ML083530789

Emergency Core Cooling System Annual 10 CFR 50.46 Report

ML083530789
Person / Time
Site:	San Onofre
Issue date:	12/16/2008
From:	Scherer A Southern California Edison Co
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
Download: ML083530789 (21)

Text

SOUTHERN CALIFORNIA A. Edward Scherer Director

_ 1 Nuclear Regulatory Affairs An EDISON INTERNATIONAL Company December 16, 2008 U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, D. C. 20555

Subject:

Docket Nos. 50-361 and 50-362, 2007 Emergency Core Cooling System Annual 10 CFR 50.46 Report, San Onofre Nuclear Generating Station, Units 2 and 3.

Reference:

Letter from A. E. Scherer (SCE) to Document Control Desk (NRC), dated December 21, 2007,

Subject:

Docket Nos. 50-361 and 50-362, 2006 Emergency Core Cooling System Annual 10 CFR 50.46 Report, San Onofre Nuclear Generating Station, Units 2 and 3 This letter transmits as Enclosures 1 and 2 the San Onofre Units 2 and 3 annual report for the 2007 calendar year required by paragraph (a)(3)(ii) of 10 CFR 50.46, "Acceptance criteria for emergency core cooling systems for light-water nuclear power reactors." This regulation requires Southern California Edison (SCE) to annually report to the NRC for San Onofre Units 2 and 3 the nature of each change to or error discovered in the Emergency Core Cooling System (ECCS) evaluation model or in the application of this model that affects the temperature calculation and estimated effects of any such changes, errors, or applications on the limiting ECCS analysis. Any significant change or error is required to be reported to the NRC within 30 days.

The previous Emergency Core Cooling System Annual 10 CFR 50.46 Report was submitted to the NRC in the above Reference.

Enclosure 1 is a description of changes or errors in the Westinghouse evaluation models (EM) used in the San Onofre Nuclear Generating Station (SONGS) Units 2 and 3 ECCS performance analyses Enclosure 2 is a summary of the effect on Peak Clad Temperature (PCT) of the errors or changes in the Westinghouse evaluation models used in the SONGS Units 2 and 3 ECCS performance Analyses. While not limiting with regard to PCT, detailed information for the Small Break Loss Of Coolant Accident (LOCA) is also included in Enclosure 2 (in accordance with Supplement 1 to Information Notice 97-15).

P.O. Box 128 San Clemente, CA 92674

Document Control Desk December 16, 2008 SCE made no changes to the LOCA evaluation models.

ODeratina Cycle Information Unit 2 and Unit 3 operation for the current reporting period is outlined below.

Unit Year Cycle 14 January 1, 2007 2 2007 through November 26, 2007 January 1, 2007 3 2007 through December 31, 2007 SONGS Units 2 and 3 Large Break LOCA Evaluation Model (EM)

- 2007 Reportina Period The Large Break LOCA (LBLOCA) analysis used the 1999 EM for Cycle 14.

The limiting Large Break LOCA PCT did not exceed the 10 CFR 50.46(b)(1) acceptance criterion of 2,200'F. This is documented in Enclosure 2 (Table 1).

There are no cumulative (sum of the absolute magnitudes of PCT changes) 1999 EM Large Break LOCA 10 CFR 50.46 model changes or model errors. This is documented in (Table 2).

SONGS Units 2 and 3 Small Break LOCA Evaluation Model (EM)

- 2007 Reporting Period The Small Break LOCA (SBLOCA) analysis uses the Supplement 2 Model (S2M) Small Break LOCA evaluation model for all of 2006.

The limiting Small Break LOCA PCT did not exceed the 10 CFR 50.46(b)(1) acceptance criterion of 2,200'F, and remained bounded by the large break LOCA PCT. This is documented in Enclosure 2 (Table 3).

Document Control Desk December 16, 2008 There are no cumulative (sum of the absolute magnitudes of PCT changes) S2M Evaluation Model Small Break LOCA 10CFR50.46 model changes and model errors. This is documented in Enclosure 2 (Table 4).

If you have any questions or need additional information on this subject, please contact Ms. Linda T. Conklin at 949/368-9443.

Sincerely,

Enclosures:

1. Description of Changes or Errors in the Westinghouse Evaluation Models used in the SONGS Units 2 and 3 ECCS Performance Analyses (Calendar Year 2007)

2. Summary of the Effect on PCT of the Errors or Changes in the Westinghouse Evaluation Models used in the SONGS Units 2 and 3 ECCS Performance Analysis (Calendar Year 2007) cc: E. E. Collins, Regional Administrator, NRC Region IV N. Kalyanam, NRC Project Manager, San Onofre Units 2 and 3 G. G. Warnick, NRC Senior Resident Inspector, San Onofre Units 2 and 3

Enclosure 1 Description of Changes or Errors in the Westinghouse Evaluation Models used in the SONGS Units 2 and 3 ECCS Performance Analyses (Calendar Year 2007)

1. 1999 EM Version of the Westinghouse Appendix K Large Break LOCA Evaluation Model for Combustion Engineering PWRs PROPRIETARY NOTATIONS IN SOURCE AND OUTPUT

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). Modifications to the source file notation and output edit headers for the major computer codes of the 1999 EM have been made to indicate the copyright status and proprietary class in conformance with Westinghouse policy, WCAP-7211.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect This process improvement has no impact on the licensed methodology of the 1999 EM and does not conflict with the SER limitation/constraints imposed on the methodology by NRC. There is no impact on PCT for 10 CFR 50.46 reporting purposes.

Page 1 of 12

IMPLEMENTATION OF OPTIMIZED ZIRLOTM CLADDING SPECIFIC HEAT

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). The evaluation of fuel designs that use Optimized ZIRLOTM cladding was described in Reference 1 and approved by NRC in Reference 2. In compliance with SER Limitation/Constraint #9 from Reference 2, the Optimized ZIRLOTM cladding specific heat has been implemented into all of the computer codes of the 1999 EM as described by Westinghouse in the response to RAI #21 in Reference 3. The Optimized ZIRLOTM cladding specific heat has been implemented as an option for any future LBLOCA analysis of a Westinghouse fuel rod design that uses this cladding.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect Since this change has already been approved by NRC, the licensed methodology of the 1999 EM is not affected. With the selection of this option for analyses using the 1999 EM, all of the computer codes are brought into compliance with the SER Limitation/Constraint imposed on the modeling of Optimized ZIRLOTM cladding. The impact of the change on PCT becomes integrated with the plant-specific and reload-specific analysis results used in the 10 CFR 50.46 reporting process.

Reference(s)

1. WCAP-12610-P-A and CENPD-404-P-A Addendum 1, "Addendum 1 to WCAP-12610-P-A and CENPD-404-P-A Optimized ZIRLOTM, LTR-NRC-03-2, February 14, 2003. (ADAMS Accession No. ML030520455)

2. Letter from H. N. Berkow (NRC) to J. A. Gresham (Westinghouse), "Final Safety Evaluation for Addendum 1 to Topical Report WCAP-12610-P-A and CENPD-404-P-A, 'Optimized ZIRLOTM' (TAC No. MB8041)," June 10, 2005.

3. Letter from J. A. Gresham (Westinghouse) to U.S. Nuclear Regulatory Commission, "Westinghouse Responses to NRC Request for Additional Information (RAIs) on Optimized ZIRLOTM Topical - Addendum 1 to WCAP-126 10-P-A," LTR-NRC-04-44, August 4, 2004.

(ADAMS Accession No. ML042240408)

Page 2 of 12

FLOW PATH ENTHALPY TRANSPORT NUMERICAL STABILITY IN CEFLASH-4A

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). Appendix I of CENPD-133 P (Reference 1) documents the flow path enthalpy transport model used in the CEFLASH-4A computer code for calculating the blowdown thermal-hydraulics response during a LBLOCA transient. An upgrade to the flow path enthalpy transport model has been implemented as an optional process improvement to give the user additional numerical stability control other than modifying time step size. The optional improvement has been implemented to automatically control numerical convergence during time periods where the system flow rates are nearly zero, such as near end of blowdown, and the stability of the calculation is more difficult to maintain.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect This process improvement has no impact on the licensed methodology of the 1999 EM and.does not conflict with the SER limitation/constraints imposed on the methodology by NRC. The improvement is made available to the user as an option and is designed to maintain a numerically stable solution. Use of the option will prevent abnormal code operation requiring manual time step adjustments with the current logic to achieve convergence. The impact on PCT for 10 CFR 50.46 reporting purposes is not significant when compared to converged results.

Reference(s)

1. CENPD-133P, "CEFLASH-4A, A FORTRAN-IV Digital Computer Program for Reactor Blowdown Analysis," August 1974.

Page 3 of 12

CORRECTION OF AN INCORRECT ERROR MESSAGE IN CEFLASH-4A

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). The editing logic in CEFLASH-4A, the blowdown thermal-hydraulics systems code of the 1999 EM, has been modified to prevent an incorrect error message from causing the utility script, which manages the operation of the 1999 EM, from an improper termination. In CEFLASH-4A, the error message should have been only a debug warning message and should not cause the utility script to stop execution.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect This process improvement has no impact on the licensed methodology of the 1999 EM and does not conflict with the SER limitation/constraints imposed on the methodology by NRC. There is no impact on PCT for 10 CFR 50.46 reporting purposes.

Page 4 of 12

UPGRADE TO THE AUTO-AXIAL POWER SHAPE OPTION IN STRIKIN-II

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). The 1999 EM is required to specify the axial power shape for LBLOCA analyses consistent with the methodology documented in Appendix A of CENPD-132 Supplement 3-P-A. Also, the specification of the axial power shape must be in compliance with an SER Limitation/Constraint on the use of this methodology.

As part of the Advanced Automated Integrated Code System (AAICS) for the 1999 EM, an option was implemented previously to automatically determine the axial power shape that complies with the core design characteristics of the analysis and also meets the methodology requirements. One aspect of the automatic process involves achieving a particular power distribution consistent with the target value for the minimum ASI, the limiting axial shape index prescribed for top peaked power distributions that are worst for LBLOCA. An upgrade to the computer logic has been implemented to improve the process for matching the prescribed, target ASI for the 1999 EM. The improvement more accurately achieves the target ASI over a-broader range of conditions than before. This improved computer logic replaces the previous logic.

However, it still remains the responsibility of the user to examine the axial power shape generated by the automatic process on a case by case basis.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect This process improvement has no impact on the licensed methodology of the 1999 EM and does not conflict with the SER limitation/constraints imposed on the methodology by NRC. There is no impact on PCT for 10 CFR 50.46 reporting purposes.

Page 5 of 12

AUTO-GENERATION OF THE STRIKIN-II BASE AND CASE DECKS

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). Computer code base and case decks provide the inputs to the operation of the 1999 EM. One of the advancements of the 1999 EM has been the auto-generation of base decks that can be created using inputs provided elsewhere in the 1999 EM code system. This eliminates some documentation and quality assurance and improves quality control by eliminating sources of error.

The hot rod heatup computer code of the 1999 EM is STRIKIN-II. A process improvement has been implemented to automatically generate the STRIKIN-II base and case decks in a manner consistent with the fuel performance data set from the FATES3B computer code, which provides initial conditions for fuel rod stored energy and rod internal pressure.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect This process improvement has no impact on the licensed methodology of the 1999 EM and does not conflict with the SER limitation/constraints imposed on the methodology by NRC. There is no impact on PCT for 10 CFR 50.46 reporting purposes.

Page 6 of 12

INCREASE THE NUMBER OF RADIAL REGIONS FOR COMZIRC

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). The core-wide cladding oxidation is calculated using the COMZIRC computer code. COMZIRC represents the core fuel pin census with a table of values of fuel rod power versus fraction of rods. The current version allows a maximum of only 12 radial intervals to represent the core. The purpose of this change is to increase the radial detail from 12 to 120 intervals, thus allowing enhanced control of the discretionary conservatism needed to bound the core pin census with the COMZIRC input table.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect This process improvement has no impact on the licensed methodology of the 1999 EM and.does not conflict with the SER limitation/constraints imposed on the methodology by NRC. There is no impact on PCT for 10 CFR 50.46 reporting purposes.

Page 7 of 12

ASSIGNMENT OF CEFLASH-4A REGIONS IN COMZIRC

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). The core-wide cladding oxidation is calculated using the COMZIRC computer code. COMZIRC represents the core fuel pin census with a table of values of fuel rod power versus fraction of rods. COMZIRC initializes the radial core regions using information provided by the CEFLASH-4A computer code for its radial distribution in the core.

The assignment of the CEFLASH-4A radial regions to the radial pin census in COMZIRC was previously a manual operation performed by the user. Computer logic has been implemented in COMZIRC to automatically confirm and/or provide the assignment of the CEFLASH-4A radial regions to the pin census.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect This process improvement has no impact on the licensed methodology of the 1999 EM and does not conflict with the SER limitation/constraints imposed on the methodology by NRC. There is no impact on PCT for 10 CFR 50.46 reporting purposes.

Page 8 of 12

IMPLEMENTATION OF CATHCART-PAWEL OXIDATION MODEL AND PRE-TRANSIENT OXIDATION MODEL

Background

A new User Control Interface (UCI) parameter ('oxidationmodel') is added to the UCI File Parameter List to facilitate the selection of the Cathcart-Pawel model for cladding oxidation as an option for non-licensing applications. The Cathcart-Pawel oxidation model is a best estimate model utilized by the industry as an alternative to the Appendix K required Baker-Just model.

This new UCI option is not permitted for licensing applications of the 1999 EM, which must use the Appendix K required Baker-Just model.

A new UCI parameter ('pretranoxidation') is added to the UCI File Parameter List to facilitate the specification of the initial oxide layer thickness through the UCI input file instead of through the base decks for non-licensing applications. This option may also be used to link the input specification to an interface output file from the HIDUTYDRV computer code, which provides maximum oxide thickness as a function of burnup (coordinated with FATES3B cycle numbers) and cladding type. In addition, this UCI parameter contains a new cladding conductivity option, which includes the impact of the oxide layer on the cladding conductivity, thereby directly linking the initialization of the fuel stored energy with the amount of pre-transient oxidation.

The initial oxide layer thickness is a user-specified required input to the evaluation model as a constraint on the acceptability of the model for licensing applications. Therefore, this new UCI option is not permitted for licensing applications of the 1999 EM, which must use the required input.

The alternate oxidation model and pre-transient oxidation model are used to study various aspects of new embrittlement criteria being suggested by NRC and to provide an alternate means to address NRC questions regarding the calculation of the peak cladding oxidation percentage in applications of the 1999 EM to CE plants. The new UCI options provide for automatic activation of input vector changes and ensure consistency among the other computer codes of the 1999 EM.

Affected Evaluation Model(s)

Non-Licensing Applications using the Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect This process improvement is for non-licensing applications. Therefore, this change has no impact on the licensed methodology of the 1999 EM and does not conflict with the SER limitation/constraints imposed on the methodology by NRC for licensing applications. For licensing applications, there is no impact on PCT for 10 CFR 50.46 reporting purposes since these changes are intended only for use in non-licensing calculations.

Page 9 of 12

IMPLEMENTATION OF FINAL SER LIMITATIONS AND CONDITIONS FOR THE OPTIONAL SPACER GRID STEAM COOLING HEAT TRANSFER MODEL IMPROVEMENT IN STRIKIN-II

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). An improved optional steam cooling heat transfer model for core reflood rates less than 1 in/sec was submitted to NRC in May 2006, to calculate the effects of spacer grids on steam cooling heat transfer mechanisms, CENPD-132, Supplement 4-P-A, Addendum 1-P. NRC acceptance of this improvement to the 1999 EM was documented in the final SER received by Westinghouse in June 2007.

The final SER from NRC requires that several changes be made to the methodology. All future licensing applications that utilize the optional spacer grid steam cooling heat transfer model will be required to implement these changes by using STRIKIN-II, Version STR.2.11 or higher. The following changes bring this version of STRIKIN-I1 into full compliance with the SER limitations and conditions on the use of the optional steam cooling model as imposed by NRC:

Grid Rewet Temperature Criterion - The final model approved by NRC requires that the spacer grid rewet temperature criterion be reset to a slightly different value than originally proposed.

Grid Model Output File - The final model approved by NRC requires that additional information be plotted and submitted for NRC review prior to the initial application of the model to a particular CE plant. To facilitate this documentation requirement, an additional output file is added to streamline generation of the needed plots.

Update Reynolds Number Formulation - The final model approved by NRC requires that the Reynolds number formulation for the wetted spacer grid heat transfer calculation be revised.

Diagnostic Edit Statements - The final model approved by NRC requires that the blockage fraction and Reynolds number ranges of applicability be confirmed for each application of the model. Diagnostic edit statements to display the range of validity checks are upgraded to alert the user if the use of the model is outside the range of applicability.

These changes have an insignificant impact on the overall calculated results with no impact on PCT. The revised methodology is activated with the selection of the user-specified option that is referred to as the "NRC-approved model." These changes have no effect on the calculated results if the optional steam cooling model is not used in the analysis.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect These NRC-required changes to the optional steam cooling model have no impact on PCT for 10 CFR 50.46 reporting purposes for any application that utilizes the optional methodology.

Page 10 of 12

UPDATE TO THE AUTO-GENERATION OF THE STRIKIN-II BASE AND CASE DECKS

Background

The Appendix K ECCS Performance Analysis for LBLOCA for CE plants is performed with the 1999 Evaluation Model (1999 EM). Computer code base and case decks provide the inputs to the operation of the 1999 EM. One of the advancements of the 1999 EM has been the auto-generation of base decks that can be created using inputs provided elsewhere in the 1999 EM code system. This eliminates some documentation and quality assurance and improves quality control by eliminating sources of error.

The hot rod heatup computer code of the 1999 EM is STRIKIN-JI. A process improvement has been previously implemented to automatically generate the STRIKIN-I1 base and case decks in a manner consistent with the fuel performance data set from the FATES3B computer code, which provides initial conditions for fuel rod stored energy and rod internal pressure. For licensing applications, an update has been made to the auto-generation of these input decks for LBLOCA.

These updates are categorized as general code maintenance changes to improve the operation and flexibility of this feature of the code.

Affected Evaluation Model(s)

Appendix K LBLOCA Evaluation Model, 1999 EM Estimated Effect This process improvement has no impact on the licensed methodology of the 1999 EM and does not conflict with the SER limitation/constraints imposed on the methodology by NRC. There is no impact on PCT for 10 CFR 50.46 reporting purposes.

Page 11 of 12

2. S2M Version of the Westinghouse Appendix K Small Break LOCA Evaluation Model for Combustion Engineering PWRs None.

Page 12 of 12

Enclosure 2 Summary of the Effect on PCT of the Errors or Changes in the Westinghouse Evaluation Models used in the SONGS Units 2 and 3 ECCS Performance Analyses (Calendar Year 2007)

2007 REPORTING PERIOD LOSS OF COOLANT ACCIDENT (LOCA) MARGIN

SUMMARY

SAN ONOFRE NUCLEAR GENERATING STATION UNITS 2 AND 3 Large Break LOCA Table 1 provides a time line of the items which could affect the Large Break LOCA (LBLOCA) peak clad temperature (PCT) for this reporting period. The 10 CFR 50.46 PCT limit of 2,200'F was not exceeded.

Table 1 Limiting Large Break LOCA PCT iF Unit 3 51 Unit 2 Unit 3 Limiting Large Break LOCA PCT End of 2006 2,170 OF 2,170 OF Changes in PCT during 2007 due to:

a ) Model changes or Model errors

• Cycle 14 0 OF 0 °F b ) Cycle Dependent Input Changes

• Cycle 14 0 OF 00 F Limiting Large Break LOCA PCT End of 2007 2,170 OF 2,170 OF Page 1 of 4

The cumulative 10CFR50.46 model changes and model errors through the end of Cycle 14 for the 1999 EM are shown in Table 2.

Table 2 10 CFR 50.46 Model Changes and Model Errors Y IA PCTI

• IiUnit2 [Unit 3 Cumulative LBLOCA 10 CFR 50.46 Model Changes and Model Errors Priorto 2007 0 OF 0 OF Changes in LBLOCA PCT due to Model Changes and Model Errors Discoveredin 2007

• Cycle 14 0 OF 0 OF Cumulative LBLOCA 10 CFR 50.46 Model Changes and Model Errors 0 OF 0 OF End of 2007

• Sum of the absolute magnitude of the 10 CFR 50.46 model changes and model errors.

Page 2 of 4

Small Break LOCA Table 3 provides a time line of the items which could affect the Small Break LOCA (SBLOCA) peak clad temperature (PCT) for this reporting period. The SBLOCA 10 CFR 50.46 PCT limit of 2200 OF was not exceeded, and the SBLOCA PCT remained bounded by the LBLOCA PCT.

Table 3 Limiting Small Break LOCA PCT Unit 3 Unit 2 Unit 3 Limiting Small Break LOCA PCT End of 2006 2,077 OF 2,077 OF Changes in PCT during 2007 due to:

a) Model changes or Model errors

• Cycle 14 0 OF 0 OF b ) Cycle Dependent Input Changes

• Cycle 14 0 OF 0 0F Limiting Small Break LOCA PCT End of 2007 2,077 OF 2,077 OF Page 3 of 4

The cumulative 10CFR50.46 model changes and model errors for the "Small Break LOCA S2M Evaluation Model" are shown in Table 4.

Table 4 Cumulative Small Break LOCA 10 CFR 50.46 Model Changes and Model Errors IA PCTI

• 11Unit 2 Unit 3 Cumulative SBLOCA 10 CFR 50.46 Model Changes and Model Errors Priorto 2007 0 OF 0 OF Changes in SBLOCA PCT due to Model Changes and Model Errors Discovered in 2007 0 Cycle 14 0 OF 0 OF Cumulative SBLOCA 10 CFR 50.46 Model Changes and Model Errors End of 2007 0 OF 0 OF

• Sum of the absolute magnitude of the 10 CFR 50.46 model changes and model errors.

Page 4 of 4