RS-11-173, Response to Request for Additional Information Related to Relief Request I3R-08

Response to Request for Additional Information Related to Relief Request I3R-08

ML113070349
Person / Time
Site:	Braidwood
Issue date:	11/02/2011
From:	Gullott D Exelon Nuclear, Exelon Generation Co
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
RS-11-173
Download: ML113070349 (6)

Text

Exelon Generation Company, LLC www.exeloncorp.com 4300 Winfield Road Nuclear Warrenville, IL 60555 RS-1 1-173 10 CFR 50.55a November 2, 2011 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-001 Braidwood Station, Units 1 and 2 Facility Operating License Nos. NPF-72 and NPF-77 NRC Docket Nos. STN-456 and STN-50-457

Subject:

Response to Request for Additional Information Related to Braidwood Station Relief Request 13R-08

References:

1. Letter from J. L. Hansen (Exelon Generation Company, LLC) to U. S. Nuclear Regulatory Commission, "Third 10-Year Inservice Inspection Interval Relief Request 13R-08, Alternative Requirements to ASME Section XI Appendix VIII (Supplements 2 and 10), Examinations of Class 1 Pressure Retaining Welds Conducted from the Inside Surface In Accordance with 10 CFR 50 .55a(a)(3)(i),"

dated April 11, 2011

2. Letter from J. L. Hansen (Exelon Generation Company, LLC) to U. S. Nuclear Regulatory Commission, "Supplemental Information Supporting Relief Request 13R-08: Alternative Requirements to ASME Code Requirements For Class 1 Pressure Retaining Welds," dated June 6, 2011

3. Letter from N. J. DiFrancesco (U. S. Nuclear Regulatory Commission) to M. J. Pacilio (Exelon Generation Company, LLC), "Braidwood Station, Units 1 and 2 - Request For Additional Information Re: Request for Relief 13R-08 (TAC Nos. ME6024 and ME6025)," dated October 12, 2011 In References 1 and 2, Exelon Generation Company, LLC, (EGC) submitted a request for relief, 13R-08, Revision 0, to the U. S. Nuclear Regulatory Commission (NRC) for review and approval.

This alternative requested relief from certain examination qualification requirements of the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code,Section XI.

EGC proposed the use of root mean square (RMS) error criteria for sizing flaws that are greater than the requirements of ASME Code Case N-695, "Qualification Requirements for Dissimilar Metal Piping Welds," and N-696, "Qualification Requirements for Appendix VIII Piping Examinations Conducted From the Inside Surface."

In Reference 3, the NRC requested additional information to complete the review of the proposed relief request. In response to this request, EGC is providing the attached information. The Attachment provides the requested responses.

November 2, 2011 U. S. Nuclear Regulatory Commission Page 2 There are no regulatory commitments contained in this submittal. Should you have any questions concerning this letter, please contact Ms. Lisa A. Simpson at (630) 657-2815.

Respectfully, David M. Gullott Manager - Licensing Exelon Generation Company, LLC

Attachment:

Response to Request for Additional Information cc: NRC Regional Administrator, Region III NRC Senior Resident Inspector - Braidwood Station NRR Project Manager - Braidwood Station Illinois Emergency Management Agency - Division of Nuclear Safety

ATTACHMENT Response to Request for Supplemental Information By letters dated April 11, 2011, and June 6, 2011, Exelon Generation Company, LLC, (EGC) submitted a request for relief, 13R-08, Revision 0, to the U. S. Nuclear Regulatory Commission (NRC) for review and approval. This alternative requested relief from certain examination qualification requirements of the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code,Section XI. EGC proposed the use of root mean square (RMS) error criteria for sizing flaws that are greater than the requirements of ASME Code Case N-695, "Qualification Requirements for Dissimilar Metal Piping Welds," and N-696, "Qualification Requirements for Appendix VIII Piping Examinations Conducted From the Inside Surface."

In a letter dated October 12, 2011, the NRC requested additional information to complete the review of the proposed relief request.

NRC Question 1:

The large differences in RMSE between non-Appendix VIII DMWs, Appendix VIII DMWs, and Appendix VIII austenitic-to-austenitic welds are influenced by changes to the surface conditions in the test mockups used to demonstrate capability, effectiveness, and skill. Unless the surface conditions of the field welds are exactly the same as test mockups, the applicability of a specific RMSE value may not be appropriate. Correlations between the essential variable "surface preparation" and RMSE values have not been developed. Therefore, the "best effort" RMSE values assigned by the Electric Power Research Institute (EPRI) - Performance Demonstration Initiative (PDI), which are referenced in the submittal for DMW and austenitic-to-austenitic welds, may not be appropriate depending on the relationship of RMSE and actual surface preparations (conditions). In the absence of any correlation between RMSE and surface conditions, the "best effort" PDI RMSE values are questionable with respect to reliability, reproducibility, and applicability to actual depth sizing of cracks identified during an examination.

Provide a discussion on the following two probe-to-component surface contact conditions:

(a) Based on the non-Appendix VIII vendor performance demonstrations performed on surfaces that are capable of maintaining good probe-to-component surface (defined by the EPRI Performance Demonstration Initiative as 1/32-inch gap or less) contact, the vendor's RMSE data suggest a reasonable probability of reproducibility. For surface conditions similar to those associated with the non-Appendix VIII performance demonstration, discuss a conservative approach for using RMSE values that would apply to both DMWs and austenitic-to-austenitic welds, i.e., one "best effort" RMSE value.

(b) In examinations performed during the last 10-year inservice inspection (ISI) interval, the licensee's vendor recorded no flaw (crack) indications for the subject welds. These examinations were performed with a combination of ultrasonic testing (UT) and eddy current testing (ET) methods. For upcoming examinations, any cracks that are detected would have occurred between the last ISI examination and upcoming examination. The most likely cracking mechanism (identified in the licensee's risk-informed program) is primary water stress corrosion cracks (PWSCC). In the event that the UT or ET examination detected PWSCC, UT depth sizing performed on surfaces exhibiting gaps greater than 1/32-inch between probe and component surface may be less conservative than the proposed alternative in 1 a, above. Because sizing accuracy is unknown for gaps greater that 1/32-inch, the application of RMSE as a tolerance to adjust crack depth is questionable. As an alternative to the adjustment approach, cracks that are Page 1 of 4

ATTACHMENT Response to Request for Supplemental Information depth sized on rough surfaces may be bounded by calculating crack growth from the last crack free examination. For calculating purposes, the initial (at the time of the crack free examination) crack depth should be assumed 5-percent through-wall (the smallest crack size that may be in a performance demonstration test). For PWSCC, the bounding crack depth can be calculated using the guidance provided in the EPRI -Materials Reliability Program (MRP) MRP-1 15, MRP-252, and/or MRP-287, as appropriate, for representative DMW and austenitic-to-austenitic welds. Provide a discussion on the disposition of cracks as determine by calculating the maximum through-wall crack depth and identify the associated assumptions, if any.

Response to Question 1(a):

The non-Appendix VIII performance demonstrations described in the June 6, 2011 letter (ADAMS ML111580106) were only associated with dissimilar metal welds having relatively smooth surfaces. As such it is not practical to extend such results (to develop a combined 'best effort' RMSE value) to an austenitic stainless steel weld configuration that is field welded and is comprised of counter-bores, localized manual grind-outs and weld shrinkage.

The most conservative approach for one 'best effort' RMS error value remains that developed in the combined Appendix VIII Supplement 2 and 10 qualifications. For EGC's inspection vendor, this combined RMS error is 0.245-inch. This value includes inspection results on dissimilar metal weld specimens having a mixture of relatively smooth surfaces associated with shop welds, and non-flat surfaces associated with field welds, and similar metal austenitic stainless steel weld specimens having non-flat surfaces. Such non-flat surfaces are considered to be greater than the 1/32-inch criteria at various points across the weld as assumed given the condition of open test specimens supposedly representative of the blind test specimens. The non-flat surfaces include an approximate 1/4-inch fit-up mismatch, counter-bores, localized grindouts, and weld shrinkage. The increase in RMS error from 0.189-inch for dissimilar metal welds to 0.245-inch for combined dissimilar and similar metal welds is most likely associated with the addition of more UT results from less flat examination surfaces since the applied UT techniques were the same.

The EGC ISI vendor has demonstrated the ability to depth size flaw indications in dissimilar metal welds with a RMS error of 0.189-inch instead of the 0.125-inch RMS error required by Appendix VIII Supplement 10 (Code Case N-695) and an RMS error of 0.245-inch instead of the 0.125-inch RMS error required by the combined Appendix VIII Supplement 2 and 10 qualification (Code Case N-696).

Response to Question 1(b):

Calculating a maximum through-wall crack depth based on guidance in MRP-1 15, MRP-252, and MRP-287 would not yield realistic results when starting with such a shallow flaw. Such calculations rely upon approximations and assumptions used in the residual stress analysis that have been shown to produce large variations in the through-wall residual stress profiles between analytical and experimental results. Uncertainty in the actual PWSCC crack growth rate and the actual crack growth behavior or morphology may not yield realistic flaw depth. The assumption of a 5% through-wall flaw may also be unrealistic given the application of eddy current techniques that although not performance demonstrated in accordance with the ASME Page 2 of 4

ATTACHMENT Response to Request for Supplemental Information Code have been demonstrated in Sweden for the detection of 0.04-inch deep by 0.25-inch long planar flaws (approximately 1 - 2% through-wall).

UT techniques can determine the through-wall size of planar flaws as evidenced by the combined RMS error for dissimilar and similar metal welds demonstrated in accordance with Appendix VIII, Supplement 14. The test specimens consisted of surface conditions representative of shop and field weld configurations. While the RMS error value is not consistent with the established ASME Code standard, it still represents a reasonable measurement of flaw through-wall depth albeit with a greater uncertainty. However this uncertainty is to be factored into the flaw size which would be used in fracture mechanics analyses in accordance with the proposed alternative in the April 11, 2011 letter (ADAMS ML111020263).

NRC Question 2:

The vendor participated in three non-PDI performance demonstrations on mockups with smooth ID surfaces (probe-to-component surface gaps of less than or equal to 1/32-inch). Each of the demonstrations used UT procedures and equipment nearly identical to those to be applied in the upcoming Braidwood, Units 1 and 2, examinations. The personnel performing the Appendix VIII examinations are required to be qualified on the specific UT procedure that will be used for the examinations. Please provide a discussion on the participation in the non-PDI performance demonstrations of the UT ultrasonic testing personnel piping qualifications that will be performing the upcoming Braidwood, Units 1 and 2, examinations.

Response to Question 2:

EGC's inspection vendor has a total of five (5) current UT analysts that have achieved consistent through-wall depth sizing results within a RMS error of 0.189-inch for dissimilar metal welds, and a RMS error of 0.245-inch for combined dissimilar metal and austenitic stainless steel welds. A subset of these UT analysts will be used for the Braidwood Unit 1 and 2 examinations.

Three of these individuals directly participated in the three non-PDI performance demonstrations related to the UT procedure (see Response to Question 3(b) ADAMS ML111580106). Two of these three individuals also participated in and successfully completed blind personnel qualifications conducted by the Swedish Qualification Centre (SQC) using the qualified procedure.

The ultrasonic test procedures and equipment demonstrated in both the PDI performance demonstrations, and the non-PDI performance demonstrations (specifically the SQC-qualified and PDI-qualified versions) are nearly identical and were developed in the same time period.

The only variations are associated with slightly different inspection performance objectives however the through-wall depth sizing logic is the same. All of the UT analysts are trained on the PDI demonstrated procedure using test data from non-PDI performance demonstrations and other investigations on cracked samples.

Page 3 of 4

ATTACHMENT Response to Request for Supplemental Information NRC Question 3:

To ensure that the examination method when performed will provide reasonable assurance of structural integrity. In the June 6, 2011 letter, the response to question 1(c), starts a sentence with, "The estimated lack of coverage for these welds...." Please provide the estimated numerical coverage values, if they exist.

Response to Question 3:

Question 1(c) in the June 6, 2011 letter, was associated with the application of OD techniques.

The phrase'The estimated lack of coverage for these welds...' referred to all Braidwood Unit 1 and 2 inlet and outlet nozzle to safe end welds. Figures 1-1 and 1-2 provided an example representative of Braidwood Unit 2 'A' inlet and outlet nozzle to safe end welds. Using this example, the estimated numerical coverage values are:

• Braidwood Unit 2 'A' Inlet Nozzle to Safe End Weld (Figure 1-1 in the June 6, 2011 letter) - 63% coverage for circumferential flaws (essentially all in the PWSCC susceptible material); 92% coverage for axial flaws (all in adjacent base metal)

• Braidwood Unit 2 'A' Outlet Nozzle to Safe End Weld (Figure 1-2 in the June 6, 2011 letter) - 94% coverage for circumferential flaws (near the buttering/carbon steel nozzle interface); 83% coverage for axial flaws (all in adjacent safe end material)

Such values represent the contour at one location of the weld. Variations may exist from area to area. Surface conditioning would be required in order to achieve the estimated coverage given above.

Page 4 of 4