ML18152B363

Safety Evaluation Supporting Eddy Current Techniques Used by VEPCO to Determine Depth of Degradation Evident in Units SG Tubing & VEPCO Approach for Dispositioning Tubes with Avb Wear Indications

ML18152B363
Person / Time
Site:	Surry
Issue date:	08/23/1999
From:	NRC (Affiliation Not Assigned)
To:
Shared Package
ML18152B362	List: ML18152B361 ML18152B363
References
GL-97-05, NUDOCS 9908300134
Download: ML18152B363 (4)

Text

e UNITED STAT*::::,

NUCLEAR REGULATOR\\ COMMISSION WASHINGTON, D.C. 20555--0001 STAFF EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION SURRY POWER STATION, UNITS 1 AND 2 VIRGINIA ELECTRIC AND POWER COMPANY DOCKET NOS. 50-280 AND 50-281

1.0 INTRODUCTION

On December 17, 1997, the NRC issued Generic Letter (GL) 97-05, "Steam Generator Tube Inspection Techniques." The focus of GL 97-05 was to verify that steam generator (SG) tube inspection practices used by licensees were consistent with existing regulatory requirements and the licensing bases for the subject plants. The requirements applicable to the inspection of SG tubing are included in Title 10 of the Code of Federal Regulations (10 CFR) Part 50 and the plant technical specifications (TS). Of particular interest to the NRC in issuing GL 97-05 was to ensure licensees are utilizing inspection techniques qualified to size the depth of identified tube degradation to determine whether degraded tubes are acceptable for continued operation with respect to the plugging limits included in the plant TS.

Virginia Electric and Power Company (the licensee), submitted a response to GL 97-05 for Surry Power Station, Units 1 and 2, on March 17, 1998, as supplemented on September 10, 1998. In addition, the NRC staff and the licensee discussed inspection practices at Surry Unit 2 on October 28, 1998, and April 28, 1999, during a telephone call. The licensee documented the plugging of certain SG tubes in a letter dated May 17, 1999. The NRC has completed its review of the licensee's responses to GL 97-05. The following documents the staff's conclusions from this review.

2.0 INSPECTION PRACTICES AT SURRY POWER STATION Surry Power Station, Units 1 and 2, are equipped with Westinghouse Model 51 F SGs which replaced the units' original SGs in the early 1980's. These generators employ a second generation of tube material (thermally treated Alloy 600) that has been proven to be more resistant to corrosion*-induced damage than materials that were originally used in the fabrication of pressurized water reactor (PWR) SGs. In accordance with the plant TS, the SG tubes are periodically inspected to ensure that adequate margins exist for the continued integrity of the reactor ~oolant pressure boundary. The plt11;ging limit applicable for Surry 1 and 2 require~ Lilat*

tubes with through-wall degradation with depths in excess of 40 percent of the nominal tube wall thickness be removed from service.

The licensee stated that only volumetric indications (e.g., anti-vibration bar (AVB) wear, mechanical wear, pitting) are dispositioned by eddy current sizing techniques during inspections.

No crack-like degradation has been observed to date. To identify tubes with possible

• - possible degradation, the licensee inspects using techniques qualified in accordance with Appendix Hof the Electric Power Research lnstitute's "PWR Steam Generator Examination Guidelines," Revisions 3 through 5 (EPRI Guidelines) Techniques are qualified using test samples to assess detection and sizing capabilities.

Two modes of SG tube degradation are dispositioned by sizing the depth using eddy current techniques at Surry Power Station, AVB wear and pitting. AVB wear is the result of mechanical contact between SG tubes and secondary support structures in the upper U-bend region of the bundle. The tube and the AVB may move relative to one another during operation at some AVB 'locations. This movement abrades the surface of the tube leading to wall loss and a reduced margin of structural.integrity for the tube. This mode of degradation is common to PWR SGs and is well understood. Because AVB wear is a mechanical rather than* corrosive (i.e., stress corrosion cracking) degradation mechanism, all tube materials are susceptible to this form of damage. The licensee uses a bobbin coil voltage amplitude sizing technique to estimate the depth of AVB wear indications. The equipment is calibrated using a standard

• replicating the damage found in the Surry SGs.

Pitting of SG tubing is a corrosive mode of degradation caused by the presence of aggressive chemical species on the outer surface of the tubes. According to the licensee, suspected pit indications are characterized with rotating eddy current probes. A volumetric geometry as observed from this examination confirms the presence of pit indications.. *.;Yolumetric geometry

efers to an eddy current response that shows no apparently crack-like (i.e., linear) flaw characteristics. Once the pit is confirmed, a bobbin coil technique is applied to estimate the depth of the indication. As indicated in the submittal dated September 10, 1998, the licensee was tracking six cold leg pit indications located above the tube sheet secondary face in four tubes in the Surry Unit 2 SGs. The Surry Unit 2 SGs contained six cold leg pit indications located above the tube sheet secondary face in four tubes prior to the Spring 1999 refueling outage. During the 1999 outage, the licensee inspected all but one of these tubes using ultrasonic test (UT) techniques and removed all four tubes from service. The results of the inspections supported the licensee's conclusion that the indications were volumetric, corrosive-induced degradation.

3.0 EVALUATION OF INSPECTION TECHNIQUES In order to disposition SG tube degradation in accordance with the depth-based repair limits in the plant TS, nondestructive examination techniques must be capable of identifying, characterizing and sizing the through-wall penetration of degraded areas. Adequately achieving each of these steps for all modes of damage observed in PWR SGs may be difficult.

For example, bobbin coil eddy current probes may be capable of detecting axially oriented stress corrosion cracking (SCC), but such probes can neither confirm that the damage mode is SCC nor provide an accurate estimation of the through-wall depth. Consequently, many utilities rely on alternate inspection probe8 (e.g., rotating probes) to address the limitations of bobbi11 probe examinations.

One step in the inspection process that is not formally addressed in current industry guidelines for the examination of SG tubing is the characterization of the mode of damage.

Characterization involves the ability to distinguish between the various forms of tube damage

~l that can be detected during nondestructive testing. Appendix H of the EPRI Guidelines only provides detailed criteria for quantifying detection and siz;:1g capabilities of inspection techniques. Eddy current inspection techniques are capable of providing the location of an indication, the morphology of the electromagnetic response of the indication and a flaw orientation (if applicable). Licensees have traditionally relied upon data provided from inspections and industry experience (e.g., pulled tube data) to characterize the nature of indications. For example, an indication detected at an AVB can be confirmed by the inspection technique to be a volumetric indication. However, without the knowledge that thousands of tubes have been observed throughout the industry with wear at these locations, a utility may not be capable of ultimately concluding that the damage mode is wear rather than some other volumetric form ofdegradation (i.e., integranular attack (IGA)).

In the licensee's submittal dated September 10, 1998, it was stated that the primary basis for concluding that several indications identified in the Surr 1 Unit 2 SG tubes were the result of pitting was from their volumetric appearance on a rotating probe terrain plot display.

Specifically, it was stated that "a pit most closely matches the features exhibited on the terrain plot for these indications." The staff is familiar with the eddy current response from a number of SG tube degradation modes including tube pitting. Pitting generally produces a volumetric terrain plot response characterized as having approximately equal axial and circumferential extent. In addition, some analysts will note that pits produce a return to null when a pancake coil is centered directly over the degraded area. However, the staff notes that signals with these same characteristics can be generated from indications of volumetric !GA One PWR licensee attempted to qualify an approach to distinguish between pit-like IGA and tube pitting, but was unsuccessful. Based on these observations, the NRC has concluded that no qualified method exists to adequately distinguish SG tube pitting from other volumetric damage modes using only eddy current inspection data. In the case at hand, the licensee has not indicated that it has completed a site-specific qualification program to distinguish SG tube pitting from other volumetric eddy current signals.

The licensee st~ted that copper deposits were present on the tubing prior to chemical cleaning in 1994. The staff agrees that copper is a potential contributor to the development of SG tube pits in Alloy 600 material. However, the Surry Unit 2 SG tubes were fabricated with thermally treated Alloy 600 material. Industry experience has demonstrated that this material is less susceptible to corrosive-induced degradation than mill annealed Alloy 600 tubing.

Consequently, comparing damage mechanisms between tubes fabricated from these different materials may be difficult. The NRC is presently unaware of any confirmed (e.g., by tube pulls) pit indications in thermally treated Alloy 600 tubing in any SGs operating worldwide.

Based on the absence of industry data supporting the licensee's conclusion that the Surry Unit 2 SG tubing contained pitting degradation and the industry's past inability to distinguish pitting from other volumetric degradation using the eddy current inspection data, the NRC concluded that the licensee had insufficient evidence to conclude that the suspected pit indications are the result of actual pits in the tubing. Therefore, the licensee's application of eddy current sizing techniques to estimate the depth of the pit-like indications in inspections conducted prior to the spring 1999 outage was considered inappropriate. The use of eddy current sizing techniques qualified for one damage mechanism to estimate the depth of a different mode of degradation may yield depth estimates well outside the expected range of scatter determined from a statistical analysis of the results from the qualification program. The staffs position on this issue was discussed during a telephone conversation on Aori! "<8, 1999.

During the Spring 1999 rer'ueling outage, the licensee examined tubes with pit-like indications using an alternate inspection technique. The licensee indicated during the April 28, 1999, telephone conversation that the primary basis for performing these inspections was to resolve the nature of the pit-like indications. The staff is encouraged by the licensee's actions. The use of alternate inspection techniques to resolve indeterminate indications is a practice that is rarely used at domestic PWR facilities. The steps taken by the licensee indicate their desire to resolve the nature of the pit-like indications and the condition of the affected tubes, and the staff commends the licensee on taking these steps to ensure the integrity of the reactor coolant pressure boundary. In addition, the use of UT inspection techniques demonstrates the licensee's understanding of the shortcomings in using a single inspection technique to assess the nature of unique indications.

The UT inspections of pit-like indications confirmed the presence of degradation in several tubes documented in the Surry Unit 2 "Steam Generator Tube Inspection Report," dated May 17, 1999. According to the report, eight tubes including the four tubes with pit-like indications known prior to the Spring 1999 outage were removed from service.

4.0 CONCLUSION

The licensee for Surry Power Station, Units 1 and 2, has applied qualified eddy current techniques to determine the depth of degradation evident in the units' SG tubing. Indications of AVB wear and pitting is dispositioned using techniques to assess whether the affected tubes are acceptable for continued service with respect to the tube repair limits included in the plant TS.

The NRG staff concluded that the licensee had an inadequate technical basis for assuming suspected pit indications contained in four tubes in the Surry Unit 2 SG prior to the Spring 1999 refueling outage were representative of actual pitting. Therefore, the application of a pitting sizing technique to estimate the depth of these indications was inconsistent with the requirements of Appendix B to 10 CFR Part 50 in the absence of additional data to support the assumption that the signals observed by eddy current methods are the result of pitting degradation. The licensee took appropriate measures to address this issue in the spring 1999 outage by inspecting the pit-like indications with alternate techniques and dispositioning the degradation in the tubes in an appropriate manner. The NRC did not identify any concerns with the licensee's approach for dispositioning tubes with AVB wear indications.

Principal Contributor: P. Rush Date: August 23, 1999