ML070250031

Technical Specification (TS) Change 06-06 - Revised Steam Generator (SG) Voltage-Based Repair Criteria - Probability of Prior Cycle Detection (Popcd)

ML070250031
Person / Time
Site:	Sequoyah
Issue date:	01/12/2007
From:	Morris G Tennessee Valley Authority
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
TS-06-06, TVA-SQN-TS-06-06
Download: ML070250031 (58)

Text

Tennessee Valley Authority, Post Office Box 2000, Soddy-Daisy, Tennessee 37384-2000 January 12, 2007 TVA-SQN-TS-06-06 10 CFR 50.90 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D. C. 20555-0001 Gentlemen:

In the Matter of ) Docket No. 50-328 Tennessee Valley Authority SEQUOYAH NUCLEAR PLANT (SQN) - UNIT 2 - TECHNICAL SPECIFICATION (TS) CHANGE 06 REVISED STEAM GENERATOR (SG) VOLTAGE-BASED REPAIR CRITERIA - PROBABILITY OF PRIOR CYCLE DETECTION (POPCD)

Pursuant to 10 CFR 50.90, Tennessee Valley Authority (TVA) is submitting a request for a TS change (TS-06-06) to License DPR-79 for SQN Unit 2.

The proposed TS change revises the Unit 2 TS requirements to include a SG voltage-based repair criteria probability of detection (POD) method using plant specific SG tube inspection results. The proposed POD method is referred to as the POPCD method. The POPCD method is to be available for use on a permanent basis for the remaining cycles for SQN Unit 2 starting with Unit 2 Cycle 15 (U2C15) refueling outage and until such time as the Unit 2 SGs are replaced.

TVA's proposed change is best implemented during a refueling outage. TVA requests NRC review and approval on a schedule to allow implementation of the proposed change during the U2C15 refueling outage that is scheduled to begin in April of 2008.

Printed on recycled paper

U.S. Nuclear Regulatory Commission Page 2 January 12, 2007 Additionally, in accordance with 10 CFR 50.91(b) (1), TVA is sending a copy of this letter and enclosures to the Tennessee State Department of Public Health.

There are no commitments contained in this submittal. If you have any questions about this change, please contact me at 843-7170.

I declare under penalty of perjury that the foregoing is true and correct. Executed on this 12th day of January, 2007.

Sincerely, G enn W. Morris Manager, Site Licensing and Industry Affairs

Enclosures:

1. TVA Evaluation of the Proposed Changes

2. Proposed Technical Specifications Changes (mark-up)

3. Changes to Technical Specifications Bases Pages

4. Technical Report cc: See page 3

U.S. Nuclear Regulatory Commission Page 3 January 12, 2006 Enclosures cc (Enclosures):

Mr. Lawrence E. Nanney, Director Division of Radiological Health Third Floor L&C Annex 401 Church Street Nashville, Tennessee 37243-1532 Mr. Brendan T. Moroney, Senior Project Manager U.S. Nuclear Regulatory Commission Mail Stop 08G-9a One White Flint North 11555 Rockville Pike Rockville, Maryland 20852-2739

ENCLOSURE 1 TENNESSEE VALLEY AUTHORITY (TVA)

SEQUOYAH NUCLEAR PLANT (SQN)

UNIT 2

1.0 DESCRIPTION

TVA's proposed change would revise the operating license to allow update of the SQN Unit 2 Technical Specifications (TSs) to use a steam generator (SG) voltage-based repair criteria probability of detection (POD) method using plant specific SG tube inspection results. The proposed POD method is referred to as the probability of prior cycle detection (POPCD) method. The POPCD method is to available for use on a permanent basis for the remaining cycles for SQN Unit 2 starting with Unit 2 Cycle 15 (U2C15) refueling outage and until such time as the Unit 2 SGs are replaced.

The Administrative Section of the SQN Unit 2 TS (Section 6.8.4.k) contains steam generator program requirements.

These requirements include voltage-based alternate tube repair criteria from NRC Generic Letter (GL) 95-05, "Voltage-Based Repair Criteria for Westinghouse Steam Generator Tubes Affected by Outside Diameter Stress Corrosion Cracking," dated August 3, 2005. The requirements of GL 95-05 apply a POD of 0.6 to previously inspected tube indications (bobbin inspection) for the determination of the beginning of cycle (BOC) indication voltage distribution for operational assessment. TVA's proposed change to use a POPCD method to determine the BOC voltage distribution is an exception to GL 95-05 and requires prior NRC review and approval.

The NRC staff has previously approved Diablo Canyon Nuclear Power Plant for a licensing amendment to apply the POPCD method to their SGs. NRC provided the Diablo Canyon TS approval by letter dated October 28, 2004.

2.0 PROPOSED CHANGE

TVA's proposed change would allow use of the POPCD method to determine the BOC voltage distribution for the SG operational assessment. TVA is proposing a change to the reporting requirements provided in SQN TS 6.9.1.16. The new POPCD reporting requirement is added as item 6.9.1.16.4 and reads as follows:

"6.9.1.16.4 For implementation of the probability of prior cycle detection (POPCD) method, for the voltage-based repair criteria at tube support plate intersections, if the end-of-cycle conditional tube rupture probability for a postulated main steam line break, the projected primary to secondary leak rate during a postulated EI-I

main steam line break probability, the projected main steam line break leak rate, or the number of indications are under predicted by the previous cycle operational assessment, the following shall be reported to the Commission pursuant to 10 CFR 50.4 within 90 days following return to service of the steam generators:

1. The assessment of the probable causes for the under prediction, proposed corrective actions, and any recommended changes to probability of detection or growth methodology indicated by potential methods assessments.

2. An assessment of the potential need to revise the alternate repair criteria analysis methods if: the burst probability is under predicted by more than 0.001 (i.e., 10 percent of the performance criteria) or an order of magnitude; or the leak rate is under predicted by more than 0.5 gallon per minute (gpm) or an order of magnitude.

3. An assessment of the potential need to increase the number of predicted low voltage indications at the beginning of cycle if the total number of as-found indications in any SG are underestimated by greater than 15 percent or by greater than 150 indications."

TVA's proposed change is an exception to SQN's Unit 2 current licensing basis for GL 95-05. The current licensing basis associated with compliance with GL 95-05, requires using 0.6 as the POD value. This POD value is not specified as a level of detail within the SQN Unit 2 SG program requirements of TS 6.8.4.k. TVA is including proposed changes to SQN's TS Bases that provide discussion of SG tube inspection methods and provide supporting references to the technical reports. The TS markup pages and Bases markup pages are provided in Enclosures 2 and 3, respectively.

3.0 BACKGROUND

3.1 SG Voltage-Based Repair Criteria The SG tubes constitute more than half of the reactor coolant pressure boundary (RCPB) area. Design of the RCPB for structural and leakage integrity is a requirement under Title 10 of the Code of Federal E1-2

Regulations, Part 50 (10 CFR 50), Appendix A. Specific requirements governing the maintenance and inspection of SG tubes are contained in the SQN TSs,Section XI of the American Society of Mechanical Engineers Boiler and Pressure Vessel Code, and Regulatory Guide 1.83. SQN also implements the requirements of NEI 97-06. These include requirements for periodic inservice inspection of the tubing, flaw acceptance criteria (i.e., repair limits for plugging), and primary-to-secondary leakage limits. These requirements have formed the basis for assuring adequate SG tube integrity.

SG tube plugging limits are specified in SQN TSs. The current SQN TSs require that flawed tubes be removed from service by plugging if the depths of the flaws are greater than or equal to 40 percent through-wall, unless the degradation is subject to voltage-based outside diameter stress corrosion cracking (ODSCC) repair criteria or W* repair criteria. TS 6.8.4.k.c repair limits ensure that tubes accepted for continued service will retain adequate structural and leakage integrity during normal operating, transient, and postulated accident conditions, consistent with General Design Criteria 14, 15, 30, 31, and 32 of 10 CFR 50, Appendix A. Structural integrity refers to maintaining adequate margins against gross failure, rupture, or collapse of the SG tubing. Leakage integrity refers to limiting primary-to-secondary leakage to within acceptable limits.

The generic criteria for voltage-based limits for ODSCC are contained in GL 95-05. These criteria rely on empirically derived correlations between a nondestructive inspection parameter, the bobbin coil voltage, and tube burst pressure and leak rate. The GL guidance ensures structural and leakage integrity continue to be maintained at acceptable levels consistent with the requirements of 10 CFR 50 and the guideline values in 10 CFR 100.

GL 95-05 focuses on maintaining tube structural integrity during the full range of normal, transient, and postulated accident conditions with adequate allowance for eddy current test uncertainty and flaw growth projected to occur during the next operating cycle. In order to ensure the structural and leakage integrity of the tube until the next scheduled inspection, GL 95-05 specifies a methodology to determine the conditional burst probability and the total primary-to-secondary leak rate from an affected SG during a postulated main steam line break (MSLB) event. The methodology in WCAP-14277, Revision 1, "SLB Leak Rate and Tube Burst Probability Analysis Methods for ODSCC at TSP Intersections," dated December 1996, is used to implement the GL 95-05 structural integrity methodology.

E1-3

A probabilistic analysis to quantify the potential for SG tube ruptures given a MSLB event is performed per WCAP-14277, Revision 1, and compared to a reporting threshold value of 1x10- 2 per cycle as required by GL 95-05. This threshold value provides assurance that the probability of burst (POB) is acceptable considering the assumptions of the calculation and the results of the staff's generic risk assessment for SGs contained in NUREG-0844, "NRC Integrated Program for the Resolution of Unresolved Safety Issues A-3, A-4, and A-5 Regarding Steam Generator Tube Integrity."

Failure to meet this threshold value indicates ODSCC confined to within the thickness of the tube support plate (TSP) could contribute a significant factor of the overall conditional probability of tube rupture for all forms of degradation assumed and evaluated as acceptable in NUREG-0844.

The calculation of conditional burst probability is, in part, a function of the POD and the resulting indication voltage distribution at BOC. The indication voltage distribution at BOC is based on consideration of all previous bobbin indications that were detected at the BOC, including those that were plugged. The POB threshold value of 1xl0-2 per cycle is contained in SQN Unit 2 TS 6.8.4.k.b.

The voltage-based ODSCC repair criteria are described in SQN's Unit 2 TS Bases. TVA's enclosed changes include a discussion of the POPCD methodology within the TS Bases to describe the more realistic application and adjustment of POD values as a function of tube indication voltage.

3.2 Purpose for Proposed Amendment The use of a constant POD of 0.6 for determination of the indication voltage distribution for the BOC is nonconservative for indications below about 0.5 volt and excessively conservative for indications above 1 volt. The POPCD method to determine POD provides a more realistic POD that is a function of voltage.

4.0 TECHNICAL ANALYSIS

The POPCD methodology is described in a technical report that is included as Enclosure 4. The SQN Unit 2 POPCD database, as described in Section 5 of the technical report, satisfies the guidelines for the minimum number of inspections and for the minimum number of indications.

However, the database does not satisfy the third guideline for the number of indications in the upper voltage range.

The SQN Unit 2 POPCD database (see Section 5.2) includes E1-4

nine rotating pancake coil (RPC) confirmed indications above 2.0 volts and three above 3.0 volts compared to the guidelines of 20 and 5 indications, respectively.

Therefore, the POPCD distribution calculated from the SQN Unit 2 data is compared to the POPCD for the industry database for 7/8-inch tubing in Section 5.3. In Section 5.3, it is shown that there is no significant difference between the industry and the SQN Unit 2 POPCD distributions and the SQN Unit 2 plant specific POPCD can be applied for alternate repair criteria (ARC) analyses.

As part of the implementation of POPCD methodology, SQN Unit 2 will apply the preventive repair guidelines in Electric Power Research Institute (EPRI) Topical Report NP 7480-L, Addendum 6, "Steam Generator Tubing Outside Diameter Stress Corrosion Cracking at Tube Support Plates Database for Alternate Repair Limits," 2004 Database Update, dated October 2004 for repair of indications found to have a

+Point voltage greater.than 1.9 volts. The objective for this preventive repair is to aid identification of potentially deep flaws that could result in a high bobbin voltage growth due to through-wall penetration in the next cycle of operation. A 100 percent +Point inspection of bobbin indications greater than or equal to 1.7 volts is performed to support this preventive repair guideline.

EPRI developed an outlier growth method for application to ODSCC ARC coincident with the use of the POPCD method as described in Enclosure 4. Upon implementation of POPCD, TVA will review future inspection results at SQN Unit 2 and Diablo Canyon SGs for the presence of outliers (i.e.,

indications greater than about 8 volts) as a means of assessing the success of the preventive repair guideline to reduce the likelihood of outlier flaws. If an outlier is confirmed at either plant, TVA will implement the outlier method. To date, no outliers have been identified at either SQN Unit 2 or Diablo Canyon since implementing the preventive repair guideline. This includes two inspections at SQN Unit 2 (End of Cycle [EOC]-12 and EOC-13) for which bobbin indications greater than 1.5 volts were +Point inspected with no +Point indications exceeding the 1.9-volt threshold for preventive repair.

Growth rates used in the operational assessments upon approval for POPCD application are obtained as the bounding growth rate of the SGs and the composite average growth over the last two cycles of operation. When the composite average growth rate increased over the last operating cycle, the difference in average growth per effective full power year (EFPY) between the last two cycles is incrementally added to the bounding growth per EFPY distribution. With the exception of the adjustment for increasing average growth, this method of developing growth distributions is the same as applied in the SQN Unit 2 last two 90-day reports for projections at EOC-13 and EOC-14.

E1-5

SQN Unit 2 benchmarking analyses were performed and are included in Enclosure 4 (Technical Report for Probability of Prior Cycle Detection). Benchmark results indicate that for the methodology used to project the EOC conditions, while not conservative in all cases, the under predictions in general were not significant, except for one instance. In this instance, an extreme growth rate was observed for one indication, and the extreme growth is considered to be an outlier or random event.

In conclusion, based on industry and SQN Unit 2 specific bobbin detection data for ODSCC within the SG TSP region, large voltage bobbin indications that can individually challenge structural or leakage integrity can be detected with near 100 percent certainty and would not be left in service. These large voltage indications should not be included in the BOC voltage distribution, other than as inferred from the voltagedependent POD, for the purpose of the operational assessment. The POPCD method approach to POD considers the potential for missing indications that might challenge structural or leakage integrity by applying the POPCD data from successive inspections. The use of the POPCD method to determine the BOC voltage distribution will improve EOC projections and lead to appropriate estimates of the margin in SG tube structural and leakage integrity.

5.0 REGULATORY ANALYSIS

TVA is requesting to amend Operating License DPR-79 for SQN Unit 2. The requested change would allow use of the probability of prior cycle detection (POPCD) method to determine the beginning of cycle voltage distribution for steam generator (SG) operational assessments. TVA's proposed change provides an exception to NRC Generic Letter (GL) 95-05 for specifying a bobbin probability of detection (POD) or the bobbin indications, which are to be used for determination of the beginning of cycle (BOC) indication voltage distribution for operational assessments.

TVA has reviewed and evaluated the proposed changes for no significant hazards consideration determination as follows:

5.1 No Significant Hazards Consideration TVA has evaluated whether or not a significant hazards consideration is involved with the proposed amendment by focusing on the three standards set forth in 10 CFR 50.92, "Issuance of amendment," as discussed below:

1. Does the proposed change involve a significant increase in the probability or consequences of an accident previously evaluated?

Response: No The use of a revised SG voltage-based repair E1-6

criteria POD method, the POPCD method, to determine the BOC indication voltage distribution.

for the SQN Unit 2 operational assessments does not increase the probability of an accident.

Based on industry and plant-specific bobbin detection data for ODSCC within the SG tube support plate (TSP) region, large voltage bobbin indications which individually can challenge structural or leakage integrity can be detected with near 100 percent certainty. Since large voltage outside diameter stress corrosion cracking ODSCC bobbin indications within the SG TSP can be detected, they will not be left in service, and therefore these indications should not be included in the voltage distribution for the purpose of operational assessments. The POPCD method improves the estimate of potentially undetected indications for operational assessments, but does not directly affect the inspection results. Since large voltage indications are detected, they will not result in an increase in the probability of SG tube rupture accident or an increase in the consequences of a tube rupture or main steam line break (MSLB) accident.

Therefore, the proposed change does not involve a significant increase in the probability or consequences of an accident previously evaluated.

2. Does.the proposed change create the possibility of a new or different accident from any accident previously evaluated?

Response: No.

The use of the POPCD method is associated with numerical predictions of probabilities for the steam generator tube rupture (SGTR) accident.

Since the SGTR accident is considered in SQN's Updated Final Safety Analysis Report, there is. no possibility to create a design basis accident that has not been previously evaluated.

Therefore, the proposed change does not create the possibility of a new or different accident from any accident previously evaluated.

3. Does the proposed change involve a significant reduction in a margin of safety?

Response: No.

The use of the POPCD method to determine the BOC voltage distribution for the SQN Unit 2 E1-7

operational assessments does not involve a significant reduction in a margin of safety. The applicable margin of safety potentially impacted is the SG tube structural and leakage criteria.

Based on industry and plant-specific bobbin detection data for ODSCC within the SG TSP region, large voltage bobbin indications that can individually challenge structural or leakage integrity can be detected with near 100 percent certainty and will not be left in service.

Therefore, these indications should not be included in the voltage distribution for the purposes of operational assessments. Since these large voltage indications are detected, they will not result in a significant increase in the actual EOC leakage for a MSLB accident or the actual EOC probability of burst. The POPCD method approach to POD considers the potential for missing indications that might challenge structural or leakage integrity by applying the POPCD data from successive inspections. If a large indication was missed in one inspection, it would continue to grow until detected in a later inspection.

Accordingly, there is no significant increase in the margin of safety.

Based on the above evaluation, TVA concludes that the proposed change present no significant hazards consideration under the standards set forth in 10 CFR 50.92(c),-and accordingly, a finding of "no significant hazards consideration" is justified.

5.2 Applicable Regulatory Requirements/Criteria For SQN Unit 2 SGs, GL 95-05 requires the application of a POD value of 0.6 for all previous bobbin indications and the determination of the indication voltage distribution for the BOC. TVA's license amendment request (LAR) proposes use of other POD values (i.e., and methodology) that requires NRC approval as a deviation from GL 95-05.

In conclusion, based on the considerations discussed above, (1) there is reasonable assurance that the health and safety of the public will not be endangered by operation in the proposed manner, (2) such activities will be conducted in compliance with the Commission's regulations, and (3) the issuance of the amendment will not be inimical to the common defense and security or to the health and safety of the public.

6.0 ENVIRONMENTAL CONSIDERATION

TVA has evaluated the proposed amendment and has determined that the proposed amendment does not involve (i) a significant hazards consideration, (ii) a significant change in the types or E1-8

significant increase in the amounts of any effluent that may be released offsite, or (iii) a significant increase in individual or cumulative occupational radiation exposure. Accordingly, the proposed amendment meets the eligibility criterion for categorical exclusion set forth in 10 CFR 51.22(c) (9).

Therefore, pursuant to 10 CFR 51.22(b), no environmental impact statement or environmental assessment need be prepared in connection with the proposed amendment.

E1-9

ENCLOSURE 2 TENNESSEE VALLEY AUTHORITY SEQUOYAH NUCLEAR PLANT (SQN)

UNIT 2 Proposed Technical Specification Changes (mark-up)

E2-1

NOTE:, This page is impacted by TVA TS change 05-09, which is currently undergoing review for approval by NRC staff.

I REACTOR COOLANT SYSTEM SURVEILLANCE REQUIREMENTS (Continued)

e. The calculated steam line break leakage from the application of tube support plate alternate repair criteria and W* inspection methodology shall be submitted in a Special Report in accordance with 10 CFR 50.4 within 90 days following return of the steam generators to service (MODE 4). The report will include the number of indications within the tubesheet region, the location of the indications (relative to the bottom of the WEXTEX transition (BWT) and TTS), the orientation (axial, circumferential, skewed, volumetric), the severity of each indication (e.g., near through-wall or not through-wall), the side of the tube from which the indication initiated (inside or outside diameter), and an assessment of whether the.results were consistent with expectations with respect to the number of flaws.and flaw severity (and if not consistent, a description of the proposed corrective action).

7

f. For implementation of the probability of prior cycle detection (POPCD) method, for the voltage-based repair criteria at tube support plate intersections, if the end-of-cycle conditional tube rupture probability for a postulated main steam line break, the projected primary to secondary leak rate during a postulated main steam line break, or the number of indications are under predicted by the previous cycle operational assessment, the following shall be reported to the Commission within 90 days following return to service of the steam generators:

1. The assessment of the probable causes for the under prediction, proposed corrective actions, and any recommended changes to probability of detection or growth methodology indicated by potential methods assessments.

2. An assessment of the potential need to revise the alternate repair criteria analysis methods if:

the burst probability is under predicted by more than 0.001 (i.e., 10 percent of the performance criteria) or an order of magnitude; or the leak rate is under predicted by more than 0.5 gallon perminute (gpm) or an order of magnitude.

3. An assessment of the potential need to increase the number of predicted low voltage indications at the beginning of cycle if the total number of as-found indications in any SG are underestimated by greater than 15 percent or by greater than 150 indications.

May 3, 2005 SEQUOYAH - UNIT 2 3/4 4-14c Amendment No. 243, 291 E2-2

ENCLOSURE 3 TENNESSEE VALLEY AUTHORITY SEQUOYAH NUCLEAR PLANT (SQN)

UNIT 2 Changes to Technical Specification Bases Pages E3-1

NOTE: This page is impacted by TVA TS change 05-09 which Insert 1I is currently undergoing review for approval by NRC staff.

REACTOR COOLANT SYSTEM BASES where VGR represents the allowance for flaw growth between inspections and VNDE represents the allowance for potential sources of error in the measurement of the bobbin coil voltage. Further discussion of the assumptions necessary to determine the voltage repair limit are discussed in GL 95-05.

The mid-cycle equation of SR 4.4.5.4.a.10.e should only be used during unplanned inspection in which eddy current data is acquired for indications at the tube support plates.

SR 4.4.5.5 implements several reporting requirements recommended by GL 95-05 for situations which NRC wants to be notified prior to returning the S/Gs to service. For SR 4.4.5.5.d., Items 3 and 4, indications are applicable only where alternate plugging criteria are being applied. For the purposes of this reporting requirement, leakage and conditional burst probability can be calculated based on the as-found voltage distribution rather than the projected end-of-cycle voltage distribution (refer to GL 95-05 for more information) when it is not practical to complete these calculations using the projected EOC voltage distributions prior to returning the S/Gs to service.

Note that if leakage and conditional burst probability were calculated using the measured EOC voltage distribution for the purposes of addressing GL Sections 6.a.1 and 6.a.3 reporting criteria, then the results of the projected EOC voltage distribution should be provided per GL Section 6.b(c) criteria.

Wastage-type defects are unlikely with proper chemistry treatment of the secondary coolant. However, even if a defect should develop in service, it will be found during scheduled inservice steam generator tube examinations. Plugging will be required for all tubes with imperfections exceeding the repair limit defined in Surveillance Requirement 4.4.5.4.a. The portion of the tube that the plugging limit does not apply to is the portion of the tube that is not within the RCS pressure boundary (tube end up to the start of the tube-to-tubesheet weld).

The tube end to tube-to-tubesheet weld portion of the tube does not affect structural integrity of the steam generator tubes and therefore indications found in this portion of the tube will be excluded from the Result and Action Required for tube inspections. It is expected that any indications that extend from this region will be detected during the scheduled tube inspections. Steam generator tube inspections of operating plants have demonstrated the capability to reliably detect degradation that has penetrated 20% of the original tube wall thickness.

Tubes experiencing outside diameter stress corrosion cracking within the thickness of the tube support plate are plugged or repaired by the criteria of 4.4.5.4.a.10.

The W* criteria incorporate the guidance provided in WCAP-14797, Revision 2, "Generic W* Tube Plugging Criteria for 51 Series Steam Generator Tubesheet Region WEXTEX Expansions." W* length is the length of tubing into the tubesheet below the bottom of the WEXTEX transition (BWT) that precludes tube pullout in the event of.a complete circumferential separation of the tube below the W* length. W* distance is the distance from the top of the tubesheet to the bottom of the W* length including the distance from the top of the tubesheet to the BWT and measurement uncertainties.

Indications detected within the W* distance below the top-of-tube sheet (TTS), will be plugged upon detection. Tubes to which WCAP-14797 is applied can experience through-wall degradation up to the limits defined in Revision 2 without increasing the probability of a tube rupture or large leakage event. Tube degradation of any type or extent below W* distance, including a complete circumferential separation of the tube, is acceptable. As applied at Sequoyah Nuclear Plant Unit 2, the W* methodology is used to define the required tube inspection depth into the hot-leg tubesheet, and is not used to permit degradation in the W* distance to remain in service. Thus while primary to secondary leakage in the W* distance need not be postulated, primary to secondary leakage from potential degradation below the W* distance will be assumed for every inservice tube in the bounding steam generator.

May 3, 2005 SEQUOYAH - UNIT 2 B 3/4 4-3a Amendment No. 181, 211, 213, 243, 291 E3-2

INSERT 1 For the operational assessment, the Probability of Prior Cycle Detection (POPCD) voltage based probability of detection (POD) method, as described in a September 15, 2006 letter from Westinghouse Electric Company to TVA (LTR-CDME-06-121, "Technical Support for Application of Probability of Prior Cycle Detection for Sequoyah Unit 2 Voltage Based Alternate Repair Criteria"), is used to determine the beginning of cycle voltage distributions. The POPCD method is an exception to the GL 95-05 guidance that requires the application of a POD of 0.6 to all previous bobbin indications.

E3-3

ENCLOSURE'4 TENNESSEE VALLEY AUTHORITY SEQUOYAH NUCLEAR PLANT (SQN)

UNIT 2 Technical Report for Application of Probability of Prior Cycle Detection (POPCD)

E4-1

Technical Report for Application of Probability of Prior Cycle Detection (POPCD) 1.0 Introduction This document provides the technical bases for Sequoyah Unit 2 application of steam generator (SG) detection probabilities for the bobbin voltage alternate repair criteria (ARC) based on the Probability of Prior Cycle Detection (POPCD). The POPCD method is described in Reference 1 for generic industry applications with NRC GL 95-05 (Reference 2) voltage based repair criteria.

GL 95-05 requires NRC approval for implementing POPCD. NRC approval of POPCD was provided for the Diablo Canyon Power Plant (DCPP) Unit 2, Cycle 12 in Reference 3 and for both DCPP units in Reference 4. The POPCD requirements and data of this report reflect Reference 1 and the NRC review reflected in the DCPP approvals of References 3 and 4.

NRC approval for POPCD is requested for Sequoyah Unit 2 and the supporting plant specific POPCD data is provided only for Sequoyah Unit 2. The Sequoyah Unit 1 SGs were replaced in 2003 and no longer apply the ARC. In addition, Unit 1 had considerably more denting than Unit 2 and the Unit 2 data are more appropriate for POPCD applications. POPCD data for Sequoyah Unit 1 is included in the industry database used to compare the industry POPCD results with the Sequoyah Unit 2 POPCD results.

The POPCD method is described in Section 2 including the classification of indications for POPCD analyses. Methods for calculating POPCD and development of the industry database POPCD distribution are given in Section 3. Growth rate considerations for POPCD applications are provided in Section 4. Section 5 provides the Sequoyah Unit 2 POPCD evaluation including comparisons with the industry POPCD evaluation, and Section 6 summarizes benchmarking results for Sequoyah Unit 2 POPCD applications. Section 7 describes guidelines for continuing POPCD assessments and reporting for POPCD. Conclusions from this report are given in Section 8 and Section 9 lists the report references.

2.0 POPCD Method and Classification of Indications 2.1. POPCD Definition POPCD is calculated as the ratio of indications reported at the prior inspection to the total indications found at the subsequent inspection (all indications reported in the prior cycle plus new indications). POPCD for the (end of cycle) EOC inspection (EOCn) is defined as:

E4-2

EOCn bobbin EOCn÷I RPC confirmed detected, RPC plus not RPC confirmed plus not inspected and bobbin + inspected and POPCD detected at EOCn repaired at EOCn EOCn÷I RPC confirmed EOCn bobbin New EOCn÷j RPC plus not RPC + detected, RPC + confirmed or only RPC inspected and bobbin confirmed plus not detected plus not RPC detected at EOCn inspected and inspected indications repaired at EOCn (i.e., not detected at EOCn)

Notes for POPCD definition:

1. Rotating pancake coil (RPC) is intended to include an RPC probe or equivalent such as a +Point coil.

2. RPC confirmed means a bobbin indication signal was detected with an RPC probe.

3. Not RPC inspected means an indication signal was detected with a bobbin probe and was not further inspected with an RPC probe.

4. Bobbin indications that were RPC no detectable degradation (NDD) at EOCn and at EOCn+1 are either, (1) RPC detected or (2) bobbin detected without RPC inspection, are evaluated as new indications (undetected at EOCn) for consistency with the licensed DCPP POPCD of Reference 4. This represents a conservative difference from the industry POPCD recommendations of Reference 1 and NRC GL 95-05 (Reference 2) wherein bobbin detected indications are treated as detected independent of RPC confirmation.

This definition of POPCD is based on the premise that all indications that can contribute significantly to burst and leakage for voltage-based repair criteria application can be confirmed by RPC inspections. The term RPC is meant to include an RPC probe or equivalent, which includes a +Point coil. The POPCD definition that is used for outside diameter stress corrosion cracking (ODSCC) ARC analyses is based on RPC confirmed indications plus indications not RPC inspected.

This application of the RPC results is applied to exclude from POPCD the probable false bobbin calls that would not contribute to tube integrity concerns at EOCn+1 for both previously reported and new indications. The use of RPC confirmation provides.a measure of confidence that irrelevant prior and new bobbin calls are not allowed to influence the POD either as an increase or a decrease in the POD. The intent is that the EOCn+1 RPC should define the significance of the bobbin indication for POD considerations such that the insignificant RPC NDD indications are excluded from the POPCD method for both previously reported and new bobbin indications. EOCn bobbin calls that were RPC NDD at EOCn but were either bobbin or RPC detected or not RPC inspected at EOCn+2 are included in POPCD as bobbin undetected indications at EOCn.

E4-3

The POPCD approach treats all new indications at an inspection as having been undetected at the prior inspection even though some of the new indications may have initiated during the operating cycle. The application of POPCD for OA then accounts for newly initiated indications as well as previously undetected indications. This differs from the conventional POD definitions based on only detection of known flaws, but is applied for the operational assessments to eliminate the need for additional terms to account for newly initiated indications.

The single-cycle basis for a POPCD evaluation assumes that large flaws would be detected during the EOCn+1 inspection. This assumption is supported by the industry POPCD database of 42 inspections representing both 3/4- and 7/8-inch *tube diameters.

Single cycle basis means that the data is evaluated for one cycle to obtain a set of POPCD detection and non-detection data. The single-cycle data sets are then added to obtain a multi-cycle POPCD database. Voltage for an undetected indication would be based on a look-back analysis of the data for the first inspection prior to being detected to maximize the voltage as compared to a multi-cycle look-back to an earlier cycle to define non-detection. The multi-cycle development of a POPCD database, over sequential inspections provides additional assurance that the potential for large undetected indications is included in POPCD. If an indication is undetected, it can be expected to grow over subsequent inspections to a detectable level. When detected, it is included at the BOC voltage as an undetected or new indication. As an example for a 3-volt undetected bobbin indication, it can be expected that further growth would lead to detection at the next inspection and the indication would be included in POPCD as a 3-volt undetected indication. The industry data includes up to five successive inspections in a plant with noise levels (i.e., support plate residuals) significantly higher than currently active SGs. Nine inspections from the two units of the plant with high noise levels account for 70 percent of the new or undetected indications above 1.6 volts, including all three of the new indications above 2.5 volts in the 7/8-inch tubing industry database. However, none of the undetected indications exceeded 3.2 volts. The largest undetected indication in the Sequoyah Unit 2 database is less than 2.0 volts.

2.2. Classification of Indications and Assignment for POPCD Analyses Table 1 identifies all potential classifications for tracking bobbin coil indications between EOCn and EOCn+1 . The classification of an indication is dependent upon whether the indication is bobbin detected at each cycle and whether the indication is confirmed by RPC inspection or not RPC inspected.

The letters in the columns of Table 1 reflect the column for entering data in the Table 2 reporting format for POPCD data E4-4

results. The Table 2 format for reporting POPCD data is revised from that of the NP 7480-L Addendum 6 POPCD table format due to differences in evaluating EOCn RPC NDD indications. The format for reporting POPCD data in Table 2 is used to develop the POPCD distributions following the guidelines described in this subsection. The upper rows of each column in Table 2 provide the source of input data based on bobbin and RPC detection at EOCn and EOCn+1 . Guidelines for assigning bobbin voltages to bobbin NDD indications are provided at the end of this subsection. The rows show the voltage bin widths of 0.1 volt used for the Sequoyah Unit 2 POPCD evaluation. The Sequoyah Unit 2 indication classifications for POPCD are the same as approved by the NRC in Reference 4 for DCPP SGs.

Indications that were not detected at EOCn or EOCn+j by the standard bobbin inspection, but that are detected during mOCn+1 as a result of RPC inspection activities, are considered as new EOCn+j RPC confirmed indications for the POPCD analyses for EOCn (column F in Table 2). As shown in Section 5.2, there are only a few small indications at Sequoyah Unit 2 not detected via the standard bobbin inspection, but detected as a result of other RPC inspection activities. These indications would continue to be included as new indications in subsequent cycles unless the indication is reported in the normal bobbin coil inspection at a cycle following EOCn+÷. Thus, the indications will be considered as undetected at a minimum of two cycles, the EOC, and EOCn+1 inspections, even if repaired at the EOCn+÷ inspection.

The following indications are considered to be false bobbin coil calls and are not included in the POPCD analyses. The numbers are tabulated in column H in Table 2.

• Bobbin indications reported at EOCn but found to be NDD by RPC inspection at EOCn+1 . EOCn bobbin indications confirmed by RPC at EOC, and found to be RPC NDD at EOCn+j are not expected to occur but would be excluded from the POPCD analyses if they would occur since the Cycle n+l inspection would expect to find a larger indication if the EOCn indication was not a false call.

" Bobbin indications reported at EOCn but not found by the bobbin inspection at EOCn+1 . These indications are classified as indications not reportable (INR) and require resolution analysis to confirm that an indication is not present at EOCn+1 . Again, the Cycle n+l inspection would expect to find a larger indication if the EOCn indication was not a false call.

• New bobbin indications reported at EOCn÷+ but found to be NDD by RPC inspection at EOCn++/-. RPC NDD indications at EOCn+1 are assumed to be false bobbin calls for POPCD applications.

E4-5

The following guidelines are to be applied for determination of detection and bobbin voltages for the POPCD analyses:

• All determinations of detection at EOCn and voltages assigned to EOCn detected indications are generally based on the inspection records for the EOCn inspection. An exception may be applied for reevaluated EOCn voltages based on the detected EOCn+1 indications when developing growth rates although this reevaluation is not generally applied unless based on assessing an unusual occurrence such as a large negative growth rate. In the latter case, the reevaluated voltages are used in the ARC analyses and the POPCD evaluation.

Determination of detection at EOCn and at EOCn+1 is always based on the inspection records.

Lookback analyses are applied to assign voltages for new indications detected at EOCn+1 when EOCn voltages are not available from the inspection records.. When the lookback analysis leads to a voltage at EOCn greater than the EOCn+1 voltage, such as can occur for a low amplitude signal, the EOCn+1 voltage can be applied for EOCn.

• For indications remaining bobbin NDD at EOCn by the lookback analyses, the EOCn bobbin voltage should be obtained by subtracting the average growth rate from the EOCn+1 voltage.

When new EOCn+1 indications are found by RPC inspection but not reported as EOCn+1 bobbin indications, the voltages for EOCn+1 may be obtained by identifying the flaw based on a review of the 200 kHz bobbin data or by applying a site specific bobbin voltage to RPC voltage correlation (guidance for these methods is provided in Section 10.1 of Reference 1, Addendum 6 to the ARC database). The 400/100 kHz mix channel is the primary detection channel for tube support plate (TSP) ODSCC. The 200 kHz raw data channel may be evaluated to assist in the detection of small ODSCC indications. If upon review of the 200 kHz bobbin data, the indication is still not present, the indication is assigned a voltage using the bobbin voltage to RPC voltage correlation. These methods should also be applied to obtain bobbin voltages for EOCn indications found only by RPC and repaired at EOCn. In 90-day reports for SG voltage-based ARC that implement the POPCD method, TVA will assess bobbin voltages assigned to indications to verify that the assigned voltages are conservative. For prior cycle indications with assigned voltages that become detectable by bobbin, the review will assess the prior cycle assigned bobbin voltages, the-current cycle actual bobbin voltages, and growth rates.

• If the RPC inspection identifies more than one ODSCC indication at the same TSP intersection for which there is no reported bobbin indication, the bobbin voltage assigned to the E4-6

TSP is estimated as the square root of the sum of squares for the bobbin voltages inferred from the RPC indications. This is an approximation to the effect on bobbin voltage of multiple indications around the tube circumference. This technique is only used if the indications are only detected by rotating probe and a review of the 200 kHz bobbin coil data does not identify a flaw. Data supporting the adequacy of this technique was given in answers to RAIs from the DCPP submittal dated August 18, 2004.

• Indications found only by RPC inspection in deplugged tubes returned to service at EOCn are included as new indications at EOCn if found by bobbin and/or RPC at the EOCn+÷ inspection.

3.0 POPCD Analyses and Calculation of the Industry POPCD Distribution 3.1. Loglogistic Fitting The preferred approach to simulating the POD is to fit a loglogistic cumulative distribution function to the empirical data based on generalized linear model regression analyses. The statistically based POD distribution such as the loglogistic provides uncertainties in the POD distribution. Considerations on including POD uncertainties in Monte Carlo analyses are discussed in Section 3.2. The functional form of the loglogistic equation is, 1

S+e-[bo +b, log(V)]

where P is the POD, V is the bobbin amplitude, and bo and bi are parameters obtained by performing a regression analysis of the empirical POD data. The equation can be easily rearranged into the log-odds form as, In l*p= bo + b,log(V) (2) where the ratio in the parentheses is the odds of detection, i.e., the ratio of the POD to the probability of nondetection.

For application to the Sequoyah Unit 2 POPCD analyses, the data are sorted into 0.1 volt bins representing various voltage levels, and the POPCD distributions used for ARC analyses are developed using fits to weighted binary data (hit/miss equal 1/0), where weighting is based on the number of indications in each bin.

3.2. Monte Carlo Techniques The following is a description of the Monte Carlo techniques that are used to apply POPCD curves. The Monte Carlo analysis consists of simulating all of the indications in a SG several E4-7

thousand times. Each simulation of all of the indications in a SG is referred to as one simulation of the SG. If POD uncertainties are needed for the ARC analysis, each simulation of a SG calculates a set of random possible parameters for the POD equation (the intercept, slope and error standard deviation) that are applied to all of the detected indications to establish a population of detected and undetected indications. For a given POD, P, the number of indications that remain in service, N, in a given bin is given by, N = ND -NP' (3)

P where ND is the number of indications detected and Np is the number of indications plugged. Because the POD is a decimal value, the fraction in the above equation will not return an integer number of indications. The result is truncated to an integer value and a random draw from a uniform distribution is used to determine if an additional indication should be added to the total. If the value obtained from the uniform distribution is greater than the value of the fractional indication, an indication representing the fractional indication is not present.

Likewise, if the value is less than the fractional indication value then a whole indication representing the fractional indication is present and is included in the analysis to determine the probability of burst and potential total leak rate for that simulation of the SG. The application of this approach results in one additional indication being present in the prediction bin for a fraction of the SG simulations that matches the average value of the predicted fractional indication. For example, 10 indications in a bin with a POD of 0.33 gives rise to predicting that 30.3 indications were originally present. If the random draw from a uniform distribution is less than or equal to 0.3 the indication is present. If the number drawn is greater than 0.3 the indication is not present.

For each of the Monte Carlo simulations of a SG, the elements of the population variance-covariance matrix for the parameters of the loglogistic equation are found using the estimated values from the regression analysis and a random value from the Chi-Square distribution corresponding to the degrees of freedom associated with the regression analysis. Once the population values for the variance-covariance matrix have been calculated, population parameters of the POD equation, Po and P, corresponding to b 0 and bl, can be calculated using two random values from the standardized normal distribution based on the assumption that they are bivariate normally distributed. Given the population parameters, the POD for any indication voltage, Vi, associated with a bin can be calculated as,

{1 exp[- P0 - P, log(V,)} (4)

E4-8

where the P values are the estimated population parameters corresponding to the regression parameters b 0 and bl. For each bin of indications found during the current outage, the number of indications present that gave rise to finding the number reported for that bin is found by substituting the calculated P1 values into Equation 3.

The methodology to employ an analytic form for the POD is essentially identical to that used to simulate indications for evaluating the probability of burst and leak. If POD uncertainties are included in the Monte Carlo analysis, the process is repeated so that each simulation of all of the indications in a SG is independent of the other simulations of all of the indications in the SG. In this manner, thousands of variations of the possible levels of degradation within the SG can be considered. The determination of the POB and the potential leak rate during a postulated steam line break event proceeds according to the methodology outlined in GL 95-05.

Sequoyah Unit 2 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. The Sequoyah Unit 2 and industry POPCD uncertainties are small at 95/50 confidence (See Figures 1 and 3) due to the large number of indications in the database. Sampling of POD uncertainties leads to the same number of simulations with more than or less than the average number of indications for the mean POD. Due to the small uncertainties and the trend for the Monte Carlo results to converge to that obtained for the average number of indications, the effects of including POD uncertainties in the Monte Carlo process are negligible. It can also be noted that when the POD uncertainties are significant, the Monte Carlo sampling process leads to changes in the mean POD distribution near 0 percent and 100 percent POD. The latter results change since applying uncertainties near 0 percent POD are cutoff. at 0 percent (smaller uncertainties applied below the mean POD than above the mean POD) with a resulting increase in the mean POD near the 0 percent value. Near 100 percent POD, the uncertainties are cutoff at 100 percent above the mean POD but not below the mean POD with a resulting decrease in the mean POD near the 100 percent value. These effects are found when Monte Carlo sampling a POD distribution with uncertainties and comparing the mean POD after sampling with the input mean POD.

For applications of the industry or Sequoyah Unit 2 POPCD, these cutoff effects would be negligible due to the small POD uncertainties.

The p-value from the POPCD regression analysis is the probability of observing a value of X 2 as small as the one calculated from the data. If the p-value is found to be greater than 5 percent (i.e., the probability of randomly observing a value as small as the one calculated would be greater than 5 percent), TVA will propose an alternate POD and submit the recommendation to the NRC for approval to apply for SQN Unit 2 operational assessments.

The associated implication would be that there could be E4-9

sufficient noise at the location of the indications to interfere with the detection of the indications. For the Sequoyah Unit 2 regression results presented in Section 5, the p-values were calculated to be effectively zero.

Single-cycle POPCD distributions can be expected to change from cycle-to-cycle dependent upon the number and size of indications in a given inspection. For inspections with a large number of indications spanning up to at least 5 volts, the differences between a single-cycle POPCD and a combined cycle POPCD would be expected to be modest. Multi-cycle POPCD distributions are applied for ARC applications. The integrated inspection experience is then included in POPCD to represent the history of undetected indications. For example, if the last inspection resulted in the largest undetected voltage indication, the last inspection results do not imply similar occurrences for the next cycle and the overall integrated history would be the best estimate for the next cycle. This conclusion is applicable as long as SG conditions at TSP intersections do not significantly degrade with operating time, which is typical of operating times after the first one or two cycles during which the residual signals form at TSP intersections. Significant increases in denting are not occurring in the Sequoyah Unit 2 SGs.

3.3. Guidelines for Applying Plant Specific POPCD Distributions Application of a plant specific POPCD requires satisfaction of minimum data requirements for which guidelines are given in the ARC database Addendum 6 (Reference 1) and described below. If a plant does not meet the minimum data requirements, the industry POPCD distribution must be applied unless it is shown that the plant specific POPCD distribution for ARC analyses yields a uniformly lower POD above 1.0 volt and has a POD less than the industry POPCD in the range of 0.1 to 0.2 volt. Alternately, a plant specific POPCD can be constructed by applying the lower of the plant specific and industry POPCD distribution above 1 volt and the lower of the two distributions below 1.0 volt, particularly near about 0.2 volt. These options permit application of conservative POPCD distributions while the plant specific database is increased to meet the guidelines for applying a plant specific POPCD distribution.

Guidelines for applying a plant specific POPCD distribution are:

Minimum Number of Inspections for Plant Specific POPCD Application Minimum of three consecutive inspections for a plant with one unit applying the ARC and a minimum of four inspections with at least two inspections in each unit for a plant with two units applying the ARC. If a two unit plant has the minimum of three consecutive inspections for one unit, that unit can apply a plant specific POPCD.

E4-10

Minimum Total Number of Indications for Plant Specific POPCD Application The total number of indications must be at least 500 for a plant specific application. This represents a very conservative population for defining a POD distribution when the data satisfies the higher voltage range requirement given below.

Minimum Number of Detected and RPC Confirmed Indications in Upper Voltage Range The POPCD database must include a minimum of 20 detected and RPC confirmed indications (1) above 1.0 volt with at least 5 indications above 2.0 volts for a plant with 3/4-inch tubing or (2) above 2.0 volts with at least 5 above 3.0 volts for a plant with 7/8-inch tubing. The minimum of 20 indications is typical of a lower range for statistical acceptability to define a total POD distribution for reasonably distributed indications, and is typical of current Electric Power Research Institute (EPRI) ETSS requirements. The requirement of at least 5 indications greater than 1 volt above the GL 95-05 ARC repair limits assures detected indications near the voltage of about 3 volts for which the POD is expected to be near unity.

The minimum of 20 indications above the repair limit also shows that the ODSCC population has matured to repairable indications to support the POD development.

If the plant specific POPCD satisfies the above minimum data requirements, there is no need to compare the plant specific and industry POPCD distributions to select the more conservative distribution. The database for POPCD would be much larger than generally applied to develop POD distributions (e.g., EPRI ETSS) with a representative sample of indications at least 2 volts above the ARC repair limits.

The Sequoyah Unit 2 POPCD database, as described in Section 5, satisfies the above guidelines for the minimum number of inspections and for the minimum number of indications. However, the database does not satisfy the third guideline for the number of indications in the upper voltage range. The Sequoyah Unit 2 POPCD database (See Section 5.2) includes 9 RPC confirmed indications above 2.0 volts and 3 above 3.0 volts compared to the guidelines of 20 and 5 indications, respectively. Therefore, the POPCD distribution calculated from the Sequoyah Unit 2 data is compared to the POPCD for the industry database for 7/8-inch tubing in Section 5.3. In Section 5.3, it is shown that there is no significant difference between the industry and Sequoyah Unit 2 POPCD distributions and the Sequoyah Unit 2 plant specific POPCD can be applied for ARC analyses.

E4-11

3.4. Noise Considerations for POPCD Applications The POPCD approach to detection probabilities considers the potential for missing indications that might challenge structural or leakage integrity. Detection is sensitive to the noise levels at the TSP intersection. Since it is shown in Section 5.3 that the Sequoyah Unit 2 POPCD database can be applied for ARC analyses based on adequately satisfying the guidelines given above in Section 3.3, the Sequoyah Unit 2 noise levels are included in the POPCD database and no supplemental noise analyses are required.

3.5. POPCD Evaluation for the Industry Database The POPCD industry database was updated in EPRI Topical Report NP 7480-L, Addendum 6 (Reference 1). The industry POPCD data for 7/8 inch diameter tubing were updated for this report to provide the latest industry POPCD for comparisons with the Sequoyah Unit 2 POPCD. Updates are included for the last inspection of Diablo Canyon Unit 1 and Sequoyah Unit 2. The POPCD data were obtained from 30 inspections in plants with 7/8-inch tubing including Sequoyah Unit 1 data up to the time of SG replacement. Table 2 provides the resulting POPCD database for 7/8-inch tubing. Since Sequoyah Unit 2 has 7/8-inch tubing, the comparison with the industry POPCD for 7/8-inch tubing is appropriate.

The bin mid-range voltages of the POPCD data tables are used to develop the loglogistic POPCD distribution. The generalized linear model (GLM) regression analyses of the data are based on weighted binary input as described in Section 3.1.

Figure 1 shows the industry POPCD loglogistic fit to the 7/8-inch data including the lower 95 percent confidence bounds on the POPCD distribution. The lower 95 percent confidence bounds show only a small reduction relative to the nominal regression fits, which supports small uncertainties for the industry POPCD distributions based on the large databases available. The updated POPCD of Figure 1 is essentially the same as reported in Addendum 6 (Reference 10) with a reduction in the POD of less than 2 percent near 0.1 volt as the only discernable change.

The industry POPCD databases are well-defined relative to the number of indications in the database up to about 3.1 volts with at least 10 indications in each 0.1-volt bin. From Table 2, all indications above 3.2 volts for the industry database are detected. However, the limited number of indications above 3.2 volts leads to a reduction in the POD below unity as seen by the reduction from 1.0 for the fraction detected at about 3.5 volts to about 0.97 for the nominal regression fit of Figure 1.

The undetected indications in the industry POPCD database above approximately 1.6 volts are dominated by the data from 2 units (9 of the 30 inspections, SGs with 7/8-inch tubing since replaced) that had very high noise levels at the TSP E4-12

intersections. The 9 inspections in these 2 units account for 70 percent of the new indications.(POPCD assumption of missed indications) above 1.6 volts in Table 2 including all 3 of the new indications above 2.5 volts. The 3 new indications above 2.5 volts and the total of 18 new indications in the 2.0- to 2.5-volt range all occurred in 1994 to 1997 inspections. This supports improved detectability since 1997 based on increased analyst experience for ARC inspections. For the SGs still operating, the industry database is expected to be conservative above about 1.6 volts.

The industry database in Table 2 above 3.2 volts includes 97 indications all of which were detected. There have been about an additional 50 indications above 3.2 volts detected and repaired at the last inspection before SG replacement, which are not included in POPCD evaluations since there was no subsequent inspection for POPCD evaluation. The large number of indications above 3.2 volts provides a good statistical database to define a complete POD distribution and helps to define the higher voltage range POD. Given the large numbers of indications above the lowest undetected indications in the database, there is no need to truncate the POPCD curve or transform the curve to a horizontal line at any voltage.

Overall, it is concluded that the upper-voltage range (above 2 volts) uncertainty is adequately addressed by the industry database and the statistical GLM regression analyses described previously.

3.6. Benchmarking of Industry POPCD Extensive benchmarking has been performed for the industry POPCD data as described in the ARC database Addendum 6 (Reference 1).

These benchmark analyses strongly support the use of POPCD for ARC analyses. Benchmarking analyses were also performed by DCPP to show the adequacy of using a DCPP POPCD distribution (References 5 and 6). Monte Carlo burst probability and leak rate projections were performed for multiple cycles using the DCPP POPCD. The results of these analyses indicate that the methodology used to predict EOC conditions were generally although not always conservative. The under predictions were generally not significant relative to the as-found inspection results. Under predictions when applying POPCD are more likely to be due to growth rate uncertainties than POPCD uncertainties.

The reference benchmarking results support adequacy of the POPCD methodology for ARC applications.

Benchmarking results for the Sequoyah Unit 2 POPCD are described in Section 6.

4.0 Growth Rate Considerations for POPCD Applications E4-13

4.1. Provision for Outlier Growth Rates Calculations applying POPCD or 0.6 as the POD will not predict flaws which result from voltage growth rates that are higher than previously seen (e.g., the DCPP R44C45-2H flaw which was detected as a 21.5-volt flaw at DCPP in the 2RII inspection and was detected as a 2.0-volt flaw in 2R10 and left in service per the ODSCC ARC repair criteria) and both POD methods can lead to underestimates of the maximum flaw size, burst probability, and leak rate. The overly conservative and arbitrary application of a POD of 0.6 has not changed any assessments for corrective actions following identification of a large growth rate indication. For smaller growth rate under predictions, the application of a POD of 0.6 can mask a real growth rate issue by leading to artificially high burst and leakage predictions with an associated conclusion that no corrective action is necessary.

A high-voltage growth rate can be expected to periodically occur and cannot be predicted for a specific cycle. Normal growth in depth can lead to a large voltage increase when the upper range of depth growth occurs for a near throughwall or short throughwall indication. This occurs as the result of the bobbin coil's exponential dependence of voltage on depth and again on throughwall length and is inherent to the voltage-based methods for tube integrity assessments. For many occurrences of large growth rates with associated large EOC voltages (i.e., greater than 8 volts), the indications have been pulled and destructively examined. When the largest indications found in ARC inspections have been destructively examined, all indications have demonstrated burst pressures exceeding APSLB and leak rates generally consistent with the ARC correlations.

Sequoyah Unit 2 applies the preventive repair guidelines in Addendum 6 (Reference 1) for repair of indications found to have a +Point voltage greater than 1.9 volts. The objective for this preventive repair is to aid identification of potentially deep flaws that could result in a high bobbin voltage growth due to throughwall penetration in the next cycle of operation. A 100 percent +Point inspection of bobbin indications greater than or equal to 1.7 volts is performed to support this preventive repair guideline.

EPRI has developed an outlier growth method for application to ODSCC ARC coincident with the use of the POPCD method as described in Reference 1. Upon implementation of POPCD, TVA will review the future inspection results at Sequoyah Unit 2 and Diablo Canyon SGs for the presence of outliers (i.e., indications greater than about 8 volts) as a means of assessing the success of the preventive repair guideline to reduce the likelihoOd of outlier flaws. If an outlier is found at either plant, TVA will implement the outlier method upon NRC approval of POPCD and the outlier methodology. To date, no outliers have been found at either Sequoyah Unit 2 or Diablo Canyon since implementing the E4-14

preventive repair guideline. This includes two inspections at Sequoyah Unit 2 (EOC-12 and EOC-13) for which bobbin indications greater than 1.5 volts were +Point inspected with no +Point indications exceeding the 1.9-volt threshold for preventive repair. Application of this guideline at Diablo Canyon SGs over five inspections has led to preventive repair of four indications.

4.2. Development of Growth Rate Distributions Conservative growth rates should be used in the operational assessments for ARC analyses. The outlier growth method noted above is one technique for adding conservatism to the growth distributions. GL 95-05 requires the use of the most limiting of the two previous growth rate distributions for the next operating cycle, and WCAP-14277, Revision 1 (Reference 7) recommends that the ARC analyses should use the more conservative growth (relative to predicting burst and leakage) between the distributions for the specific SG and for all SGs collectively for the two previous growth rate distributions.

Upon implementation of POPCD, Sequoyah Unit 2 will apply a bounding growth distribution based on bounding the growth distribution of all SGs over the last two cycles. This bounding growth distribution will be applied for all SGs. This provision is applied based on the observation that the limiting growth distribution including the largest growth values often shifts between SGs in successive cycles. This bounding approach for developing growth rates has been applied for the last two operational assessments at Sequoyah Unit 2. Upon implementation

• of POPCD, if the composite average growth rate for all SGs shows an increase for the just completed cycle compared to the prior cycle, the incremental increase in average growth per EFPY will be added to each point in the bounding distribution. This provision provides an allowance for potential progressive increases in growth for the next cycle.

GL 95-05 notes that a single growth distribution in terms of AV rather than percent AV can be used provided the conservatism of this approach continues to be supported by operational experience. This requires an assessment for the potential onset of voltage dependent growth (VDG). The onset of VDG can generally be seen from plots of voltage growth versus BOC voltage. The occurrence of a higher frequency of large growths or increased growth rates with increased BOC voltage is an indication of the presence of VDG. Methods for calculating voltage dependent growth rate distributions are described in Reference 1. These methods define techniques for determining break points in voltage bins with each bin defining a separate growth distribution. When applying POPCD, the assessment for the onset of VDG should be performed and the methods of Reference 1 applied when growth rates show a dependence on the BOC voltage.

E4-15

Up to the last EOC-13 inspection at Sequoyah Unit 2, the growth rates do not show the presence of voltage dependent growth (Reference 8 includes the EOC-13 assessment for VDG). The assessment for VDG will be included in future 90-day report operational assessments and the VDG methods will be applied if VDG is found for the growth data.

5.0 POPCD Evaluation for Sequoyah Unit 2 5.1. Inspection Considerations for the Sequoyah Unit 2 ARC The following inspections are performed at Sequoyah Unit 2 for ODSCC indications at TSPs in support of the voltage based ARC:

• 100 percent bobbin inspection from the hot leg tubesheet to the cold leg tubesheet.

• 100 percent +Point inspection of all bobbin DSI indications greater than or equal to 1.7 volts. The 1.7 bobbin volt inspection threshold is established for preventive repair considerations based on the guidelines of Section 8.3 of Addendum 6 (Reference 1). All +Point indications greater than 1.9 volts will be preventively repaired independent of the bobbin volts.

0 A 100 percent +Point inspection of all hot leg dents greater than or equal to 2 volts.

• +Point inspection of mixed residual indications identified by lead analyst review as potentially masking a 1.0 volt ODSCC indication.

• A +Point inspection at EOCn÷+ will be performed for +Point confirmed indications at EOCn (either bobbin detected or bobbin NDD) that are bobbin NDD at EOCn÷+.

The +Point inspection at EOCn+1 for +Point confirmed indications at EOCn that are bobbin NDD at EOCn÷1 is added per the footnotes to the NRC request (Table 1, Note 3).

Up to the last Sequoyah Unit 2 inspection at EOC-13, no circumferential cracks (SCIs) have been found at dented TSPs.

Denting at TSPs in Sequoyah Unit 2 is modest compared to that found in Sequoyah Unit 1. On the order of 2040 dents have been found in Sequoyah Unit 2 at the EOC-13 inspection compared to the order of 15000 dents in Sequoyah Unit 1 through the EOC-10 inspection. The Sequoyah Unit 2 dents include about 1175 hot leg TSP intersections, out of a total of 23716 hot leg intersections and about 865 cold leg TSP intersections. The largest number of dents at any one TSP is about 156 at H07 in SG 3. Based on the modest denting in Sequoyah Unit 2, the occurrence of axial primary water stress corrosion cracking (PWSCC) at dented TSPs has been negligible. The PWSCC ARC is not applied at Sequoyah Unit 2 and indications found to have axial PWSCC or SCIs are repaired. There have been no. occurrences of axial ODSCC and PWSCC at the same TSP intersection.

E4-16

Upon implementation of POPCD, all TSP bobbin flaw indications will be assigned the flaw code DSI representing ODSCC unless a prior or current +Point inspection has identified a volumetric flaw. When +Point inspection follows the bobbin inspection, the bobbin flaw code will be changed if volumetric, PWSCC or circumferential cracks are found. This bobbin flaw code assignment conservatively assumes that all bobbin calls at TSPs are axial ODSCC unless confirmed by +Point to be other than axial ODSCC. All bobbin calls in peripheral cold leg TSP locations have been +Point inspected to confirm the tube damage mechanism.

Similarly, new bobbin indications in this area will be +Point inspected to confirm the tube damage mechanism. A PCT code is assigned to bobbin calls confirmed by +Point to be volumetric indications.

To date, there have been only six occurrences of indications found by +Point that were not detected by the bobbin coil since the ARC was implemented in 1996 at EOC-7. The +Point volts for these six indications ranged from 0.16 to 0.78 volt. All ODSCC indications detected only by +Point were repaired when detected.

Figure 2 shows the bobbin to +Point volts correlation developed for EOCs 8 to 13 that is applied in the POPCD analyses to obtain bobbin volts for indications detected only by +Point. The correlation shows a high degree of scatter in bobbin voltages for

+Point volts less than 0.5 volt and has limited data above 1.0 +Point volt. Given the few number of indications found only by +Point at Sequoyah Unit 2, the uncertainty in the bobbin to

+Point voltage correlation would have negligible impact on the burst and leak rate analyses. This correlation will be applied to obtain bobbin voltages for determining the need for tube repair and for inclusion in the ARC condition monitoring and operational assessments. The correlation will be updated in the ARC 90-day report following each outage. Upon implementation of POPCD, for indications found only by +Point that are left in service based on the bobbin to +Point volts correlation and later become detectable by bobbin, a review in the 90-day report will assess differences between voltages inferred from the correlation and the measured bobbin voltages.

5.2. Sequoyah Unit 2 POPCD Data and Evaluation Sequoyah Unit 2 POPCD data were evaluated based on inspections at EOCs 7 to 13 which assess the composite POD at EOCs 7 to 12. The ODSCC ARC was implemented at EOC-7 in 1996. The evaluation uses nondestructive examinations (NDE) history files of final inspection results for EOCs 7 to 13 to develop the POPCD data.

An evaluation of POPCD at EOC-12 was also performed to compare the last completed cycle POPCD with the composite POPCD for all six completed ARC cycles. This evaluation is described at the end of this section.

The composite POPCD data are provided for the tracking table in Table 3, for the POPCD data table in Table 4 and for a plot of E4-17

the resulting POD in Figure 3. The Table 3 tracking table provides two sub tables for all volts and for less than or equal to 1.0 volt. The weighted GLM analysis for the POPCD distribution of Figure 3 is based on the data in Columns I and J of Table 4 for detected and undetected indications at EOCn.

Table 7 provides the parameters for the log logistic functions used for the POPCD distributions. The regression p-values for each of the POPCD distributions in Table 6 are less than

2. 9xi0- 7 .

The POPCD distribution of Figure 3 shows a POD less than 0.6 below about 0.3 volt. At 1.0 volt, the POD is about 0.88 and increases to about 0.97 at 3 volts. The overall average POD is about 0.7 as shown in the lower right hand corner of Table 4 based on total detected and undetected indications. The Sequoyah Unit 2 POD is well above the GL 95-05 guideline of 0.6 for significant.bobbin indications. From Figure 3, it is seen that the POPCD uncertainties are small even at the 95 percent confidence level.

The largest missed bobbin indication was found for SG 3, R32C17 H01, which had a bobbin indication of 2.03 at EOC-10 and was bobbin NDD at EOC-9. No reevaluation of the EOC-9 bobbin data was performed so the assigned bobbin volts was obtained by subtracting the average growth of 0.10 volts from the EOC-10 measurement to assign 1.93 volts for the missed indication at EOC-9. The second largest missed indication was 1.62 volts for SG 2, R9C51 which, was found only by +Point with SAIs of 0.63 and 0.40 volt for which a bobbin voltage of 1.62 volts is inferred from the bobbin to RPC voltage correlation. The largest missed indication based on bobbin reanalysis was 1.44 volts for SG 4, R23C54 at EOC-12, which had a reported DSI of 1.54 volts at EOC-13.

The composite POPCD data for indications found only by RPC in Column F of Table 4 shows 11 occurrences although there were only 6 actual occurrences. The RPC only detections are counted as missed indications in two successive cycle POPCD evaluations even though they were repaired upon RPC detection. The last occurrence was in the EOC-13 inspection and the second occurrence for the composite evaluation is not yet included in Table 4.

The results of the POPCD evaluation for the last operating cycle (evaluation at EOC-12 based on EOC-13 inspection results) are shown in Table 5 for the database tracking table, Table 6 for the POPCD data table and Figure 4 for the plot of the POD distribution. It is seen in Figure 4 that POPCD for the last cycle represents an improvement in detection compared to the composite data for EOC-7 to EOC-12. The increase in POD below about one volt is primarily due to fewer new indications than found for the historical database. The slight decrease in POD above 1.0 volt is more a reflection of raising the low volt POD E4-18

for a log logistic function than any changes in POD above 1.0 volt. Although the six cycle composite POPCD is considered more appropriate for future operational assessments, the EOC-12 POPCD results indicate probable improvement in POD over the last cycle of operation.

Tables 3 and 5 provide the NRC required POPCD data tracking tables for the EOC-7 to EOC-12 and the EOC-12 POPCD evaluations, respectively. The upper parts of the tables provide the data for all indications and all voltages and the lower parts of the tables provide the data for bobbin voltages less than or equal to 1.0 volt. These tables break down the indication counts to reflect all possible combinations of bobbin and inspection results between the two successive cycles included in a POPCD evaluation. For example, the results show that there were 32 historical occurrences of indications bobbin detected but RPC NDD at EOCn that were bobbin detected at EOCn++/- without RPC or confirmed by RPC. These indications are included as new undetected indications in the POPCD analyses although included in ARC operational assessments as detected indications.

5.3. Comparisons with Industry POPCD Figure 5 compares the Sequoyah Unit 2 POPCD distribution of Figure 3 with the generic industry 7/8-inch tubing POPCD distribution of Figure 1. It is seen that the two distributions are essentially identical above about 0.7 volt. Below 0.7 volt, the Sequoyah Unit 2 POD is slightly higher than the industry value reflecting slightly smaller occurrence of new low voltage indications. Since the Sequoyah Unit 2 POPCD database does not satisfy all the guidelines of Section 3.3 (guideline for numbers of indications above 2.0 volts are not satisfied as described in Section 3.3) for applying plant specific POPCD distributions, the guidelines recommend that the lower of the industry and plant specific POPCD above 1.0 volt be applied together with the lower POD in the range of 0.1 and 0.2 volt. Since there is no discernable difference in the industry and Sequoyah Unit 2 POPCD above 1.0 volt, the Sequoyah Unit 2 POPCD satisfies the guidelines in this voltage range. In the 0.1- to 0.2-volt range, the Sequoyah Unit 2 POPCD is less than 5 percent higher than the industry POPCD. The higher Sequoyah Unit 2 POPCD in this range is largely due to the improved low voltage POD found for the EOC-12 inspection compared to the prior cycle values. The small difference between the industry and Sequoyah Unit 2 POPCD distributions, particularly at the lower 95 percent confidence level, does not warrant application of the industry POPCD for ARC analyses. The use of the Sequoyah Unit 2 POPCD distribution permits continued refinement of the data with future inspection results.

5.4. Comparisons with DCPP POPCD A comparison of the Sequoyah Unit 2 POPCD distribution with the plant specific DCPP POPCD (Reference 10) approved by the NRC in Reference 4 is shown in Figure 6. The Sequoyah Unit 2 POPCD is E4-19

notably higher than that for DCPP in the low voltage range and slightly lower above one volt. The difference in the low voltage range reflects a trend for fewer new indications at Sequoyah Unit 2 than at DCPP. For the range above one volt that is important for operational assessments, the differences are negligible.

6.0 Benchmarking of POPCD for Sequoyah Unit 2 6.1. Benchmark Analysis Methods and Data Analyses to benchmark the POPCD methods were performed to compare projections with the as-measured results at EOCs 11 to 13. The benchmarking analyses use the same burst and leakage correlations used in the 90-day reports (References 9 to 11) to calculate the as-measured steam line break (SLB) burst probabilities and leak rates. The indications left in service at the start of. each cycle (without POD correction) were also taken from the 90-day reports.

As noted in Section 4.2, growth rates used in the operational assessments upon approval for POPCD application are obtained as the bounding growth rate of all SGs and the composite growth over the last two cycles of operation. When the composite average growth rate increased over the last operating cycle, the difference in average growth per EFPY between the last two cycles is incrementally added to the bounding growth per EFPY distribution. With the exception of the adjustment for increasing average growth, this method of developing growth distributions is the same as applied in the last two Sequoyah Unit 2 90-day reports for projections at EOC-14 and EOC-15.

Figure 7 shows the trend in average growth rates at Sequoyah Unit 2. For Cycle 10, the average growth decreased compared to Cycle 9 and no growth rate adjustment is applied for the EOC-I1 predictions. For Cycles 11 and 12, the increases in average growth were 0.023 volt and 0.016 volt, respectively, and these increases are added incrementally to each point in the bounding growth rates used for the EOC-12 and EOC-13 projections.

The POPCD distribution developed in Section 5.2 at the lower 95 percent confidence level and given in Figure 3 was used for all cycles in the benchmark analyses. Since Sequoyah Unit 2 did not satisfy the requirements of Section 3.3 in prior cycles, the guidelines would have required use of the industry POPCD in prior cycles which is essentially the same as the Sequoyah POPCD as shown in Section 5.3. The use of the Sequoyah Unit 2 POPCD permits a more direct assessment of the POPCD methodology application.

6.2. Benchmark Analysis Results Comparisons of projected SLB burst probability, SLB leak rate and numbers of indications are compared with the as-found results for E4-20

EOC-1I to EOC-13 in Table 8. The upper part of Table 8 provides the POPCD results and the low part of the table provides the analysis results for POD equal to 0.6 as reported in the 90-day reports.

From Table 8, it is seen that the POPCD projections for EOC-ll are conservative for SGs 1 to 3 compared to the values obtained from the as-found voltage distribution. For SG 4, SLB leakage and the number of indications are conservatively predicted compared to the as-found results although burst probability and maximum voltage are slightly under predicted. The burst probability under prediction of 2x10- 5 is negligible and much less that the guideline value in Section 7.0 of ix10- 3 for assessing the potential need to revise the analysis methods.

At EOC-12, an outlier indication of 9.76 volts was found in SG 3 that could not be predicted from the prior cycle inspection results. This led to a significant underestimate in the SLB burst probability and leakage for SG 3 that is independent of the choice of POD for the operational assessment. This is the only Sequoyah Unit 2 inspection that found an unusually large voltage indication including the later Cycle 13 inspection results. The number of indications in SGs 1 to 3 is also significantly underestimated by both POPCD and POD equal to 0.6. The number of indications found in these three SGs essentially doubled at EOC-12 compared to EOC-ll, which would not be predicted by any POD under consideration for ARC analyses. This large increase in the number of indications was not found in Cycles 11 or 13. The large number of new indications for Cycle 12 contributes to a reduction in the low voltage POPCD for Sequoyah Unit 2. Even though the number of indications is under estimated in SGs 1 and 2, the burst and leak results are conservative compared to the as-found results. Since the EOC-13 POPCD predictions for the number of indications are essentially equal to or greater than the as-found results, there is no further need for a methods adjustment for the number of indications. As a result of finding the large growth rate at EOC-12, the methods for applying voltage growth were modified in the 90-day reports for Cycles 13 and 14 and also applied for the POPCD growth rate guideline. This change is to develop the growth rate for operational assessments based on the bounding growth rate of all SGs and to apply this bounding growth rate in the operational assessments for all SGs.

At EOC-13, all predictions are conservative compared to the results obtained from the as-found voltage distribution. This result is largely'due to the inclusion of the large voltage growth rate found at EOC-12 and no large voltage indication found in the EOC-13 inspection.

Overall, it is concluded that the benchmark results indicate that for the methodology used to project the EOC conditions, while not conservative in all cases, the under predictions in general were not significant except for one instance. In this instance, an extreme growth rate was observed for one indication, and the extreme growth is considered to be an outlier or random event.

E4-21

In addition, Sequoyah Unit 2 has committed to perform a rotating probe inspection of all bobbin indications above 1.7 volts to address the potential cause for the outlier growth rate. Since the methods have not always been conservative, Sequoyah Unit 2, upon implementation of POPCD, commits to perform an assessment of the adequacy of the methodology, under certain conditions, for performing operational assessments as defined in Section 7.0.

6.3. Projections for EOC-14 The operational assessment projections for EOC-14 to be completed in the fall of 2006 are given in Table 9 for POPCD and for POD equal to 0.6. The results for POD equal to 0.6 are the same as reported in the 90-day report of Reference 8.

The growth rates used for POPCD and for POD equal to 0.6 are the same bounding growth distribution for Cycles 12 and 13. The large growth rate found in Cycle 12 SG 3 is included in the projections for all SGs since the same bounding growth rate is applied for each SG projection. As would be expected, the projections for burst probability, leakage, and number of indications applying POPCD are slightly lower than obtained with a POD of 0.6.

7.0 Continuing Assessment and Reporting for POPCD Upon implementation of POPCD, if the EOC conditional SLB burst probability, the SLB leak rate or the number of indications are under predicted by the previous cycle operational assessment, the following guidelines will be applied to assess the need for methods adjustments and the assessments will be documented in the ARC 90-day report:

" An assessment of the probable causes for the under predictions, proposed corrective actions, and any recommended changes to the probability of detection or growth methodology indicated by potential methods assessments.

" An assessment of the potential need to revise the ARC analysis methods if: the burst probability is under predicted by more than 0.001 (i.e., 10 percent of the reporting threshold); or the SLB leak rate is under predicted by more than 0.5 gallons per minute (gpm) or an order of magnitude.

" An assessment will also be made of the potential need to increase the number of predicted low voltage indications at the beginning of cycle if the total number of as-found indications is underestimated by greater than 15 percent or by greater than 150 indications.

Historically, there have been no ARC cases where undetected indications (POD effects) have led to a challenge to structural or leakage integrity. These cases have been associated with E4-22

under predictions in growth rates. It is expected that growth rates would be the first potential cause examined for ARC under predictions. Potential POD effects as the cause for under predictions would also be assessed if the probable cause for the low predictions is a larger than anticipated undetected indication or due to cumulative numbers of indications above about 1 volt. The 90-day report will document any recommended changes to POD or growth methodology indicated by the assessments.

Upon implementation of POPCD, the following additional information shall be included in the 90-day report:

• To assess POPCD for potential changes over time, the 90-day report will compare the multi-cycle POPCD distribution applied for the last operational assessment with the POPCD distribution obtained for only the last operating cycle.

Differences in the two POPCD distributions should be assessed relative to the potential for significant changes in detection capability.

" For RPC confirmed indications at EOCn that are RPC NDD at EOCn+1 , an assessment is required for the cause for the "disappearing flaws" if the +Point voltage is greater than 0.5 volt. If there are a significant number of occurrences (such as more than three occurrences in any one SG) of these "disappearing flaws," the cause will be evaluated independent of the +Point voltage.

" The POPCD data reported in the 90-day report will include the plant specific historical results and the last cycle results reported in the format of Tables 1 and 2. A graphical comparison will be provided of the POPCD distribution obtained from the composite historical results recommended for ARC applications with the POPCD distribution obtained from only the data for the just completed cycle.

In support of this evaluation, Table 1 requires an RPC inspection at EOCn+1 for RPC confirmed indications at EOCn (either bobbin detected or bobbin NDD) that are bobbin NDD at EOCn+1 . This inspection is necessary to ensure that all known ODSCC indications are included in the condition monitoring and operational assessments as well as properly categorized for the POPCD evaluation.

8.0 Conclusions The current licensed ARC methodology of using a uniform POD value of 0.6, based on GL 95-05, results in overly conservative and counter intuitive estimates of the number and severity of indications remaining in the SGs following the inspection.

Results of POPCD evaluations support a high POD for bobbin E4-23

indications above about 1 volt and lead to near unity for POD above about 5 volts. All POPCD and voltage dependent POD results show that use of a constant POD of 0.6 is non-conservative below about 0.3 volts and very conservative above 1 volt, which leads to excessively conservative probability of burst and leakage predictions. The POD averaged over all indications, as shown in the lower right hand corner of Table 4 for Sequoyah Unit 2, is about 0.7, which is approximately consistent with the GL 95-05 value of 0.6 also developed independent of voltage. The application of the POPCD method is justified and appropriate for ARC analyses.

POPCD distributions are developed using loglogistic functions, which are commonly applied in tube integrity analyses for POD distributions. Uncertainties in the resulting POD distributions are shown to be negligible due to the large number of indications included in the composite POPCD evaluation. Sequoyah Unit 2 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. In addition, the Sequoyah Unit 2 POPCD results for bobbin coil detection are shown to be in good agreement with POD results obtained for the generic industry POPCD and the DCPP POPCD results.

Based on industry and Sequoyah Unit 2 specific bobbin detection data for ODSCC within the SG TSP region, large voltage bobbin indications that can individually challenge structural or leakage integrity can be detected with near 100 percent certainty and would not be left in service. These large voltage indications should not be included in the BOC voltage distribution, other than as inferred from. the voltage dependent POD, for the purpose of the operational assessment. The POPCD approach to probability of detection considers the potential for missing indications that might challenge structural or leakage integrity by applying the POPCD data from successive inspections. The database used to develop the Sequoyah Unit 2 POPCD values includes data from six successive inspections. If a large indication was missed in one inspection, it would continue to grow until finally detected in a later inspection. Therefore, the use of the POPCD method to determine the BOC voltage distribution will improve EOC projections and lead to appropriate estimates of the margin in SG tube structural and leakage integrity.

The benchmark results indicate that for the methodology used to project EOC conditions, while not conservative in all cases, the under predictions in general are not significant except for one instance. In this instance, an extreme growth rate was observed for one indication, and the extreme growth is considered to be an outlier or random event. In addition, Sequoyah Unit 2 has committed to perform a rotating probe inspection of all bobbin indications above 1.7 volts to address the potential cause for E4-24

the outlier growth rate. Since the methods have not always been conservative, Sequoyah Unit 2, upon implementation of POPCD, also commits to perform an assessment of the adequacy of the methodology, under certain conditions as defined in Section 7.0, for performing operational assessments.

9.0 References

1. EPRI Topical Report NP 7480-L, Addendum 6, "Steam Generator Tubing Outside Diameter Stress Corrosion Cracking at Tube Support Plates Database for Alternate Repair Limits", 2004 Database Update, dated October 2004.

2. Generic Letter 95-05, "Voltage-Based Repair Criteria for Westinghouse Steam Generator Tubes Affected by Outside Diameter Stress Corrosion Cracking," dated August 3, 1995

3. NRC Letter, G. S. Shukla (NRC) to G. M. Rueger (PG&E),

"Review of Diablo Canyon Nuclear Power Plant, Unit No. 2 -

Issuance of Amendment - Revised Steam Generator Voltage-Based Repair Criteria Probability of Detection Method for Diablo Canyon Unit 2 Cycle 12 (TAC No. MB9742)," dated October 21, 2003.

4. NRC Letter, G. S. Shukla (NRC) to G. M. Rueger (PG&E),

"Diablo Canyon Power Plant, Unit Nos.l and 2 - Issuance of Amendment RE: Permanently Revised Steam Generator Voltage-Based Repair Criteria Probability of Detection Method (TAC Nos. MC2313 and MC2314)," dated October 28, 2004.

5. PG&E Letter DCL-03-121 to the NRC, "PG&E Response to NRC Questions on 2R11 Steam Generator Tube Inspections," dated September 30, 2003.

6. PG&E Letter DCL-04-119 to the NRC, "DCPP Unit 1 Voltage-Based Repair Criteria Benchmarking and EOC-12 Projections," dated March 15, 2004.

7. WCAP-14277, Revision 1, "SLB Leak Rate and Tube Burst Probability Analysis Methods for ODSCC at TSP Intersections," December 1996.

8. SG-SGDA-05-29, "Condition Monitoring and Operational Assessment: GL795-05 Alternate Repair Criterion End of Cycle 13 90 Day Report Sequoyah Unit 2," August 2005.

9. SG-SGDA-03-55, "Condition Monitoring and Operational Assessment: GL-95-05 Alternate Repair Criterion End of Cycle 12 90 Day Report Sequoyah Unit 2," February 2004.

10. SG-SGDA-02-15, "Condition Monitoring and Operational Assessment: GL-95-05 Alternate Repair Criterion End of Cycle 11 90 Day Report Sequoyah Unit 2," July 2002.

11. 01-TR-FSW-002, "Sequoyah Unit 2 End of Cycle 10 Condition Monitoring and Operational Assessment 90 Day Report GL 05 Alternate Repair Criterion," January 2001.

12. Letter from G. Srikantiah of EPRI to J. Riley of NEI, "Beaver Valley Date Report", March 28, 2002

13. E-mail from R. F. Keating of Westinghouse to G. Srikantiah of EPRI, "Error in Updated ODSCC LR Correlation", April 22, 2002 E4-25

Table 1: Data Table for Tracking Indications Between EOCn and EOCn+1 BIDD at EOCn+1 BND at EOCn+1 EOCn BIDIDw/o RPC BIDID WRID BODIDw/RND BND w/o RPC BND WRIDD BND w/RND Plugg~ed 1Not JPlugged 1Not Plugged JPlugged] Plugged JNot Plugged ,Plugged Not JPlugged Plugged ýPuge Not Not Plugged ýPlugged Plugged C .. ___X_................

BIDIDw/oRPC Not Plugged B B A A H H H H A A H H B. ............ D.. ....... ..

at.... Pl g e C BO w. RO ...... ._

NotIPlugged C, ....

...... H.H..H..H.A. A..H.H BNOI w/o R PCI Not...........

Plugged........E......0......H...H..No....Count.....No....Count......F..F...Count......Count...

a OnNot Plugged E E A 0 2 1 1 F F2 H (2) H(ý2)

Pleud.

gg ........

BN.

N.O. O.

Not. Plge ......R... ...... .... ....................... ...........H.....NNo....

....Con

......... ..............N . ....o.......Cono.. ....

H: NoCon.FF.ont General Notes:

The column letters correspond to the column letters in POPCD Table 2.

No Count = Indications not detected by bobbin or RPC are not counted or needed for POPCD.

BDD = Bobbin detected indication BND = Bobbin NDD intersection RDD = RPC detected indication RND = RPC NDD intersection Specific Notes:

1) For EOCn bobbin indications that are confirmed by RPC or detected only by RPC, EOCn+1 RPC should be performed when bobbin is NDD and the number in this category will be "0" for future inspections.

2) If indications are RPC confirmed at EOCn but RPC NDD at EOCn+1 and the +Point voltage is greater than 0.5 volts, the causative factors for this change in RPC detection should be discussed in the ARC 90 day report. If there are a significant number of these occurrences in this category, independent of the +Point voltage, the cause, will be evaluated in the 90 day report.

3) EOC, bobbin indications that were RPC NDD at EOCn and at EOCn+1 are either RPC detected or bobbin detected without RPC inspection, are treated as undetected EOCn in accordance with NRC request.

E4-26

Table 2. Generic 7/8" Industry POPCD Evaluation Col...II A I B I I E I F I u H IJ K Evaluation for POPCD for Plants with 718" SG Tubes Combined Data from 30 Post-92 ('93 and later) Inspections Detection at EOC, No Detection at EOCn (N~w Indlcatlonn) 1 EICI PC Con.-.d .t EOC I EOC, RPC Not In.pect.datt EOC. Bobbi nd. Repaired at New EOC_.÷ Bobbin OPC New EOC,.,Bobbin NotRPC tnd.Found Only by RPC at EIC,.. or EDO. RPC NOD Bobbin Total, for POPOD Dootirmeda EDO EDO, RPO ~ Eat EDCoid Car cinletd atE Pl u

Plggd.,ol-'

EOI atindication.

i 0/

Exciuded from POPCOD LLEeuto 0D0/RDD"=BDDORDD BDODW.R/OPC-- BDDW/oRPC BDDIRDD

• Plugguduti BND/,oRPC "0BDDI/RDOD BNDwoIRPC 0DDw/oIRPC BNDo.oRPC OND/RDD B ODDORND -BDDl.oRPC All RNIDAT EOC'.,

ODD/ROD -=BND/RDD BODD/IRDO--- *BDD/o/RPC ODDwý0o RPC -p Plouuodni BND/ODD -*0DD/ODD BNDIRDD -.- BDDwI RPC BNDIROD . BND/RDD ODD/RND -- pB0/ODD 01/BND . R00 PC 0/O. N. POPCD for Voltage Bin ODD/RND -*0BDDI/DD BODD/IRND---*ODDW/oRPC 1BND/RND .-- *DDiRDD BND/RND -.. ODDw/o RPC BND/RND -. 0ND/RDD O ODDI/RND --.

• OND/RDD MtEOCn Voltage Bin (No/t 2) ODD/I/ND "=4BND/RDD BND / RDD -4 Plugged at EOCn Mt EOCn (No/t 1)

IDD to RPC -0D0DIORDD

... *nPr .NI D I) POPPCDfor 0Ich voltage bil calculated

/ as (Detection at EOCn)l(Dotection at EOCn + No Detection at EOCn). 0 column. POPCD = (A+BC)I/(AoB+C+D+E+F).

2/ Plan/spe0ifi/ POPCDto bebnsd/pono.tt/g/binab os.olvolt. Indos//0 POPCODdatbase ralyuseO .2Ovolt bin, duo 1 dificultyfofdjusting /)dsbngdatabase tonsmall/r bi/ns,

3) Inoludosin/ditoo/,,ns /(t2E0 plugged 0/ .OCn OndflwindicOtions 0t EOC 1,00mopor/nd/ inthbobbin innpection, nndfound 0nty byIRPC onspec/ton 0/fdon/n, iedresdunls oronerresons for the RPC inspeaon.

1)ODD = 00bbin detected indication; BND = Bobbin NDD intersection; ROD = RPC detecotd indication; IND = RPC NODintersection E4-27

Table 3. Sequoyah Unit 2 Database Tracking Table for POPCD at EOC-7 through EOC-12

________________________Data Table for All Indications and All Voltages BDD at EOCn+1 _ _____BND at EOCn+1______

BOO w/o RPC BOO wIRDO BOO w/RND BND wlo RPC BND W/RDD BND wIRND Not Not Not Not0 Not m=

_____ EOCn Plugged Plugged Piugged IPlugged Plugged Plugged Plugged Plugged Piugged Plugged Plugged Plugged BOO Wlo Plugged 45 ......  : ::::;;;::......

RPC Not Plugged 43 3769 23 252 0 27 2 331 0 0 0 17 at.......... O....... wi...lugged..23 n.. ROD.No.lugg d.2.22.5.14.0.2. ............ . 4..0..0.0..0 EO a RODed Not......lugged

- Dnw 23 ...0... 0...0...0...

........ 0...0...0....0...0....0...0...0.

______ RNO Not Plugged 02 2__ 5_148 0 00 oCutN on 4 0 No Con NoCon

________ BD w/ nPugd Pugd Pugd Puge Plugged lge lge lge lge Plge Plugged........ Plugged.....

EOn ROO Not Plugged 1 183 0 36 0 01 0 10 0 0 0 0 BNO w/o PluggedI

..... . ..."1:*' . .::..

NO w Nota Plugged 2.....

01 No Count No Count 2 0 N o.C.o.u.nt ..No C.o..u..n.t..

_ __ _ __ R Nt N ot Pl g e 01 8 04 E4-28

Table 4. Sequoyah Unit 2 Composite Evaluation for POPCD at EOC-7 (1996) through EOC-12 (2003)

Column A B C D F IGI H I I J K Sequoyah-2 ,

Y r Detection at EOC. o Detection at EOCn (New Indications)

I EOC, Bobbin Ind. RPC EOC. Bobbin Ind. Not RPC EOC, Bobbin Ind. Repaired at New EOC-, Bobbin RPC New EOC.,. Bobbin Not RPC EOC., or at EOC. & Plugged

.1 EOC.RPCNDDBobbin iEOC RPC Nd D Bobbin Excluded from Totals for POPCD Confirmed at EOC.,, Inspected at EOC.,, EOC. Contirmed Inspected a Indicationsr' POPCD Evaluation

- P)

BDDw/o RPC -* BDD wio RPC BSD IROD Plugged etEOCn BNODw/o RPC - B ' RDDOD BND wlo RPC -* BO w/o RPC BNDw/oRPC -4 BND/RDD BDDIRND

• BD0wSoRPC AlIRNDAT EOCn.

BO '(IROD BDD IRDD

• Bn IRD BND/RDD DD/RDD --11 BDDW/oRRPC ISBDRwOoD- -* u BNDIRNO B/DDRD BNDIRDD -i- BDDw/oRPC BNDI ROD -4. BND/RDD BODIRND --m BO/ROD A"IBNDw/oRPC Detection No POPCD for BintaOe BwO/RPC -BD/RD BB O BND/I NDOD I BO/ROD BND/RND -m BOOw/oRPC BNO /RND -4 BND(RDD SOD / RND BND / ROD at EOCn+I at EOCn Detection Voltage Bin BDDw/o RPC --

• BND/RDD BNDO/RD - Plugged at at EOCn Note 0' I

EOCn BDDIRND/Plugged at EOCn 0 1 4 25 0 66 138 256 0 130 501 404 0 112 758 398 0.41-0.50 20 0 80 768 269 0.51-0.60 0 _L4 649 191 0.61-0.70 101 0 L9- 510 136 0.71-080 38 3 L4_ 383 60 0.81-0.90 32 2 34 283 47 0091-1.00 18 181 30 10 114 21 1.11-1.20 6 0 88 13 1.21-1.30 1 44 6 1.31-1.40 10 2 0 30 6 1.71-1.80 1241-1,50 0 0 24 1 2

1.51-1.60 0 15 2 1

1 .01-1. 70 2 0 0

1.71-1.800 1 0 0 0 1.01-1.900 0 0 0 1.01-2.00 0 0 0 2.0 1-2.10 0 0 0 0 0

-L 0 0

O 0 0 0 0

-L 0 0

-r 0 0 0

-r 0

-r 0 0

0 3.11-3.20 0 0 0

0

• lotes:

t) POPCD for each voltage bin calculated as (Detection et EOCn)/(Detection at EOCn + No Detection atEOCn). By coumn, POPCD = (A+B+C)I(A+B+C+D+E+F+G).

1) EOCn RPC NDD bobbin indicatimonsare treated as new indications per NRC request I) Includes indicetions at EOCn plugged at EOCn and new indications at EOCn+I, not reported in the bobbin inspection, and found only by RPC inspecton of dents, mixed resduals aoother reasons for the RPC inspection.

t) BOD = Bobbin detected indication; BND = Bobbin NDD Intersection; ROD = RPC detected indication; RND = RPC NDD intersection E4-29

Table 5. Sequoyah Unit 2 Database Tracking Table for POPCD at EOC-12 Data Table for All Indications and All Voltages BDD at EOCn+1 BND at EOCn+1 BDD w/o RPC BDD w/RDD BDD w/RND BND w/o RPC BND w/RDD BND w/RND NNot ot Not Not Not Not EOCn ____ Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged BOO w/o Plugged 16 * ** * ===**==**=`***********``; `;*`; ``=* ** I'* *  ; ',

RPC Not Plugged 9 1191 4 78 0 9 0 5 0 0 0 *0 BOOI Plugged0 Pug BlNgwoPe d:::::::::::::::::::::::::::

........................... :::::::::::::::::::::::::::::: q..................

EOCn ROD BNO w/o Not Plugged Plugged 1 1

................. 63 4 102 :::::

0 1 0 0 ....... 0 0 0 ... 0 RND Not Plugged 0 14 0 3 0 3 0 26 0 0 0 0 RPC Not Plugged 2 284 0 56 1 22 No Count No Count 0 0 No Count No Count EOCn ROD Not Plugged 0 0 0 0 0 0 0 0 0 0 0 0

___ RND Not Plugged 0 0 0 0 0 0 No Count No Count 1 0 No Count No Count BOO w/ Plugged 0 Data Table . ...for.All .... Indications

....... Less

........... than .. or Equal to 1.0 Volt_______ .......

______BOO at EOCn+l BND at EOCn+l _____

BOO w/o RPC BOO w/RDD BOO w/RND BNO wlo RPC BNO w/RDD BND w/RND SNot Not Noto Noto EOCn ____Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged Plugged RPC Not Plugged 9 1191 4 78 0 9 0 5 0 0 0 0 EOCn RDD Not Plugged 0 42 0 13 0 0 0 0 0 0 0 0 RND Not Plugged 0 10 0 1 0 1 0 13 0 0 0 0 B ND.w/o Plu g g e d  ;.**...X -'-:::::

-:.:.:::":::ii:i RPC Not Plugged 2 281 0 51 0 16 No Count No Count 0 0 No Count No Count aOCn ROD Not Plugged 0 0 0 0 0 0 0 0 0 0 0 0 BNO wi Plugged

_______ RND Not Plugged 000000 NoCutNCon 10 Noont oCut E4-30

Table 6. Sequoyah Unit 2 Evaluation for POPCD at EOC-12 Column A B C D I E F G H I I I K Seguoyah-2 Specific POPCD Data Table Detection at EOC, No Detection at EOCn (New Indications) inO. IouflO] Inly ay Kr-.. a1 EOC, Bobbin Ind. RPC EOC, Bobbin Ind. Not RPC EOC, Bobbin Ind. Repaired at EOCý New EOC, 1 Bobbin RPC New EOC,, Bobbin Not RPC EOC,. 1 or at EOC, & Plugged EOC, RPC NDDBobbin Excluded from Totals for POPCD Confirmed at EOCn Inspected at EOC , Confirmed Inspected at POC *n Indications ) POPCD Evaluation BDD/RDD -*BDD/RDD BDDw/oRPC -'BDDw/oRPC BDD/RDD -Plugged at EOCn BNDw/oRPC -- ' BDD/RDD BNDw/oRPC - BDDw/oRPC BNDw/oRPC -0"8ND/RDD BDD/RNO -0BDDw/oRPC AtIRNDATEOC_

Voltage BDD/RDD -p BND / RDD BDD/RDD -BOD w/o RPC BDDw/o RPC -4.Plugged at EOCn BND/RDD -- BDD/RDD BND/RDD -- BDDw/o RPC BND/RDD -o BND /RDD BDD/RND -4BDD/RDD AlIBND w/o RPC Detection No POPCD for Bin BDDw/oRPC -4.BDD/RDD BND/RND -+.BDD/RDD BND/RND -4. BDDw/o RPC BND/RND --p. ND RDD BDD/RND --.*BND/RDD at EOCn+1 at EOCn Detection Voltage Bin BODw/o RPC -4 BND / RDD BND / RDD Plugged at BDD/RND/Plugged at EOCn Note n)

EOCn at EOCn 0.01-0.10 0 0 0 0 7 0 0 0 0 7 0.000 0.11-0.20 2 38 1 2 50 0 0 0 41 52 0.441 0:21-0.30 4 161 2 4 60 0 1 1 167 65 0.720 0.31-0.40 1 263 3 5 64 0 0 6 267 69 0.795 0.41-0.50 7 232 3 11 36 0 2 6 242 49 0.832 0.51-0.M0 5 175 2 6 34 0 0 8 182 40 0.820 0.61-0.70 8 153 1 8 19 0 4 6 162 31 0.839 0.71-0.80 18 112 3 7 5 1 3 5 133 16 0.893 0.81-0.90 29 72 0 4 4 1 0 10 0.918 0.91-1.00 21 36 1 4 4 0 1 3 0.866 1.01-1.10 26 11 0 1 2 0 2 7 37 5 0r8 1,11-1.20 25 9 2 2 1 0 2 3 36 5 0.878 1.21-1.30 15 1 1 0 0 0 1 5 17 1 0.944 1.31-1.40 8 1 1 1 0 0 1 3 10 2 0.833 1.41-1.50 10 0 2 1 0 0 0 2 11 0.92 1.51-1.60 6 0 2 0 0 0 0 0 8 0 .0 1.61-1.70 3 0 0 0 0 0 0 0 3 'I.000 1.71-1.80 0 0 2 0 0 0 0 0 2 0 1.000 1.81-1.90 0 0 0 0 0 0 0 2 0 0 1.91-2.00 0 0 0 0 0 0 0 0 0 0 2.01-2.10 0 0 1 0 0 0 0 0 1 0 1.000 2.11-2.20 0 0 1 0 0 0 0 0 2.21-2.30 0 0 0 0 0 0 0 0 0 0 2.31-2.40 0 0 0 0 0 0 0 0 0 0 2.41-2.50 0 0 2 0 0 0 0 0 2 0 1.000 2.51-2.60 0 0 0 0 0 0 0 0 0 0 2.61-2.70 0 0 0 0 0 0 0 0 0 0 2.71-2.80 0 0 0 0 0 0 0 0 0 2.81-2.90 0 0 0 0 0 0 0 0 2.91-3.00 0 0 0 0 0 0 0 0 0 0 3.01-3.10 0 0 0 0 0 0 0 0 0 0 3.11-3.20 0 0 0 0 0 0 0 0 0 0 3.21-3.30 0 0 0 0 0 0 0 0 0 0 3.31-3.40 0 0 0 0 0 0 00 -0 3.41-3.50 0 0 0 0 0 0 0 0 3.51-3.60 0 0 1 0 0 0 0 0 9.71-9.80 0 0 1 0 0 0 0 0 1 0 1.000 Total 188 1264 2 1756 0.804 Nlotes:

1) POPCD toreach voltage bin calculated as (Detection at EOCn)/(Detection at EOCn + No Detection at EOCn). Bycdumn, POPCD = (A+B+C)/(A+B+C+D+E+F+G).

2) EOCn RPC NDDbobbin indications are treated as new indications per NRC request

3) Includes indications at EOCn plugged at EOCn and new indications at EOCn+1, not reported in the bobbin inspection, and found only by RPC inspection of dents, mixed residuals or other reasons for the RPC inspection.

4) BDD= Bobbin detected indication; BND = Bobbin NDDintersection; RDD = RPC detected indication; RND = RPC NODintersection E4-31

Table 7 Sequoyah Unit 2 and Industry POPCD Log Logistic Distribution Parameters Parameter Sequoyah Unit 2 Sequoyah Unit 2 Industry POPCD POPCD POPCD 30 Inspections at EOC-7 to At EOC-12 EOC-12 Number of Data 6393 5926 46454 Points b0 (Intercept) 1.9947 2.0005 1.9628 bl (Slope) 2.9920 2.8274 3.1433 Vll 3.530E-03 3.788E-03 3.742E-04 V1 2 6.826E-03 7.385E-03 7.215E-04 V2 2 1.728E-02 1.904E-02 2.077E-03 E4-32

Table 8. Sequoyah Unit 2 Prior Cycle Benchmarking Results for POPCD Projecte SG Projected Results As Found Results POB & Leak Growth Rate POD d Outage POB Leak No. Max. POB Leak No. Max. Rate Used Used Rate Ind. Volts*') Rate Ind. Volts*1 ) Correlations EOC-II 1 1.21x10 0.253 159 2.2 1.2x0 0.155 129 1.61 Addendum 4 for All SGs: Seq. 2 5 Burst, Leak Bounding of POPCD 2 1.21x10- 0.429 176 2.4 1.2x10 5 0.197 138 1.69 Rate all SGs & Fig.

5 Correlations Composite 3 1.90x10- 0.603 186 2.7 1.9x10- 0.392 151 2.26 from Ref. 12 Over Prior 5 and 13. Two Cycles 4 3.10x10- 1.91 755 2.4 5.3x105 1.33 625 3.35 APsLB = 2560 Plus 5 psi Increase in EOC-12 1 5.26xi0- 0.240 183 2.6 6.2xi0-5 0.209 247 1.88 Addendum 5 Average 5 PsLB = 2560 Growth Over 2 9.72xi0- 0.279 186 2.6 5. 3xi0- 0.236 252 1.81 psi the Last 5 Cycle 3 1.86x100 0.530 196 2.9 3.6xi0- 1.08 307 9.76 4

4 2.22x10- 1.580 848 3.0 1.4x10-4 0.965 739 3.55 4

EOC-13 1 1.05x10- 0.421 342 8.5 1.90x10- 0.072 294 1.95 Addendum 5 13 4 ZPsLB = 2405 2 1.02x10 0.429 342 8.5 3. 10x10- 0.076 305 1.97 psi 3

9 3 1.23xi0 0.574 406 8.7 5.26xi0-5 0.224 412 2.36 3

4 2.78xi0 1.230 969 8.8 9.24xi- 0.285 836 1.74 3

Prior Analyses for POD = 0.6 from 90 Day Reports References 8 to 11 EOC-11 1 1.05x10 0.345 180 2.1 1. 2x1 0.155 129 1.61 Addendum 4 for SG Specific 0.6 5

Burst, Leak Cycles 9 &

2 1.69x10 0.634 214 2.4 1.2x10-5 0.197 138 1.69 Rate 10 5 Correlations Bounding,

3. 3.61x10 0.900 237 2.7 1.9xI105 0.392 151 2.26 from Ref. 12 Table 7.2, 5 and 13. Ref. 11 4 1.82xi0 2.52 899 2.3 5.3xi0-' 1.33 625 3.35 APSLB = 2560 5 psi EOC-12 1 1.2X10-5 0.519 217 2.4 6.2x105 0.209 247 1.88 Addendum 5 SG Specific 2 4.2x0-5 0.634 227 2.7 5.3xi0-5 0.236 252 1.81 LPsLB = 2560 Cycle 11 3 5.8x105 1.05 259 2.9 3.6xi- 1.08 307 9.76 psi Bounding, 4 7. 3xi0- 3.40 1024 3.7 1.4x10-4 0.965 739 3.55 Table 7.2, Ref. 10 EOC-13 1 1.52xi0 0.546 411 8.6 1.90x10-T 0.072 294 1.95 Addendum 5 All SGs:

3 4 LPSLB = 2405 Bounding 2 l.41x10 0.568 416 8.7 3.10x10-5 0.076 305 1.97 psi Cycles 11 &

3 9 12, Fig. 3-3 3.09x0-3 0.985 506 10.4 5.26Xi0-5 0.224 412 2.36 16, Ref. 8 4 3.92xi0- 1.69 1211 9.0 9.24xl0- 5 0.285 836 1.74 3

Notes:

1. Voltage where projected tail accumulates to 0.3 ind.

E4-33

Table 9. Sequoyah Unit 2 EOC-14 ARC Projections for POPCD and POD = 0.6 Projected Results POB & Leak POD Steam Generator SLB POB SLBLeak Rate-gpm No.

Indications Max.

Volts(1) Rate Correlations Growth Rate Used POPCD 1 1.73xi0 0.400 401 9.3 Addendum 6 Bounding 2 1.80x10- 0.409 412 9.3 LPsLB = 2405 Cycle 12, 3 2.23xi0 - 0.667 531 9. psi Fig. 3-16, 4 4.57xi0 -1 1.12 1116 9.6 1 ps Ref. 8 Prior Analyses for POD = 0.6 from 90 Day Report Reference 8 0.6 1 22.20x10-T 0.519 489 F 9.5 Addendum 6 Bounding 2 2.47xi0-s 0.530 501 9.5 LPsLB = 2405 Cycle 12, 3 3.41xi0 - 0.874 680 9.8 psi Fig. 3-16, 4 6.51xi0O. 1.47 1387 9.8 Ref. 8 Notes:

1. Voltage where projected tail accumulates to 0.3 indication E4-34

Figure 1. Generic Industry POPCD Distribution for 7/8-Inch Tubing Generic Industry 7/8" Tubing POPCD as Function of Bobbin Amplitude Weighted Generalized Linear Model Loglogistic Solution 100%

-]-

I I I POD Regression I I I I lIT ~

90% ... LogLogistic 5.0 % Conf + + +-+-~-~ I I I -

/fol 80%

I I H 01 70%

60%

o 50%

40%

30%

20% 00f, 10%

0%

0.0 1 0.1 1 1 1+111 1 Lj 10 Bobbin Amplitude (Volts)

Figure 2. Sequoyah Unit 2 Bobbin Volts to +Point Volts Correlation E4-35

Sequoyah-2 Bobbin to +Point Volts Correlation Data from Inspections at EOC-8 to EOC-13 with Single +Point SAIs 4.00 y 0.0134X2 + 0.9445x + 0.6468 3.00- Rt2 0.9999

.3 0"

.5 2.00-0 Ep 0 NDE Data 1.00- Regression Line

--- Upper 95% Confidence

- Polynomial H~tto Upper 95% Confidence Curve 0.00.0 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50

+Point Volts E4-36

Figure 3. Sequoyah Unit 2 POPCD Distribution for EOC-7 through EOC-12 Sequoyah-2 POPCD as Function of Bobbin Amplitude Weighted Generalized Linear Model Loglogistic Solution 100%

- -- Sequoyah-2 POD Regression 90% Sequoyah-2

-- Lower 95% Confidence -- _ -"

80%

70%.

0 60%

0 50%

40%

30%

20% -Ile 10%

0%

0.01 0.1 I 10 Bobbin Amplitude (Volts)

Figure 4. Comparison of Sequoyah Unit 2 POPCD for EOC 7-12 with EOC-12 E4-37

Comparison of Sequoyah-2 EOC 7-12 and EOC 12 POPCD Weighted Generalized Linear Model Loglogistic SolutiQn 100% 1 1 1 1 1 1 1 1 1 Sequoyah-2 EOC 7-12 POPCD - -

90% -- - Sequoyah-2 EOC 12 POPCD - -

Sequoyah-2 EOC 7-12 POPCD at Lower 95%

80% -- - Sequoyah-2 EOC 12 POPCD at Lower 95% X . I 70% / -

60% - - - - -

o 50% -.-

• 40% _

30%

I-d 30%,

10%

0.01 0.1 10 Bobbin Amplitude (Volts)

E4-38

Figure 5. Comparison of Sequoyah Unit 2 POPCD with Generic Industry POPCD Comparison of Sequoyah-2 and Industry 7/8" Tubing POPCD Weighted Generalized Linear Model Loglogistic Solution 100%

Sequoyah-2 POPCD Regression 90%

-.- Sequoyah-2 POPCD at Lower 95%

- A- Industry 7/8 POPCD Regression 80%

- - Industry POPCD at Lower 95%

70%

0 60%

'4-0 50%

".0" 40%

0 30%

20%

10%

0%

0.01 0.1 I 10 Bobbin Amplitude (Volts)

Figure 6. Comparison of Sequoyah Unit 2 POPCD with Diablo Canyon POPCD E4-39

Comparison of Sequoyah-2 EOC 7-12 and DCPP Through 1R13 POPCD Weighted Generalized Linear Model Loglogistic Solution 100% I I I [ [ [ L r I 7L~¶i~t~

I I I I I I I I I

- Sequoyah-2 EOC 7-12 POPCD 90% - A DCPP Through I RI 3 POPCD 80%

A 70%

60%

0 50%

40%

30%

20% 1 -4 0ý 10%

0%

0.01 0.1 I 10 Bobbin Amplitude (Volts)

E4-40

Figure 7. Sequoyah Unit 2 Average Growth Rate and Average Bobbin Volts Changes over Cycles 8 to 13 Sequoyah-2 All SG Cycle Dependence of Average Growth and Average Volts 0.09 0.7

- Average Voltage Growth - All SGs

-Average Bobbin Voltage - All SGs 0.08 0.6 0.07 0.5 0.06

_0.4 0.05 0

0.04 0.3 U 0.03 0.2 0.02 0.1 0.01 0 0 8 9 10 11 12 13 14 Operating Cycle E4-41