ML051320232
| ML051320232 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse, Oconee, Arkansas Nuclear, Crystal River, PROJ0693, Crane  |
| Issue date: | 03/25/2005 |
| From: | Crawford H, Holm J B & W Owners Group |
| To: | Justin Fuller Document Control Desk, Office of Nuclear Reactor Regulation |
| References | |
| BAW-2308 (NP), NRC:05:019, OG:05:1869 | |
| Download: ML051320232 (6) | |
Text
Duke Energy Corporation Oconee 1.2.3 AmerGen Energy Company. LLC TMI-1 Entergy Operations. Inc.
ANO-ITho B&W FirstEnergy Nuclear Operating Company D-B Progress Energy, Florida Crystal River3 Owners Group FramatorneANP. Inc. (FANP)
Working Together to Economically Provide Reliable and Safe Electrical Power March 25, 2005 NRC:05:019 OG:05: 1869 Document Control Desk U.S. Nuclear Regulatory Commission Washington, D.C. 20555-0001 Supplemental Material for Review of BAW-2308(NP), Revision 1, "Initial RTflDT of Linde 80 Weld Materials" Ref. 1: Letter, James F. Mallay (Framatome ANP), to Document Control Desk (NRC), 'Request for Approval of BAW-2308(NP), Revision 1, 'Initial RTtDT of Linde 80 Weld Materials',"
NRC:03:052, OG:03:1835, August 19, 2003.
On behalf of the B&W Owners Group Reactor Vessel Working Group, Framatome ANP (FANP) requested the NRC's review and approval for referencing in licensing actions the topical report BAW-2308, Revision 1, 'Initial RTNDT of Linde 80 Weld Material" (Reference 1).
In a December 2004 conference call with the NRC, the staff stated a concern over the continued use of RTTO =
To + 35 F. This concern arose principally from the evaluation of a large population of laboratory test specimen fracture toughness data which showed more data lying below the Code Case(s) N-629/N-631 curve than the ASME Kk curve. The basic concern is whether RTTO, as defined by the Code Case(s) N-6291N-631, makes RTTO functionally equivalent to RTNDT. This issue is addressed in Attachment A under the heading 'Heat Specific RTT,.'
In a conference call with the NRC on March 4, 2005, the NRC stated that an additional 201F bias is needed for the generic Linde 80 RTTO to maintain functional equivalence. The B&WOG Reactor Vessel Working Group is willing to accept an additional 206F bias for the generic Linde 80 weld RTT. as described in Attachment A to this letter under the heading 'Generic Linde 80 RTTO."
The timely completion of the review of this topical report is appreciated.
Sincerely, Jerald S. Holm, Director Howard Crawford, Chairman Regulatory Affairs B&W Owners Group Steering Committee Framalome ANP. Inc. B&W Owners Group 3315 Old Forest Road Lynchburg, VA 24501 Phone: 434-832-3635 Fax: 434-832-4121
Enclosures cc:
D.G. Holland Project 693 Reactor Vessel Working Group
Document Control Desk NRC 05O01910G:05:1869 March 25, 2005 Page A-1 Attachment A Heat Specific RTTO A functional equivalence comparison of RTTO relative to RTNDT can be made by comparing the number of points that lie below the ASME K1c curve with the number of points that lie below the Code Case N-629 curve for a large dataset. A comparison is shown in Figure 1 (from WRC Bulletin 457' ). The comparison shows that about 7 points, in the transition temperature range, lie below the ASME Kg, curve out of 1509 (0.5%). These RTNoT values are measured values. The same data can be adjusted to 1T equivalent toughness, plotted against T-To and compared to the ASME Code Case N-629 curve. The comparison in Figure 2 (also from WRC Bulleting 457) shows that significantly more data lie below the Code Case N-629 curve, about 35 points (2.3%). A comparison with the measured toughness (without size adjustment), shows that about 40 points (2.6%) fall below the Code Case N-629 curve (See Figure 3). The ASME Kk curve and the ASME Code Case N-629 are not functionally equivalent when compared on this basis.
There is a difference, however, in the application of the data. When using measured RTNDT values per 10CFR50.61, 0 (zero) initial margin is used (a,). Since the ASME Kt is a lower bounding curve, an initial margin is already built-in. When applying RTTO to PTS evaluation, the NRC set a precedent of including an initial margin as established in the Kewaunee SER2. This initial margin is based on data variation using a Monte Carlo analysis. A Monte Carlo analysis is also used in BAW-2308, with the addition of ASTM El 921-97 To sample size standard deviation to the initial margin term. The heat specific a, values in BAW-2308 vary from 1OF to 13F compared to the a, of 14F in the Kewaunee SER.
Based on the precedent set in the Kewaunee SER, the NRC will likely not accept a master curve submittal with cg = 0. The comparison in Figures 2 and 3 are with a, = 0.
This does not reflect the actual application which includes an initial margin term. Figure 4 compares the 1T equivalent toughness data to the Code Case N-629 + 2a, curve, where a, is 1OF (the lowest heat specific a, value in BAW-2308, Revision 13). About 9 points lie below this curve or 0.6%. This is functionally equivalent to Figure 1 (RTNDT)-
In BAW-2308, Revision 1, Code Case N-629 combined with the initial margin term provides functional equivalence with the RTNOT based 1 OCFR50.61 for the seven heats for which test data is presented.
'VanDerSluys. W. A., et al.. "Fracture Toughness Master Curve Developmenl: Fracture Toughness of Ferrilic Steels and ASTIM Reference Temperature (TO)," Welding Research Council Bulletin 457, Dcccmbcr 2000.
2US Nuclear Regulatory Commission, ".Sqfety Evaheation 1Xy the Office of Niular Reactro RegulaItioi Regardkiqg Antednent of the Kewaunee Nuclear Power Plant License to liclule die Use of at Aflaster CLurve-Blasedl ehimodologyfor Reactor Pressure i essel Ilntegrir iAssessment:, May 2001.
3 Yoon, K. K., "Initial RTNi,r of Linde 80 W'eld Materials," BAW-2303, Revision 1. August 2003.
Document Control Desk NRC:05 01910G:05:1869 March 25, 2005 Page A-2 Generic Linde 80 RTT 10CFR50.61 states that when a measured value of RTNDT is unavailable a generic mean value for the material class may be used if there are sufficient test results to establish a mean and standard deviation. BAW-2308 establishes a generic mean RTT7 and a standard deviation for the Linde 80 class of materials based on the fracture toughness data on 7 Linde 80 heats. This was done in a manner consistent with the requirements of 10CFR50.61. Therefore, there is no extra initial margin that would give the same functional equivalence as in the heat specific case. For the generic Linde 80 RTTO, it may be prudent (and it is acceptable to the B&WOG RVWG) to add an additional 20°F margin to obtain functional equivalence with 10CFR50.61.
ALL DATA 1509 DATA POINTS 600_ -ASME KIc a 1: 72W X
A 2:73W x 3: A53381 x
4: A533B1 t
5: PXle 13A x
X 500
- 6: GKSS o 8: WF70 Benline
- 9: WF70 Nozzle 10:A508d3 x
o 11:A508c13
- 12:A50803MPC RRkx x
I x
o 13:A508d3A
- 14:A508d3BX x x
C 400 M 15:A508ci2 A 16:A533Bcl1 X xx X
X 17. A508c11 A 18: HSST02 X
xx x
o 19:A533 Bd1l
- 0:A533Bdt10 X
0 30 o 21:A533c11 X
M:A53I I
- X a
300 23:A508cI3 x O
- 4508*c J I D
E zoo 1i Jo 100 50
-150
-100
-50 05 0
5 T-RTNDT, r Figure 1: 0.5% of data lie below K,c curve
Document Control Desk March 25, 2005 NRC:05:O191OG:05:1869 Page A-3 STATIC DATA, 1525 POINTS 600 500 C
,,, 400 0) i_ 300
- ~200 C.
I-100 00
-150
-10 D 50 0
so 100 150 200 T-Ta. F Figure 2: 2.3% of data lie below the Code Case N-629 curve ALL STATIC DATA, 1528 POINTS 600 500 c 400 c 300 I 200 100 o _
-150
.100 0
100 150 200 T.To. F Figure 3: 2.6% of data lie below the Code Case N-629 curve
Document Control Desk March 25, 2005 NRC.05:0190G:05:1869 Page A-4 STATIC DATA, 1528 POINTS 600 500 C
v, 400
.)c C
300 i= 200 C.
LU 100 0
-150
-100
-50 0
50 100 150 200 TITo, F Figure 4: 0.6% of data lie below the Code Case N-629 + 2a, curve where 2a, = 20F