22 leak from a water line or whatever inaide of containment, i
23 and then one day you see a tenth of a g increase, do you 24 speculate'the leak from the water line is just getting 25 bigger, or do you speculate there's a leak in the primary
, si 3;
ACE FEDERdL REPORTERS, INC.
,202 347-3700 Nationwide Coverage 800-336-6646 ,
~ ,. _, , , . . . . .-.,,_,....__,_._._,.._.._~_,.._.-...~-.-_,,J
s[b ' "
i, '
30335.0 .
28 B2T !
'* c y gl-'
b 1 pressure line? ' i x
/ f~ ;
(3 g 2 MR. ELLIOT: We have two sources, identified and 3 unidentified -- i
- t. F 4 DR. SHEWMON: You changed the rule. A minute f u "5 ago you said a large leak.
L 6 MR.'fMICHELSON: From the water line.
.1 7 DR.1 SHENMON: Part of it is how long they can 8 operate that way?
9 [MR. MICHELSON: I think they can operate with 10 CC ,
gallons per mi'nute coming all the time; is that right? Is sl that your leakage on identifiable? Yes. 10 gallons per i
- d. 12 minute.
'. x 13 I MR. ELLIOT: If we identify the leak, it then
. 11 'goes into the identified category..
15 MR. MICHELSON: But identified leaks don't 16 remain constant necessarily. A .1 of an increase in a 10 gpm 17 l'eak is not surprising, because the pressures change from 18 water; pressure changes could cause that.
19 MR. ELLIOT: What we are looking for is a leak 20- -over the long term. They monitor this leak to keep track 3 L.I, of it, and as it changes over the long term, they'd have to
, '22 keep track of the unidentified leak and identified leak.
23 When the unidentified leak approached .5 gallons per minute, 124 they'would have to go into containment and identify it.
25 And if it's identified as valves or whatever, it then r~s.
N-] l 24CE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6 4 6 ,
, I
, 'n,'
- 30335.0 29
.BRT 1 becomes identified.
.~ (%J']
2 ,
MR. MARTIN: This is Roger Martin of Duquesne
\
3 Light Company, maybe I can assist from our experience on 4 Unit' l~,A Beaver' Valley.
h.iQ ,
5 We did have a small crack of a drain line which 6 was welded to either a instrumentation or drain line.
7 Let's say a small, 2-inch line on the reactor coolant
~8 '
system in Unit 1.
s 9 We did experience an indication of a rise in the 10 radia' tion monitor that monitors the environment, the air in 11 the containment. We did experience an indication whether ,
! 12 that small leak occurred, and we did identify the~ leakage 13 from the" water that was evident in the reactor containment.
o 14 We also do have the PAS system, post-accident l 15 " sampling system, which monitors the water in the sump.
l 16 That's. continuously monitoring the chemical content and 17 radiation level if it exists, or what level it might exist 18 at when the water enters the sump. So I think we have some 19 ' ' experience from Un'it 1 that might help answer your question.
1 20 MR. MICHELSON: How often do you do that? q ,
21 MR. ETHERINGTPSt T1 3se systems really are for 22 environmental protection.. L.c activity depends on how 1 23 clean the water is, so it doesn't look like a very good 24 quantitative leak detection system.
25 MR. MARTIN: We know the background of what is O
ACE FEDERAL REPORTERS, INC.
! 202-347-3700 Nationwide Coserage 800-336-6646
30335.0 30
-BRT.
[}. 1 going into the sump. If there is some identified leakage.
2 MR. ETHERINGTON: A change.
3 MR. MARTIN: Yes. Wo would see a change.
4 MR. ETHERINGTON: It doesn't look like a good 5 primary measure of detection.
6 MR. LEFAVE: Primary detect it. But quantifying 7 it, then they would have to do a, mass balance to quantify 8 the amount of leakage. They won't lump it in with the 10 gpm 9 that may be coming from this known leak, if it gets to a 10 half a gpm they have to go in and inspect these positions 11 .that fall within the leak-before-break category to make 12 sure it is not from there, and identify where it is coming 13 from. ,
14 MR. MICHELSON: If at any time there is a change 15 in the leakage to the sump, whether it's identified or not, 16 you have to go in and find out; is that correct?
17 MR. LEFAVE: Right. If it gets to a half a gpm.
18 MR. MICHELSON: Even though you suspect it's 19 just a change to your identified leakage, you have to go in 20 and verify what it is.
21 MR. LEFAVE: You can't guess. You may think so, 22 but if it gets to a half a gpm difference, you have to make 23 sure of that.
24 MR. MICHELSON: It's always unidentified until 25 verified; is that right?
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 I - ..
30335.0 31
-BRT MR. L2FAVE: Yes. 'Right.
^ J/ } l i- 2 MR. MICHELSON: That would make it a little 3 -better. That's in the tech specs, I assume?
4 MR. LEFAVE: For 1 gpm it's in the tech specs.
5 Half a gpm it's in the procedures.
6 MR. MICHELSON: I see a shake of the head --
7 DR..SHEWMON: Any change at all still has to be 8 quantified.
, 9 MR. MICHELSON: That's no problem. These are 10 quantified -- -
11 DR. SHEWMON: A minute ago you said any change.
12 You want them to admit to any change, and they are trying 13 to admit to a change of half a gpm.
(2) 14 MR. LEFAVE: That's the th'reshold where they go 15 into containment, half a gpm.
16 MR. MICHELSON: Right. Any change that much.
17 .But they must go into containment and find out why, verify 18 it's just increasing in the identified leakage or whether 19 it is a new leak?
20 MR. LEFAVE: Right.
21 MR. MICHELSON: That is in the tech specs?
22 MR. LEFAVE: Under administrative procedures, 23 not under tech specs.
24 MR. MICHELSON: Thank you.
25 MR. SULLIVAN: The final points I would like to
/~T
, V t
ACE-FEDERAL REPORTERS, INC.
- 202-347-3700 Nationwide Coverage 800-336-6646
30335.0 32 BRT cover are just the Staff conclusions.
) l 2 (Slide.)
3 As I said previously, none of the lines in the 4 WHIPJET program, from a review of the screening criteria, 5 are believed to have a potential for unacceptable 6 degradation.
7 The Duquesne Light Company WHIPJET program 8 followed all of the NRC guidelines for how to perform the 9 analysis and what margins to apply in performing the 10 analysis.
11 They used appropriate materials data which they 12 are going to be going'into in more detail. We just got
-3 13 finished talking about the leak monitoring and detection
! o
~
14 procedures. The computer programs they used were 15 benchmarked'. They used programs for leafyge'through the 16 cracks that was part of the@ analysis and the fracture 17 mechanics stability codes; confirmatory stability 18 calculations by the Staff verified the WHIPJET results.
19 We got the loads and so forth, the pipe material 20 properties, and performed our own fracture mechanics 21 calculations which verified the WHIPJET results.
~
f 22 And so, in conclusion, what we have recommended 23 is that an exemption be granted to eliminate the dynamic 24 effects of the postulated pipe breaks in the five systems 25 that are part of the WHIPJET scope.
1
,r%
k l
2
/\CEJFEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800 336-6646
30335.0 33 ERT
('T 1 DR. SHEWMON: Thank you.
v 2 MR. MICHELSON: Can I ask a general question?
3 What you've done here is taken out certain restraints and 4 so forth, because you'll be able to' detect the leaks before 5 pipes break. This must have caused some small' change in 6 the probability of having -- I guess it's beyond the design 7 basis accident -- in other words, our design basis now 8 shifts from accommodating breaks to not accommodating 9 breaks, because we took the restraints out that normally 10 would accommodate it. So we are going into a regime of 11 design basis accidents, if we should for some reason have a 12 break anyway. That is in the severe accident policy that 13 we treat these kinds of accidents.
_ C_'>
14 To what extent now, from.the viewpoint of the 15 severe accident policy statement, have you reviewed this 16 program to assure that you still don't have possibility for 17 catastrophic accidents of the kind that ruptures the type 18 as well as containment, and getting into what we call 19 severe accidents? Have you looked at it from that 20 viewpoint at all?
21 MR. BOSNAK: Well, the probability of having 22 pipe breaks, this was one of the tenets in the broad scope 23 rule, is such that we are well beyond the range of a severe 24 accident.
25 MR. MICHELSON: You are beyond the 10 to the O
V ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0- 34 BRTL
( l .minus 6frange probability?
2 DR. SHEUMON: 10 to-the minus 12.
3 MR. BOSNAK:- Somewhere up in that range.
4 DR. SHEWMON: That's why they came back-and said, 5' hey, if the steam generator falls over, that's serious, but 6 if you have cracks --
a_ 7 MR. MICHELSON: There are PRAs, where you
'8 initiate breaks in the probability area of 10 to the minus 9 4 --
10 MR. BOSNAK: Those are still there, the low 11 probability, we are talking about the areas where we have 12 not eliminated breaks. We have retained all of the 13 environmental qua,lifications that are required to deal with
.o 14
~
such things.
15 What we don't have now, and we believe we don't
, 16 need them, we have improved the safety of the plant, 17 are these large pipe whip restraints, jet impingement 18 barriers --
19 MR. MICHELSON: These were protecting primary 20 containment, in some cases. -
21 MR. BOSNAK: But only if the pipe broke in that 22 exact location.
23 MR. MICHELSON: In some cases, that's a long 24 range, like the main steam line wrapping around containment.
25 There are quite a few changes of position and so forth,
()
- /\CE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
. - _ . - - - . . , - , - _ . - . . - - - . - . _ . ~
30335.0 35 ERT
- /s 1 elbows and transitions and whatever.
G 2 MR. BOSNAK: Of course the probability of having 3 an instantaneous guillotine break for which --
4 MR. MICHELSON: That's why I'm asking. Have you 5 done these probability numbers, such that show that, even 6 under the severe accident policy, it's not a problem?
7 MR. BOSNAK: The work that some of our research 8 consultants did does confirm that these are very low 9 probability kinds of things.
10 MR. MICHELSON: I would probably agree with that.
11 I'm just trying to get a feel, though -- see, there's more --
12 you have to look at the probability of the occurrence and 13 then.the consequence of the occurrence. The' consequences 14 in some cases would be very severe. For instance, main 15 steam line rupturing containment at the same time, which 16 might have been initiated by a steam tube rupture to begin 17 with, and steam generator overfill, and from the severe 18 accident policy viewpoint I think these accidents are less i 19 than probable than the one in a million --
20 MR. BOSNAK: That's right.
21 MR. MICHELSON: And you have gone back and 22 assured yourself that even with the restraints and so forth 23 removed, they are still in that low probability range?
24 MR. BOSNAK: That's correct.
25 MR. MICHELSON: Okay. Thank you.
l ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 36 BRT
(')
%./
1 MR. : SULLIVAN: Roger?
2 MR.. MARTIN: My'name is Roger Martin and I'm 3 manager of regulatory affairs.for Duquesne Light Company.
4 We appreciate the opportunity to have members of the 5 Subcommittee visit the Beaver Valley site last Tuesday.. I 6 think it gave us an opportunity to share with them the 7 magnitude of some of the pipe break restraints that we were 8 able to eliminate under this program.
9 Our wish is that we had been able to undertake 10 this program earlier than we did; in the past 12 months or 11 so that.we have been having conversations with the 12 Commission, we have all learned a great deal, I believe. I 13 hope the benefit of this increased, knowledge will be O 14 utilized throughout the industry.
15 At the risk of being a little redundant, I'll 16 quickly go through some of the highlights which I would 17 like to address.
18 (Slide.)
19 I think that we have discussed that we feel the 20 WHIPJET program justifies exemption f rom the GDC-4, the ,
l 21 double-ended guillotine break, and that it follows the ,
l 22 guidelines of NUREG-1061.
23 The program used deterministic fracture 24 mechanics analysis in the leak-before-break methodology.
25 A critical crack size was determined, stable O
V ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
~130335.0 .37
.BRT > .,
- ' l' crack size , and the . associated _ leakage was identified.
2 This tied in:withLthe safety margin of 10, so that if you 3 had a detection capability of one half gallon gun: minute, f
-4 you needed to assure yourself that you would have a crack 5 size of 5 gallons- per minute, so that you would be sure to
}
6 detect this' leak and .have adequate amount of leakage.
7 That's why some of the smaller pipe sizes were eliminated 8 from the program.
9 We addressed only austenitic pipe inside 10 containment, 304 and 316 in the final phase. I.think the 11 situation that you realized was that we were continuing to y
. 12 build the plant. We had our construction on schedule.'
13 Some of the restraints had already been installed, and we
(:)
14 felt that economically, and for the benefit of all who were 15 trying to come to an acceptance of this proposal, that it' 16 would be best to confine oursolves to_those items which 17 could be addressed most readily.
18 The line sizes considered were from 6 to 14 19 inches. There's.no change in the considerations that 20 result for environmental qualification when you consider 21 the LOCA, and the ECCS requirements still being the same.
22 Industry test data for the austenitic feel was 23 used for the fracture toughness reference, and the items
- 24 previously addressed by Mr. Sullivan were f atigue, 25 corrosion, water hammer, the erosion, creep, age, flow
()
l ACE-FEDERAL REPORTERS, INC.
1 202-347-3700 Nationwide Coverage 800-336-6646
- -- ~ ,,,-,,,-,,,-.w. .. ,n., ,-,ne--,--,,--e,n.,---,r.,,,,e.,r--,->,,+,,,,,--r.r-am+nn*-,,,,.e,m4,,-,,.n.wgrn,,r. n vam---,.<-
30335.0 38 BRT
~
l stratification, and stable crack growth was assumed.
a 2 (Slide.)
3 Briefly, some of the benefits that we feel might 4 occur from this proposal are that the engineering and 5 design process is simplified. I think your visit at the 6 site indicated the complexity of some of these restraints.
7 It reduced the effects on building design, of 8 course. There was a direct cost reduction. There was 9 reduced engineering ef fort, including hazards analysis 10 consideration. We originally would have had to install 67 11 protective devices. Under the WHIPJET program, 27 devices 12 are required to be installed.
gs 13 DR. SHEWMON: A pro device is a shield or 14 restraint?
15 MR. MARTIN: Shield and restraints, that 16 includes both jet shields and restraints.
17 Fewer complex intermediate structures based upon 18 your device reduction. Of course the construction effort 19 would be reduced, and there was a marginal net cost 20 reduction for Beaver Valley 2. But anyone undertaking it 21 at an early stage in plant construction, I think, would 22 realize appreciable savings above what we were able to 23 realize.
24 The indirect cost reductions, of course there 25 are reduced obstructions; the facilitation of the O
(_/
ACE-FEDERAL REPORTERS, INC.
l 202 347 3700 Nationwide Coserage 800-336-6646
'30335.0 39 BRT l . in-service inspection program; there was improved ALARA
(}
2 ' exposure for normal _ operation and maintenance.
3 One of the things that we have experienced is 4 the maount of manpower required during refueling outages
- 5 for the in-service inspectiop program; that's a major 6 requirement for personnel to inspect welds under the 7 program that we have in place, many of those areas are 8 somewhat inaccessible, and it takes time for the inspection 9 personnel to move their equipment there and reach these 10 areas that need to be reviewed.
11 There would be elimination of spurious 12 mechanical interference, between the conduit, the piping,
.- 13 the cable trays, both the mechanical and electrical, devices, 14 all must be fitted into the containment cubical areas. And 15 cost reduction in the engineering against the causative 16 factors which are better than assuming failures. That i
17 would be an indirect cost, the design effort required for 18 that particular activity.
19 DR. SHEWMON: Before you leave that, let's go E 20 back to the top again.
21 MR. MARTIN: All right.
22 DR. SHEWMON: If you were going to do the same 23 thing for the limited scope rule, anything you say there
! 24 you could probably say again but, in number 2, the number ;
25 of devices would be smaller, but they would all be bigger, ;
! (
4
/\CE FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800 336-6M6
- . _ __i
30335.0 40 BRT
(~ l or what other changes would there be as you go through 2 there?
3 MR. KARTIN: Under the reactor coolant limited 4 scope application?
3 DR. SHEWMON: Yes. ,
6 MR. MARTIN: The fracture mechanics 7 consideration was not the basis for that determination.
8 Maybe I don't understand your question.
9 DR. SHEWMON: Maybe I don' t know how to pose it.
10 There were two stages of this, at least in my 11 mind. There was the limited scope gdc-4, which applied to 12 the primary system. ,
,- 13 MR. MARTIN: Primary; yes.
( )
'~
14 DR. SHEWMON: And there's the WHIPJET, which you 15 are primarily interested in talking about today.
16 MR. MARTIN: Yes.
17 DR. SHEWMON: If we went back and we went 18 through what changes there were in your plant when the 19 limited scope rule started being applied, you would have 20 less engineering effort to do hazards evaluations? Or 21 doesn't that -- isn't that a significant change there?
22 MR. MARTIN: Yes. That's right. You would have 23 fewer.
24 DR. SHEWMON: Fewer protective devices?
25 MR. MARTIN: Fewer protective devices would be o
s_ ,
1 ACE. FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-6646 l
,30335.0 41-
- CRT l required.
2 DR. SHEWMON: Probably the numbers wouldn' t' be
.' 3 a.4 big, but the protective devices would be bigger?
~4 -MR. MARTIN: The protective devices basically on 5 -the primary coolant system would be quite massive. They 6 'would be larger than some of the WHIPJET restraints that 7 had been eliminated.
8 DR. SHEWMON: What's an intermediate structure?
]
9 MR. MARTI'N: Those intermediate -- the 10 supporting devices for -- in between the --
- 11 DR. SHEWMON: The restraints and the structure.
12 Fine. I guess I see what I need.
13 MR. MARTIN: You need to find a wall to' tie in L
o 14 the base plate, and also your structural steel, which in l 15 turn holds your restraints. So you have to find a place to j
- 16 tie it in, to a beam or concrete wall location.
t 17 DR. SHEWMON: Thank you.
l 18 MR. MICHELSON: Fewer protection devices, the
- 19 ones that you removed, were any of them protecting primary 20 containment as a target?
21 MR. MARTIN: Some of them served a dual function, 22 which would be pipe support and break restraint. Those >
23 that had to be in there for pipe support would remain.
< 24 MR. MICHELSON: That wasn't quite my question.
i- 25 My question was, of the devices you removed, did any of
,i O
I'
- ACE FEDERAL REPORTERS, INC.
j 202-347 3700 Nationwide Coverage 800 336-6646
3033520 42 BRT l them protect primary containment?
'}
2 MR. MARTIN: No.
3 MR. MICHELSON: These are inside the shield wall, 4 so primary containment wasn't a question?
5 MR. MARTIN: Yes.
6 MR. MICHELSON: Okay. Thank you.
7 (Slide . )
8 MR. MARTIN: Where we believe the benefits from 9 this program can be realized, in the long term. One of the 10 benefits is that the program provided technical data in 11 support of the NUREG-1061, and implementation for all high 12 energy lines. We feel that this would enable the Staff to
, 13 review any future applications and use the technical data
- l - .
14 that has been provided to answer the questions that have 15 been addressed here previously.
16 It did justify the elimination of 32 restraints, 17 pipe restraints -- break restraints, and 8 jet shields.
18 Plants with longer construction lead time would 19 expect a larger benefit, we feel. And there is the 20 potential for application to carbon steel lines outside 21 containment.
22 Once again, in the interest of completing the 23 activity, we moved ahead with the construction program on a 24 risk basis during the year of discussions 'and providing of 25 technical data and the development of the test program that (b ,
kJ l
l l
/\CE-FEDERAL REPORTERS, INC. l 202 347 3700 Nationwide Coverage m) 336-6M6 l l
30335.0 43 BRT
-1 EPRI provided for us. ;
} ,
2 There was a'need to go ahead with.the design and 3 installation of some restraints.
4 Where we could retire a restraint in place we 5 did, and that would provide an advantage during the 40-year.
6 life of the plant, even though the internal structural
. 7 configuration might have been complicated by a restraint 8 that remained in place, the shims and the need to check 9 that restraint at each refueling outage was eliminated, and 10 that would reduce the operator maintenance personnel 11 exposure to irradiation in entering the plant during the 12 refueling outages.
13 Once again, I think that we have mutually 4
O 14 benefited from this program and 'we hope its information 15 will be valuable to the industry.
16 MR. MICHELSON: Question. Your last item
- 17 suggests that later on you might consider going outside of
- l. 18 containment.
19 In you go beyond what's proposed in front of us 20 at the moment, do they have to come back to NRC with each 21 and every addition, or is there kind of a blanket --
22 MR. BOSNAK They would have to come back.
L 23 MR. MICHELSON: I'm talking about just this 24 Applicant.
- 25 MR. BOSNAK
- Even this Applicant, if they wanted I
! /\CE. FEDERAL. REPORTERS, INC.
202 347 3700 Nationwide Coserage 800-336-6M6
_m ,. ,.,,...,,.._,e~. r....-y,.,-,3- , m.y,,,,_ , , - . . , .e% ,.,,-..._,w,.,,.,,,.y,~. w-,-on-,e-+.w.,-..-.,.,.,..-..,,,m,.,,,-,_-.+,-----,---rw-w-y-e-.-~
30335,0 44 BRT 1 to go further than they have, they'd have to come back.
2 MR. MICHELSON: They have identified which ones 3 they are removing and that's it?
4 MR. BOSNAK: That's it for now.
5 MR. MICHELSON: You'd have to come back and .
6 review anything further, including outside of containment?
7 MR. BOSNAK: Including outside. Yes.
8 DR. SHEWMON: How would you handle leak rate 9 measurements outside containment?
10 MR. MARTIN: That was one of the reasons for the 11 EPRI study. Maybe Jim Szyslowski could address this? Do 12 you want to come up here, Jim? Jim Szyslowski, Duquesne 13 Lig ht .
t
'a 14 MR. SZ YS LOWS KI: EPRI ran.the tests for us where 15 we could detect .1 gpm outside containment if we wanted to.
16 DR. SHEWMON: How?
17 MR. SZYSLOWSKI: Visually. The test showed not 18 only water was collecting at the bottom on the floor, but 19 also you could see the steam escaping from the seams of the 20 insulation. So that was what we were going to do when we 21 had the steam generator blowdown lines in the WHIPJET 22 program.
23 But then it got later and later in the program 24 and there were other problems with trying to cover them, 25 like with the crosion, corrosion considerations as well as n
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 800 336-6646
30335,0 45 BRT
. (~) 1 economic. So eventually we decided to drop that from the v
2 program.
3 DR. SHEWMON: Okay. Thank you.
4 MR. ETHERINGTON: You haven' t justified 5 elimination of restraints on a branch pipe, do you consider 6 a break in that branch pipe as imposing a load on the main 7 pipe?
8 MR. MARTIN: Let me refer that to Stone &
9 Webster.
10 MR. ETHERINGTON: Was the answer no?
11 DR. SHEWMON: He referred it to Stone & Webster.
12 MR. MARTIN: Would you repeat the question?
f- 13 MR. BOSNAK: I think I can answer that. If the
(') 14 load is there -- if the break is there, the loads still 15 have to be assumed. In other words the break is not 16 eliminated --
17 MR. ETHERINGTON: I feel that should be the case.
18 I wasn't sure whether it was taken into account.
19 DR. SHEWMON: Okay. Thank you.
20 MR. MARTIN: Ih clarification to your comment 21 about would we consider carbon steel outside containment, 22 in view of our intent to load fuel May 21st, I feel that 23 this may be a future consideration, but at the present time 1
24 we need to wrap up this particular phase.
25 Thank you.
O ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646
3033520 46 BRT 1 MR. ROEMER: I'm Bob Roemer, Stone & Webster.
2 (Slide.)
3 Basically, my presentation is now redundant 4 almost totally. I just want to make some clarifications on 5 questions and points that have been brought up before.
6 I think the first pertinent clarification is the 7 scope of pipe rupture hardware throughout the plant, and 8 where the various elimination ef forts have come to bear and 9 what the summary results of all the efforts have been.
10 This is a pie chart that represents all the 11 potential pipe rupture hardware that would have been in the 12 plant had we not implemented the limited GDC-4 exemption,
.s 13 the arbitrary or intermediate break relief, and the WHIPJET
) .
14 program. -
15 The various sectors of the pie represent the 16 various reductions as a result of that.
17 Now, it is, perhaps, misleading. You'll notice 18 the largest sector here is the sector for the arbitrary 19 intermediate breaks, which is shown to have potentially 20 eliminated 127 pieces of pipo rupture mitigation hardware, 21 either restraints or jet impingement shields.
22 This was considered at the time, before we had 23 done very detailed evaluation of pipe rupture target 24 interactions from the safety implications, and it's 25 probably a very generous projection of what would have been
(~'y
's j Ace FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-(M6
3033520 47 BRT 1 eliminated, much of that probably would have been
. x_
f~'/i 2 eliminated downstream by more detailed analysis.
3 As you can see, in the primary loop we did 4 eliminate 12 pipe rupture restraints as a result of the 5 narrow scope rule, program.
6 In answer to a question that Chairman Shewmon 7 had about the relative size and number of pieces of 8 hardware that were eliminated, the number, as you can see 9 in the primary loop, is relatively small. The size was 10 massive, very massive indeed.
11 In addition to simply being very, very large and 12 among other things providing an enormous heat sink,
- 13 eliminating the thick insulation.that would have been O 14
- appropriate at the points at which they were connected, and l
l 15 also having very, very small gaps, they were a source of 16 great confusion to maintenance personnel.
17 The remaining what we call balance-of-plant 18 restraints originally, as Roger Martin described, were 67, 19 encompassing the red sector and the blue sector, of which
- 20 40 were eliminated by the WHIPJET program.
21 The number is larger and on the average the size 22 is smaller, but some of the intermediate structures, not 23 the pipe rupture restraints themselves, were as massive in 24 some cases as some of the primary loop restraints. So 25 there's a very, very significant effect in terms of simply
(
/(CE. FEDERAL REPORTERS, INC. l
, 202 347 3700 Nationwide Cmerage 800 336-(M6 j l l
30 335.0 48
~
BRT
(')
x_-
I hardware mass there.
2 We are lef t total in the plant with 27 devices, 3 which is, as you can see, a reduction of about a factor of 4 8 from where we originally projected to start off.
5 DR. SHEWMON: Tell me about,the blue? The 6 balance-of-plant is things which are inside containment but 7 outside the primary? Or what?
8 MR. ROEMER: Both. Inside containment and 9 outside containment -- those that remain on the main steam 10 and feed water lines outside containment and inside other 11 buildings where, for a plethora of reasons, it was not 12 appropriate, because of time, economics, and technology, 13 and' regulatory posture,' to eliminate.
I) 14 DR. SHEWMON: Tell me what I'm supposed to be 15 seeing here? The arbitrary intermediate breaks are the 16 total number there or eliminated?
17 MR. ROEMER: None that were eliminated. All 18 were eliminated. All arbitrary intermediate breaks at all 19 locations in the plant were eliminated by that.
20 DR. SHEWMON: Are those in blue over there?
21 Were they eliminated?
22 MR. ROEMER: No. Those remained using the old 23 3.6.2 criteria.
24 DR. SHEWHON: The blue remains?
25 MR. ROEMER: Correct. 26 restraints total.
O Ace FEDERAL reporters, INC.
202 347 3700 Nationwide Coserage 80r) 334 6646
30335 0 1 49 BRT
? .- 1 MR. MICHELSON: I think you said balance-of-plant includes outside containment,f and you made 2
3- it clear you are not going 'outside of containment.
4 MR. ROEMER: That's confusing. The 5' balance-of-plant hero.in the red, all of those 6 WHIPJET-eliminated pipe rupture mitigating devices were 7 inside containment. The remaining blue here were both 8 inside and outside containment.
9 Now, as far as the process is concerned, the 10 method of the overall program of pipe rupture-tying you 11 would melt into WHIPJET, we used a traditional hazards 12 evaluation.
13 (Slide.)
C) 14- 'We first, started with high energy systems in 15 accordance with 3.6. 2, worked our way through the pipe 16 stress analysis to establish the break locations, did a 17 target review, and this was the determination, development 18 - of the determination of the need for hardware, did a 19 shutdown review to determine if the safety related 20 equipment were required to mitigate the consequences of 21 that particular accident in conjunction with an ECCS review, 22 and that determined where we needed hardware.
23 Based on that, that was the criteria on which we 24 went into the WHIPJET program on a system basis to 25 eliminate the various hardware. Once we get into the O
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Cmerage 800 3366M6
3033520 ,
50 B. T 1 WHIPJET program, we did as both Ted Sullivan and Roger 2 Martin mentioned; we did a screening review. First of all, 3 material review, transient review, vibration review, 4 corrosion review, and then we went into the fracture 5 mechanics analysis, the crack growth analysis, which 6 ultimately led in the particular case of a leakage size 7 crack into a comparison with the leak detection 8 capabilities and the leak rate analysis.
9 Now, if all of that was found acceptable, then 10 we were able to eliminate on a system basis the restraints 11 inside containment, and that eliminated the red sector I 12 showed on the ' previous pie graph.
13 (Slide.)
14 Again, to summarize, where we started in the 15 orig'inal candidates for the WHIPJET program and where we
- 16. .
ended up, these were the 12 systems we started off with.
17 These were the candidate systems that we looked at, because 18 there were potentially hardware on them, and we had not 19 done detailed analysis, and they seemed like appropriate 20 systems in which to implement leak-before-break.
21 We had, basically, three methods of determining 22 their acceptability. First of all, if there were no 23 unacceptable interactions as a result of pipe rupture, then 24 the discussion of leak-before-break became totally academic, 25 there was no need for hardware, thus no need to eliminate m
ace. FEDERAL REPORTERS, INC, 202 347 3700 Nationwide Coserage 800 336-6M6
30335,0 51 BRT 1 hardware. So we have shown on the various systems in which 2 we indicated no unacceptable interactions, they immediately 3 dropped out the program, and there was no need to go into a 4 screening analysis. This was cost-ef fective Harper logic 5 because we had very little time to go into the program.
6 On the BDG, we discussed that previously, it had 7 two significant problems. One of which was it was carbon 8 steel, and it was outside containment and inaccessible in 9 terms of visual detection of leaks outside containment.
10 DR. SHEWMON: Before you leave that, why is 11 carbon steel a strike against it?
12 MR. ROEMER: Because of the data we had at the 13 time. It was the first time we had introduced carbon steel 14 fracture mechanics. Obviously at the time we started the 15 program, there was not a significant amount of data on that.
16 DR. SHEWMON: But there's nothing -- okay. It's 17 not an inherent -- more corrosion or toughness --
18 MR. ROEMER: Not at all. It's something, 19 obviously, both the industry and the NRC will proceed on 20 discussing and investigating.
21 The feedwater system, basically a concern about 22 water hammer as a result of a large number of events and 23 NUREGs that have been issued, most notably 0582.
24 Main steam system again was carbon steel and, as 25 Roger mentioned, there was a very important economic reason.
/m !
ACE.FEnERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 800336Wi46
l 3033530 52 BRT 1 We basically had the hardware in place. Which left us with 2 the last three systems that are shown here.
3- I guess it would be appropriate at this point, 4 since what I have beyond here is essentially redundant to 5 everything that has come before, to introduce Bill Server 6 of R.L. Cloud, who is responsible for taking the stress 7 results and determining the fracture mechanics 8 characteristics.
9 Bill?
10 MR. SERVER: This flow chart gives you an 11 overview of the program. The top part before the blue 12 section is pretty much what Bob talked about, a definition
- 13 of what are the breaks that would need protective hardware, l'4 what happens when you go through the target interaction 15 studies to help eliminate some of those pieces of hardware, 16 or the break associated with the hardware; and then the 17 screening on an industry experience, economics and leak 18 detectability margins that are required.
19 ( Slide . )
20 What I would like to talk about next is what 21 would be in the blue portion, which is the start of the 22 fracture mechanics portion, where we'll first discuss what 23 the highest stressed locations are in terms of minimum 24 material proporties. We'll go into a little bit about the 25 f atiguo crack growth calculations that were run, and then ACE FEDERAL REPORTERS, INC.
l 202 347 3700 Nationwide Coverage mn336-6646
30335.0 53 BRT
~
/ '3 1 we'll get into actual leak rate estimations, calculations
'v' 2 that were performed for normal operating loads; then into 3 the crack stability evaluation, where we'll evaluate those 4 contracts associated with a detectable leakage for higher 5 loads, normal plus SSE loads, and then a final crack 6 stability check, where we'll take the leak rate crack size 7 and apply higher loads than normal plus SSE, that would be 8 where the square root of 2 comes, normal plus SSE.
9 DR. SHEWMON: It sounds very interesting, and I 10 think we could devote more attention to it if we took at 11 least the scheduled break in here, although we are ahead of 12 schedule.,
Let's take a 10-minute break and come back to it.
13 (Recess.)
0
14 DR. SHEWMON: It was always my impression that 15 breaks in the middle of the meetings were somehow more 16 useful than the meetings themselves, but let us get on --
17 present company excepted, though. Please go ahead.
18 MR. SERVER: Well, I certainly needed the break.
19 (Slide.)
20 Let's start off talking about the highest 21 stressed location within the lines and how that was 22 de te rmined , relative to the minimum material properties 23 that were then used to evaluate the stability in leak rates 24 for that particular line.
25 The approach taken was to use the procedure O
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 80>336-6M6
C l 30335.0 54 BRT {
fj 1 outlined in NUREG-1061, of going through the line, anchor h 2 to - anchor, and determining the highest stressed - location, 3 where " highest stress" is defined as the highest. normal 4 plus SSE bending. loads creating a stress, a circumferential 5 cracking stress at that location.
6 Then, associated with that in a particular line, 7 you could have different material properties. You could 8 have base metal; you could have shielded metal arc welds or 9 submerged arc welds.
10 On'e approach to take relative to this'would be 11 to go through the lines and determine all of the submerged 12 arc weld locations and determine the highest stressed 13 location for those submerged arc welds; likewise doing it 14 for the base metal and if there are shielded ' metal arc -
15 welds, whatever, determining for an individual type of 16 material the highest stressed location.
17 What we chose to do here is go through the line 18 and pick the highest stressed location throughout the whole 19 line and then evaluate it for all of the different possible 20 materials that could occur in the line. So we did an 21 anchor-to-anchor evaluation. It ends up the highest 22 stressed location always occurred within the class 1 23 portion. Then we evaluated everything for base metal and 24 welds. It was only the particular welds that would occur 25 in that type of line.
O ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646
[
if I 30335.0 55 iBRT- ..
s r N -1 MR. RODABAUGH: Bill, at this point let me get 'i" i i (r~J ,
-u
[ 2 one of my two questions out of the way. .
! 3 MR. SERVER: Yop said you 5bil had one, Ed.
4 s MR. RODABAUGH: On your 14-inch pressurized 5 surge line you have a line, 14-inch line,,etainless ateel Someplace i 6 line coming into.the ferritic steel pressurizer. s.
s 7 in there you have a transition joint between ferritic/sEeel d 8 and austenitic" steel,,wh.'.ch is sometimes troublescue-la q t
l 9 industry,. I wonder if you would explain for the benpfit of '
R
's ,.
10 the Committee how you decided that that was not 1 cri'ti~ cal 3 11 -- critical..-
..s' ,
< ~
i l 12 MR. SERVER: Ed's question, basica.lly is didlMe 13 consider dissimilar metal welds. In terms of the $rsetura '
0-
,y .
14 mechanics analysis, no, we did not. The transitibn.frem 1
.s - ,
the stainless steel safe end to the ferric steel itself 15 '
involves an Inconel weld. That's actually considered 7part 16 17 of the vessel in the'way we approach {.he problem, not part -';
18 of the piping. Even though the otherpend of the safe, is i s T welded to the vessel itself and' that'.s a sta nie.is r.tegl
~
l 19 i-
. i 20 weld, indicative of the products we have evaluated here.
21 He have gone bacx cnd looked at these;cransition-22 pieces, the dissimilar metal wclds. I guoss a couple of f ,
- 23 comments there uculd be, first ofall that location doesn't ,
24 happentobethehigheststrbssed' location. There are. i 25 other locations in the line that have higher strasses, no
(
Ace FEDERAL ReeonTens, INC.
202 347 3700 Nationwide Cosense R00 336-6646
l' 3033530 56 BRT l !
[ '
&~
1 it isn' t the highest stressed location.
2 The transition into the base metal properties
?,
- ' involves material that starts getting very thick, okay?
3
\
4 The' nozzle portion from the safe end immediately gets very 5 thick, so the stresses there, excluding the fact that you
\
6 have thermal discontinuity there because of the two i \
7 different materials, the stresses decrease significantly 8 because of bigger thickness.
9
, 3 DR. SHEWMON: They decrease in the dissimilar 10 weld or 'right af ter?
11 MR. SERVER: Right after it.
12 DR. SHEWMON: Is that any help?
,, 13 MR. SERVER: It's helpful in terms of the
( "' !
14 proporties of the ferritic base metal because the stresses 15 are going down, if that's your concern. And then the m 16 ferritic base metal itself is class 1 -- well, in the surge 17 line it cras a class 1 material and it's a good grade 18 material with a very high RDNDT, so you are definitely on 19 the upper shelf and you don' t have any problem with low 20 temperatures and plus the surge line is always low 21 temperatures.
22 MR. RODABAUGH : The dissimilar weld itself, I'm 23 repeating Paul's question, the pipe itself, I think, is 21 1.312-inch wall. But what's the thickness of the 25 dissimilar metal weld? Is it actually greater than that?
O L) s
\
- /\CEJFEDERAL REPORTERS, INC.
/ i 202 347 3700 Nationwide Coverage 800-336-6646 4
e o
a '
'30335 0 :
, 57
-BRT q b
I
( 'l MR. SERVER: As I remember it's greater than T 2 that. I think it was like 1-3/4 inches, as I remember.
3 MR. RODABAUGH: That's the reassuring part of 4 the information.'
5 DR. SHEWMO,N: Let me see- what the unreassuring ;
+
n 6 part was. I guess my experience with this has been in - ,
7 high-temperature corrosion aspects, say stainless against 8 2-1/4 chrome, where things happen you don't like. But here 9 you are not worried -- are you worried about corrosion or i 10 initial toughness as fabricated or what?
s 11 MR. RODABAUGH: More mechanical. The ferritic 4
~
12 steel wants to expand at a certain rate with relation to 13 temperature, the austenitic at a different rate. If
() 14 there's a temperature difference in the fluid it creates a 15 high stress.
' 16 DR. SHEWMON: Is there any fatigue usage they
\ ,
17 should take account of here then with cycling? -
18 MR. RODABAUGH: Yes. And I think Bill can Yell 19 you that they have.
20 MR'. SERVER: Yes. It's designed for the class 1
- b. \
21 analysis.
22 MR. ETHERINGTON: That is a terminal end , isn' t 23 it?
24 MR. SERVER: Yes, sir. '<
25 MR. ETHERINGTON: So the pipe would be O
ACE FEDERAL REPORTERS, INC.
4 202-347-3700 Nationwide Coserage 800 D 6-6646 L.- ..
30335.0 58 BRT
(} l restrained at that point?
2 MR. SERVER: Normally there would be a restraint 3 there; yes.
4 DR. SHEWMON: Okay.
5 MR.. SERVER: Let's talk about the material 6 properties.
7 (Slide.)
8 What we have first had to show, we didn' t want 9 to do a testing program. We wanted to use industry data 10 that was already available to qualify the materials, so the 11 first thing we had to do was to show that the materials in 12 Beaver Valley Unit 2 were representative o,f the industry 13 data; that there was enough of a match between the O- 14 materials in the plant versus what is in the -- can be 15 collected as a data base outside from industry tests, to 16 show a similarity there.
17 It ends up, based on looking at the tensile 18 properties, we assured ourselves that, hey, we did have a 19 representative material; that the materials in the plant 20 were essentially the same as what had been used out in the 21 field for industry data.
22 As we said earlier the materials are 304, 304-N, 23 and 316 base metal. And then shielded metal arc welds and 24 a very few submerged arc welds.
25 DR. SHEWMON: Those "very few" were in the few O
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6646
30335.0 59 BRT
(~x 1 lines were you concerned with, namely the biggur ones?
d 2 MR. SERVER: Yes, sir. The 10-inch lines and 3 bigger had some submerged arc welds. Predominantly 4 shielded metal are but some submerged arc.
5 For fatigue --
6 DR. SHEWMON: Let me stop you. You say there 7 are only a few. Does -that mean only a certain percentage 8 in these large-diameter, of the welds in the large-diameter 9 pipe were sub' arc or does it mean all of the ones'in the 10 large-diameter lines were sub arc?
11 MR. SERVER: Some welds are the only ones sub 12 arc. Some of the.ones that come out to the field have some 13 sub arc with them. In the field everything'is shielded tO
\-/ 14 metal arc welds; so you could have a few shop welds th$t 15 were fabricated earlier.
16 DR. SHEWMON: Of those you'd have only those in 17 the bigger lines? .
18 MR. SERVER: That's right. 10-inch lines and 19 larger.
20 For the fatigue crack growth studies the data 21 that we used was based upon a correlation that James / Jones 22 had developed.
23 (Slide.)
24 They came up with a crack growth relationship as 25 & function of temperature and R ratio and all the important ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6616
1 l
30335.0 60 l BRT fi J
1 parameters. The only thing they didn' t have was an 2 adjustment for PWR water environment. And I'll show you in 3 a minute a slide that represents how we got around that 4 issue.
5 For the stress / strain properties we used 6 industry data that was -- had been generated primarily for 7 EPRI or by NRC research. We used two types of evaluations.
8 For the stability assessments we always used 9 minimum, the lowest stress / strain properties that existed 10 for that type of material. For leak rate estimates we 11 always used an average value. It ends up the reason we 12 went to the average value for leak rate estimates was to be 13 slightly more conservative. Had we used the lower bound we
(-
14 would have estimated lower, so we wanted to be conservative.
15 In terms of the JR curves we always used the minimum values 16 associated with the particular materials of interest.
17 (Slide.)
18 On the crack growth data, this represents the 19 James / Jones relationship, where you have crack growth 20 versus the applied stress intensity, the delta K. We have 21 it for different values of R ratio plotted here. Then we 22 have also plotted on this graph some actual PNR water data, 23 crack growth data.
24 What we are trying to show here is in the 25 original relationship, James / Jones used what they called a O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 61 BRT
')
i s_/
1 frequency factor of 1.
2 What we've done is taken their correlation and 3 adjusted that frequency factor to a value of 2. As you can 4 see, comparing the data to the different lines there, we 5 have pretty well bounded all of the a,vailable data for PWk 6 water environment.
7 Typically, our R ratio for the fatigue crack 8 growth calculations that were done were in the range of .3 9 to .8.
10 DR. SHEWMON: Remind me again, if I have an R 11 for -- full reversal of stress, is that R equals zero or R 12 equals I?
13 MR. SERVER: That would be minus 1; yes. Full C:)
14 reversal is.minus 1.
15 DR. SHEWMON: What's R equals zero?
16 MR. SERVER: Zero to maximum. '
17 DR. SHEWMON: All right.
18 MR. SERVER: Then to give you a feel for the 19 lower bound stress / strain and J-R curve properties.
20 (Slide.)
21 Let's look at top, which is the stress / strain 22 cu rves . The 304 and 316 materials are pretty similar in 23 terms of stress / strain properties.
24 Shielded metal arc welds, SMAWs -- the SMAUs 25 here have higher yield strength properties and higher A
V ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 62
.BRT 1 ultimate strength properties, . and the submerged; arc welds 2 are significantly higher, higher -ultimate properties.
3 Because of the way they want to match the weld .
4 with.the base metal, it has higher tensile properties..
5 When we look at the J-R curves, what we'll find 6 is the opposite effect. This is J versus' crack extension.
7 This dashed line represents J-1-C, initiation value. For 8 the 4752 we used a common curve and it has very high 9 toughness properties.
10 When we go look at the weld metals we see they 11 are reduced significantly. This is lower bound. This is 12 . based on the' lowest data that anybody in the industry has 13 collected that we are aware of, so these are lower-bound 14 properties.
15 DR. SHEWMON: Is there any data at all for J-1-C 16 in a dissimilar metal weld?
17 MR. SERVER: Like the Inconel material?
18 DR. SHEWMON: The ones you have in your system;
't 19 yes.
20 MR. SERVER: There are some data available. My
. 21 recollection is that it is going to be better than the 22 stainless steel welds. ,
23 MR. RODABAUGH: What you could be talking about, i- 24 Bill, the dissimilar metal weld at one edge could be a l
25 Inconel carbon steel --
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6M6
-. .-.-. - - - . . - . _ _ - - _ _ - . . - . - . - . . . - - , - - . . . . - - _ _ _ . . . . _ . . . ~ - . - , - - - , ,
30335.0' 63 BRT MR. SERVER:
!() 1 .If you evaluated the weld metal 2 itself.
3 MR. RODABAUGH: 'That's not the significant thing.
4- MR. SERVER: Discontinuity itself.
i 5 MR. RODABAUGH: And the different alloys as you 6 go from one part to the other part of the weld.
7 DR. SHEWMON: This is the' weld metal which has 8 properties appreciably lower than the base metal.
.9 MR. RODABAUGH: Yes.
10 DR. SHEWMON: I'm partly concerned about that, 11 or partly interested, at least.
12 MR. RODABAUGH: They presumably took that into 13 account.
14 ,
MR. SERVER: Definitely, on the fracture 15 mechanics analysis when we evaluate the material for the 16 shielded or submerged arc weld we actually.took these 17 vast -- the curves.
18 DR. SHEWMON: The question was how you treated 19 the dissimilar metal welds, and the answer was you didn' t 20 because you didn' t have to?
21 MR. SERVER: That's right.
22 MR. RODABAUGH: That 304, 316, does that 23 correspond to 30,000? This is room temperature tests.
24 MR. SERVER: These are at 550.
25 MR. RO DABAUGH : 550, I'm going to the true
($)'
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335,0 64 BRT 1 strain --
2 MR. SERVER: Those are 550 also.
3 MR. RODABAUGH : They show .2 -- no, that's 2 4 percent. Okay. I was wondering how these curves compared 5 with the code minimum values.
6 MR. SERVER: Significantly higher.
7 MR. RO DABAUGH: Is that conservative or 8 unconservative to assume they are significantly higher than 9 code minimum values?
10 MR. SERVER: I guess the question would be --
11 they are higher but not significantly higher, I guess would 12 be a better way to put them. I don't think -- these are
~s 13 based upon actual materials. Some have been taken out of
(#i
'~
14 plants. It's the best data we~have.
15 MR. RO DABAUGH: Okay. But your yield strengths 16 show quite a scattered range, for example , when you show it 17 in your report.
18 MR. SERVER: Throughout the industry; yes.
19 (Slide.)
20 For the fatigue crack growth evaluation, what we 21 did is we took the acceptance criteria from ASME Section 11 22 of the code as defining our depth of initial crack which 23 typically turns out to be around 11 percent. Assuming an 24 aspect ratio of 6:1, meaning it's essentially 6 times 25 longer than deep, we then went through and evaluated the O
ACE FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-6646
-30335.0 65 BRT
[ ).
s 1 fatigue crack growth response for the transients that would 2 exist in each line.
3 -
MR. RODABAUGH: Would it have made any 4 significant difference if you assumed that this 10 percent 15 allowance exists all the way tround in your fatigue ,
6 analysis?
7 MR. SERVER: A 6:1 ratio is pretty conservative.
8 It really wouldn't.
9 MR. RODABAUGH : Okay.
10 MR. SERVER: So, because these were all class 1, 11 high-stress locations, we then had. all the class 1 12 information to be able to do the cyclic fatigue crack 13 growth studies. . So we took them out to. a full 40-year 14 lifetime and we also used the full stress indices, that 15 would normally have been used in Section 3 of the code to 16 increase the stresses, to be very conservative in the crack
- 4. 17 -growth study.
i 18 The criteria for the crack growth studies was 19 that, starting with this 10 percent flaw, we couldn' t 20 propagate it to more than 60 percent through the wall so 21 our limit was 60 percent as being a failure criterion for 4 22 fatigue crack growth. We also had some other requirements 23 relative to the cyclic plastic zone size that existed ahead 24 of the crack at any time during the studies and then crack 25 studies relative to the surface length of the crack keeping
(
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
_ _ . . ~ -
30335.0 '66 BRT (j 1 the 6:1 ratio as a constant.
2 Basically as you can see down in the results for 3 the different lines, we started with an initial A/T of 4 around 10 to 11 percent, .1, .11; as you can see hardly --
5 a lot of them didn' t grow at all and it was only in the 6 14-inch pressurizer surge line that we saw any significant 7 crack growth at all, and it was around 0.3 crack growth.
8 It didn't get halfway -- just about halfway to our 9 criterion.
10 So crack growth was definitely not an issue for-11 these lines.
12 (Slide.)
13 Some of this is a bit redundant, talking about
-14 the limiting detectable leak rate and then applying the 15 margin to that.
16 As was stated earlier we can take a 0.5 gpm 17 detected leak inside containment, the EPRI test that 18 segmented that 0.1 was easily detected visually that gave 19 rise to the idea you can go inside containment and at least 20 detect something leaking at .1, and then, with the 0.5, and 21 apply a margin of 10, we come up with the 5 gpm ; we take 22 the 5 dpm, take a crack size associated with 5 gpm, and 23 then use that crack size in the fracture mechanics 24 evaluations.
25 (Slide.)
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 67 BRT.
() 1 For' the leak rate calculations to determine the 2 crack size for 5 gpm leakage, we only used the normal l
3 operating loads at this highest stress location.
4 What we did was take the loads that would open 5 up a circumferential crack under normal operating 6 conditions and added those up. So it would be the effect 7 of dead weight, thermal expansion, and pressure.. We added 8 those algebraically so we found out that the true amount of 9 opening that would occur. Taking this algebraic sum we 10 took a computing code developed by EPRI called PICEP -- yes?
1 11 DR. SHEWMON: Let me back up to thermal 12 expansion. Thermal expansion has to be pushing against m 13 something. In the case of dissim-ilar welds I can
- (J 14 understand what this is. In other systems is this a 15 thermal stress in a transient or against some restraint 16 down the line or what?
17 MR. SERVER: Your piping system is built. Then 18 you have to get it up to temperature. When you get it up 19 to temperature then you expand the pipe. Okay?
20 That will move the pipe and create some stresses 21 because of that. And that's what this thermal expansion 22 stress represents. It can be both an axial stress, if it 23 is impinging on something that can't move, or can result 24 also in a bending moment.
25 DR. SHEWMON: And this is partly because your O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 .
68 BRT
() 1 vessel and steam generator are fixed and therefore things 2 will bend when you -- okay. Fine.
3 MR. SERVER: The EPRI PICEP code -- I can't 4 remember what it stands for, pipe cracking something or 5 other -- was benchmarked against laboratory data that EPRI 6 and other people generated and some actual service data.
7 We feel the accuracy of PICEP is about plus or minus 25 8 percent.down at the level of 0.1 gpm, and it gets better as 9 we move up into higher leak rates.
10 Plugging is not a problem for the leak rates 11 above 0.1 or for the pipe materials of interest here. It's 12 one issue you might have to consider with carbon steel, p- 13 would be potential corrosion products blocking up the crack.
14 Here in the stainless steel material we don't feel that's 15 an issue.
16 Taking the 5 gallon per minute crack size and 17 putting into the PICEP code the average stress / strain 18 properties for the base metal we get 5 gpm crack sizes a 19 shown below, in inches.
20 Typically we are talking about a four or five-inch
! 21 crack size except for the 12-inch RHS line which had 22 significantly lower stresses so it required a larger crack 23 to leak at 5 gallons per minute.
24 MR. RODABAUGH: These are the circumferential 25 cracks?
(
l ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 i
30335.0- 69 BRT
'( )' 1 MR. SERVER: Exactly.
2 (Slide.)
3 And just to give you a feel of what you get out 4 of the leak rate calculation for the 12-inch RHS line, this
- 5. is leak rate as a function of crack length. What it does 6 is it determines -- the.PICEP code generates this as a 7 function of crack size so you'd come off at the 5 gpm some 8 level and this would represent your 5 gpm crack size.
9 (Indicating.)
10 MR. RO DABAUGH: That's associated with some 11 existing internal pressure and some set of normal loads?
12 MR. SERVER: The most normal operating rs 13 conditions for that line are used to determine this curve.
\)
~
14 (Slide.)
15 When we go to the crack stability evaluation, 16 what we are going to do is take this 5 gallon per minute 17 crack size and we are going to do things with it to show 18 that things are stable.
19 One of the things we are going to do is we are 20 going to increase the loads that we apply to show stability.
21 We are going to go and take the normal operating loads 22 which we j ust talked about, but we are also going to add 23 the safe shutdown earthquake loads, but we combine those 24 absolutely.
25 We take the normal loads we had from before that 24CE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
~-p
- 30335.0 70 BRT-() I we used in the leak rate calculation and add the safe 2 shutdown earthquake and just add them absolutely, because 3 they can be plus or minus; we just add them, plus, to give 4 1the worst case.
5 In evaluating the crack stability we used 6 another EPRI code. This is code FLET. I can't remember 7 what it stands for. This was verified against existing 8 data generated by the industry plus comparisons to some 9 other codes that have been used; also make the comparison 10 with existing data; and we used an approach out of the FLET 11 method that's called the DPFAD, it's deformation plasticity 12 failure assessment diagram. It is no different than the
, 13 stanaard J-T approach, it's just a different way of 14 plotting it and utilizing the same information.
15 The reason we happened to use that from the EPRI 16 perspective was that their J-T methods did not allow for 17 crack growth, extension, going above J-1-C, whereas the JDFAD 18 method that was included in the code did.
19 For this evaluation of crack stability we then 20 take and put a margin on the crack size to check for 21 stability. In the NUREG-1061, it suggested a value of 2.0.
22 We used a range of 1.8 to 2.0 to show this assessment. It 23 ends up there's only one line that was not evaluated for 2; 24 it was evaluated for.l.8.
25 We also used the lower-bound J-R curve t
l
(
I ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
_ . . _ . . _ . _ ~ . . - . - . - - - . _ . . . _ _ _ _ - _ _ _ ,_ ,
30335.0 71 BRT
()
- 7. ,
1 properties and the lower-bound stress / strain properties 2 which gives us the most conservative condition.
3 The results, then, show that for the six-inch '
4 line where we used a factor.of 1.8 on the crack size margin 5 rather than 2, that the base, if we evaluated the base 6 metal condition and the shielded metal arc welds that 7 existed, there were no sub arc welds in that particular 8 line, you'll see that for this thing to be stable this has 9 to be a value of 1.0 or greater. Okay?
10 As we go down you can see the effect of all --
11 all these lines were based on a crack margin of 2. You can 12 see that everything here is at least 1.0 or greater.
13 ,
Yes?
14 DR. SHEWMON: If I didn' t know how conservative 15 eng'ineers were tur nature, I would say you are very much on the ragged edge at eight inches because there's no safety 17 margin at all.
18 MR. SERVER: This is the safety margin that's 19 built into this calculation, is the margin on crack size.
20 Okay?
21 DR. SHEWMON: Okay.
22 MR. SERVER: This is an evaluation just to show 23 if you meet the criteria. It ends up you are right, 24 exactly for this 8-inch line using the margin of 2 we just 25 meet it. Had we used the margin of 1.8 here on the 8-inch ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-66J6
30335.0 72 BRT I 1 we would have made it significantly.
s 2 MR. RO DABAUGH: We have been hearing over the 3 past couple of years how terrible this submerged arc weld 4 is. Here, if I'm reading your picture right, it's better 5 than the base metal?
6 MR. SERVER: Yes. That's what these numbers 7 show.
8 MR. PODABAUGH : Your J curves are much lower.
9 MR. SERVER: If you were thinking of fracture in 10 terms of linear elastic conditions, you wouldn' t expect 11 that. But when you go into the elastic / plastic regime 12 approaching fully plastic conditions, things aren' t exactly 7- 13 the same, they don' t work exactly the same.
Q_, .
14 It ends up the controlling feature in these 15 elastic / plastic assessment is more the stress / strain 16 properties than the J-R curves. If you'll remember the 17 stress / strain properties were significantly higher and that 18 more than compensates for the decrease in the J-R curve.
19 MR. RO DABAUGH: I had something like a bolt 20 steel that I quenched and did very little tempering on it, 21 so I got extremely high yields -- and even though the J 22 might drop way down it still would be highly likely to --
23 if a pipe were made out of such material, it would be very 24 likely to leak before it broke. Somehow --
25 MR. SERVER: It seems kind of strange. That's
/^
O)
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6646
30335.0 73 BRT- .
rig ht . If you go to the fully plastic condition where you (v^) 1 2 are basing 'something strictly on failure of reaching a 3 limit load condition, in assessing the limit load if you 4 had a higher flow stress or ultimate stress it means it 5 wouldn't fail until you got up there. Okay? And because 6 even though these weld materials have lower toughnesses 7 than the base metal we are still in material that is still 8 tough. Okay? It is just lower than the base metal.
9 MR. RODABAUGH: We used submerged arc welds 10 anyway, 11 DR. SHEWMON: Are you saying because it's 12 stronger it doesn' t yield as early or the crack doesn' t
- 13 advance as early as it does in the base metal?
'14 MR. SERVER: Yes. Because the plasticity is 15 constrained there ; yes.
16 MR. ETHERINGTON: This is complicated, too, 17 because the crack is not really in the base metal or weld; 18 it's mostly in the HAZ, isn't it?
19 MR. SERVER: It can be; yes. The heat-affected 20 zones always present problems because it's difficult to 21 measure real properties there.
22 (Slide.)
23 The final stability check, once we've done the 24 check on taking essentially twice the 5 gallon per minute 25 leak rate crack size and checked it for stability, now O
v ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
'30335.0 74 BRT
() 1 let's do an additional check and let's keep the 5 gallon 2 per minute crack size constant, let's not put a margin on 3 it, but now let's put a margin on the loads that'we are 4 going.to apply to assessability.
5 This is where the value of 1.414, the square 6 root of 4 times the normal plus SSE loads are utilized.
7 Again'we used the EPRI FLET code, the 5 gallon 8 per minute crack size with no margin on the crack size 9 itself. Again use lower-bound stress / strain properties and 3
10 J-R curve properties.
11 What you'll see here is that we have 12 significantly higher margins here than we had when we took 131 twice the crack size and used normal plus SSE 1oads. ,
This 14 , shows one feature of why things are different from linear 15 elastic conditions to this elastic-plastic case. Because 16 these margins are significantly higher here. In linear 17 elastic terms it would have been almost the same test.
18 That's basically it. That's the result of all 19 the calculations.
20 DR. SHEWMON: Thank you very much. Is that the 21 end of the presentation, then?
22 It has been very interesting. Do you want 23 something out from us? When do we hear from each other 24 next?
25 MR. TAM: This is Peter Tam speaking . I believe O f f
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6M6
30335.0 75
~BRT y
[ 1~ that we are here today because of your request in your 1985 2 letter. I really don' t h' ave any idea what --
- 3 DR. SHEWMON: Do you have any other comments?
4 Okay. Fine. Thank you.
5 Let's take a short break, and when we convene 6 again in five or 10 minutes we'll be talking about 7 NUREG-0313.
8 (Recess.)
9 DR.-SHEWMON: 0313 is getting rather venerable
-10 at this point in time. It has been around many years and 11 seen many incarnations, and actually revision 2 has been 12 around .for longer than Warren would care to admit. This 13 'has gone out for public comments. We were one of those who O. 14 wrote public comments and we are hearing a presentation 15 from the Staff on what they propose to do with the public
-16 comments and go on from there.
17 With that limited production I'll turn it over 18 to you, Warren.
19 MR. HAZELTON: All right. You did touch on the 20 antiquity of this stuff.
21 What I was planning to say was, as we all know, 22 in the beginning God created the heavens and earth; shortly 23 thereafter Hazelton started working on BWR pipe cracks.
24 (Laughter.)
25 I keep telling people it is neither my intent O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
J
'30335.0 -
76 BRT
() 1 nor my desire to make a life's career of th'is subject, 2 contrary to appearances.
3 What I would like to do here is to go through 4 some of the early history very quickly. I think it might 5 be relevant to do that because it has been going on so long 6 some people may not remember exactly where we are in this-7 business.
8 However, I intend to cover these things very 9 quickly. If you have any questions, sound out. Otherwise i 10 .I'm going to go as fast as I can.
l 11 Earlier, IGS CC was found at Dresden 1 in 1985, 12 attributed to improper welding. Furnace-sensitized 13 camponents also were found to be cracked. As a matter of I
14 fact, a rule ' requiring low-carbon material in BWRs was "
15 seriously considered in 1972.- That 'was before I got here.
16 Then, Reg. Guide- L,44 was issued, instead. It 4
17 was called " control of the use of sensitized stainless 18 steel" and high-carbon material continued to be accepted, 19 but control of welding parameters was used as a basis. As 20 we now know, these controls proved to be inadequate.
. 21 DR. SHEWMON: Control of welding parameters i.
22 meant low heat input or what?
23 MR. HAZELTON: Yes. But that's not -- really 1
24 not a simple subject. Yes. They were talking about 100 KJs .
25 per inch. That's not the only parameter that's important .
()
i i ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646
_ _ _ _..~.._ _ , , _ . . _ . . _ . . _ . . , . _ _ _ , _ . _ , . . ~_ .. _ - _ _ . . . . - _ __
?30335.0 77 BRT 1 Really, we' feel if that's the only parameter you control
-([ )
2 'you better be down more like- 20 than 100.
Okay.
3 DR. SHEWMON:
4 MR. HAZELTON: In 1974 cracks were found in
'S recirculation bypass lines, those were 4-inch lines, and a 6 bulletin was issued and required inspections within 60 days.
7 Then another bulletin was issued tightening the leakage 8 limits.
9 In 1975 the task ~ force was formed, they called 10 it a study group at that time.
11 (Slide.)
4 12 During this time period, as a result of the 13 inspection requirements and so forth, and'. finding leaks, .
14 they continued to find cracks in core spray lines. Two 15 more bulletins were issued, requiring inspections within 20 16 days. Then the study group report was issued and the 17 ancient and venerable NUREG-0313 was issued.
18 It set Staff positions on materials and 19 processes, augmented in-service inspection requirements and 20 augmented leak detection requirements.
'l 21 Leaks and cracks continued to be found in small l
22 lines, that is 12 inches and smaller, generally less --
23 there's some some arguments on whether there's any 12-inch 24 ones. In 1978 cracks were found in a large pipe in Germany.
i
! 25 In 1978, therefore, a new pipe crack study group .
l
(
l ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
30335.0 78 BRT 1 was formed, and in 1979, their report was issued. In 1980, 2 NUREG-0313 was duly revised and issued, and it was supposed 3 to have been implemented by a generic letter but the 4 responses to the generic letter indicated poor performance.
5 I guess that's an un,derstatement.
6 (Slide.)
7 Now we come up somewhat to the present, recent 8 history.
9 In 1982 leaks were found, presumably by water 10 dripping on the head of a bald NRC inspector, at Nine Mile 11 Point Unit 1. Because it was 28-inch pipe we got concerned, 12 so information notice was issued. Then, almost immediately 13 thereafter, bulletin 82-03 was issued. The idea was to j x
14 check and see whether other* plants have the same problem or 15 whether it was just confined to Nine Mile, because it was a 16 different -- had a different type of'recirc system and a 17 very early plant.
18 It required a sample inspection of the recirc 19 piping at nine plants and it also required that the people 20 doing the examination give a demonstration, how well they 21 could find cracks in some of the Nine Mile Point piping.
22 As a matter of fact they showed that they 23 weren't very good at it, but they went out. They found a 24 mess of cracks.
25 Bulletin 83-02 was issued which required the 1
(~/'\
w ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646 l
30335.0 79 BRT
/~T 1 inspection of the remaining plants and upgrade the
\J 2 capability demonstration tests to say, essentially, that 3 you had to be at least this good, and it wasn't all that.
4 tight but they had to find a certain number of cracks.
5 They tried to limit the number of overcalls, that is, areas 6 called cracked that weren't cracks. But the requirements 7 were not very stringent.
8 Then some orders were issued, when more cracks 9 began to be found, orders were issued to five little plants 10 to inspect a little earlier than their next scheduled 11 outage.
12 In 1983 -- well, there's something not in there.
13 In 1983, NUREG-0313 was revised again. But, before it was 14 issued the piping review committee and pipe crack task 15 group was formed and they said hold up on revision until we 16 see what these people say. However, we had to be making 17 day-by-day decisions, what was being done, whether it was 18 adequate and so forth. So we had, you can call it ad hoc 19 Staff requirements. So, SECY 83-267(c) described these, 20 were presented to the Commission and so forth. This was 21 then considered the short range plan.
22 (Slide.)
23 The document was implemented by generic letter 24 84-11 and we developed a long-range plan for dealing with 25 it. And the Commission asked us for one, so, in SECY 84-301, J
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
30335.0 80 BRT
() 1 described this plan identified as generic issue 86, and the 2 plan essentially was to issue the pipe task group report 3 1061 volume 1 for comment, and then revise NUREG-0313, and 4 also to include comments on 1061, volume 1; and as a result 5 of the reinspections going on consider code modifications 6 that were going on, and other in'ustry d developments like 7 EPRI tests and so forth, and then to implement it by 8 generic letter.
9 dkay. So where are we? I hope you have this.
10 I know you can't read it off of this, but I just wanted to 11 talk about it. I just recently got this from General 12 Electric, they kindly gave it to me. It gives a count to-13 the number of cracks that have been found.' This has a-0 14 little detail of the different piping systems.
15 Then, as a cumulative number of cracks found by 16 year, and you see we are down to nearly 1200 cracked welds.
17 That is the number we know about.
18 There were several plants that, after they found 19 a few cracks, decided the heck with it and replaced the 20 piping without doing a complete inspection, so there were a 21 lot -- a lot? I assume there were quite a few cracks in 22 that old replaced piping, that are not in this tally.
23 MR. MICHELSON: Question, reactor water cleanup.
24 I assume that includes both inside and outside of 25 containment?
O ace. FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
'30335.0 81
'BRT
(~') 1 MR. HAZELTON: I tried to find that out and s_/
2 couldn't. That's a sensitive issue that I'll talk about a 3 little later, as a matter of fact.
4 MR. MICHELSON Okay. For now you are not sure.
5 MR. HAZELTON: No. I asked General Electric 6 that question on this and they didn' t have the answer.
7 MR. MICHELSON: Okay.
8 MR. RODABAUGH: Warren, you said this is the 9 number of cracked welds. Suppose there's three or four 10 cracks in one weld. Would that be reflected --
11 MR. HAZELTON: No. It is intended to be the 12 number of cracked welds.
13 MR. RODABAUGH: Okay. Thank you.
O 14 MR. HAZELTON: ' However, General Eledtric thinks 15 that there may have been some cases where they had cracks 16 on both sides of the weld and they called that two. But, .
17 in other words, this isn't a very accurate thing. It gives 18 you an overall picture. The intent was not to call that 19 two cracked welds but in some cases they think they did.
20 DR. SHEWMON: On the positive side, the control 21 rod drive problem tends to have been fixed, one might hope 22 from looking at these numbers?
23 MR. HAZELTON: Yes. Right. A lot of actions 24 have been taken.
25 MR. MICHELSON: Is part of this just simply ACE FEDERAL REPORTERS, INC.
202-347-3700 Nat;onwide Coverage 800-336-6646
30335.0 82
~BRT j ) 1 'because they are just looking further out and so forth? 'Is 2' that why we are seeing more cracks? Or are they going over 3 the same material again and finding more?
4 MR. HAZELTON: Both.
5 MR. MICHELSON:' Both. You . can' t sort that out?
i
-6 MR. HAZELTON: No. 84-11 did not specifically 7 require 100 percent' inspection, but it did have a sample 8 expansion program that, if they kept finding cracks, would
. 9 have essentially gotten them to a complete inspection. But 4 10 there are still some welds out there that have never been 11 inspected.
- 12 ( Slide . )
13 So, as I said, the purpose of the NUREG Rev. 2 i .Os '
14 was originally to provide BWR licens,ees with alternative 1
~
15 mitigative actions or combinations considered acceptable by 16 the Staff. I'll talk about those.
17 These include, of course, obviously: pipe 18 replacement; weld residual stress improvement processes; 19 water chemistry improvements; weld repair and evaluation 20 criteria; and augmented in-service inspection schedules.
j 21 MR. MICHELSON: What puzzles me a little bit is,
. 22 like reactor water cleanup, a large portion of that system
. 23 is -- most of it is outside of containment.
< 24 MR. HAZELTON: Yes.
25 MR. MICHELSON: Most of it is nonsafety-related i
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
1 ,
='
- ,/
30335.0 i 6. b l' 83
, -BRT ~
, [ s
(~' 1 by classification. 9 V
2 MR. HAZELTON: By definition. s 3 MR. MICHELSON: ' lt may be built, t and likely, in ..
4 fact, older plants are built to 31-1.
5 If it is indeed experiencing an accelerated rates, -
6 of cracking, then it becomes very1 serious because that1c 7 system is, you know, potentially very hazardous becausel 8 it's on-line, full pressure, full temperature. '
9 If it doosd't isolate, if it b"reaks and doesn't isolate you are in bad shape.;<
10 }
11 MR. HAZELTON: Yes. I agree. And if you wait 12 until a little later on I'm going to be talking about that, s
s 13 because that is a sensitive subject.
' O- 14 MR. MICHELSONr Oh, good. Okay.
15 MR. HAZELTON: The point that we are making is 16 that no new requirements are igposed by the NdREG. Of~
s s 17 course the NUREG can' t impose requirements'. The" -
c .
18 requirements are j ust essentially -- not 6xactly imppsed, 19 but suggested by;the generic letter. The' NUREG just has 20 recommendations in.ih,1but the generic letter that should 21 implement these recommendations calls them Staff positions.
22 The import'an: thing is that Rev. 2 gives the 23 utility more options than Rev. 3 did.
, 24 ( Sli'db . )
25 Some of the other differences here between Rev.
O
~
Ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6 )
9
I 30335.0 x 84 BRT l' and Rev . 2, Rev. 2 -- and just remember, whenever I say
~
( ; 1 w;
2 Rev. 2 I'm talking about "and the associated generic letter" 3 -- includes all stainless piping regardless of class. Rev.
4 1 just had a limited scope. It just picked certain systems.
5 MR. MICHELSON: You made an interesting 6 statement. You said it includes all piping regardless of 7 class. You meant regardless of which Section 3 class it 8 was in? Because you said class 1, 2 and 3. Do you mean 9 A&S 1, 2, and 3? Or Section 3?
10 MR. HAZELTON: I would talk about regardless of 11 code clascification.
12 ,
MR. MICHELSON: Including B-131 piping?
- 13 MR. HAZELTON: Yes. That's the way it is 14 currently written.
15 Remember, all of this is legally predecisional.
16 l Some of these things can change before it hits the street i
17 fo rmally.
18' MR. MICHELSON: And it includes safety- and 19 nonsafety-related systems? Because reactor water cleanup --
20 MR. RAZELTON: If you wait until I cover the 21 subject --
22 MR. MICHELSON: I'll wait.
23 DR. SHEWMON: On the last slide, weld repair is 24 overlay; is that right? The previous slide, I'm still 25 trying to read that. You don' t have to put it back unless i
n ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
Qi
\i 30335.0. 85 BRT' N i-
, .L '
[lJ'l you need - to, but you had water chemistry improvement was f]
2 one option, weld repair was another, weld repair is'the 3 same as overlay? m 4 MR. RAZELTON: Basically, that's right. They 5 could repair it by a standard method of cutting the weld 6 out and putting a new weld in or putting in a piece, but 7 basically weld rep.tf r -is what's practically always l'nvolved.
s 8 DR. SHEWMON: Yes.
9 MR. HAZELTON: One important thing heres is that l0 the new Rev. 2 and generic letter requires the formal 1
!- 11 qualification of NDE examiners, and procedures -- this I
'L 12 touched on -- was' initiated by the bulletins, but then has
, 13 been developed since then with very close cooperation by -
u .
14 the industry that is primarily the b' oiling water reactors 15 owners group and EPRI, and the NRC. We have an official 16 document called the " Coordination Plan,$ that we are 17 working on to coordinate the efforts here. We have a 18 strong qualification program. ,
- 19 Whereas Rev. 1 just hadi the , strong caution that
,j" 20 code methods were probably worth less and suggested they vi <
21 trycto use better UT techniques than were commonly being 22 used.
23 The new documents provide quantitative benefits 24 for : proven mitigation methods, that is if you do this you
.25 get so many Brownie points on the augmentation of your >
O ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6686
30335,0 86 BRT
('N, 1 in-se rv ice inspection. Whereas Rev. I had no specific 2 provisions. It just said if you do stress improvement ,
3 it's probably a good idea, but didn' t give them any Brcwnie
\
4 points.
5 One thing here that you have to remember is that 6 Rev. 2 provides guidelines for evaluation and repair of 7 cracked welds, together with the augmented in-service 8 inspection program required. Whereas Rev. 1 very simply 9 said: If you find a crack in a weld, you replace the pipe.
10 (Slide.)
11 This is all rather obvious, but we'll go through 12 it and I'll touch on a lot of these thihgs in a little more
, 13 detail later.
s) 14 So, the new documents generally follow the 15 recomraendations of the piping \ review comm'ittee, recognizing 16 that they met and didstheir deliberations and wrote their 17 reports when a lot of this stuff was in a state of flux.
18 So we had to update some of the concepts, in light of later 19 developments. But, basically it generally follows the 20 recommendations.
21 It recommends the use of resistant materials and 22 recommends replacement of subacceptable piping as the best 23 choice and gives credit for residual stress improvement, 24 thinks that's a good idea, and strongly recommends improved 25 water chemistry.
fm L]
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6646
m 30335.0 87-BRT l It provides specific inspection schedules
. x_-
(~
2 considering~the type of material, the type of processing, 3 kind of water chemistry and repair and cracking conditions.
4 It provides guidelines for analytical evaluation of crack 5 evaluation and repairs, and it upgrades the leakage limits 6 and the monitoring.. That is, it upgrades it over what most 7 people -- I think, right now, it upgrades it over what some 8 people still have, but most utilities have been using the 9 upgraded requirements.
10 I don't know that I have to cover the Staff 11 position on resistant materials. It's pretty easy. Staff 12 position on processes -- I might put up something like that. l 13 (Slide.) -
'O 14 Obviously, if you do solution heat treatment 15 af ter welding you are in good shape. It not only 16 eliminates the sensitization from welding, but relieves the 17 residual stress from welding.
18 Now, heat sink welding is a process that was 19 used on some BWRs in a transition period, where they 20 realized they needed to do something , but didn' t have the 21 time to go out and buy new pipe, so they made some of these 22 welds by having water inside. This does two things, it 23 reduces the possibility of extensive sensitization and also 24 improves the residual stress pattern.
25 Stress improvement processes, the first was O
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M6
J30335.0 88 BRT
, ("T
.J 1 ' induction heating stress induction process. Now it appears 2 we have agreed that the mechanical stress' improvement 3 process is also as acceptable. The important thing here is 4 that the weld, as welded, has high tensile stresses on the
, 5 inside' surface and then the complex residual stress 6 distribution through the weld depending on size and exactly 7 how it was made; but the important thing is .that we 8 determined that that tensile stress on that inside surface 9 is -- may well be the most important single factor in 10 initiating and propagating IGFCC.
11 So the stress improvement process essentially 12 reversed the stress distribution and really helped the-13 situation. -
O . 14 The last pass heat sink welding is something 15 like that. When they get through making the weld, they put 16 something like a wash basket on the outside which does 17 something like the induction heating stress process. That 18 is, it's intended to reverse the stresses.
19 There has been some question as to whether it's i
! 20 really effective or not.
21 DR. SHEWMON: How is last pass heat sink 22 different from heat sink?
23 MR. HAZELTON: Oh, the last pass heat sink, heat
- . 24 sink will make the route, make it watertight, then they'll 25 put water inside the pipe and make the rest of the weld O
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
/
30335.0 89
=BRT
(~)
V I with water inside. The last pass heat sink weld means they 2 put the water in, then finish up and put a wash basket on 3 it.
4 MR. MICHELSON: In case there's a weld that 5 becomes defective because the heat-affected zone of the 6 base metal has become sensitized from the original welding 7 process, in a case like that, if they want to elect to 8 grind out the weld and redo it, how much of the base metal 9 do they have to remove?
10 MR. HAZELTON: All of it. All the way to the 11 inside.
12 MR. MICHELSON: No, I mean in terms of width.
13 MR. HAZELTON: The heat-affected zone is
() 14 probably maximum an eighth to a quarter of an inch.
15 MR. MICHELSON: Do they have to take that out if 16 they want to replace the weld?
17 MR. HAZELTON: Yes. They have to. That's the 18 part we are worried about.
19 MR. MICHELSON: Yes. That's right. So what you 20 do is you have to grind all of that away or machine it away 21 or whatever, and then you have to put in a very massive 22 weld to fill the gap again?
23 MR. HAZELTON: Right.
24 MR. MICHELSON: In the process of doing that you 25 hopefully don't sensitize the base metal again?
O V
i l
l ACE FEDERAL REPORTERS, INC.
202-347-3700 Natienwide Coverage 800-336-6M6
30335.0 90 BRT 7 ~y 1 MR. HAZELTON: That's right.
(_)
2 MR. MICHELSON: It's a hard thing to do, I guess.
3 MR. HAZELTON: .These kinds of reasons that you 4 are talking about are exactly why people don't do it.
5 MR. MICHELSON: I just wanted to be sure I was 6 thinking right.
7 MR. HAZELTON: You are.
8 MR. RO DABAUGH : Warren, while you are on this
~
9 part of the document, do you have the latest draft of it up?
10 Do you have the latest draft, the one handed out to Dee, 11 draft 2? Have you got this in front of you? There's a 12 typo earlier that keeps being repeated in draft after draft 13 after draft.
(~3 -
\/
14 MR. HAZELTON: What's that?
15 MR. RO DABAUGH: Look on page 2-5.
16 MR. HAZELTON: Yes?
17 MR. RODABAUGH: Under induction heating, stress 18 improvement, the following sentence, "no subparagraph is 19 identical to the following paragraph."
20 Something is editorially wrong there. That's 21 the point. j 22 DR. SHEWMON: You have omitted the paragraph 23 that should be under ISHI?
24 MR. MICHELSON: That's right. That's a big one.
25 It's more than typographical.
(7 U
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33 & 6646
. . . . . _ . ~ . .
30335.0 -91'
'BRT' 1 MR. HAZELTON: As a result of some comments from 2 other people in the agency, I rearranged the paragraphs' 3 because they thought it would be neater.that way and so, .
4 -what the girl.did was, as you see it..
5 MR.- RODABAUGH: I'll buy that.
6 DR. SHEWMON: We trust in the next version it
- l. 7 will-be different,~with luck.
8 MR. HAZELTON: I just have one comment to make, 9 that 5520 system is real great, but you can' t imagine the 10 problems I have when I want to go through and change a
- 11 comma to a semicolon or something like that, and end up 12 doing that properly, but losing entire sentences or 13 something else.
.. O 14 If I g'o through and proofreali to see whether 4
15 they made my appropriate changes they did, and that's what l
4 16 I did there. I just looked at the title of the paragraph.
'17 DR. SHEWMON: Okay.
[
18 MR. HAZELTON: Thank you.
19 ( Sl ide . )
20 Water chemistry, you are going to hear an awful 21 lot about that, so I'm not going to harp on it. Just to ,
.22 say if people use what we will consider good water
- 23 chemistry, improved water chemistry, we'll give them credit 24 for reduction in augmented in-service inspection 25 requirements.
J l
t ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
30335,0 92 l I
LRT 1 DR. SHEWMON: A year ago or six months ago the 2 Staff was still having a lot of trouble deciding what it 3 was they were going to do to change the specification of 4 water chemistry. Can you tell me what they have come down 5 with now?
6 MR. HAZELTON: Right. Well, it just so happens 7 I lucked out. I've got the guy right behind you that is 8 right up on the latest, so I'll let him describe it.
9 MR. WITT: Frank Witt, W-i-t-t. You asked 10 whether we are planning to change the technical 11 specifications?
12 DR. SHEWMON: Well, the Staff, for years, has
-s 13 sort of said we know water chemistry is important , but we I ) .
14 don't know where'to specify it down in there and so they 15 have maintained that one micromole is as low as they want 16 to go after the owners group started trying to go lower.
17 Now Warren suggests that, indeed, if they get better water 18 chemistry, it reduces stress corrosion cracking and I -
19 wanted to know what the Staff was now willing to specify on 20 that because last time I heard the explanation they hadn' t 21 changed things.
22 MR. WITT: We still aren' t changing things. But 23 we are encouraging all BWRs to abide by the owners group 24 recommendation which is less than .3. And, also we are 25 reviewing right now the guidelines for permanent (3
V ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6M6
- 30335.0 93 BRT 1 installation'for hydrogen water chemistry.
)
2 DR.RSHEWMON: The tech spec will still say_1 3 micromole, but they are still encouraged to do better; is 4 'that it?
5 MR. WITT: Right. Most of the plants are doing-
.6 that. Peach Bottom has improved very well, they are down 7 below .2 micromole.
8 DR. SHEWMON: So when you come back to.saying 9 whether or not you'll give them credit for it, you can't do 10 it on the basis of their tech specs. Do you do it on the ,
11 basis of their performance in the last year or management 12 commitment or what?
13 MR. HAZELTON: I could be flip and say I will do 14 it on the basis of what Frank Witt tells me, but the real '.
15 answer to your question is we don't know.how we are going 16 to do that. That's why you'll see the words in here 17 regarding how much credit we give are weaseled. We say 18 we'll do it on a case-by-case basis; okay? I don't think 19 we have determined exactly how we are going to quantify 20 that and how it is going to be handled.
21 MR. WITT: We have one example , on Peach Bottom, 22 where they want to extend operation beyond the midcycle 23 shutdown and we looked very carefully at the water 24 chemistry and are very well satisfied that they had 25 improved it, had the management commitment to keep the O.
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6M6
3033520 94 BRT 1 chemistry down below, way below tech spec values. And they 2 are doing a lot of changes in their cleanup system and 3 making major changes in the chemistry area including who 4 the chemists report to, the line of command which is much 5 better now; the checklist is listened to now and if he 6 needs some work done, he has management support.
7 DR. SHEWMON: Okay.
8, MR. HAZELTON: Crack evaluation criteria, 9 basically that's when you leave a crack in there, is it 10 safe? For how long do we think it is going to be okay?
11 Our criteria are consistent with the intent of NUREG-1061, 12 volumes 1 and I think that should be 3. But the technology 13 in this area has been improving and solidifying , so
( )
14 basically, we are now simplifying our criteria' to say that 15 it should be in line with our IWB 3600, of the 1986 edition 16 of the code. The 1986 edition picks up what was in the 17 1985 regarding flux welds and picks up what was in some of 18 the earlier editions regarding the other types of welds.
19 The 1986 edition now has the thing all cleaned up in one 20 version.
21 (Slide.)
22 At the time this was prepared, we were still 23 fussing around a little bit with it. But it's all I
24 clarified now.
25 DR. SHEWMON: Has that IWB 3600, as that has
()
'v' ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-3,5-6646 L
30335.0 95 BRT r-
~~
1 evolved, you and some others have had some complaints with 2 it. That's all long settled?
3 MR. RODABAUGH: Yes.
4 DR. SHEWMON: Okay.
5 MR. HAZELTON: To use IWB 3600 for a case of a 6 nonoverlaid crack weld, the code, of course, says you have 7 to calculate how fast the crack is going to grow and factor 8 that into your calculations. But it doesn't tell you h:7w 9 to do it.
10 (Slide.)
11 The NUREG does have the methodology in there, in 12 the appendix, that tells people how to do the crack growth 13 analysis. This is pretty standard. Everybody seems to be
? \ .
\ happy with it. It was presented at the last SMERT 14 15 conference and it's there for the help of people.
16 MR. MICHELSON: If I have a sipe that's in a 17 cracked condition, which I may either be going along with 18 the cracks or I may have done some kind of a basic or 19 temporary repair, what -- have you done any evaluation of 20 what you think the effect is going to be on the probability 21 of failure of that pipe, given the condition it is now in?
22 The probability of pipe failure? Have you done a number?
23 Clearly it's going to be somewhat greater --
24 MR. HAZELTON: I'm glad you asked the question.
25 Let me just give you -- here is what we are talking about l'h O
ACE-FEDERAL REPORTERS, INC. .
202 347-3700 Nationwide Coserage 800 336-6646 l
30335.0 96 '
BRT on weld overlays. Some people have been confused about 4([ l 2 what do we mean about a " weld overlay," so we are talking 3 about weld passes all the way around and the length of it 4 and the thickness are calculated. We pay most attention to 5 the thickness of it.
6 (Slide . )
7 We say if you have a crack in there, your final 8 "as overlaid" joint, has to meet IWB 3600. Okay?
9 Now we have several kinds of weld overlays. One 10 is-commonly called the " standard overlay" and that does the 11 calculations for 3600 as if you had a crack all the way 12 through the original pipe for 360 degrees.
- 13 Then there is something that we now refer to as
'k )
14 a designed overlay that only assumes the crack is all the 15 way through for the measured length of the crack. But even 16 if the crack is halfway through, it assumes it is all the 17 way through for that length.
18 We have been having some work done, as some of 19 you know, out at Battelle, Columbus on the degraded pipe 20 program. It's kind of neat that they just finished some 21 tests on weld overlaid pipes. There is a draft report out 22 that obviously you can get. I have a copy of it. I think 23 there's -- you still -- it still needs some editing. J 24 (Slide . )
25 Here is what happened when they tried to break a O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-335 6646
s30335.0- 97 8RT 1 16-inch pipe with a weld overlay on it. I. hope you can --
2 gee, I don't know whether it goes that way or this way, but 3 I don't think it matters.. I have been looking at it this 4 way.
5 You can see the cloud of steam here obscuring 6 the trees. I hope you can.
7 That's how far they had to bend this hunk of 8 pipe, to cause it to spring a leak in that overlaid weld.
9 The way that was formed was they made an EDM notch, 10 I think the number was 17 percent on some of these, of the 11 circumference, halfway through the wall. Then cycled it in 12 f atigue and grew a crack 180 degrees all the way through 13 the wall.
' Then they had a weld' overlay appl'ied and --
14 15 designed and applied by the people that do this every day.
16 And there it is.
17 That happens to be a 16-inch and it was schedule 18 100, so the pressure was a little more than -- you know --
19 had to be designed a little differently, but consistent 20 with 3600.
21 Here is what happened to the six-inch pipe.
22 Same conditions, same type of preparation.
23 DR. SHEWMON: This was four-point loading in 24 some way or what?
25 MR. HAZELTON: I guess you'd call that O
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 800-336-6M6
. - . . , . . - __ r - _ . _ _ , , ,__,-. < _=_,+. m.. . ._-
30335.0 98 BRT
(~ ; 1 four-point loading. ,
2 DR. SHEWMON: But the loads are applied way out 3 here someplace and the supports are on either side of what 4 we see there?
5 MR. HAZELTON: Yes. This is pressurized to 6 essentially BWR design conditions; design temperature, and 7 then bent. All right?
8 One can conceive other types of tests where one 9 increases the pressure beyond and so forth. But this is 10 what they did.
11 (Slide.)
12 What made me happy I kind of figured as a
,_s 13 blacksmith that this was going to happen, that you'd have f ) .
14 to bend that pipe a lot more than you can conceive of .
15 pending it in a real plant to make this thing pop loose.
16 As I said, the report will be out shortly. I 17 think you'll find it very interesting reading.
18 DR. SHEWMON: Is the research program doing 19 anything on the rate of stress corrosion crack growth in 20 overlays?
21 MR. HAZELTON: Yes. Yes. As a matter of fact 22 Bill Shack and company at Argonnt have stuf f going .there.
23 And on a mock-up of that type of thing.
24 They thought for a while they were getting crack 25 into the overlay. It turned out that the crack -- this is
('M C1 ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646
30335.0 99 BRT 1 -very high stresses, you know.
(
2 DR. SHEWMON: Yes.
3 MR. .HAZELTON: An' aggressive environment. When 4 ;it hit the overlay it stopped and started going along the 5 interface between the overlay and original pipe the way' you 6 would think it would be because you'd-have some 7 sensitization there. But that's what's happening.
8 DR. SHEWMON: So your impression is the overlay.
9 is of a composition which is resistant to growth plus it 10 has a stress --
11 MR. HAZELTON: Since day 1 we have required that.
12 MR. MICHELSON: What's the present policy on how 13 long you can operate with' overlay? ,
(:)
14 MR. HAZELTON: Well, let's go to the next slide.
15 (Slide.)
- 16 Basically, 1061, volume 1, limited the use of 17 overlays to two fuel cycles unless an inspection could be 18 developed.
19 When those words came out why that, shall we say, 20 galvanized the industry into action and the owners group 21 funded the NDE center for a lot of work on that. We 22 consider now that effective methods are available. They 23 can detect any cracking into the overlay and they also can 24 detect cracking within 1/4 thickness of the overlay. But 25 it requires attention to the surface contour and surface O
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Cmerage 800-336-6M6
30335>0 100 BRT
) 1 finish.
2 DR. SHEWMON: I don' t understand the 1/4 3 thickness, after it has gone through the overlay that it 4 can be detected?
5 MR. HAZELTON: It hasn't hit the overlay yet, 6 but it's 3/4 of way through the original wall, from there 7 on they can detect it and they can detect it through the 8 overlay.
9 DR. SHEWMON: Fine.
10 MR. HAZELTON: And we require overlay welds be 11 inspected every other refueling outage.
12 DR. SH EWMON: All of them or a sample?
,-, 13 MR. HAZELTON: All of them.
/ >
14 DR. SHEWMON: In the areas where the cracks were 15 found before or 360 degrees?
16 MR. HAZELTON: Inspect the overlay. The whole 17 thing.
18 We don't necessarily think that they have found 19 c11 the cracks in that weld before they overlaid it.
20 DR. SHEWMON: This was each inspection or every 21 other inspection? Tell me again.
22 They must be inspected how often?
23 MR. HAZELTON: Every other outage.
24 MR. RO DAB AUGH : Every other fueling cycle.
25 MR. HAZELTON: Every other fueling cycle.
(~b V
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
30335.0 -
101 JBRT' MR. RODABAUGH:' That's six years or something
{= 1 2 -like that.
3 MR. HAZELTON: Basically, it varies with the 4 refueling time.
5 DR. SHEWMON: More like three times. A year and 6 a half is about --
7 MR. RODABAUGH: Is that right?
A 8 Warren, I wonder if _ I could follow up on one of 9 these questions with respect to a statement in the NUREG.
. 1 10 It says, "the Staff recommends no austenitic material be
! 11 considered to be resistant to cracking in the presence of a i
12 crevice such as formed by a partial penetration weld where I 13 the crevice is exposed to reactor coolant" which is
' O 14 precisely what we have in the overlay.
15 MR. HAZELTON: You are right.
16 MR. RO DABAUGH I don't know that I've seen any 17 contradiction, but that sentence meant to me gosh, you t
18 don't want any crevices underneath anything.
19 DR. SHEWMON: Where are you reading?
! 20 MR. RODABAUGH z Page 2-3 of the NUREG. It's a 21 little list of numbers go'ing down to 6. Then there's a
- 22 separate little paragraph.
23 MR. HAZELTON: Right. That was aimed at this 24 joint between thermosleeves and the safe ends in plants 25 that had been finding cracks there, even when the safe end
(
i j
ace. FEDERAL REPORTERS, INC.
l 202 347-3700 Nationwide Coverage 800-336-6646 i
3033530 102 BRT 1 material was 304-L.
a 2 If you want to ask me my personal opinion on 3 whether overlays are going to last the life of the plant, I 4 guess my answer is, well, I think most of them will, but I 5 would not be surprised to have some of them end up cracking.
6 I can't prove it, but I think that if we require them all 7 to be inspected every other refueling outage, we'll find 8 out if there is a problem.
9 MR. MICHELSON: Right now there is no limit on 10 the life expectancy of an overlay?
11 MR. HAZELTON: No. No.
12 MR. RODABAUGH : That sketch you showed of the
,_ 13 overlay showed all the beads -- can they inspect that way?
( "' l 14 MR. HAZELTON: I can't cover everything. But 15 there's strict GA requirements on it. It has to be done 16 completely in accordance with code requirements and they 17 inspect it -- but that isn't to say there can't be some out 18 there with problems. Nothing is perfect.
19 DR. SHEWMON: What was your question? They 20 wouldn't lay it down circumferential1y?
21 MR. RO DABAUGH : Warren's sketch showed how it 22 would be when you finished the welding, but I don' t know 23 that you can inspect it very well witnout now going back 24 and grinding off the beads.
25 You have to come back and finish off that rough p.
\_J ACE-FEDERAL REPORTERS, INC.
~02-347 3700 Nationwide Coserage 800-336-6646
30335.0 103 BRT Q l stuff.
b 2 MR. HAZELTON: Those weld beads are emphasized 3 so you can see they are weld beads in the picture.
4 (Slide.)
5 We had a lot of them that looked something like 6 that original.
7 I can go into it.in more detail, but the 8 inspection requirement carries with it a requirement on the 9 smoothness of the surface. The NDE center has specific 10 s amples . It says it has to be this good, and so forth.
11 This is part of the inspection criterion, part 12 of the examination criteria.
13 So when they are out inspecting them, they don't O
14 look like that anymore. They are pretty nice.
15 MR. HERNAN: Are these inspections always 16 ultrasonic testing or --
17 MR. HAZELTON: It's all ultrasonic. They had to 18 develop some special techniques for this, but basically, 19 the use of diffracted longitudinal, has come through with 20 flying colors here. Of course, at the NDE center, they 21 taught more people who are going to inspect overlays, and 22 there is a qualification procedure.
23 (Slide.)
24 On the primary coolant leakage limits, it has 25 been bouncing around. The limits that we are talking about n
v ACE.FEoERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
30335.0 104 BRT 1 now are essentially, I think they are identical to those
(~')
U 2 that were in 0313, Rev. O. And almost all, to the best of 3 my knowledge, of the operating BWRs are operating to this, 4 either in the tech specs or in some administrat.ive control.
5 That is we are talking about sump measurements.
6 MR. MICHELSON: I thought when I asked the 7 people at Beaver Valley, I thought they told me they were 8 doing it every six hours or 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />.
9 MR. HAZELTON: Beaver Valley is not a BWR.
10 MR. MICHELSON: You put different requirements 11 on PWRs on this, if they have this problem?
12 MR. HAZELTON: Yes. They have to monitor their 13 leakage every four hours and the plant has to be shut down O *
- 14 for action when unidentified leakage rate increases by --
15 that should be a 2 in 'there, 2 gpm' in a 24-hour period or 16 when the total _.u_nidentified leakage reaches 5 gpm.
17 1061 volume 1 recommended a total limit of 3 gpm, 18 but af ter beating this around and trying to figure out we 19 could justify that we decided we'd leave it 5 gpm.
20 Apparently some plants do have problems {
21 identifying the source of leakage and we determined that 22 these 5 gpm some would be adequate.
23 Coming back to what we've done here, we've 24 developed Staf f positions on the following subjects.
25 Essentially, recommendations in the NUREG and there's Staf f O
s4CE FEDERAL REPORTERS, INC. ,
202 347-3700 Nationwide Coverage 800-336-6M6 l
l
3033520 105 BRT
~'
1 positions in the implementing generic letter which you 2 don't have, but you should have a copy of something called 3 " Attachment A," which is the attachment to the generic 4 letter that givea details on all these Staf f positions.
5 (Slide.)
6 DR. SHEWMON: How about attachment l?
7 MR. HAZELTON: That's probably right.
8 DR. SHEWMON: Okay.
9 MR. HAZELTON: I think what I have been talking 10 about indicates pretty well what the Staff positions are, 11 but there's some stuff I might talk about.
12 Staff position on materials is low carbon, good 13 stuff.
t )
14 Staff position on processes has to do with 15 dissolution heat treatment and stress improvement.
16 Water chemistry we talked about.
17 We talked about the Staf f position on weld 18 overlay reenforcement.
19 Staff position on partial replacement, that is 20 if you want to put in a pup piece, fine, but when you end 21 up replacing partial replacement your inspection schedule 22 is going to depend on the materials and processes and so 23 forth that you used in there. We have had some arguments 24 with utilities on that. That's why it is there. It 25 appeared to be obvious.
,fm O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
3033530 106 BRT 1 Then there's Staff position on stress 2 improvement of cracked weldments. At the moment, we are 3 saying two processes are qualified, both have very 4 extensive procedures for doing it and controls on them.
5 That's the conduction heat.
6 Induction heat stress improvement and the 7 mechanical stress improvement -- recently developed by 8 O'Donnell Associates. We looked at it. It looked pretty 9 good. We had lots of questions for O'Donnell and he 10 answered them. We specifically asked Argonne to take a 11 look at it and give us a recommendation.
12 Argonne did that and came to the conclusion that
,_s 13 it was at least as effective, as good as induction heating
( )
14 stress hnprovement and recommended that it be completely 15 accepted and a research information letter has been 16 formally written to us to tell us the results of that. So 17 that's one change from the earlier version. We are now 18 accepting it.
19 Clamping devices, yes, you can still use it, but 20 each case has to be evaluated separately.
21 People aren't thinking seriously -- well, maybe 22 some people are thinking in some very temporary cases, but 23 we haven' t had any real examples come to us for approval 24 yet.
25 Repair characterization and repair criteria,
(~T U
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336 4 46
_q 4
30335.0 107 l BRT 1 r's 1 basically IWP 3600.
U Staff positions on inspection methods and 2
3 personnels -- bdy, oh boy. I just want to make one, I 4 guess, apology for our typists. They are in the midst of 5 an agonizing reorganization; they don' t know where they are 6 going to go and they are busy typing up lists of people and 7 they are being interviewed and these poor girls have a 8 great deal of difficulty in really paying attention to what 9 they are doing on the 5520. I'm saying that normally these 10 typists don't make as many errors as this.
11 MR. MICHELSON: Doesn't that have a dictionary 12 in it?
13 MR. HAZELTON: No.
- 14 You have already found a few cases. I spent the 15 last week finding them and getting those changes. Then the 16 version that comes out has some other glitches in it'that 17 weren't in the original one. It's frustrating.
18 Anyway, Staff position on inspection schedules, 19 we'll cover in a moment.
20 The Staff position on sample expansion, it just 21 says: All right, you do your inspection and you find a 22 crack or unexpected growth of a crack, you have to take 23 another sample and go on like that.
l l 24 And the Staff position on leak detection that we 25 just covered.
l ace.FEoERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-136 6M6
30335:>0 108 BRT 1 As I said, you have the documents there for the 2 complete writeup on each one of these Staff positions.
3 There may be some questions on this.
4 (Slide.)
5 That's sort of the heart of the whole thing, the 6 inspection schedules. We can talk about this.
7 The changeup here from the last version you saw, 8 the one that went out for public comment. One went out for 9 public comment had some words up there that I put in, 10 frankly, with my tongue in my cheek, saying they had to 11 inspect all dissimilar metal welds at all terminal ends 12 plus 25 percent of the other welds. This is essentially --
,_ 13 that is essentially what the code says. But I didn' t say 14 that. Yes -- plus 25 percent of the other welds.
15 The code says you look at all dissimilar metal 16 welds and all terminal ends, and then you pick enough other 17 welds to bring the total to 25 percent. That, I didn' t 18 agree with because it turns out, in the case of a 19 rede sig ned , replaced, recirculation system, the number of 20 welds goes down from about 105 to about 55 to 60 and the 21 number of terminal ends and dissimilar metal welds stays 22 the same. It turns out the number of dissimilar metal 23 welds, terminal ends, is essentially 25 percent of the 24 welds in the whole system, so they would never look at any
\
25 other welds.
I O
\_
l l
ace FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6M6
30335.0 109 BRT
,~) 1 In discussing this with the owners group who 2 canplained I was asking them for too much and then I' 3 pointed out why and they said: Oh, you have a good 4 suggestion. We would never think of not inspecting a good 5 sample of the other welds also. They were in this kind of 6 problem, the code is currently looking at whether they 7 really mean that they should look, you know, concentrate 8 100 percent on dissimilar metal welds and terminal ends.
9 It turns out in our case here in the BWR pipes, we have had 10 much lower percentage of welds crack that were dissimilar 11 metal welds in terminal ends. Smaller percentage.
12 If we are worried about IGSCC, that's not the 13 place to look. It turns out that on terminal end or a 14 dissimilar metal weld, usually i't's a weld to a component, 15 you have a different weld geometry, you are welding it to a 16 heavier section, and you end up with, shall we s.ay, more 17 favorable residual stress patterns. So that's probably why 18 there have been fewer cases of IGSCC.
19 The inside surface of that joint isn' t in high 20 tensile -- isn' t high tensile stress. So I guess it 21 depends on what you are looking for. If you are looking 22 for f atigue, why, maybe that's the place to look. Maybe.
23 But looking for IGSCC, it's clearly not the obvious place 24 to look.
25 In any case, I changed this and I have some v,
ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336 6M6
-l 30335.0 110
'BRT
. (^N 1 words saying the utilities should take a look at their
&J 2 system. With what they know about the stresses and 3 condition of each of the welds, they should make a 4 selection, 25 percent that they'll look at every 10 years.
5 Because the code tells them to do something else, they may 6 have a legal problem, but we have a very thorough paragraph 7 in 10 CFR 50-55-A, that says if they want to use the '74 8 edition of the code to make this selection of welds they 9 can do that if they ask us.
10 I'm telling them if you think it's not prudent 11 to do exactly like the code says, you figure out what you 12 want to do and we can agree with'you.
13 DR. SHEWMON: I don' t see water chemistry 1:) .
14 anyplace-in there.
15 MR. RAZELTON: No. That's in the other --
16 that's in the Staff position on water chemistry.
17 DR. SHEWMON: So this presupposes they are 18 meeting the Reg. Guide? -
19 MR. HAZELTON: No. This presupposes they have 20 not had water chemistry.
21 DR. SHEWMON: That's not what I said, dammit. I 22 said water chemistry and the owners group has water 23 chemistry specs separate from this. And if you use crappy 24 enough water you cal crack resistant material, as you know.
25 So, it seems to me there ought to be something O
i
/\CE FEDERAL REPORTERS, INC.
! 202-347 3700 Nationwide Coverage 800-336-6646
- _ . _ _ _ _ _ _ . _ , . _ , . . _ - _ . _ , _ . ~ . _ . - - - _ - . - _ . . - - - _ - _ _ _ _ - , _ . . - -
e%
v 3
- 39335.9 111
.BRT .' '
1 abodt the owners gro water chemistry or what assumptions 2 you are mak4ng on hoW much sulfate they are dumping in this'
, (
3 thing nith resin beds,'doing nothing on -- you know..
~
a, ,
4 MR. HAZELTON:- You may have a good point..;-I 5 can't'really comment on it at the moment. The only comment 6 I could make is y think these inspection-schedules are sort 1 .
7 l of based on our past 4xperience or our pest worst
- 8 experience including crack growth rates and so forth, that
- ~
9 .are the worst, thatr"eallyrypresentcasesofutilities
' ~
le that have had bad water chemistry in the past.
ti 11 DR. SHEWMON: i 'You mean it's only Kassner and his 12 crowd'that can crack resistant materials or(have a history 13 of cracking resistant materials? Is that what.you ara s
i,s 14 telling me? . ,
\ 3 15 'MR. HAZELTON: Are you talking about this or all 16 of it? ' ,
17 TR..SHEWMON: Let's talk with the first one. If 18 you don't have any limits on water chemistry at all, or you 19 don't give any dredit for water chemistry at all or --
20 where are we in water chemistry? s.
21 You sort of get a warm feeling if they use the 22 owners group' spec and you have h he'. on part of yc# handout
\ \ ,
20 ppm and .3 nicromble it will prevent IGSCC6 but 23 at L 24 that's not in the tech specs. I; \' 1 'l
.' l',.
'\.
25 MR. HAZELTdN: And thati'= 1sn ' t.: assumed in here..
s , .
c s v s l.
Ace. FEDERAL REPORTERS, .!NC. 3 202 347 3700 Nationwide Coserage r 800 336-664, ; .l
__i.
c . t, % .
5
- 4
30335.0 112 BRT l DR. SHEWMON: Then I don' t know that I believe
~}
2 that I like your first line up there. Because if the water 3 is crappy enough, that's really not very conservative.
4 MR. HAZELTON: Talking about new materials that 5 are low carbon that were tested -- essentially tested in 6 the GE test rigs.
7 DR. SHEWMON: The Swedes have cracked their 8 stainless steel, and that was low carbon. I don' t know if l
9' it's low enough to meet your current spec. And they run 10 pretty good water.
11 I don't know what it was, but I'll bet it was 12 . pretty good, knowing the Swedes and their concern here. So L
7 13 if you .get some of the American reactors, which, in the
/ )
14 past, have run a lot dirtier, then I don' t see the basis 15 for waiting 10 years.
16 MR. HAZELTON: We are saying they should do at 17 least half of those in six years.
18 DR. SHEWMON: That's an improvement. I hadn't 19 seen that.
20 MR. HAZELTON: We don't want them to wait 10 21 years before they look at any. Because, admittedly they 22 might , unless we had something in there, we put that in.
23 DR. SHEWMON: It seems to me what you are 24 telling me is you haven't got any credit for water 25 chemistry here unless they are talking hydrogen water lO O
ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
.30335.0 , 113 BRT , ,.
/~'3 1 chemistry?
V 2 MR. HAZELTON: That's right.
l l 3 DR. SHEWMON: I think that's a poor way to l 4 motivate people when you know the difference between crappy l .
! 5 water and reasonably good w'ater is beneficial.
6 MR. HAZELTON: How am I going to quantify this, 7 if I say hey, if you use -- I've got two ways to go.
8 Either, if they don' t use the owners group, make 9 this 50 percent every 10 years,' or if they do use owners 10 group, make it 12 percent every -- you know we can' t do 11 that because that's less than the code.
12 DR. SHEWMON: Tom wants t.o comment.
13 MR. KASSNER: It's difficult for Rick to 7,..
- . 14 quantify the benefits with regard to water purity under 15 normal water chemistry conditions. In other words, even 16 under'the purest water chemistry conditions you can have 17 cracking.
1 18 DR. SHEWMON: You can?
19 MR. KASSNER: You can. And as you get the water
'20 less and less pure you aggravate the situation more and 21 more. But if you want to make a reg or something they 22 could use to quantify, it's really in the case of sulfate, 23 aort of like a continuous thing. You don' t know at what 24 levdl of impurity PPD -- you could give them credit for.
25 It's too damn low, to be honest with you.
() *
-- ACE-FEDERAL REPORTERS, INC.
) 202-347 3700 Nationwide Coverage 800-336-6646
- 30335.0 114
. BRT - ,
3) 1 All of the utilities know that there is a
- (])
2 benefit in getting down and maintaining high water purity, 3 so it really isn' t necessary, I don' t think, for them to 4 get into it and try and quantify this.
5 All the plants are improving.
6 DR. SHEWMON: So here we think the common sense 7 of the utilities will make them accept the owners group?
8 Recommendations?
~ ~
9 MR. HAZELTON: I think that's right. I haven' t 10 talked in detail with Frank here recently, but I think in 11 talking with the utilities, every time they keep coming in, 12 it is clear that they are all now strong believers in water
- 13. chemistry. They are really trying to take steps.
14 Whether or not we should require it is another 15 question. That clearly would be a backfit, whether we 16 could justify it on the basis of reduction in core melt 17 probability, I don't know.
18 DR. SHEWMON: You know, one arbitrary way would 19 be to say if they don' t meet the BWR owners group, 20 everything moves down one level. And since everybody meets 21 the BWR owners group, or you think they do, then -- this is
- 22 the owners group water chemistry -- then he it is not going 23 to hurt them.
24 MR. KASSNER: They all don't meet it yet.
25 There's a couple of outliers, two or three plants.
I- (
4 ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage S00-336-6M6
30335.0 115 BRT f'; 1 DR. SHEWMON: The thing that bothers me about v
2 these meetings is I think the people who come in and talk
'- 3 to us are, what I'll call the top quarter of the class.
4 What concerns me is the bottom quarter of the class that we 5 don' t hear from.
6 Okay. I think that has to do, actually, with 7 one of our comments in the letter we had. I wish there was 8 some way you could do something to make a positive benefit 9 for accepting the BWR owners group, even if that isn't a 10 guarantee of no problems.
11 MR. HAZELTON: Frankly, I don't know how I can 12 do it; under the rules that we are operating.
13, Yes, Frank?
14 MR. WITT: One thing I'm aware of is the 15 utilities are required to commit to the BWR owners group 16 water chemistry guidelines and they are keeping track of 17 that situation.
18 DR. SHEWMON: That's not part of an SALP.
19 MR. WITT: No. This.is INPO.
20 DR. SHEWMON: But it's a typed kind of grade 21 card we give the utilities. I know there's a difference 22 between the NRC and the utility let's call it a grade card.
23 I know that's not the way they like to be classified. I 24 understood what you said. My question was when people do 25 SALP ratings, do we ever get into that kind of thing?
O ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 116 BRT c
() 1 MR. HERNAN: In a SALP rating, the area this 2 might come into, they are rated in licensing activity. One
. 3 of those is response to NRC initiatives.
4 DR. SHEWMON: But the NRC doesn't seem to have 5' any initiative on this. They haven't changed the tech spec 6 requirement. They aren't putting it here.
7 MR. HERNAN: But they are getting to the fact 8 that each utility is going to have to come in with a 9 program; is that correct, Warren? The quality of the 10 program that they commit to with the NRC, as far as-I'm 11 concerned, will be graded as part of their SALP grading.
12 0 That's not really a direct incentive, I agree.
13 DR. SHEWMON: I don' t se,e anything about -- or I
[.
14 see disjointed things about water chemistry. If I look at 15 the Staf f position. here , this is only hydrogen water 16 chemistry which is separate from this general background 17 that we are talking about.
18 MR. HAZELTON: That's right.
19 DR. SHEWMON: Maybe I haven' t heard yet about 20 what you are talking about, or I don't remember hearing 21 about it. But at least from the handout I've got so far it 22 is not clear that it even would be part of that.
23 MR. HAZELTON: No. You are right.
p 24 DR. SHEWMON: I think that's too bad. We ought 25 to, at least, be out there trying to help encourage people l r
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
.30335.0 117 9RT ,
to use the owner group criteria. INPO is doing it on their (a~) 1 2 own plus the utility common sense. Sometimes common sense
'3 doesn' t help you move the board of directors, the guys with 4 the purse strings.
5 MR. WITT: The NRC inspectors out at the Regions 6 check into it, too, and I've seen quite a few reports where 7 they say that they are committed to the BWR owners group 8 chemistry guidelines and that they have the management 9 commitment.
10 I just saw one on Peach Bottom and it was a very 11 good report.
12 DR. SHEWMON: Now we get back into it. Tell me 15 about the bot, tom quarter of the class. You know, you are 14 saying a few of them are doing it and I. believe you. But 15 the top quarter of the class isn't the ones that get you in 16 trouble.
17 MR. WITT: I think we'd just have to make sure 18 that all the Regions are looking at the water chemistry and 19 inquire whether they are meeting BWR water chemistry 20 guidelines.
21 DR. SHEWMON: But it's not part of your 22 regulations.
23 MR. WITT: It's in their inspection modules, 24 this type of thing.
25 DR. SHEWMON: Why, if it's not part of your ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0- 118 BRT (J 1 regulations? I would have thought they would inspect the 2 reg ulations , the rules you guys thought enough of to put
-3 out as a rule instead of these other things.
4 I mean if it's true, it's nice, but I'm not sure 5 I believe you. I'm sure you get it on some, but it's not 6 clear to me why the Regions should give this any priority 7 if you don' t -- or the home office doesn' t in terms of any 8 regulation or rule that I can find.
9 MR. HERNAN: I guess I'd have to put it in the 10 category of performance indicators. As you know one thing 11 the NRC is doing, which is beyond the regulations per se, 12 is potentially trying ' to assess th'e utility's performance 13 and management r,esponsibil,ity by things that they.have some 14 control over.
15 MR. MICHELSO ': This is not one.
16 MR. HERNAN: I don't know if this is 17 specifically one of them, but it seems to me it's in that 18 category.
19 DR. SHEWMON: It could be in that category.
20 MR. MICHELSON: It could be. I just never heard 21 that one, that is potentially a performance indicator.
22 MR. HAZELTON: I think you are right. I think 23 it has only been just in the past several years, maybe only 24 two, that we really have become aware of the importance of 25 some of these slight variations in water chemistry. It is O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
30335.0 119
-BRT
(~') 1 kind.of new. It certainly wasn' t considered that important, v
2 originally.
3 I think whether or not it is important enough to 4 hnpose some kind of a requirement, it is not clear to me.
5 I think in today's climate, if we are_ going to put on some 6 requirement like that then we are right into the new 7 backfit rule which, as you probably realize, is pretty 8 tough. You have to really show by something like a 9 regulatory analysis, and usually using a probabilistic risk 10 assessment, that it really makes a difference. And you 11 have to show that basically quantitatively.'
12 DR. SHEWMON: Well, do you have to show that 13 stress improvement is much more effective than water
-O 14 chemistry to put this ranking of the first two you have .
15 here? What sort of a backfit analysis do you have to do 16 for stress huprovement?
17 MR. HAZELTON: The importance and effectiveness 18 of the stress Lmprovement,has been discussed a lot here 19 before. It's pretty hard to quantify it if you can' t get a 20 stress huproved weld to crack, you know?
21 DR. SHEWMON: If I look at the first two there l
22 and I hear your story on water chemistry, I say: Well, why 23 can't you make the same case for water chemistry that you 24 can for stress improvement?
25 Stress improvement is black and white. You can O
l ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M6
1 30335.0 120
-BRT
() 1 say we did it or we didn't. Whereas Tom says you can' t put 2' a threshold on water chemistry. You just- know that cleaner 3 is better.
4 MR. HAZELTON: What we've said in the Staff 5 position on water chemistry is that if you are go'ing with 6 improved water chemistry and it includes hydrogen water 7 chemistry, we give you a factor of 2 on credit on schedule.
8 So we are specifically saying they are going to get credit 9 if they use better water chemistry, including hydrogen 10 water chemistry.
11 The way I understand it you are trying to make 1
12 me determine the differentiation between so-so water 13 chemistry and owners group rec,ommendations.
14 DR. 'SHEWMON: Yes.
15 MR. HAZELTON: I don' t think I can do that.
16 DR. SHEWMON: ' You don' t think there's enough 17 difference to make it a defensible case?
18 MR. HAZELTON: No. I don't have a quantitative 19 story to justify a difference.
20 DR. SHEWMON: Whereas, with water chemistry, we 21 haven't worked with that very long and we think that's 22 going to be good?
23 MR. HAZELTON: You mean stress improvement?
24 DR. SHEWMON: No, the hydrogen water chemistry.
25 MR. RAZELTON: The hydrogen water chemistry, yes, O
1.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
'30335.0. 121 BRT
{} l we think hydrogen water chemistry with the control of the others, we think is going to be good and that's what we are 2
3 talking about.when we'use the term " hydrogen water 4 chemistry."
5 MR. KASSNER: One quick comment about water 6 chemistry in terms of conductivity which has been talked 7 about here. People are cutting it a little bit finer.
8 They are now asking when you have, let's say, .3 micros, 9 they are asking what is in the water that's contributing to 10 it? In all the things, in some cases what is contributing 11 to it could be very bad. In other cases, it could be 12 rather innocuous. From the EPRI standpoint, they are 13 looking to see if people.are putting that in and can we 4-O 14 the past rules about cor.ductivity and close chloride limits, 15 is history. So when you look to see what a limit should be 16 or where you have higher conductivity, you then have to go 17 back and look and see what is contributing to it and will 18 it, in fact, affect cracking.
19 DR. SHEWMON: And the instrumentation they have 20 is something which wi11 operate reliably at the plant yet 21 will identify what some of those sources are?
22 MR. KASSNER: Yes. At low levels. Sulfate, for 23 example, is bad. Something else, chloride, for example , at 24 the same level, it isn't so bad. So industry is taking a 25 lead in this area through the owners group and EPRI. They O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 122 BRT
() 1 are trying to incorporate not just conductivity, but what 2 is there.
3 DR. SHEWMON: Fine. Okay. Go ahead.
4 MR. HERNAN: Warren, you made the statement that 5 -about a factor of 2 credits is given for hydrogen water 6 chemistry in the inspection schedule.
7 MR. HAZELTON: Eight.
8 MR. HERNAN: Why does it show up on the schedule?
~
9 MR. HAZELTON: No. It's not in this. It's 10 separate. It's in the Staff position on water chemistry.
11 MR. RODABAUGH : I think you say more than that, 12 Warren.
13 MR. HAZELTON: Yes, I did.
O's .
14 MR. RODABAUGH : You say "if improved water 15 chemistry control including hydrogen additions is 16 implemented" --
17 MR. HAZELTON: Yes. I have been continually 18 saying better water chemistry control including hydrogen 19 additions, if you do that, we'll take a look at the 20 situation case by case, hopefully Frank Witt will be 21 looking at that and see whether he thinks they can really 22 do it. And we'll give them, hopefully, about a f actor of 2 23 on schedule here.
24 I could have added that detail in each one of 25 these, with or without water chemistry, but I didn' t think O
l l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 123
- BRT' 11 .it was' worthwhile.
' g(S 2 There's nothing much else here different except
- 3- in the Staff position on this, I note some utilities have 4 some overlays on some very small' cracks, but the' overlays 5 don't meet our Staff position for either standard or 6 designed overlays. There's just-not enough overlay on'it.
7 The owners group and others suggested that maybe 8 after a few inspections, if no change is seen, we could-9 upgrade it to E.- So that's a change I've made. I think I 10 said after 34 inspections, if you see no change, then you 11 can upgrade it to E, which means do it every other outage 12 instead of every outage. ,
13 I think that's basically the only changes except cc:) 14 hopefully cleaning up the language somewhat and making the 15 thing more clear. \
16 DR. SHEWMON: Let me come back to that. , You are 17 saying if you have a crack there and they think, for some 18 reason, it's not going to grow or it's small or something, 19 they can inspect.it and every time and leave it there?
20 MR. HAZELTON: No. What I said is, or tried to 21 say: If the weldment is classified category F because it 22 has inadequate weld overlay -- not just because it cracked 23 and no repair -- but if the reason for the classification 24 is that the overlay isn' t thick enough to meet the 25 requirement, in that case after looking at it four times, O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
-30335.0 124
..BRT
() 1 then I think it is safe for them to start looking every 2 other outage. That's all.
3 DR. SHEWMON: This presupposes they can talk 4 about -- or that that crack, indeed, is inspectable through 5 the overlay.
6 MR. HAZELTON: Yes. That's the conclusion that 7 we've come to.
8 DR. SHEWMON: But what you have presented to us 9 before was if it comes through -- 3/4 of the way through 10 and outside, then we can see it.
11 MR. HAZELTON: That's right.
12 DR. SHEWMON: There could be some growth up to 13 that level, but that's a no-never mind, I guess.
O. .
14 MR. HA2ELTON: I can go into more detail.
15 Most of the cases that would come under this 16 situation would be cases where they had an axial crack that 17 was essentially through a wall, or we considered it to be 18 through a wall. But just a short axial crack.
19 So all they did was put enough overlay on, 20 really, to make a leakage barrier. Of course, it also 21 helps in case they didn't find any other cracks. But they 22 put on enough for a leakage barrier and then decided they 23 wouldn't put any more on so it is that kind of a situation 24 where most of these are where they have just a little bit 25 of overlay.
O ACE. FEDERAL REPORTERS, INC.
l 202-347-3700 Na'ianwids Coverage 800-336-6646
30335.0 125 BRT
~^
g 1 DR. SHEWMON: Okay.
(U 2 M R. HAZELTON: The last time --
t 3 (Slide . )
4 MR. RODABAUGH : Do you have any repairs made by 5 gouging out the crack? .
6 MR. HAZELTON: I know none that were done other 7 than a little bit of touchup before the overlay. They 8 generally grind down a little bit and then kind of peen it 9 shut. Then hand weld over it to repair it and make it leak 10 tight and then put the overlay on.
11 Other than that -- no.
12 Okay. The last time you had three comments in
,_s 13 the letter. One was obviously a technically significant
14 typo, it came out .08 and should have been .03. That's 15 probably because my writing is bad.
16 The important one here is that you suggested '
17 that we require two mitigations for elimination of the 18 augmented inspection, taking it to category 1, that is 19 using resistant material and either stress resistant or 20 hydrogen water chemistry.
21 There's no way I can stand here and disagree 22 that that isn' t a good idea for them to do that. We think 23 obviously, if it were my plant and put in new material, I'd 24 put in stress improvement on every weld and I would also go 25 to hydrogen water chemistry as soon as I could. However, O
i ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
' l l
30335.0 126 BRT
(; I we are in a situation where, how do we quantify the value 2 you'd get from it? This question was kicked around in the 3 pipe crack task group. It was kicked around for, oh, 4 several days. Not totally on that. But it was looked at 5 very carefully and it was decided that if you read 1061, 6 you see that it strongly recommended that you use belt and 7 suspenders and safety pins. But it only required that you 8 have a belt on to go to -- go back to code inspection.
9 That is, it only required resistant material.
10 Part of -- there was a lot of thinking and 11 discussion that went on on that. This was the conclusion 12 that they came to, and I saw no reason to do any different
,, 13 when I wrote this.
( )
14 Although, clearly, the more you do the better, 15 but it's a matter of what's necessary and whether something 16 should be a requirement or not. And that's the state we 17 are in.
18 I might point out that Rev . 50 and Rev. 1, both 19 said replace the material with resistant material and you 20 are home free. Okay? And that's been our statement all 21 along.
22 If now I'm going to say no, you have 'to do 23 something else, that's now a new requirement. It's a 24 backfit. And I don' t know how I can justify it on any kind 25 of a regulatory analysis. Although, clearly, it's better.
p V
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6646
-30335.0 127 BRT
(} l . So, we decided that we wouldn' t put that in.
The next one, and, again, it is something we all 2
3 would like to have, is if they have a lot of cracks we 4 think they ought to replace the piping. However, the pipe 5 crack task group didn't say so and we couldn't find any 6 real technical justification for it that we could quantify 7 and we had not been imposing such a limit and we thought it 8 would be very difficult to impose. So we were not going to
~
9 include it. But the CRGR suggested that we put it in to 10 get public comment on it, so we did. And we got a lot of 11 strong public comments. My ear was hurting from telepnone 12 calls.
13 Let me just say the problem that I have with it 73 O 14 is, certainly, not philosophic. In the proposed generic 15 letter and NUREG, we have statements currently that say 16 that we,-the Staff, believe that the most certain way to 17 stay away from crack problems is to replace the piping. If 18 you have seriously degraded pipes. There's no question 19 that's the way we feel.
20 One of the problems that I had, when this was 21 suggested, was how do we define the crack? W'll, e I think 22 it was the definition, the way I interpreted it, it meant 23 anything that would not be acceptable by Section 11 without 24 evaluation. In other words, the 3500 criteria. Anything 25 beyond that would be counted against the 25 percent whether l
l (~
V)
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
~30335.0 128 BRT 1 repaired by overlay or not.
s 2 The problem with that that I have, the main 3 problem,.is that then we start emphasizing very accurate 4 and precise measurement of crack depth. IWB 3500, 5 depending on length of crack and everything, is talking 6 about anything up to about 10 percent of the wall is okay.
7 Beyond that, then you have to do an evaluation or you 8 repair it so the cutoff point is, call it for simplicity, 9 10 percent.
10 Well, the boys have made a lot of improvements 11 in accuracy of crack depth measurement, but believe me 12 there's no way they can tell the difference between, say, 9 13 and 11 percent. In fact, they'd be very lucky to tell the
('
difference between 5 and 15 percent.
14 15 DR. SHEWMON: I'll grant you that part. I guess 16 I hadn't thought as I said to you privately, if you end up 17 with repairs on a lot of your welds, you are building a 18 system that you sure wouldn't be able to get past code if 19 you started in that way.
20 MR. HAZELTON: That hasn't proven to be true.
l 21 DR. SHEWMON: You mean the code is allowing 22 socket welds now?
23 MR. RODABAUGH: Partial penetration welds.
l l 24 MR. HAZELTON: The code has allowed cracked l
25 welds in IWB 3600.
l ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
30335.0 129 BRT
/~T 1 MR. RODABAUGH: But not the construction code.
V 2 Not section 3.
3 MR. HAZELTON: That's right.
4 (Discussion off the record.)
5 MR. HAZELTON: I don' t know whether I got my 6 point across. What I'm saying is --
7 DR. SHEWMON: You got your point on the 8 inspectability limit, but I think that's a nit. I'd 9 cheerfully go back to saying 25 percent repaired welds, but 10 you wouldn't accept that either for the same reasons. 5'd 11 let you out of that one and say a cracked weld was one that 12 was so badly cracked it had to have an overlay on it.
MR. HAZELTON:
13 Then I'll tell you what some
']
'- 14 utilities have already told us; that is, heyf we've gone 15 back and reviewed the UT measurements and we find out that 16 that one we put an overlay on really didn't need it, it was 17 cnly 5 percent. Or we find out it wasn't even a crack.
18 That's the kind of argument, if we had a 25 percent limit 19 on, you can be sure that everybody would find a whole mess 20 of them that could be analysed as 9 percent deep instead of 21 10 or 11.
22 Because we had already been having that kind of 23 argument from the utilities or discussion from the 24 utilities, I recognize we'd be in a continual argument with 25 the utility about whether they had 24 percent or 26 percent O
Ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
_ . ~ . . _ . - . . . . - , _ . . - . __
30335.0 130
-BRT cracked. So, although I don' t -- I personally think when
( ) I 2 they have a number like that, why they ought to replace the
.3 piping. But I can' t make a hard enough argument on the 4 basis of Lmprovement in core melt probability to say that 5 I'd require it.
6 DR. SHEWMON: Ev?
7 MR. RODABAUGH: Warren, looking at your sheet 8 here, I see lots of cracked welds. Are you aware of any 9 plant that would be anywhere near close to this 25 percent?
10 MR. HAZELTON: Oh, yes. Depending on, I think, 11 this is important -- whether you -- you made your 12 recommendation and I couldn't quite see how we could 13 implement it. It needed to be-more definitive. So I was 14 .
told to make it more definitive, what I thought was 15 reasonable and what I thought was the intent of the ACRS.
16 So did you -- we are talking about, say, a total of maybe 17 225 welds in the plant. If you talk about the total numcer 18 of welds or you are talking about system by system, see, it 19 appears to me that system by system or pipe run by pipe run, 20 it makes more sense if we are talking about being concerned 21 about unusually high stresses in it. So that's the way I l
22 worded it, hoping that that was, you know, the kind of 23 thing you were intending. And, yes, there are plants that 24 have more than 25 percent of the welds overlaid in certain 25 runs or certain systems. But, again we have to be a little O
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6616
s 30335.0 131 BRT careful. A plant might have two welds in, let's say the RHR (V) 1 2- suction, that are stainless steel and the rest of carbon 3 steel.
4 If one of those had a little crack and they put 5 an overlay on it to be eafe, 50 percent of them are cracked.
6 You get into a complex auditing thing that is 7 pretty hard to come up with any simple requirement that 8 doesn't have problems associated with it, particularly when 9 we are talking about: Yes, there are plants out there that 10 have 25 percent or more of certain systems and certain 11 lines cracked and overlaid, i
12 The other, if you want to call it the 13 bureaucratic. question, would be: All right; if.we had 14 indicated that we really wanted those overlays to be safe, l5 which we did in many cases, we argued with them because of 16 uncertainties on their measurement and so we convinced them 17 they should put overlays on and now in we come back and say, 18 aha, we made you put all of those on, but now we are going 19 to make a requirement that you can only have so many --
20 that's a real difficult position to justify unless you've 21 got some very hard facts and good quantitative analysis, 22 which we don't think we have.
23 DR. SHEWMON: Okay. No other questions.
24 MR. HAZELTON: Public comments.
25 ( Slide . )
O ACE FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coverage 800-336-6646
~l 30335.0' 132 BRTL The resolution of_the public comments are still, a
)- 1 2 I-guess you can say, technically predecisional. However, 3 there are only several that are at all likely to be 4 different when it finally comes out. You have the details 5 on all of these there. If you need a copy of the owners 6 group comments so you can follow what I wrote a little 7 better, why, that's available, but I think you have that.
8 You had that earlier.
9 Basically they had a problem with casting. Cast 10 valve and pump bodies. When they replace pipe, they 11 normally don' t replace the pumps and valves that were in 12 there.
13 The pumps and valves that were in there were 14 made of casting that had high enough carbon, low enough t
15 ferrite, so they would not meet the current Staff position 16 for resistant materials. Therefore, the owners group was
- 17 concerned that if they replaced the pipe, they'd still end 18 up with every pipe to valve and pipe to pump joint being a
, 19 category D weld and requiring quite of ten inspections, 20 whereas they didn't think that this was justified on the 21 basis of experience.
22 I had already come to the conclusion that I 23 wasn't going to assume that they were nonresistant welds, 24 but I hadn' t put that in the documents. It just slipped 4
25 through the cracks.
)
i I /LCE FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coserage 800-336-6646
- _ _ . . _ - _ _ - . . _ . _ . . . _ _ . _ __ _ _ . _ _ _ _ , . ~ . _ . , _ . . _ - . . _ _ _ _ . _ .
'30335.0 13S BRT
(~') 1 So we decided to discuss that. I had a group, u
2 including Bill Shack, we got together and talked about it, 3 and Bill was able to point out some very important things.
4 Even though the tests that had been run and reported on by 5 General Electric would indicate that materials with carbon 6 content and ferrite content comparable to these actual 7 pieces were subject to sensitization, it was clear that 8 they were not nearly as susceptible to sensitization as, 9 say, normal 304. It was somewhere in the middle between 10 resistant and nonresistant materials. You know, you can't 11 make a sharp gradation.
12 Also, the important thing that Bill Shack and I 13 thought of, and we agreed on, was that because it is a
)' 14 heavier section, in the bump, the residual stress pattern 15 is bound to be better, or I thought it was. So Bill
^
16 started looking around and found results of calculaticns 17 and so forth that showed it was far better. That joint 18 doesn' t have the high residual tensile stress on the inside 19 of it, just because of the geometry and general dynamics of 20 the welding process. So I came to the conclusion that 21 unless there has been something done, like a lot of weld 22 repairs to louse up that favorable stress distribution, 23 that we could consider them to be resistant joints for 24 purposes of the inspection.
25 Another question that they had was that some of d
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
30335.0 134 ;
BRT l
'N-f') 1 the welds they had been using, and I think possibly, ,
2 although we weren't clear, in putting in replacement piping, '
3 replace it with low carbon material and low carbon weld 4 metal. But they didn't control the ferrite content to as 5 high levels as was recommended and included in our Staff 6 position. Their comment, like 5 percent ferrite instead of 7 8. They complained that we might consider some code-approved 4 8 materials requiring augmented in-service inspection.
9 Of course, this doesn't bother me because the 10 code has a lot of approved materials that are subject to 11 cracking problems and the code says the owner is supposed 12 to look out for that kind of thing.
13 At an'y rate, I have writt'en some words to say
.( 14 that it is certainly true if the carbon content is low 15 enough, 5 percent ferrite is sufficient to keep you out of 16 trouble. Therefore, we'll look at it on a case basis, but 17 I can' t, at the moment, quantify the synergisms between the 18 two enough to give a table or fonnula. Maybe -- that might 19 be done in the past, but what I'm saying is we'll look at it.
20 One Lmportant thing was originally I had that we l
21 were going to: class 1, 2, and 3 piping systems were to be 22 included in here. But, remember, we are only talking about 23 systems with stainless steel piping larger than four inches 24 that contains reactor coolant with temperatures over 200 25 Fahrenheit.
O ACE FEDERAL REPv.o ERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646
c30335.0' 135 BRT..
Now, it turns out, people complained that -- we
. ( )' 1 2 were going to make them do an awful-lot of inspections of' 3 class 3 things that .weren' t wanted and so forth, so I 4 started asking questions. What have you got out there in 5 class 3 that would meet these other requirements.
6 Well, I kept getting sheepish answers back in a 7 week or so: Hey, we don' t have any. Sorry.
8 But it turns out some plants, I knew, did have.
9 Some older BWRs have the part of the reactor water cleanup 10 system outside of. containment, past the second valve. That 11 is stainless steel. Most of the new plants, it's all 12 carbon steel. But there are a few old plants and I think 13 there,are like five or six plants in that, but I haven' t
~O- 14 been able to determine that. It.is hard to get an answer i' '
15 on that question.
16 I have discussed this situation with several of 17 the utilities. In fact, some of it is unclassed and' 18 sometimes it's class 3, for reasons that I don' t understand 19 and neither does the utility.
20 MR. MICHELSON: It's also nonseismic, in some 5
21 cases.
22 MR. HAZELTON: Yes.
23 In talking with some of the utilities they -- oh,
- 24 one problem was we don't want to inspect that because it's f
. 25 so hard to inspect, it originally was made to B31.1, and
(
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6646
30335.0 136 BRT
/
1 there wasn' t any radiography required, there wasn' t any
[]
2 inspection required and the welds were made by people that 3 don't make neat welds and they look bad. They are just not 4 suitable for inspection by UT.
~
5 My feeling is that all those things are also the 6 things that make me worry about whether it's going to be 7 subject to IGSCC.
8 Without trying to make any big safety argument, 9 which I can't, it just appears to me that it would be 10 prudent to look at those. I don' t think it involves that 11 much work for that many utilities. So my current words say 12 that the scope, as you see, is going to be, regardless of 13 code class, if it's made of stainless steel and carries 14 reactor coolant during normal operation over 200 Fahrenheit 15 and over four inches, you ought to look at it.
16 ~~ I don' t know whether it is going to stay that 17 way as it goes through CRGR.
18 MR. MICHELSON: Why did you do the cut at over 19 four inches?
20 MR. HAZELTON: That's historical.
21 MR. MICHELSON: It certainly has no relationship 22 to safety then, I guess, because a four-inch pipe is a 23 pretty dangerous pipe break outside of containment.
24 MR. HAZELTON: No, four inches and up.
25 MR. MICHELSON: I thought you said over 4 inches.
( -
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
30335.0 137
-BRT
(~N 1 MR. HAZELTON: It includes four inches.
%)
2 M R. MICHELSON: Because much of that reactor 3 water cleanup is four-inch water piping and some smaller 4 and there's a little bit of six.
5 But you can certainly make safety arguments, if 6 you need some help, I think you can find help to make the 7 safety argument.
8 MR. HAZELTON: You have a copy of my resolution 9 of that public comment.
10 MR. MICHELSON: I don' t have it, I don't believe.
11 Is it in the package?
12 MR. HAZELTON: Yes.
13 MR. MICHELSON: Okay. -
<:) 14 MR. HAZELTON:
I think I make a pretty strong .
15 argument, but I don't know if it's strong enough.
16 MR. MICHELSON: There's no trouble making some 17 very interesting safet.y arguments about the failure of that 18 piping, including the qualification of the valves which 19 have to close in order to assure you don' t blow the reactor 20 down outside of containment. It's a small break LOCA of 21 very significant proportions.
22 MR. HAZELTON: Even if the valves do close, you 23 can' t ever run very long without that reactor water cleanup 24 system or you are going to violate tech specs.
25 MR. MICHELSON : For a situation like this, you O
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
30335.0 138 BRT
() 1 don' t worry about violating tech specs.
2 MR. HAZELTON: They'd have to shut down.
3 SR. MICHELSON: Well, I would hope anyway, but 4 it's an accident of interesting proportions.
5 ,
MR. HAZELTON: I'm glad you agree with me.
6 Let's see where we go on that.
7 MR. RODABAUGH: You end up saying even with P31 8 part 1, it is still covered?
9 MR. HAZELTON: Yes.
10 MR. RO DABAUGH : If it's four inches or over.
11 MR. HAZELTON: From one other standpoint, and I 12 have been talking with a few ancient and honorable code s 13 people with both the ANSI and the boiler code , and asked
.gJ 14 them the' question: If you had a piece of pipe' originally 15 designed and built IWB 31.1, and now you know it has a 16 crack in it, could you say that it still meets the original 17 code which was the original design basis for the plant?
18 That's a rhetorical question, and it gets 19 messier the more you think about it, but I'm saying it 20 doesn't meet the original design basis for the plant.
l 21 There is no original licensing basis --
l 22 MR. RODABAUGH: I agree. In the old codes in 23 the welding requirement, there's something that says you 24 shouldn't have anything like a crack.
25 MR. HAZELTON: Exactly. Now, if you know you O
l ACE FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-6646
- g. . .. -. - - .
7-y sb '*
- w. O - i7 c
g- s' *. v . s;
% . .g ,
vr 3033'5.'0; ,'- s ;
'\ O. /
.139 i BRT ;,
.s f,( . h_ s )
4 i 1 have cracks,'but you a ko,tjoing to look fd23,Sem becad 0 , ,.
2 you might find them,dI do
~
thjhk that's real meetirs-It[. -
a .- u < t 3 MR. . MICHELSO.;,) Another problem. you "might be ,) .
s-4 running into is that' out in that area of being ,6jnonsafety 6 '
y :-
v 3
5 rela'ted-e,ystem, non-QA and so forth, they also paerenyy#as
., s 6 careful with sensitization of the matirials.as they were '
Q .
7 doing the welding ard whatever.- Thei may beypretty '
8 sensiti :dd materials because there was no ra\pirement-
.9 otherwise.
l t%.
4
\'
.10 MR. HAZELTON: Exactly. , Exactly. /
. s l* > b 11 MR. MICHELSON: In some plantsi, ddpending on how ', r-
.a > '%.,U ,, z.
12 they arranged to veht the compartments 'and'soiforth where'
<, g- m 6, Ml -
! 13' these breaksica'n occur, tnis could be an extremelp M4 l- t /f y N g \m iQ,'
l 14 dangerous breaV. I hav9heen~ narping on b$ at one for' a % C '.
15 w
long time, we ?really need tio look at those systems closely v
4
, ?
- ) )
16 because they are essentially an extension of full preissure -
4 s ,
i 17 boundary, f ull temperature, full pressure, you are fully \
{>
18 dependent on detection'and isolation of the break, you are
- (
19 dependent on the valves.' capscity
- to closuunder thohe 7 w
' i 20 conditions, they have not be'an bought that way, the old 3 l 21 plants, there's no data that shows they;have been tested;
! 6 / ,
- - 22 there's no;way to do an in-service inspection to assure the
- s ), ,
23 valves woUld close under break -- it's just < a very bad
\
24 situation and it's a very bad break. It's not<very good 7 .,
v -
25 material by comparicon with what we've done ondhe fprimary Lo s s,
i a ACE-FEDERAL REPORTERS, INC. ,
202 347-3700 Nationwide Coverage . 800-336-6M6 Y
! s L. -. -. _- - . . . .-
i _.,.
- - _ _ .___ L;
30335.0 140 BRT
- \Qi]
- 6['I .1 . break boundary.
- t -) .
(<
I'm not sure, by the way -- I'm not sure you
/ N' 2
\ .-
' ^ Jlf -
3f have provided much inspection service inspection out there 4 either, beyond the isolation valves. I don't think there's
"\ 5 any.
, ( 6 MR. HAZELTON: You have to look visually at some
- t. ,3 kg 7 of them.
8 MR. MICHELSON: But there's no requirement. Is h there a requirement to even look at it?
I 4 10 ,
MR. HAZELTON: 'The code has a requirement that c ti
!, 11 class 3 be looked at initially.
epr'>h s . g I ' \' 12 MR. MICHELSON: This is not class 3.
13 s MR. HAZELTON: Some of it is and some is unclassed.
. 14 t MR. MICHELSON: And it's only through the second U lsolation valve, that's class 3, that they look. The other, 15 3 5'
- i. 1 16 you don't ever have to go look at it, not unless the room i .
Q7 starts steaming up and that's the point where you start
!Y g, 18 looking to see where it's leaking.
i 19 MR. HAZELTON: That's right. At any rate, if we 20 don't make them look at it and one has a great big leak 21 that gets a lot of media attention, I just want to be on
( 22 record for saying that I said they ought to look at them.
23 MR. MICHELSON: Those can also be activity 24 levels way beyond what's in the water itself, too, because 25 there's all the resin beds and whatever out there that will O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
7_ _y . _ _ _ _ _ - _ _ _ _ _ _ - _ _ _ - _ _ _ _ _ _ _ _ - - - _ _ -
- p. =
$30335.9 141 ERT,p .,
t
~
r'~yC 1 bl'ow back to the leak.
G y; 2 MR. HAZELTON: This was stated as an objection 3 to looking at them because there's high radiation.
4 MR. MICHELSON: Yes. It's hotter than a pistol 5 out there and it's a ppoblem all right,3 but that can't
~
6 relieve us of.the safety. concern.
7 MR. HAZELTON: That's probably the most 8 sensitive point in the resolution of public comments'. "
l 9 ,
A simplified -- just use a fuel cycle basis for 10 'ISI. schedules instead of trying to go with 3 1/3 years, the 11 ,
-way the code was working, because right now the code
's 12' l~ perio'ds don't jibe with hhat the utilities really do, l-(131 , trying to make the 3 1/3 jibe with what the code says.
/ , ,
, 114 Three 3 1/3 is 1e < years and so forth.
- l. '
l '15 Extended exams for category A, I discussed that 16 earlier in detail. They said they picked.the 25 percent of 17 welds' they think are most important to look at, and they in 18 make a judgment, not us. Credit for MSIP, I discussed that.
!/ 19 There was one comment that some people felt 20 rather strongly'about. That is that once they replaced the l
21' piping, then the inspections shouldn't have to be done by 22 people qualified to find cracks. I said that's ridiculous s 23 and puts us back where we were before. So my resolution is:
- t 24 m Leave it that way. I think anybody inspecting things for 25 cracks should have to prove he can find cracks.
M ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-236-6 4 6
_L _ _ _ _ _ _ __ __
30335."0 142 BRT' ,
1 ~MR. MICHELSON: What were they thinking they D) 2 were inspecting for, if they aren't looking for cracks, 3 what were they looking.for?
4 MR. ETHERINGTON: Absence of cracks.
5 MR. MICHELSON: They will assure that, if they 6- are not qualified.
7 MR. .HAZELTON: The code says you turn the 8 machine on and you find a notch in the calibration block.
9 Then you go on the pipe and you rub it thusly. (Indicating.)
10 And they did.
11 DR..SHEWMON: About five years ago some of the 12 industry people were saying, you know, if they define what 13 they are looking for, v: En find it with more cert-ainty.
14 I weht back to some of the Staff and the Staff said: No, 15 we don' t want to be too specific because we may be losing 16 out something else. To say we are looking for the end of
! 17 cracks is, indeed, progress.
! 18 MR. MICHELSON: Yes. If they stick to their 19 guns.
20 MR. HAZELTON: So where are we now, right?
21 (Slide.)
l 22 What we are going to do is turn out NUREG-0313, 23 Rev. 23, which will be transmitted to licensee by generic 24 letter. Really the generic letter, as I pointed out, is 25 the important thing. The generic letter will request the O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
?30335.0 143
., ; 'BRT
~N' licensee to provide their plans, ask them whether they plan (J 1 2 to1 follow the Staff - positions, if not, what they plan to do 3 that they think is just as good, and propose appropriate 4 changes. These words are not completely accurate because
, 5 there have been some changes in the thinking. ,
6 DR. SHEWMON: On a generic letter like this,
'7 every BWR owner has to write back something?
8 MR. HAZELTON: Yes. In he doesn't, he'll wish 9 he had, I guess is a better way to put it. They don' t have 4
10 to do anything.
11 (Slide.).
12 Currently, where we are is down here talking to 13 the ACRS. I can say as far as I'm concerned, any ACRS G
\~# 14 support on this is -- would be welcome. As of yesterday 15 we -- it looked like we might even get a CRGR meeting in 16 April, maybe around the middle of the month. That's not 17 certain by any means yet.
18 I drafted a Commission paper and if we get 19 through CRGR, hopefully we'll send the Commission paper, 20 .this whole package to the Commission, and we expect there 21 will be a Commission meeting. It is going to be 22 interesting because the Commission meeting will almost 23 certainly occur after the reorganization and we'll have a 24 whole new set of actors on the subject.
25 DR. SHEWMON: When is the reorganization plan i
- ($)
ACE FEDERAL REPORTERS, INC.
l 202-347 3700 Nationwide Coverage 800 336-6646
(
l 30335.0 144 ORT 1 due out?
~j 2 MR. HAZELTON: April 12th.
3 DR. SHEWMON: As long as it's not April 1st.
4 (Laughter.)
5 Well, thank you very much for your presentation.
6 MR. HERNAN: Dr. Shewmon, as Warren indicated, 7 we appreciate your support, but we are not asking for a 8 full Committee letter unless the full Committee feels moved 9 to write a letter. We are required to give a full report 10 to the full Committee April 9, and as far as we are 11 concerned the documentation and the full Committee minutes 12 will be enough for us to say we went back to ACRS and
,_s 13 discussed the resolution.
('~)
14 DR. SHEWMON: Fine. Ok'ay. Is that it?
15 Gentlemen, I guess we are on schedule. That 16 doesn' t sound right.
17 MR. MICHELSON: Exactly on schedule.
18 We'll break for lunch, we'll come back and if 19 you run out of things to do at lunch, if you go find John 20 McKinley, you'll find a draf t of this letter that I'd like 21 to take up with you between 3: 30 and 4: 00 or 4: 30. Well, 22 you have to find Al Igne. Why don't you pass it out.
23 Adjourn for an hour, then.
24 (Whereupon, at 1:00 p.m., the hearing was 25 recessed, to reconvene at 2: 00 p.m. , this same day.)
g
_/
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 145 BRT
'^
l AFTERNOON SESSION (2:00 p.m.)
m 2 DR. SHEWMON: This afternoon, the remaining item 3 on the agenda, we will get to hear about the status report 4 on the hydrogen water chemistry from Jeff Cass of G.E.
5 MR. CASS: I'll give you a chance to get the 6 copies passed around. Thank you very much for giving me a 7 chance to update you. I think last time I was here was in 8 1984.
9 (Slide.)
10 I have also asked my colleague, Dr. Robin Jones 11 from EPRI, to join us. There may be some questions you may 12 ask that it may be more appropriate for him to answer or
, 13 comment on.
- '~ 14 The focus of my talk is going to be on IGSCC 15 prevention by use of hydrogen water chemistry. I'm a lot 16 more familiar with the materials aspects of this than I may 17 be on the chemical -- chemistry aspects.
18 (Slide.)
19 Just to give you a veer brief overview here, in 20 -general, hydrogen water chemistry offers an opportunity for 21 utilities to avoid major replacements of structural 22 components. It works by changing the radiolysis in the 23 core through small additions of hydrogen to the feedwater 24 which supplies radiolytic production of oxygen and other 25 oxidizing species in the core.
/~^ .
L l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 146 BRT 7'N 1 If proper amounts of hydrogen are added we can ci 2 reduce the oxidizing power below the threshold where IGSCC 3 is observed.
4 We do have very strong positive evidence that 5 hydrogen water chemistry inhibits intergranular stress 6 corrosion cracking from fundamental studies, fundamental 7 modeling studies and electrochemical studies; an awful lot 8 of laboratory testing at laboratories throughout the world, 9 full scale pipe tests and now an increasing amount of 10 direct, in-reactor test data.
11 We have looked at a broad range of structural 12 materials that could be affected by this addition program 13 and we know that it will arrest existing stress corrosion I
() 14 cracking in piping and vessel materials as well'as the 15 reactor vessel internals.
16 We have some limited data which we hope will 17 shortly be significantly expanded to show that hydrogen 18 water chemistry offers the promise for mitigation of stress 19 corrosion cracking in highly irradiated stainless steels 20 and there is a confirmatory test program which will reach a 21 very important stage here in one of the next few weeks at 22 one of our operating plants.
23 So, the reason hydrogen water chemistry is 24 important is because it offers the potential to protect the 25 structural materials throughout the plant.
O ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
30335.0 147 BRT c'.
) 1 (Slide.)
v 2 Just another introductory chart, I'm sure you 3 are familiar with this, but the way hydrogen water 4 chemistry works is to suppress the radiolytic decomposition 5 products that cause increases in the oxidation potential of 6 the coolant. If we have excess hydrogen present, then 7 those oxidizing species are Gettered.
8 The Gettering mechanism relies on the presence 9 of short-lived, free radicals that would catalyze the 10 combination.
11 DR. SHEWMON: If we can stop with that first 12 reaction which, you'll forgive me -- it would be nice if 7- 13 the hydrogen balanced, for example . But what you have done
('~') .
14 is normally free hydrogen is given off as molecular 15 hydrogen, therefore you --
16 MR. CASS: It's very simplified.
17 Pr7fessor Shewmon, this is not intended to be an 18 exact balanced equation. The number of equations and 19 processes that go on in the core are very complicated, 20 there's a very large number of reactions that actually 21 occur. I'm trying to simply show you schematically what 22 happens, and that's true for this case and this case.
23 There's no intent here to make an exact true chemical 24 equation. Just trying to make a repre sentation of the 25 processes.
O L;
ACE FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coserage 800-336-6646
30335.0 148 BRT.
'DR. SHEWMON: Okay.
]' l 1 2 MR. CASS: In principle, the' process works like 3 this.
4 (Slide.)
5 Looking at oxygen as an oxidizing species, as
~
6 you know there are many oxidizing species that are 7 available, but we know that as we suppress the oxygen, the 8 corrosion potential decreases and we have found empirically 9 that when that corrosion potential is suppressed below 10 about minus a quarter of a volt, then we have no IGSCC.
11 DR. SHEWMON: This quarter of a volt is measured 12 in some reference electroid, that as much as an equilibrium 13 process; is that right?
' * - - Thare's a process 14 MR. CASS: That's correct.
15 that we use and other organizations use the same or similar 16 kinds of processes. Most people will use at least one 17 reference electroid, silver chloride, silver-silver 18 chloride measured against stainless steel, and we will use 19 a second, typically copper-copper oxide, again measured 20 against stainless steel. Most people will also use 21 platinum, and a reversed equation as a crosscheck. As long 22 as the platinum behaves in a reversible fashion, we are 23 comfortable we are getting the right answer.
24 (Slide.)
25 There are some significant complications O
14CE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M6
1 30335.0 149 ERT-
- ('~T - 1 resulting from hydrogen water chemistry implementation.
U 2 The most obvious is the increase in main steam line 3 radiation during operation and I'll give you some examples 4 of the extent of main steam line radiation that has been --
5 increase -- that has been observed in the significant 5 number of plants that have conducted some tests to 7 determine that.
8 ItLis possible to turn the hydrogen off, in 9 order to do turbine maintenance, for example. If a utility 10 was worried about maintenance that had to be done in the 11 turbine area, that's an area where a radiation increase 12 resulting from hydrogen water chemistry would be 13 s ign if.icant , you can just turn the hydrogen off and when O 14 that happens, due to the relatively short half-life of 15 these species, the radiation levels return to normal very 16 quickly.
17 We have to have very careful operation of the 18 off-gas and recombiner. We are no longer producing large 19 amounts of oxygen radiolytically and so we have to add 20 oxygen deliberately in order to make up for the amount 21 that's not produced in the core.
22 We have to have good water chemistry operational 23 practices. It turns out if the conductivity of the coolant 24 exceeds several tenths of a microsiemen, then the benefits 25 of the electrochemical improvement from oxygen suppression
(
l l
l l
l ACE FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Cearage 800-336-6646
30335.0 150 BRT
.1 is lost. Operation at very small amounts of oxygen with a V(~T.
2 microsiemen per centimeter, for example , of sulf ate , will 3 result in the same amount of IGSCC if you had the normal 4 200 ppb oxygen.
5 Of course, there has to be a hydrogen addition 6 system. There have been a number of systems that have been 7 devised. The amount of hydrogen that has to be added 8 varies from plant to plant. The experience to date is 9 about .3 to 1.5 ppm. That, of course, can increase the 10 plant's operational costs. Then, of course, there would 11 have to be some sort of a capital investment for an 12 injection system.
13 What I'm trying to get to here is that we have 7s 14 made a point to the utilities that managing hydrogen water 15 chemistry operations is not just a plumbing job. It 16 requires more than just installing a gas addition system.
17 DR. SHEWMON: On the second item, would you.
18 expand on that a little bit? There's already some hydrogen 19 coming off, " requires careful operation of off-gas and 20 recombiner." This is additional hydrogen coming off?
21 MR. CASS: No. The amount of oxygen that is 22 produced in the core, radiolytically, is decreased.
23 DR. SHEWMON: Yes.
24 MR. CASS: And the amount -- while the vunt of 25 hydrogen produced in the core is the same. Of course, we
) !
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646
30335 0 151 BRT e', 1 are adding additional hydrogen.
~_
2 So we have to balance that when we get to the 3 recombiner, with an extra addition of an injection, a 4 gaseous injection of oxygen so we have the correct 5 proportions.
6 DR. SHEWMON: Can you inject air for that or do 7 you need to inject oxygen?
8 MR. CASS: I would say, I guess, you could 9 inject air. I'm trying to think of what the nitrogen would 10 do to the recombiner operation over a long period of time.
11 DR. SHEWMON: Common practice is to add pure 12 oxygen?
~
13 MR. CASS: The common practice is to add pure
()
\> 14 oxygen because that's a parallel to what has been done in 15 the past. I'm trying to think of what the nitrogen might .
16 do to the catalysts that are present in the recombiner.
17 Robin, can you help me?
18 MR. JONES: To keep the volumes similar to the 19 design of the off-gas system, it's preferable to use oxygen 20 because you are not carrying the other along.
21 MR. MICHELSON: I had two questions. The first 22 one deals with, could you tell me the size, roughly, and 23 location and routing and so forth of the hydrogen addition 24 system?
25 MR. CASS: Yes. There are a number -- first on (7
LJ l
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
30335.0 152 BRT
/~Y l the routing. I'll get to the size.
(/
2 The routings are really simple. We just inject 3 just upstream of the pump to assure the best --
4 MR. MICHELSON: Which pump? Upstream of the 5 pump?
6 MR. CASS: The feedwater pumps.
7 MR. MICHELSON: Main feedwater -- there are 8 several feedwater pumps, condensate pumps, booster -- are 9 you talk about upstream of the main feedwater?
10 MR. JONES: The typical point is the suction 11 side of the booster pump, because the pressure is low there.
12 That's a convenient spot.
13 MR. MICHELSON: That's one injection point.
n\~' 14 What is the --
15 MR. CASS: That's a typical injection point.
16 MR. MICHELSON: What does the apparatus consist 17 of at that injection point? Just bubbling it into the 18 stream or hew is it done?
19 MR. CASS: That's basically it. You basically 20 bubble into the stream. There is a hard line in any 21 permanent installation, it would be a small diameter hard 22 line that would be routed suitably through the reactor 23 building.
24 MR. MICHELSON: We aren' t dealing reactor 25 building yet; this is in the turbine building.
O ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
~30335.0 153 BRT
(~) 1 MR. CASS: Okay. Through the turbine building,
\_/
2 to the source of the hydrogen.
3 MR. MICHELSON: Yes.
4 MR. CASS: There are at least three basic ways 5 to inject the hydrogen. The first is to do it with gas 6 trucks.
7 MR. MICHELSON: Pressurized gas.
8 MR. CASS: The second is cryogenic tank; and the 9 third is an electrolytic.
10 MR. MICHELSON: Are most people considering 11 cryogenic?
12 MR. CASS: No. I think there's a variety. Most 13
' people are -- you see, the decision will be economic.
I)
' 14 MR. MICHELSON: .There's a little bit of safety 15 involved here, too, you can't take hydrogen too lightly 16 when you start talking about building boilers for it and so 17 forth.
18 MR. CASS: But the decision that is made as to 19 what kind of system is used will depend partly on the 20 amount of injection required and will also depend on 21 whether the utility -- depends on the layout of the site.
22 Is the site big enough to easily put a cryogenic tank in?
23 Is it easy to put an electrolytic system in? There are 24 electrolytic considerations, obvious differences in balance.
25 There are about 3/8 inch or half inch lines.
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 80 4 336-6646 ,
l
- _ _ _ _ . _ . _ _ _ , . _ _ _ _ . . . _ _ _ __ _ _ _ _ - . . . . _ _ _ _ _ _ , _ _i
30335.0 154 BRT je"S 1 DR. SHEWMON: . What do you do to -- atomize it is V
2 not the right word -- but it's a problem of not having the 3 stuff bubble on through. You want to get it dissolved.
4 MR. CASS: The injection point is chosen so you 5 are so close to the booster pump that we would get very 6 ef fective mixing.
7 DR. .SHEWMON: That still may not be dissolution.
8 What I'm concerned about is whether you try to make this 9 into small bubbles or --
10 MR. CASS: No. There is no mechanism beyond the 11 pump action to get it dissolved.
12 MR. JONES: The quantities are very small, and 13 the actual going into solution.is very fast indeed. We 14 have done some studies to see if there was any kind of 15 cavitation problem with the pumps and there isn't. It 16 dissolves almost instantaneously.
17 MR. CASS: In the initial programs, there were a 18 number of different injection locations evaluated, and a 19 number of different injection schemes evaluated. Such 20 considerations as pump cavitation and optimal dissolution 21 of the gases was considered and the result of that study 22 was the current mechanism.
23 MR. MICHELSON: What flow rate are we talking 24 about?
25 MR. CASS: It will vary from plant to plant. It O
ACE. FEDERAL REPORTERS, INC.
202 147 4 700 Nationwide Coserage 800 336-6646
30335.0 155 BRT I will vary from 6 to 40 or 50 standard cubic feet per minute.
(mv) 2 MR. MICHELSON: You are not going to go 40 or 50 3 with a 3/8 inch line, are you?
4 MR. CASS: Then it's a half inch line. It's not 5 a six-inch pipe. It's a small diameter pipe.
6 MR. MICHELSON: I realize it isn't a six-inch 7 pipe. 3/8 inch is a pretty small line. First of all, 8 nobody in their right mind would be a hydrogen line in that 9 small because the first time anybody steps on it, it would 10 be exciting. You want something physically sound, 11 generally you wouldn't go less than one inch, double extra 12 strong, or something, for this.
13 MR. CASS: Can I suggest here we are getting a 14 little ahead.
There's a place in this presentation where I 15 wanted to talk about the owners group guidelines. There 16 have been two reviews with other ACRS committees to 17 consider such things as fire safety and the safety --
18 MR. MICHELSON: You are going to cover that 19 later.
20 DR. SHEWMON: Fine. We'll wait. Go ahead.
21 MR. CASS: A place where we'll get to that.
22 MR. MICHELSON: I have a second question 23 unrelated to the first. The second is an area in which I 24 know little or nothing, but, looking at the Surry event and 25 so forth,'I kind of got the bnpression there was a problem O
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336 6646
3033520 156 BRT 1 with extremely low oxygen water?
2 MR. CASS: There's a section in this 3 presentation where I will specifically address it.
4 MR. MICHELSON: You are going to tell us why 5 this won' t start corroding carbon steel -- okay. Go ahead.
6 MR. CASS: Yes. There's a section in this 7 presentation where I will specifically say that.
8 MR. MICHELSON: Thank you.
9 MR. CASS: The Dresden 2 plant in the U.S. has 10 been operating for over two fuel cycles now on hydrogen 11 water chemistry.
12 (Slide.)
13 I just wanted to give you a little update on its 14 status. This chart is just a little out of date. The 15 Dresden 2 plant has completed its second cycle.
16 We did a very extensive plant materials test . s 17 program on it, ran a large number of constant extensive 18 rate tests, some with deliberate intermittent hydrogen.
19 The test times, all told, covered about 2000 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />. We also have about 2000 hours0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> of precision crack 21 growth rate testing, which I'll talk more about later, and 22 there have been periodic ultrasonic inspections of cracked 23 safe ends and recirculation pipe.
24 There has been a detailed fuel examination af ter.
25 one cycle with no deleterious effects observed and the fuel p
L.)
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6M6
30335.0 157 BRT
~
1 ' examination, af ter the second cycle is in progress. The (v ') -
2 initial results are very similar. That fuel exam is not 3 quite ccmplete, but so far, the fellows have not seen any 4 kind of deleterious effects.
5 LMR. MICHELSON: What concentration were they 6 using?
7 MR. CASS: In terms of hydrogen injection?
8 MR. MICHELSON: Yes.
9 MR. CASS: The injection at Dresden is about 1. 4 10 parts per million. That's the highest injection rate of 11 any plant that has been measured to date.
12 M R. JONES: You might want to add, Jeff, that 13 the core doesn't see 1.5 ppm O 14 MR. CASS: The reactor water hydrogen 15 concentrations are typically, well, for Dresden 2 it is 16 about, something like 150 ppb, 150 or 175 ppb.
17 DR. SHEWMON: When people look at the fuel 18 examination, do they actually -- is this all visual? Or do 19 they actually get to bend something?
20 MR. CASS: It is a combination of visual and 21 destructive evaluation.
22 DR. SHEWMON: Okay.
23 MR. CASS: There is a very careful look at 24 corrosion products, hydrogen pickup, crud scrapings and the 25 like.
O ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
I: ~i
.30335.0 158 l
~BRT 1 MR. MICHELSON: I~ guess you are inferring that
- (f-4 2 they actually analyze some of the materials to see if there ,
3 was a hydrogen pickup.
~4 MR. CASS: That's correct.
5 MR. MICHELSON: Apparently there was none 6 significant?
-7 MR..CASS: There's a normal amount of hydrogen f-pickup and the amount measured in these studies fits right 8
9 in with the data band for the normal water chemistry for 10 this kind of exposure.
11 As a result of this, I think you'll see from 12 charts I'll develop in a little bit here, there is a 13 significantly expanded application of hydrogen water O 14 _ ' chemistry in progress here and in Europe, in this country 15 and in Europe.
16 (Slide.)
17 This chart I'm showing you now is sort of a 18 checklist, . without going into great detail, of the J
19 materials testing that has been conducted both in the j 20 laboratory and in reactor.
21 Here I've listed the material: Sensitized
[ 22 stainless steel, Inconel 60, Inconel 182, carbon and low 23 alloy steel; high strength materials; and irradiated
- 24 stainless steels.
i 25 You can see that we have evaluated all of these i
(}
/\CE. FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6646
30335.0 159 BRT
(') I conditions in the laboratory. The irradiated stainless 2 steel tests were conducted in a hot cell with a quite 3 active gamma source.
4 We have also done quite a bit of testing of 5 these different materials in reactor.
6 We have observed beneficial effects for all of 7 these structural materials in every case that we have 8 examined so far.
9 MR. RODABAUGH : Does high strength materials 10 include bolt materials?
11 MR. CASS: The highest strength materials that 12 we have looked at are the austenitic steels, those are the
,_ s
. 13 highest strength steels that we use in the boiling water
( 1
~' 14 reactors, and those strengths exceed those of the pressure 15 vessel bolt materials.
16 Other bolting materials have been evaluated for 17 water chemistry. Inconel 750s are used for bolts in 18 boiling water reactors and those have also been evaluated.
19 DR. SHEWMON: G.E. has issued a suggestion about 20 a year ago or more about high, you call it, radiation 21 enhanced stress corrosion cracking or something like that?
22 MR. CASS: Yes.
23 DR. SHEWMON: Is it your continuing feeling that 24 the hydrogen water treatment reduces that, too?
25 MR. CASS: I would like to answer that question
/'N
)
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
-30335.0 160 BRT
(~' in two parts.
'd' 1 2 (Slide.)
-3 I have a detailed chart on that, just a little 4 bit higher.
5 (Slide.)
S Professor Shewmon, this is just one chart shead 7 if you are looking in the handout.
8 DR. SHEWMON: I don' t mean to get you out of 9 order.
10 MR. CASS: That's quite all right. That's. fine.
11 We conducted some constant extension rate test 12 on yery highly irradiated stainless steel to assess gamma 13 flux effects. And also to assess effects o,f electr5 chemical O 14 control.
15 We had these pieces of stainless steel cut out 16 of reactor components that were irradiated to 3 times 10 to 17 the 21. We tested them in 32 ppm oxygenated water, because 18 that probably is a match to -- probably an underestimate to 19 the fact, to the kind of ECPs one would find in core. We 20 compared that to tests in .2 ppm, and then tests were 50 ppb.
21 We also ran tests with and without high gamma 22 flux. Basically what we found is without the gamma flux 23 that we could steadily decrease the amount of IGSCC by 24 suppression of the oxygen.
25 You see here with 32 ppm oxygen we had almost O
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800 336-6M6
. . _ _ . _ . _ . ~ . _ _ _ _ _ _ ~ . - _ _ . _ . _ _ _ _ , _ _ _ . _ _ _ _ - _ _ . . , _ _ . . . . - _ . ,
3033530 161 BRT
/ 1 1 entirely intragranular fracture; with 200 it was reduced to 2 60 percent IGSCC, and then down to nothing when we got the 3 oxygen concentration low enough.
4 We then added a gamma flux source of around 10 5 to the 9th r per hour. That's a factor of somewhat around 6 50 less than is calculated in cora, but still a significant 7 gamma source; again on the highly irradiated base metal.
8 Here you can see that with 2 times 10 to the 6 9 r/ hour; I'm sorry, not to the 9th, it's 2 times 10 to the 6 10 r/ hour, 'we had 85 percent IGSCC. You can see the 11 importance of the gamma source, here without from 60 12 percent to 85 percent; and then we took out the oxygen,
,_., 13 reducing from 200 ppb to 50 ppb, and we eliminated the
( )
14 IGSCC.
15 So, in principle, here, if we can, by suitable 16 hydrogen injection, suppress the ECP of the core below a 17 quarter of a volt, which is what this corresponds to, below 18 this minus 230 or 250 millivolts, we think that we can 19 prevent the radiation assisted stress corrosion cracking.
20 We don't know, however, how much hydrogen will be required 21 to do that.
22 We presently, as you know, measure the ECPs in 23 the recirc loop. We don't measure it in core.
24 We have a special program at Nine Mile Point 25 which I will describe here in a few minutes, that is s - ;
1 ACE FEDERAL REPORTERS, INC.
202-147 3700 Nationwide Coserage 800-334 6646
. ~. . _ .. . _ . . . . _. ._. _
1 [ 3'0'335.0' ~ 1162= ,
'BRT
{
~
1 designed to measure the ECP in core and ~ also to assess T
2 directly, IGSCC performance of' highly irradiated material 3 in the core.
-4 So, in the laboratory it is promising. But we.
5 don't know how much hydrogen will be required to suppress ' '
We do have a program 'I 6 the ' ECP in core to the desired value.
7 :to address that.
8 -Now, on the Surry question, the next. couple of ,
f 9 charts here, I think -- I guess we can concentrate on this
- j. 10 chart here. I think that will help to address that. e i
11 (Slide.)
12 I think all of us are aware of the literature i 13 data which indicates that, at very low dissolved oxygen :
3 14'
~
concentrations, it is possible to have accelerated general I I
i x .
15 corrosion;. less than 10 ppb, and'certainly.in the 1 ppb, 16 which I am told was the-case as Surry, one would have t j 17 expected very high general corrosion conditions which, when 18 coupled with a design situation such as was the case at 19 Surry where there were two flow perturbations that were 20 very, very close together, would lead to erosion / corrosion.
L 21 We measured in the laboratory, the general ,
i' 22 corrosion rates in normal water chemistry and in hydrogen 23 water chemistry. Of course, initially, these oxygen ;
l 24 concentrations are specified to be between 20 and 50 ppb i
25 oxygen. That's a specific number. I'll come back to that
.i
!(1) i '
AcEJFEDERAL REPORTERS, lNC.
l 202 347 3700 Nationwide Coverage 800-336-6M6 !
30335.0 163 BRT
(~T 1 in just a second because that becomes a very important LJ 2 number.
3 But, initially we had a fairly high rate of 4 corrosion and, as with all of these kinds of processes, the 5 general corrosion rate then flattened out to give, in 6 general, fairly moderate rates.
7 There is no question that, if in the feedwater 8 system we allowed the oxygen concentration to decrease to 1 9 ppb, we would have the potential for high general corrosion.
10 However, we have made the oxygen specification for the 11 feedwater system exactly the same as it is for normal water 12 chemistry. In normal water chemistry it's between 20 and 13 50 ppb, and in hydrogen water chemistry it's between 20 and .
14 50 ppb.
i-15 DR. SHEUMON: I'm sorry, what is the hydrogen
-16 content?
17 MR. CASS: If a plant has a very tight system 18 and the oxygen concentrat' ion falls below the specification, 19 then it is required to inject oxygen into the feedwater 20 system. People sometimes get confused, you are injecting 21 hydrogen, you are injecting oxygen. Let me point out, by 22 comparison the amount of radiolytic gases produced in the 1 1
23 core are about 100 standard cubic feet per minute and of l
24 course a third of that, roughly a third of that or so would l 25 be oxygen.
O ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-334 6646 j l
i 30335.0 164 BRT ,
l l
l If we had no oxygen in the feedwater system and
(~')
v 1
2 we had to inject enough to get the full 50 ppb, we would be 3 injecting .1 cubic feet per minute. So the amount of 4 oxygen that would possibly be required to maintain the 5 specification is negligible in comparison to the amount of 6 oxygen that is produced radiolytically that we have to 7 suppress.
8 DR. SHEWMON: Could you tell me where, in going 9 from the condenser to the reactor, that you have carbon 10 steel and where you have stainless steel?
11 MR. CASS: Yes. I think I can. I'll probably 12 need some help from Dr. Jones on this.
13 In the reactor vessel in general, inside the O -
14 reactor vessel, that i s ', it is all stainless steel. Of -
15 course, the pressure vessel is low alloy steel, but it is 16 clad with stainless steel weld metal.
17 The recirculation pipe, reactor water cleanup 18 pipe, of all stainless steel.
19 The feedwater pipe is carbon steel.
20 The main steam, which is two phased -- so far we 21 have only been talking about one-phase, single-phase 22 because that's the case with Surry -- the feedwater pipe,
- 23 one, single-phase system is carbon steel. The main steam 24 pipe is carbon steel.
25 Then, beyond that, when we get into the balance O
i ace FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 800-336-6 4 6
30335.0 165
- BRT (O 1 of the plant, beyond the nuclear steam supply system the
%J 2 amount 1 that's stainless steel and carbon steel will vary 3 from plant to plant.
4 But I think the main point I'm trying to address 5 here is that for the feedwater system, single-phased system 6 in the NSSS, we have made sure that we wi.11 not have -- and 7 by the way, this specification was developed long before 8 Surry, it was developed three or four years ago -- we had 9 made sure that we will not have an erosion / corrosion 10 situation.
11 In the main steam we still measure significant 12 amounts of oxygen. In normal water chemistry we typically
. 13 would measure maybe seven or so ppm, and in hydrogen water 14 chemistry we measure one or two ppm, so there's still a 15 significant amount of oxygen in the main steam. And I 16 really don't think that we could get into a Surry situation 17 specifically by the addition of hydrogen water chemistry.
18 I do think, though, that it is not impossible --
19 in fact, we have looked through our failure analysis 20 history, to see are there examples of erosion / corrosion-type 21 failures and, of course, we have found some in the two-phased 22 systems.
23 Everyone that has been found that I'm aware of 24 is associated with a design situation where -- associated 25 with a flow perturbation, something like Surry where there O
i ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646 e
30335.0 166 BRT
. /'s 1 were two flow perturbation elements that were pretty close V
2 together.
3 DR. SHEWMON: You are talking. steam lines?
4 MR. CASS: When I'm talking about oxygen 5 addition, I'm talking about oxygen addition to the feedwater.
6 That's carbon steel.
7 DR. SHEWMON: When you are talking about 8 Two-phase --
9 MR. CASS: When I'm talking about the steam 10 system, steam pipe, that's, of course two-phase.
11 MR. MICHELSON: That's a single phase, it's all 12 steam, all gas phase.
, 13 Two-phase means you've got both liquid and gas. ,
\'-)
14 MR. CASS: Fine. Fine.
15 MR. JONES: In a BWR you have a lot of wet steam.
16 MR. MICHELSON: That's not generally thought of 17 as two-phase --
18 DR. SHEWMON: The droplets are what wears the 19 holes more than the --
20 MR. MICHELSON: That 's possible ; yes .
21 MR. CASS: In the steam system, we still have 22 under hydrogen water chemistry at least a ppm of oxygen 23 coming off.
24 MR. MICHELSON: The oxygen injection is right 25 beside the hydrogen injection, I guess? Do you bubble both O
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 800 136-6646
Y,' y ,
'~~30335.0 - ,l 167 ;
= BRT : V;s 1 ,
j .,
.s 'p/
i 1 in:at the~same, time, then?
pndfofvarying.theflbwrates?
-2 MR. CASS: Yes. That's right.
x '3 DR. ' SHEWM0id . But\ tg '4 oxygen- is inj ected farther ' ;
/ y, $. ' A
- 4 back closer to,the condenser, isn't it? ,
o; e gq 5 MR. JONES: Itdouldbeinjectedatanyofthe-[4 '
6 low pressure points of the system. You need to get it.into j,
the region where the temperatures are below abouti300 F.
7
.\ .,
8 MR.IMICHELSON: You have ' to go ba'ck further than ,
t ,
9 the booster pump then, probably.
,j j i t
10 DR.:SHEWMON: Let me come back to the- steam line.
11- You talked -- you have had problem in the steam lines, but 12 these are where you have perturbations? Or whatever you 13 called . the.n? '
. s l t 14 ,
MR. CASS: There have been examples of -
, I.
.15 erosion / corrosion in BWRs. They have been Ebdurved in 16 condenser systems. I think there has been '6ne in a steant ~,, t;
\;
17 line. What I'm trying to explain to y u is each one of .'f j 18 those was associated with a flow perturbation.
(
19 MR. MICHELSON: Do you think BWRs are missing i
\., ,
20 perturbations in the steam ines? 'j \
21 MR.,CASS Each of the erosichs t .,
we have found ,
pq 3
(
22 were associated with flow perturbations and there's been a i
23 recommendation made to every utility by EPRI, and GE has ,
- s.
24 participated in some of these evaluations, to study the }*$
25 plant layouts to pick out design situations that might lead O x: ,
' t ,' G ACE FEDERAL REPORTERS, INC.' },
202 347-3700 Nationwide Coverage 800 336-6646' /
('
y . .
A.+*
1
!~ l 30335.0 168 l I
BRT s 8 ,j.
-f] 1 to eroe. ion / corrosion. And I suspect that some will be v
m, t2 found.
3 .DR. SHEWMON: I guess I'm bothered by your p ,
J
- s. 4 calling it ", erosion / corrosion," because I have bumped into
'5 ,
that for a single-phase flow such as the Surry. You are Lx
.(
( 6 h using it under conditions where, presumably, you have 7 t'wo-phase erosion.
I would think, there, any corner would s ,
be,a potential for water -- an'y elbow.
~
8 Yet if you call
\
w 9 that a " perturbation," then systems inevitably -- they must 10 have corners.
i 11 MR. CASS: I think that the studies that are "j .
g 12 going on now to try to list 'and identify each of the cases 13 that are in any way associated with wall thinning of carbon O, :[. 14 steels will shed seine light on whether these are general r,
15 corrosion type or whethed these are erosion / corrosion type 16 failures, e
17 There is a conference that has been scheduled in
, 18 mid-April to review some of these things. An EPRI -
- , 19 sponsored conference.
4 20 DR. SHEWMON: Who is running theqconference?
21 M R. CASS: EPRI.
12 MR. MICHELSON: I thought I derived some comfort k 23 y I'm a statement,'but maybe I misunderstood. I thought you s ,
24 " said on the steam side of the PWR you were up to one or two So
[ 25 parts per million and, the re f ore , it is a nonproblem.
10 .
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coscrage 800-336-6M6
I!l 30335.0 169 BRT
() I why are we worrying about it on the steam side?
2 MR.' JONES: 'Can I make a couple of comments? In 3 a single-phase water. system, there's a certain amount of 4 oxygen that~you need dissolved to suppress IGSCC at neutral y, 5 pHs. When;you get to wet steam systems tne oxygen tends to 6 partition. You have almost oxygen-free liquid phase in 7 those systems. That's why you still get erosion / corrosion 8 problems in boiling water r'eactors, even though the steam 9 normally has ppm quantities of oxygen in it. The water l l
10 doesn' t --
11 MR. MICHELSON: It separates out that quickly?
f 12 MR. JONESr.'Yes.
13 MR. ETHERINGTON: In addition to effects on the
' *(
r 14 blading, have you looked -- well, I'm sure you have looked ,
15 at -- can you tell us something about what possible effect
? 16 it, the hydrogen, might have on the disk problems that your 17 competitor is having and you are having a little, in the 18 turbines . mean?
19 MR. CASS: Right. There have been studies of 20 stress corrosion evaluations conducted in laboratories on 21 the disk materials with and without hydrogen water 22 chemistry. You know, I guess I may be more f amiliar with 23 the boiling water reactor systems than I am with the 24 pressurized water reactor systems. There's a crevice and a 25 notch associated with the boiling water reactor disk O L ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
30335.0 170 BRT
()
v 1 problems.
2 The studies that have been conducted that I'm 3 aware of clearly show that the hydrogen water chemistry 4 helps the situation.
5 MR. JONES: A point you might want to make, Jeff, 6 is that the change on the steam side that we are talking 7 about is really quite minor. You go from a somewhat 8 oxidizing condition to another somewhat oxidizing condition.
9 We don' t really expect to see any major perturbations.
10 MR. CASS: We would really not expect to see a 11 major effect on the turbine situation or in the behavior of 12 the main steam piping. The place where we would expect to 13 see it is in the vessel and in the piping, stainless steel p
14 ' piping and the reactor internals and, of course, tlhe f'eedwater 15 pump, the place where we looked. \ '
16 DR. SHEWMON: When you inject this hydrogen, if 17 I got an answer, I've forgotten it already, you are not 18 injecting it into stainless steel piping?
19 MR. JONES: That's right. Carbon steel.
20 DR. SHEUMON: It's into carbon steel.
21 Now pretty soon after that it must convert to 22 stainless steel pipings, doesn't it?
23 MR. JONES: No, carbon steel all the way up 24 through the feedwater system.
25 DR. SHEWMON: It's only then after you get into O
v i
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
30335.0 171 BRT
- 1 the vessel that you get this 10
- 1 dilution?
w' 2 MR. CASS: Right. Rig ht .
3 MR. JONES: Just to close on the steam 4 erosion / corrosion question, problems attributed to that 5 phenomenon have been seen in all kinds of turbines, wet 6 steam, piping, extraction line, that sort of thing. We 7 can' t expect the hydrogen water chemistry is really going 8 to change that situation. The utilities have got 9 inspection programs in place now to examine that piping and, 10 where necessary, replace it. And replace it with materials 11 that are much more. resistant that are now being used and 12 are now available.
,, 13 .D R . SHEWMON: These are basically low alloy
-(')
14 steel with chrome bearings? Half chrome?
'15 MR. JONES: Half chrome, 1 chrome, is typical.
16 MR. CASS: I mentioned to you that we now have 17 quite a bit of experience with a number of different plants.
18 (Slide.)
19 This table is sort of a brief summary of the 20 plants that have been studied just an update here on Hatch, 21 the test is ongoing. We are just about done. We will be 22 finished in just a couple of days and I can read in for you 23 what some of these values are.
24 These are the hydrogen water chemistry 25 parameters where we have achieved less than minus a quarter
)
(^'J L
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M6
P 30335.0 172 BRT l of a volt ECP in the research system. We see that the
(
2 feedwater hydrogen varies from around .3 ppm to 1.4. I'll 3 come back to that in a minute.
4 The resultant oxygen varies from as low as it 5 can be measured, 2 ppb , to 15 to 20 ppb. The recirc 6 hydrogens vary from 30 to 200 or so ppb. And the effects 7 on the main steam radiation increase also vary quite a bit 8 from very small values to quite significant values.
9 DR. SHEUMON: Where is it you measure your 10 suppression potential?
11 MR. CASS: We measure the suppression potential 12 in the recirc loop at the present time.
13 DR. SHEWMON: So after it has gone through the 7.s 14 vessel.once and come out and been recirculated?
15 MR. CASS: That's exactly right.
16 Now, we are also -- we also have a test program 17 that I'll describe to you here shortly where we will be 18 measuring the ECPs and also the materials performance in 19 the core bypass water, directly. We have an in-core --
20 DR. SHEWMON: Obviously, a substantial 21 difference in the ef fect of construction on plant, in the 22 amount of hydrogen you need to put in; is that right?
23 MR. CASS: Yes. These differences in feedwater 24 hydrogen are a reflection of the size of the core. It is 25 the power of the match -- the catalysts for their reaction;
<x C
ACE FEDERAL REPORTERS, INC. I 202-347 3700 Nationwide Coserage 800-336-6646
30335.0 173 BRT g 1 the residence time of the downcomer, width of the downcomer, 2 those kind of considerations.
3 It's due to plant design considerations. It's 4 possible to look at the power of the core and to l
5 rationalize these numbers quite nicely. They all fit.
6 DR. SHEWMON: Dresden 1 can't be --
7 MR. MICHELSON: That's kind of small by i l
8 comparison with Peach Bottom. l l
9 DR. SHEWMON: It's one of the old ones. l l
10 MR. CASS: I'm sorry, I couldn't hear what you 11 said.
12 DR. SHEWMON: I'm sorry, I'm mumbling. Peach 13 Bottom 3 is a bigger plant?
7_),
\
14 MR. CASS: That's right. But Peach Bottom 3 is 15 also a copper plant, which interferes somewhat with the 16 process.
17 MR. MICHELSON: You mean copper in the condenser?
18 MR. CASS: Yes. That's right.
19 Aside from the copper plants, it's possible to 20 make all of these plants fit on a curve that is dependent 21 on the considerations I mentioned to you, the power of the 22 core, core density, residence time in the downcomer and the 23 width of the downcomer.
24 DR. SHEWMON: So if you wanted to design a plant 25 or have a plant where you could have a minimum amount of f3 0
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
f 30335.0 174 BRT radiation in the steam, then you want --
(( ) 1 2 MR. CASS: You want a small, hot core. And an 3 appropriate residence time.
4 That's what these plants. are. They are high 5 density, high power density.
6 DR. SHEWMON: So when you talk about newer 7 plants which are likely to be bigger plants, that means you 8 are likely to have more turbine shine; is that right?
9 MR. CASS: In general that's probably true.
10 There's also plant-specific considerations 11 associated with exactly the way the turbine is shielded in 12 its current configuration at that entry. But, in general, 73 13 you are right.' In general, you ar'e right.
14 DR. SHEWMON: Does that mean also that people 15 have gone to -- driven by Appendix K to have lower heats 16 per linear foot, that they are also going to have the 17 higher steam radiation if they put hydrogen chemistry in? .
18 MR. CASS: I don't know. I don't know how to 19 answer that.
20 DR. SHEWMON: It's going to be lower specific 21 power , I think.
22 MR. MICHELSON: You were saying you were going 23 to tell us about the Hatch numbers. Did I miss it?
24 MR. CASS: I beg your parden. Thank you.
25 The Hatch numbers are, le t's see , I'll take it O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
30335.0 175 l BRT
(~')
I this way here.
2 Main steam radiation increase is close to 4, a 3 little less than 4.
4 MR. MICHELSON: 4X?
5 MR. CASS: Yes, a little less than 4X.
6 The recirc hydrogen is about, oh, 170 ppb, 180 ppb.
7 The recirc oxygen is about zero.
C And the feedwater hydrogen is 1.3 or 1.4 ppm.
9 Those numbers may change as time goes on because 10 the plant has a significant copper inventory. It has 20 ppb 11 copper in the recirc as a result of the Admiralty brass 12 condenser.
13 As I will explain in some more detail later on, 0[
14 the hydrogen injection program,had the additional benefit l
15 of suppressing the general corrosion, suppressing, 16 apparently, the general corrosion of the Admiralty brass in 17 the condenser. The copper went way down and the soluble 18 copper in the recire, also, went way down.
19 MR. MICHELSON: What-is Brunswick using for 20 condenser tubes?
21 MR. CASS: Brunswick is not a copper plant.
22 MR. MICHELSON: No, I wouldn't think so.
23 MR. CASS: Brunswick is not a copper plant.
24 MR. JONES: I think Brunswick retooled with 25 titanium.
. }
l ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
r-230335.0 176
- BRT--
(} l -
MR. MICHELSON: But Peach Bottom is a copper 2 plant?
3 MR. CASS: .Yes. Peach Bottom and Hatch and also 4 Oskarshamm 1 in Sweden are copper plants..
5 One would hope that as the -- if these plants 6 continue to run on hydrogen water chemistry, gradually the 7 amount of copper in the system will decrease, it will 8 become cleaned out and we'll have a little more effective 9 use of our hydrogen and so the amount of hydrogen required 10 will decrease.
11 Of course, the main steam radiatior. will also 12 decrease.
13 ( Slide . )
() -
14 But I think that's an important side benefit for 15 you gentlemen t'o consider l'n terms of fuel performance.
16 The decrease in copper could turn out to be very beneficial 17 for fuel performance.
18 MR. MICHELSON: You mean because of the fluff in 19 the fuel?
20 MR. CASS: The so-called "CILC failures."
21 MR. MICHELSON: Which?
22 MR. CASS: Copper-induced localized corrosion --
23 I'm sorry, crud-induced localized corrosion attributed to 24 copper in the water.
25 Here's some data from these different plants. I O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
30335.0^ 177 BRT ,
1 l
plotted for you recirc oxygen versus feedwater hydrogen.
1
-( }-
2 Just to show you that there is quite a bit of variability 2
3 from plant to plant and the response, again, a reflection 4 of the plant design considerations we just discussed.
5 (Slide.)
6 This --
7 DR. SHEWMON: Did the oxygen you plotted there 8 correlate well with the electrochemical potential you used 9 other times?
, 10 MR. CASS: Yes. As we have gained experience 11 with, in particular, the reference electrodes, our 12 electrochemical measurements have gotten better and better.
13 One thing we rely on .as a crosscheck is the use of platinum
(:) 14 as a reversible reference.
15 MR. MICHELSON: This oxygen must be injected, 16 then, at two points: One in the feedwater and one just 17 before the recombiner?
18 -
MR. JONES: There aren't any U.S. plants that 19 are injecting into the recirc point. They have sufficient 20 leakage --
21 MR. CASS: But I think there definitely will be.
22 There are a number of plants that you'll see in a minute 23 that will be coming on to determining hydrogen water 24 chemistry in the next few months, and a number of those are 25' plants that have very tight condensors, and they will have O
ACE FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-6646
)
L30335.0 178
, -BRT
[v'] f
.2 MR. KASSNER: Do you plan, in the near future, 3 to report on the' geometric configurations that you 8
4 mentioned with regard to the downcomer and all that power 5 ,
level as to what the correlation is?
6 MR. CASS: Yes. Bob Cowlin, I'm not going to 7 present it in here --
8 MR. KASTENBERG: I wondered if you were going to
! 9 report at sometime?
10 MR. CASS: Bob Cowlin -- a number of people have -
11 -different, but obviously related, kinds of correlations and 12 I believe Bob Cowlin will be wanting to report on that.
13 -
DR. SHEWMON: -
Earlier you had talked about 1.4 O. -
i 14 bringi~ng Dresden 2 down to the same electrochemical 15 potential as the others. Yet here the oxygen potential is ,
16 an order of magnitude higher.
17 MR. CASS: Let me just explain here, the numbers s
18 that I quoted to you are the oxygen concentrations where 19 the ECP is less than minus a quarter of a volt.
20 In some cases the plots of ECP versus feedwater 21 hydrogen are such that you have gradual decrease and then a 22 very, very steep decrease. So you might have, for example, 23 here, Dresden 2 operating at -- they have, it turns out 24 Dresden 2 has a fairly gradual change, gradual and smooth ,
25 change of ECP versus injection. So they will operate at, o
i ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
30335.0 179 BRT 1 say, minus 2850 or so. On the other hand, Peach Bottom and
. n.f'))
2 Fitzpatrick and a number of the other plants have this 3 gradual increase followed by an very steep decline. So 4 they will operate at minus 330 or, in the case of Nuclenor, 5 even 400. So it's really not an apples to apples 6 presentation.
7 These are the main steam line results, 8 responding to the oxygen results I showed you earlier.
9 We have conducted a number of constant extension J
10 rate tests.
11 (Slide.)
12 Would-you like me to explain -- review what a 13 constant extension rate test is? Are you famili'ar with it?
O 14 MR. MICHELSON: That's all right. I don' t need 15 to get educated on it --
16 MR. CASS: I'll do it anyway. I know Paul is 17 familiar with it. I know Paul and 'iom are. A constant 18 rate extension test is where we take a small sample of very 19 heavily sensitized stainless steel and slowly stretch it so 20 we put about 20 percent strain in over a seven-day period.
21 MR. MICHELSON: At what temperature?
22 MR. CASS: At the operation temperature. This 23 is done in a sample line taken from the recire group, done 24 at the plant at 288 centigrade.
25 W'e will examine the little sample af ter it has O
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-66:6
30335.0 ~180 BRT
-()- 1 this 20 percent strain or so applied, to see if there's any 2 intergranular cracking.
3 These are the results of the in-plant tests that 4 we have conducted to date.
5 Here I've plotted for you average crack 6 propagation rate, that's just the depth of the deepest 7 crack found divided by the time on test versus the measured 8 electrochemical potential.
9 Ue do see this minus 230 millivolt border here.
10 These are the two recent Hatch results that I 11 described to you earlier. It is a pretty steep curve.
12 The steepness of this curve may be a reflection 7s 13 of the fact that we put 20 percent strain in over a one-week d 14 time period. If we had a slower applied strain, I would 15 expect that this would have a more gradual transition.
16 But this does give us a lot of confidence that 17 our minus quarter of a millivolt is a pretty good number.
18 Yes?
19 MR. JONES: Just leave that up for a minute.
20 The important point to notice is that all the points along 21 the axis there showed no crackings. Plotted them at that 22 point --
23 MR. CASS: I plotted them at 10 to the minus 8, 24 but when we examined them in 250x in the microscope we 25 can't find any cracks at all. Thank you, tO V
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
-30335.0 181
-BRT 1 (Slide.)
2 Another way to express'that is to show ECP here 4
3 and the oxygen here, this just illustrates it's the recirc-4 oxygen that's not the controlling parameter, it's the ECP 5 that is the controlling' parameter.
6 (Slide.)
7 A number of laboratories have conducted 8 systematic studies of coolant impurity ef fects. Argonne, 9 Dr. Kassner and Dr. Shack and his colleagues and a number 10 of other laboratories have conducted studies of the effects 11 of coolant impurities.
12 The bottom line of these studies is it's
-. 13 essential to control not only the electrochemical potential, ,
14 but the coolant conductivity. There is no question that 15 the benefit of hydrogen water chemistry will be iost if 16 very strict control is not maintained overs the coolant 17 conductivity.
18 Part of the study was an evaluation of what 19 irapurities might be present and what the sources of those 20 impurities might be.
21 I have listed them here, air in-leakage for the 22 CO2 and CO3; resin decomposition for sulf ates. Silica is a 23 precursor. We also have some nitrate coming through from 24 those processes.
25 Of course, from radiolysis we have peroxide.
O l
ACE FEDERAL REPORTERS, INC.
f 202-347-3700 Nationwide Coverage 800-336-6646
,_ _ __ - . _ . _ ~ _ _ - _ _ . . . _ . _ , _ . - - _ _ - __ __
~
130335.0 182 BRT
- ) j_
~
1 The.others, I think, are fairly obvious here; the brass
, 2 condensors or for copper and oils and lubricants can give 3 organics.
4 (Slide.)
5 Most of these impurities are deleterious. As 6 you can see here there's an ongoing study. I'm just 7 reporting to you in a simplified form for our results to 8 date. With the exception of carbonate and nitrate, most of 9 these are, in fact, deleterious.
10 The most important ones of these impurities are 11 probably the copper and the sulfates from resin fines.
12 MR. ETHERINGTON: Is that H 203? In the resin;
- f-( 13 H 2SO3?
- O 14- MR. CASS: There 's a range.
15 DR. SHEWMON: H2SO3 is one, then?
16 MR. CASS: Yes.
17 MR. MICHELSON: The resin fines are coming from 18 the main line as well as the reactor cleanup?
19 MR. CASS: Primarily reactor water cleanup.
20 MR. MICHELSON: Why primarily reactor water 21 cleanup?
22 MR. CASS: Because the history is such that the 23 operational practices lead to problems with that particular 24 resin system. The other is a little easier to maintain.
25 MR. MICHELSON: The full flow resin bed is doing O
l 4 ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 i
30335.0 183 BRT
() 1 all right you are saying, but the cleanup resin bed isn' t?
2 MR. CASS: Yes. That, of course, will vary from 3 plant to plant. I'm just giving you a very broad overview 4 of an awful lot of plant data.
5 MR. MICHELSON: What.does the hydrogen do to the 6 reactor water cleanup system? Any effects there?
7 MR. CASS: It will protect it.
8 DR. SHEWMON: Protect it from what?
9 MR. CASS: Its ECP will be below -- it will be 10 essentially the same as the recire.
11 MR. JONES: You asked me about the operation of 12 the resins --
, 13 MR. MICHELSON: I'm also interested, does it 14 interfere with the operation of the resins at all?
15 MR. CASS: No. It thought you meant would it 16 protect it. It won't have anything to do with th'e --
17 MR. MICHELSON: The water being circulated, 18 though, in that system is the recirc waters in the 150 ppb 19 kind of water; right?
20 MR. CASS: Yes.
21 MR. JONES: Yes.
22 MR. MICHELSON: So the hydrogen concentrations 23 are quite low in that system?
24 MR. JONES: Yes.
25 MR. MICHELSON: Are they high enough to protect O
ACE FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-6646
30335.0 184' BRT.
() l it? Dealing with liquid phase now, not with gas phase?
2 MR. JONES: The cleanup system in most plants is 3 stainless steel.
4 MR. MICHELSON: Well, that depends on what you 5 mean-by "most plants." What plants do you think are 6 stainless?
7 MR. CASS: The cleanup systems in all systems in j 8 all plants that I'm aware of on the vessel side of the 9 first isolation valve are all stainless steel.
10 MR. MICHELSON: Oh, yes, on the vessel side. We 11 are talking about the cleanup system itself which is where 12 this water is circulating through. Is this hydrogen t
- 13 . helping to protect it from stress corrosion cracking? Is 14 there enough? Is the concentration enough to do any good 15 in.the liquid phase?
16 MR. JONES: If you are asking about stress 17 corrosion cracking, the answer is yes. If you are asking 18 about. erosion / corrosion --
19 MR. MICHELSON: Mostly stress corrosion cracking.
20 There is some carbon steel over there also.
21 MR. CASS: In terms of stress corrosion cracking, 22 there would be enough to protect at this time.
23 MR. MICHELSON: Okay.
24 MR. JONES: Enough in the Swedish -- in the 25 Asaya design plant, the answer would be no because their ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6 4 6
30335.0 185 BRT l(f 1 ' cleanup system comes off the top of the core wh'ich is much ,
2 more oxidized a location.
3 MR. MICHELSON: Typically the cleanup system for 4 BWRs in this country, BWR-3, -4 series, where does it come 5 off from?
6 MR. JONES: Off the recirc system.
7 MR. MICHELSON: Also returns to the feedwater 8 line; doesn't it? Or doesn't it?
9 MR. JONES: bo you know the answer to that?
10 MR. CASS: Can you repeat the question, please7 11' MR. MICHELSON: Where does the reactor water 12 cleanup return, typically, in a boiling water reactor?
- 13 MR. CASS: I guess I don't have that with m'e. I 14 can't answer that.
15 MR. MICHELSON: My vague recollection is it 16 returns to the feedwater line.
17 MR. CASS: No, I don't think that's right. I --
18 MR. MICHELSON: You think it returns back to a 19 recirc loop?
20 MR. CASS: No. I guess I can get that 21 information for you, but I don't have it.
22 MR. MICHELSON: It comes off the recirc loop as 23 I recall, but it returns, I thought, to a feedwater line, 24 but I'm not sure. I was hoping you'd clarify it.
25 If it comes back to the feedwater, of course,
(
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-66:6
30335.0 186 BRT.
'i 1 then -- well, it's probably a small fraction.
(V 2 MR. CASS: There's a set of water chemistry 3 guidelines that have been prepared or that are, rather, in 4 the process of being prepared for the BWR owners, and this 5 is sort of a status. This document is not complets yet.
6 There is a committee that meets periodically to develop the 7 document. It will be issued in about the middle of 1987 8 and some of these numbers here that I have up on this chart 9 here will be fine-tuned so as to be in line with all of the 10 available data at the time. But I just wanted to give you 11 a representation of the kinds of things that are being 12 considered.
13 First of all, the guidelines incorporate all of Cs) the control parameter values that we have for normal water 14 ,
15 chemistry guidelines. There is a set of normal water 16 chemistry guidelines that has been issued.
17 In addition, and specifically in the normal 18 water chemistry guidelines the objective is to have a 19 utility respond if the conductivity of the plant exceeds 20 about .3 of a microsiemen. There are actions that must be 21 taken and there are various times allowed before action 22 must be taken and they go all the way from diagnostics to 23 shutting the plant down depending on what the value is of
?4 the conductivity.
25 In addition to that conductivity control, there O
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
39335.9 187 BRT
() 1 is an electrochemical control that is being considered that 2 is being considered as a main part of this guideline.
3 That's a. reflection of a minus a quarter of a volt that I 4 mentioned to you earlier and there are various actions that 5 must be taken to, if this minus a quarter of a volt is 6 exceeded.
7 These numbers, I'm sure, will change before the 8 document is issued, but what I'm trying to get to you -- to 9 explain to you is that electrochemical control is being 10 added to the conductivity control.
11 It is also required to conduct a confirmatory 12 test program. It can be either periodic constant extension
' 13 rate testing or it can be on-line crack growth and I'll
-14 explain that in a little bit.
15 The on-line crack growth has advantages over the 16 constant extension rate approach because it is continuous 17 and because it is a precision measurement and because it 18 provides a hard documented record that IGSCC has been 19 stopped over a long period of time and I'll get into that 20 in just a few charts from now.
21 DR. SHEWMON: Is normal chemistry guidelines the 22 owners group guidelines?
23 MR. JONES: Yes. It's being produced by the 24 same committee.
25 MR. CASS: The normal water chemistry is issued, O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6 4 6
30335.0 188 BRT
- 1 and this one will be issued, in a few months.
2 DR. SHEWMON: What percentage of the operating 3 BWRs in the country have committed to the BWOG owners group 4 chemistry; do you know?
5 MR. JONES: The current version of the 6 guidelines all of the members of the owners group are 7 committed to them. It's 100 percent.
8 DR. SHEWMON: Of the owners group. But I asked 9 aboitt all the operating plants.
10 MR. JONES: I think there are maybe four or five 11 plants who have not been. I don't know the status of their 12 commitment.
s 13 DR. SHEWMON: If I were to ask the man sitting
\
]
14 next to you after the meeting, he probably could identify 15 who those four or five plants were.
16 MR. CASS: I think it is important that we take 17 into consideration the very dramatic improvement in 18 conductivity performance by all of the BWRs in the last two 19 years, in particular in the last year. There has been a 20 very vigorous campaign to get the message of the importance 21 of conductivity control to the utility executives.
22 The plant managers and the utility vice 23 presidents.
24 DR. SHEWMON: That's everybody? Or that's the 25 BWOG owners groups?
%/
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
30335.0 189 BRT j} l MR. CASS: Everybody. -Everybody.
2 There are statistics that we keep for the plants 3 on conductivity and performance. I'm just give you some 4 representative numbers. These are not exact numbers.
5 A few years ago, less than 25 percent of the 6 plants would meet the current guidelines. Today, greater 7 than 80 percent meet the current guidelines, based on 8 performance over:the last 12 months.
9 The plants that, at present, do not meet those 10 guidelines, it's interesting that a number of them had 11 hydrogen water chemistry minitests and, as a result of 12 those hydrogen water chemistry minitests have had some work 13 done to, at least over the time p,eriod, one month or so or
'O 14 two months of the minitest, get the conductivity down 15 because otherwise the minitest won't work.
16 In general they have carried on to maintain that 17 good water chemistry from the minitest to the subsequent 18 operation.
19 So there has been a very dramatic improvement in 20 plant conductivity quality.
21 HR. MICHELSON: Not being that familiar with 22 what goes on as far as cleaning up the water, what does go 23 on? How do they improve it with present equipment and so 24 forth? Just attention to detail? Or do they have to spend 25 more money on resins or something? What is it?
O ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
- - - - . . _ _ _ _ __ .- - ~ . , _ , _ _ _ _
30335.0 190 BRT
/ 1 MR. JONES: Actually it turns out you spend less 2 money on resins, so the water is cleaner.
3 MR. MICHELSON: What do they do?
4 MR. CASS: The most pervasive item is a 5 condensor leak. And what is done, what has been done in --
6 the most dramatic cases is to have a very vigorous program 7 of finding and immediately fixing condensor leaks.
8 MR. MICHELSON: I see. So it is possible by 9 just care in operation to get the water chemistry well in 10 control.
11 MR. CASS: Yes. It definitely is.
12 MR. MICHELSON: If I were able to keep my water
,- 13 chemistry in good control, apparently I still need the
)
14 hydrogen? -
15 MR. CASS: Yes, you do.
16 MR. MICHELSON: Why?
17 MR. CASS: Because you cannot prevent IGSCC just 18 by conductivity control. IGSCC will occur in water of 19 essentially theoretical purity with normal water chemistry.
20 The electrochemical power of the water, electrochemical 21 potential of the water is sufficient to produce IGSCC.
22 If we look back in the pipe crack records and 23 look at what plants have experienced pipe cracking and 24 compare that to their water chemistry, we can see that some 25 of the very best plants in the world that had life time N)
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
'30335.0 191 i BRTE
/~'i 1 conductivities approaching a tenth of a microsiemen per V,
2 . centimeter, experienced significant amounts of pipe crack.
3 So it's really not possible.to prevent IGSCC unless you 4 suppress the ECP and the conductivity both.
5 DR. SHEWMON: What plants were those that you 6 would put in that class?
7 MR. CASS: KKM, Cooper, I would put in that 8 plant.
9 DR. SHEWMON: KKM7 10 MR. CASS: A Swiss plant. Cooper is a plant --
11 DR. SHEWMON: Both of these had normal 304, not' 12 304 L?
13 MR. CASS: That's correct. That's correct.
()
, _s 14 Kayorso in Italy is another plant that had just 15 terrific coolant conductivity performance and they had some 16 pipe-cracks.
17 There's a number of elements, in addition to the 18 conductivity control, we have the 19 electrochemistry control, we have the checking with the 20 materials test, and we also have diagnostics of the water 21 chemistry specifically dissolved oxygen and hydrogen 22 measurements.
23 MR. MICHELSON: Those diagnostics will have to 24 be installed in the recirculation systems?
25 MR. CASS: Yes, that's correct.
i.
(~
(J ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 i
'30335.0 192 BRT
[)
N/
1 MR. MICHELSON: Is that pretty tricky business?
2 MR. CASS: No. They are typically Orbisphere, 3 typically Orbisphere hydrogen and oxygen meters.
4 MR. MICHELSON: Unfortunately I don't know what 5 that is.
6 MR. CASS: Just a small standard off-the-shelf 7 device.
8 MR. MICHELSON: Where you take a side stream 9 from the recirc loop, pass it through the instrument?
10 MR. CASS: That's right.
11 MR. MICHELSON: How big an instrument are we 12 thinking of?
,- s _ 13 MR. CASS: They'will typically be a half a liter
'~
14 a minute kind of side stream.
15 There are, typically, existing sample lines that 16 will, from the recirc loop, that will take five or 10 17 liters a minute.
18 MR. MICHELSON: Must be a little pump with this 19 apparatus or just a natural delta P somewhere?
20 MR. CASS: Natural delta P. Five liters a 21 minute and then a side stream of a half liter.
22 MR. MICHELSON: That's an off-the-shelf device 23 at these temperatures and pressures?
24 MR. CASS: That's correct. Well, excuse me, the 25 half liter a minute has to be cooled to 25 degrees ,
(
l
. ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 l
39335.9 193 BRT 1 centigrade.
(~)T u
2 MR. MICHELSON: You have a little cooler in the 3 system, just a little bit of claptrap with it.
4 MR. CASS: It is precision,s you know, precision 5 equipment.
6 MR. MICHELSON: Oh, yes, I'm sure.
7 DR. SHEWMON: By the time you get done paying 8 for it you are convinced it is.
9 MR. MICHELSON: Very precise.
10 Let me pursue the instrument just a wee bit 11 further.
12 MR. CASS: Sure. Sure.
13 MR. MICHELSON: Let's speculate, I guess you are (2) 14
~
only going to put in one.such instrument or are you going 15 to put two such instruments in?
16 MR. CASS: There are guidelines that have been 17 presented to ACRS committees and to the Regulatory 18 Commission for that kind of instrumentation and there are 19 duplicate instruments for everything.
20 There's a prime and a backup.
21 MR. MICHELSON: What is the significance of, for 22 whatever reason, failure of control devices feeding the 23 hydrogen or whatever.
24 What is the significance of getting too much 25 hydrogen or too much oxygen in for short periods of time?
O ACE FEDERAL REPORTER 1', INC.
202-347-3700 Nationwide Coverage 800-336-6646
- 30335.9 194- '
BRT
(} l MR. CASS: In terms of the materials performance-2 it would be -- I guess we are not aware of any. You have 3 .to remember that a pressurized water reactor operates at
- 4 couple of ppm. Couple of ppm in.the recirc or what the
. 5 equivalent of the recirc would be and here we are talking 6 about a couple of hundred ppb.
7 So the consequence of having a least an order of 8 magnitude difference --
9 MR. MICHELSON: It would make no difference.
10 Okay.
11 MR. CASS: I mentioned the conductivity.
12 ( Slide .')
13 The conductivity requirements -- there is, in O 14 addition, special focus on certain impurities which are 15 particufarly aggressive.
16 , . Sulfates, the Committee-is considering a i 17 limitation of around 20 ppb, because above that there is an 18 effect. 20 or 25, some number in that vicinity. There's 19 going.to be a special limit. Of course, there are existing 20 special limits for chlorides.
21 I would suspect that one more information about 22 plants with Admiralty brass condensors is developed, there 23 may be some guidelines for those plants in terms of copper.
24 We talked about the feedwater dissolved oxygen ,
25 issue. We don't have to go back over that.
O I
i ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
, . _ _ _ _ = _ . _ _ _ ~ , - _ , - , . - - .. . _ _ ____ _ -_-_____--_.. - . , _ _
30335.0 195 BRT'
?-
- .'" m 1 Here is my little discussion of the amount of v[^)
2 radiolytic-gas, 100 standard cubic feet per minute versus .1, 3 standard cubic feet per minute that would be required to 4 maintain 50 ppb oxygen in the feedwater.
5 So it's a spec that has a lot of aspects to it 6 and has received an awful lot of consideration.
7 (Slide.)
8 This next chart will give you an idea of the 9 extent of implementation. There are a lot of plants, and 10 I've listed them throughout the world here, that have 11 minitests completed or planned.
12 There are a large number that have completed and f
13 really quite a few that will be conducted in the near t:) .
14 future. Chinshan is the plant in Taiwan.
15 DR. SHEWMON: Is that the one that had the fire 16 from the hydrogen leak not long ago?
17 MR. MICHELSON: Yes.
18 -
MR. CASS: If it is, I'm not aware of it.
19 DR. SHEWMON: Your lawyers know about it.
20 (Laughter.)
21 MR. CASS: For plants with permanent high 22 general water chemistry; Dresden 2 is under way, Nuclenor 23 in Spain, four Swedish plants, and one Japanese heavy water 24 plant that are presently operating on hydrogen water 25 chemistry.
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6 4 6
'39335.0 196
-BRT~
s.
3
/~') 1 MR. MICHELSON: How long have they been using it?
.X s.
2 MR. CASS: Two fuel cycles for Dresden 2, abcut 3 six months for Nuclenor; up to two fuel cycles for these 4 Swedish plants. Really, from six months up to two fuel 5 cycles.
6 And the one Japanese plant, one fuel cycle.
,7 Here, where I listed plants that are committed it I'm counting them as committed if they have the injectio.'
9 systems in the process of being installed. I'm not 10 counting -- if some utility executive says I'm going to put 11 it on, they are not on this list. It's only when they have 12 written purchase orders and bought equipment that they are p 13 xon this list.,
t) 14 Yes, sir?
15 DR. SHEWMON: The Japanese plant that has 16 already~~got it on --
17 MR. CASS: It's Fuega. A heavy water plant.
18 Not a BWR.
19 DR. SHEWMON: Fine. And this other one over 20 here is a BWR?
21 MR. CASS: .Chinshan is a boiling water reactor.
22 DR. SHEWMON: No, the top one, you have one 23 Japanese plant.
24 MR. CASS: Same plant. They completed a is 25 minitest and also are under way permanently.
t ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
m'ki 33335.9 ., 197 BR1F ,
j f~'y. 1- DR. SHEWMON: I'm sorry. I didn't realize they v-2 overlap.
3 MR. CASS: I want you to realize.in the.next few 4 months here, there will be a significant addition to the 5 number of plants that are permanently operating a hydrogen 6 water chemistry. In the next few months, all these fellows, 7 many of these plants are currently in refueling outages and 8 when they come up to power after their refueling outage, I 9 they will be on hydrogen water chemistry.
10 (Slide.)
11 Just some housekeeping here with this chart.
12 (Slide.)
13 There was a meeting on October 18 with ACRS
'O 14 Subcommittee on fire protection and radiological effects 15 for the purpose of information exchange on hydrogen water 16 chemistry.
17 The topics discussed included the minitest 18 results; the impact on main steam line radiation levels; 19 hydrogen leak prevention and detection; siting 20 considerations for hydrogen and oxygen storage and the 21 off-gas steam impact. Many of the subjects that we 22 discussed today were specifically oriented towards those 23 discussions, whereas this talk is specifically oriented 24 towards IGSCC prevention.
25 In the view of the industry people who attended O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
r 13s335.s 198 BRT..
l the meeting, the information was favorably received by the
({][ .
2 ' subcommittees. There was a general consensus that in terms 3 of ALARA,. hydrogen water chemistry is effective and that 4 the hydrogen storage and distribution concerns appear to be 5 adequately addressed by these guidelines. ,
6~ They will be considered in the preparation of 7 the. appropriate NRC documentation.
8 (Slide.)
- 9 I would like to change the subject just a little
-10 bit.- I talked about some of the monitoring. I just want 4
11 to take a few minutes and then I'll close out here -- to 12 talk to you about some in-plant monitoring systems that we 13 have.
'O 14 This is a schematic that describes some of the 15 three different kinds of monitoring systems that we have 16 developed. They are based on reversing D.C. concepts, 17 where we can, with great. precision, tell if and how fast a 18 crack is growing.
19 I refer to the one-pipe system, that is where we
-:2 0 take a sample, five liters a minute or so from the recirc 21 line. We also have a system that is similar that is i
22 installed in core. These systems feature fracture I
23 mechanics specimens and electrochemical sensors.
! 24 Finally, there is an additional application 25 where we can directly monitor a cracked component, such as :
i 1 1E)
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
. - - ,-.--_ __ _ _ -,._. - ~._ . , . _ , _ . . . . _ . _ . , _ . _ , _ . . _ . _ _ . _ , _ _ _ _ , . . . . , , , . .. . ..
30335.0 199 BRT
/~N 1 a pipe with a weld overlay on it.
kJ 2 I'm going to describe a little bit the 3 performance of the system described here in blue. I'm 4 . going to show you some data from Peach Bottom. These are 5 the kind of systems I referred to when I mentioned 6 precision crack growth. These are the kind of systems that 7 are being put to demonstrate that the system is providing 8 crack arrest.
9 (Slide . )
10 This crack arrest verification system has an
- 11. IGSCC precracked, and autoclave system attached to the 12 recirc loop, it provides continuous crack growth monitoring
- 13. and can verify crack growth predictions. If there's a O 14 transient in the water chemistry, and it's reasonable to 15 expect them to occur, what's the ef f ect o'f those transients.
16 So it can demonstrate the control _o,f cracking by water 17 chemistry.
18 (Slide.)
19 Here's a sample. You can see the platinum leads 20 attached for the precision monitoring.
21 (Slide.)
22 Here's a picture of a typical system that is 23 attached to the sample line in the recirc loop. There is 24 an autoclave here, it's a 1.5 gallon autoclave. Typically 25 we might have three specimens, one heavily sr.isitized 3C4; O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
-30335.0 200 BRT f'T 1 one Inconel 182; one Inconel 600.
V 2 MR. MICHELSON: Where are you going to put that 3 inside containment? I assume it goes inside containment?
4 MR. CASS: Well, the sample line is taken 5 through the dry well.
6 MR. MICHELSON: You are going to come outside of 7 containment then?
8 MR. CASS: You know we take the -- depends on 9 what you mean by " containment."
10 MR. MICHELSON: You know what containment is, 11 primary containment.
12 MR. CASS: It is in the reactor building, but f_s 13 it's through the dry well.
(J 14 MR. MICHELSON:
What kind of isolation is on 15 that line?
16 MR. CASS: There are double isolation valves on 17 either side.
18 MR. MICHELSON: What si2e line is it?
19 MR. CASS: Half-inch lines. Right here.
20 MR. MICHELSON: Do they have excess flow checks 21 in it?
22 MR. CASS: Yes. Yes. It is fire resistant --
23 there's a computer that's associated with it.
24 (Slide.)
25 There's fire resistant electrical lines that are O !
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 201 LBRT
.r"T 1 used with it.
V 2 This is Laura Harris who works in my group.
3 She's the engineer who built the system for Fitzpatrick.
4 There's an on-line PC to develop the data.
5 'The computer can be located remote from the 6 system or right by the system, as the. utility desires.
7 MR. KASSNER: One quick question. How do you 8 control the flow that goes through the system? In the 9 sense if you are taking water off the recirc line to the 10 half-inch lines going to the system, where do you dump the 11 water since there's a pressure differential --
12 MR. CASS: We dump it to the low pressure side
., 13 of the RWCU. You need about 100. psi, at long as you have
( )
14 100 psi delta P, that's a typical dump.
15 MR. KASSNER: So that's how it is controlled.
16 Thank you.
17 MR. CASS: Some data.
18 (Slide.)
19 This is data from Peach Bottom, Peach Bottom 3 20 here, Peach Bottom 2 here.
21 This is change in crack length of the fracture 22 mechanics specimen. Do you see the precision, that's 5 23 mils. If we blow the scale up we'll get very fine as you 24 can see in just a second.
25 This is time here, of course. We have about i
/~T l V
i i
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 I
30335.0 202 BRT
{) 1 7000 hours0.081 days <br />1.944 hours <br />0.0116 weeks <br />0.00266 months <br />' worth of data I'm showing you.
2 What you can see is that we can relate the crack 3 growth in these heavily sensitized fracture mechanics 4 specimens to the plant conductivity.
5 For Peach Bottom 3, during the initial part of 6 the startup, the conductivity was fairly high and ,so the 7 crack growth rates are correspondingly high.
8 l Later on conductivity got under control and the f
9 I crack growth rates flattened out.
10 l There are a couple of places, one here and one 11 l spectacular one here, where there were resin intrusions and 12 the crack growth rates increased. ,
s , 13 Peach Bottom 3 over this 7000-hour time period 14 had an average conductivity of about .3 of a microsiemen.
15 There were eight transients exceeding 1 microslemen per 16 centimeter.
17 Towards the end of this time period the 18 conductivity is excellent and there are no transients and 19 the crack growth rates are really quite good.
20 DR. SHEWMON: The largest step was with a 21 transient in the resin?
22 MR. CASS: Yes, sir.
23 DR. SHEWMON: Interesting.
24 MR. CASS: Here Peach Bottom 2, average 25 conductivity I only have about 1000 hour0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br />; of data, but I l
l ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M6
'30335.0 203 ,
BRT
(~3 1 think you can see the effect anyway.
v.
2 MR. MICHELSON: How long a duration was the 3 resin intrusion before it was cleaned up?
4 MR. CASS: .This one is about 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />.
5 MR. MICHELSON: And its effect was reflected for 6 hundreds of hours thereafter?
7 MR. CASS: Yes, it will. Because it takes time, 8 there's a lag time from the time that the intrusion starts 9 until you see the effect at the crack tip. You know, it's 10 a liquid diffusion type consideration. And, of course, 11 there's a lag time before you see the benefit of the 12 cleanup.
gS 13 These lag times have been observed also at ,
\')
14 Argonne National Labs and other places.
15 I was saying here you can see the benefit of the 16 improved conductivity, here .15, no transients, crack 17 growth rates are quite a bit less.
18 (Slide.)
19 Here is not the most spectacular transient, but 20 still one that was significant to show you how well we can 21 monitor with this device.
22 Again note the resolution here is a mil here.
23 We had two resin intrusions here, fairly closely spaced at 24 2500 hours0.0289 days <br />0.694 hours <br />0.00413 weeks <br />9.5125e-4 months <br />, crack growth rates double.
25 MR. MICHELSON: How big was this resin intrusion ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 .
30335.0 204 BRT 1 how extensive?
2 MR. CASS: About 5 microsiemen per centimeter; 3 on that order. It's not necessarily --
4 MR. MICHELSON: Was this just a small hole in 5 the bed? What caused the intrusion?
6 MR. CASS: It had to do with the details of how 7 the resins, the treatment of the septa whether the resin 8 beds were changed and there was a detail that when the 9 resin beds were changed allowed some resin fines to sneak 10 through backflow and get into the system and that procedure 11 was changed in the last part of this -- that was identified 12 as a problem and changed here. And after that everything 13 was fine. But it took them a little bit of time to figure
('- ') .
14 out what was going on and how'to fix it.
15 DR. SHEWMON: And they have committed on Peach 16 Bottom 3, which is the one which has had more problems; is 17 that it?
18 MR. CASS: Both plants. Both plants. Both 19 plants will be operated on hydrogen water chemistry. Unit 20 2 is starting up on hydrogen water chemistry in May when it 21 comes out of its outage; and Unit 3 will be having an 22 outage in the fall and they will also be on hydrogen water 23 chemistry.
24 There are a number of plants, and I've listed 25 them here, that have adopted this crack arrest verification p
V ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 800-336-6646
'39335.9 205
'BRT
(} l system and will have the capability for this kind of 2 precision monitoring.
< 3 _Both the Peach Bottom plants, Pilgrim, 4 Fitzpatrick, Monticello and the Nine Mile Point,- and there 5 may be more as the time goes.on.
6 Those are most of the plants that_will be 7 operated on hydrogen water chemistry permanently.
8 (Slide.)
9 I mentioned a few times-this in-core program. I 10 -don't want to go too-far beyond my time limit. We have 11 this in-core test program. The objective is to measure how 12 much benefit would be' gained by the highly irradiated 13 internals as a result of hydrogen injection,-to determine
. O_ 14 if hydrogen water chemistry can suppress the corrosion 15' potential.to the desired value.
16 It's a program that is jointly funded by EPRI 17 and ESSERCO, which is sort of a New York State EPRI and 18 being performed at Nine Mile Point.
19 We have in-core precision crack growth of those 20 fracture mechanics, specialized version of them designed to 21 fit in an in-core tube and ECP electodes, again in an in-core 22 tube.
23 The probe was inserted in the spring of 1986.
24 We have lots of normal water chemistry data and the 9
25 hydrogen water chemistry program is in progress now. We O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
139335.9' 206 tBRT' 1 expect that. hydrogen will be added in April.
]
2 (Slide.) ,
3 It will be.very interesting to us to see if we 4 have established' crack growth rates of a specific nature in
~
5 normal water chemistry. It will be very interesting to see 6 if we can stop that crack growth in hydrogen water 7 chemistry.
8 This is just a schematic of the system. In the 9 core here, Nine' Mile Point has some spare LPM slots and we 10 have put to in-core tubes with electrochemical sensors and 11 these crack growth specimens in as well. Then, of course, 12 we have the recirc pipe sample to complement.
13 DR. SHEWMON: What is a LPRM?
- ~14 MR. CASS: Low power cange monitor, an in-core 15 instrument.
16 MR. KASSNER: What kind of reference electrodes 17 are you using?
18 MR. CASS: Very specially hardened silver 19 chlorides.
20 MR. KASSNER: Okay.
21 MR. CASS: I'm sure Maurice has talked to you 22 about this. Of course, in hydrogen water chemistry, we are 23 going to rely on the platinum.
24 I only have two more charts here.
. 25 MR. SHEWMON: I only see one.
ACE FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-6646
.30335.0 207 BRT l
/~N 1 MR. CASS: There's an additional one on copper x_/ i 2 that I thought I'd show because Tom expressed some interest 3 in the Hatch results. I will give you hard copy of it as 4 soon as -I can find it.
5 (S lide . )
6 At Hatch, we did succeed in reducing the total 7 copper by greater than 50 percent. In the recirc system, 8 the hot well data shows that the source term was reduced 9 even further. The change in insoluble copper was 10 relatively small.
11 These are similar results to that found at 12 Oskarshamm. To date the OCG, the utility that operates
_ 13 Oskarshamm, has not been s,uccessful in producing the
~
14 desired ECPs. Perhaps they haven't waited long enough or 15 injected enough hydrogen. As I mentioned, we did succeed 16 getting good extension rate test at Hatch and this 17 reduction in copper could lead to better fuel performance.
18 DR. SHEWMON: Does the copper catalyze the 19 removal of hydrogen in some way?
20 MR. CASS: Well, yes, I think it interferes with 21 the effective suppression of the radiolytic products.
22 DR. SHEWMON: Okay.
23 MR. CASS: The last chart is a conclusion chart 24 of what we think the status is now.
25 (Slide.)
A V
Acu-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30335.0 208 BRT 1 It's pretty clear to us that IGSCC is suppressed q
J')N 2 when both the electrochemical potential and conductivity 3 are. suitably controlled. It is necessary to suppress all 4 of the oxidizing power of the water, not just the oxygen, 5 it's the ECP that is the' controlling parameter.
6 As I mentioned, the specification for the 7 feedwater is unchanged, specifically to address any 8 potential for accelerated general corrosion. We need some 9 oxygen in that feedwater system.
10 The in-plant materials monitoring technology 11 that I described to yoa really offers the potential for 12 improved plant operation.
13 Just to put that in perspective, one thing I CE) 14 would like to point out.to you is that the Dresden fuel 15 plant over its two fuel cycle history has operated about 90 16 percent of the time with the oxygen -- with the l'/ conductivity in specification; and a little less than 80 18 percent of the time with the oxygen in specification.
19 DR. SHEWMON: That's after they started the 20 hydrogen water treatment program?
21 MR. CASS: Yes. Yes. In other words, they had 22 an oxygen specification of 15 or 20 ppb and a conductivity 23 specification of .3. They met the conductivity goal 90 24 percent of the time and they met the oxygen goal a little 25 less than 80 percent of the tin.e.
I O
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6M6
30335.0 209 BRT
/^ l Over this two fuel cycle time period, they V}
2 certainly haven't discovered any new cracks. The 3 monitoring that we have done, the 2000 hours0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> _of crack 4 extension rate testing and 2000 hours0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> of this precision 5 crack growth, where the prototype-for this crack-growth was 6 done, showed no IGSCC.
7 So there certainly seems to be some tolerance
-8 for being off the specification without having a measurable 9 effect on the IGSCC performance of the stainless steels.
10 I doubt that we will be able to really quantify 11 how many hours in a given period of time are allowable. We 12 collect some of the data from these crack arres.t 13 ' verification systems that are,being installed at the plants.
I,>s\
\ 14- It might be 10 percent, that might be a safe number. But I 15 think we need some more data to get a precise number.
16 MR. KASSNER: With regard to the hydrogen water 17 chemistry implementation, for example, Commonwealth Edison 18 has a lot of experience with Dresden 2.
19 I was wondering, it doesn't appear on your chart, 20 but a utility like that might decide to put it on Dresden 3, 21 without going through the CAV system and all this other 22 stuff. Have you had any discussions with them whether to 23 go that way?
24 MR. CASS: We have had discussions with them 25 with respect to Quad 2. I think Dresden 3 has had a pipe
! (_)
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6M6
- - - . . _ - _ , _ , _ . . . . ~ _ , . _ _ , - - _ _ _ - ..- ,, .. .. , _ ._ .-- .. - , , . , _ , - _ , , . . , .
30335.0 210 BRT
,'" l replacement program. The Quad plants have not. So I think s
\
2 in terms of IGSCC protection -- prevention by water 3 chemistry control, I don' t know this for a fact, but 4 perhaps Commonwealth Edison's focus might be first on the 5 Quad plants and then on Dresden.
6 MR. KASSNER: I see.
7 MR. CASS: We know Commonwealth Edison is on 8 this path for the Quad plants. I haven't listed them 9 because, at least to my knowledge they are not currently 10 installing. They may be installing soon.
11 We are in discussions with Commonwealth Edison 12 on these monitoring systems, however, specifically for the 13 ' Quad plants. They will probably make a decision in that 14 regard in one or two months.
15 DR. SHEWMON: Thank you very much. It has been 16 quite informative and you obviously have been making 17 progress in the last couple of years.
18 Gentlemen, why don't we stretch. That's the end 19 of the meeting then, for today, and the end of what we'll 20 have for you.
21 (Whereupon, at 3:45 p.m., the hearing was 22 concluded.)
23 24 25 t'
L)%
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6646
l CERTIFICATE OF OFFICIAL REPORTER rm
.iv)
This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the matter of:
NAME OF PROCEEDING: ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON METAL COMPONENTS, DOCKET NO.:
PLACE: . WASHINGTON, D. C.
O- -
V DATE: THURSDAY,' MARCII 26, 1987 were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission.
/
(sig (TYP )
JOEL BREITNER Official Reporter ACE-FEDERAL REPORTERS INC.
Reporter's Affiliation, O
N.i .
f])l ,
Technical Presentation
-to the ACRS Subcommittee on Metal Components March 26, 1987 l
O Leak-Before-Break Applied to Balance-of-Plant Piping at BVPS-2 Duquesne Light Company Stone & Webster Engineering Corporation Robert L. Cloud Associates, Inc.
Electric Power Research Institute O
=
/~;
- V WHIPJET PROGRAM HIGHLIGHTS
- 1) JUSTIFIES EXEMPTION FROM GOC-4 -- DEGB
- 2) FOLLOWS Gul0ELINES OF NUREG 1061
- 3) USES DETERMINISTIC FRACTURE MECHANICS ANALYSIS LEAK BEFORE BREAK-METHODOLOGY- 4
- 4) CRITICAL CRACK SIZE / ASSOCIATED LEAKAGE 10ENTIFIED
- 5) SAFETY MARGIN OF 10 APPLIED TO LEAK OETECTION CAPABILITY (0.5 GPM)
- 6) ONLY AUSTENITIC-PIPE INSIDE CONTAINMENT CHOSEN (304; 316)
- 7) LINE SIZE 6" T014" QUALIFIED FOR RESTRAINT EXCLUSION 8)= NO CHANGE IN LOCA - ENVIRONMENTAL QUALIFICATION - ECCS REQUIREMENTS Ou/ 9)
USED INDUSTRY TEST DATA FOR FRACTURE TOUGHNESS 4
- 10) ADDRESSED FATIGUE, CORROSION, WATER HAMMER, EROSION, CREEP, AGING, FLOW STRATIFICATION
- 11) STABLE CRACK GROWTH ASSUME 0 O
.1 BV-2 WHIPJET 'l SEQUENCE OF EVENTS l
- p l v -
9/16/85 INFORMED NRC 0F INTENDED PROGRAM
_9/24/85 PROGRAM DESCRIBED T0 ACRS SUBCOMMITTEE 11/1/85 STATUS PRESENTATION TO ACRS FULL' COMMITTEE 3/86 -!9/86 SEVEN (7) REVIEW MEETIN35 IN BETHESDA 10/23/86 ' SUBMITTED PROGRAM REPORT 1/16/87 REVIEW MEETING IN BETHESDA 2/2/87 SUBMITTED REVISED FINAL REPORT 2/13/87 SUPPLIED ADDITIONAL INFORMATION TO NRC 3/16/87 SER ISSUED - SUPPLEMENT NO. 4 O.
O-
.c , j LEAK DETECTION INSIDE CONTAINMENT i
l O-LEAK DETECTION SYSTEMS (REGULATORY GUIDE 1.45) i
- DIVERSE & REDUNDANT.
- SEISMIC CATEGORY I
- CLASS lE
- CONTROL ROOM ALARM /
INDICATORS
.i l
LEAKAGE EXCEEDS DETECTABLE RATE OF O - 0.5 GrM i
YES n
i CONTAINMENT ENTRY AND INSPECTION f-f :
0.I GPM- s.
VISUALLY DETECTABLE 4
4:
I BENEFITS OF LEAK BEFORE BREAKf(LBB)
- SIMPLER ENGINEERING AND DESIGN PROCESS
- REDUCED EFFECTS ON BUILDING DESIGNS
- DIRECT COST REDUCTION
- 1. REDUCED ENGINEERING EFFORT -
INCLUDING HAZARDS EVALUATION
- 2. FEWER PROTECTION DEVICES 27 VS 67-
- 3. FEWER COMPLEX INTERMEDIATE STRUCTURES
() ! BASED UPON DEVICE REDUCTION
- 4. REDUCED CONSTRUCTION EFFORT
- 5. MARGINAL NET CONSTRUCTION COST REDUCTION ,
FOR BVPS-2
- INDIRECT COST REDUCTION
- 1. REDUCED OBSTRUCTIONS
- 2. FACILITATION OF ISI
- 3. IMPROVED ALARA EXPOSURE FOR NORMAL OPERATION AND MAINTENANCE
[
- 4. ELIMINATION OF SPURIOUS MECHANICAL INTERFERENCES v
e
! 5. ENGINEER AGAINST CAUSATIVE FACTORS
, () BETTER THAN ASSUMING FAILURE
J.J . .
BV-2 WHIPJET PROGRAM 2
BENEFITS J-1
- 1) 'PROVIDED TECHNICAL DATA IN SUPPORT OF NUREG 1061 IMPLEMENTATION FOR ALL
- i. HIGH ENERGY LINES I
- 2) JUSTIFIED THE. ELIMINATION OF 32 RESTRAINTS AND 8 JET SHIELOS IN BV-2 J) 3)' PLANTS WITH LONGER CONSTRUCTION LEAD TIME COULD EXPECT LARGER BENEFIT POTENT 1.^L FOR APPLICATION TO CARBON STEEL' LINES OUTSIDE CONTAINMENT l 4) t i
O
O- BVPS-2 HARDWARE REQUIREMENTS Pipe Rupture Restraints and Jet Shields-BALANCE-OF-PLANT (SRP 3.6.1 CRITERIA) 27 (13.1%)
40 (19.4%)
BALANCE-OF-PLANT hlNATED BY WHIPJET) 127 (e1.7%)
ARBITRARY INTERMEDIATE 12 (5.8%)
PRIMARY LOOP-O 1
~ ^
bl
[
'- HIGH ENERGY SYSTEt1S PIPE STRESS t1ATERIAL ANALYSIS REVIEW BREAK TRANSIENT ANALYSIS REVIEW TARGET VfBRATION REVIEW REVIEW SHUTDOWN CORROSION REVIEW REVIEW CRACK GROWTH ECCS REVIEW ANALYSIS
- LUHIPJET LEAKAGE RATE PROGRAM ANALYSIS LEAK v RESTRAINTS i
& SHIELDS DETECTION O
Pipe Rupture Reulew
%-, .--.m. - . - ._ . _ - 7-...., _ ._., .-..,,,.,.__.__.-___,-r ,,,,,.__,,._,,,,_,,--w -.----c-,,- ,- - - - - - - - ,
Lino 3 SRP 3.6.2 Requiring 6-- (Except atbitrary
. Protection ..
intermediate breaks) 1 r
< Target and Interaction Evaluations
- mr Screening:-
Industry Experience, -
Economics Leak Detectability Margin Material Highest Stress :
Properties ; ;
Locations (Normal +SSE)
With Minimum Material Properties 1 r Fatigue Crack l . Growth Evaluation 1 .O
, m r >
l Limiting Detectable Leak Rate Leak Rate x Margin ? Calculations : Normal Loads 1
l l 3 r l
Margin on m Crack Stability 2 Normal +SSE ' l l Crack Size "
Evaluat!io.n Loads i
l 1 r Stability Check For Excessively 0 1.414(Nor+SSE)
High Loads Loads -
O l
( ) WHIPJET PROCESS
SUMMARY
OVERVIEW INITIAL INTERMEDIATE FINAL WHIPJET SYSTEMS SCREENING EXEMPTION SCOPE 1
t
- 1. ASS -[ NO UNACCEPTABLE INTERACTIONS
- 2. BDG ----------------------- CARBON STEEL, INACCESSIBILITY FOR DETECTION ,
i
- 3. BRS -[ 1H) UNACCEPTABLE INTERACTIONS ;
I i I
- 4. CHS -[ NO UNACCEPTABLE INTERACTIONS I 1 .
- 5. DGS -[ NO UNACCEPTABLE INTERACTIONS .
( ) l
- 6. FWE -[ NO UNACCEPTABLE INTERACTIONS I
l
- 7. FWS -----------------------[ INDUSTRY EXPERIENCE l (THERMAL, FATIGUE, WATER HAMMER)
- 8. GNS-[NOUNACCEPTABLEINTkRACTIONS l
- 9. MSS -----------------------[ CARBON STEEL l '
-[8", 14"
- 10. RCS I
-[ 10", 12" l
1 1 . RH S - - - - - - - - -- --- - - - -- - - - - -- - - - - - - - - - - - - - - - - --
i l 6", 12"
- 12. SIS -------------------------------------------[
l l 6
DUE TO CONCERNS ABOUT FATIGUE CRACKING RELATED TO l SOCKET WELDED FITTINGS
{ )
1 l
PIPING ELIMINATED FROM WHIPJET SCOPE BY SCREENING OR OTHER CONSIDERATIONS ,
l l
SYSTEM RUPTURE PIPE RETAINED REASON FOR LOCATION SIZE HARDWARE -
ELIMINATION (IN)
FWS CONTAINMENT 16 5 INDUSTRY EXPERIENCE RCS CONTAINMENT 1.5, 2 2 INDUSTRY EXPERIENCE MSS CONTAINMENT 32 9 ECONOMICS MSS OUTSIDE 32 3 ECONOMICS CONTAINMENT BDG TUNNEL - 3 5 INACCECSIBILITY OUTSIDE O. CONTAINMENT BDG CABLE VAULT - 3 3 LEAK DETECTION &
OUTSIDE ECONOMICS CONTAINMENT TOTAL 27 1
l C:)
i 1
1
- - L_-.
I
( FINAL PIPING SYSTEMS FOR WHIPJET ANALYSIS i
t MATERIAL BREAKS HARDWARE PIPING PIPE -------------
(IN) (1) (2) :
l' E-SA 376, 20 7 5 -
SIS 6 TYPE 316 6 6 0 8 -SA 376, RCS TYPE 304 i
n SA 376, 1 0 (3) 0 j RSS 10 .
TYPE 316 }
l.
O TYPE 316 i
NOTES:
PRR = Pipe rupture restraint
( ) ((1) 2) JIS = Jet impingement shield (3) Break requires SIS restraint .
O. O O-i I betIPJET PROGRfWI PIPIMS DATR RRTERIAL TEPP PRESS INSULRTIONe Tiet LOCRTION 2 LINE MuleER 00 MRL1.
2515-006-012-1 6.625 0.718 SR376 TYPE 316 595 2327 REFLECTIVE 3.5 CS206 2515-006-269-1 6.625 0.718 SR376 TYPE 316 105 2327 BARE O.0 CS206 ;
2515-006-015-1 6.625 0.718 SR376 TYPE 316 545 2327 REFLECTIVE 3.5 CS204 I
2515-006-270-1 6.625 0.718 SR376 TYPE 316 105 2327 BARE O.0 CS206 25I5-006-016-1 6.625 0.718 SR376 TYPE 316 595 2327 REFLECTIVE 3.5 CS202 2515-006-271-1 6.625 0.718 SR376 TYPE 316 105 2327 BRRE O.0 CS206 2515-006-026-1 6.625 0.718 SR376 TYPE 316 613 2327 REFLECTIVE 3.5 CS206
! 25I5-006-268-1 6.625 0.718 SR376 TYPE 316 105 2327 BRRE.- 0.0 CS206 25I5-006-029-1 6.625 0.718 SR376 TYPE 316 613 2327 , REFLECTIVE 3.5 CS209
! 105 BARE O.0 CS206 2S15-006-266-1 6.625 0.718 SR376 TYPE 316 2327 2 SIS-OO6-025-1 6.625 0.718 SR376 TYPE 316 613 2327 REFLECTIVE 3.5 CS202 25I5-006-267-1 6.625 0.718 SR376 TYPE 316 105 2327 BRRE O.0 CS206 2RCS-OOS-420-1 9.625 0.906 SR376 TYPE 309 613 2327 REFLECTIVE 3.3 CS206 2RCS-006-021-1 8.G25 0.906 SR376 TYPE 309 59L 2327 REFLECTIVE 3.3 CS206 l
2RCS-OOS-090-1 8.625 0.906 SR376 TYPE 304 613 2327 REFLECTIVE 3.3 CS204 J
2RCS-OOS-041-1 8.625 0.906 SR376 TYPE 309 545 2327 REFLECTIVE 3.3 CS209 i
2RCS-006-060-1 8.625 0.906 SR376 TYPE 309 613 2327 REFLECTIVE 3.3 CS202 l 2327 REFLECTIVE 3.3 CS202 j 2RCS-OOS-061-1 8.625 0.906 SR376 TYPE 309 595 2RHS-010-023-1 10.750 1.125 SR376 TYPE 316 105 2327 REFLECTIVE 3.7 CS202
' 2327 REFLECTIVE 3.7 CS202 2RHS-010-024-1 10.750 1.125 SR376 TYPE 316 105 l
12.750" 1.312ee SR376 TYPE 316 613 2327 REFLECTIVE 3.7 CS206 l
2RHS-012-001-1 REFLECTIVE 3.7 CS206 2RHS-012-056-1 12.750 1.312 SR376 TYPE 316 613 2327 l
i 12.750 1.312ee SR376 TYPE 316 595 2327 REFLECTIVE 3.7 CS206 2S15-012-299-1 REFLECTIVE 3.7 CS206 2515-012-067-1 12.750 1.312 SR376 TYPE 316 105 2327 12.750 1.312 SR376 TYPE 316 105 680 ArutE O.0 CS206 2515-012-066-2 680 SARE O.0 CS102 2 SIS-012-250-2 12.750 0.375 SR376 TYPE 309 105 1.312e* SR376 TYPE 316 545 2327 REFLECTIVE 3. 7 - CS209 2 SIS-012-280-1 12.750 2327 REFLECTIVE 3.7 CS209 25I5-012-071-1 12.750 1.312 SR376 TYPE 316 105 SR376 TYPE 316 105 680 BARE O.0 CS209 2S15-012-070-2 12.750 1.312 SR376 TYPE 309 105 680 BARE O.O CS105 25I5-012-252-2 12.750 0.375 2327 . REFLECTIVE 3.7 'CS202 1.312ee SR376 TYPE 316 545 1
I 2S15-012-287-1 12.750 2327 REFLECTIVE 3.7 CS202 25I5-012-069-1 12.750 1.312 SR376 TYPE 316 105 l SR376 TYPE 316 105 680 BRRE O.0 CS202 I 2 SIS-012-068-2 12.750 1.312 105 680 BRRE O.0 CS107 2515-012-251-2 12.750 0.375 SR376 TYPE 304 1.406ee SR376 TYPE 309 656 2327 REFLECTIVE 3.7 CS203 2RCS-014-Oe9-1 14.000 NOTE:
- Reflective insulation has tsoth an inner and outer steintess steel lever ee Westin@ouse nozzle Fitted with Sch 140 weld prep, connects to Sch 160 piping, countertnered to match s
s
thele 3.4 LeetPJET PaoGksul PIPE SPEAKS AND enestraaHRE TYPE ufLD esfWtDunstE List sRaeEda SPEAK easebER (Fs pare 3.13 12-asoch SIS to Cetd Leg AD e og 253 5-en,3-C4 Ese Drs 2515-PetR-406
<ss ,.re 3.8 e-sace. to Fu 2535-PRR-SSO 2515-412-2e9- 3 Srl 25854RR-tu6 2585-et06-012-3 25tS-wet-C-C IE 25tS-Peta-eSO 2515-005-C-4. Ese are 25I5-PER-406 25IS-803-C-C EA Fu 253 5-407-C-L EA 25IS-IS2-C-L Ese Fu 258 5-JI S-eks 25I S-eu9-C-e. EA Su 25854RR-oo.
."$I F RS3-C-C EA tu 25tS-Pp2-eSt 253 5-430-C-4' Ese Su 25tS4RR-ec6 25tS-LS4-C-L ta Fu 25IS-JIS-4St 2515-013-C 4 Ese fu 25I 54ets-eus 25f s-3SS-C-C ise tu 25154PR-458 2515-4 3 3-C -L Em 5u 25I5-PRR-ton Faa 2535-J15-8"Bt 2515-412-C-C Ese fu 25I S-tam-tos,4G1 25I 5- ESo-C -L ito 2515-lSe-C-L te Fu 2585-JI5-413 2515-032-4,67-E taa A550C1 ATEo seARounkE 2585-se3-C-L EA FM 2515-Ja5-458 25I5-032-06e-2 tat ee5SOCleTED seHRDuRkt 2535-006-266-5 udd stS50CthTE8 teAPDuRAE 2585-et2-2SG-2 ans se55GCleelte seestuusekE Gagure 3.2 e-s ce. 515 4. Cold Leg e) 25tS-JIS-eeO (Fignare 3.14 12-tech SIS to Cete Le4(E84 Fu 25IS-JIS-oo3 2585-east- O tS- t 2535-162-C 4 TE Fu 25tS-JIS-eeO 2545-012-28e-t '2535-osa4-C Su 2515-485-e05 25I5-144-C-L En Fu 2515-037-C-C ER 25IS-tae-C-L ER ru 2515-JI S-eed 25ES-9te-t-L E se Su 25I S-JI S-eOS 25IS-368-C-L EA Faa 25tS-JIS-see 253 5-020-C4 Ese Su 2515-JIS-es#5 25IS-422-C 4 EM Su 25I S-JI S-40S 253 5-021-C4 Ese su 2515-PAR-etS.Ste 2585-026-C-L Ese en 25154PR-417 en 25tS-esa-e!4
&s pre 3.3 m-se.ce. 515 to Cold Leg O TE Fu 25tS-PRA-471,JIS-471 258 5-02s-C-L Ese 2585-ch6-03e-1 253 5- 17 7-C-C 2515-012-071-1 eso A550CI ATED seRRouukE 25tS-t?9-C-C EA Fu 25ts-PRA-e72 25I5-042-070-2 eau se550CleEED seHkananmE 253 5-a04-273-3 esse 2550CI ATED sene@netaf 2135-012-252-2 see RSSOCI ATED seHMbeWif (Fa gure 3.15 12 sence SIS to Cete Leg C)
~
et5 isetED setRounat 258 5-e32-C A Esb Su 25IS-J15-60G 25I5-012-287-5 EA Sta 2515-JIS-e(s3
.2515-ease-2ee- t ens et550CIATED esses @uhRE 2585-454-C-L 25I54RR-424 2535-435-C-C ER Su J Su 25t $-JIS-e(s)
', G a gnere 3.S e-sach 515 to oest Leg 3) 25IS-ste-C-C TE FM 21354RA-460 2515-036-C-L 2515-437-C-C ER EA FM 25t S-Past-427 2515-eass-624- 1 EA rna 2535-PkR-468 fu 25t $-esta-a27 25t h ate-C-C 25I5-038-C-L Ese 25I5 4Nat-624 eso R550CIATED esAkonstitf 2515-439-C-C EA ett 25ts-oG6-2ee-1 ** Ort 2515-PRa-424 2535-010-C-L Ese 25I5 4 am-626 (F6 psre 2.6 e-sach SIS to seet E.g OTE 252 5-eve n-C-C Ese Su 2515-147-C-C Fu 25tS-PRR-470 2515-eg2-C-L Ese Su 25tS-PRR-827 2515-ems 6-42S- 8 Fu 2585-J45-e70 25IS-Pah-625 2515-t*I4-C-L EA 25t h-@43-C-C Ese Saa eso stS50CleerED sethtaahAE 2515-044-C-L Ese su 253 5-esta-e27 2515-006-267-t . . . . . . ,, . sea A550CIATED #eHRos4 APE see se550CI ATED seHetatetPK 2515-412-2S t-2 see M550CI ATED seekouAkE 2RCS-eJO4-C-C FE Fu 2RC54RR-eS3 JAC5-ense-02 4 - 1 (Figure 3.14 14-sanch 5 rge Lae.e to edet Leg C3 2stC5-ema-834,e tS TE FW (re pr. 3.8 e-sach Leap 9 Dypees* Fu 2RCS-PRA-420 2stC 5-O t+-OS4- 1 2RC S- 241-C -C Su 2stC 54Ra-e14 2nC5-eMae-440-3 2RCS-ube-C-C TE 2MC 5-242=C -C Ese TE Fu 2RC54RR-eS2 2pC5- 2<s34 -4. Efe SW 2stCS-esta-a ge JDC5 m t-4 2PCS-csDS-C-C 2RC 5-2444-C EA Su 2stC 5-Pan-62S Ese Su 20tC54sta 834>
' G ignere 3.9 t-sace. Leap C Dypeest 2AC5432-42e 2PC S- 2-eS-C -L 2mC54aR-e34,ets i
2DC5-eJUS~$s40- 1 2RCS-612-C s TE Fat 2kC 5-2-es-C-C Est ert OstC5-uo9-C-C IE Fu 2kCS PkR-eSt 2mCS- 247-C -4. En ers 2stC 5-estF-e tS,e te 2RCS-eue-66 3- 1 Em en 2mC5-PRR-e25,ete 2RC5-2-ee-C-C g Gnepare 3.30 to-se.cen ase5 Da mes erge, C=emeste S) 2kC 5- 299-04 Ea het 2aCS-fast-e R4 2RMS-es01-C-C TE tu 25tS-PRA-et? 2kC5-2SG-C-C Em are 2mC54stst-e25,e tS
' 2*ns-elo-023-1 2stC 5-Paa-S ES,e l4,826 2 sac 5-2524-C En ers 2NC 54Ra-415,82S ,8 th l
Ge w . 1.53 tek s2Ree5-uG2-C-C ren Rees Descenerge, Cubscle CD TE eu 25tS-PRR-426 2kCS-254-C-C 2kC5- 2S6-C -C Ese TE Fu Fte 2MC 5+etR-oo?,eae l
2kses-ete-424-1 e 3.12 32-a cs= stHS Seesteen fre seet rea Le9 As 2stees-Pfa-e01 Jass5-uo3-C-C TE f _ _ _ ,12-003,8 esG AS5dCIssTED
- e ug i *- I2-8SE - t i
TE = lermanet End Greets
! 42 = Eseceese to.e $srese end/or Fatsgnse Facter ThreshetJ ter Breets CEsDE s fu = a raeld feeracesed We3J se nede et Thas Peetesteted Pape Rupt.sre Lesettese Su = se sen.sp Fema scoted 68.14 s e no.se et Tense Peetestete PareRus,ture en - seee tietet. rnere s. e 44 .e ts.s e Pe.t t ete d Pape Lesetseen R pture L.es.esen I
i t
1 i
I 1
0
- Q Q l.
1 -
G" CSS l CL EL 74 l'- 5" l '
i 2 SI S-179- C- C Ih I 7 251S-006-271-1 2 SIS-17 7- C- C
~
i '
G~ SI5 l
LI I IN 6 2 SIS-OOG-16-1 j 2 SIS-PRR872
. L0 CATION i -
NODE IBD I
1,, '
.([2 SIS-PRR871 s P R O B. X.lo S F ,
l i *
] 1 j
1 2RCS-27S-9 -1 SC2
! ~' '
f, CL EL 732'-33I 16
, (COLD LEG) l ..
MoocAATc g,WsECaY l n aure 3 3 6' SIS PIPING TO COLD LEG C CUBICLE C EAST LOOP i
l .
~
l .
1
1 o
i .
O O -
l I
1 4 >
8VPS-2 HIGH STRESS LOCRTIONS FOR HHIPJET LB8 INRLYSIS p f
i l
1 LIE CHLC LORDING CASE FORCES (1bs) MOPENTS (f t.-lbs)
J
- REU CIDO1TIONS ND. Fx Fy Fz Mw My Ma i
1 2 -1067 -10
-5 10 -43 135 515 6-INCH X108F DEROMEIGHT 721 11223 REV 2 TERMFL 14 -16 -843 80 1135 SSE 10 169 172 283 732 2720 1610 2 1115 -1967 114 285 52 4344 8-ITCH X70B DEFElHEIGHT 2 TERIR. 13 -2128 -763 -884 -3804 -4314 11098 SSE 10 1258 1735 3% 1219 854 1761 2 7 -2949 9 -7 -179 -4015 RHS 10-INCH DEFEINEIGHT 2 TERMFL 13 48 -2318 -80 2162 714 -31215 i l 5194 4998 7868 SSE 10 2566 3338 2017 J
X70R DERONEIGHT 2 -77 4500 318 -6440 -1555 6306 RHS 12-INCH 361 -17014 2 TERin. 13 -1319 -2369 -1679 15713 ,
SSE 10 5210 3639 2077 7147 20757 29201 2 2807 579 38 1031 -4231 -2405 5 8 12-INCH X70R DEFElNEIGHT 4866 -38393 -22826 38127
.'
- p 2. TERfe. 13 -737 3706 f '
10 1894 1109 1676 2847 10799 9045 i
SSE DEf0 HEIGHT 2 105 2135 59 -2237 -1729 10360 RCS 14-IPCH X70C 161404 47651 TERMFL 13 11592 3479 -14015 2951 2 1525 12814 8969 SSE 10 1791 1241 1529
\
\ - _ . - . .-- -
. - LinOO SRP 3.6.2 47 Requiring 6 (Except arbitrary
. Protection .
intermediate breaks)
O 1 r Target and Interaction Evaluations
- u screenings i
Industry Experience, Economics Leak Detectability Margin i
1 Material Highest Stress Properties ; 4 Locations (Normal +SSE)
With Minimum Material Properties i r Fatigue Cracir
! , Growth Evaluation
- l F
l 1 r 4
Limiting Detectable Leak Rate Normal Loads
- Leak Rate x Margin ? Calculations
I n Margin on m Crack Stability 2 Normal +SSE
- Crack Size "
Evaluat' ion Loads i
l i u l
i Stability Check i For Excessively 1.414(Nor+SSE)
High Loads Loads -
l O
l 1
.__-n--,,.,. - - - , - - , -
,,n-------_ _ _,_ _-_
O HIGHEST STRESS LOCATIONS (NOR + SSE LOADS) WITH l
i MINIMUM MATERIAL PROPERTIES O
o ANCHOR-TO-ANCHOR EVALUATION l o ALWAYS WITHIN CLASS 1 PORTION i
o EVALUATED FOR BASE METAL AND WELDS IN SPECIFIC LINE O
O i
MATERIAL PROPERTIES o BVPS-2 MATERIALS SHOWN TO BE REPRESENTATIVE OF INDUSTRY DATA BASE (TENSILE PROPERTIES AND WELDING PROCEDURES)
TYPE 304, 304N, AND 316 SS BASE METAL SMAW AND A FEW SAWm O o FATIGUE CRACK GROWTH --
FORM OF
! JAMES / JONES CORRELATION ADJUSTED FOR PWR ENVIRONMENT o STRESS-STRAIN CURVES MINIMUM FOR STABILITY ASSESSMENTS AVERAGE FOR LEAK RATE ESTIMATES o J-R TOUGHNESS CURVES MINIMUM FOR STABILITY ASSESSMENTS 4
i
Q -
ax(NPaVE) ,
10
' r 100 I I i lIlli I i I t ,4 fl ser a=o.2 a=o. 7 ,?
. o w 4 ,
,,-2
, o-G a -
FonGEo 304 O E a
g roacco sean e . -
w, R _
=
g to-4 - T=550 F (288 C) - -
2 g R O
}
P=2000 psi (14MPa) - I,
.- .c
. a -
.- 'l" w .
_ 10-2 O<
r -
g da ;
i.
j - =
g - F C S (aK)3.3 , -
g
.dn '
. g O! ,,.,
r-2.
n -
- M e
g -
e .
u -
- u
. . 4
. *e 10 "
Bamford, ASME JPVT
! Feb. 1979 p. 73 Z
,o-s I l i 1/ l i ll i I I I lill 10 . 100 aK (KSI M )
- 1 O
---m-------- - - - ,
STRESS - STRAIN CURVES
.O V
120 -
T = 550' F 800 100 -
g lii
- 600 80 - SAW "
m g SMAW y 60 -
400 0 m
40 g Type 316- . .
200 Type 304 0
7 ,
!0 ' ' ' ' .
Q 0 0.02 0.04 TRUE STRAIN 0.06 0.08 Oil J-R CURVES da (tum) 0 1 2 3 4 5 1000 , , , ,
- T = 550' F 8000 1.5 Type 304/316 c.
n 1.0 s6 60,00 -
y c
4752 (
,R 4000 -
.o T - SMAW 0.5 a
U 2000
- SAW 0 i . - - I . . . . 0 0 0.1 0.2 A a (in.)
FATIGUE CRACK GROWTH EVALUATION o POSTULATE INSIDE SURFACE CRACK EQUAL TO BECTION XI ACCEPTANCE STANDARDS (ASPECT RATIO - 6) o EVALUATE FOR GIGNIFICANT GROWTH BASED ON EXISTING CLASS 1 STRESS / CYCLE INFORMATION (40-YEAR LIFE AND FULL STRESS INDICES) l o CRITERIAs CRACK GROWTH LESS THAN h o"A l THROUGH THE WALL PLASTIC ZONE BIZE LESS THAN REMAINING LIGAMENT OTHER CRITICAL LENGTH j ABBESSMENTS j o RESULTea (ALL OTHER CRITERIA MET) !
! l LINE EINe6 S41 ea SIS O.1s Ba RCS o.1s 10" RHS O.11 O 1=~ RH= o.11 O.21 12" SIS l 14" RCS O.32 J
___. ._ . _, -. ___,._ _ _._._. .._ .,,__ _ . . . c_y . , . , ,_,_.-%.__m.,__ . . _ . _ - _ - .
LIMITING DETECTABLE LEAK RATE AND MARGIN I
o BVPS-1 EXPERIENCE INDICATES LESS TH'AN O.5 GPM UNIDENTIFIED
- LEAKAGE WILL BE DETECTED AND i IDENTIFIED IN A TIMELY MANNER INSIDE CONTAINMENT LABORATORIES SPON ORED BY EPRI DEMONSTRATE THAT LEAK RATES EQUAL TO AND ABOVE O.1 GPM CAN BE READILY DETECTED VISUALLY
- o THEREFORE, 0.5 GPM IS THE l LIMITING LEAK DETECTION j CAPABILITY l
o MARGIN OF 10 ON LEAKAGE CRACK SIZE TO LEAK DETECTION CAPABILITY RESULTS IN A 5 GPM LEAK RATE CRACK SIZE TO BE ASSESSED FOR STABILITY O
LEAK RATE CALCULATIONS o NORMAL OPERATING LOADS COMBINED ALGEBRAICALLY DEADWEIGHT THERMAL EXPANSION PRESSURE o EPRI PICEP COMPUTER CODE BENCHMARKED USING LABORATORY AND SERVICE DATA 25% ACCURACY WHICH IMPROVES i WITH LEAK RATES ABOVE O1 GPM o PLUGGING IS NOT A PROBLEM FOR THE LEAK RATES AND PIPE MATERIALS OF CONCERN t
o 5 GPM LEAK RATE CRACK BIZE 1
o AVERAGE STRESS-STRAIN PROPERTIES 1
l FOR BASE METAL l
l RESULTem o
() 14" RCS 4.4
NP..S 2._R. _$ 12.I.N01 LIE, .M...M. L LOH LB. MG.I,_C.I.MM . ..
o ig,g L .__.,.._.....3 : -
E i A
X 8.N ~~ - --'-'-
- j
~~ * -
R ,
i A : i :
6.N + * + -+i. .
+ - + + --: '
E 4,g ,.........i.........L.....
C
... .... .. .l P j N
2.N - -
'e +-' - *
.00 2.00 4.00 6.00 8.00 10.'N CMCX LINGTH (INCHt$)
( ~ .NP$ - .2 .M$.12. !..N_C.H .L.I..E,.. MM_.L L4.0 LDM.CI, A..XI.AL iggg , ...........r - * - - - -
'7 E
I A
X 8.00 '- - - -
R
'A 6.00- '
' +
- I ' + - ' '
E .
4.00- '
+- ' -'-
P M *-
2.N - +- - -
'+l4 2
.00 2.00 4.00 6.00 8.00 10.00
\
(MCX LINGtH (INCHis)
._m - . _ , . _ _ _ _. _ - _ _ _ . _ , . . . _ _ _ _ _ _ _ _ . _ _ _ _ . - - _ , - _ _ _ . _ _ _ . _ , _ _ . . _ _ . _ _ _ _ _ _ _ . , . _ _ _ _ . - - . - _ _ _ - .
CRACK STABILITY EVALUATION o NORMAL + SSE LOADB COMBINED ABSOLUTELY a EPRI FLET COMPUTER CODE VERIFIED USING EXT.BTING DATA PLUS COMPARISON TO OTHER CODES USED DPFAD APPROACH WHICH INCLUDES CRACK GROWTH o CRACK SIZE MARGIN OF 1.5 -
2.0 FOR 5 GPM CRACK SIZE o USED LOWER BOUND STRESS-STRAIN AND J-R CURVE PROPERTIES FOR BASE METAL AND WELDS o RESULTes DEEen Bel 1Q LINE MORE EdeW E8W 6" BIBC1.5) 1.1 1.6 -
i B" RCS 1.0 1.5 -
i 10" RHS 1.6 2.4 2.1 12" RHS 1.1 1.5 1.5 12" BIS 1.6 2.4 2.0 14" RCS 1.4 2.0 1.7
(
1 asis-012-289-1 mor 449 nera (mmt + sso cast rYPE 316 O 1.s 0.99 -
9.80 -
va c=6t-
- sta" g,yg, ,
9.60- -
8- 0.50- -
Stable Q,(Q,, Region Leading Line DPFAD Wargin = (b/a)
- 11. 3 0
^ " " '"" * ' '"* /
9JB- ,
O 0.10- .
,' Y..91
/ -
9.89 1.69 2.49 h
O
I l
STABILITY CHECK FOR ,
EXCESSIVELY HIGH LOADS o 1.414 x (NORMAL + SSE) LCADS o EPRI FLET COMPUTER CODE o 5 GPM CRACK BIZE (NO MARGIN ON CRACK BIZE) o USED LOWER BOUND STREBB-STRAIN AND J-R CURVEB FOR BASE METAL AND WELDS l
o RESULTBs l QEEen BOI1Q l
uxus mens amew new 6" sis 1.6 2.4 -
B" RCS 1.9 3.0 -
10" RHG 2.1 3.2 2.8 12" RHB 2.0 3.1 2.6 12" sis 2.0 3.0 2.6 O 14" RCe 1.4 2.1 1e
BEAVER VALLEY UNIT 2 APPLICATION OF LEAK-BEFORE-BREAK TECWiOLOGY TO CERTAIN PIPING O September 6,1985 Initial proposal by Duquesne Light Company to use LLS technology October 10, 1985 Additional infomation from applicant on proposal November 13, 1985 ACRS letter, " ...... departure from previous practice and could have benefits" March 3,1986 Letter to applicant giving preliminary "go ahead" with WMPJET program March 4,1986 First progress report meeting March 12,1986 Second progress report meeting April 10,1986 Third progress repm ; meeting April 20,1986 Fourth progress re , ort meeting July 10,1986 NRC staff site visit to observe piping at tv-2 August 13, 1986 Special meeting with NRR management and NRC legal staff regarding WEPJET August 27, 1986 Fifth progress report meeting September 12, 1986 Staff position on WMPJET review September 16 and 1/, NRC staff and consultant audited piping stress '
IIO0 procedures in Stone and Webster offices October 9,1986 NRC staff observed EPRI leak detection tests October 23, 1986 WEPJET Report submitted January 16, 1987 Meeting to discuss review of report February 2,1987 App 1tcant provided additional infomation and !
O , fomally requested exemption from GDC 4 '
I March 1987 SER Supplement 4 issued with staff review results i
()
REGULATORY MILESTONES IN LEAK-BEFORE-BREAK i
! - MAY AND JUNE 1983 ACRS ON LBB RESOLUTION OF USl A-2, ASYMMETRIC LOCA LOADS,
- WINTER 1984-85 ACRS ON NUREG 1061, VOL, 3.
l OCTOBER 1985 BV-2 EXEMPTION ON DYNAMIC EFFECTS OF POSTULATED l PIPE BREAKS IN PRIMARY COOLANT LOOPS, FEBRUARY 1986 ACRS ON STATUS OF BEAVER VALLEY 2 WHIPJET PROGRAM, APRIL 1986 PUBLICATION OF MODIFICATION OF GDC-4 ALLOWING l LEAK-DEFORE-BREAK FOR PWR PRIMARY COOLANT LOOP PIPING.
O -
APRIL /MAY 1987 ACRS ON HEAVY COMPONENT SUPPORT REDESIGN / FINAL BROAD SCOPE GDC-4 WITH ITS SPP, l
I i
l il b__- .
O CRITERIA FOR REVIEW 0F BV-2 WHIPJET
- NUREG-1061, VOL, 3 " REPORT OF THE V.S. NUCLEAR REGULATORY COMMISSION PIPING REVIEW COMMITTEE, EVALUATION OF POTENTIAL FOR PIPE BREAKS," NOV 198ti.
- 51 FRN 12502, april 11, .1986 - FINAL NARROW SCOPE REVISION TO GDC-4.
- 51 FRN 26393, JULY 23, 1986 - PROPOSED BROAD SCOPE REVISION TO GDC-4.
O O
1
v Q NRC SCREENING CRITERIA WHIPJET CANDIDATE LINES WERE ASSESSED TO DETERMINE P0TENTIAL FOR SIGNIFICANT DEGRADATION DURING SERVICE
- FATIGUE
- THERMAL AGING
- CORROSION / EROSION
- CREEP O
i i
i O
p
- ,Q-PIPING INCLUDED IN ORIGINAL SCOPE OF WHIPJET PROGRAM RCS - BYPASS' LINES SAFETY INJECTION SYSTEM (SIS)
ACCUMULATOR INJECTION LINES PRESSURIZER SURGE LINE RESIDtIAL HEAT REMOVAL FEEDWATER RCS - SMALL DIAMETER PIPING MAIN STEAM STEAM GENERATOR BLOWDOWN O INCLUDED FERRITIC AND AUSTENITIC, INSIDE AND OUTSIDE CONTAINMENT, PIPE SIZES FROM 1.5" TO 16",
i i
l 1
O
l O
s../
SUMMARY
OF HARDWARE ELIMINATED AS A RESULT OF WHIPJET SYSTEMS PIPE SIZE PRR JIS RCS-BYPASS LINES 8" 6 0 6" 7 5
.. SIS ACCUMULATOR LINES 12" 10 3 PRESSURIZER SURGE 14" 8 0 RESIDUAL HEAT REMOVAL 12" 1 0 0
([) RESIDUAL HEAT REMOVAL 10" BREAK IN 10" RHS MITIGATED BY RESTRAINTS ON CONNECTING ACCUMULATOR LINES, LINES ARE' WROUGHT TYPES 304, 304N, OR 316 AUSTENITIC STEEL WITH SMAW AND/0R SAW WELDS, LINES ARE ALL INSIDE CONTAINMENT, O
1 O
LEAK-BEFORE-BREAK FROCEDURES AND MARGINS THESE PROCEDURES AND MARGINS ARE THE SAME AS THOSE USED FOR PRIMARY COOLANT PIPING, PREVIOUSLY REVIEWED BY ACRS PROCEDURES FOR LBB CONSIST OF:
- DETERMINE AS-BUILT LOADS IN PIPING SYSTEM
- CHARACTERIZE MATERIAL PROPERTIES
- DETERMINE LIMITING LOCATION
.
- DETERMINE MINIMUM DETECTABLE LEAK
- APPLY MARGIN OF 10 ON MINIMUM DETECTABLE LEAK TO DETERMINE LEAKAGE SIZE CRACK
- VERIFY STABILITY OF CRACK TWICE THE LEAKAGE SIZE CRACK -
- VERIFYSTABILITYOFLEAKAGESIZECRACKFOR/2XMAXIMUMLOADS
([)
f I
)
_ . _ . - . _ . _ _ . - _ . , - . _ _ . , _ _ _ _ _ _ - . ~ . . . _ _ _ . . _ . . . _ . . _ _ _ . . . - . - _ . . . _ _ , , _ _ . , _ . . _ _ _ _ _ _ . . _ _ _ _ . . . _ . . . _
O .
LEAK DETECTION CONSIDERATIONS CONTAINMENT LEAK DETECTION SYSTEMS MEET R.G. 1.45 MINIMUM DETECTABLE LEAK 0.5 GPM PRIMARY SYSTEM EMPLOYED FOR LEAK DETECTION - CONTAINMENT SUMP .
SUMP FLOW RATE MONITORED ONCE A SHIFT AVERAGE UNIDENTIFIED LEAKAGE FOR BV-1 IS APPROXIMATELY 0.3 GPM OPEP.ATING PROCEDURES REQUIRE CONTAINMENT ENTRY AND INSPECTION FOR UNIDENTIFIED' LEAKAGE EXCEEDING 0.5 GPM (2) 9 O
t
1.
, f ::
}
STAFF CONCLUSIONS WHIPJET-ANALYSIS NO LINES IN WHIPJET WITH POTENTIAL FOR UNACCEPTABLE DEGRADATION FOLLOWED NRC GUIDELINES ON ANALYSIS AND MARGINS USED APPROPRIATE MATERIALS DATA EMPLOYS ACCEPTABLE LEAK MONITORING AND DETECTION PROCEDURES USED BENCHMARKED LEAKAGE AND STABILITY CODES CONFIRMATORY STABILITY CALCULATIONS VERIFIED WHIPJET RESULTS STAFF RECOMMENDS EXEMPTION BE GRANTED TO ELIMINATE DYNAMIC EFFECTS OF POSTULATED PIPE BREAKS IN FINAL WHIPJET SCOPE O
O
d) ,' .
HISTORY OF PIPE CRACK ACTIONS EARLY HISTORY IGSCC FOUND AT DRESDEN 1 IN 1965 ATTRIBUTED TO IMPROPER WELDING FURNACE SENSITIZED COMPONENTS ALSO WERE FOUND TO BE CRACKED.
A RULE REQUIRING LOW CARBON f1ATERIAL WAS SERIOUSLY CONSIDERED IN 1972 REG GUIDE 1,44 " CONTROL 0F THE USE 0F SENSITIZED STAINLESS STEEL" WAS ISSUED.
(])
- HIGH CARBON MATL CONTINUED TO BE ACCEPTED.
CONTROL OF WELDING PARAMETERS USED AS BASIS.
THESE CONTROLS PROVED TO BE INADEQUATE.
1974 CRACKS IN RECIRC BY PASS R. O BULLETIN 74-10 ISSUED RE0VIRED INSPECTIONS WITHIN 60 DAYS IE BULLETIN 74-10B ISSUED TIGHTENING LEAKAGE LIMITS 1975 PIPE CRACK STUDY GROUP FORMED 1
EARLY HISTORY (CONTINUED) 1975 CRACKS IN CORE SPRAY LINES FOUND
- BULLETINS 75-01 AND 75-01A ISSUED
- REQUIRED INSPECTIONS WITHIN 20 DAYS 1975 STUDY GROUP REPORT NUREG 75/067 ISSUED 1977 NUREG 0313 ISSUED STAFF POSITIONS SET FORTH ON MATS & PROCESSES AUGMENTED INSERVICE INSPECTION REQUIREMENTS AUGMENTED LEAK DETECTION REQUIREMENTS
(]) LEAKSANDCRACKSCONTINUEDTOBEFOUNDINSMALL(jf12") LINES 1978 CRACKS FOUND IN LARGE (22")
PIPES IN GERMANY, 1978 NEW PIPE CRACK STUDY GROUP FORMED 1979 STUDY GROUP REPORTED ISSUED 1980 NUREG 03]3 REVISION 1 ISSUED IMPLEMENTED BY A GENERIC LETTER RESPONSES INDICATED POOR CONFORMANCE O
4 (I RECENT HISTORY 1982 LEAKS AND CRACKS FOUND IN 28" PIPE AT NINE MILE POINT UNIT 1 INFORMATION NOTICE 82-39 ISSUED BULLETIN 82-03 ISSUED REQUIRED SANPLE INSPECTION OF RECIRC PIPING AT 9 PLANTS REQUIRED DEMONSTRATION OF EXAMINATION CAPABILITY l
()- 1983 CRACKS FOUND IN 7 0F PLANTS l'
BULLETIN 83-02 ISSUED REQUIRED INSPECTIONS OF REMAINING PLANTS UPGRADED CAPABILITY DEMONSTRATION TESTS ORDERS ISSUED TO 5 PLANTS TO INSPECT EARLIER, 1983 PIPING REVIEW COMMITTEE AND PIPE CRACK TASK GROUP FORMED SECY 83-267C DESCRIBED STAFF REQUIREMENTS FOR REINSPECTION AND REPAIR OF CRACKED PIPING
()
O IMPLEMENTED BY GENERIC LETTER 84-11 (SHORT RANGE PLAN)
. STAFF LONG RANGE PLAN FOR DEALING WITH STRESS CORROSION CRACKING IN BWR PIPING.
-AFTER ISSUING SECY 83-267C COMMISSION REQUESTED A LONG RANGE PLAN SECY 84-30L DESCRIBED THIS PLANT-(IDENTIFIED AS GENERIC ISSUE 86)
ISSUE NUREG-1061 VOL. 1 FOR PUBLIC COMMENT O *
, - REVISE NUREG 0313 TO INCLUDE RECOMMENDATIONS OF 1061 VOL. I CONSIDER:
f-COMMENTS ON 1061 VOL. 1 RESULTS OF REINSPECTIONS CODE MODIFICATIONS ONG0ING INDUSTRY DEVELOPMENTS IMPLEMENT BY GENERIC LETTER O
l
( )- PURPOSE OF NUREG 0313 REV. 2 THE PURPOSE OF NUREG 0313 REV. 2 IS TO PROVIDE BWR LICENSEES WITH ALTERNATIVE MITIGATIVE ACTIONS OR COMBINATIONS CONSIDERED ACCEPTABLE BY THE STAFF THESE INCLUDE:
PIPE REPLACEMENT WITH RESISTANT MATERIAL WELD RESIDUAL STRESS IMPROVEMENT WATER CHEMISTRY IMPROVEMENT
() WELD REPAIR AND EVALUATION AUGMENTED INSPECTION SCHEDULES TECHNICAL BASES ARE AS RECOMMENDED BY THE PIPING REVIEW COMMITTEE, NU:?EG 1061 VOL.1 NO NEW REQUIREMENTS ARE IMPOSED BY NUREG 0313 REV. 2 THE BASIC REQUIREMENTS ARE THE GENERAL DESIGN CRITERIA REV. 2 PPOVIDES THE LICENSEES WITH A BROADER SELECTION OF ALTERNATIVE WAYS TO MEET THE GDC'S THAN REV. 1 O
d
,)
NUREG 0313 REV. 2 EXPANDS REV. 1 C0VERAGE INCLUDES ALL STAINLESS PIPING (CL 1, 2, 3)
REV. 1 HAD LIMITED SCOPE USES FORMAL QUALIFICATION OF NDE EXAMINERS AND PROCEDURES INITIATED BY BULLETINS 82-03 8 83-02 AND DEVELOPED BY NRC-INDUSTRY C0 ORDINATION.
REV. 1 JUST RECOMMENDED THAT IMPROVED UT PROCEDURES BE USED l -
C)
PROVIDES QUANTITATIVE BENEFITS FOR PROVEN MITIGATION METHODS (HWC AND SI) i REV. 1 HAD N0 SPECIFIC PROVISIONS l PROVIDES GUIDELINES FOR EVALUATION AND REPAIR OF CRACKED WELDS 1
-REV. 1 REQUIRED REPLACEMENT OF CRACKED WELDS
/
(_) NUREG 0313 REV. 2 GENERALLY FOLLOWS RECOMMENDATIONS OF PIPING PEVIEW COMMITTEE NUREG 1061 VOL, I RECOMMENDS:
USE OF IGSCC RESISTANT MATLS REPLACEMENT OF SUSCEPTIBLE PIPING PROCESSES FOR RESIDUAL STRESS IMPROVEMENT IMPROVED WATER CHEMISTRY O PROVIDES SPECIFIC INSPECTION SCHEDULES CONSIDERING:
5 MATERIAL IGSCC RESISTANCE STRESS IMPROVEMENT PROCESSING WATER CHEMISTRY IMPROVEMENT REPAIRS AND CRACKING CONDITION PROVIDES GUIDELINES FOR CRACK EVALUATION AND REPAIRS UPGRADES LEAKAGE LIMITS AND MONITORING O
9 STAFF POSITION ON RESISTANT MATERIALS RESISTANT MATERIALS ARE:
LOW CARBON WROUGHT AUSTENTIC STAINLESS STEELS E.G. 304L, 304NG, 3L6L, 3L6NG, 347NG STAINLESS STEEL WELD METAL WITH LOW CARBON AND HI FERRITE CAST AUSTENITIC STAINLESS STEEL WITH LOW CARBON AND HI FERRITE O
INCONEL 82 WELD METAL LOW STRENGTH LOW CARBON STEEL NICKEL BASE ALLOYS ON A CASE BASIS i
NOTE THAT MOST AUSTENTIC ALLOYS ARE SUSCEPTIBLE TO IGSCC IN CREVICE AREAS POSITIONS ARE CONSISTENT WITH 1061 VOL. 1
O STAFF POSITION ON PROCESS SOLUTION llEAT TREATMENT (SHT)
ELIMINATES PRIOR SENSITIZATION RELIEVES RESIDUAL STRESS FROM WELDING HEAT SINK WELDING (HSW)
REDUCES SENSITIZATION FROM WELDING REDUCES RESIDUAL STRESS FROM WELDING O
STRESS IMPROVEMENT PROCESSES AS WELDED CONDITION HAS TENSILE STRESS ON ID SI REVERSES THIS STRESS DISTRIBUTION LAST PASS HEAT SINK WELDING MAY BE HELPFUL TO RESIDUAL STRESS IF DONE PROPERLY WILL BE EVALUATED ON A CASE BASIS POSITIONS ARE CONSISTENT WITH 1061 VOL, 1 O .
WATER CHEMISTRY BWR COOLANT IS ESSENTIALLY PURE WATER EVEN SMALL AMOUNTS OF IMPURITIES WILL CAUSE IGSCC IMPURITIES ARE OF TWO TYPES:
INCREASE THE OXIDIZING POTENTIAL INCREASE THE CONDUCTIVITY BOTH MUST BE LOW T0 KEEP THE ELECTROCHEMICAL
(]) POTENTIAL LOW 0XIDIZING P0TENTIAL CAN BE REDUCED WITH HYDR 0 GEN ADDITIONS CONDUCTIVITY CAN BE REDUCED WITH ATTENTION TO WATER CLEAN UP AND CONDENSER LEAKS MAINTAINING 0 2 TO BELOW 20 PPB AND CONDUCTIVITY TO BELOW 0.3 MICR0 SIEMENS WILL PREVENT IGSCC INSPECTION FREQUENCY MAY BE REDUCED WITH HWC O
l'~h CRACK EVALUATION CRITERIA CONSISTENT WITH NUREG 1061 VOLS I AND 11]
BASED ON LIMIT LOAD ANALYSES TAKES ACCOUNT OF LOW TOUGHNESS WELDS MOST WELDS ARE SAW OR SMAW fi.L STRESSES ARE INCLUDED FACTOR OF 3 AGAINST FAILURE IS REQUIRED.
2 WILL RECOGNIZE REVISED CODE IWB 3600
(])
l
! OLD IWB 3600 MAY BE USED IN INTERIM, HOWEVER:
CRACK DEPTH IS LIMITED TO 2/3 IWB SECONDARY STRESSES MUST BE USED AS-0VERLAID CRACKED WELDS MUST MEET SAME CRITERIA i
l VERY FEW CRACKED WELDS ARE NOT OVERLAID l
O
. 1 l
l 0 CRACK GROWTH CALCULATIONS METHODOLOGY AND PARAMETERS REVIEWED ,
AND ACCEPTED IN NUREG 1061 VOL, 1 -
PRESENTED AT LAST SMIRT CONFERENCE BASIS IS THE ITERATIVE FRACTURE MECHANICS K vs GROWTH RATE APPROACH STAFF POSITION SPECIFIES:
O RESIDUAL STRESS DISTRIBUTION ALL SUSTAINED STRESSES INCLUDED K vs RATE BASED ON RELEVANT TESTS RATES BOUND FIELD DATA s.
O
t O EXTENDED USE OF 0VERLAY REPAIRS 1061 VOL. 1 LIMITED USE TO TWO FUEL CYCLES UNLESS INSPECTION METHOD WERE TO BE DEVELOPED.
BWROG II FUNDED EPRI NDE CENTER EFFORT EFFECTIVE METHODS NOW AVAILABLE CAN DETECT ANY CRACKING INT 0 O' LAY CAN DETECT CRACKING WITHIN 4 T OF O' LAY REQUIRES ATTENTION TO SURFACE FINISH O
0313 REV. 2 REQUIRES INSPECTIONS EVERY 2 REFUEL CYCLES I
EPRI HAS LONG TERM PIPE TESTS IN PROGRESS O
HYDR 0 GEN WATER CHEMISTRY
-O REDUCED INSPECTION NUREG 1061 VOL 1 SPECIFIC FACTOR OF 2 ON INSP, SCHED.
NO CREDIT ON REPAIRED WELDS NUREG 0313 REV. 2 "ABOUT" A FACTOR OF 2 ON SCHEDULE, ON A CASE BASIS REDUCED FREQUENCY FOR REPAIRED WELDS-
- O REASONS 4
DATA FROM PLANT HWC TESTS SHOWS WIDE PLANT SPECIFIC i VARIATION HWC WILL ALSO BENEFIT REPAIRED WELDS I
l
(
O l
?
'O PRIMARY C0OLANT LEAKAGE LIMITS NUREG 0313 REV, 2 SAME AS NOW IMPOSED UNIDENTIFIED LEAKAGE MONITORED EVERY 4 HRS, PLANT SHUT DOWN FOR ACTION WHEN:
1 UNIDENTIFIED LEAKAGE RATE INCREASES I
BY GPM IN 24 HR PERIOD
()
OR TOTAL UNIDENTIFIED LEAKAGE REACHES 5 GPM i
1061 VOL,1 RECOMMENDED A TOTAL LIMIT OF 3 GPM REASONS FOR DIFFERENCE SOME PLANTS HAVE PROBLEMS IDENTIFYING SOURCE OF LEAKAGE CURRENT LIMITS JUDGED ADE00 ATE O
I I
L
TABLE 1
)
SUMMARY
OF INSPECTION SCHEDULES FOR BWR PIPING WELDMENTS IGSCC INSPECTION DESCRIPTION OF WELDMENTS NOTES CATEGORY EXTENT & SCHEDULE Resistant Materials A 25% every 10 years
, (at least 32% in 6 years)
Nonresistant Matls (1) B 50% every 10 years SI within 2 yrs of (at least 25% in 6 years) operation (1)
Nonresistant Matis (1) C All within the next 2 refueling SI after 2 yrs of cycles, then all every 10 years operation Non Resistant Matl (1) D All every 2 refueling cycles No SI Cracked (1)(2) E 50% next refueling outage, then Reinforced by weld overlay all every 2 refueling cycles or mitigated by SI Cracked (2) F All every refueling outage
, O' Inadequate or no repair Non Resistant (3) G All next refueling outage Not Inspected Notes:
(1) All welds in non-resistant material should be inspected after a stress improvment process as part of the process. Schedules shown should be followed after this initial inspection.
(2) See recommendations for acceptable weld overlay reinforcements and stress improvement mitigation.
(3) Welds that are not UT inspectable should be replaced, " sleeved", or local leak detection applied. RT examination or visual inspection for leakage may also be considered, i
, O
O ACRS COMMENTS ON DRAFT PACKAGE
- 3. A LIMIT, SUCH AS 25%, ON THE NUMBER OF CRACKED WELDS
.O
O ACTIONS ON ACRS COMMENTS
- 1. TYP0 FIXED - CHANGED TO 0.04
- 2. TWO MITIGA" IONS CONSIDERED NOT NECESSARY, IN LINE WITH PIPING REV EW COMMITTEE RECOMMENDATIONS (NUREG 1061 VOL. 1)
- 3. A PROPOSED LIMIT WAS INCLUDED IN THE DRAFT FOR COMMENT GENERIC LETTER DBL DID NOT AGREE WAS DONE AT CRGR'S REQUEST.
ESSENTIALLY ALL COMMENTERS OBJECTED O DBL CONCLUDED
&lin'A!ibhb WOULD BE COUNTERPRODUCTIVE TO CONSERVATIVE UT CALLS LIMIT IS NOT INCLUDED IN PRESENT LETTER O
t O
RESOLUTION OF PUBLIC COMMENTS WRITTEN DOCUMENT GIVES DETAILS.
IMPORTANT SUBJECTS
- 1. CASTINGS
- 2. MARGINAL FERRITE WELDS
- 3. INCLUSION OF ALL CODE CLASSES IN SCOPE l 4. FUEL CYCLE BASIS FOR ISI SCHEDULES
- 5. EXTENT OF EXAMS FOR CATEGORY A
! 6. SAMPLE EXPANSION PLAN 7, 25% LIMIT ON CRACKED WELDS
- 8. CREDIT FOR MSIP ,
l
( 9. PERSONNEL QUALIFICATION FOR CATEGORY A WELDS I
I l
O i
O' IMPLEMENTATION PLAN FINAL APPROVED VERSION OF NUREG-0313 REV. 2 WILL BE TRANSMITTED TO LICENSEE BY GENERIC LETTER GENERIC LETTER WILL REQUEST THE LICENSEE T0:
PROVIDE TilEIR PLANS REGARDTNG PIPE REPLACEMENT OR ALTERNATIVE ACTIONS
' PROPOSE A CHANGE IN THEIR TECH SPECS O
INCORPORATING AN INSPECTION PROGRAM CONSISTENT WITH 0313 REV, 2 THE STAFF WILL REVIEW RESPONSES AND EVALUATE THEIR ADE0VACY O
r l
- l t
1 O .
I l
r i
9 4
, PRESENT PLAN I ACRS SUPPORT IS WELCOMED.
t i
I COMMISSION PAPER DRAFTED. :
I i ASSUME COMMISSION MTG REQUIRED.
i O
1 4
1 J
b i
l l 1
b f
4 I
I I
{
t
-j i
l 4
i t
) i l I a
1 e f
1 i
i +
I
.r_____.____ ,_._._ _ __,.._...__--..__ _.--_ _... _ _, _ __.-_ _ ._,,.
k Document Name: l HAZELTON-3 O . Requestor's ID:
i DORSEY Author's Name:
W.HAZELTON ,
Document Comments:
RESOLUTION OF PUBLIC COMMENTS ;
O .
l
?
l I
l l
O :
l i
i
o UNITED STATES T o NUCLEAR REGULATORY COMMISSION y ,I WASHINGTON. D. C. 20666 u
m s.
,..... /
RESOLUTION OF PUBLIC COP 94ENTS TO DRAFT GENERIC LETTER, "NRC POSITION ON IGSCC IN BWR AUSTENITIC STEEL PIPING," AND NUREG 0313, REV. 2 Introduction The Draft Generic Letter "NRC Position on IGSCC in BWR Austenitic Stainless Steel Piping" and the Draft NUREG 0313, Revision 2" Technical Report on Material Selection and Processing Guidelines for BWR Coolant Pressure Boundary Piping" were issued for public comment in the July 21, 1986 edition of the Feder# Jegister. The formal period of public comments ended September 20, 1986, although comments also were received considerably after that date.
Formal comments were received from 12 utilities licensed to operate BWRs, General Electric, two consulting firms, and the Boiling Water Reactor Owners Group for IGSCC Research. Altogether, well over 200 comments were received, but most of these were essentially duplicates or very similar in content and thrust. Most of the comments were essentially editorial, requests for clarification, suggested re-wordings, etc, that did not represent any disagree-ment with the technical or licensing aspects of the documents. Nevertheless, several subjects did require further review and technical resolution.
In the following discussion of the staff resolution of comments, it was N cc,nvenient to group them by subject, as treating every one individually would be unwieldy, as well as confusing to the reader.
Most of the important comments were already combined into one package by the BWR Owners Group for IGSCC Research. These will be discussed first, and will cover the major subjects. The order of coments follows the numbering used by the Owners Group. Individual coments not covered by the Owner's Group will be addressed later. To the extent possible, the coments are simplified to reflect the salient issue. In many cases, clarification of the issue was accomplished by telecon with the originator of the comment or a representative of the Owners Group.
Owners Group Coments
, 1. (a) Coment:
l l
The staff position on IGSCC resistance of castings would require l augmented inspection of welds joining piping to cast valve and pump bodies, unless the castings contained a maximum of 0.035 carbon and a minimum of 8% ferrite. It was felt that this would not be warranted, in view of the generally good experience at such joints.
Discussion:
The basic concern is that when piping is replac with resistant material, the original pumps and valves are retained. These rally are made of material
> 'o O ~'ta c rbo" ce"t "t or o os to o oa. #o < rrit 'ev i Because the staff position on castings would classify these as nonresistant 6 55-material, the joints would be classified as IGSCC category B, C, or D, depending on when and if the joint was given an SI treatment, and would require augmented inspections.
G V Information in the technical literature available to the staff indicates that high carbon, low ferrite castings are not as resistant to sensitization and IGSCC as other materials classified as resistant by the staff.
However, such castings are more resistant than, for example, wrought type 304.
There has been at least one known case of IGSCC in a cast body in a BWR.
This case (Monticello) was apparently unusual because the joint had been repeatedly weld repaired, resulting in probable sensitization and unfavorable residual stresses. The particular weld joints involved have a different configuration than simple pipe-to pipe butt welds. The heavier section of the pump or valve side of the joint results in a more favorable residual stress pattern, which, in combination with the relatively mild degree of sensitization, makes this specific type of joint resistant to IGSCC.
Weld repairs to this joint may increase the probability of significant sensitization and modify the residual stress. distribution, so must be taken into account.
Resolution:
The staff has concluded that unless there has been a history of weld repair to an individual joint, that pipe-to pump and pipe-to-valve joints may be classified as ICSCC category A, and augmented inspections schedules are not required, if the piping is of resistant material. The Generic Letter and NUREG have been modified to reflect this position.
- 1. (b) Comment:
Plain carbon steel is immune to sensitization and IGSCC, therefore it should not be addressed in the Generic Letter or NUREG.
Discussion:
Historically, plain carbon steel has been included in the NUREG list of resistant materials. Because of concerns about the suitability of high strength steels for BWR piping, the NRC followed the lead of General Electric and referred to the material as " low strength, high toughness carbon steel".
The Owners Group apparently do not want specifications for strength and toughness for carbon steel in the document related to control of IGSCC in austenitic stainless steels.
Resolution The Scope of the Generic letter has been specifically re-written to exclude
Although the staff would have concerns if high strength ferritic materials were to be used as piping replacem?nt material (because they are subject to types of stress corrosion cracking) this concern does not have to be addressed in these documents.
- 2. Comment The concern here apparently is that the staff positions would classify weld metal with above 5 but below 8% ferrite as non-resistant, and would mean that Code-approved material would require augmented inspection.
I (m) Discussion Unfortunately, the Code does not specify requirements needed for control of environmental problems. It specifically states that this is the responsibility of the Owner. The fact that the Code only requires 5% ferrite in austenitic weld metal is not the relevant issue. The staff position on weld metal is that if it has a maximum of 0.035 carbon and a minimum of 8% ferrite, it is resistant. The resistance to sensitization and IGSCC of weld material depends on several interrelated factors; carbon content, ferrite content, size, metallurgical structure, and perhaps others. It is certainly possible to have a weld containing only 5% ferrite that is adequately resistant to sensitization, depending on these other factors. At the present time, the staff cannot provide a quantification of the synergisms involved.
Certainly, welds with low or marginal ferrite content might be shown to be resistant, and may be evaluated as such as a special case by the staff.
Resolution:
The staff position has been modified to specifically allow for evaluation of the resistance of weld metal on an individual case basis.
- 3. Comment Wrought material solution heat treated after welding should be considered to O 6 re i t #t.
Discussion:
This is the intent of the staff, as indicated in the position on processes.
Resolution:
To provide clarification, solution heat treated material is now also covered under the Staff Position on Materials.
- 4. Comment This comment points out that the resistant material type 347 modified may have carbon content up to 0.04.
Discussion:
Because type 347 is stabilized with niobium, it can contain higher carbon content and still be resistant to sensitization.
Resolution:
The maximum carbon content for type 347 has been increased to 0.04 in the Staff Position on Materials.
O
4_
O 5. Comment This comment provided a suggested rewrite of the Staff Position on
~~
Materials.
Discussion:
For several reasons, the words suggested do not reflect the exact Positions
. intended by the staff, but differences are minor, or as discussed in the previous comments.
Resolution:
The Staff Position on Materials has been rewritten, and incorporates most of the suggestions.
- 6. Comment Suggested deletion of the words "new or replacement" in the Staff Position on Processes.
Discussion:
This represents a typographical type of error.
)
Resolution:
The words "new or replacement" have been deleted.
- 7. Comment The Staff Position on Water Chemistry should be updated to reflect recent experience, and should be confined to piping.
Discussion:
Recent experience and results of research and development have provided more confidence in water chemistry control.
i Resolution:
The Staff Position has been reworded to provide a stronger position, and references to components other than piping have been deleted because they are not relevant to the piping issue.
- 8. Comment Under the Staff Position on Inspection Methods and Personnel, there may be confusion regarding the applicability of the requirement for a qualification
() program.
(3 V
Discussion:
It is the intent of this Generic Letter to require a formal qualification program only for examinations performed on piping covered'by the Scope of the letter. It is not the intent of this Generic Letter to require such qualification for other Code required examinations.
Resolution:
The staff position has been specifically reworded to include only those examina-that come within the Scope of the letter.
- 9. Comment The issue involved in this comment has to do with the inclusion of Class 3 piping in the Scope. Many commenters discussed this issue.
Discussion:
This is an important issue, with several ramifications. Many discussions and telecons were held with the various commenters to clarify their concerns and determine the real extent of the problem. As the Scope is presently written, only austenitic piping 4 inches or larger normally containing reactor coolant at temperature over 200*F is included. The salient facts appear to be the O' following:
- 1. The only piping that is either class 3 or unclassed that is covered by the Scope that was identified by the discussions with commenters is that part of the reactor water cleanup system past the second isolation valve on some plants.
- 2. Several of the commenters, after review, reported that they had no class 3 or unclassed piping that would be subject to the Generic Letter. This is because later plants have carbon steel in these i
systems.
- 3. A major concern was that the Class 3 and unclassed piping was fabricated and erected to less stringent codes; welds were not originally inspected, or even designed and fabricated to facilitate examination according to the requirements of the Generic Letter.
- 4. It was agreed that the reactor water cleanup system past the second isolation valve was not directly related to reactor safety, nevertheless
- the system is required to maintain adequate reactor coolant chemistry control. Thus its function is to protect the reactor coolant system, including piping, from environmental degradation.
l
- 5. If weld failure or excessive leakage in this system occurred, it could be isolated, but the plant could not operate for very long before the specification on water chemistry intended to protect the O core and other components would be violated. Thus the continued operability of this system is directly related to plant availability and core integrity.
O 6. there is no effective leak detection system for this piping, thus increasing the probability that a large, critical-size crack could develop without prior knowledge.
- 7. The same factors stated by the commenters to cause difficulty in examining the welds (lack of close control of welding and weld configuration, lack of original inspections, etc) are factors that are also likely to increase the probability of initiating and growing IGSCC.
- 8. The piping system in question has experienced considerable cracking.
The staff feels that when there is a high probability of cracks, the piping does not meet the original Code or licensing basis unless periodic effective examinations are performed, because neither the ASME Code or ANSI B31.1 permit cracks.
Resolution:
After reviewing these factors, the staff concludes that it is prudent to apply the principles covered by the Generic Letter to all piping, regardless of class, that is covered by the Scope of the Generic Letter. Cases involving undue hardship may, of course, be reviewed on an individual basis.
- 10. a Comment:
Some inspection schedules are intended to apply every other refueling outage. Commenters suggested that with present extended cycles, this period could be up to 4 years, rather than the 31/3 years as shown in the table in the Draft Generic Letter. (The 3 1/3 year interval in the Table was supposed to synchronize with the Code schedule of 3 1/3, 6 2/3 and 10 years). Commenters suggested the wording "Every two refueling cycles or 4 years, whichever is greater."
Discussion:
The staff believes that basing the inspection schedule on refueling cycles is acceptable, even if the refueling cycle is extended to 2 years. It is felt that the industry's stated concern can be more simply accommodated by basing inspections on refueling cycles alone, with no specificed time period. As the inspection period for Category A & B welds are based on Code, they should stay on a 10 year basis.
Resolution:
Except for IGSCC Category A and B welds, inspection schedules will be based only on refueling cycles.
- 10. b Comment:
(N The commenters picked up a discrepancy between the text and the Table related to the inspection requirements for IGSCC Category E welds during the next outage.
i
,,,--n . - . - . - . , _ - , . - - - - _ . . . - - . . , , . . - - - - - -- . . - - . - _ - . . . - . - . - - , - - - . . -- -
.q)
Discussion:
This was typographical type of error. The intent was to have the table match the text.
Resolution:
i The Table has been changed to match the text.
10.c Comment:
The extent of inspection shown in Table 1 of the draft Generic Letter indicated that all terminal ends and dissimilar metal welds and, in addition, 25% of other welds, should be examined. This was correctly shown to be more than required by the Code. It was suggested that the Code extent be used directly.
Discussion: ,
The extent of examination shown in the Table was intended to reflect the fact l
that for a replaced, redesigned recirculation system, following the' Code sample selection process would, in many cases, result in a non-representative sample.
This is because the.1977 and later Codes require that 100% of all terminal ends i
and dissimilar metal welds must be examined, plus enough other welds to bring the total sample to 25%. However, the new design of recirculation systems has only about 50-60% of the total number of welds as the older design, whereas it has the same number of terminal ends and dissimilar metal welds. This would result in an examination sample essentially entirely of terminal ends and dissimilar metal welds, because they would constitute approximately 25% of the total number of welds.
In reviewing the distribution of cracking that has occurred in recirculation that' terminal ends and dissimilar metal welds had a systems, the significantly lower perce staff noted,htage of cracked welds than the other welds. Clearly, a sample based entirely on these would not be a meaningful or representative sample. The staff believes that the sample selection should be based on good engineering judgement, and if bias toward certain types of welds is warranted
- for technical reasons, this should be reflected in the sample selection. The staff feels that the licensee-is in the best position to ' apply his knowledge of i stress levels, weld configuration, mitigation effectiveness and similar factors in selecting a meaningful sample.
Discussions with several Code representatives indicated that they agree in principle with this approach, and are now considerinkschanges in the Code to p'ermit other sample selection bases. Until such changes are effected, M. . licensees may use the provisions of 10 CFR 50.55a, (b),(2)(ii), which permits the use of the 1974 Edition of the Code for sample selection.
s s
.- , _ . , . . . - . - . . _ _ . . , . . . - - _ . _ _ . . . , _ , . , _ , . . _ , , . . - . . . _ .. _\ ._ . ._ - , . - . _ _ _
,. - ._ ._ _ . _ ~ __. __, . _ _.
O Resolution:
t '
The table and the text of the Generic Letter have been rewritten to require that the licensee select-the sample of 25% for IGSCC Category A and the 505 sample for IGSCC Category B based on his judgement in selecting a meaningful and representative sample. The 1974 Edition of the Code permits this, and the provisions of 10CFR50.55a, b, (2),(11) are referenced. -
-11. Comment:
4 Concern was expressed regarding the requirement that the leakage
! detection systems must conform to Regulatory Guide 1.45, because 4
although many do not, they have been accepted by the NRC on an L individual basis, Discussion:
J
-The adequacy of the leakage detection systems was addressed for each plant i during the licensing process, and should not be re-opened as part of the IGSCC issue. The wording used in the draft Generic Letter was copied directly from NUREG-0313 Rev. 1, and was considered non-controversial.
Resolution:
, (I The wording of the Generic Letter has been modified by adding the words
"--- or as otherwise approved by the NRC" to the sentence containing the requirement.
. 12. Comment:
There are three items included in this comment. Two are essentially editorial
, and non-controversial. The third involves the details of how a second sample
, of welds to be examined is selected, when one or more cracked welds is found in the group of welds constituting the required sample.
Discussion:
! It is agreed that the sample expansion plan included in the original draft NUREG-0313 Rev. 2 (it was not included in the draft Generic Letter) was confusing as written, and included some provisions that, upon closer review, were redundant or not relevant.
Resolution:
The entire position on requirements for sample expansion has been rewritten, simplified, and included in the Generic Letter as part of the Staff Position on Sample Expansion. The staff believes that this new section will resolve the concerns expressed by the commenters, while still providing a sample expansion plan that is consistent with past staff and Code practices.
r
O~
- 13. Comment:
Most of the commenters disagreed strongly with the proposed 25% limit on number of cracked welds in a piping system. Because this is a significant issue, the comment submitted by the Owners Group is included here in its entirety:
"A proposed addition to the draft generic letter sets a limit on the number of cracked weldments in piping, irrespective of repair or evaluation.
The basis for this limit is not discussed in draft NUREG-0313. Rev. 2 and appears to be arbitrary. The position of the BWROG-II is that measures for evaluation, mitigation, reinforcement and inspection as identified in the draft NUREG revision are adequate to control IGSCC in BWR piping systems, and that the additional arbitrary limitation is not necessary or beneficial.
Meaningful technical discussion is difficult because the basis for the proposed limit has not been disclosed. We speculate that a possible motivation is a desire to minimize shrinkage stress resulting from a large number of weld overlay repairs. However, shrinkage and fitup stresses are best quantified by analysis and controlled as part of the overlay design process, as discussed in the draft NUREG revision. Shrinkage stress is not sensitive to the number or proportion of welds repaired and is not limited by the proposed 25% limit. For example, analysis would show that a single overlay repair in a recirculation system riser causes higher shrinkage stress
- than a symmetric arrangement of repairs in each riser associated with one ring header. The arbitrary 25% limitation is not an effective way to limit or reduce shrinkage and fitup stresses."
l "We question whether an analysis of costs and benefits associated with the proposed limit would support its implementation. The proposed limit requires !
BWR owners to apply increased resources and man-rem exposure toward inspection l and sizing of small cracks which can have only small effects on system integrity. An arbitrary limit on the total number of small cracks could force replacement of lines which are structurally sound and reliable, again at high cost radiation exposure. The proposed limit precludes use of alternatives described in the draft NUREG revision which could be more cost effective in
- l. maintaining system integrity. For these reasons, we recommend that the five paragraphs in the draft generic letter under the heading Limits on Number of Cracked Weldsents in Piping be deleted in their entirety."
f Discussion:
l
- This proposed limit on the number of welds (regardless of repair or mitigation
- action) with cracks exceeding the Code size criterion for acceptability j without evaluation (IWB3500) was recommended by the ACRS in a memorandum, Ward to Stello dated March 18, 1986. Although staff reviewers felt that the limit was not necessary from a technical standpoint, and that placing such a limit on welds with cracks could be counter productive to the inspection program, the proposed limit was included for public comment at the suggestion of the I CRGR. The major difficulty that would be encountered if the limit were to be imposed is related to the fundamental problem that crack sizing is inherently l
l l
O L'
inaccurate in the size range of concern. According to the ACRS proposal, cracks of a size acceptable under IWB3500 of the Code would not be included in the limit criterion. This size is approximately 10% of the wall thickness, depending on the length of the crack. Thus, a weld with a crack only 9% of the wall in depth would not be counted against the 25% limit, whereas if it were to be reported as 11%, it would be. Although the technology for depth sizing has recently been developed to the point where the accuracy is acceptable for evaluational purposes, it is still not possible to consistently size to an accuracy of, for example, plus or minus 5% of the nominal wall thickness of BWR piping. Further, shallow cracks, in the range of 5 to 15% of the the wall, are even more difficult to size accurately than deeper cracks.
Thus, a crack actually about 10% of the wall in depth (just at the Code limit) could easily be sized anywhere from 5 to 15%. Without the propcsed 25% limit, conservative depth calls have no severe consequences. With a limit on the number of welds with cracks sized deeper than 10%, there certainly will be more reluctantance on the part of the examiner and Licensee to make conservative depth estimates. Staff reviewers believe that these considerations would mean that the imposition of a limit on the number of welds with cracks requir-ing evaluation could be counterproductive to the overall effectiveness of the inspection program.
Resolution:
The staff has concluded that adequate control of IGSCC can be achieved through the implementation of the Staff Positions in the Generic Letter without the need for a limit on number of cracked welds in the piping. Therefore, the portion of the Generic Letter pertaining to this limit has been deleted.
- 14. Comment:
The draft Generic Letter required that the inservice inspection plan be incorporated in the plant Technical Specifications. It was pointed out that this would be a lengthy addition to the Tech Specs, and because it would have to be updated very often, perhaps every outage in some cases, it would be inappropriate to include the detailed program in the Tech Specs.
Discussion:
The staff agrees that incorporating the inspection plan in the Technical Specification would be unweildy and necessitate excessive paperwork.
Nevertheless, the staff feels that linkage with the Tech Specs may be advisable from a regulatory standpoint.
Resolution:
The Generic Letter has been changed to require only that a statement be included in the Technical Specification to the effect that the inservice inspection program for piping covered under the Scope of the Generic Letter will be in compliance with the staff positions and Table 1 of the Generic l
Letter.
4
' l 1
l l
O INDIVIDUAL COW 4ENTS General Discussion:
There were many comments of an editorial nature that were incorporated into the final documents if deemed appropriate. There also were several requests for clarification, or comments apparently based on a mis-reading of the documents. These were handled by direct communication with the commenter, and usually did not warrant any modification of the text.
Specific Comments
- 1. Comment:
The reduction in augmentation of inspection schedules that may be allowed by the use of HWC should not result in less inspection than that required for.IGSCC Category A welds, which is the basic code requirement.
Discussion:
This was covered in the draft NUREG 0313 Rev. 2, and as written would not permit the situation of concern. -This detail is now also included in the Generic Letter.
O- aeeoiotioa:
Credit for HWC in reducing the frequency of augmented inspections is Iteited to welds in IGSCC categories B,C,D, and E by the Generic Letter.
- 2. Comment:
The recently developed SI method, Mechanical Stress Improvement Process (MSIP) should be considered to be fully effective, and the same credit for mitigation should be applied as that given IHSI.
Discussion:
Since the publication of the Draft documents, additional research, develop-ment, and confirmatory work has been done on the MSIP process. Confirmatory work at Argonne National Laboratory, funded by the NRC, is reported in a Research Information Letter, RIL 147 " Evaluation of Mechanical Stress Improvement Process". This RIL recommends that MSIP be considered a fully effective Stress Improvement process.
Resolution:
With the publication of the RIL referred to above, the staff has modified the Generic Letter (and NUREG) to include MSIP as a fully effective SI treatment.
O 1
/^s V
- 3. Comment:
There were several comments and suggestions related to evaluation of the extent of mitigation provided by weld overlays and IHSI, and the resulting degree of inspection augmentation.
Discussion:
The comments covered here all relate to whether a given weld should be classified as IGSCC Category E or F, with the proposed changes resulting in upgrading of some marginal welds to Category E. The staff reviewed the suggestions and determined that the present criteria should be retained.
Nevertheless, there could be situations for which less conservative criteria would be adequate. The staff believes that these might best be proven by service experience.
Resolution:
The staff has concluded that if an IGSCC Category F weld (which must be examined every outage) shows no adverse change through four refueling outages, it may be upgraded to IGSCC Category E. This has been detailed in the Generic Letter under Staff Position on Inspection Schedules.
p 4. Comment:
b It was requested by one commenter that IGSCC Category A weldments be excluded from the requirement that examiners be qualified to detect IGSCC.
Discussion:
As specifically indicated by the wording used throughout the Generic Letter and the NUREG, resistant materials, although resistant to IGSCC, are not immune. Clearly, one purpose of an inservice examination is to detect IGSCC should it occur as a result of inadvertant use of improper material, excessive weld repairs or similar problems. It is now realized from experience and shown by industry and NRC programs that unless proper examination methods are used by personnel specifically trained for the task, the examinations prescribed by the Code are almost useless for detection of IGSCC.
Further, the detection of transgranular stress corrosion cracks, and even the detection of fatigue cracks in austenitic stainless steel require the same degree of attention to detail and training.
The staff believes that it is not improbable that cracks may develop in some IGSCC Category A welds, and because the sample size (only 25% of the welds in 10 years) is small, the examinations that are performed should be effective.
Resolution:
() The requirements for training and qualification of examination personnel for those examining IGSCC Category A weldments will remain in the Generic Letter.
. ATTACHMENT 1 O Scope This Generic Letter applies to all BWR piping made of austenitic stainless
- steel that is four inches or larger in nominal diameter and contains reactor ;
coolant at a temperature above 200*F during power operation regardless of Code classification. It also applies to reactor vessel attachments and appurten-ances such as jet pump instrumentation penetration assemblies and head spray and vent components.
This Generic Letter does not apply to piping made of carbon steel classified
) as P-1 by the ASME Boiler and Pressure Vessel Code.
~
Staff Position on Materials i
Materials considered to be resistant to sensitization and IGSCC in BWR piping ~
! systems are:
. (1) Low carbon wrought austenitic stainless steel, which includes types 304L,
! 304NG, 316MG and similar low carbon grades with a maximum carbon content i of 0.035%. Type 347, as modified for nuclear use, will be resistant with somewhat higher carbon content, the usual maximum of 0.04% is adequate.
4 These materials must be tested for resistance to sensitization in accord- ,
ance with ASTM A262-A or -El or equivalent standard.
4 O c2) L- carbon -id -tai . inciudin. types >=. mL, 3m and siniiar grades, with a maximum carbon content of 0.0355 and a minimum of'7.5%
(or FN) ferrite may be sufficiently resistant, depending on carbon content and other factors. These will be evaluated on an individual case basis.
Welds joining resistant material that meet the ASME Boiler and Pressure Vessel Code requirement of 5% (or FN) ferrite, but are below 7.5%
t (or FN) may be sufficiently resistant, depending on carbon content and other factors. These will be evaluated on an individual case basis.
- - (3) Piping weldsents are considered resistant to IGSCC if the weld heat
- affected zone on the inside of the pipe is protected by a cladding of I
resistant weld metal. This is often referred to as corrosion resistant cladding (CRC).
(4) Cast austenitic stainless steel with a maximum of 0.035% carbon and a .
minimum of 7.5% (or FN) ferrite. Weld joints between resistant piping and cast valve or pump bodies that do not meet these requirements are !
[ considered to be special cases, and are covered in the Staff Position on j Inspection Schedules below.
i (5) Austenitic stainless steel piping that does not meet the requirements of
~
I (1) above is considered to be resistant if it is given a solution heat
! treatment after welding.
!O i -
O V (6) Other austenitic materials, including nickel base alloys such as Inconel 600, will be evaluated on an individual case basis. Inconel 82 is the only commonly used nickel base weld metal considered to be resistant.
It is the staff position that no austenitic material is resistant to cracking in the presence of a crevice, such as formed by a partial penetration weld, where the crevice is exposed to reactor coolant.
Staff Position on Processes The processes considered to provide resistance to IGSCC in BWR piping welds are:
(1) Solution Heat Treatment (SHT)
(2) Heat Sink Welding (HSW)
(3) Stress Improvement (SI)
- a. Induction Heating Stress Improvement (IHSI)
- b. Mechanical Stress Improvement Process (MSIP)
Although last pass heat sink welding (LPHSW) is not now considered to be fully effective, specific cases may be evaluated individually.
O Staff Position on Water Chemisir, The use of hydrogen water chemistry, together with stringent controls on conductivity, will inhibit the initiation and growth of IGSCC. However, the responses of BWRs to hydrogen injection differs from plant to plant, and the development and verification of a generic HWC specification is not yet complete. For these reasons, reductions in piping inspection frequency based on the use of HWC will be considered on an individual case bases at the present time. If fully effective HWC is maintained, a factor of two in reduction of inspection frequency may be justified for IGSCC Categories B, C, 0, and E weldments. (See Table 1)
Staff Position on Weld Overlay Reinforcement Cracted weldments that are reinforced with weld overlay are acceptable for short-term operation, and may be considered for longer ters operation provided:
(1) The overlayed weldments are in conformance with the criteria of IWB 3600 of Section XI of the 1986 Edition of the ASME Boiler and Pressure Vessel Code, and (2) they are inspected in conformance with the Staff Position on Inspection Methods and Personnel, by UT examiners and procedures qualified to inspect overlayed welds.
O
i O Staff
- sition on Partiai Re.iacement If portions of cracked piping are replaced in the course of repair, the replaced portions will be subjected to inservice inspection requirements that will depend on the materials and processes used. All relevant staff positions
, of this Generic Letter will apply.
Staff Position on Stress Improvement (SI) of Cracked Weldsents Stress Improvement is also considered to be an effective mitigation process when applied to weldsents with short or shallow cracks. Specifically, welds j with cracks that are no longer than 10% of the circumference, and are no deeper than 30% of the wall thickness will be considered to be mitigated by SI.
l SI is only considered to be effective if it is followed by a qualified UT examination, and if cracks are found, they must be sized both in depth and length, by procedures and personnel qualified to perform sizing evaluations:
Staff Position on Clampina Devices l
- Clamping devices may be used for temporary reinforcement of cracked weldsents.
Each case must be reviewed and approved on an individual basis.
Staff Position on Crack Characterization and Repair Criteria 3
O aethods and criteri < r cr c= ch r cteriaatioa and repair shouid 6e ia conformance with IWB-3600 of Section XI of the 1986 Edition of ASME Boiler
) and Pressure Vessel Code.
Because detailed sections of the Code are still under development, methods of analysis and acceptance criteria described in detail in NUREG-0313, Rev. 2 are ,
considered acceptable to the staff. '
Staff Position on Inspection Methods and Personnel I
Examinations performed under the Scope of this letter should comply with the applicable Edition and Addenda of the ASME Code,Section XI, as specified in
, paragraph (g), " Inservice Inspection Requirements" of 10CFR50.55a, codes and j Standards, or as otherwise approved by the NRC.
! In addition, the detailed procedure, equipment and examination personnel shall l' i be qualified by a formal program approved by the NRC such as that being 4 conducted in accordance with the NOE' Coordination Plan agreed upon by NRC, I
EPRI, and the Boiling Water Reactor Owners Group for IGSCC Research, being implemented at the EPRI N0E Center in Charlotte, North Carolina.
Staff Position on Inspection Schedules A summary of the Staff Position on Inspection Schedules is given in Table 1.
Additional details and definitions are provided below. NUREG-0313, Rev. 2, Section 5 provides background information and technical bases.
1 4
O (1) Weids of resistant materiai, IGSCC Cate.ory A, shouid as a minimum be examined according to an extent and frequency comparable to that intended by the applicable provisions of Section XI of the ASME Boiler and Pressure Vessel Code, which is a 25% sample every 10 years. The selection of specific welds to be included in this sample is the responsibility of the Licensee, and should include considerations of stress levels, piping configurations, weld details, etc, and should represent his best judgement regarding selection of a representative ,
and meaningful sample.
The provisions of 10CFR50.55a, (b),(2),(ii) may be invoked if it is determined necessary to use the 1974 edition of the Code to permit a meaningful sample selection.
(2) Although castings with higher carbon content than 0.035% are not considered to be resistant to sensitization, welds joining such castings (in the form of pump and valve bodies) to piping have been relatively free of IGSCC. This may be attributed to a favorable residual stress distribution, as calculations have indicated. For this reason, welds joining resistant material to pumps and valves will be considered to be resistant welds, and included in IGSCC Category A. If extensive weld repairs were performed the residual stress may be unfavorable, in which case such welds should be included in Category D.
(3) Welds that have been treated by SI or reinforced by weld overi n that are O classified as IGSCC Category F because they do not meet the applicable staff positions may be upgraded to Category E if no adverse change in crack condition is found after 4 successive examinations.
Staff Position on Sample Extension If one or more cracked welds in IGSCC Categories A, B, C, or 0 are found !
during an inspection, an additional sample of the welds in that category should be inspected, approximately equal in number to the original sample.
This additional sample should be similar in distribution (according to pipe size, system, and location) to the original sample, unless it is determined that there is a technical reason to select a different distribution. If any ,
cracked welds are found in this sample, all of the welds in that IGSCC Category should be inspected.
If significant crack growth or additional cracks are found during the inspection of an IGSCC Category E weld, all other Category E welds should be examined.
a) Significant crack growth for overlayed welds is defined as crack extension to deeper than 75% of the original wall thickness, or for cracks originally deeper than 75% of the pipe wall, evidence of crack l
growth into the effective weld overlay.
I i b) Significant crack growth for SI mitigated Category E welds is defined as growth to a length or depth exceeding the criteria for SI mitigation.
(either 10% of circumference in length or 30% of the wall in depth).
I i
O Q Staff Position on Leak Detection Leakage detection systems should be in conformance with Position C of Regulatory Guide 1.45 " Reactor Coolant Pressure Boundary Leakage Detection Systems, or as otherwise approved by the NRC.
- 1. Plant shutdown should be initiated or inspection and corrective action when, within any period of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> or less, any leakage detection system indicates an increase in rate of unidentified leakage in excess of 2 gpa or its equivalent, or when the total unidentified leakage attains a rate of 5 gpm or equivalent, whichever occurs first. For sump level monitoring systems with fixed-measurement-interval methods, the level should be monitored at approximately 4-hour intervals or less.
- 2. Unidentified leakage should include all leakage other than (a) leakage into closed systems, such as pump seal or valve packing leaks that are captured, flow metered, and conducted to a sump or collecting tank, or (b) leakage into the containment atmosphere from sources that are both specifically located and known either not to interfere with the operations of unidentified leakage monitoring systems or not to be from a throughwall crack in the piping within the reactor coolant pressure boundary.
- 3. For plants operating with any IGSCC Category D, E, F, or G welds, at least one of the leakage measurement instruments associated with each sump shall be operable, and the outage time for inoperable instruments shall be limited to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, or immediately initiate an orderly shutdown.
O
t TABLE 1
,)
SUMMARY
OF INSPECTION SCHEDULES FOR BWR PIPING WELDMENTS IGSCC INSPECTION !
DESCRIPTION OF WELDMENTS NOTES CATEGORY EXTENT & SCHEDULE Resistant Materials A 25% overy 10 years (at least 12% in 6 years)
Nonresistant Natis (1) 8 50% every 10 years SI within 2 yrs of (at least 25% in 6 years) operation (1)
Nonresistant Natis (1) C All within the next 2 refueling SI after 2 yrs of cycles, then all every 10 years operation Non Resistant Matl (1) D All every 2 refueling cycles No SI Cracked (1)(2) E 50% next refueling outage, then Reinforced by weld overlay all every 2 refueling cycles or mitigated by SI Cracked (2) F All every refueling outage Inadequate or O no repair Non Resistant (3) G All next refueling outage Not Inspected Notes:
(1) All welds in non-resistant material should be inspected after a stress improvment process as part of the process. Schedules shown should be followed after this initial inspection.
(2) See recommendations for acceptable weld overlay reinforcements and stress improvement mitigation.
(3) Welds that are not UT inspectable should be replaced, " sleeved", or local leak detection applied. RT examination or visual inspection for leakage may also be considered.
O v