ML073130385

Request for Additional Information Regarding Revised Probability of Prior Cycle Detection Model

ML073130385
Person / Time
Site:	Sequoyah
Issue date:	11/20/2007
From:	Moroney B NRC/NRR/ADRO/DORL/LPLII-2
To:	Campbell W Tennessee Valley Authority
Bowen, Jeremy NRR/ADRO/DORL/LPL2-2
References
TAC MD4110
Download: ML073130385 (8)

Text

November 20, 2007 Mr. William R. Campbell, Jr.

Chief Nuclear Officer and Executive Vice President Tennessee Valley Authority 6A Lookout Place 1101 Market Street Chattanooga, TN 37402-2801

SUBJECT:

SEQUOYAH NUCLEAR PLANT, UNIT 2 - REQUEST FOR ADDITIONAL INFORMATION REGARDING REVISED PROBABILITY OF PRIOR CYCLE DETECTION MODEL (TAC NO. MD4110)

Dear Mr. Campbell:

By letter dated January 12, 2007, Tennessee Valley Authority (the licensee) submitted technical specification change request TS-06-06 to revise the probability of prior cycle detection model used in implementing the steam generator tube voltage-based repair criteria discussed in the technical specifications for Sequoyah Nuclear Plant, Unit 2.

In order for the U.S. Nuclear Regulatory Commission staff to complete its review of the information provided by the licensee, we request that TVA provide responses to the enclosed request for additional information (RAI). Based on discussions with your staff, we understand that you plan to respond to the enclosed RAI by December 14, 2007. If you have any questions about this material, please contact me at (301) 415-3974.

Sincerely,

/RA/

Brendan T. Moroney, Project Manager Plant Licensing Branch II-2 Division of Operating Reactor Licensing Office of Nuclear Reactor Regulation Docket No. 50-328

Enclosure:

RAI cc w/enclosure: See next page

ML073130385 NRR-088 OFFICE LPL2-2/PM LPL2-2/PM LPL2-2/LA CSGB/BC LPL2-2/BC NAME AHiser JBowen BMoroney CSola TBoyce by memo DATE 11 / 16 /07 11 / 19 /07 11 / 16 /07 08/01/07 11 / 20 /07 REQUEST FOR ADDITIONAL INFORMATION SEQUOYAH NUCLEAR PLANT, UNIT 2 TECHNICAL SPECIFICATION CHANGE REQUEST TS-06-06 REVISED STEAM GENERATOR VOLTAGE-BASED REPAIR CRITERIA PROBABILITY OF PRIOR CYCLE DETECTION METHODOLOGY DOCKET NO. 50-328 By letter dated January 12, 2007 (Agencywide Document Access and Management System Accession No. ML070250031), Tennessee Valley Authority (the licensee) submitted a request to revise the probability of prior cycle detection (POPCD) model used in implementing the steam generator (SG) tube voltage-based repair criteria discussed in the technical specifications for Sequoyah Nuclear Plant, Unit 2 (SQN-2). In order for the U.S. Nuclear Regulatory Commission (NRC) staff to complete its review, responses to the following questions are requested.

1. Regarding your proposed reporting requirements for implementation of the POPCD methodology, please update your proposal to reflect your new SG technical specification format. In addition, please discuss your plans to clarify when the report would be submitted since as currently written, you would be required to submit a report 90-days after the SGs were returned to service regardless of whether SG tube inspections were performed. For example, you could specify that consistent with Technical Specification 6.9.1.16.1, the report would be submitted 90-days after the initial entry into MODE 4 following completion of an inspection performed in accordance with Technical Specification 6.8.4.k, ASteam Generator Program.@

Please note that the discussion in Section 2.0 of Enclosure 1 to your January 12, 2007, submittal does not match the wording in your proposed Technical Specifications in Enclosure 2. The staff assumed that the proposed wording in the Technical Specifications is what you intended to propose.

2. Regarding the changes you intend to make to your Bases, the NRC does not have a copy of the Westinghouse letter that is referenced. As a result, it can not comment on whether this change is appropriate. The NRC expects that if the POPCD methodology is approved that it would be implemented consistent with the methodology provided to the staff to support its approval. The methodology in the referenced Westinghouse letter may have been modified as a result of responses to the NRC staff=s request for additional information. Please discuss your plans to ensure that the Bases reflect the actual POPCD methodology that is approved by the NRC staff.

3. On pages E1-5 and E4-14 of Enclosure 1 to your January 12, 2007, submittal, you indicated that if an indication grows by an extreme amount (i.e., an outlier) that you will implement a methodology intended to address this issue. Given the text on these two pages are different, it is not clear whether the methodology would be submitted for NRC approval. Please confirm that this outlier methodology would be submitted for NRC approval prior to implementation. The NRC staff notes that its previous approval of POPCD relied on (1) modifications to the inspection and repair criteria to limit the potential for indications with high voltage growth; (2) other calculations performed by the Enclosure

licensee; and (3) additional reporting requirements that essentially require an assessment if the under predictions were significant (as discussed in the technical specifications).

4. On pages E1-5 and E4-15 of Enclosure 1 to your January 12, 2007, submittal, you indicated that you performed rotating probe inspections of bobbin indications greater than 1.5 volts and no indications exceeded the 1.9 volt threshold for preventive repair.

Since the actual practices in prior inspections may affect the potential for large voltage growth rates, please discuss whether any indications were preventively repaired as a result of these inspections (i.e., regardless of whether they exceeded the 1.9 volt threshold). If indications were repaired (to address the potential for large voltage growth rates), discuss whether similar practices will be employed in all future inspections.

5. You indicated that the number of high voltage indications in the POPCD database for SQN-2 does not satisfy the minimum requirements specified by the industry. As a result, you compared the SQN-2 database to the industry database. Please discuss what the effect would be on the SQN-2 POPCD curve if the SQN-1 [Sequoyah Nuclear Plant, Unit 1] and SQN-2 POPCD databases were combined. Although there may be differences in the noise levels (e.g., from denting) between the two units, there may be some insights gained on potential performance for higher voltage level indications.

In addition, please provide an enlarged view of the data above 1.0 volt in Figure 5.

Your justification for using the SQN-2 POPCD curve instead of the more limiting of the composite POPCD or SQN-2 POPCD curve is largely qualitative and relies on the Asmall@

differences between the curves. Since the POPCD curve is used to calculate the probability of burst under steam line break conditions and the amount of leakage under steam line break conditions, it would appear that a more appropriate comparison would be to evaluate the effects of the different POPCD curves on the structural and leakage integrity of the SG tubes. Please provide this assessment.

Since this is a one-time assessment, please discuss your plans to submit a similar analysis until a sufficient number of Ahigh@ voltage data points are present at SQN-2.

Alternatively, discuss your plans for using the most limiting of the POPCD curves.

6. On page E4-6 (1st bullet), you indicate that the end of cycle (EOC) inspection (EOCn) bobbin voltages are Agenerally@ based on the inspection records for the EOCn. Please discuss your plans to report (in the 90-day report) any instances in which sizing is based on a reanalysis of the voltages previously reported. The staff notes that detection will always be based on past inspection records (as indicated in the first bullet).

7. On page E4-6 (2nd bullet), you indicate that the EOCn voltage for new EOCn+1 indications will be based on lookback analyses when the EOCn voltages are not available from the inspection record. Please clarify Awhen@ an EOCn voltage would be available for a flaw not reported in EOCn (or should the Awhen@ in this sentence really be Asince@). In addition (if Awhen@ is the correct word in this sentence), please discuss why it is acceptable (in this case) not to confirm that the reported EOCn voltage actually corresponds to the flaw reported at EOCn+1.

8. On page E4-7, you indicated that data supporting the adequacy of using the square root sum of the squares method for determining the inferred bobbin voltage from multiple rotating probe indications at a tube support plate was provided by another utility (in an August 18, 2004 letter). Since this information was specific to this utility, please provide a similar plot demonstrating that the approach is currently adequate for SQN-2 (future assessments will be provided in the 90-day report).

9. Figure 2 provides a correlation for assigning an inferred bobbin voltage to indications detected only with a rotating probe. A similar curve is provided in Figure 3-13 of your March 20, 2007 letter, which submitted the SQN-2 Cycle 14 90-day SG report. Please discuss the differences in these curves. Please discuss if there have been significant changes in this correlation with time. If so, discuss whether a composite curve is more appropriate or whether a conservative lower bound to any single cycle data is appropriate. In addition, please confirm that there are no data from axially oriented outside diameter stress corrosion cracking indications not detectable by bobbin (AONDBs) such that an assessment of the adequacy of this correlation can currently be assessed. The staff notes that it appears that the prior practice was to plug AONDBs on detection (page E4-17).

10. On page E4-9, you indicated that if the p-value is greater than 5 percent, you will propose an alternate probability of detection (POD) model and submit the recommendation to the NRC for approval. Given that it may take time for the NRC to review any changes to the POD model, please discuss the model to be used in the interim pending staff review of the alternate model. The NRC staff notes that it previously accepted the use of a default value of 0.6 for POD as an acceptable alternative.

11. On page E4-15, you indicate that an assessment for the onset of voltage dependent growth Ashould@ be performed and the methods of Reference 1 applied when growth rates show a dependence on the beginning of cycle voltage. Please confirm that this assessment Awill@ be performed. In addition, discuss whether there are any differences in the voltage dependent growth methodology discussed in Reference 1 and the methodology approved by the NRC staff in prior POPCD approvals. If there are any differences, please justify them.

12. On page E4-16, you indicated that a rotating probe (+Point) inspection would be performed for +Point confirmed indications at EOCn that are not detected by bobbin at EOCn+1. Please confirm that this is Note 1 in Table 1 (rather than Note 3 in Table 1 as indicated in the text).

13. On page E4-17, you indicate that you assign a through-wall depth (i.e., a percent code) to bobbin indications confirmed with a rotating probe to be volumetric indications. Please confirm that this applies only to volumetric wear indications and thinning/wastage indications rather than to volumetric indications attributed to intergranular attack or closely spaced cracks. Please confirm that the voltage based repair criteria are applied to volumetric outside diameter cracking indications (including intergranular attack). If not, please confirm that these indications are plugged on detection.

14. On page E4-18, you indicated that no reevaluation was performed of the EOC 9 data for a 2.03 volt indication that was detected at EOC 10. Please confirm that future assessments of POPCD that use previous data will be re-evaluated to determine the voltage of the potentially missed indication. In addition, please discuss what growth rate was assigned to this indication if the previous data were not reviewed (this is important from a benchmarking standpoint).

15. On page E4-18, you indicate the composite POPCD data contains 11 indications found only with a rotating probe (Column F of Table 4) although there were only six occurrences. You further indicate that the rotating probe only detections are counted as missed indications in two successive cycle POPCD evaluations even though they were repaired on detection. Please clarify this discussion. If an indication was initially detected at EOCn+1 and subsequently plugged, it does not appear (to the staff) that the indications should be counted twice based on a review of Table 1.

16. Please provide an enlarged view of the data above 1.0 volt in Figure 4. In addition, please discuss what POPCD curve is required to be used in the assessments given that the POD above 1.0 volt for the cycle 12 data is less than the corresponding POD for the composite dataset.

17. Regarding Table 7, please confirm that the number of data points in the POPCD curve for EOC 7 through 12 is only slightly greater than the number of data points from EOC 12. Since a number of datapoints should be present from year to year, the staff would have expected a large number of datapoints for the EOC 7 through 12 composite curve and a significantly less number of datapoints for any one cycle (unless a significant number of new indications were identified in EOC 13).

18. Please confirm that no adjustment to the voltage growth rate distribution is performed when the average growth rate decreases from one cycle to the next.

19. In the benchmark analysis, you used the composite POPCD curve from Cycle 7 through 12. This does not appear to be appropriate since some of the data was not present at the time of the inspections (and using Afuture@ data to predict past trends would normally be expected to provide reasonable results). As a result, please repeat the benchmark analysis with POPCD data available at the time the inspections were performed (use of the latest burst and leakage correlations and actual cycle lengths is acceptable). In addition, this benchmarking should not include the datapoint that exhibited Aextreme voltage growth@ since actions were taken to limit the possibility of such extreme voltage growths and inclusion of this datapoint can skew the results. This datapoint should not be included in the beginning of cycle voltage distribution or the growth rate data (i.e., the average growth rate or the distribution of growth rates). As part of this benchmarking analysis, please include an analysis of the EOC 14 inspection data completed in Fall 2006. Please include in this response the database used for assessing the integrity of the tubes, the actual cycle length, and the cycle length assumed in the benchmarking analysis. Lastly, discuss whether there has been any significant preventive plugging of tubes such that the benchmarking analysis (or growth rate distributions) may have been skewed.

20. Regarding your continuing assessment of the inspection results, please confirm that an assessment will be performed for any underpredictions in order to assess the probable cause. As currently written, it is not clear to the staff whether the assessments would only be performed when the quantitative criteria listed in Section 7.0 of your submittal are exceeded. In addition, please confirm that the assessment of the underprediction of the number of indications will include the potential need to increase the number of indications regardless of whether the indications that need to be increased are low or high voltage. Lastly, please confirm that you will provide an update (based on the results of your inspection) to Table 7 for the composite and one-cycle POPCD curve for SQN-2 in your 90-day report.

21. On page E1-5, you indicate that upon approval of the POPCD methodology the growth rates used in the operational assessments will be obtained as the bounding growth rate of the SGs and the composite average growth over the last two cycles of operation.

Please clarify this statement. For example, is the composite average growth, the composite from all SGs over one cycle (or two cycles)? If the composite is from all SGs, discuss whether it is necessary to assess whether the average growth rate from one SG is increasing at a rate greater than the other SGs (implying that this SG may need a larger increase in the average growth rate than the Acomposite@ growth rate would suggest).

22. Please provide a step-by-step description of the rationale that will be used for selecting the growth rate distribution and the POPCD curve used in the EOC projections. Please confirm that this methodology was used in performing the benchmarking analysis. If not, please justify any differences. Please confirm that voltage dependent growth is assessed on a SG basis rather than just a composite of all SGs.

23. You indicate that you will address POPCD uncertainties by either applying POPCD at the lower 95 percent confidence level or including an uncertainty analysis for POD in the operational assessment (Section 8.0 of your January 12, 2007 letter). Please clarify whether the Auncertainty analysis for POD in the operational assessment@ is equivalent to the uncertainty analysis approved for use at Diablo Canyon (in their approval to use POPCD). If not, please clarify this statement.

In addition, since the POPCD curve is used to calculate the probability of burst under steam line break conditions and the amount of leakage under steam line break conditions, it would appear that an assessment of the adequacy of using the lower 95 percent confidence level to address uncertainty (rather than the Monte Carlo approach approved for use at Diablo Canyon) would be to evaluate the effects of the different approach for modeling uncertainty in the POPCD curves on the structural and leakage integrity of the SG tubes. Please provide this assessment.

Since this is a one-time assessment, please discuss your plans to submit a similar analysis in each of your 90-day reports. In the event that the results using the lower 95 percent confidence level are non-conservative, discuss your plans to use the uncertainty modeling approach approved for Diablo Canyon.

William R. Campbell, Jr.

Tennessee Valley Authority WATTS BAR NUCLEAR PLANT cc:

Mr. Gordon P. Arent Ms. Beth A. Wetzel, Manager New Generation Licensing Manager Corporate Nuclear Licensing and Tennessee Valley Authority Industry Affairs 5A Lookout Place Tennessee Valley Authority 1101 Market Street 4X Blue Ridge Chattanooga, TN 37402-2801 1101 Market Street Chattanooga, TN 37402-2801 Mr. Ashok S. Bhatnagar Senior Vice President Mr. Masoud Bajestani, Vice President Nuclear Generation Development Watts Bar Unit 2 and Construction Watts Bar Nuclear Plant Tennessee Valley Authority Tennessee Valley Authority 6A Lookout Place P.O. Box 2000, EQB 1B 1101 Market Street Spring City, TN 37381 Chattanooga, TN 37402-2801 Mr. Michael K. Brandon, Manager Mr. James R. Douet Licensing and Industry Affairs Vice President Watts Bar Nuclear Plant Nuclear Support Tennessee Valley Authority Tennessee Valley Authority P.O. Box 2000 3R Lookout Place Spring City, TN 37381 1101 Market Street Chattanooga, TN 37402-2801 Mr. Michael J. Lorek, Plant Manager Watts Bar Nuclear Plant Mr. H. Rick Rogers Tennessee Valley Authority Vice President P.O. Box 2000 Nuclear Engineering & Technical Services Spring City, TN 37381 Tennessee Valley Authority 3R Lookout Place Senior Resident Inspector 1101 Market Street Watts Bar Nuclear Plant Chattanooga, TN 37402-2801 U.S. Nuclear Regulatory Commission 1260 Nuclear Plant Road Mr. Michael D. Skaggs, Site Vice President Spring City, TN 37381 Watts Bar Nuclear Plant Tennessee Valley Authority County Executive P.O. Box 2000 375 Church Street Spring City, TN 37381 Suite 215 Dayton, TN 37321 General Counsel Tennessee Valley Authority County Mayor 6A West Tower P. O. Box 156 400 West Summit Hill Drive Decatur, TN 37322 Knoxville, TN 37902 Mr. Lawrence E. Nanney, Director Mr. John C. Fornicola, Manager Division of Radiological Health Nuclear Assurance Dept. of Environment & Conservation Tennessee Valley Authority Third Floor, L and C Annex 3R Lookout Place 401 Church Street 1101 Market Street Nashville, TN 37243-1532 Chattanooga, TN 37402-2801 Mr. Larry E. Nicholson, General Manager Performance Improvement Tennessee Valley Authority 4X Blue Ridge 1101 Market Street Chattanooga, TN 37402-2801