ML20211E627
| ML20211E627 | |
| Person / Time | |
|---|---|
| Site: | Brunswick  |
| Issue date: | 06/10/1986 |
| From: | Zimmerman S CAROLINA POWER & LIGHT CO. |
| To: | Muller D Office of Nuclear Reactor Regulation |
| References | |
| NLS-86-189, NUDOCS 8606160204 | |
| Download: ML20211E627 (4) | |
Text
c
~
C9&L Carolina Power & Light Company SERIAL: NLS-86-189 Director of Nuclear Reactor Regulation Attention:
Mr. Dan Muller, Director BWR Project Directorate #2 Division of BWR Licensing United States Nuclear Regulatory Commission Washington, DC 20555 BRUNSWICK STEAM ELECTRIC PLANT, UNIT NO. 2 DOCKET NO. 50-324/ LICENSE NO. DPR-62 ADDITIONAL INFORMATION CONCERNING IGSCC INSPECTION RESULTS - BRUNSWICK-2 RELOAD 6
Dear Mr. Muller:
On May 16 and May 20,1986, conference calls were held with members of your staff to discuss the results of the Brunswick-2 IGSCC inspections performed during the Reload 6 outage. Enclosure i provides documentation of those discussions.
Please refer any further questions regarding this matter to Mr. Stephen D. Floyd at (919) 836-6901.
Yours very truly, huffettan S.
. Z' imerman l
anager Nuclear Licensing Section l
SRZ/ MAT /crs (3933 MAT)
Enclosure cc:
Mr. W. H. Ruland (NRC-BNP)
Dr. J. Nelson Grace (NRC-RII)
Mr. E. Sylvester (NRC) i l
8606160204 860610 PDR ADOCK 05000324 G
PDR 411 Fayettevitte Street
- P O Box 1551
- Raleigh. N C 27602 hh N/
t
ENCLOSURE 1 TO SERIAL NLS-86-189 QUESTION 1 The May 8,1986 submittal states that the entire Class 1,6-inch RWCU system piping was replaced from the RHR system tie-in to the outboard primary containment isolation valve with nuclear grade stainless steel piping. What is the nuclear grade of this piping?
RESPONSE
The 6-inch RWCU system piping was replaced with 316 control chemistry (.02 percent maximum carbon,.06 to.1 percent nitrogen) stainless steel piping.
QUESTION 2 How many cracked welds were found during the pre-lHSI examination? How many cracked welds were found during the post-lHSI examination? How many of the post-IHSI cracked welds were also determined to have cracks during the pre-lHSI examination?
RESPONSE
l Forty-eight welds were examined prior to performing IHSI,28 of which were found to l
contain cracks. Twenty-seven of the cracked welds were then overlayed with no IHSI performed. The IHSI treatment was performed on the remaining weld before repair by weld overlay. Four additional welds were found to have cracks during post-lHSI examinations and these welds were subsequently overlayed.
QUESTION 3 There are only seven 12-inch inconel butter welds listed in Table 4 of the May 8 submittal; are there actually ten 12-inch inconel butter welds? If so, provide a justification for not inspecting the additional three welds.
RESPONSE
j There are ten 12-inch inconel butter welds. A 20 percent sample was selected from each pipe size due to the high radiation levels near the reactor nozzles. The original 12-inch sample did not yield IGSCC indications. An additional five 12-inch welds were also examined and found to be free of IGSCC. The remaining three 12-inch welds were not examined due to ALARA considerations.
QUESTION 4 With regard to the two 28-inch suction nozzle inconel weld butters containing the shallow axial indication, provide information concerning the sizing of the axial cracks.
RESPONSE
The 28-inch nozzle indications were sized using EPRI supplied 1 MHz 45,60, and 70 degree refracted longitudinal transducers using the TIP diffraction technique. Manual re-examinations were made with eight refracted longitudinal transducers ranging from 31 to 70 degrees, a 0 degree longitudinal transducer, and 45 degree shear wave transducer.
(3933 MAT /pgo )
The indications were sized by two GE Level III inspectors. Comparisons were made between a 10 percent notch in the mockup inconel butter and the two 28-inch nozzle butters using a Parametric 2002 Epoch and CRT printer. These comparisons confirmed the sizing technique.
QUESTION 5 Explain Note 8 on page D.2 of Appendix D of the NUTECH Report.
RESPONSE
Note 8 indicates that the as-built dimensions listed in Attachment D of the NUTECH Report do not include the first layer overlay thickness except for the 28 inch welds.
These 28 inch pipe weld joints were repaired with leak barrier rather than full structural overlays. CP&L believes that two layer overlays are adequate for leak barrier protection. This was identified as an exception to the guidelines of Generic Letter 84-11 and the staff concurred with the Company's position.
QUESTION 6 What crack growth equation was used for the crack growth analysis described on page 2 of the GE Fracture Mechanics Analysis?
RESPONSE
IGSCC crack growth data for Inconel 182 weld material have been developed by GE. The data indicates that atpigh stress intensity factors, the growth rate plateaus to a value of approximately 5 x 10- in/hr. The weld residual g). stress at the weld creates a large stress intensity factor at the crack tip (>50 ksi in Thus the plateau growth rate is applicable and was assumed in the analysis for the 12,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> of further operation.
(da/dt)IGSCC x time + a nitial = afinal (5 x 10-5 in/hr)(12,000 hrs) + 0.25 inch = 0.85 inch QUESTION 7 Provide an explanation of the following statement taken from page 2 of the GE Fracture Mechanics Analysis: "The normal operation sustained hoop stresses at the location of the indications consists of 9 ksi pressure and approximately 70 ksi weld residual. Using a combined stress of 80 ksi, stress intensity factors were calculated by the method outlined in Appendix A to Section XI."
RESPONSE
The stresses which contribute to IGSCC crack growth of an axial crack in a pipe are the normal operating sustained hoop stresses. The only primary hoop stress is created by pipe pressure. For the analysis, the design pressure of 1250 psi was conservatively assumed, creating a hoop stress of approximately 9 ksi. Under normal operation, secondary thermal hoop stress is essentially negligible. The other sustained stress is the weld residual stress. For large diameter pipes, finite element studies have been performed by GE to determine the weld residual stress distributions in dissimilar stainless steel-to-low alloy steel full penetration welds with Inconel 182 butter. The hoop stress in the butter was determined to be approximately 70 ksi. The combined (3933 MAT /pgp )
sustained hoop stress is, therefore, (9 + 70) or approximately 80 ksi. This stress was used i
with the initial flaw size of 0.25 inch to determine the applied stress intensity factor (K ). The formulation for K which was used is described in Appendix A to Section XI of g
g the ASME Code, for part through flaws.
QUESTION 8 Section (a) on page 3 of the GE Fracture Mechanics Analysis discusses the linear elastic fracture ghanics assessment. Where did the available crack arrest toughness of 200 ksi in come from?
RESPONSE
Figure A-4200-1 of Appendix A to Section XI of the ASME Code provides conservative lower bound crack arrest toughness for low alloy steel as a function of temperature. The RT f r the low alloy material is on the order of 50*F. Therefore, the T-RTNDT NDT durmg normal operation will be approximately 500*F.gis is beyond the range ot 4
Figure A-4200-1; however, a value of KIA = 200 ksi in is certainly conservative, and was therefore used in the leak before break analysis.
QUESTION 9 The limit load assessment in the GE Fracture Mechanics Analysis concluded that even if I
crack extension into the low alloy steel nozzle occurs, a significant leak before break margin will be maintained. How was this analysis performed? What table was used for
. the low alloy steel?
RESPONSE
The evaluation of the indications in the Inconel 182 butter considered the flaw to be in i
the austenitic safe end and in the low alloy steel nozzle. For the safe end, compliance with Table IWB-3641-3 was demonstrated while for the low alloy nozzle, sizable leak before break margin was calculated.
i l
The ASME Code Section XI does not provide an acceptance criteria for ferritic steel components less than 4 inches in thickness and the criteria for components greater than 4 inches in thickness (IWB-3610) is too conservative. Therefore, the leak before break margin was assessed by considering two failure mechanisms: brittle fracture and limit load (gross plasticity). The limit load calculation made use of the expression for gross plastic yielding of a cylinder with an axial through wall crack. This expression is shown on page 3 of the GE Fracture Mechanics Analysis, and was obtained from the Battelle Columbus Laboratories Report BMI-1908, June 1971, " Investigation of the Initiation and Extent of Ductile Pipe Rupture," R. J. Eiber. The hoop stress was set equal to the l
pressure stress of 9 ksi(away from the weld there would be no residual stress), and the equation was solved for half crack length "a". The limit load critical flaw size of 56 inches is greater than the brittle fracture critical flaw size of 30 inches. There is a i
considerable leak before break margin by either failure mode. It should be noted that IGSCC mechanisms would not extend the crack into the low alloy nozzle.
I I
(3933NAT/pgo) l
- _, - - -- _ - - - _._