05000270/LER-2002-003
| Docket Number | |
| Event date: | |
|---|---|
| Report date: | |
| 2702002003R00 - NRC Website | |
EVALUATION:
BACKGROUND
Duke Power provides this report as a voluntary Licensee Event Report.
Oconee Nuclear Station (ONS) uses the Babcock and Wilcox (B&W) Nuclear Steam Supply System [EIIS:AC], which includes two Once Through Steam Generators (SGs) [EIIS:SGI], for primary to secondary heat transfer per unit. Each SG is a vertical, straight tube heat exchanger. Inside the SG shell, there is an upper tube sheet, 15 tube support plates, a lower tube sheet, and 15531 tubes per steam generator. The tubes are nominally 0.625 inches in outside diameter. During manufacture of the tubes, minor dents and manufacture burnish marks were considered acceptable. However these are now recognized as potential stress concentration points which can contribute to expected tube degradation due to age.
During each refueling outage, in-situ pressure testing is conducted on selected SG tubes as part of the SG In-Service Inspection. The tubes are selected for this test based on eddy current indications observed during the current outage inspections. By normal practice, any tube selected for in-situ pressure testing will be removed from service by plugging regardless of the test result.
The acceptance criteria are that the tubes should not burst at either 3.0 times normal steady state pressure for full power operation or 1.4 times the transient pressure during a limiting design basis accident concurrent with a safe shutdown earthquake.
For ONS the procedural limiting value, corrected for test conditions and instrument issues, is 4250 psi. The test is conducted using a test pump with a nominal 3 gpm capacity.
Prior to this event, Unit 2 was defueled during a Refueling outage (No Mode) with no safety systems or components out of service that would have contributed to this event.
7
EVENT DESCRIPTION
During October 2002, in-situ pressure testing of SG tubes was conducted as part of the Oconee Unit 2 refueling outage (EOC 19) In-Service Inspection. A total of 21 tubes were tested (ten tubes in SG "A" and eleven tubes in SG "B"). All ten of the SG "A" tubes and ten of the eleven SG "B" tubes showed no leakage during the pressure testing. However, one SG "B" tube began to leak prior to reaching the highest test pressure.
Specifically, eddy current testing of SG "B" Tube 37-27 had shown a defect measured as a 2" long single axial indication, 95% through wall (maximum depth), in a dent located just above the 15th support plate. Therefore the tube was selected for in-situ testing.
During in-situ pressure testing conducted on 10-31-02, SG "B" Tube 37-27 began to indicate leakage at approximately 3900 psig. The tube did not reach the full 3 times normal operating delta-p pressure (4250 psig) and thus did not meet the test criterion. The leak rate equaled or exceeded the capacity of the test pump, approximately 3 gpm, such that pressure could not be maintained.
The test was aborted and appropriate personnel, including Operations, were notified.
Additional eddy current testing and video inspections were conducted to evaluate the leak. The leak was found to be at the defect mentioned above. SG "B" Tube 37-27 was subsequently removed from service by plugging.
Following the evaluation of the leak and eddy current data from the current outage, data from prior outage inspections on Unit 2 were re-evaluated. This review found that recordable volumetric and dent indications had been observed at this location as far back as 1993. However, the indications were not interpreted as indicative of a crack. Reviewing the data in retrospect, it appears that as early as 1998 the data was ambiguous and could reasonably be interpreted either as indicative of a manufacture burnish mark (MBM) superimposed on a dent, to be resolved accordingly, or as degradation associated with the dent.
and the most recent outages on Oconee Units 1 and 3 were re- evaluated. The review for Unit 2 found 28 additional tubes with locations where dent and MBM indications overlapped. These tubes locations where dent and MBM indications overlapped appeared to contain possible defects as was seen on 27. Unit 3 revealed no similar overlapping dent/MBM found 13 but none SG 2B tube 37- indications.
CAUSAL FACTORS
The immediate cause mechanism was determined to be expected component aging. Axial indications have been seen in dents in the past and are anticipated in the ONS SGs. Therefore the root cause of this event is related to the failure to properly analyze the flaw and remove it from service sooner. The root cause of the event was evaluated by knowledgeable individuals and reviewed by outside consultants from EPRI and the industry.
The root cause of this event is human error (i.e. guidance document not followed correctly) due to the difficulty associated with interpreting eddy current test results during prior inspections.
While it is reasonable, in retrospect, to state that previous inspections provide evidence of the presence of degradation, the combination of signals (MBM and dent) resulted in masking the defect.
Investigation and detailed examination of the leaking tube found that there was a dent at 5.41 inches above the 15th tube support plate. The dent had axial length. Also there is a volumetric flaw superimposed over the end of the dent at this location. The volumetric flaw is believed to be a manufacture burnish mark. The axial flaw was essentially a crack inside the dent and the area of the MBM. This was the point where the structural failure occurred during the in-situ pressure test.
A review of prior eddy current test data shows that the bobbin data has not changed since 1993, which indicates that the dent is unchanged. However, rotating coil data is more capable of detecting a crack. The review of rotating coil data found that an axial indication that was marginally detectable in 1998 has been slowly degrading since then. The current analysis guidelines were reviewed and determined to be adequate to have identified the axial indication during the outages since 1998. Therefore, in hindsight, it appears that the axial flaw could have been detected earlier.
However, because these flaws were in essentially the same location as the dent, the potential for small cracks at this location was masked, making interpretation of indications from earlier inspections difficult.
As a result of this finding, guidance will be clarified so that all indication of dents and volumetric flaws in close proximity to each other will be considered a precursor signal and masking combination that affects detectability of potential flaws. Oconee will revise criteria so that, for the current ONS Steam Generators, any tubes displaying this combination of signals in the future will be removed from service as a preventative measure. All three ONS units will undergo SG replacement, expected to begin during with the next Unit 1 refueling outage.
CORRECTIVE ACTIONS
Immediate:
1. The affected tube was removed from service by plugging.
Subsequent:
1. Test data from this Unit 2 outage and the preceding outages on Units 1 and 3 were reviewed for similar indications. As a result, 28 additional tubes were preventatively plugged on Unit 2. Thirteen overlapping dent and MBM indications on Unit 1 were found and dispositioned as having no indication of cracking.
Unit 3 revealed no similar overlapping dent and MBM indications.
Planned:
1. Revise analyst guidelines to include specific guidance on dents with complex indications that potentially could mask degradation.
2. Revise dispositioning guidelines for the current ONS Steam Generators so that any dent indication with evidence that other signals could be present shall be plugged since it could be masking more serious degradation.
3. Indications similar to the pre-2002 eddy current results for B 37-27 indication shall be included in future analyst training 7 and testing. Verification that analyst are sensitive to complex signals contained at dented locations shall be demonstrated.
None of these corrective actions are considered NRC Commitment items. There are no other NRC Commitment items contained in this LER.
SAFETY ANALYSIS
The in situ pressure test indicated the axial flaw met the limiting pressure requirements for normal operation and worst case accident conditions. The SG tube did have a loss of required margin, but there is no presumption that the tube would have failed during an actual event. Therefore this event does not represent significant SG tube degradation that would have a safety concern or operability issue for operation in the previous cycle. For that reason, this event is not considered a safety system functional failure.
However, per NEI 97-06 rev 1 guidance, maintaining the required margin is a maintenance rule function. Since the required margin was not maintained, this event is a maintenance rule functional failure and is considered reportable under the Equipment Performance and Information Exchange (EPIX) program. The affected component was the 2B Steam Generator [EIIS:SGI], manufactured by Babcock and Wilcox [manufacturer's code: B015].
Due to previous operational assessments, the ONS units are limited to only one cycle of operation between required 100% inspections which insures that other potentially significant degradation in this geometry will be detected prior to exceeding the safety requirements for protection against burst and/or leakage due to most limiting accident condition and during normal operation.
Therefore, there was no actual impact on the health and safety of the public due to this event.
ADDITIONAL INFORMATION
There have not been any previous indications of dents with flaws that have approached structural limits. Therefore this event is not a recurring event. Since 1997, several hundred axial 7 indications have been in-situ pressure tested with no problems; though of these, only a few have been associated with a dent. This indicates that there has not been a generic problem with detectability or growth rate of defects contained within dented locations. There have not been any indications of volumetric flaws approaching structural limits and therefore the detectability of MBM indications is not in question.
There were no releases of radioactive materials, radiation exposures or personnel injuries associated with this event.