ML18153A332

Forwards Results of Failure/Metallurgical Analysis,As Committed to in Licensee 980327 Ltr.Results Include Structural Integrity Evaluation of RHR Piping.List of Util Commitments,Encl

ML18153A332
Person / Time
Site:	Surry
Issue date:	09/21/1998
From:	Hartz L VIRGINIA POWER (VIRGINIA ELECTRIC & POWER CO.)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
98-177B, NUDOCS 9809290153
Download: ML18153A332 (9)

Text

'r

• e e VIRGINIA ELECTRIC AND POWER COMPANY RICHMOND, VIRGINIA 23261 September 21, 1998 United States Nuclear Regulatory Commission Serial No. 98-177B Attention: Document Control Desk NL&OS/GDM R2 Washington, D. C. 20555 Docket No. 50-280 50-281 License No. DPR-32 DPR-37 Gentlemen:

VIRGINIA ELECTRIC AND POWER COMPANY SURRY POWER STATION UNIT 1 RESIDUAL HEAT REMOVAL PIPING LEAK EVALUATION REPORT In a letter dated March 23, 1998 (Serial No.98-177), Virginia Electric and Power Company requested relief from certain ASME Section XI requirements associated with the repair of a through-wall leak in the Residual Heat Removal System piping. In a subsequent letter dated March 27, 1998 (Serial No. 98-177A), we committed to

1) perform a failure analysis of the Residual Heat Removal (RHR) System pipe leak,*

including a metallurgical analysis of a pipe sample taken from the leak area, and

2) provide the results to the NRC. The failure/metallurgical analysis has been completed and the results are provided in the attachment. Our assessment also includes a structural integrity evaluation of the RHR piping considering the results of the failure/metallurgical analysis.

It has been determined that the RHR piping leak resulted from intergranular attack in the pipe to flange weld heat affected zone that had progressed to the point that a leak path was formed. The area of the leak was localized to the heat affected zone with no discernable indication of any through-wall flaw that would be considered a discrete crack.

As discussed in the attachment, the RHR system is considered fully operable and capable of performing its *design function:** *However,* dueto *the particular conditions observed at the leak location and for long term assurance of reliability, that portion of RHR line 12"-RH-19-602 from the affected weld to the next upstream weld will be replaced during the next Unit 1 refueling outage, currently scheduled for October 1998.

The condition of the removed piping will be evaluated to determine further corrective action. -- '. k\ n '..'°\, "'1, ,,..

~l

(-------------.---~

9809290153 980921 PDR ADOCK 05000280 P PDR .

Visual walkdown inspections to detect any boric acid accumulation and/or unidentified leakage will continue to be conducted on both units at the current frequency of each scheduled cold

• shutdown. Furthermore, in addition to the required ASME Code Section XI pressure testing performed each Code period, portions of the RHR ASME Class 2 600 pound stainless steel piping will be placed in a supplemental nondestructive examination (NOE) program for both units. This program will select welds for examination at a frequency three times that required by the ASME Code for Class 2 piping, and the NOE inspections performed on the selected welds will be the same as that required for Class 2 piping.

If you have any questions or require additional information, please contact us.

Very truly yours, C>>f514r;-

L. N. Hartz Vice President - Nuclear Engineering and Services Commitments contained in this letter:

1. That portion of RHR line 12"-RH-19-602 from the affected weld to the next upstream weld will be replaced during the next Unit 1 refueling outage, currently scheduled for October 1998.

2. The condition of the removed piping will be evaluated to determine further corrective action.

3. Visual walkdown inspections to detect any boric acid accumulation and/or unidentified leakage will continue to be conducted on both units at the current frequency of each scheduled cold shutdown.

4. Portions of the RHR ASME Class 2 600 pound stainless steel piping will be placed in a supplemental nondestructive examination (NOE) program for both units. This program will select welds for examination at a frequency three times that required by the ASME Code for Class 2 piping, and the NOE inspections performed on the selected welds will be the same*as that required*for Class 2 piping.

cc: United States Nuclear Regulatory Commission Region II Atlanta Federal Center 61 Forsyth Street, SW, Suite 23T85 Atlanta, Georgia 30303 Mr. R. A. Musser NRG Senior Resident Inspector Surry Power Station

• ATTACHMENT e

RESIDUAL HEAT REMOVAL SYSTEM PIPING LEAK EVALUATION REPORT SURRY POWER STATION UNIT 1 Introduction On March 22, 1998, minor leakage was detected on the Unit 1 Residual Heat Removal (RHR) Class 2 piping. The leak was identified in line 12"-RH-19-602 upstream of the flow element 1-RH-FE-1605 and downstream of the RHR heat exchangers' discharge piping in the common header that feeds both Loops B and C in the Reactor Coolant System cold legs. A detailed characterization of the through-wall flaw could not be performed by available ultrasonic (UT) technique, and no flaw was determined by UT measurement at the leak location. A flaw size of 1/32" was conservatively estimated based on visual observation, since there was no clear indication of a crack-like flaw at the location. An evaluation, which included a quantitative structural analysis, determined that the RHR piping was capable of performing its intended design function, however, the piping did not meet ASME Section XI Code acceptance criteria.

Therefore pursuant to 10 CFR 50.55a(a)(3)(ii), a one-time relief was requested from the NRC, both verbally and in a letter dated March 23, 1998 (Serial No.98-177), from the ASME Section XI pressure boundary leakage flaw evaluation and corrective action requirements to permit the RHR system to be considered operable even with the identified flaw. The NRC granted verbal relief on the same day. In a subsequent letter dated March 27, 1998 (Serial No. 98-177A), Virginia Electric and Power Company committed to perform a failure analysis of the RHR pipe leak, including a metallurgical analysis of the pipe sample taken from the leak area, and to provide the results to the NRC. In a letter dated April 28, 1998, the NRC documented their verbal approval of the requested relief.

A small portion of piping (approximately 2" in diameter) was cut out to remove the flawed portion of the pipe to facilitate a Code repair and to provide a pipe sample at the flaw location to support the failure/metallurgical analysis. The Code repair was performed in accordance with an approved design change package.

Failure/Metallurgical Analysis and Structural Integrity Evaluation The initial evaluation of the removed sample was conducted onsite. Figures 1a and 1b show the outside and inside surfaces, respectively, of the sample removed from line 12"-RH-19-602. The initial evaluation indicated that the inside surface had experienced extensive intergranular attack as shown in Figures 2a and 2b. No discernable leak path could be detected on either the inside or outside surfaces.

Page 1 of 4

e Therefore, the sample was sent to Westinghouse Electric Company for further evaluatior.1 and detailed failure analysis. The failure analysis of the sample, performed by Westinghouse, used a number of evaluation techniques and measurements.

Leakage from the line was attributed to a small area of intergranular attack in the heat affected zone of the piping-to-flange butt weld. The metallurgical evaluation concluded that a sensitized structure existed in the sample. The piping material in that location had been sensitized by prior thermal history, including but not necessarily limited to stress relief at 1600°F after cold bending and prior to subsequent welding. This sensitized material allowed the intergranular attack under the environmental influence of stagnant, highly borated reactor coolant with uncontrolled oxygen content and residual stresses associated with the welding. Eventually the attack progressed to the point that, even though no through-wall flaw could be discovered with nondestructive examination (NOE) techniques, such as liquid penetrant and ultrasonic inspection, a through-wall path of diffuse, interconnected, grain boundaries allowed slow diffusion of fluid through the pipe wall.

Based on a review of the results of the failure/metallurgical analysis, certain conclusions can be drawn regarding the condition of the piping material left in service after the repair of the leak. Hardness testing performed on the piping material sample yielded results that indicate a tensile strength of the pipe material on the order of 90,000 to 95,000 ksi which is probably near the high end of the range of expected strength for the material. The minimum tensile strength requirement of the pipe material is 70,000 ksi.

The additional carbide precipitation connected with the sensitization of the material is known to have a role in elevating the strength, but usually a small effect. The sensitization process is also known to have a more pronounced effect on the impact strength of the material, the degree of which is dependent upon the temperature at which sensitization occurs. Since the sensitization of this material largely occurred at the relatively high temperature of 1600°F, any effect on impact properties would have been minor at most (Reference 1). Based on the above facts and the observation that even in the area suspected to be the leak location the intergranular attack was diffuse rather than dense, it is anticipated that the affected material would exhibit the specified base metal or weld metal mechanical properties.

A necessary input to a structural analysis of the as-left p1pmg in question using traditional fracture mechanics techniques would be to characterize the flaw, assuming a flaw were present. The characterization involves assessment of the size and shape of the flaw. However, in this case the affected area was removed by the repair undertaken to stop the leakage. Even in the small sample removed for failure analysis the area of probable leakage was very small compared to the volume of material in the sample and was significantly more degraded than material just a short distance from the leak location. The average depth of penetration of the intergranular attack over the bulk of the sample was only about 1 grain deep and would be conservatively estimated to be about 5% of the wall. Given the localized nature of the severest damage (which has been removed) and that no flaw characterization could be performed at the leak location, the shallow depth of the bulk of the intergranular attack, and the diffuse and Page 2 of 4

random nature of the attack, the piping remaining in service can be considered to be free of defects that could affect structural integrity.

A structural integrity evaluation relative to the identified leak in the RHR piping was previously discussed in our relief request dated March 23, 1998 (Serial No.98-177) and the NRG safety evaluation report included in their April 28, 1998 letter. As noted in these letters, the structural integrity evaluation of the RHR piping was conducted using an area reinforcement analysis, a limit load analysis, and a fracture mechanics evaluation. The evaluation demonstrated that:

Ductile tearing is unlikely to occur at the leak location when the piping is subjected to the design pressure of 600 psig.

The limit load analysis shows there is adequate margin against a ductile rupture.

The fracture mechanics analysis shows that the calculated stress intensity factor is well below the allowable stress intensity factor. A failure due to brittle fracture is therefore unlikely.

These analysis conclusions demonstrate that the worst case loading is the design pressure for the type of assumed flaw. As an additional assurance of system operability and integrity, the repaired RHR piping section that previously contained the affected area was pressure tested at operating pressure and no leakage was identified.

Conclusion Because our evaluation of the affected area could not find any discrete flaw and, therefore, no conclusive flaw evaluation could be performed, definitive assessment of the factors that may have contributed to the cause of the leakage is difficult. However, we have concluded that given 1) the localized nature of the severest damage (which has been replaced), 2) the shallow depth of the bulk of the intergranular attack, and

3) the diffuse and random nature of the attack, the RHR system piping remains operable and free of defects with the potential for propagation of structurally significant flaws. Specifically, based on the failure/metallurgical analysis, the RHR system is considered fully operable and capable of performing its design function.

Due to the particular conditions observed at the leak location and for long term reliability, that portion of RHR line 12"-RH-19-602 from the affected weld to the next upstream weld will be replaced during the next Unit 1 refueling outage, currently scheduled for October 1998. The condition of the removed piping will be evaluated to determine further corrective action.*

Visual walkdown inspections to detect any boric acid accumulation and/or unidentified leakage will continue to be conducted on both units at the current frequency of each Page 3 of 4

e scheduled cold shutdown. Furthermore, in addition to the required ASME Code Section X.I pressure testing performed each Code period, portions of the RHR ASME Class 2 600 pound stainless steel piping will be placed in a supplemental nondestructive examination (NOE) program for both units. This program will select welds for examination at a frequency three times that required by the ASME Code for Class 2 piping, and the NOE inspections performed on the selected welds will be the same as that required for Class 2 piping.

Reference:

Davis, J. R. (ed.), ASM Specialty Handbook - Stainless Steels, ASM International, 1994, pp 40, 48 - 50 Page 4 of 4

• e Figure 1a: Photograph of the outside surface of the sample removed from 12-RH 602 as received at the SRF. The suspected leak location is bounded by* the mark indicated by the arrow. Magnification about 1.5X.

Figure 1 b: Photograph of the inside surface of the sample removed from 12-RH 602 as received at the SRF. No evidence of a leak site was found. The ink marks are artifacts of efforts to transfer the suspected OD leak location to the ID. Magnification about 1.5X.

r

~.

Figure 2a: Electron micrograph of the inside surface of the sample showing evidence of extensive IGA which was typical over the entire surface and suggested the material was sensitized . Magnification about 1OOX.

I Figure 2b: Electron micrograph of the inside surface of the sample showing the IGA at higher magnification. Some general attack of the surface is also apparent.

Magnification about 400X.