ML13081A019

Response to Request for Additional Information Regarding License Amendment Request for Permanent Use of Areva Fuel and for Permanent Exemption to Use M5 Cladding

ML13081A019
Person / Time
Site:	San Onofre
Issue date:	03/19/2013
From:	St.Onge R Southern California Edison Co
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
TAC ME6820, TAC ME6821, TAC ME6822, TAC ME6823
Download: ML13081A019 (74)

Text

I EDISON SOUTHERN CALIFORNIA Richard 1. St. Onge Director, Nuclear Regulatory Affairs and E O Emergency Planning An EDISON INTERNATIONAL Company Request to Withhold from Public Disclosure Under 10 CFR 2.390(a)(4) and (6)

March 19, 2013 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001

Subject:

Docket Nos. 50-361 and 50-362 Response to Request for Additional Information Regarding License Amendment Request for Permanent Use of AREVA Fuel and for Permanent Exemption to Use M5 Cladding (TAC Nos. ME6820, ME6821, ME6822, AND ME6823)

San Onofre Nuclear Generating Station, Units 2 and 3

References:

1. Letter from N. Kalyanam (NRC) to P. T. Dietrich (SCE) dated August 1, 2012;

Subject:

San Onofre Nuclear Generating Station, Units 2 and 3 License Amendment Request RE: Use of AREVA Fuel (TAC Nos.

ME6820, ME6821, ME6822, AND ME6823)

2. Letter from D. Bauder (SCE) to J. Sebrosky (NRC) dated July 29, 2011;

Subject:

San Onofre Nuclear Generating Station, Units 2 and 3 Proposed Permanent Exemption Request and Proposed Change, Number (PCN) 600, Amendment Application Numbers 261 and 249, Request for Unrestricted Use of AREVA Fuel.

Dear Sir or Madam:

By email dated February 20, 2013, the Nuclear Regulatory Commission issued a Request for Additional Information (RAI) regarding unrestricted use of AREVA fuel and permanent exemption to use M5 cladding. The RAI requested a response on March 20, 2013. The enclosure of this submittal contains the response to the request for additional information.

Enclosure 2 of this submittal contains information that is proprietary to SCE or AREVA.

SCE requests that this proprietary enclosure be withheld from public disclosure in accordance with 10 CFR 2.390(a)(4). Enclosure 1 provides notarized affidavits from SCE and AREVA which set forth the basis on which the information in Enclosure 2 may be withheld from public disclosure by the Commission and addresses with specificity the considerations listed by paragraph (b)(4) of 10 CFR 2.390. Enclosure 3 provides the non-proprietary version of Enclosure 2.

Notice: This document is decontrolled when separated from Enclosure 2

• pDl P.O. Box 128 San Clemente, CA 92672

Request to Withhold from Public Disclosure Under 10 CFR 2.390(a)(4) and (6)

Document Control Desk March 19, 2013 There are no new regulatory commitments contained in this letter. If you have any questions or require additional information, please contact Mark Morgan, Licensing Lead, at (949) 368-6745.

Sincerely,

Enclosures:

1. Notarized Affidavits Proprietary Enclosures

2. Response to Request for Additional Information (RAI) Part 2 regarding use of unrestricted usage of AREVA fuel and permanent exemption to use M5 cladding Non-Proprietary Enclosures

3. Response to Request for Additional Information (RAI) Part 2 regarding use of unrestricted usage of AREVA fuel and permanent exemption to use M5 cladding cc: E. E. Collins, Regional Administrator, NRC Region IV B. Benney, NRC Project Manager, San Onofre Units 2 and 3 R. Hall, NRC Project Manager, San Onofre Units 2 and 3 G. G. Warnick, NRC Senior Resident Inspector, San Onofre Units 2 and 3 Notice: This document is decontrolled when separated from Enclosure 2

Request to Withhold from Public Disclosure Under 10 CFR 2.390(a)(4) and (6)

ENCLOSURE 1 NOTARIZED AFFIDAVIT Notice: This document is decontrolled when separated from Enclosure 2

AFFIDAVIT STATE OF CALIFORNIA )

) SS.

CITY OF SAN CLEMENTE)

1. My name is Vickram Nazareth. I am employed by Southern California Edison Company ("SCE"). My present capacity is Manager, Nuclear Fuel Management, for the San Onofre Nuclear Generating Station ("SONGS"), and in that capacity I am authorized to execute this Affidavit.

2. SCE is the operating agent for SONGS. I am familiar with the policies established by SCE to determine whether certain SCE information is proprietary and confidential, and to ensure the proper application of these policies.

3. I am familiar with SCE information in the following document: "Response to Request for Additional Information (Set 2) Southern California Edison, San Onofre Nuclear Generating Station Units 2 and 3, Proposed License Amendment Request for Unrestricted Use of AREVA Fuel and for Permanent Exemption to Use M5 Cladding, Docket Nos. 50-361 and 50-362, TAC Nos: ME6820, ME6821, ME6822, AND ME6823.

4. SCE has classified the information contained in this document as proprietary and confidential in accordance with SCE's policies.

5. Specifically, SCE applied the following criteria to determine that the information contained in the document should be classified as proprietary and confidential:

(a) SCE has a non-disclosure agreement with Westinghouse Electric LLC

("Westinghouse") and AREVA NP ("AREVA") (the NDA is referred to as the "Westinghouse-AREVA-SCE NDA"), under which Westinghouse and AREVA have provided to SCE certain proprietary and confidential information contained in the document.

(b) The information reveals details of Westinghouse's, SCE's, and/or AREVA's research and development plans and programs, or the results of these plans and programs.

(c) The information includes test data or analytical techniques concerning a process, methodology, or component, the application of which results in a competitive commercial advantage for Westinghouse, SCE, and/or AREVA.

(d) The information reveals certain distinguishing aspects of a process, methodology, or component, the exclusive use of which provides a competitive commercial advantage for Westinghouse, SCE, and/or AREVA on product optimization or marketability.

(e) The unauthorized use of the information by one of Westinghouse's, SCE's, and/or AREVA's competitors would permit the offending party to significantly reduce its expenditures, in time or resources, to design, produce, or market a similar product or service.

(f) The information contained in the document is vital to a competitive commercial advantage held by Westinghouse, SCE, and/or AREVA, would be helpful to their competitors, and would likely cause substantial harm to the competitive position of Westinghouse, SCE, and AREVA.

6. The information contained in the document is considered proprietary and confidential for the reasons set forth in Paragraph 5. In addition, the information contained in the document is of the type customarily held in confidence by AREVA, Westinghouse, and SCE, and not made available to the public. Based on my experience in the nuclear industry, I am aware that other companies also regard the type of information contained in the document as proprietary and confidential.

7. In accordance with the Westinghouse-AREVA-SCE NDA, the document has been made available to the NRC in confidence, with the request that the information contained in this document be withheld from public disclosure. The

request for withholding the information from public disclosure is made in accordance with 10 CFR 2.390. The information qualifies for withholding from public disclosure under 10 CFR 2.390(a)(4) "Trade secrets and commercial or financial information."

8. In accordance with SCE's policies governing the protection and control of proprietary and confidential information, the information contained in the document has been made available, on a limited basis, to others outside Westinghouse, SCE and AREVA only as required in accordance with the Westinghouse-AREVA-SCE Nondisclosure Agreement.

9. SCE's policies require that proprietary and confidential information be kept in a secured file or area and distributed on a need-to-know basis. The information contained in the document has been kept in accordance with these policies.

10. The foregoing statements are true and correct to the best of my knowledge, information, and belief, and if called as a witness I would competently testify thereto. I declare under penalty of perjury under the laws of the State of California that the above is true and correct.

Executed on (-I-

• ot 0(.*

Date Vickr zareth

California All-Purpose Acknowledgment 2008 Code Section 1189 Compliant State of Californ County of Q,_-/*

)JeL On Ma rrA/ & .J)10 before me,

• 42'1/27,n406- /

(here insert name and title of the officer) /

P/rrc av)9i, personally appeared &/mJj.c2!2i who proved to me on the basis of satisfactory evidence to be the person(s) whose name(s) is/are subscribed to the within instrument and acknowledged to me that he/she/they executed the same in his/her/their authorized capacity(ies), and that by his/her/their signature(s) on the instrument the person(s), or the entity upon behalf of which the person(s) acted, executed the instrument.

I certify under PENALTY OF PERJURY under the laws of the State of California that the foregoing paragraph is true and correct.

WITNESS my hand and official peal.

Comm. 9t94267 Signature If~otaryPub~lic.-C.Hsi" No.San Diego County (Seal)

OPTIONAL INFORMATION Law does not require the information below. This information could be of great value to any person(s) relying on this document and could prevent fraudulent and/or the reattachment of this document to an unauthorized document(s)

DESCRIPTION OF ATTACHED DOCUMENT Title or Type of Document:

Document Date: Number of Pages:

Signer(s) if Different Than Above:

Other Information:

CAPACITY(IES) CLAIMED BY SIGNER(S)

Signer's Name(s):

o Individual O Corporate Officer (Title(s))

o Partner O Attorney-in-Fact O Trustee

" Guardian/Conservator O Other:

SIGNER IS REPRESENTING:

Name of Person(s) or Entity(ies):

© 2008 Notary Public Seminars wwwv.notarypublicseminars.com

Request to Withhold from Public Disclosure Under 10 CFR 2.390(a)(4) and (6)

ENCLOSURE 3 Response to Request for Additional Information (RAI) Part 2 regarding use of unrestricted usage of AREVA fuel and permanent exemption to use M5 cladding (Non-Proprietary)

Notice: This document is decontrolled when separated from Enclosure 2

SOUTHERN CALIFORNIA EDISON RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION SET 2 SOUTHERN CALIFORNIA EDISON SAN ONOFRE NUCLEAR GENERATING STATION, UNITS 2 AND 3 PROPOSED LICENSE AMENDMENT REQUEST FOR UNRESTRICTED USE OF AREVA FUEL AND FOR PERMANENT EXEMPTION TO USE M5TM CLADDING DOCKET NOS. 50-361 AND 50-362 TAC NOS. ME6820, ME6821, ME6822, AND ME6823)

SUBJECT PAGE RA I#1 ............................................................................................................................................. 3 RA I #2 ............................................................................................................................................. 4 RA I#3 ............................................................................................................................................. 6 RA I#4 ............................................................................................................................................. 7 RA I #5 ........................................................................................................................................... 13 RA I #6 ........................................................................................................................................... 14 RA I #7 ........................................................................................................................................... 16 RA I #8 ........................................................................................................................................... 18 RA I #9 ........................................................................................................................................... 20 RA I #10 ......................................................................................................................................... 21 RA I#11 ......................................................................................................................................... 22 RA I#12 ......................................................................................................................................... 24 RA I #13 ......................................................................................................................................... 25 RA I #14 ......................................................................................................................................... 26 RA I #15 ......................................................................................................................................... 28 RA I #16 ......................................................................................................................................... 30 RA I #17 ......................................................................................................................................... 31 RA I#18 ......................................................................................................................................... 32 RA I #19 ......................................................................................................................................... 34 RA I #20 ......................................................................................................................................... 35 Page 1 of 66

RA I #2 1......................................................................................................................................... 38 RA I #22 ......................................................................................................................................... 39 RA I #2 3 ......................................................................................................................................... 40 R A I #2 4 ......................................................................................................................................... 41 R A I #25 ......................................................................................................................................... 43 R A I #2 6 ......................................................................................................................................... 44 RA I #2 7 ......................................................................................................................................... 46 R A I #2 8 ......................................................................................................................................... 48 R A I #2 9 ......................................................................................................................................... 49 R A I #30 ......................................................................................................................................... 50 RA I #3 1......................................................................................................................................... 51 R A I #32 ......................................................................................................................................... 54 R A I #33 ......................................................................................................................................... 61 R A I #34 ......................................................................................................................................... 62 RA I #3 5......................................................................................................................................... 63 R A I #36 ......................................................................................................................................... 65 Page 2 of 66

RAI #1

1. The licensee has requested a permanent exemption from requirements of 10 CFR 50.46 and 10 CFR 50 Appendix K to use M5TM clad AREVA fuel in SONGS Units 2 and 3. The license amendment also requests unrestricted use of AREVA fuel in SONGS Units 2 and 3. In section 3.2.3.2 of Enclosure 2, the licensee stated that "The changes consist of changes to SONGS Technical Specification 5.7.1.5 (CORE OPERATING LIMITS REPORT (COLR)) methodology reference list that are required to support the core design and implementation of unrestricted use of the potential alternative vendor fuel assemblies." Please explain what is meant by "implementation of unrestricted use of potential alternative vendor fuel assemblies."

RESPONSE

"Implementation of unrestricted use of potential alternative vendor fuel assemblies" means that SCE may design and implement cores using either the existing Westinghouse fuel design, or the AREVA CE-HTP fuel without lead test assembly (LTA) limitations, based on the methods described in the approved licensed methods for SONGS, SCE-9801-P-A, as updated per this LAR.

Page 3 of 66

RAI #2

2. (Section 8.0 References) Reference No. 8.8, BAW-10227(P)-A, Revision 0 is the only version of the topical report that describes the "Evaluation of Advanced Cladding and Structural Material M5)" that was docketed at the NRC. Reference No.

8.9 lists a Revision 1 of BAW-1 0227(P)-A. Please clarify this apparent discrepancy

RESPONSE

Revision 0 is correct and should have been cited. Revision I will be removed.

A replacement page is provided on the following page.

Page 4 of 66

LICENSEE'S EVALUATION SONGS PCN-600 Request for Unrestricted Use Of AREVA Fuel

8.0 REFERENCES

8.1 Letter From SCE (Short) to NRC dated January 30, 2009, "Request for Temporary Exemption from the Provisions of 10 CFR 50.46 and 10 CFR 50, Appendix K for Lead Fuel Assemblies... for San Onofre Nuclear Generating Station, Units 2 and 3," (ADAMS Accession Number ML090360738).

8.2 Letter from NRC (Hall) to SCE (Ridenoure) dated December 17, 2009, "SONGS, Units 2 and 3 - Temporary Exemption from the Requirements of 10 CFR Part 50, Section 50.46 and Appendix K for Lead Fuel Assemblies (TAC No. ME0602 and ME0603),"

(ADAMS Accession number ML090860415).

8.3 ANP-2839(P), Revision 1, "San Onofre Nuclear Generating Station Lead Fuel Assemblies Fuel Design Criteria Review," September 2009 8.4 Southern California Edison Company (SCE), "Reload Analysis Methodology for the San Onofre Nuclear Generating Station Units 2 and 3," Topical Report SCE-9801-P, November 1998 and approved version Topical Report SCE-9801-P-A, June 1999.

8.5 Southern California Edison Company (SCE), "PWR Reactor Physics Methodology Using Studsvik Design Codes," Topical Report SCE-090 1-A, December 2009.

8.6 Letter from NRC (Hall) to SCE (Ridenoure) dated December 15, 2009, "San Onofre Nuclear Generating Station Units 2 and 3 - Issuance of Amendments Revising Technical Specification 5.7.1.5, Core Operating Limits Report (COLR)" (TAC Nos. ME0604 and ME0605)" (ADAMS Accession Number ML093220105) 8.7 CENPD-404-P-A,"Implementation of ZIRLOTM Cladding Material in CE Nuclear Power Fuel Assembly Designs," November 2001.

8.8 Same as Reference 8.9.

8.9 BAW- 10227(P)-A, Rev 0, "Evaluation of Advanced Cladding and Structural Material (M5 TM) in PWR Reactor Fuel," approved by the NRC in February 2000. This Reference contains the NRC Safety Evaluation and approval letter identified in Reference 8.13.

8.10 BAW-10240(P)-A, "Incorporation of M5TM Properties in Framatome ANP Approved Methods," approved by the NRC in May 2004.

8.11 BAW- 10241 (P)(A), Revision 1, "BHTP DNB Correlation Applied with LYNXT,"

approved by the NRC in July 2005.

8.12 XN-NF-82-06 (P)(A), Revision 1 and Supplements 2, 4 and 5, "Qualification of Exxon Nuclear Fuel for Extended Burnup."

Page 162 of 166 Page 5 of 66 Enclosure 2 to SONGS PCN-600

RAI #3

3. (Section 4.2.2) Thermal-HydraulicTreatment of Rod Bow Penalty, states that rod bow evaluation is performed using References 8.16 (CENPD-225-P-A, June 1983 for CE fuel) and 8.18 (XN-75-32(P)(A), February 1983) which are legacy methodologies, which, since their publication have revised with later updates. The NRC staff would refer the licensee to Section 3.9 of Topical Report (TR), BAW-10227P-A, February 2000 and Section 6.1.6 of TR, BAW-10240(P)-A, Revision 0, May 2004 for rod bow analysis updates. Please explain why SONGS did not use the latest methodology for rod bow analysis and show that rod bow penalty listed in Table 4.2 of Enclosure 2 of license amendment request (LAR) is still accurate despite the use of the older methodology.

RESPONSE

XN-75-32(P)(A)

The AREVA rod bow methodology was originally developed in Topical Report XN-75-32(P)(A) for Zircaloy-4 (Zr-4) clad fuel rods. This methodology was extended to fuel rod bumups _<62 GWd/MTU in Topical Report XN-NF-82-06(P)(A) Revision 1 and Supplements 2, 4, and 5.

Rod bow for AREVA M5TM clad fuel rods was generically addressed in Section 3.9 of Topical Report BAW-10227(P)(A). This topical report concluded that rod bowing in M5TM clad fuel rods would be (( )) the rod bow of equivalent Zr-4 clad fuel rods. Topical Report BAW-1 0240(P)(A) establishes the link between the M5 assessment in BAW-10227(P)(A),

and the specific rod bow methodology from XN-75-32 (P)(A).

Section 6.1.6 in BAW-1 0240(P)(A) notes that:

"To evaluate the effect of rod bow on thermal margins, FRA-ANP uses a conservative rod bow projection, which is (( ))with bumup."

The conservative, (( )) projection of rod bow with burnup noted in BAW-10240(P)(A) is the relationship that was developed in topical report XN-75-32(P)(A). This is confirmed through the citation of XN-75-32(P)(A) as Reference 12 in Section 9.0 of BAW-10240(P)(A).

SCE referenced the base topical report (XN-75-32(P)(A)) rather than the subsequent topical reports which verify that this topical report remains applicable for M5TM cladding.

The AREVA rod bow penalty listed in Table 4.2 of Enclosure 2 of the LAR is valid and conservative.

CENPD-225-P-A CENPD-225-P-A (including Supplements 1, 2, and 3) as referenced in SCE-9801-P-A remains the applicable Rod Bow Methodology for Westinghouse manufactured CE fuel used at SONGS.

Page 6 of 66

RAI #4

4. (Section 4.3.2) The licensee states that "SCE will be generatingfuel rod behavior analysis data per Reference 8.25 to support the non-LOCA [loss-of coolant accident] safety analyses and calculations that support SCE setpoints analyses (perReference 8.4). Specifically, ff 1]

Section 3.4 of AREVA TR, BAW-1 0240(P)(A), Revision 0, "Incorporation of M5 Properties in Framatome ANP Approved Methods," lists several NRC-approved AREVA topical reports. BAW-10240(P)(A) listing discusses how each report and the associated methodology is impacted by the incorporation of M5 TM cladding. The list consists of codes and methodologies for fuel performance, fuel mechanical design, fuel rod bow, fuel generic design, neutronics, thermal hydraulics/mixed core analysis, non-LOCA Chapter 15 analyses, DNB correlation, statistical setpoint methodology for Westinghouse plant types, statistical methodology for CE plant types, and seismic analyses all of which are AREVA codes and methodologies.

AREVA recommends its own methodologies and codes to analyze M5TM clad fuel.

Contrary to what AREVA's recommendation, the licensee is proposing to use CE/W fuel performance code, FATES3B, for several analyses for the new M5TM clad fuel.

The FATES3B code is approved by the NRC for analysis of CE/W fuel. Provide a detailed justification for the use of the FATES3B code to analyze AREVA fuel or alternatively provide the AREVA fuel design criteria evaluation using an approved AREVA methodology.

RESPONSE

AREVA does not have a NRC approved transient analysis methodology for CE plants with COLSS/CPC monitoring and protection systems.

As discussed in the license amendment request (LAR) (Enclosure 2, Section 4.3.2), AREVA will be performing all fuel mechanical design analyses for Chapter 4 of the Standard Review Plan and LOCA analyses, including the fuel rod initial conditions for the AREVA LOCA analyses.

SCE will use existing approved methodology, SCE-9801-P-A, to establish the initial rod conditions to support the non-LOCA safety analysis. To achieve this, ((

)). Justification for this approach is provided in Section 4.3.2 of Enclosure 2 to the LAR.

Page 7 of 66

Non-LOCA Transient Analyses As described in the LAR, Enclosure 2, Section 4.3.2, SCE will use FATES3B calculations to provide (( )) values for input to the transient analyses, and to provide (( )) the CEA Ejection transient analysis.

Depending on the non-LOCA transient being evaluated, it is appropriate to model the event with either a ((

)), respectively. The specific fuel performance output parameters described below are consistent with the SCE Reload Analysis Methodology topical, described in SCE-9801-P-A, Section 3.3.2, and are appropriate considering the similarity of AREVA fuel to the existing Westinghouse/CE fuel.

((

1]

The CEA ejection transient analyst selects initial conditions ((

)) as calculated by the FATES3B code.

The Doppler effect mitigates the CEA ejection transient. To (( )) the Doppler effect the analyst chooses initial conditions ((

To (( )) heat retention within the fuel rod, the analyst chooses initial conditions associated with the ((

LOCA Transient Analyses As described in the LAR, Enclosure 2, Section 4.3.2, the FATES3B code is not used in the AREVA LOCA Transient Analyses. AREVA will use their approved fuel behavior codes and methodology to perform all of the fuel mechanical design calculations for input to their LOCA analysis for the SONGS units. The AREVA fuel mechanical design analyses will be originated using codes and methods that have been NRC approved for their intended purposes as described in Enclosures 3 and 4 of PCN-600.

Page 8 of 66

Fission Gas Release for Rod Internal Pressure Calculations (i.e., no-clad liftoff)

As described in the LAR, Enclosure 2, Section 4.3.2, the FATES3B code is not used in the AREVA fuel rod operating pressure calculations. AREVA will use their approved fuel behavior codes and methodology to perform all of the fuel mechanical design calculations for input to their LOCA analysis for the SONGS units. The AREVA fuel mechanical design analyses will be originated using codes and methods that have been NRC approved for their intended purposes as described in Enclosures 3 and 4 of PCN-600.

Power to Fuel Centerline Melt The fuel performance output parameter of Power to Fuel Centerline Melt is consistent with the SCE Reload Analysis Methodology topical, described in SCE-9801-P-A, Section 3.3.2, and is appropriate considering the similarity of AREVA fuel to the existing Westinghouse/CE fuel.

The FATES3B code ((

Conservatism in the fuel behavior analysis results is accomplished by the means described in the NRC Question 3.A response as presented in CEN-193(B)-P Supplement 2-P. The response states that the code results ((

)) Consideration of this margin, and the previously discussed conservatisms in input and modeling, it is acceptable for FATES3B to calculate power to fuel centerline melt.

Page 9 of 66

Clad Strain and Fatigue The FATES3B code is not used in the clad strain and fatigue analyses for AREVA fuel, which will be performed by AREVA as part of their fuel rod mechanical design.

Use of FATES3B with AREVA Fuel Combustion Engineering letter LD-89-055 (referenced in CEN-161(B)-P, Supplement 1-P-A) specifically requests generic approval of the FATES3B code:

"The purpose of this letter is to request generic approval of the FATES3B [Reference (A)] fuel performance code for analysis of pressurized PWR fuel in C-E plants."

Additionally, the FATES3B generic request letter enclosure states in its discussion the generic applicability of the FATES3B and other fuel performance codes:

"The FATES3B fuel performance code is generically applicable to PWR fuel as evidenced by its development history summarized above. The various versions of FATES have historically been evaluated against an extensive variety of fuel rod designs and power histories. The code applicability extends to all zircaloy clad UO 2 fuel pellets currently used in PWRs."

It is worth noting that the FATES3B code has since been modified and approved to model the Westinghouse ZIRLO cladding material (CENPD-404-P-A) in much a similar fashion that SCE is proposing to include the AREVA M5TM cladding, and that both ZIRLO and M5TM are substantially similar to Zircaloy-4 for which FATES and many other fuel performance codes were originally designed to model.

The letter's discussion continues to describe the database used for the model development:

"The data base for model development and for overall code verification has been broad in the range of fuel designs and irradiation histories covered, beginning with the initial version of FATES (Reference 1) and continuing through FATES3B (Reference 10). The broad nature of the data base is typical of the high burnup fission gas release data used to develop and verify the gas release model in FATES3B. The verification data base includes fuel designs ranging in diameter from smaller than the C-E 16x1 6 designs to larger than the C-E 14x14 designs. This includes fuel fabricated by other U. S. fuel vendors (e.g., Westinghouse) as well as foreign fuel vendors (e.g., KWU) and also includes fuel rods irradiated in both commercial cores and special test reactors."

The discussion reiterates that fuel performance codes often have wide applicability as evidenced by comparing to other fuel performance codes:

"The adequacy of the fuel performance model has also been established by comparisons with similar models. ... The practice of comparing fuel performance code results demonstrates a belief that all fuel codes tend to have generic applicability."

Page 10 of 66

This is true of the NRC's own code, FRAPCON-3.4 (NUREG/CR-7022, Vol. 1) which states even more broadly, "FRAPCON-3 is a Fortran 90 computer code that calculates the steady-state response of light-water reactor fuel rods during long-term burnup." In practice, FRAPCON-3 has been used acceptably for a wide variety of vendor fuel (e.g., C-E, Westinghouse, AREVA, GE) for both PWR and BWR fuel rods.

Similarly, the discussion on FATES3B continues to describe how the code may be used on several different designs of fuel:

"Generic applicability of FATES3B is facilitated through user input options. The FATES3B input was designed such that it can be used to describe fuel design features that may differ (such as diameters, fill gas pressure, etc.) and to describe specific fuel fabrication characteristics and/or microstructure that will influence behavior during irradiation (such as fuel grain size, porosity, surface roughness, densification, etc.)."

Additionally, the fuel rod data base used to model the fuel behavior within the FATES code relied not only on Combustion Engineering fuel, but also Kraftwerk Union (KWU) designed fuel used in the Obrigheim (KWO) and Stade (KKS) reactor (see CEN- 161(B)-P-A, Section 9.0).

These rods were selected for their similarity to the C-E designed fuel rods, much as AREVA has designed fuel rods with similar specifications to the existing C-E/Westinghouse designed fuel in use by SCE (see LAR Section 5.1.3.1).

The Staff's evaluation letter with subject "Generic Approval of C-E Fuel Performance Code FATES3 B (CEN- 161 (B)-P, Supplement l-P)" states, in part:

"On May 19, 1989, you requested NRC review and generic approval of the C-E fuel performance code FATES3B described in the topical report CEN-161 (B)-P, Supplement I-P. The FATES3B code was approved for licensing applications for Calvert Cliffs Units I and 2 in an NRC safety evaluation dated February 4, 1987. Based on your submittal and review of the previously approved SER, we conclude that the FATES3B code is not necessarily plant-specific for Calvert Cliffs I and 2 and therefore FATES3B can be applied generically to other C-E plants. ... In summary, the NRC staff approves the generic applicability of FATES3B for licensing applications. Our evaluation applies only to matters described in the topical report."

The statements in the Staff s evaluation letter permit the FATES3B code to be used generically on C-E plants within the limits of applicability of topical CEN- 161 (B)-P, Supplement 1-P-A. As described above, CEN-161 (B)-P, Supplement 1-P-A and its supporting letter (LD-89-055), and topicals (e.g., CEN- 193(B), Supplement 2-P, CEN-345(B)-P, CEN- 139-P-A, CENPD-275-P-A, etc.) modeled and approved FATES3B for UO 2 fuel with Erbia and Gadolinia (also see SCE's response to RAI Set #1 question #2) with a fuel database that was not specific to Combustion Engineering. SCE is requesting AREVA provide SCE with U0 2 fuel and U0 2 with Gadolinia that has been specified to be substantially similar to the existing fuel in use by SCE. SCE has modified the FATES3B code to incorporate the physical properties of the M5TM cladding and submitted this change to the NRC. Given the Staff s generic approval of the FATES3B code for Page II of 66

licensing analyses for fuel in C-E type plants and that the basis for modeling urania fuel within the code encompassed a variety of PWR fuel designs - including those by other manufacturers -

it is acceptable to use the FATES3B code to analyze AREVA fuel.

Page 12 of 66

RAI #5

5. (Section 4.3.3.1.1) The licensee states that it used M5T cladding fuel rod growth M

equation for the mean (AL/Lo) and justifies the action by the statement that "The use of a nominal projection of M5 cladding fuel rod growth is consistent with the use of M

T a nominal projection in the currently approved FATES3B methodology, because the local fast neutron flux is biased to the low side in licensing calculations (

Reference:

CEN-1 93(B))."

BAW-1 0240(P)-A, "Incorporationof M5 Propertiesin FramatomeANP Approved Methods" describes how the NRC-approved M5 TM material properties (

References:

BAW-1 0227P-A and BAW-1 0231 P-A) in to a set of AREVA-(Formerly FANP) approved mechanical analysis, SBLOCA, and non-LOCA methodologies. The licensee has indicated that the local fast neutron flux will continue to drive the cladding growth in the FATES3B w/ MST M cladding model and the overall calculation of cladding growth will be conservative. Provide a detailed justification for the use of the FATES3B code for their fuel mechanical analysis or alternatively provide the analysis using an approved AREVA methodology.

RESPONSE

FATES3B will not be used for the fuel mechanical analysis of AREVA fuel as discussed in the LAR. AREVA will use their approved fuel behavior codes and methodology to perform the fuel mechanical design calculations and for input to their LOCA analysis for the SONGS units. The AREVA fuel mechanical design analyses will be originated using codes and methods that have been NRC approved for their intended purposes as described in Enclosures 3 and 4 of the LAR.

AREVA fuel mechanical design analysis is discussed in the LAR Enclosure 2, Sections 4.3.2, 4.7 and 5.1, including fuel rod growth.

Page 13 of 66

RAI #6

6. Section 4.3.3.1.5 describes how the M5 thermal conductivity is calculated. As M

T described in the LAR, the AREVA's equation for M5TM thermal conductivity in LAR is taken from Section 4.1.3 of Reference 8.10 (BAW-1 0240(P)-A) (Equation 4.3) which is valid for a temperature range of 273 K (32°F) to 1600 K (2421OF). The licensee has transformed the quadratic relationship for M5TM thermal conductivity to a linear relationship as shown in Section 4.3.3.1.5 of the LAR and changed the range of applicability of temperature to a limited range 400°F to 1000°F in order to satisfy the requirement in FATES3B which is fuel performance code designed for CE/W fuel. Provide justification for the use of CEIW fuel performance codes for the treatment of M5 thermal conductivity or alternatively use an alternate NRC M

T approved methodology for the analysis. In addition, provide a justification based on analytical treatment for the use of a linear function as opposed to a quadratic function, and a justification for the change in the temperature range from that discussed in BAW-10240.

RESPONSE

Use of FATES3B with M TM Cladding Thermal Conductivity Properties The treatment of M5TM properties was previously approved for use by AREVA in fuel performance analysis. The LAR describes the proposed incorporation of this treatment of M5TM properties into the FATES3B code. The thermal conductivity for M5TM cladding was taken from the NRC-approved topical BAW-10240(P)-A, Section 4.1.3, and is identical to Equation 10-30 in NRC-approved topical BAW- 10231 P-A. AREVA (Framatome ANP) has used the M5TM thermal conductivity equation in several codes, including COPERNIC (BAW- 10231 P-A),

RODEX2-2A (BAW- 10240(P)-A), and S-RELAP5 (BAW- 10240(P)-A).

Additionally, BAW-10227P-A, Section 4.1 notes that "...the thermal conductivity is not affected by changes in minor alloying constituents and will not vary between M5TM and Zircaloy." Since thermal conductivity of a metal alloy (such as Zircaloy or M5TM) is a generic material property and is not specific to any safety analysis methodology or code, AREVA adopted the existing Zircaloy thermal conductivities for M5 TM (BAW-10227P-A, Section A.2.2.2).

SCE concludes it is appropriate to use these correlations within the FATES313 code for modeling fuel with M5 TM cladding.

Use of the Linear Thermal Conductivity Equation with Reduced Temperature Range in FATES3B The modeling of M5 TM clad fuel in the FATES3B code requires an equation to define the approximate M5TM thermal conductivity over the temperature range experienced by the M5TM cladding. In the FATES3B code other cladding material thermal conductivity equations were modeled linearly with respect to temperature. To minimize changes to the FATES3B code structure, the M5TM thermal conductivity equation was modified from a quadratic relationship to Page 14 of 66

a linear relationship. Since the FATES3B code is designed to model steady-state conditions, the range of applicability of temperature for the M5TM thermal conductivity equation was changed to a limited range of (( )), which bounds the expected temperature range from approximately 535°F (corresponding to the minimum reactor coolant inlet temperature) to 8007F (typical maximum cladding temperature). LAR Enclosure 2, Figure 4.3.1, documents the close similarity (( )) between the two curves over the equation's applicable temperature range.

Page 15 of 66

RAI #7

7. With respect to M5T thermal expansion coefficient (Section 4.3.3.1.6 of LAR), the NRC staff notes that the treatment is not approved by the NRC since (1) AREVA fuel performance code RODEX2-2A is replaced with CE/W code, FATES3B for analyzing AREVA clad properties, and (2) the temperature range of applicability is not fully covered in the altered treatment. Provide a justification for the use of the CE/W code, or alternatively submit the analysis using an NRC approved methodology.

RESPONSE

Use of FATES3B with M5TM Cladding Thermal Expansion Properties The modeling of M5TM clad fuel in the FATES3B code requires an equation to define the approximate M5TM thermal expansion coefficients in the axial and tangential directions over the temperature range experienced by the M5TM cladding.

The coefficients listed in LAR Section 4.3.3.1.6 are from NRC approved topical BAW-10240(P)-A, "Incorporation of M5TM Properties in Framatome ANP Approved Methods." As noted by the Staff, this topical specifically limits the use of M5TM and its properties in FANP (AREVA) approved methods. SCE is requesting use of AREVA's M5TM properties in the non-AREVA code, FATES3B.

The thermal expansion coefficients for M5TM listed in BAW- 10240(P)-A, Section 4.1.7 are from the M5TM topical BAW-10227P-A, "Evaluation of Advanced Cladding and Structural Material (M5) in PWR Reactor Fuel." BAW-10227P-A, Appendix K.2.4, states the analysts, "have experimentally determined the coefficients of thermal expansion for M5TM." These correlations are independent of any safety analysis methodology.

Further, the inherent similarities of M5TM to other zirconium based cladding alloys is apparent in Framatome ANP's initial plan to use standard the standard MATPRO data for Zircaloy-4 with M5TM. As stated in BAW-10227P-A, Appendix I on page 1-70:

"The thermal expansion model developed in BAW-10227 was based on the expectation that the expansion of M5TM would be similar to that of Zircaloy. Following the issue of BAW- 10227, dilatometry testing of M5TM, of as-manufactured tubing, has been completed. Figure 1-5-1 shows the resultant base thermal expansion correlations and the average to be implemented in the LOCA and Safety Analysis. ((

1]",

Thus, Framatome ANP (AREVA) ((

)) as described in BAW-10227P-A, Appendix K.2.4.

Page 16 of 66

It is also worth noting that in Westinghouse's implementation of its zirconium based alloy ZIRLO into FATES3B, ((

)). This was recognized and accepted by the Staff in the Safety Evaluation of CENPD-404-P, Section 2.3.

SCE has elected to model the thermal expansion coefficients of M5TM based on the experimentally derived correlations presented in BAW-10240(P)-A, Section 4.1.7. Given the similarities of the materials already modeled by FATES3B to M5TM, and given the experimental data in BAW- 10227P-A, Appendix K.2.4 is independent of safety analysis methodology, SCE believes it is appropriate to use these correlations within the FATES3B code for modeling fuel with M5 TM cladding.

Use of the Reduced Temperature Range Correlation in FATES3B The temperature range experienced by the M5TM cladding is expected to range from approximately 5357F (corresponding to the minimum reactor coolant inlet temperature) to 800 0 F.

As discussed in the License Amendment Request (LAR) (Enclosure 2, Section 4.3.3.1.5), the M5TM thermal expansion coefficients over the expected fuel clad temperature range are bounded within the AREVA M5TM thermal expansion coefficient equations at the low temperature range of (( )). Therefore, programming only the low temperature range into the FATES3B code is appropriate.

Page 17 of 66

RAI #8

8. What is the technical basis for the three acceptance criteria listed in Section 4.3.3.2.1 of the LAR, namely, (1) selection of M5T creep behavior, (2) the acceptance range of +/- 0.7 mil in the computed creep behavior of M5TM clad fuel rods, and (3) the selection of axial growth. Describe the "fuel rod test cases that have been used to verify the overall performance of FATES3B w/ M5 and to M T

demonstrate compliance with the acceptance criteria." Describe what are the "hypotheticalM5TM fuel rod cases"mentioned in item number 2 on page 37 of the LAR.

RESPONSE

Description of Acceptance Criteria As discussed in the License Amendment Request (LAR) (Enclosure 2, Section 4.3.3.2.1),

Acceptance Criterion I is not related to M5TM creep behavior. Acceptance Criterion 1 requires that the addition of M5TM material property models will have no impact on the existing cladding material property models or the results of analyses using Zircaloy-4 and ZIRLOTM cladding materials. The technical basis for Acceptance Criterion 1 is to satisfy the requirement that the results of the calculations for Zircaloy-4 and ZIRLOTM clad fuel rods using FATES3B w/M5TM must produce the same numerical results as the unmodified version of FATES3B when executed with the same input deck on the same computer platform. Compliance with this requirement demonstrates that the code modifications made for the addition of M5TM cladding material property models have not affected the calculations for Zircaloy-4 and ZIRLOTM cladding materials.

As discussed in the LAR (Enclosure 2, Section 4.3.3.2.1), Acceptance Criterion 2 requires that the computed creep behavior of M5TM clad fuel rods shall be within a mil m] acceptance range. This acceptance range was selected ((

I))

As discussed in the LAR (Enclosure 2, Section 4.3.3.2.1), Acceptance Criterion 3 requires that the computed axial growth behavior of M5TM clad fuel rods shall be within the tolerance defined by the acceptance criteria based on available measured axial growth data. AREVA Topical Report BAW- 10240(P)-A Section 6.1.7 and Figure 6.1 uses bounding empirical models to compute the irradiation growth of the clad and fuel assemblies. The bounding models are established by a 95/95 statistical evaluation of the growth measurements. These models provide a nominal projection for fuel rod growth that has been encoded into the FATES3B w/M5TM code.

The technical basis for Acceptance Criterion 3 ensures that the calculated axial growth data for the M5TM licensing cases generated with FATES3B w/M5TM coincide with the best fit curve and are within the 95/95 tolerance interval for the measured M5TM fuel rod empirical growth data.

Page 18 of 66

Description of Fuel Rod Test Cases As discussed in the LAR (Enclosure 2, Section 4.3.3.2.1), fuel rod test cases have been used to verify the overall performance of FATES3B w/M5TM and to demonstrate compliance with the acceptance criteria. As discussed in the following paragraphs, the test cases include three types of fuel rods:

1. Zircaloy-4 and ZIRLOTM fuel rod cases obtained from previously originated SONGS core design analyses (e.g., Unit 3 Cycle 16) that used the unmodified version of FATES3B.

These test cases are referred to as "Zircalov-4 and ZIRLOTM licensing cases".

2. Hypothetical M5TM fuel rod cases created by changing the clad material in the Zircaloy-4 licensing cases described in Item 1. These test cases are referred to as "MSTM licensing cases".

3. Test cases based on six M5TM fuel rods that have been irradiated between I and 3 cycles.

These test cases are referred to as the "six M5Tmfuel rod cases". These same test cases were evaluated in AREVA Topical Report BAW-10240(P)(A), Revision 0, "Incorporation of M5 Properties in Framatome ANP Approved Methods".

Zircaloy-4 and ZIRLOTM licensing cases [Acceptance Criterion 1]: The Zircaloy-4 and ZIRLOTM fuel rod Licensing Cases were obtained from SONGS Units 2 and 3 reload calculations that used the unmodified version of FATES3B. The output from the Zircaloy-4 and ZIRLOTM fuel rod Licensing Cases produced from the modified FATES3B code were compared to output files generated with the unmodified FATES3B code. These case comparisons demonstrated that the code modifications incorporated into FATES3B have not adversely affected the coding in the unmodified version of FATES3B. Criterion I as discussed in the response to above was satisfied when the modified FATES3B code produced the same numerical results as the unmodified FATES3B code when executed with the same input deck on the same computer platform.

M5TM licensing cases [Acceptance Criterion 3]: The sole difference between the M5TM licensing cases and the Zircaloy-4 and ZIRLOTM licensing cases described in the preceding paragraph is that the M5TM licensing cases activate the M5TM cladding material option in FATES3B to model M5TM cladding rather Zircaloy-4 or ZIRLOTM cladding. The M5TM Licensing Cases were evaluated based on a review of the overall fuel rod behavioral trends with respect to the Zircaloy-4 and ZIRLOTM Licensing Case results presented in LAR Enclosure 2 Figures 4.3.6 through 4.3.8. Further, the M5TM licensing case results have been used for comparison to the M5TM axial growth behavior (i.e., Criterion 3 as discussed in the response above). Demonstration of compliance with Criterion 3 was achieved when the calculated axial growth data for the M5TM licensing cases generated with FATES3B w/M5TM was found to coincide with the best-fit curve and are within the 95/95 tolerance interval for the measured M5TM fuel rod growth data from AREVA Topical Report BAW-10240(P)-A (Section 6.1.7.1, page 6-6; Figure 6.1, page 6-25).

Six M5TM fuel rod cases [Acceptance Criterion 21: The results from the six M5TM fuel rod cases have been evaluated by comparison to the available measured post-irradiation data for M5TM fuel rods and the applicable acceptance criteria. Criterion 2 as discussed in the response to above was satisfied when the cladding creep strains calculations for the M5TM fuel rod cases were found to be within a range of (( )) of applicable measured data.

Page 19 of 66

RAI #9

9. Sections 4.3.3.3.1 through 4.3.3.3.3 describes fuel rod behavior comparisons for fuel temperature, power-to-melt (PTM) and internal hot gas pressure. Provide details of the methodology, computational procedure and computer codes used in the development of Figures 4.3.4 through 4.3.8

RESPONSE

The License Amendment Request (LAR) (Enclosure 2, Section 4.3.3.3) compares AREVA UO2 fuel rod behavior when modeled with ((

)).

Output for each of the three cases includes fuel average temperature as shown in Figure 4.3.6, power-to-centerline melt (PTM) as shown in Figure 4.3.7, and rod internal pressure as shown in Figure 4.3.8.

The calculation for fuel average temperature uses the FATES3B w/M5TM code. This code calculates fuel average temperature using the same methodology and computational procedure described in Topical Report CENPD-139-P-A as improved by Topical Report CEN-161 (B)-P-A Supplement I-P-A, with additional methodology clarifications provided in Topical Report CEN- 193(B)-P Supplement 2-P.

The calculation for PTM uses the FATES3B w/M5TM code. This code calculates fuel average temperature using the same methodology and computational procedure described in Topical Report CENPD-139-P-A as improved by Topical Report CEN-161(B)-P-A Supplement 1-P-A, with additional methodology clarifications provided in Topical Report CEN-193(B)-P Supplement 2-P.

The calculation for rod internal pressure uses the FATES3B w/M5TM code. This code calculates fuel average temperature using the same methodology and computational procedure described in Topical Report CENPD-139-P-A as improved by Topical Report CEN-161 (B)-P-A Supplement 1-P-A, with additional methodology clarifications provided in Topical Report CEN-193(B)-P Supplement 2-P.

Page 20 of 66

RAI #10

10. Thermal conductivity of U0 2 fuel deteriorates with fuel burnup. Has the thermal conductivity degradation (TCD) with burnup been accounted for in the fuel temperature, power-to-centerline melt and the internal hot gas pressure calculations? If the licensee has used a code that has no explicit treatment of TCD with burnup, explain what compensatory measures were implemented for TCD in the calculations.

RESPONSE

Please see response to RAI #10 (Set #1) as submitted by SCE letter dated September 14, 2012.

Page 21 of 66

RAI #11

11. In Section 4.3.4 of Enclosure 2 of LAR, it is stated that the "hot rod" is a composite rod and represents several different rods at a given burnup and may represent many different rods during the course of burnup from the beginning of life (BOL) to end of life (EOL)." Provide a detailed explanation for this statement. (Details on composite rod, several different rods at a given burnup, many different rods during the course of burnup from the beginning of life (BOL) to end of life (EOL), etc.)

RESPONSE

CEN-193(B)-P Supplement 2-P contains RAI responses supporting the CE submittal CEN-161 (B)-P, "Improvements to the Fuel Evaluation Model," regarding changes to the FATES computer code. This topical has been approved by the staff and issued as CEN- 161 (B)-P-A.

Response to RAI question 3A in CEN-193(B)-P Supplement 2-P states, ((

Page 22 of 66

1]

This methodology is the same used for Westinghouse fuel and remains unchanged for AREVA fuel.

Page 23 of 66

RAI #12

12. Section 4.4.3 Application of AREVA Burnup Limits for M5 fuel SONGS indicates M

T that "The addition of gadolinia will impact the thermal conductivity and melt temperature, but only incrementally from BOL and is not burnup dependent." Please provide justification based on technical analysis.

RESPONSE

The technical analysis of the impact of addition of gadolinia on thermal conductivity is addressed in Topical Report CENPD-275-P, Revision l-P, Supplement 1-P-A, Section 2.2.5. This topical report (( )). Thermal conductivity degradation as a function of burnup is addressed in the response to RAI Set #1 Question #10 as submitted by SCE letter dated September 14, 2012.

LAR Enclosure 2, Section 4.10, provides a discussion related to Safety Limit 2.1.1.2 which specifies fuel centerline temperature variation with burnup and its adjustment for the presence of burnable poison. The adjustment for gadolinia burnable poison is a function of weight percent poison. The adjustment for gadolinia burnable poison is independent of burnup. There is no additional burnup dependency on melt temperature beyond that discussed in Enclosure 2, Section 4.10 for U0 2 ((

Page 24 of 66

RAI #13

13. Provide details of the utility code, INTEG and how this code is used to calculate the results listed in Tables 4.5.1 and 4.5.2.

RESPONSE

Details on the [NTEG utility code are provided in Reference 8.24 of Enclosure 2 to the LAR. In particular, the response to Question 3 in Appendix A of Reference 8.24 describes how INTEG is used for the mechanistic assessment of clad ballooning.

The LAR Enclosure 2 Tables 4.5.1 and 4.5.2 present the results of the use of INTEG to replicate a series of the EDGAR tests on pressurized M5TM clad specimens of the same outer radius and wall thickness as used in SONGS Units 2 and 3. M5TM clad specimens at various initial clad temperatures and pressures were subjected to various uniform clad temperature ramp rates to determine time of clad rupture and the clad rupture pressure. The

[R 1] EDGAR series of tests were selected for replication using INTEG since the duration of these tests exceeded the maximum expected time in DNB ((

)) noted in LAR Reference 8.24 (CEN-372-P-A, "Fuel Rod Maximum Allowable Gas Pressure," May 1990). Ability to replicate the EDGAR tests justifies the use of INTEG for (( )) since both involve a time-dependent strain resulting from differential pressure and changing clad temperature.

Some of the parameters used by the INTEG code are not available from the EDGAR test documents. Consequently, the INTEG replication analysis used the existing Zircaloy-4 properties in the code model. This is justified a posteriori given the good agreement between the INTEG results and the EDGAR tests.

Page 25 of 66

RAI #14

14. Section 4.5.4.1 describes licensee's analysis for the deposited energy (fuel pin enthalpy) acceptance criteria. NRC Information Notice 2009-23 dated October 8, 2009, notified licensees of nuclear power reactors of the degradation of fuel thermal conductivity of uranium fuel pellets with increasing burnup (TCD). The significance of this effect was not included in the fuel thermal-mechanical performance codes approved prior to 1999.

The licensee is using a legacy code, STRIKIN-Il, for heat transfer treatment of U0 2 fuel in its CEA Ejection analysis to calculate the deposited energy in the fuel. Thermal conductivity of U0 2 fuel deteriorates with burnup, and as such, each fuel vendor must have an explicit model to generate burnup dependent fuel thermal conductivity in their analyses to simulate transients and accidents.

Explain how the impact of TCD with burnup on the non-LOCA transients and postulated accidents analyses have been implemented in, specifically but not limited to, the spectrum of control rod ejection accident analyses.

RESPONSE

The impact of TCD with burnup on the analyses of non-LOCA transients and postulated accidents was addressed in the responses to RAI# 10 (Set #1) as submitted by SCE letter dated September 14, 2012.

The CEA Ejection event is analyzed to verify the acceptance criteria of fuel rod centerline deposited energy and radial average fuel rod deposited energy. In the STRIKIN-I1 code, the fuel rod heat transfer model includes the effects of the fuel pellet thermal conductivity and gap conductance.

The fuel performance data used in STRIKIN is generated from the FATES3B analysis (see RAI

1. 4). The value of gap conductance is conservatively determined to challenge the acceptance criteria. For the average rod model in STRIKIN, the goal is to maximize the radial average deposited energy. In this case the maximum gap conductance is used to allow fast heat transfer from the rod, thus reducing the Doppler feedback effect and maximizing the core power rise.

TCD reduces the fuel pellet thermal conductivity and causes a slower heat transfer from the rod.

It is therefore conservative to not consider the effect of TCD for the radial average deposited energy criterion.

For the hot rod model in STRIKIN, the goal is to maximize the rod centerline deposited energy.

In this case the minimum gap conductance is used to minimize the heat transfer from the rod.

With the minimum gap conductance, the fuel rod heat transfer is dominated by the pellet-clad gap thermal resistance. In addition, the minimum gap conductance is selected from the plateau of the relative power density fall-off curve at a burnup typically less than 10 GWD/MTU. At this low bumup the TCD effect is negligible. Based on the discussion above, it is acceptable to not consider the effect of TCD in the hot rod calculation in CEA Ejection analysis.

Page 26 of 66

Other key parameters used in the STRIK[N analysis include the maximum ejected CEA worth, maximum radial power peaking factor Fr, least negative Doppler reactivity curve, and minimum delayed neutron coefficients. These parameters are conservatively selected to assure that margin is maintained in the CEA Ejection analysis.

Page 27 of 66

RAI #15

15. In Section 5.1.2 of Enclosure 2 of the LAR, the licensee states that the reload assemblies planned for San Onofre Unit 3 have 2 component changes to improve seismic margin. Describe AREVA fuel assembly structural response to seismic and LOCA loads. The details should include the methodology of analysis, uncertainty allowances and combination of loads from natural phenomena and accident conditions (i.e., Seismic and LOCA).

RESPONSE

The methodology of the lateral seismic and LOCA analyses (Reference 2) is in compliance with BAW-10133PA, Revision-i and its Addenda 1 and 2. Beyond the methodology defined above, additional measures have been taken to address the issues identified in NRC Information Notice 2012-09: Irradiation Effects on Fuel Assembly Spacer Grid Crush Strength. The irradiation effects discussed in this Information Notice fall into two categories: (1) spacer grid strength and (2) fuel assembly stiffness.

With respect to spacer grid strength, AREVA has determined that the primary effects of irradiation are due to changes in irradiation hardening of the grid material and* relaxation of the strip faces holding the fuel rods. Testing was performed with solid pins to simulate the irradiated condition of the fuel rod in which solid contact between the pellet and cladding is known to exist.

These effects were simulated in spacer grid testing to determine irradiated spacer grid properties, including strength. This testing was performed in addition to normal "beginning-of-life" (BOL) spacer grid testing. Both BOL and simulated "end-of-life" (EOL) spacer grid characteristics were considered in seismic and LOCA analyses for SONGS.

With respect to fuel assembly stiffness, AREVA has determined that the primary effect of irradiation is due to relaxation of the grip forces holding the fuel rods. This effect is simulated by building and testing a fuel assembly constructed with pre-relaxed spacer grids. A design specific test fuel assembly was constructed in this condition for the San Onofre application. This testing was performed in addition to normal BOL fuel assembly testing. Both BOL and irradiated fuel assembly characteristics were considered in seismic and LOCA analyses for SONGS.

It should be noted that while Addendum I of BAW-10133PA, Revision 1 defines an allowable spacer grid load based on the point of instability (i.e., buckling) of the spacer grid, this definition is conservatively modified for the application of HTP spacers at SONGS. In the beginning-of-life condition, the allowable load for the HTP spacer is based on that load which results in a uniform permanent deformation of 1mm, which is well below the point of instability for the grid.

A deformation of 1mm, uniformly distributed across the spacer grid is not significant enough to challenge the underlying requirements for faulted conditions (i.e., coolability and control rod insertability).

Page 28 of 66

Uncertainty Allowances In accordance with the Standard Review Plan 4.2, Appendix A, Section 11.3, a sensitivity study was performed for the case which provided the highest grid loading condition for both the OBE and SSE loading conditions. It was verified that the grid impact loads obtained from the forcing functions shifted by ((

))of the nominal case. Therefore no factor was needed to be applied to the calculated grid impact loads. Hence, these studies satisfied the guidance of the NUREG-0800, Standard Review Plan 4.2, Appendix A, Section 11.3 on the uncertainty allowances sensitivity study.

Combination of Loads With respect to the combination of loads from natural phenomena and accident conditions (i.e.,

seismic and LOCA), this was performed in accordance with BAW-10133PA, Revision 1, Addendum 1 and the NRC guidance provided in NUREG-0800, Standard Review Plan 4.2, Appendix A. In accordance with these methods, the seismic and LOCA cases are analyzed independently. The bounding impact load from all seismic cases is then combined with the bounding load from all LOCA cases by means of the square root of the sum of squares (SRSS),

regardless of time of occurrence.

Page 29 of 66

RAI #16

16. Provide details of the evaluation of SONGS AREVA fuel assembly structural response to externally applied forces (seismic and LOCA) and show how the acceptance criteria per Standard Review Plan (SRP, NUREG-0800, Chapter 4.2, Appendix A,Section IV) is satisfied.

RESPONSE

NUREG-0800, Standard Review Plan 4.2, Appendix A discusses two primary criteria that apply for LOCA: (1) fuel rod fragmentation must not occur as a direct result of the blowdown loads, and (2) the 10 CFR 50.46 temperature and oxidation limits must not be exceeded. A third criterion of control rod insertability must also be satisfied for combined LOCA plus seismic loads. These three criteria are discussed below:

Fuel Rod Fragmentation For fuel rod cladding, the stress categories and allowed stresses intensities were per the M5TM topical report. The stress intensity criteria provided in these M5TM topical reports are more conservative than the stress intensity limits defined in ASME B&PV Code, Section Ill, Appendix F. Thus, the analyses performed with the allowed stress intensity criteria for M5TM fuel rod cladding ensure that the fuel rod fragmentation will not occur.

10 CFR 50.46 Temperature and Oxidation Limits and Control Rod Insertability As noted above, the amount of permanent deformation corresponding to the load limit for the HTP spacer grid is not significant enough to challenge the coolability of the fuel assembly or control rod insertability. This is consistent with the guidance provided in NUREG-0800, Standard Review Plan 4.2, Appendix A, in which it is stated that if the combined grid loads (i.e.,

seismic plus LOCA) remain below P (crit) (i.e., allowable grid load) then significant deformation of the fuel assembly would not occur and thereby not interfere with control rod insertion or impede fuel assembly coolability.

Page 30 of 66

RAI #17

17. Provide a detailed summary of thermal and hydraulic compatibility analysis to show that the AREVA 16 x 16 fuels is compatible with the coresident CE 16 x1 6 fuel. The analysis should demonstrate that the hydraulic and thermal margin performance of the core will not significantly be impacted by the introduction of the new AREVA 16x16 fuel assemblies in to the SONGS core. Hydraulic compatibility should include the assessment of the impact of AREVA fuel on core flow distribution. Thermal compatibility analysis should evaluate the impact on DNBR calculations due to the introduction of AREVA fuel.

(Note: Section 5.1.3 of Enclosure 2 says, "Overall, the mechanical compatibility evaluations performed for the LFA program, and the evaluations of the changes planned for the reload AREVA fuel have confirmed that the AREVA fuel assemblies are compatible with the SONGS reactor components and the co-resident fuel in the SONGS core." The NRC staff notes that the compatibility evaluations for LFAs are not sufficient for the reload core since the LFAs were limited in number and they were approved for non-limiting locations in the SONGS core. Therefore for the reload core, a new compatibility analysis is needed.)

RESPONSE

Thermal and hydraulic compatibility analyses were performed by both Westinghouse and AREVA for the AREVA CE-HTP Lead Fuel Assemblies (LFAs).

These analyses considered bounding conditions, not just the actual LFA locations in the core design. The two minor changes being implemented for SONGS-3 Cycle 17 AREVA fuel are not significant changes for thermal and hydraulic compatibility analyses. As discussed in the LAR Section 5.1, the LFA compatibility analyses were reexamined in light of the minor design changes, and the fuel with the minor design changes remains compatible with the host reactor core internals, handling equipment, storage racks, and co-resident fuel.

The impact of the core inlet flow distribution is determined by the explicit modeling of each assembly in the TORC computer code. This impact assessment is performed for each core design which uses different fuel assembly types. DNBR calculation is performed for the specific DNBR limiting assembly, using the appropriate licensed CHF correlation for that fuel type. This process is demonstrated in detail in the LAR Enclosure 2 Section 7.2.

The mechanical compatibility analyses address the interface between an AREVA assembly and the upper and lower core support plates, the CEAs, the handling equipment, and the elevations (e.g. spacer grids, fuel column, end fittings, etc.) and envelopes in relationship to the co-resident fuel. These evaluations are not affected by the number of assemblies and will continue to be applicable for the reload core.

Page 31 of 66

RAI #18

18. Table 5-1 of Enclosure 2 lists reload fuel assembly nominal mechanical design features and compare them with the LFA mechanical features. Please provide a similar table with the mechanical design features of the resident fuel (CE 16x1 6) along with the new AREVA fuel. Table should include details of the relative axial positions of spacer grids and straps, upper and lower tie plates, and the top and bottom of active fuel length of the two fuel designs.

RESPONSE: The AREVA CE-HTP fuel for SONGS is designed to closely match the specifications for the existing CE 16 fuel from Westinghouse. Spacer grid locations are matched to the same centerline. Top and bottom of the active fuel length is identical. A comparison of the fuel designs is provided below: ((

t +/-

t -I- I-

-I- -I-4 +

I 4. 4.

.4 I.

.4 I. I.

Page 32 of 66

A mechanical compatibility analysis was performed by AREVA for the lead fuel assemblies at SONGS. These results remain valid for the VQP fuel design. The conclusions of this analysis are summarized below:

The AREVA fuel assembly design is mechanically compatible with co-resident fuel design at SONGS. The envelopes of the components of the AREVA design are bounded by the envelopes of the coresident fuel. The fuel column lengths and elevations are comparable, the spacer grid positions align axially and maintain overlap throughout the design life, and the elevations of the upper and lower tie plates are comparable.

Page 33 of 66

RAI #19

19. Please explain why a mixed core characterization evaluation was not performed.

RESPONSE

The mixed core impact has been examined for both the mechanical aspect (LAR Section 5.1),

and also for the thermal-hydraulic reload analysis impact (LAR Section 7.2). See the response to RAI #17 in this document (Set 2) for a description of these analyses.

The response to RAI #17 in this document (Set 2) discusses the physical and mechanical similarity between the standard CE16 design at SONGS, and the CE-HTP Lead Fuel Assemblies and the CE-HTP design planned for SONGS-3 Cycle 17. The CE-HTP is designed to closely match the dimensions of the standard CE16 design in order to be a direct replacement option.

For the thermal-hydraulic aspect, the pressure drop differences in the CE 16 fuel and the CE-HTP are driven by the grids (HID-1L and HID-2L for CE16, and HTP for CE-HTP) and the lower end fitting (GUARDIAN for CE16 and FUELGUARD for CE-HTP).

For the grids, the AREVA CE-HTP grid pressure drop ((

)). The differences in the pressure drop in the lower end fitting are analyzed in the Inlet Flow Distribution calculation.

Page 34 of 66

RAI #20

20. Table 5-2 lists generic mechanical design criteria for AREVA fuel rod and fuel assembly designs. In addition to the listed criteria, the licensee should add the following criteria under the listed categories in accordance with SRP (NUREG-0800)

Section 4.2:

Fuel Rod Failure Criteria

• Overheating of cladding

• Overheating of fuel pellets

• Excessive fuel enthalpy

• Fuel failure due to bursting Fuel System Criteria

• Fatigue

• Fretting wear

• Rod bow Fuel Coolability

• Cladding embrittlement

• Violent expulsion of fuel

• Fuel ballooning

RESPONSE

Table 5-2 addresses the generic mechanical design criteria addressed by AREVA in their scope of work. Other design criteria are addressed under the existing SCE methodology (SCE-9801-P-A) as discussed below. San Onofre's Licensing Basis is NUREG-75/087, not NUREG-0800.

Fuel Rod Failure Criteria Overheating of cladding Per the CPC Setpoints methodology, as described in SCE-9801-P-A Sections 3.4 through 3.5, accidents in which the overheating of cladding is a concern are evaluated against acceptance criteria described in SCE-9801-P-A as a normal part of the reload process in accordance with SCE methodology. Based on these results, the CPC Setpoints are generated and utilized in the plant. Incorporation of the BHTP heat flux correlation for use with AREVA fuel is added as discussed in LAR, Enclosure 2, Section 4.2. Supporting thennal-hydraulic calculations demonstrating the acceptable cooling of the cladding are also discussed in LAR Enclosure 2, Section 4.2.

Page 35 of 66

• Overheating of fuel pellets Per the CPC Setpoints methodology, as described in SCE-9801-P-A Sections 3.4 through 3.5, accidents in which the overheating of fuel pellets is a concern are evaluated against acceptance criteria described in SCE-9801-P-A as a normal part of the reload process in accordance with SCE methodology. Based on these results, the CPC Setpoints are generated and utilized in the plant. There is no change to this methodology. A discussion on the fuel rod performance is in the LAR Enclosure 2, Section 4.3. Supporting fuel performance calculations demonstrating the acceptable behavior of the fuel are also discussed in the LAR Enclosure 2, Section 4.3.

" Excessive fuel enthalpy Per the CPC Setpoints methodology, as described in SCE-9801-P-A Sections 3.4 through 3.5, accidents in which excessive fuel enthalpy is a concern are evaluated against acceptance criteria described in SCE-9801-P-A as a normal part of the reload process in accordance with SCE methodology. Based on these results, the CPC Setpoints are generated and utilized in the plant. Incorporation of M5TM properties is discussed in LAR, Enclosure 2, Section 4.3. A discussion of the CEA Ejection event is provided in LAR, Enclosure 2, Section 4.5.4.

" Fuel failure due to bursting Per the CPC Setpoints methodology, as described in SCE-9801-P-A Sections 3.4 through 3.5, accidents in which fuel failure due to bursting is a concern are evaluated against acceptance criteria described in SCE-9801-P-A as a normal part of the reload process in accordance with SCE methodology. Based on these results, the CPC Setpoints are generated and utilized in the plant. Additionally, fuel rod burst criteria are addressed in the lead fuel assembly design analysis performed by AREVA (LAR Reference 8.3, ANP-2839(P)) for LOCA considerations.

Fuel System Criteria

" Fatigue Fatigue is evaluated as discussed in item 3.3.2 in Table 5-2 of Enclosure 2 to the LAR.

• Fretting wear Fretting wear is evaluated as discussed in item 3.3.3 in Table 5-2 of Enclosure 2 to the LAR.

" Rod bow The mechanical aspects of rod bow are covered in the vendor's mechanical design analysis described in Section 5.1 of Enclosure 2 of the LAR. The Rod Bow penalty is analyzed through the thennal-hydraulic analysis and verified to satisfy the Technical Specification DNBR limit. Rod Bow is discussed in Section 4.2.2 of Enclosure 2 of the LAR. RAI #3 in this document also discusses Rod Bow penalty.

Page 36 of 66

Fuel Coolability

• Cladding embrittlement Cladding embrittlement is evaluated using the fuel vendor LOCA methodology (see Section 4.6 of Enclosure 2 to the LAR).

" Violent expulsion of fuel Per the CPC Setpoints methodology, as described in SCE-9801-P-A Sections 3.4 through 3.5, accidents in which the violent expulsion of fuel is a concern are evaluated against acceptance criteria described in SCE-980 1-P-A as a normal part of the reload process in accordance with SCE methodology. Based on these results, the CPC Setpoints are generated and utilized in the plant. Incorporation of M5TM properties is discussed in LAR, Enclosure 2, Section 4.3. A discussion of the CEA ejection event is provided in LAR, Enclosure 2, Section 4.5.4.

" Fuel ballooning Fuel ballooning is evaluated using the fuel vendor LOCA methodology (see Section 4.6 of Enclosure 2 to the LAR). Additionally, fuel rod burst criteria are addressed in the lead fuel assembly design analysis performed by AREVA (LAR Reference 8.3, ANP-2839(P)) for LOCA considerations.

Page 37 of 66

RAI #21

21. Briefly describe the methodology and computer codes that were used in generating key physics parameters (Table 7.1.1, Table 7.1.2 and Table 7.1.3)

RESPONSE

As discussed in the LAR Section 7.1, the physics parameters documented in Tables 7.1.1, 7.1.2, and 7.1.3 were generated following SCE's approved reload analysis methodology provided in Reference 8.4 (SCE Reload Analysis Methodology, Sections 3.1 and 4.1). The data were generated using Studsvik Scandpower computer codes CASMO-4 and SIMULATE-3. The approved methodology incorporating CASMO-4/SIMULATE-3 into the SCE reload design process is documented in Reference 8.5 (PWR Reactor Physics Methodology Using Studsvik Design Codes) (See LAR Section 4.1).

Page 38 of 66

RAI #22

22. Since the maximum Unit 3 Cycle 18 fuel enrichment is planned for 4.95 w/o which is greater than the maximum approved enrichment for spent fuel criticality of 4.8 w/o, is the licensee planning to submit license amendment request for new spent fuel criticality analysis?

RESPONSE

The 4.95 w/o fuel enrichment for the proposed Unit 3 Cycle 18 core design was chosen to exercise the SONGS reload analysis process using a bounding maximum enrichment. Actual core designs that are implemented will conform to SONGS licensing bases.

Page 39 of 66

RAI #23

23. Provide definitions for 'core flow factors' that appear in Figures 7.2.4 through 7.2.7.

RESPONSE

The core flow factors shown in the LAR Figures 7.2.4 through 7.2.7 represent the fraction of the core flow seen by each assembly at the core inlet. The inlet flow factors are normalized to 1.0 (i.e., they average out to 1.0). It serves as a boundary condition for TORC, when TORC is performing DNBR calculation for each assembly (SCE-9801-P-A Section 3.2.1.1). See CENPD-206-P-A, "TORC Code Verification and Simplified Modeling Methods" Section 3.1 for a discussion of inlet flow factors and TORC modeling.

The core flow factors for a mixed-core have already accounted for the differences in the hydraulic properties of neighboring assemblies. Having a core flow inlet distribution (i.e,,

assembly-specific flow factor) along with the assembly-specific power distribution allows the TORC code to determine the limiting location (e.g., the limiting assembly) in the core with respect to DNBR. This is the basis of why a mixed-core penalty is not needed for SONGS. The TORC model used for DNBR calculations models each individual assembly, and DNBR is determined for each assembly based on assembly-specific properties (e.g., assembly flow and power).

Page 40 of 66

RAI #24

24. NRC staff's safety evaluation report (SER) for Exxon Nuclear Company (Currently AREVA NP) Topical Report, XN-NF-82-21(P)(A) requires that "an adjustment of

(( )) on minimum DNBR must be included for mixed cores containing hydraulically different fuel assemblies." Section 7.2.2.1 of Enclosure 2 of LAR states that since "the SCE method explicitly models the mixed core configuration to determine the flow impacts; there is no need for an additional mixed core penalty." The licensee's position of 'no mixed core penalty' contradicts with the fuel vendor's position of

(( )) mixed core penalty. Provide clarification for this position.

RESPONSE

The (( )) penalty enacted by the SER to XN-NF-82-21 (P)(A) was based on sensitivity studies to address "mixed core reloads where the geometric differences between fuel assemblies are greater than those found in Prairie Island". Since XN-NF-82-21 (P)(A) developed the mixed core methodology, the (( )) penalty was enacted to bound future mixed core analyses which did not fall under the mixed core analyses performed for XN-NF-82-21 (P)(A). SCE has specifically addressed the mixed core impact in the LAR.

This LAR is not subject to a (( )) mixed core penalty, because the mixed core impact has been specifically analyzed for the CE16 and CE-HTP fuel at SONGS. The LAR is not a mixed core methodology submittal.

Table 3.1 of XN-NF-82-21 (P)(A) shows the difference in loss coefficients between the Prairie Island Exxon and Westinghouse assemblies of ((

As part of the compatibility analyses performed for the LAR, AREVA calculated loss coefficients by the same methods for CE16 and CE-HTP fuel using flow tests of both assemblies.

The resultant loss coefficients are used to calculate a total assembly loss coefficient to compare to the (( )) difference in XN-NF-82-21(P)(A).

Page 41 of 66

Er The loss coefficient difference between the CE16 fuel and the CE-HTP fuel at SONGS at the core average Reynolds' Number of 472,000 is less than (( )). This shows that there are no significant hydraulic differences between the CE 16 and CE-HTP fuel assemblies.

Second, the analyses performed in support of XN-NF-82-21 (P)(A) were based on ((

)). This approach cannot accurately predict crossflow impact between adjacent assemblies of different designs.

The current thermal-hydraulic approach at SONGS utilizes a quarter-core full assembly model where no lumped channels are used. As a result, each individual assembly is modeled with appropriate grid loss coefficients for the specific grid design. This modeling scheme can predict crossflow between assemblies much more accurately than a lumped channel model.

Additionally, a (( )) TORC code uncertainty penalty has already been built into the SAFDL DNBR Limit through the Modified Statistical Combination of Uncertainties process at SONGS (SCE-9801-P-A, Section 3.2.1.1).

The above shows that the (( )) penalty applied in XN-NF-82-21(P)(A) should not apply to SONGS due to the hydraulic similarity of the CE 16 and CE-HTP designs, and since the mixed core is explicitly modeled in SCE's thermal-hydraulic analyses.

Page 42 of 66

RAI #25 25.A notification of discovery was issued and presented to the NRC staff by AREVA dated November 15, 2011, regarding (( )) defined on page 2-3 of Section 2.2 Fuel Design Factorin BAW-1 0241 (P)(A), Revision 1, BHTP DNB CorrelationApplied with LYNXT. The CE16 fuel design ((

)). AREVA initiated a condition report (CR) (CR2011-8304) and assessed the impact of the (( )) and ((

)) resulting in a small CHF penalty. Provide details of this penalty on CHF due to the (( )).

RESPONSE

The wetted perimeter ratio (WPR) term is a term that feeds into the CHF correlation and it

[I .)) AREVA's analysis of the CHF correlation M/P data demonstrated that the use of a (( )) provides a conservative prediction of CHF. The analysis concluded that for SONGS, use of a WPR up to

(( )) provides a conservative CHF prediction and is considered acceptable.

This means that for a ((

)). This results in a conservative prediction of CHF.

Page 43 of 66

RAI #26

26. Section 7.2.3 Thermal-Hydraulic Tuning and Benchmarking of Enclosure 2 describes CETOP-D code benchmarking analysis "to determine the ((

)) Provide (a) details of the adjustment factors, (b) the basis for the adjustment factors, and (c) how they are applied in the CETOP-D benchmarking analysis.

RESPONSE

This question is best suited to be answered in the following order:

b) the basis for the adjustment factors -

The basis for adjustment factor lies in the requirement that CETOP-D code must calculate a smaller (i.e., more conservative) DNBR than the TORC code for the same set of input parameters. The CETOP-D code is used in the majority of transient analyses, setpoint analyses and is the basis of the models installed in COLSS and CPCS at the plant.

a) details of the adjustment factors -

Details of the overall thermal-hydraulic analysis process at SONGS are provided in SCE-9801-P-A Section 3.2. The process discussed below remains unchanged for the addition of AREVA fuel.

The starting point for evaluating CETOP-D conservatism with respect to TORC is the CETOP-D model (i.e., input deck). The heat flux adjustment factors are called the overpower penalties on CETOP-D. They are calculated as follows: a multitude of TORC and CETOP-D runs are performed over the entire operating condition range at various reference DNBRs.

The heat flux for these TORC and CETOP-D runs are iterated upon until the calculated DNBR converges to the reference DNBR value. In this manner, the difference in the calculated DNBR between TORC and CETOP-D has been converted to the difference in the iterated heat flux. A larger iterated heat flux for CETOP-D than the iterated heat flux for TORC code indicates that CETOP-D is non-conservative with respect to TORC since it took more heat to reach the same DNBR as TORC given the rest of input were the same.

Conversely, the smaller heat flux for CETOP-D than TORC indicates CETOP-D conservatism with respect to TORC and no adjustment is needed. To quantify the CETOP-D non-conservatism, the ratio of TORC heat flux to CETOP-D heat flux is calculated for all the cases considered in the analysis. The smallest (i.e., most conservative) ratio is selected and transmitted as the overpower penalty on CETOP-D code. This maximum conservative overpower penalty is applied to all thermal hydraulic conditions in non-LOCA transients and COLSS/CPC setpoints.

Page 44 of 66

c) how they are applied in the CETOP-D benchmarking analysis -

The overpower penalties (OPP) are calculated by the CETOP-D benchmarking analysis. They are applied in the downstream analyses which use the CETOP-D code to calculate DNBR. The application of OPP is as follows: for a given CETOP-D calculation, the input heat flux is divided by the OPP. For an OPP value less than one, the effect of application of OPP to the heat flux is analogous to raising the inputted heat flux, thus decreasing the calculated DNBR. As mentioned above, a CETOP-D OPP is calculated based on a condition where the CETOP-D was most non-conservative with respect to TORC. For most downstream CETOP-D cases, the operating condition is not the one that led to calculation of the OPP. This is an inherent conservatism in the SONGS methodology in application of the OPP. Also, it should be noted that the OPP greater than 1.0 indicates that CETOP-D was conservative to TORC at all conditions. However, it has not been SONGS practice to credit the use of an OPP greater than 1.0.

Note that the discussion above is on the existing Thermal-Hydraulic methodology from SCE-9801-P-A. No change to this methodology is being implemented in the LAR. The only changes are the process and code changes required to implement an ((

Page 45 of 66

RAI #27

27. Section 7.4.1.2, Pre Trip MSLB Event DNBR Propagationcontains a sentence that reads "The clad strain for this event is well below (( )) clad strain limit described in Section 4.4.1." The licensee should correct this statement to read "The clad strain for this event is well below (( )) clad strain limit described in Section 4.5.1."

RESPONSE

This typo has been corrected. A replacement page is provided on the following page.

Page 46 of 66

LICENSEE'S EVALUATION SONGS PCN-600 Request for Unrestricted Use Of AREVA Fuel Table 7.4.3 1[

4 4

)) Therefore, DNB propagation will not occur for M 5 TM fuel during this event and coolable geometry will continue to be maintained.

7.4.1.3 Pre-Trip MSLB Event Analysis Dose Results The Pre-Trip MSLB analysis is also performed to determine the dose consequences as a result of the implementation of AREVA fuel. For the VQP cycle 17 core described in Table 7.21.1, the Pre-Trip MSLB event predicted fuel failure values with ((

II]

Page 152 of 166 Page 47 of 66 Enclosure 2 to SONGS PCN-600

RAI #28

28. Please provide a summary of the analysis where the licensee determined that ((

1].

RESPONSE

An Increased Main Steam Flow with Loss of AC Power (IMSF+LOAC) event was analyzed to compare the fuel failure impact resulting from using the (( )) correlation with that from the CE-I correlation. The "slow-trip" case was run as it provides a degree of fuel failure rates and would be sensitive to changes in the (( )) used. Each analysis used the identical NSSS transient input to the respective fuel failure. The CE-1 correlation was applied using the CETOP-D code conservatively benchmarked to the TORC code so as to give conservative DNBR results relative to what the TORC code would provide. This is in accordance with SONGS NRC-approved reload methodology. The ((

)) to obtain DNBR results.

The DNBR results obtained above were then compiled into (( )) needed for fuel failure determination per the SONGS NRC-approved reload methodology. The convolution method was used in each fuel failure analysis. The results for each are as follows:

This indicates that a CETOP-D model, tuned by a CETOP-D/TORC benchmarking process where the (( )), is conservative for use with AREVA fuel.

Page 48 of 66

RAI #29

29. Please provide details of the methodology and computer codes used for the pre-trip MSLB analyses described in Section 7.4.1 of Enclosure 2.

RESPONSE

The pre-trip MSLB analysis described in Section 7.4.1 of Enclosure 2 is consistent with the SONGS Reload Analysis Methodology (SCE-9801-P-A) Section 3.4.2.1.2 and SONGS 2/3 UFSAR Section 15.10.1.3.1. The objective of the analysis is to demonstrate that the dose consequences for this event do not exceed the acceptance criteria based on the alternative source term (AST) guidance of Regulatory Guide 1.183. If the acceptance criteria are violated, the dose consequences are reduced by either providing a COLSS required overpower margin (ROPM) and/or a CPCS power penalty to reduce the event-initiated fuel failure and consequently to reduce the dose consequences.

Conservative assumptions are used in the analysis to maximize potential for degradation in fuel performance. The initial conditions and event initiators are chosen to result in the most adverse pre-trip power excursion and fuel performance degradation. The least negative Doppler temperature coefficient is used to minimize the amount of negative reactivity feedback during the power excursion. Reactor protection trip setpoints are conservatively modeled to account for uncertainties which result in conservative response time.

The RPS trips available to mitigate this event are the low steam generator pressure (LSGP) and low level (LSGL) trips, the CPC trips and variable overpower (VOPT) trips. The insertion of the CEAs upon reactor trips terminates the power excursion. Loss of off-site AC power (LOAC) is assumed to occur during the MSLB event, which results in a coast down of the reactor coolant pumps (RCPs).

The Nuclear Steam Supply System (NSSS) response to the pre-trip MSLB event is simulated using the CENTS computer code. The transient response from CENTS is then analyzed using the CETOP-D code to calculate ROPM. The most limiting ROPM data will be implemented in COLSS to maintain initial margin to DNBR SAFDL and be used to evaluate fuel failure. The NRC approved the use of the statistical convolution technique to determine fuel failure for events that assume a loss of flow and that fail fuel. Therefore, the statistical convolution technique may be used in fuel failure evaluation for this event.

Page 49 of 66

RAI #30

30. Please provide details of methodology, computer codes used, and the analysis procedures for the CEA Ejection Analysis for ((

)) that are presented in Section 7.4.2 of Enclosure 2.

RESPONSE

The CEA Ejection analysis described in Section 7.4.2 of Enclosure 2 is consistent with the SONGS Reload Analysis Methodology (SCE-9801-P-A) Section 3.4.2.1.4 and SONGS 2/3 UFSAR Section 15.10.4.3.2. Conservative assumptions are used in the CEA Ejection analysis.

The most conservative ejected rod worth as a function of power level and power dependent insertion limits (PDIL) is used as well as the most conservative radial power peaking factor.

Conservative, normalized pin census are used for fuel failure calculations. Reactor shutdown is initiated by the high power level trip or the CPC variable overpower trip (VOPT).

The STRIKIN computer code is used to simulate the fuel pin response during the event (see RAI

1. 14). The energy deposited in the fuel rod by an ejected CEA at various plant conditions and power levels are calculated. Both HFP and HZP cases are analyzed to demonstrate that the worst case CEA ejection accident will not cause limits on coolable geometry to be exceeded.

The analysis process is the same for Westinghouse fuel with ZIRLO fuel cladding and AREVA fuel with M5 M fuel cladding. For AREVA fuel, two input changes are required. ((

Conservative criteria are used to determine the amount of fuel failure. For CEA Ejection, all fuel pins that experience DNB are assumed to fail. All fuel pins with greater than 250 cal/gm total centerline enthalpy are assumed to fail. All fuel pins with greater than 200 cal/gm average enthalpy of the hottest fuel pellet are assumed to fail. The dose consequences are evaluated subsequently to demonstrate that the acceptance criteria based on the alternative source term (AST) guidance of Regulatory Guide 1.183 are met. These criteria are unchanged from the approved criteria for SONGS as described in SCE-9801-P-A and SONGS 2/3 UFSAR Section 15.10.4.3.2. They are applied to both Westinghouse and AREVA fuel.

Page 50 of 66

RAI #31

31. Please provide details of the SCE procedure, (e.g., why and how), the following events have been dispositioned for the SONGS reload analyses during the fuel transition. (Attachment C, Enclosure 2)

(a) Increase in main steam flow (IMSF)

(b) Single reactor coolant pump shaft seizure (c) Single reactor coolant pump sheared shaft (RCPSS)

(d) Uncontrolled CEA withdrawal (CEAW) from a subcritical or low power condition (e) Uncontrolled CEA withdrawal (CEAW) at power (f) Steam generator tube rupture (SGTR)

(g) Total loss of forced reactor coolant flow (TLOF)

RESPONSE

The SONGS reload analyses process is managed by the SONGS Core Reload Analyses and Activities Checklist procedure S023-XXXVI-2.10 (available for audit). Every UFSAR Chapter 15 non-LOCA accident is first reviewed in the Summary of Transient analysis to evaluate if the existing analysis of record (AOR) is applicable to the current reload cycle. A determination is made as to whether the event needs to be reanalyzed, or needs to be reevaluated for a certain acceptance criterion, or the existing AOR(s) remain valid. The IMSF and CEAW events are typically reanalyzed in the reload analyses. The single reactor coolant pump shaft seizure event is bounded by the RCPSS event. The RCPSS event is typically reevaluated for fuel failure in the reload analyses. The SGTR and TLOF events are typically not reanalyzed in the reload analysis.

These events are analyzed for both AREVA and Westinghouse fuel using the methodology in SCE-9801 -P-A.

IMSF+SF This is a cool down event, therefore, the peak pressure criteria and the peak LHR criterion are not challenged. The IMSF+SF event is analyzed during reload to calculate the amount of fuel failure and to verify there is no DNB propagation.

The IMSF+SF event is simulated by the CENTS computer code. The difference from introduction of AREVA fuel is the M5TM cladding material. As explained in the LAR Section 4.5.3, the existing Zircaloy-4 properties in the CENTS code adequately model the M5TM properties. Therefore, there is no need to change CENTS for the introduction of AREVA fuel.

The only key input parameter related to fuel clad is the fuel rod gap conductance. Since a conservatively bounding gap conductance is used in the analysis, the response for AREVA fuel is unchanged from Westinghouse fuel.

For the purpose of evaluating coolable geometry, the existing CE DNB propagation methodology is applicable to AREVA fuel with M5TM clad (LAR Section 4.5.1) in which M5TM Page 51 of 66

is modeled as Zircaloy-4. As such, introduction of AREVA fuel does not change the method of evaluation for maintaining coolable geometry.

As shown in the Pre-Trip Steam Line Break analysis (LAR Enclosure 2, Section 7.4.1), fuel failure with Westinghouse fuel bounds the fuel failure with AREVA fuel, due to the better thermal performance of AREVA fuel. Therefore, dose consequences with AREVA fuel will not exceed those for Westinghouse fuel reported in the UFSAR and the dose and/or a CPCS power penalty to reduce the event-initiated fuel failure and consequently to reduce the dose consequences currently reported in UFSAR remain bounding.

RCP seized rotor/sheared shaft Since these events do not result in a large pressure increase in the RCS or the secondary side, the peak Pressure Criteria are not challenged.

The main objective of the analysis is to demonstrate that coolable geometry is maintained and the radiological dose consequences remain within the acceptance criteria. For the purpose of evaluating coolable geometry, the existing CE DNB propagation methodology is applicable to AREVA fuel with M5TM clad (LAR Enclosure 2, Section 4.5.1) in which M5TM is modeled as Zircaloy-4. As such, introduction of AREVA fuel does not change the method of evaluation for maintaining coolable geometry.

As shown in the Pre-Trip Steam Line Break analysis (LAR Enclosure 2, Section 7.4.1), fuel failure with Westinghouse fuel bounds the fuel failure with AREVA fuel, due to the better thermal performance of AREVA fuel. Therefore, dose consequences with AREVA fuel will not exceed those for Westinghouse fuel reported in the UFSAR and the dose consequences currently reported in UFSAR remain bounding.

CEAW The CEAW transients are simulated by the CENTS computer code. The only difference is the introduction of AREVA fuel M5TM cladding material. As explained in LAR Enclosure 2, Section 4.5.3, the existing Zircaloy-4 properties in the CENTS code adequately model the M5TM properties. Therefore there is no need to change CENTS for the introduction of AREVA fuel.

The only key input parameter related to fuel clad is the fuel rod gap conductance. Since conservatively bounding gap conductances are used in the analyses, the response for AREVA fuel is unchanged from Westinghouse fuel. See the responses to RAIs #15, 16, 19 and 25 in this document (Set 2) for more discussion.

The transient thermal-hydraulic responses are calculated by the CETOP-D code. As explained in LAR Enclosure 2, Section 4.2, the CETOP-D code is tuned and benchmarked to yield results that are conservative with respect to the TORC code results. Therefore, the CETOP-D code does not require modification as a result of the use of AREVA fuel. See the response to RAI #49 in this document (Set 2) for a discussion of CETOP/TORC Benchmarking.

Page 52 of 66

SGTR SGTR is a system response event. There is no important input parameter related to fuel rod geometry or material so introduction of AREVA fuel has no impact on the analysis of this event.

TLOF Since this event does not result in a large pressure increase in the RCS or the secondary side, peak Pressure Criteria are not challenged.

The reactor core response of this transient is modeled by the HERMITE computer code. The difference is the introduction of AREVA fuel M5TM cladding material. As explained in Section 4.5.3.2, the existing Zircaloy-4 properties in the HERMITE code adequately model the M5TM properties. Therefore there is no need to change HERMITE for the introduction of AREVA fuel.

The transient DNBR response is determined by the CETOP-D code. As explained in LAR , Section 4.2, the CETOP-D code is tuned and benchmarked to yield results that are conservative with respect to the TORC code results. Therefore, the CETOP-D code does not require modification as a result of the use of AREVA fuel. See the response to RAI #49 for a discussion of CETOP/TORC Benchmarking.

Page 53 of 66

RAI #32 32.ANP-2975(P), Table 3-4 describes the responses to the methodology (EMF-2103) safety evaluation report (SER) conditions and limitations. Please provide more details than given in Table 3-4 for the response for limitation number 3.

RESPONSE

The relevant SER condition/limitation follows - IfAREVA NP applies the RLBLOCA methodology to plants using a higherplanarlinear heat generation rate (PLHGR) than used in the current analysis, or if the methodology is to be applied to an end-of-life analysisfor which the pin pressure is significantly higher, then the needfor a blowdown clad rupture model will be reevaluated. The evaltation may be based on relevant engineeringexperience and should be documented in either the RLBLOCA guideline orplant specific calculationfile.

The premise of this condition is the magnitude of the linear heat generation rate (LHGR) of the SONGS RLBLOCA application, Summary Report ANP-2975, in comparison with the relevant sample problem (in this case the CE design problem) presented to the Nuclear Regulatory Commission (NRC) in the Evaluation Methodology (EM). The SONGS analysis was performed for the nominal core power of 3438 MW and a peak LHGR of 12.8 kW/ft; equivalent to a maximum FQ of 2.37. The sample problem was performed with a nominal core power of 2300 MW and a maximum FQ of 2.62. Power peaking applied in the SONGS Unit 3 RLBLOCA analysis (12.8 kW/ft) is bounded by the power peaking limits of the sample problem (15.7 kW/ft). As a result, no additional discussion is required by the SER condition.

AREVA has examined the fuel conditions resulting from the limiting case of the SONGS RLBLOCA analysis and determined that the hot pins will not exhibit a blowdown rupture. This review includes the limiting pins at both the first and second burnup considered in the review extend to 33.2 GWd/MTU, the burnup from statistical cycle (ANP-2975, Table 2-1). The assessment concludes that for the SONGS RLBLOCA analysis, the cladding rupture prior to the initiation of reflood does not occur in the limiting case for the first and second fuel cycles. The following information outlines the process used in this assessment.

Hot Pin Blowdown Rupture Assessment, First and Second Cycles Hot pin peak temperature and pressure were extracted from the limiting case reported in ANP-2975 and used in a rupture model. The model was applied from the initiation of the transient through the beginning of core recovery, which is the condition of lower plenum having been refilled and liquid flow into the core being re-established. This condition occurs at 25.6 seconds for the limiting case. The hoop stress is given by the thin shell formula in the following equation:

0.0005( +:  : t' --

Page 54 of 66

Where H, is the dimensionless cladding heatup rate rco and rci are the outer and inner cladding radii, respectively Pgap and Pc00o are the fuel gap and coolant pressure, respectively The cladding rupture temperature (T,., in 'C) for M5TM cladding is computed by the following correlation:

The rupture temperature predicted by the correlation, for both the first and second cycle conditions, was compared with the cladding temperature for each rod analyzed in this limiting case. A gas pin pressure reduction of 30% was used to account for the increase in gas rupture temperature was calculated both with and without pressure reduction. cladding and rupture temperatures shows that rupture does not occur prior to the beginning of core recovery for either the first or second cycle fuel rods.

End-of-Life (EOL) Considerations The SONGS RLBLOCA analysis, discussed in ANP-2975, covered a burnup range over the first two cycles of operation. Time-in-life statistical sampling occurs over a range of burnups from 0 to 31.2 GWd/MTU for the first cycle and 17.2 to 53.3 GWd/ MTU for the second cycle. AREVA fuel may be burned up to a limit of 62 GWd/ MTU. Fresh, once, and twice burned fuels contain rods with burnups that have significant overlap related to assembly placement in the core. Many of the rods in a third cycle are effectively covered by the AREVA RLBLOCA analysis which has tenrmed the burnup ranges considered "first" and "second" cycle.

What follows is a discussion of PCT behavior at burnups beyond the second cycle of fuel operation. The discussion concludes that fuel pins with burnups in the third fuel cycle range, near the EOL will not result in the limiting PCT. It also validates that the first and second fuel cycle blowdown rupture review above is applicable for the third cycle as well.

EOL Core Power A significant factor in LOCA PCT response is bundle power. A new (or fresh) assembly typically has a higher power than an average assembly in the core. It is possible that the power of a fresh assembly at BOL may be lower than an average assembly due to gadolinia content or other fuel design aspects. However the limiting case of the RLBLOCA analysis considered both the first cycle (fresh fuel) and the second cycle (once-burned fuel) of operation out to 53.3 GWd/MTU, regardless of the assembly power. For this reason, gadolinia content and fuel design are covered by the RLBLOCA analysis into the second cycle of operation.

Within the RLBLOCA analysis, pin power is controlled for the second cycle as a function of burnup. For each fuel design pin powers are examined against this limit to assure the Page 55 of 66

conservatism of this approach. Operation beyond 53.3 GWd/MTU will start during the third cycle and continue until 62 GWd/MTU, the upper limit on fuel assembly operation. In each core design effort, third cycle fuel will be examined to assure that the pins in that cycle, beyond the second cycle burnups considered by the RLBLOCA analysis, are characterized by energy potentials that are incapable of producing limiting results. This assures that the RLBLOCA analysis results remain bounding in PCT with respect to EOL fuel that is not explicitly included in the analysis.

The power generated during a LOCA is directly related to the operating power levels prior to the moment that a LOCA initiates via decay heat. As the operating power in a fuel assembly drops, the related decay heat generated in a LOCA event also drops. From the core design calculations performed for the RLBLOCA analysis, the assembly power beyond 53.3 GWd/MTU is bounded by the operating region from 0 to 53.3 GWd/MTU.

The RLBLOCA analysis sets the radial peaking at the technical specification limit, but samples the total peaking, FQ. Total peaking is ranged from a high bounded by the technical specification limit to a low bounded by the maximum peaking factor from the power history data. As a result, both the radial and total peaking used in the analysis bound the core design. This bounding approach coupled with the fact that the most limiting peaking conditions are captured within the first 53.3 GWd/MTU of operation means that the peak power will be covered in the RLBLOCA analysis.

Core Thermal Hydraulic Conditions As discussed earlier, the limiting power conditions will be realized prior to 53.3 GWd/MTU. As a result of the lower power conditions post-53.3 GWd/MTU, the amount of boiling in a sub-channel is reduced leaving more liquid in the colder fuel assemblies. More liquid in the sub-channel will provide higher convective heat transfer coefficients than would be predicted in the hotter/fresh assemblies. Accordingly, the magnitude of the post-LOCA core heatup transient (and PCT) would be expected to decrease.

Another result of reduced steam generation in the colder region of the core is a reduced pressure drop in the region. A reduction in steam flow results in reduced velocity, which is directly related to the induced pressure drop as the steam leaves the core to find its way into other parts of the NSSS. Lower powered fuel assemblies would allow water from the top of the core to penetrate downward earlier than higher powered assemblies because of the reduction in steam exiting the top of the core. Water from the lower plenum will also enter the lower powered regions of the core sooner than it would enter the higher powered regions.

Therefore, more water will be available to the assemblies with lower power, which results in higher heat transfer coefficients. All of these factors result in lower PCTs (as a result of higher heat transfer coefficients, earlier top down fluid penetration, and earlier reflooding). Therefore, the assemblies analyzed in the first 53.3 GWd/MTU will have more limiting thermal-hydraulic conditions than those with higher bumups. Some of the second cycle assemblies will be located next to higher powered fresh assemblies.

Page 56 of 66

The question can be asked: What if these assemblies suffer similar thermal-hydraulic conditions as the hotter fresher fuel due to their proximity to the hotter fresher assemblies? In this case, while it is possible that the sub-channel conditions may be similar to the hotter core regions, PCTs in the hotter regions will continue to be bounding due to the reduction in power and peaking in the lower powered assemblies, which have operated past 53.3 GWd/MTU.

Fuel Rod Conditions Fuel rod conditions also impose significant and important contributions to the PCT. Gap conditions (between the fuel pellet and the inner cladding surface) change over the life of the fuel rod by starting with an initial gap via the initial manufacturing parameters. The fresh/cold fuel pellet-to-clad gap changes after a rod transitions from cold to operating conditions. Clad creep and fuel pellet mechanical changes during operation at full power also contribute to changes in the pellet-to-clad gap size.

The fuel rod gap at normal full power operating conditions will be open at beginning-of-life (BOL), but eventually the gap will close due to pellet mechanical changes and the cladding will be in contact with the fuel pellet. At higher burnups, when the gap is closed (i.e., the thermal gap no longer changes due to mechanical changes in the fuel rod), the thermal resistance between the fuel pellet and the clad is significantly reduced. The result is that the average pellet temperature will decrease in a closed gap operating condition due to a reduced thermal resistance between the pellet and the bundle sub-channel (through the cladding).

Therefore, a closed gap condition (during normal operation) is considered less limiting (in terms of clad heatup and PCT) because the stored energy of the fuel pellet is reduced. Therefore, the RLBLOCA analysis will capture the limiting time frame in terms of gap conditions.

The power profile of the fuel pellet is also important when considering the local stored energy and decay heat at high powered axial location along a fuel rod. As mentioned earlier, the maximum average bundle power will be captured by the RLBLOCA analysis since it occurs prior to 53.3 GWd/MTU. However, the power coupled with peaking (power profile) can potentially result in a stored energy and local power condition that may be more limiting later in life. The limiting combination of both the average bundle power and the peaking conditions in the fuel will be realized prior to the 53.3 GWd/MTU and are, therefore, accommodated by the RLBLOCA analysis.

Page 57 of 66

Pellet stored energy is a function of power generation, geometry, thermal conductivity network of thermal resistances between the pellet and the fluid sub-channel (including gap, clad, and heat transfer coefficients). The Halden Project' shows that stored energy decreases as a function of burnup; fuel centerline temperatures decrease as a function of burnup (temperature is directly related to stored energy). Fuel pellet stored energy, then, is lower after 53.3 GWd/MTU than it will be for higher burnups and the most limiting condition is accommodated by the RLBLOCA analysis.

The most limiting pellet stored energy, decay heat, and power conditions will all occur within the first 53.3 GWd/MTU of operation at SONGS. The RLBLOCA analysis considers burnup to a limit of that burnup and, as a result, captures the limiting fuel rod conditions.

Swelling and Rupture The RLBLOCA methodology does not model fuel rod swelling and rupture behavior during a LOCA as a conservatism with respect to PCT. Swelling increases the heat transfer area and reduces the flow area in the region where it occurs. This results in higher local heat transfer due to the enlarged cladding surface area and higher fluid velocities at the convective boundary. The reduced area also increases the droplet shattering effect, which improves interfacial heat transfer in the sub-channel between the droplet and steam phases. Such an increase in interfacial area will boil the droplets faster and accelerate the rate at which super-heated steam temperatures are reduced.

The issue of swelling and rupture becomes important in this discussion for lower powered and older fuel assemblies due to pin pressure increasing as a function of bumup. As fuel burnup increases, the gas release from the fuel pellets contributes to an increase in the pin pressure response. Also, the pin power has a direct effect on pin pressure. As the power in a rod decreases (due to increasing burnup), the average pellet temperatures decrease and the gas pressures also decrease. Therefore, the pin pressure is a function of pellet gas generation and thermal power as it evolves through the life cycle of each individual pin. This leads to the question: What is the effect of swelling and rupture on pins that may exceed the pressure as analyzed within the first 53.3 GWd/MTU window of the RLBLOCA analysis?

If the fuel pins at higher burnups have higher internal pin pressures, a pin will swell and rupture earlier than in a lower burnup pin, resulting in flow area changes. The flow area changes would occur earlier during cycle operation for the fuel pins at a higher burnup than in lower burnup pins (hotter sub-channels).

I The Halden Project is an international project utilizing the Norwegian reactor at Halden and the support facility capable of design and fabrication of samples for experiments. Generally, samples are prepared in capsules and inserted into the core for irradiation or transient testing. Notable is the testing of high burnup fuel to determine fuel pellet thermal conductivity degradation and the high burnup LOCA testing to assess fuel pellet relocation after clad swelling and rupture.

Page 58 of 66

The combination of reduced power, reduced stored energy, and more liquid available in the sub-channels in the higher burnup fuel assemblies (as previously mentioned) coupled with the increase in heat transfer area of the swelled region will reduce PCTs below those calculated for the more extreme conditions in the hotter regions of the core (fuel burnups less than 53.3 GWd/MTU). Therefore, swelling and rupture in the lower-powered assemblies is regarded as less limiting than in the hotter assemblies.

Swelling and rupture during the reflooding phase of an LBLOCA generally improves cooling near the rupture zone. However, rupture occurring during the middle of the blowdown period can be detrimental to core cooling. The increased gap size and the presence of superheated steam act to insulate the fuel from the cladding. This allows the stored energy in the fuel pellet to increase.

This increase in stored energy (fuel temperature) can be challenging later in the transient when the clad heat removal rate is minimal. During the later stages of blowdown and during the lower plenum refill period, the reduced clad cooling allows the fuel pin to heat up the cladding, to the extent that it approaches equilibrium with the fuel temperature. In extreme cases, the clad temperature can reach values where significant metal-water reaction energy is generated and the clad becomes at risk of exceeding the 2200 'F IOCFR50.46(b) acceptance criteria.

For mid-blowdown ruptures to occur, there needs to be a convergence of two parameters-high internal pin pressure and initial fuel temperature. A high pin pressure is required to create an outward hoop stress and a high initial fuel temperature, which is associated with high power peaking, is necessary to promote elevated cladding temperatures. All things being equal, pin pressure increases with burnup. This adversely influences (reduces) rupture temperature (computed from fuel rod models such as GALILEO or RODEX3A). However, the lower power levels that are involved in the burned fuel pins near EOL more than compensates for this effect by reducing the cladding temperatures to values below the rupture temperatures until later in the transient, thus, removing midblowdown rupture as a concern.

Based on the discussions in this section, it is concluded that fuel pellet stored energy, gap conditions, thermal-hydraulic conditions, decay heat, peak power and bundle power will be limiting in the first 53.3 GWd/mtU; the burnup period covered by the RLBLOCA analysis.

Consequently, the PCTs and fuel pin heatup transient computed by the RLBLOCA analysis will remain limiting. In addition, assessments of blowdown rupture performed for the limiting SONGS RLBLOCA case bound those expected for third cycle fuel pre-empting the need to explicitly examine EOL fuel.

Page 59 of 66

Ratio Page 60 of 66

RAI #33

33. By AREVA methodology prescription, U-235 enrichment for gadolinia rods is decreased by 5% per 1 wt% of gadolinia in the fuel rod. (See footnote for Table 7.1.1 of Enclosure 2). ANP-2975(P) that describes The SONGS Realistic Large Break LOCA (RLBLOCA) analysis, determined that the limiting peak clad temperature (PCT) (16050 F) occurred for a fresh 6% Gad rod. In spite of a combined effect of fuel burnup and reduction in uranium-235 enrichment adjustment, the limiting PCT occurred in (( )) Gad rod rather than in a fresh U0 2 rod. Please provide the details of the analysis that resulted in a limiting PCT in a Gad rod.

RESPONSE

In AREVA's letter to the NRC (see the LAR Attachment 1, Item 2), AREVA discusses the modeling of gadolinia-bearing fuel rods. The discussion indicates that the power of the gadolinia rods is limited, relative to U02 rods, to account for neutron poisons. Gadolinia content also affects the thermal conductivity of the fuel pellets and is simulated in the S-RELAP5 model. The affect is a proportional reduction in conductivity with increasing gadolinia content. The discussion points out that gadolinia rods may characterize higher PCTs. It is worth noting that the transient temperature response is similar for all of the hot rods in the limiting case.

For the limiting case, Case 7 reported in ANP-2975, the limiting PCT occurs in the first cycle

(( )) gadolinia rod. Closer examination of the results shows that, for the fresh hot rods PCTs for all of the gadolinia rods exceed those of the U02 rod. These results correspond to higher fuel temperatures of the gadolinia rods. Table 4-1 gives a summary of fuel temperature and PCT for each fresh hot rod. The reduction in rod power counters the reduction in thermal conductivity but not to the rod to be PCT-limiting in this case.

Limiting Case First Cycle Hot Rod Fuel Temperature and PCT II Page 61 of 66

RAI #34

34. (Section 4.6 and Section 5.2.3 of Enclosure 2,Long Term Core Cooling Analysis(LTC)) Section 4.6 indicates that "The current Westinghouse post-LOCA Long Term Core Cooling analyses will remain applicable for both AREVA and Westinghouse fuel." An examination of updated SONGS FSAR Section 6.3.3.4, Post-LOCA Long Term Cooling indicates that "the LTC for San Onofre Units 2 and 3 LTC was performed using the NRC approved cods and methods documented in CENPD-254-P-A. By letter dated August 1, 2005 from Robert Gramm (NRC) to John Gresham (Westinghouse), the NRC suspended the NRC approval for use of Westinghouse Topical Report CENPD-254-P, Post-LOCA Long Term Cooling Model due to the discovery of non-conservative modeling assumptions (Reference 3).

Technical evaluation that was attached to this letter identified the subject areas regarding post-LOCA long term cooling and boric acid precipitation behavior that required Westinghouse to require revision of the subject topical report. Please justify the continued use of the topical report, CENPD-254-P by SONGS for LTC.

RESPONSE

As described in UFSAR Section 6.3.3.4, the post-LOCA Long Term Cooling (LTC) analysis is based on the methodology described in CENPD-254-P. As discussed in letter ML053220569 (D.S. Collins (USNRC) to G.C. Bischoff (Westinghouse) dated November 23, 2005), the NRC determined that there is sufficient safety margin for CE-designed plants that reference CENPD-254-P. This letter included four requirements that need to be addressed on a plant-specific basis in any future submittals regarding post-LOCA LTC. As discussed in the LAR Enclosure 2 Section 5.2.3.4, no changes to the SONGS LTC analysis are being proposed, nor is a power uprate being proposed. Therefore, the existing LTC analysis and methods remain applicable. The Pressurized Water Reactor Owners Group evaluation (letter ML061720175, PWROG letter OG-06-200, dated June 19, 2006) of the existing SONGS LTC analysis with respect to the four NRC requirements described in letter ML053220569 remains valid. This evaluation concluded that sufficient margin exists to address the NRC concerns, such that SONGS remains in compliance with the regulations and its design basis. Based on the preceding, the continued use of CENPD-254-P based LTC analysis to show compliance with IOCFR50.46 is appropriate.

Page 62 of 66

RAI #35

35. Please provide details of the boric acid precipitation analysis. The response should include a list of all of the key parameter inputs and results from the analysis.

RESPONSE

The long-term cooling mechanism for a hot leg break is forced convection to liquid. Once cooling is established, and a positive core flow is assured, boron precipitation is not an issue, and no further consideration is necessary. For cold leg breaks, there is no forced flow through the core. The liquid head balance between the core and the downcomer prevents ECCS water from entering the core at a rate faster than core boil-off. Extra injection simply flows out the break and spills to the containment floor. With no core flow, core boiling acts to concentrate boric acid adding to the potential for precipitation and core blockage. To eliminate boron precipitation and any accompanying core blockage, operator action is required to establish hot and cold leg re-circulation (positive core flow).

Hot and cold leg injection is initiated to provide long-term cooling; this induces a positive core flow, capable of controlling the concentration of boric acid. The timing and effectiveness of the hot leg injection is established by demonstrating that the in-vessel concentration of boric acid is below solubility limits.

The key input parameters for the boric acid precipitation analysis are:

PARAMETER VALUE Reactor power level 3458 MWt Boron source concentrations (the maximum RCS = 1.831 wt % = 3,200 parts per million (ppm) possible boric acid concentration is assumed from RWST = 1.831 wt % = 3,200 ppm each of these sources) SIT = 1.831 wt % = 3,200 ppm BAMU= 3.5 wt%= 6,119 ppm Boron source volumes (the water volumes from RCS = 11,289 ft3 - maximum each source are determined such that the effect of RWST = 508,000 gal - maximum injection into the RCS maximizes the boric acid SIT = 7,300 ft3 - maximum concentration in the RCS). BAMU = 23,600 gal - maximum SG Tube plugging (per SG) <779 tubes (8%)

Tube plugging differential between SGs <779 tubes (8%)

The analysis results show that the 9.8 ft2 cold leg break is the limiting break in regard to long-term, boric acid accumulation in the inner vessel region. The initiation of simultaneous cold/hot side HPSI pump injection flow of at least 270 gpm per side by three hours post-LOCA provides a substantial and time increasing core flushing flow, as shown in UFSAR Figure 6.3-21. UFSAR Figure 6.3-22 shows that even with no core flushing flow, the boric acid would not begin to precipitate until after seven hours post LOCA. It is clearly demonstrated that boron precipitation Page 63 of 66

will not occur following the postulated loss-of-coolant accident, since the boric acid concentration is maintained well below the precipitation limit.

The boron precipitation analysis is not dependent on the fuel element design since boron concentrations depend on ECCS injection rate, RCS geometry, and core power level. Since the AREVA fuel does not alter these factors, the current evaluation remains valid and is equally applicable to AREVA 16x1 6 HTP fuel. Emergency operating procedures provide guidance to address the boric acid precipitation issue and ensure that long-term cooling is maintained.

Page 64 of 66

RAI #36

36. Explain how the fuel TCD due to burnup and addition of Gadolinium (Gd) for the U0 2 fuel is modeled in RODEX3A. Explain the impact of the TCD and addition of Gd on the following characteristics of the fuel:

(a) Fuel temperature calculation (b) Initial stored energy in the fuel (c) Fission gas release (d) Fuel-cladding gap conductance.

RESPONSE

RODEX3A models the effect of Gd 20 3 additive on the thermal conductivity of the gadolinia fuel by the generally used approach of an additive term to the denominator of the phononic component of the thermal conductivity, therefore decreasing the thermal conductivity of gadolinia fuel in comparison to that of U0 2 fuel, in proportion with the Gd 20 3 content. On the other hand, RODEX3A does not account for burnup degradation of thermal conductivity for either UO 2 or gadolinia fuels. This led to under-predicting the measured temperatures in the Halden tests for exposures greater than (( 11, indicative of a general under-prediction of RODEX3A of fuel temperature at medium to high burnup.

This was recognized and the NRC approved unconditional use of RODEX3A only for BOL best-estimate fuel initialization for input to LOCA analyses for PWR fuel designs. Because of this limitation of RODEX3A temperature and consequently its fission gas release predictive capability, RODEX3A has never been used for fuel thermal-mechanical licensing analyses.

RODEX3A was used for item (b), LOCA initialization, in conjunction with the AREVA RLBLOCA methodology. The bias off-set correlation of RODEX3A has been developed based on benchmarks to Halden tests, which were not originally included in the RODEX3A topical report. Among these new cases, the IFA-515.0O-A2 test measured the thermal conductivity degradation (TCD) of a gadolinia fuel. This test was used for RODEX4 in order to confirm that the TCD term of the [I JI In terms of RODEX3A applications, the good agreement obtained for IFA-5151 O-A2 after applying the bias off-set correction, demonstrate the applicability of RODEX3A to gadolinia fuel temperature evaluations for LOCA initialization in the frame of the RLBLOCA methodology.

Fission gas release (FGR) is predicted by RODEX3A and is important as it pertains to LOCA fuel pin initialization and fuel enthalpy prediction. FGR is primarily dependent on [

Page 65 of 66

Regarding the fuel-cladding gap conductance, the RODEX3A model is biased low, which causes an over-prediction of fuel temperatures at low exposures. The over-prediction is not sufficient to compensate for the lack of TCD due to exposure at mid-range burnups. With the inclusion in the RODEX3A benchmarks of the additional Halden tests in the RODEX3A bias, the fuel temperature predictions, including gap conductivity and pellet thermal conductivity degradation effects, are properly calculated throughout the licensed burnup range.

Page 66 of 66