ML082690016

Browns Ferry, Units 2 & 3 - Specifications (TS) Change TS-418 - Extended Power Uprate (EPU) - Supplemental Response to Request for Additional Information (RAI) Round 3 & 18 & Response to Round 20 Fuels Methods RAIs

ML082690016
Person / Time
Site:	Browns Ferry
Issue date:	09/19/2008
From:	Langley D T Tennessee Valley Authority
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
TAC MD5263, TAC MD5264, TS-418, TVA-BFN-TS-418
Download: ML082690016 (100)

Text

{{#Wiki_filter:Tennessee Valley Authority, Post Office Box 2000, Decatur, Alabama 35609-2000 September 19, 2008 TVA-BFN-TS-418 10 CFR 50.90 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Mail Stop OWFN, P1-35 Washington, D. C. 20555-0001 Gentlemen: In the Matter of ) Docket Nos. 50-260 Tennessee Valley Authority ) 50-296 BROWNS FERRY NUCLEAR PLANT (BFN) -UNITS 2 AND 3 -TECHNICAL SPECIFICATIONS (TS) CHANGE TS-418 -EXTENDED POWER UPRATE (EPU) -SUPPLEMENTAL RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION (RAI)ROUNDS 3 AND 18 AND RESPONSE TO ROUND 20 FUELS METHODS RAIs (TAC NOS. MD5263 AND MD5264)By letter dated June 25, 2004 (ADAMS Accession No. ML041840301), TVA submitted a license amendment application to the NRC for EPU operation of BFN Units 2 and 3. The pending EPU amendment increases the maximum authorized power level by approximately 14 percent from 3458 megawatts thermal (MWt) to 3952 MWt.On July 17, 2008, NRC issued a Round 18 RAI (ML081700102) regarding AREVA fuel methods used in support of Units 2 and 3 EPU operations. Round 18 consists of 32 RAI questions, SRXB-91 through SRXB-122. To facilitate the review of TS-418, a meeting was held on August 7, 2008, with NRC staff to review draft responses to SRXB-91 through SRXB-116. Subsequently, on August 15, 2008, TVA submitted a partial response (ML082330187) to Round 18; specifically to RAIs SRXB-92, 93, 95, 96, 97, 99, 100, and 102 through 116. This submittal responds to the remainder of the Round 18 RAIs.Additionally, NRC staff conducted an audit of AREVA fuel methods from August 18through August 28, 2008, at the AREVA engineering facilities in Richland, Washington. As a result of the audit, TVA agreed to provide supplemental responses to a number of the August 15, 2008, Round 18 RAI responses and also to Round 3 RAIs SRXB-A.34 and SRXB-A.42. The original TS-418 Round 3 response was submitted on March 7, 2006 (ML060680583). Lastly, this submittal also provides responses to the five fuels related RAIs, SRXB-123 through SRXB-127, from the NRC Round 20 RAI dated September 16, 2008.Tennessee Valley Authority, Post Office Box 2000, Decatur, Alabama 35609-2000 September 19, 2008 TVA-BFN-TS-418 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk . Mail Stop OWFN, P1-35 Washington, D. C. 20555-0001 Gentlemen: In the Matter of Tennessee Valley Authority ) ) 10 CFR 50.90 Docket Nos. 50-260 50-296 BROWNS FERRY NUCLEAR PLANT (BFN) -UNITS 2 AND 3 -TECHNICAL SPECIFICATIONS (TS) CHANGE TS-418 -EXTENDED POWER UPRATE SUPPLEMENTAL RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION (RAI) ROUNDS 3 AND 18 AND RESPONSE TO ROUND 20 FUELS METHODS RAls (TAC NOS. MD5263 AND MD5264) By letter dated June 25,2004 (ADAMS Accession No. ML041840301), TVA submitted a license amendment application to the NRC for EPU operation of BFN Units 2 and 3. The pending EPU amendment increases the maximum authorized power level by approximately 14 percent from 3458 megawatts thermal (MWt) to 3952 MWt. On July 17, 2008, NRC issued a Round 18 RAI (ML081700102) regarding AREVA fuel methods used in support of Units 2 and 3 EPU operations. Round 18 consists of 32 RAI questions, SRXB-91 through SRXB-122. To facilitate the review of TS-418, a meeting was held on August 7,2008, with NRC staff to review draft responses to SRXB-91 through SRXB-116. Subsequently, on August 15, 2008, TVA submitted a partial response (ML082330187) to Round 18; specifically to RAls SRXB-92, 93, 95, 96, 97, 99,100, and 102 through 116. This submittal responds to the remainder of the Round 18 RAls. Additionally, NRC staff conducted an audit of AREVA fuel methods from August 18 through August 28,2008, at the AREVA engineering facilities in Richland, Washington. As a result of the audit, TVA agreed to provide supplemental responses to a number of the August 15,2008, Round 18 RAI responses and also to Round 3 RAls SRXB-A.34 and SRXB-A.42. The original TS-418 Round 3 response was submitted on March 7, 2006 (ML060680583). Lastly, this submittal also provides responses to the five fuels related RAls, SRXB-123 through SRXB-127, from the NRC Round 20 RAI dated September 16,2008. U.S. Nuclear Regulatory Commission Page 2 September 19, 2008 As discussed with the NRC Project Manager for BFN, Ms. Eva Brown, on September 17, 2008, responses to remainder of the Round 20 RAIs related to steam dryers, along with supplemental responses to Round 19 RAIs EMCB.147 and EMCB.192/150 pertaining to steam dryer analyses will be provided at a later date.Enclosure 1 is a proprietary response to the subject RAIs and contains information that AREVA NP, Inc. (AREVA) considers to be proprietary in nature and subsequently, pursuant to 10 CFR 9.17(a)(4), 2.390(a)(4) and 2.390(d)(1), AREVA requests that such information be withheld from public disclosure. Enclosure 2 is a redacted version of Enclosure 1 with the proprietary material removed and is suitable for public disclosure. Enclosure 3 contains an affidavit from AREVA supporting this request for withholding from public disclosure. TVA has determined that the additional information provided by this letter does not affect the no significant hazards considerations associated with the proposed TS changes. The proposed TS changes still qualify for a categorical exclusion from environmental review pursuant to the provisions of 10 CFR 51.22(c)(9). No new regulatory commitments are made in this submittal. If you have any questions regarding this letter, please contact me at (256)729-7658. I declare under penalty of perjury that the foregoing is true and correct. Executed on this 19th day of September, 2008.Sincerely, 0 .T. L ngle Site Licensing and Industry Affairs Manager

Enclosures:

1. Supplemental Response to Request for Additional Information (RAI) Rounds 3 and 18 and Response to Round 20 Fuels Methods RAIs (Proprietary Information Version)2. Supplemental Response to Request for Additional Information (RAI) Rounds 3 and 18 and Response to Round 20 Fuels Methods RAIs (Non-Proprietary Information Version)3. Affidavit U.S. Nuclear Regulatory Commission Page 2 September 19, 2008 As discussed with the NRC Project Manager for BFN, Ms. Eva Brown, on September 17,2008, responses to remainder of the Round 20 RAls related to steam dryers, along with supplemental responses to Round 19 RAls EMCB.147 and EMCB.192/150 pertaining to steam dryer analyses will be provided at a later date. Enclosure 1 is a proprietary response to the subject RAls and contains information that AREVA NP, Inc. (AREVA) considers to be proprietary in nature and subsequently, pursuant to 10 CFR 9.17(a)(4), 2.390(a)(4) and 2.390(d)(1), AREVA requests that such information be withheld from public disclosure.

Enclosure 2 is a redacted version of Enclosure 1 with the proprietary material removed and is suitable for public disclosure. Enclosure 3 contains an affidavit from AREVA supporting this request for withholding from public disclosure. TVA has determined that the additional information provided by this letter does not affect the no significant hazards considerations associated with the proposed TS changes. The proposed TS changes still qualify for a categorical exclusion from environmental review pursuant to the provisions of 10 CFR 51.22(c)(9). No new regulatory commitments are made in this submittal. If you have any questions regarding this letter, please contact me at (256)729-7658. I declare under penalty of perjury that the foregoing is true and correct. Executed on this 19th day of September, 2008. Sincerely, Site Licensing and Industry Affairs Manager

Enclosures:

1. Supplemental Response to Request for Additional Information (RAI) Rounds 3 and 18 and Response to Round 20 Fuels Methods RAls (Proprietary Information Version) 2. Supplemental Response to Request for Additional Information (RAI) Rounds 3 and 18 and Response to Round 20 Fuels Methods RAls (Non-Proprietary Information Version) 3. Affidavit U.S. Nuclear Regulatory Commission Page 3 September 19, 2008

Enclosures:

cc (Enclosures): State Health Officer Alabama State Department of Public Health RSA Tower -Administration Suite 1552 P.O. Box 303017 Montgomery, Alabama 36130-3017 Ms. Eva Brown, Project Manager U.S. Nuclear Regulatory Commission (MS 08G9)One White Flint, North 11555 Rockville Pike Rockville, Maryland 20852-2739 Eugene F. Guthrie, Branch Chief U.S. Nuclear Regulatory Commission Region II Sam Nunn Atlanta Federal Center 61 Forsyth Street, SW, Suite 23T85 Atlanta, Georgia 30303-8931 NRC Resident Inspector Browns Ferry Nuclear Plant 10833 Shaw Road Athens, Alabama 35611-6970 u.s. Nuclear Regulatory Commission Page 3 September 19, 2008

Enclosures:

cc (Enclosures): State Health Officer Alabama State Department of Public Health RSA Tower -Administration Suite 1552 P.O. Box 303017 Montgomery, Alabama 3(3130-3017 Ms. Eva Brown, Project Manager U.S. Nuclear Regulatory Commission (MS 08G9) One White Flint, North 11555 Rockville Pike Rockville, Maryland 20852-2739 Eugene F. Guthrie, Branch Chief U.S. Nuclear Regulatory Commission Region II Sam Nunn Atlanta Federal Center 61 Forsyth Street, SW, Suite 23T85 Atlanta, Georgia 30303-8931 NRC Resident Inspector Browns Ferry Nuclear Plant 10833 Shaw Road Athens, Alabama 35611-6970 U.S. Nuclear Regulatory Commission Page 4 September 19, 2008 JEE:BCM:BDL cc (w/o Enclosures): G. P. Arent, EQB 1B-WBN W. R. Campbell, Jr., LP 3R-C S. M. Douglas, POB 2C-BFN R. F. Marks, Jr., PAB 1C-BFN D. C. Matherly, BFT 2A-BFN L. E. Nicholson, LP 4K-C L. E. Thibault, LP 3R-C R. G. West, NAB 2A-BFN B. A. Wetzel, LP 4K-C S. A. Vance, WT 6A-K E. J. Vigluicci, ET 11A-K NSRB Support, LP 5M-C EDMS WT CA-K, s:licensing/lic/submit/subs/EPU/RAI Round 3 and 18 and Round 20 Fuels Methods FAIs/Supplemental Response to Request for Additional Information (RAI) Rounds 3 and 18 and Response to Round 20 Fuels Methods RAIs u.s. Nuclear Regulatory Commission Page 4 September 19, 2008 JEE:BCM:BDL cc (w/o Enclosures): G. P. Arent, EQB 1B-WBN W. R. Campbell, Jr., LP 3R-C S. M. Douglas, POB 2C-BFN R. F. Marks, Jr., PAB 1C-BFN D. C. Matherly, BFT 2A-BFN L. E. Nicholson, LP 4K-C L. E. Thibault, LP 3R-C R. G. West, NAB 2A-BFN B. A. Wetzel, LP 4K-C S. A. Vance, WT 6A-K E. J. Vigluicci, ET 11 A-K NSRB Support, LP 5M-C EDMS WT CA-K, s:licensing/lic/submitlsubs/EPU/RAI Round 3 and 18 and Round 20 Fuels Methods FAls/Supplemental Response to Request for Additional Information (RAI) Rounds 3 and 18 and Response to Round 20 Fuels Methods RAls NON-PROPRIETARY INFORMATION ENCLOSURE 2 TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT (BFN)UNITS 2 AND 3 TECHNICAL SPECIFICATIONS (TS) CHANGE TS-418 EXTENDED POWER UPRATE (EPU)SUPPLEMENTAL RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION (RAI)ROUNDS 3 AND 18 AND RESPONSE TO ROUND 20 FUELS METHODS RAIs (NON-PROPRIETARY INFORMATION VERSION)This enclosure provides TVA supplemental responses to Round 3 RAIs SRXB-A.34 and SRXB-A.42, a supplemental response to NRC's July 17, 2008, Round 18 RAI, and a response to the five fuels methods related RAIs, SRXB-123 through SRXB-127, from NRC's September 16, 2008, Round 20 RAI.7 NON-PROPRIETARY INFORMATION ENCLOSURE 2 TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT (BFN) UNITS 2 AND3 TECHNICAL SPECIFICATIONS (TS) CHANGE TS-418 EXTENDED POWER UPRATE (EPU) SUPPLEMENTAL RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION (RAI) ROUNDS 3 AND 18 AND RESPONSE TO ROUND 20 FUELS METHODS RAls (NON-PROPRIETARY INFORMATION VERSION) This enclosure provides TVA supplemental responses to Round 3 RAls SRXB-A.34 and SRXB-A.42, a supplemental response to NRC's July 17, 2008, Round 18 RAI, and a response to the five fuels methods related RAls, SRXB-123 through SRXB-127, from NRC's September 16, 2008, Round 20 RAI. ,/ NON-PROPRIETARY INFORMATION NRC staff conducted an audit of AREVA fuel methods from August 18 through August 28, 2008, at the AREVA engineering facilities in Richland, Washington. As a result of the audit, TVA agreed to provide supplemental responses Round 3 RAIs SRXB-A.34 and SRXB-A.42. The previous Round 3 responses were originally submitted on March 7, 2006 (ML060680853). A revised response to SRXB-A.34 was also submitted on May 11, 2006 (ML061360148). NRC RAI SRXB-A.34 (From Round 3)Describe qualitatively the cross-section reconstruction process incorporated in CASMO-4 and MICROBURN-B2. The response should reflect the information provided in the slides (1-35) of the August 4 presentations, including high void fraction effects and accuracy. Provide flow chart(s), road map(s) and any other means to demonstrate the process, starting from the gathered raw void fraction data, how that data is used by CASMO-4 to generate the required cross-sections. In addition, briefly describe the development of the void fraction correlation and associated uncertainties. Supplemental Response to SRXB-A.34 MICROBURN-B2 versions prior to 2003 treated cross section dependency on spectral history differently between the fuel nuclide depletion module and the neutron flux calculation module.The fuel nuclide depletion module used [] while the neutron flux iteration calculation module used a []. This inconsistency was remedied starting in 2003 by changing the depletion module to the[ ]. Starting from 2006, both modules were converted to the [].These changes over the years were mainly due to code maintenance concerns and did not impact any result due to the [3. Unlike the cross section dependency on the instantaneous void, the [] is rather weak. This is shown in Figure SRXB-A.34.1 for Pu-239 and in Figure SRXB-A.34.2 for Pu-240. The []. At the high end of [ ], the difference between the []. This kind of difference is entirely within the uncertainty of nuclear cross section measurement and its evaluation process including the CASMO-4 lattice code.' It has no observable effect on the reactor nodal power distribution and the reactor criticality evaluation as has been verified in the code maintenance record of MICROBURN-B2. E2-1 NON-PROPRIETARY INFORMATION NRC staff conducted an audit of AREVA fuel methods from August 18 through August 28, 2008, at the AREVA engineering facilities in Richland, Washington. As a result of the audit, TVA agreed to provide supplemental responses Round 3 RAls SRXB-A.34 and SRXB-A.42. The previous Round 3 responses were originally submittecj on March 7, 2006 (ML060680853). A revised response to SRXB-A.34 was also submitted on May 11,2006 (ML061360148). NRC RAI SRXB-A.34 (From Round 3) Describe qualitatively the cross-section reconstruction process incorporated in CASMO-4 and MICROBURN-B2. The response should reflect the information provided in the slides (1-35) of the August 4 presentations, including high void fraction effects and accuracy. Provide flow chart(s), road map(s) and any other means to demonstrate the process, starting from the gathered raw void fraction data, how that data is used by CASMO-4 to generate the required cross-sections. In addition, briefly describe the development of the void fraction correlation and associated uncertainties. Supplemental Response to SRXB-A.34 MICROBURN-B2 versions prior to 2003 treated cross section dependency on spectral history differently between the fuel nuclide depletion module and the neutron flux calculation module. The fuel nuclide depletion module used [ ] while the neutron flux iteration calculation module used a [ ]. This inconsistency was remedied starting in 2003 by changing the depletion module to the* [ ]. Starting from 2006, both modules were converted to the [ ]. These changes over the years were mainly due to code maintenance concerns and did not impact any result due to the [ ]. Unlike the cross section dependency on the instantaneous void, the [ ] is rather weak. This is shown in Figure SRXB-A.34.1 for Pu-239 and in Figure SRXB-A.34.2 for Pu-240. The [ 1 At the high end of [ ], the difference between the [ ]. This kind of difference is entirely within the uncertainty of nuclear cross section measurement and its evaluation process including the CASMO-4 .lattice code. It has no observable effect on the reactor nodal power distribution and the reactor criticality evaluation as has been verified in the code maintenance record of MICROBURN-B2. . E2-1 NON-PROPRIETARY INFORMATION r" Figure SRXB-A.34.1 PU-239 sigma-1 Dependence on Spectral History at 20 Gigawatt-days per ton (GWd/T)r--U Figure SRXB-A.34.2 PU-240 sigma-1 Dependence on Spectral History at 20 GWd/T E2-2 r r NON-PROPRIETARY INFqRMATION . Figure SRXB-A.34.1 PU-239 sigma-1 Dependence on Spectral History at 20 Gigawatt-days per ton (GWd/T) Figure SRXB-A.34.2 PU-240 sigma-1 Dependence on Spectral History at 20 GWd/T E2-2 ..J NON-PROPRIETARY INFORMATION NRC RAI SRXB-A.42 (From Round 3)In August 30, 2004, General Electric Nuclear Energy (GENE) issued a 10 CFR Part 21 report (ADAMS ML042720293), stating that using limiting control rod blade patterns developed for less than rated flow at rated power conditions could sometimes yield more limiting bundle-by-bundle MCPR distributions and/or more limiting bundle axial power shapes than using limiting control rod patterns developed for rated flow/rated power in the SLMCPR calculation. The affected plants submitted amendment requests increasing their SLMCPR value. The staff understands that Framatome did not issue a Part 21 reporting on the SLMCPR methodology that addresses the calculation of the SLMCPR at minimum core flow and off-rated conditions similar to GENE's Part 21 report.Reference the applicable sections of the ANF-524P-A SLMCPR methodology that specify the requirement to calculate the SLMCPR at the worst case conditions for minimum core flow conditions for rated power. Demonstrate that the SLMCPR is calculated at different statepoints of the licensed operating domain, including the minimum core flow statepoint and that the calculation is performed for different exposure points.Supplemental Response to SRXB-A.42 AREVA NP 1 performs the safety limit Minimum Critical Power Ratio (SLMCPR) analysis on a cycle-specific basis. As discussed in the original response to SRXB-A.42 (Reference SRXB-A.42.1), the core power distributions used in the SLMCPR analyses are obtained from the MICROBURN-B2 cycle-specific design basis step-through calculation. SLMCPR analyses are performed with these power distributions at the minimum and maximum core flow allowed at rated power.The SLMCPR analyses supporting the BFN Unit 2 EPU submittal were performed for an ATRIUMTM-10 2 equilibrium cycle that assumed Maximum Extended Load Line Limit Analysis plus (MELLLA+) operation. The BFN EPU SLMCPR analyses considered the minimum and maximum flow at rated power for planned MELLLA+ operation. The cycle-specific SLMCPR analyses supporting current operating cycles for BFN Units 2 and 3 were performed consistent with the currently allowed power/flow maps for these cycles and did not include the MELLLA+flow window. Future cycle-specific BFN SLMCPR analyses will be performed consistent with the allowable power/flow map for the cycle.The AREVA SLMCPR methodology uses a design basis core power distribution. The criteria for selecting the design basis power distribution are specified in Reference SRXB-A.42.2 and state that analyses be performed with power distributions that "...conservatively represent expected reactor operating states which could both exist at the MCPR operating limit and produce a MCPR equal to the MCPR safety limit during an anticipated operational occurrence." Candidate design basis power distributions are obtained from the cycle-specific design step-through. Thedesign step-through reflects the cycle design energy and operating strategy planned by the utility and is the best projection of how the cycle will operate.The design step-through is required to have margin to the operating limit MCPR (OLMCPR).Flatter (less peaked) radial power distributions are conservative for the SLMCPR analysis. The radial power distributions from the cycle step-through are flatter than the radial power 1 AREVA NP Inc. is an AREVA and Siemens company.2 ATRIUM is a trademark of AREVA NP.E2-3 NON-PROPRIETARY INFORMATION NRC RAI SRXB-A.42 (From Round 3) In August 30,2004, General Electric Nuclear Energy (GENE) issued a 10 CFR Part 21 report (ADAMS ML042720293), stating that using limiting control rod blade patterns developed for less than rated flow at rated power conditions could sometimes yield more limiting bundle-by-bundle MCPR distributions and/or more limiting bundle axial power shapes than using limiting control rod patterns developed for rated flow/rated power in the SLMCPR calculation. The affected plants submitted amendment requests increasing their SLMCPR value. The staff understands that Framatome did not issue a Part 21 reporting on the SLMCPR methodology that addresses the calculation of the SLMCPR at minimum core flow and off-rated conditions similar to GENE's Part 21 report. Reference the applicable sections of the ANF-524P-A SLMCPR methodology that specify the requirement to calculate the SLMCPR at the worst case conditions for minimum core flow conditions for rated power. Demonstrate that the SLMCPR is calculated at different statepoints of the licensed operating domain, including the minimum core flow statepoint and that the calculation is performed for different exposure points. Supplemental Response to SRXB-A.42 AREVA NP1 performs the safety limit Minimum Critical Power Ratio (SLMCPR) analysis on a cycle-specific basis. As discussed in the original response to SRXB-A.42 (Reference SRXB-A.42.1), the core power distributions used in the SLMCPR analyses are obtained from the MICROBURN-B2 cycle-specific design basis step-through calculation. SLMCPR analyses are performed with these power distributions at the minimum and maximum core flow allowed at rated power. The SLMCPR analyses supporting the BFN Unit 2 EPU submittal were performed for an ATRIUMTM-10 2 equilibrium cycle that assumed Maximum Extended Load Line Limit Analysis plus (MELLLA+) operation. The BFN EPU SLMCPR analyses considered the minimum and maximum flow at rated power for planned MELLLA+ operation. The cycle-specific SLMCPR analyses supporting current operating cycles for BFN Units 2 and 3 were performed consistent with the currently allowed power/flow maps for these cycles and did not include the MELLLA+ flow window. Future cycle-specific BFN SLMCPR analyses will be performed consistent with the allowable power/flow map for the cycle. The AREVA SLMCPR methodology uses a design basis core power distribution. The criteria for selecting the design basis power distribution are specified in Reference SRXB-A.42.2 and state that analyses be performed with power distributions that" ... conservatively represent expected reactor operating states which could both exist at the MCPR operating limit and produce a MCPR equal to the MCPR safety limit during an anticipated operational occurrence." Candidate design basis power distributions are obtained from the cycle-specific design step-through. The design step-through reflects the cycle design energy and operating strategy planned by the utility and is the best projection of how the cycle will operate. The design step-through is required to have margin to the operating limit MCPR (OLMCPR). Flatter (less peaked) radial power distributions are conservative for the SLMCPR analysis. The radial power distributions from the cycle step-through are flatter than the radial power 1 2 AREVA NP Inc. is an AREVA and Siemens company. ATRIUM is a trademark of AREVA NP. E2-3 NON-PROPRIETARY INFORMATION distributions that would result from adjusting the control rod patterns until the core OLMCPR is reached. These control rod adjustments would result in a more peaked radial power distribution and increased margin to the SLMCPR. The design margin to the OLMCPR ensures that the power distributions from the cycle step-through are conservative relative to the power distributions that may occur during actual operation of the cycle.Figure SRXB-A.42.1 provides a comparison of the core radial power distribution from the design step-through and from actual operation for a BWR/4 at EPU. The power distributions are at the cycle exposure that was limiting for the SLMCPR analysis. The figure shows that the actual power distribution had a higher radial power distribution and is less flat than the design step-through power distribution. In addition, for actual operation there was still 5.1% MCPR margin. These comparisons demonstrate that the radial power distribution used in the SLMCPR analysis is conservative relative to the required SLMCPR design basis power distribution and bounds actual operation.

Reference:

SRXB-A.42.1 SRXB-A.42.2 Correspondence, W.D. Crouch (TVA) to U.S. Nuclear Regulatory Commission, "Browns Ferry Nuclear Plant (BFN) -Units 2 and 3, Response to NRC Round 3 Requests for Additional Information Related to Technical Specifications (TS) Change No. TS-418 -Requests for Extended Power Uprate Operation (TAC Nos. MC3743 and MC3744)," March 7, 2006 (ML060680583). ANF-524(P)(A) Revision 2 and Supplements 1 and 2, ANF Critical Power Methodology for Boiling Water Reactors, Advanced Nuclear Fuels Corporation, November 1990 E2-4 NON-PROPRIETARY INFORMATION distributions that would result from adjusting the control rod patterns until the core OLMCPR is reached. These control rod adjustments would result in a more peaked radial power distribution and increased margin to the SLMCPR. The design margin to the OLMCPR ensures that the power distributions from the cycle step-through are conservative relative to the power distributions that may occur during actual operation of the cycle. Figure SRXB-A.42.1 provides a comparison of the core radial power distribution from the design step-through and from actual operation for a BWRl4 at EPU. The power distributions are at the cycle exposure that was limiting for the SLMCPR analysis. The figure shows that the actual power distribution had a higher radial power distribution and is less flat than the design step-through power distribution. In addition, for actual operation there was still 5.1 % MCPR margin. These comparisons demonstrate that the radial power distribution used in the SLMCPR analysis is conservative relative to the required SLMCPR design basis power distribution and bounds actual operation.

Reference:

SRXB-A.42.1 SRXB-A.42.2 Correspondence, W.D. Crouch (TVA) to U.S. Nuclear Regulatory Commission, "Browns Ferry Nuclear Plant (BFN) -Units 2 and 3, Response to NRC Round 3 Requests for Additional Information Related to Technical Specifications (TS) Change No. TS-418 -Requests for Extended Power Uprate Operation (TAC Nos. MC3743 and MC3744)," March 7, 2006 (M L060680583). ANF-524(P)(A) Revision 2 and Supplements 1 and 2, ANF Critical Power Methodology for Boiling Water Reactors, Advanced Nuclear Fuels Corporation, November 1990 E2-4 NON-PROPRIETARY INFORMATION 1.45 1.4* Design Step-Through

• Actual Operation 1.35 1.3 U. 1.25 0-a 1.2 1.15 1.1 1.05 250 300 350 400 0 50 100 150 200 Assembly Rank Figure SRXB-A.42.1 Design vs.Actual Radial Power Factors E2-5 is u "' 1.45 r 1.4 r --135 K-1.3 " -NON-PROPRIETARY IN F ORMATION
• Design Step-Through <> Actual Operat io n u.. 1.25 +---'-' ...

c;; 1.2 =s "' a

1.15 +--------------

-"""11 .. -------1.1 --1.05 f-==----------- o 50 100 150 200 Assembly Ran k 250 Figure SRXB-A.42.1 Design vs. Actual Radial Power Factors E2-5 300 350 400 NON-PROPRIETARY INFORMATION On July 17, 2008, NRC issued a Round 18 RAI (ML081700102) regarding AREVA fuel methods used in support of Units 2 and 3 EPU operations. Round 18 consists of 32 RAI questions, SRXB-91 through SRXB-122. To facilitate the review of TS-418, a meeting was held on August 7, 2008 with NRC staff to review draft responses to SRXB-91 through SRXB-1 16.Subsequently, on August 15, 2008, TVA submitted a partial response (ML082330187) to Round 18; specifically to RAIs SRXB-92, 93, 95, 96, 97, 99, 100, and 102 through 116. Below are responses to the remainder of the Round 18 RAIs. Additionally, NRC staff conducted an audit of AREVA fuel methods from August 18 through August 28, 2008, at the AREVA engineering facilities in Richland, Washington. As a result of the audit, TVA agreed to provide supplemental responses to a number of the August 15, 2008, Round 18 RAI responses, whichare also provided below as indicated. NRC Introduction to Round 18 RAI Table 1.3 in Enclosure 5 to the letter dated June 25, 2004, indicates that the COTRANSA2 Version AAPR03 computer code was used to evaluate the anticipated transient without scram (ATWS) -overpressurization event. The licensee cites a May 31, 2000, letter from the Nuclear Regulatory Commission (NRC) to Framatome (now AREVA) to support the use of COTRANSA2 for the ATWS-overpressurization abnormal operating occurrence (AOO).NRC RAI SRXB-91 In Enclosure 1 of the letter dated March 7, 2006, Tennessee Valley Authority (TVA) provides information in support of the use of the Ohkawa-Lahey void quality correlation against ATRIUM-10 test data in response'to SRXB-A.35. The Ohkawa-Lahey void quality correlation appears to under-predict the void fraction for the majority of the thermodynamic qualities tested at 6.9 Megapascal (MPa). The void reactivity coefficient is sensitive to the instantaneous void fraction, generally becoming more negative with increasing void fraction.Provide a quantitative determination of the impact of the bias in the void fraction in COTRANSA2 on ATWS overpressure analysis results for the bottom head peak pressure. This should include a comparison of the impact of the void bias to the margin between the peak calculated pressure and the American Society of Mechanical Engineers Boiler & Pressure Vessel Code (ASME) acceptance criterion of 1500 pounds per square inch gage.In addition, address how known biases are taken into account for future cycle specific calculations and for bundle designs other than ATRIUM-1 0.Clarifications Provided by the NRC following a meeting on August 7, 2008Address the void bias for both the anticipated transient without scram (ATWS) overpressure as well as ASME overpressure. Response to SRXB-91 AREVA performs cycle-specific ATWS analyses of the short-term reactor vessel peak pressure using the COTRANSA2 computer code. The ATWS peak pressure calculation is a core wide pressurization event that is sensitive to similar phenomena as other pressurization transients. Bundle design is included in the development of input for the coupled neutronic and thermal E2-6 NON-PROPRIETARY INFORMATION On July 17,2008, NRC issued a Round 18 RAI (ML081700102) regarding AREVA fuel methods used in support of Units 2 and 3 EPU operations. Round 18 consists of 32 RAI questions, SRXB-91 through SRXB-122. To facilitate the review of TS-418, a meeting was held on August 7, 2008 with NRC staff to review draft responses to SRXB-91 through SRXB-116. Subsequently, on August 15, 2008, TVA submitted a partial response (ML082330187) to Round 18; specifically to RAls SRXB-92, 93, 95, 96, 97, 99, 100, and 102 through 116. Below are responses to the remainder of the Round 18 RAls. Additionally, NRC staff conducted an audit of AREVA fuel methods from August 18 through August 28, 2008, at the AREVA engineering facilities in Richland, Washington. As a result of the audit, TVA agreed to provide supplemental responses to a number of the August 15, 2008, Round 18 RAI responses, which are also provided below as indicated. NRC Introduction to Round 18 RAI Table 1.3 in Enclosure 5 to the letter dated June 25,2004, indicates that the COTRANSA2 Version AAPR03 computer code was used to evaluate the anticipated transient without scram (ATWS) -overpressurization event. The licensee cites a May 31,2000, letter from the Nuclear Regulatory Commission (NRC) to Framatome (now AREVA) to support the use of COTRANSA2 for the ATWS-overpressurization abnormal operating occurrence (AOO). NRC RAI SRXB-91 In Enclosure 1 of the letter dated March 7, 2006, Tennessee Valley Authority (TVA) provides information in support of the use of the Ohkawa-Lahey void quality correlation against ATRIUM-10 test data in response to SRXB-A.35. The Ohkawa-Lahey void quality correlation appears to under-predict the void fraction for the majority of the thermodynamic qualities tested at 6.9 Megapascal (MPa). The void reactivity coefficient is sensitive to the instantaneous void fraction, generally becoming more negative with increasing void fraction. Provide a quantitative determination of the impact of the bias in the void fraction in COTRANSA2 on ATWS overpressure analysis results for the bottom head peak pressure. This should include a comparison of the impact of the void bias to the margin between the peak calculated pressure and the American Society of Mechanical Engineers Boiler & Pressure Vessel Code (ASME) acceptance criterion of 1500 pounds per square inch gage. In addition, address how known biases are taken into account for future cycle specific calculations and for bundle designs other than ATRIUM-10. Clarifications provided by the NRC following a meeting on August 7. 2008 Address the void bias for both the anticipated transient without scram (ATWS) overpressure as well as ASME overpressure. Response to SRXB-91 AREVA performs cycle-specific ATWS analyses of the short-term reactor vessel peak pressure using the COTRANSA2 computer code. The ATWS peak pressure calculation is a core wide pressurization event that is sensitive to similar phenomena as other pressurization transients. Bundle design is included in the development of input for the coupled neutronic and thermal E2-6 NON-PROPRIETARY INFORMATION hydraulic COTRANSA2 core model. Important inputs to the COTRANSA2 system model are biased in a conservative direction. The AREVA analysis methods and the correlations used by the methods are applicable for both pre-EPU and EPU conditions as discussed in responses (ML060680583) to previous RAIs (SRXB-A.15, SRXB-A.26 through SRXB-A.29, and SRXB-A.35). The transient analysis methodology is a deterministic bounding approach that contains sufficient conservatism to offset biases and uncertainties in individual phenomena. For bundle designs other than ATRIUM-10, the void-quality correlation is robust as discussed in the response (ML082330187) to RAI SRXB-93 for past and present fuel designs. For future fuel designs, the void-quality correlation would be reviewed for applicability, which may involve additional verification and validation. A sensitivity study was performed for the limiting ATWS pressurization event for BFN Unit 3 Cycle 14 with EPU to assess the bias between the ATRIUM-10 test data and the void-quality correlation. The event was a pressure regulator failure-open (PRFO), which is a depressurization event, followed by pressurization due to main steam line isolation valve (MSIV)closure. The neutronics input included the impact of the fuel depleted with the changes in the void-quality correlation. To remove the bias in the MICROBURN-B2 neutronics input, the[ ] void-quality correlation was modified. To address the bias in the Ohkawa-Lahey void-quality correlation for the COTRANSA2 code, the void-quality relationship was changed to a [ ]. Additionally, the sensitivity study was repeated without depleting the fuel with the changes in the void-quality correlation (the change in the void-quality correlation was instantaneous at the exposure of interest). The reference ATWS case had a peak vessel pressure of 1477 pounds per square inch gauge (psig). The change in the void-quality correlations resulted in a 10-psi increase in the peak vessel pressure. The results for an instantaneous change in the void-quality correlation showed the same impact. A study was also performed for the ASME overpressure event for BFN Unit 3 Cycle 14 with EPU. The event was the MSIV closure with flux scram. The change in the void-quality correlations resulted in a 7 psi increase in the peak vessel pressure. The impact of a change in the bias of the void-quality correlations on peak pressure is expected to be more than offset by the model conservatisms. However, until quantitative values of the conservatisms can be demonstrated, TVA has directed AREVA to include a 10-psi increase to the peak vessel pressure for the EPU ATWS overpressure analysis and a 7-psi increase to the peak vessel pressure for the EPU ASME overpressure analysis.NRC RAI SRXB-94 The initial steam flow rate at extended power uprate (EPU) conditions is higher than at pre-EPU conditions, and the transient power pulse is expected to be higher during the pressurization. The suppression pool temperature for Units 2 and 3 is based on an analysis for GE14 fuel.Provide a discussion on the means used to confirm that the results of the GE 14 analysis are bounding for ATRIUM-10 fuel. This justification should contain qualitative discussion regarding the impact of the differences in nuclear characteristics and should consider the timing and nature of the transient power response during pressurization, relief, and boration.Response to SRXB-94 The higher initial steam flow at EPU conditions will result in a slightly higher power pulse during the initial relatively short pressurization phase of the ATWS event. However, the total energy released to the suppression pool is dominated by the later much longer phase of the event where power is reduced after the recirculation pumps trip and the core power is slowly reduced E2-7 NON-PROPRIETARY INFORMATION hydraulic COTRANSA2 core model. Important inputs to the COTRANSA2 system model are biased in a conservative direction. The AREVA analysis methods and the correlations used by the methods are applicable for both pre-EPU and EPU conditions as discussed in responses (ML060680583) to previous RAls (SRXB-A.15, SRXB-A.26 through SRXB-A.29, and SRXB-A.35). The transient analysis methodology is a deterministic bounding approach that contains sufficient conservatism to offset biases and uncertainties in individual phenomena. For bundle designs other than ATRIUM-10, the void-quality correlation is robust as discussed in the response (ML082330187) to RAI SRXB-93 for past and present fuel designs. For future fuel designs, the void-quality correlation would be reviewed for applicability, which may involve additional verification and validation. A sensitivity study was performed for the limiting ATWS pressurization event for BFN Unit 3 Cycle 14 with EPU to assess the bias between the ATRIUM-1 0 test data and the void-quality correlation. The event was a pressure regulator failure-open (PRFO), which is a depressurization event, followed by pressurization due to main steam line isolation valve (MSIV) closure. The neutronics input included the impact of the fuel depleted with the changes in the void-quality correlation. To remove the bias in the MICROBURN-B2 neutronics input, the [ ] void-quality correlation was modified. To address the bias in the Ohkawa-Lahey void-quality correlation for the COTRANSA2 code, the void-quality relationship was changed to a [ ]. Additionally, the sensitivity study was repeated without depleting the fuel with the changes in the void-quality correlation (the change in the void-quality correlation was instantaneous at the exposure of interest). The reference ATWS case had a peak vessel pressure of 1477 pounds per square inch gauge (psig). The change in the void-quality correlations resulted in a 10-psi increase in the peak vessel pressure. The results for an instantaneous change in the void-quality correlation showed the same impact. A study was also performed for the ASME overpressure event for BFN Unit 3 Cycle 14 with EPU. The event was the MSIV closure with flux scram. The change in the void-quality correlations resulted in a 7 psi increase in the peak vessel pressure. The impact of a change in the bias of the void-quality correlations on peak pressure is expected to be more than offset by the model conservatisms. However, until quantitative values of the conservatisms can be demonstrated, TVA has directed AREVA to include a 10-psi increase to the peak vessel pressure for the EPU ATWS overpressure analysis and a 7 -psi increase to the peak vessel pressure for the EPU ASME overpressure analysis. NRC RAI SRXB*94 The initial steam flow rate at extended power uprate (EPU) conditions is higher than at pre-EPU conditions, and the transient power pulse is expected to be higher during the pressurization. The suppression pool temperature for Units 2 and 3 is based on an analysis for GE14 fuel. Provide a discussion on the means used to confirm that the results of the GE 14 analysis are bounding for ATRIUM-1 0 fuel. This justification should contain qualitative discussion regarding the impact of the differences in nuclear characteristics and should consider the timing and nature of the transient power response during pressurization, relief, and boration. Response to SRXB*94 The higher initial steam flow at EPU conditions will result in a slightly higher power pulse during the initial relatively short pressurization phase of the ATWS event. However, the total energy released to the suppression pool is dominated by the later much longer phase of the event where power is reduced after the recirculation pumps trip and the core power is slowly reduced E2-7 NON-PROPRIETARY INFORMATION as boron injection occurs. The ATWS analyses performed for BFN Units 2 and 3 included the impact of the higher initial steam flow at EPU conditions. As shown in Table 9-4 of Reference SRXB-94.1, the impact of EPU operation on the maximum suppression pool temperature is not significant (<1 OF). This supports the conclusion that the initial power pulse, which is higher for EPU operation, is not significant relative to the total energy transferred to the suppression pool.The suppression pool temperature analyses were performed for BFN Units 2 and 3 with GE fuel (Reference SRXB-94.1). An evaluation was performed to compare fuel neutronic parameters important for the ATWS analysis (void coefficient, boron worth) for ATRIUM-1 0 and GE fuel.The boron worth characteristics of ATRIUM-1 0 were slightly better while the void reactivity characteristics were slightly worse relative to the impact on the ATWS suppression pool temperature analysis.Additional analyses were performed to assess the impact of the difference in fuel assembly reactivity characteristics on the suppression pool temperature during an ATWS. [I All fuel types in the core designs including the GE fuel were explicitly modeled in the above analyses consistent with the approved methodology. The GE fuel was modeled with a level of detail equivalent to that used for the ATRIUM-10 fuel. CASMO-4 analyses explicitly modeled the water rod configuration of the GE fuel. MICROBURN-B2 was used to calculate the core reactivity characteristics provided to the COTRANSA2 analysis. The GE fuel assemblygeometric and nuclear characteristics (enrichment and gadolinia distribution) were based on design data provided to AREVA by TVA. The hydraulic characteristics for the GE fuel assemblies were based on GE fuel assembly pressure drop tests performed by AREVA.The BFN ATWS analyses described above were performed for cycles operating at pre-EPU power levels. However, as shown in Table 9-4 of Reference SRXB-94.1, the impact of EPU operation on the maximum suppression pool temperature is not significant. Therefore, thetrends observed for ATRIUM-10 fuel in the above analyses are equally applicable for EPU operation. The analyses described above confirm that the suppression pool temperature analysis documented in Reference SRXB-94.1 is slightly conservative for ATRIUM-10 fuel. In addition, the analyses show that the difference in reactivity characteristics between ATRIUM-1 0 and GE fuel do not have a significant impact relative to the large margin to the suppression pool temperature limit shown in Reference SRXB-94.1. The conclusions of the Reference SRXB-94.1 suppression pool temperature analysis are applicable for ATRIUM-1 0 fuel and the acceptance criteria will be met for BFN Units 2 and 3 EPU operation with ATRIUM-10 fuel. E2-8 NON-PROPRIETARY INFORMATION as boron injection occurs. The ATWS analyses performed for BFN Units 2 and 3 included the impact of the higher initial steam flow at EPU conditions. As shown in Table 9-4 of Reference SRXB-94.1, the impact of EPU operation on the maximum suppression pool temperature is not significant <<1°F). This supports the conclusion that the initial power pulse, which is higher for EPU operation, is not significant relative to the total energy transferred to the suppression pool. The suppression pool temperature analyses were performed for BFN Units 2 and 3 with GE fuel (Reference SRXB-94.1). An evaluation was performed to compare fuel neutronic parameters important for the ATWS analysis (void coefficient, boron worth) for ATRIUM-10 and GE fuel. The boron worth characteristics of ATRIUM-10 were slightly better while the void reactivity characteristics were slightly worse relative to the impact on the ATWS suppression pool temperature analysis. Additional analyses were performed to assess the impact of the difference in fuel assembly reactivity characteristics on the suppression pool temperature during an ATWS. [ ] A" fuel types in the core designs including the GE fuel were explicitly modeled in the above analyses consistent with the approved methodology. The GE fuel was modeled with a level of detail equivalent to that used for the ATRIUM-10 fuel. CASMO-4 analyses explicitly modeled the water rod configuration of the GE fuel. MICROBURN-B2 was used to calculate the core reactivity characteristics provided to the COTRANSA2 analysis. The GE fuel assembly geometric and nuclear characteristics (enrichment and gadolinia distribution) were based on design data provided to AREVA by TVA. The hydraulic characteristics for the GE fuel assemblies were based on GE fuel assembly pressure drop tests performed by AREVA. The BFN ATWS analyses described above were performed for cycles operating at pre-EPU power levels. However, as shown in Table 9-4 of Reference SRXB-94.1, the impact of EPU operation on the maximum suppression pool temperature is not significant. Therefore, the trends observed for ATRIUM-10 fuel in the above analyses are equally applicable for EPU operation. The analyses described above confirm that the suppression pool temperature analysis documented in Reference SRXB-94.1 is slightly conservative for ATRIUM-10 fuel. In addition, the analyses show that the difference in reactivity characteristics between ATRIUM-1 0 and GE fuel do not have a significant impact relative to the large margin to the suppression pool temperature limit shown in Reference SRXB-94.1. The conclusions of the Reference SRXB-94.1 suppression pool temperature analysis are applicable for ATRIUM-10 fuel and the acceptance criteria wi" be met for BFN Units 2 and 3 EPU operation with ATRIUM-10 fuel. E2-8 NON-PROPRIETARY INFORMATION SRXB-94.1 NEDC-33047P Revision 2, Browns Ferry Units 2 and 3 Safety Analysis Report for Extended Power Uprate, June 2004. (ML041840301) Table SRXB-94.1 Energy Release to Suppression Pool NRC RAI SRXB-98 It appears that COTRANSA2 has two centrifugal pump models, the first pump model neglects the inertia and the second pump model is based on homologous input. Identify which model option is used. If the second model option is used, verify that it is used to model the dual recirculation pump trip during ATWS evaluations. Verify that the homologous input for the recirculation pumps for the Unit 2 analyses have been benchmarked against operational data at Unit 2.Response to SRXB-98 The second pump model based on homologous input is used to model the dual recirculation pump trip during ATWS evaluations. The homologous curves are from the pump manufacturer.The pump speed and flow are initialized from operational plant data. Frictional torque and pump inertia are tuned to model the plant coastdown rate.E2-9 NON-PROPRIETARY INFORMATION SRXB-94.1 NEDC-33047P Revision 2, Browns Ferry Units 2 and 3 Safety Analysis Report for Extended Power Uprate, June 2004. (ML041840301) Table SRXB-94.1 Energy Release to Suppression Pool [ ] NRC RAI SRXB-98 It appears that COTRANSA2 has two centrifugal pump models, the first pump model neglects the inertia and the second pump model is based on homologous input. Identify which model option is used. If the second model option is used, verify that it is used to model the dual recirculation pump trip during ATWS evaluations. Verify that the homologous input for the recirculation pumps for the Unit 2 analyses have been benchmarked against operational data at Unit 2. Response to SRXB-98 The second pump model based on homologous input is used to model the dual recirculation pump trip during ATWS evaluations. The homologous curves are from the pump manufacturer. The pump speed and flow are initialized from operational plant data. Frictional torque and pump inertia are tuned to model the plant coastdown rate. E2-9 NON-PROPRIETARY INFORMATION NRC RAI SRXB-100 Section 2.1 of ANF-913(P)(A) states that cross sections are interpolated based on both controlled and uncontrolled states at [[ ]] void fraction.These void cases appear to not be consistent with the void cases used to develop cross section response surfaces for MICROBURN-B2 [[ ]], explain this discrepancy. Supplemental Response to SRXB-100 In order to produce the COTRAN neutronic parameters, a series of MICROBURN-B2 calculations are performed. These successive calculations are: (1) Nominal initial conditions (2)E2-10 NON-PROPRIETARY INFORMATION NRC RAI SRXB-100 Section 2.1 of ANF-913(P)(A) states that cross sections are interpolated based on both controlled and uncontrolled states at [[ ]] void fraction. These void cases appear to not be consistent with the void cases used to develop cross section response surfaces for MICROBURN-B2 [[ ]] , explain this discrepancy. Supplemental Response to SRXB-100 In order to produce the COTRAN neutronic parameters, a series of MICROBURN-B2 calculations are performed. These successive calculations are: (1) Nominal initial conditions (2) E2-10 NON-PROPRIETARY INFORMATION i The 11/2 energy group diffusion equation in steady-state can be written as':a2) eP 1+ +/- V fIN+ E-2. V"f 2 (I), 2:a2=0 keff The first term is a leakage.following figure.This equation is integrated over the cylindrical node depicted in the H H H 01j+1 fý Dr"+1 j 1 1 D 1 ,i Dl 1 ,i-1 The leakage term is approximated as: 3 2DI,iDI,j(0 1 ,i- l,j) A j=1 (DO,i + OD,j) HV E2-11 NON-PROPRIETARY INFORMATION ] The 1 Y2 energy group diffusion equation in steady-state can be written as The first term is a leakage. This equation is integrated over the cylindrical node depicted in the following figure. H H H The leakage term is approximated as: 3 2D] .0] .(r!J]. -r!J] -) A _ L ,I,} ,I,} __ j=] (Dl,i + D],j) HV E2-11 NON-PROPRIETARY INFORMATION where D 1 ,i = D for plane of interest Dj = D for the nodes adjacent to the plane of interest 01,i = flux in the plane of interest 014j = flux in the regions adjacent to the plane of interest A = surface area between nodes i and j H = distance between nodes i and nodes j V = node volume E2-12 where [ NON-PROPRI ETARY INFORMATION Do = 0 for plane of interest 01,j = 0 for the nodes adjacent to the plane of interest = flux in the plane of interest ([J1,i C/J1,j = A flux in the regions adjacent to the plane of interest H V = surface area between nodes i and j = distance between nodes i and nodes j = node volume E2-12 NON-PROPRIETARY INFORMATION E2-13 NON-PROPRIETARY INFORMATION E2-13 NON-PROPRIETARY INFORMATION E2-14 NON-PROPRIETARY INFORMATION I , E2-14 NON-PROPRIETARY INFORMATION These final one-group cross section and leakage parameters are used in a new 1-dimensional flux solution and the axial power distribution is updated for the next thermal hydraulic solution.Iterations between the 1-dimensional flux solution and the thermal hydraulic solution are repeated until converged results are obtained for core power, power distribution, temperature distribution, and density distribution. E2-15 NON-PROPRIETARY INFORMATION These final one-group cross section and leakage parameters are used in a new 1-dimensional flux solution and the axial power distribution is updated for the next thermal hydraulic solution. Iterations between the 1-dimensional flux solution and the thermal hydraulic solution are repeated until converged results are obtained for core power, power distribution, temperature distribution, and density distribution. E2-15 NON-PROPRIETARY INFORMATION r..Figure SRXB-100.1 Comparison of Scram Bank Worth for[I E2-16 NON-PROPRIETARY INFORMATION r .J Figure SRXB-100.1 Comparison of Scram Bank Worth for [ ] E2-16 NON-PROPRIETARY INFORMATION NRC RAI SRXB-101 The Doppler coefficient is stated to be dependent on the broadening of the fast group cross section and to be a function of fuel temperature.

• MICROBURN-B2 calculates the nodal fuel temperature based on quadratic fitting function.Provide this function.

Discuss how the initial nodal fuel temperature is calculated. Provide a comparison of the quadratic function predicted nodal fuel temperature to results predicted using a more sophisticated thermal rod conduction model and heat transfer coefficient, such as XCOBRA-T.* Expand on the discussion provided in ANF-913(P)(A) and describe what combination of calculations is performed to determine the reactivity contribution from Doppler for ATWS overpressure analysis, for example, specify if a lattice calculation is performed to determine a coefficient relating microscopic cross sections to average fuel temperature.

• Discuss whether the rod temperatures in Section 2.1.3 of ANF-913(P)(A) are calculated based on a nodal average rod or for each rod in the node. Clarify how the transient nodal average fuel temperature is calculated.
• Provide a description of any differences between the COTRANSA2 thermal conduction models, including material properties, and the RODEX2 models. Discuss whether the RODEX2 code was used to develop input for COTRANSA2 similar to XCOBRA-T.Response to SRXB-101 E2-17[ NON-PROPRIETARY INFORMATION NRC RAJ SRXB-101 The Doppler coefficient is stated to be dependent on the broadening of the fast group cross section and to be a function of fuel temperature.
• MICROBURN-B2 calculates the nodal fuel temperature based on quadratic fitting function.

Provide this function. Discuss how the initial nodal fuel temperature is calculated. Provide a comparison of the quadratic function predicted nodal fuel temperature to results predicted using a more sophisticated thermal rod conduction model and heat transfer coefficient, such as XCOBRA-T.

• Expand on the discussion provided in ANF-913(P)(A) and describe what combination of calculations is performed to determine the reactivity contribution from Doppler for ATWS overpressure analysis, for example, specify if a lattice calculation is performed to determine a coefficient relating microscopic cross sections to average fuel temperature.
• Discuss whether the rod temperatures in Section 2.1.3 of ANF-913(P)(A) are calculated based on a nodal average rod or for each rod in the node. Clarify how the transient nodal average fuel temperature is calculated.
• Provide a description of any differences between the COTRANSA2 thermal conduction models, including material properties, and the RODEX2 models. Discuss whether the RODEX2 code was used to develop input for COTRANSA2 similar to XCOBRA-T.

Response to SRXB-101 E2-17 NON-PROPRIETARY INFORMATION E2-18 NON-PROPRIETARY INFORMATION E2-18 NON-PROPRIETARY INFORMATION E2-19 NON-PROPRI ETARY INFORMATION E2-19 NON-PROPRIETARY INFORMATION I The RODEX2 computer code provides initial input information relative to core average fuel-to-cladding gap heat transfer coefficients for the COTRANSA2 computer code. As such, RODEX2 uses steady-state heat conduction models. The heat conduction model employed by COTRANSA2 includes transient terms. The fuel thermal conductivity correlations used by COTRANSA2 are equivalent to the RODEX2 models.COTRANSA2 computes a fuel temperature for each axial plane in the core. Based onthe assumption of a core composition primarily consisting of uranium dioxide, COTRANSA2 does not account for gadolinium in the fuel thermal conductivity calculation. Heat capacities of fuel components (uranium dioxide, gadolinium, and cladding) are not required for the RODEX2 steady-state calculations, but are used in the COTRANSA2 transient calculations. The fuel pellet-to-cladding gap heat transfer coefficient used in COTRANSA2 is the product of a RODEX2 calculation. E2-20 NON-PROPRIETARY INFORMATION

• The RODEX2 computer code provides initial input information relative to core average fuel-to-cladding gap heat transfer coefficients for the COTRANSA2 computer Gode. As such, RODEX2 uses steady-state heat conduction models. The heat conduction model employed by COTRANSA2 includes transient terms. The fuel thermal conductivity correlations used by COTRANSA2 are equivalent to the RODEX2 models. COTRANSA2 computes a fuel temperature for each axial plane in the core. Based on the assumption of a core composition primarily consisting of uranium dioxide, COTRANSA2 does not account for gadolinium in the fuel thermal conductivity calculation.

Heat capacities of fuel components (uranium dioxide, gadolinium, and cladding) are not required for the RODEX2 steady-state calculations, but are used in the COTRANSA2 transient calculations. The fuel pellet-to-cladding gap heat transfer coefficient used in COTRANSA2 is the product of a RODEX2 calculation. E2-20 NON-PROPRIETARY INFORMATION Figure SRXB-101.1 RODEX Evolution of the Doppler Effective Fuel Temperature for SPC Fuel at Constant Power-E2-21 r NON-PROPRIETARY INFORMATION Figure SRX8-101.1 RODEX Evolution of the Doppler Effective Fuel Temperature for SPC Fuel at Constant Power* E2-21 ..J NON-PROPRIETARY INFORMATION Figure SRXB-101.2 RODEX Evolution of the Doppler Effective Fuel Temperature for SPC Fuel vs. LHGR and Burnup E2-22 r NON-PROPRI ETARY INFORMATION Figure SRXB-101.2 RODEX Evolution of the Doppler Effective Fuel Temperature for SPC Fuel vs. LHGR and Burnup E2-22 ..J NON-PROPRIETARY INFORMATION Figure SRXB-101.3 MICROBURN-B2 Correlation Evolution of the Doppler Effective Fuel Temperature for SPC Fuel vs. LHGR and Burnup E2-23 r NON-PROPRIETARY INFORMATION Figure SRXB-101.3 MICROBURN-B2 Correlation Evolution of the Doppler Effective Fuel Temperature for SPC Fuel vs. LHGR and Burnup E2-23 NON-PROPRIETARY INFORMATION NRC RAI SRXB-103 Provide the relationship of the term Feff to the S-factor. If axial integration is required to determine the S-factors, specify how this is performed. Address whether the S-factors are sensitive to the bundle void distribution. Describe how the S-factors are determined for conditions typical (or bounding) for operation at EPU conditions. Supplemental Response to SRXB-103 Evaluations were performed to assess the impact on ACPR of a change in Feff resulting from the variation in the lattice void fraction during a pressurization event. MICROBURN-B2 analyses were performed using the nominal void correlation and an adjusted void correlation to assess the change in Feff as void changes. The MICROBURN-B2 cases were run to reflect aninstantaneous change in core average void fraction of +0.05. For the limiting MCPR bundle in the core, the changes in void, local peaking factor (LPF), and Feff were: Avoid = +0.0441 (node 24)Avoid = +0.0456 (node 23)ALPF = -0.0026 (node 24)ALPF = -0.0030 (node 23)AFeff = 0.0000 (assembly) For other potentially limiting bundles (10% highest powered bundles) in the core, the change in Feff was between -0.0002 and +0.0011 for a +0.05 core average Avoid. In general, an increase in void fraction resulted in an increase in Feff for high power, low exposure (end of first cycle)assemblies and a decrease in Feff for low power, high exposure assemblies. A decrease in Feff during the transient will improve the CPR during the transient and result in a reduced ACPR. The converse is true for an increase in Feff during the transient. The sensitivityof MCPR to Feff is about 2 to 1; therefore, the sensitivity of ACPR is about twice the AFeff during the transient. The change in ACPR would be between 0.000 and +0.002 for a +0.05 core average Avoid.During a pressurization event, the core void will initially decrease followed by an increase in core void. Therefore, the effect of the change in void on fuel rod peaking factors (and Feff) will tend to be offset during the transient. The assessment above for the impact of a void change on AFeff and A(ACPR) is based on assuming the nuclear power is instantly converted to surface heat flux. Because the time of MCPR (-1.25 sec) is less than the fuel rod thermal time constant (- 5 sec), the actual impact on Feff and ACPR from the void change will be much less. At the boiling transition plane, there is an insignificant change in void until after the time of peak power. Because the increase in void and the corresponding increase in Feff occur close to the time of MCPR, the slight change in rod power will not significantly change the rod heat flux at the time of MCPR. Therefore, the effecton ACPR will be much less than estimated based on the MICROBURN-B2 analyses.In summary, the above results show that the effect of the variation in void fraction during a transient on the Feff has an insignificant effect on ACPR.E2-24 r NON-PROPRI ETARY INFORMATION NRC RAI SRXB-103 Provide the relationship of the term Feff to the S-factor. If axial integration is required to determine the S-factors, specify how this is performed. Address whether the S-factors are sensitive to the bundle void distribution. Describe how the S-factors are determined for conditions typical (or bounding) for operation at EPU conditions. Supplemental Response to SRXB-103 Evaluations were performed to assess the impact on of a change in Feff resulting from the variation in the lattice void fraction during a pressurization event. MICROBURN-B2 analyses were performed using the nominal void correlation and an adjusted void correlation to assess the change in Feff as void changes. The MICROBURN-B2 cases were run to reflect an instantaneous change in core average void fraction of +0.05. For the limiting MCPR bundle in the core, the changes in void, local peaking factor (LPF), and Feff were: = +0.0441 (node 24) = +0.0456 (node 23) = -0.0026 (node 24) = -0.0030 (node 23) = 0.0000 (assembly) For other potentially limiting bundles (10% highest powered bundles) in the core, the change in Feff was between -0.0002 and +0.0011 for a +0.05 core average In general, an increase in void fraction resulted in an increase in Feff for high power, low exposure (end of first cycle) assemblies and a decrease in Feff for low power, high exposure assemblies. A decrease in Feff during the transient will improve the CPR during the transient and result in a reduced The converse is true for an increase in Feff during the transient. The sensitivity of MCPR to Feff is about 2 to 1; therefore, the sensitivity of is about twice the during the transient. The change in would be between 0.000 and +0.002 for a +0.05 core average During a pressurization event, the core void will initially decrease followed by an increase in core void. Therefore, the effect of the change in void on fuel rod peaking factors (and F eff) will tend to be offset during the transient. The assessment above for the impact of a void change on and is based on assuming the nuclear power is instantly converted to surface heat flux. Because the time of MCPR (-1.25 sec) is less than the fuel rod thermal time constant (-5 sec), the actual impact on Feff and from the void change will be much less. At the boiling transition plane, there is an insignificant change in void until after the time of peak power. Because the increase in void and the corresponding increase in Feff occur close to the time of MCPR, the slight change in rod power will not significantly change the rod heat flux at the time of MCPR. Therefore, the effect on will be much less than estimated based on the MICROBURN-B2 analyses. In summary, the above results show that the effect of the variation in void fraction during a transient on the Feff has an insignificant effect on E2-24 NON-PROPRIETARY INFORMATION NRC RAI SRXB-105 Verify that the Unit 2 transient analyses were performed using input options for closure relationships that are consistent with the NRC approval of XCOBRA-T. This includes specifying the Levy subcooled boiling model, the Martinelli-Nelson two phase friction multipliers, the two phase component loss multiplier, the wall viscosity model, and thermodynamic properties from the ASME steam tables.Revised Response to SRXB-105 The BFN Units 2 and 3 EPU transient analyses used the default models of XCOBRA-T. The default models include the Levy subcooled boiling model, the Martinelli-Nelson two phase friction multipliers, the two phase component loss multiplier, and the heated wall viscosity correction model. [] as discussed in a meeting with the NRC on May 4,1995, (Reference SRXB-105.1). Thermodynamic properties from the ASME steam tables were used. The code provides a message if the default models are not used. Per AREVA's licensing analyses requirements, use of default models is required.

Reference:

SRXB-105.1 Correspondence, R.A. Copeland (Siemens) to R.C. Jones (NRC), "ATRIUM-10 Presentations," RAC:95:080, May 4, 1995 (38-9091703-000). NRC RAI SRXB-107 Address how the wall friction and component loss coefficients were determined for Unit 2.Address whether these parameters were input in the analysis to account for friction. Provide these parameters and the technical basis for their selection. Relative to pre-EPU conditions, channel flow tends to redistribute at EPU conditions as there are fewer low resistance bundles in the core. Address whether the friction parameters were selected to be consistent with this expected trend.Supplemental Response to SRXB-107 During the NRC audit of AREVA codes and methods in Richland, Wa., from August 18 through August 28, 2008, the NRC requested additional information regarding the background and process that [Spacer Pressure Drop Testing The Portable Hydraulic Test Facility (PHTF) is used by AREVA to obtain single phase loss coefficients for the spacers. The friction factor correlation is based on previous tests performed at the PHTF that remain applicable for current fuel designs (rods and channel have a consistent surface condition). The pressure drops across the spacers are measured in the PHTF for each new fuel design. The PHTF has pressure taps just upstream of the spacers so that the flow will be fully developed. The component of pressure drop due to friction is calculated and subtracted from the total measured pressure drop. The remaining pressure drop is due to the spacers and is used to determine the spacer pressure loss coefficients. E2-25 NON-PROPRIETARY INFORMATION NRC RAI SRXB-105 Verify that the Unit 2 transient analyses were performed using input options for closure relationships that are consistent with the NRC approval of XCOBRA-T. This includes specifying the Levy subcooled boiling model, the Martinelli-Nelson two phase friction multipliers, the two phase component loss multiplier, the wall viscosity model, and thermodynamic properties from the ASME steam tables. Revised Response to SRXB-105 The BFN Units 2 and 3 EPU transient analyses used the default models of XCOBRA-T. The default models include the Levy subcooled boiling model, the Martinelli-Nelson two phase friction multipliers, the two phase component loss multiplier, and the heated wall viscosity correction model. [ ] as discussed in a meeting with the NRC on May 4,1995, (Reference SRXB-1 05.1). Thermodynamic properties from the ASME steam tables were used. The code provides a message if the default models are not used. Per AREVA's licensing analyses requirements, use of default models is required.

Reference:

SRXB-105.1 Correspondence, R.A. Copeland (Siemens) to R.C. Jones (NRC), "ATRIUM-10 Presentations," RAC:95:080, May 4, 1995 (38-9091703-000). NRC RAI SRXB-107 Address how the wall friction and component loss coefficients were determined for Unit 2. Address whether these parameters were input in the analysis to account for friction. Provide these parameters and the technical basis for their selection. Relative to pre-EPU conditions, channel flow tends to redistribute at EPU conditions as there are fewer low resistance bundles in the core. Address whether the friction parameters were selected to be consistent with this expected trend. Supplemental Response to SRXB-107 During the NRC audit of AREVA codes and methods in Richland, Wa., from August 18 through August 28,2008, the NRC requested additional information regarding the background and process that [ ]. Spacer Pressure Drop Testing The Portable Hydraulic Test Facility (PHTF) is used by AREVA to obtain single phase loss coefficients for the spacers. The friction factor correlation is based on previous tests performed at the PHTF that remain applicable for current fuel designs (rods and channel have a consistent surface condition). The pressure drops across the spacers are measured in the PHTF for each new fuel design. The PHTF has pressure taps just upstream of the spacers so that the flow will be fully developed. The component of pressure drop due to friction is calculated and subtracted from the total measured pressure drop. The remaining pressure drop is due to the spacers and is used to determine the spacer pressure loss coefficients. E2-25 NON-PROPRIETARY INFORMATION Preliminary ATRIUM-10 Spacer Loss Coefficients Development of the ATRIUM-10 fuel design took place in Germany. Because PHTF pressure drop testing was not complete, single phase pressure drop data for ATRIUM-10 was obtained from the German development effort. For the use in preliminary ATRIUM-10 design assessments, the German data was used to develop single phase spacer pressure loss coefficients appropriate for use with Richland hydraulic models. Analyses using these singlephase losses resulted in an under prediction of the pressure drop data as shown in Figure SRXB-107.1. The spacer loss coefficients (K) used to generate the results presented in Figure SRXB-1 07.1 are of the form K=A +BReC where A, B, and C are constants and Re is the Reynolds number based on local fluid conditionsand geometry. Until PHTF data was available for the ATRIUM-10 design, a means of adjusting the German-based pressure loss coefficients to better predict the pressure drop data using Richland methods was developed. [] are shown in Figure SRXB-107.2. The spacer loss coefficients (K) used to generate the results presented in Figure SRXB-107.2 are of the form where [ ] for the ATRIUM-10 design.Further development of ATRIUM-1 0 spacer loss coefficients was subsequently performed based on PHTF ATRIUM-10 pressure drop data.PHTF ATRIUM-10 Based Spacer Loss Coefficients The ATRIUM-10 PHTF tests form the basis for the single phase loss coefficients currently used for design and licensing analyses supporting domestic BWRs. PHTF data was reduced to determine single phase losses for the spacers in the lower (fully-rodded) region of the bundle, the spacer in the transition (end of part-length rods) region of the bundle, and the spacers in the upper (partially-rodded) region of the bundle.Assessments of the predicted pressure drop relative to measured two phase pressure drop data confirmed the applicability of the [ ] for use with spacer pressure loss coefficients based on PHTF data. Results of analyses for each region of the bundle (lower, transition, upper) when using the PHTF spacer loss coefficients [ ] are shown in Figures SRXB-107.3, SRXB-107.4, and SRXB-107.5. NRC Interactions On May 4, 1995, a meeting was held with the NRC to describe the ATRIUM-10 design and the application of the approved AREVA methodology for the design. Two view graphs extracted from those presented at the meeting are provided in Figures SRXB-107.6 and SRXB-1 07.7.A summary of the May 4, 1995 meeting was provided to the NRC in Reference SRXB-107.1. E2-26 NON-PROPRIETARY INFORMATION Preliminary ATRIUM-10 Spacer Loss Coefficients Development of the ATRIUM-10 fuel design took place in Germany. Because PHTF pressure drop testing was not complete, single phase pressure drop data for ATRIUM-10 was obtained from the German development effort. For the use in preliminary ATRIUM-10 design assessments, the German data was used to develop single phase spacer pressure loss coefficients appropriate for use with Richland hydraulic models. Analyses using these single phase losses resulted in an under prediction of the pressure drop data as shown in Figure SRXB-1 07.1. The spacer loss coefficients (K) used to generate the results presented in Figure SRXB-107.1 are of the form K=A + B Re c where A, B, and C are constants and Re is the Reynolds number based on local fluid conditions and geometry. Until PHTF data was available for the ATRIUM-10 design, a means of adjusting the German-based pressure loss coefficients to better predict the pressure drop data using Richland methods was developed. [ ] are shown in Figure SRXB-107.2. The spacer loss coefficients (K) used to generate the results presented in Figure SRXB-107.2 are of the form [ ] where [ ] for the ATRIUM-10 design. Further development of ATRIUM-10 spacer loss coefficients was subsequently performed based on PHTF ATRIUM-10 pressure drop data. PHTF ATRIUM-10 Based Spacer Loss Coefficients The ATRIUM-10 PHTF tests form the basis for the single phase loss coefficients currently used for design and licensing analyses supporting domestic BWRs. PHTF data was reduced to determine single phase losses for the spacers in the lower (fully-rodded) region of the bundle, the spacer in the transition (end of part-length rods) region of the bundle, and the spacers in the upper (partially-rodded) region of the bundle. Assessments of the predicted pressure drop relative to measured two phase pressure drop data confirmed the applicability of the [ ] for use with spacer pressure loss coefficients based on PHTF data. Results of analyses for each region of the bundle (lower, transition, upper) when using the PHTF spacer loss coefficients [ ] are shown in Figures SRXB-107.3, SRXB-107.4, and SRXB-107.5. NRC Interactions On May 4,1995, a meeting was held with the NRC to describe the ATRIUM-10 design and the application of the approved AREVA methodology for the design. Two view graphs extracted from those presented at the meeting are provided in Figures SRXB-107.6 and SRXB-107.7. A summary of the May 4, 1995 meeting was provided to the NRC in Reference SRXB-107.1. E2-26 NON-PROPRIETARY INFORMATION Applicability for EPU Operation The ATRIUM-10 hydraulic models have been verified over a range of conditions that bound both pre-EPU and EPU operating conditions. The applicability of the models is described and supported by data presented in the thermal hydraulics section of the response to RAI SRXB-A.15 (Reference SRXB-107.2).

References:

SRXB-107.1 Correspondence, R.A. Copeland (Siemens) to R.C. Jones (NRC), "ATRIUM-10 Presentations," RAC:95:080, May 4, 1995 (38-9091703-000). SRXB-107.2 Correspondence, W.D. Crouch (TVA) to U.S. Nuclear Regulatory Commission,"Browns Ferry Nuclear Plant (BFN) -Units 2 and 3, Response to NRC Round 3 Requests for Additional Information Related to Technical Specifications (TS)Change No. TS-418 -Requests for Extended Power Uprate Operation (TAC Nos.MC3743 and MC3744)," March 7, 2006 (ML060680583). E2-27 NON-PROPRIETARY INFORMATION Applicability for EPU Operation The ATRIUM-10 hydraulic models have been verified over a range of conditions that bound both pre-EPU and EPU operating conditions. The applicability of the models is described and supported by data presented in the thermal hydraulics section of the response to RAI SRXB-A.15 (Reference SRXB-1 07.2).

References:

SRXB-107.1 Correspondence, R.A. Copeland (Siemens) to R.C. Jones (NRC), "ATRIUM-10 Presentations," RAC:95:080, May 4, 1995 (38-9091703-000). SRXB-107.2 Correspondence, w.o. Crouch (TVA) to U.S. Nuclear Regulatory Commission, "Browns Ferry Nuclear Plant (BFN) -Units 2 and 3, Response to NRC Round 3 Requests for Additional Information Related to Technical Specifications (TS) Change No. TS-418 -Requests for Extended Power Uprate Operation (TAC Nos. MC3743 and MC3744)," March 7, 2006 (ML060680583). E2-27 NON-PROPRIETARY INFORMATION Figure SRXB-107.1 ATRIUM-10 Bundle Pressure Drop I I E2-28 NON-PROPRIETARY INFORMATION r .J Figure SRXB-107.1 ATRIUM-10 Bundle Pressure Drop [ ] E2-28 NON-PROPRIETARY INFORMATION Figure SRXB-107.2 ATRIUM-10 Bundle Pressure Drop[]E2-29 NON-PROPRIETARY INFORMATION r ..J Figure SRXB-107.2 ATRIUM-10 Bundle Pressure Drop [ . ] E2-29 NON-PROPRIETARY INFORMATION Figure SRXB-107.3 ATRIUM-10 Lower Region Spacer Pressure Drop Using PHTF Loss Coefficients I I E2-30 NON-PROPRIETARY INFORMATION r Figure SRXB-107.3 ATRIUM-10 Lower Region Spacer Pressure Drop Using PHTF Loss Coefficients [ ] E2-30 .J NON-PROPRIETARY INFORMATION Figure SRXB-107.4 ATRIUM-10 Transition Region Spacer Pressure Drop Using PHTF Loss Coefficients I I E2-31 NON-PROPRIETARY INFORMATION r ..J Figure SRXB-107.4 ATRIUM-10 Transition Region Spacer Pressure Drop Using PHTF Loss Coefficients [ ] E2-31 NON-PROPRIETARY INFORMATION Figure SRXB-107.5 ATRIUM-10 Upper Region Spacer Pressure Drop Using PHTF Loss Coefficient I I E2-32 NON-PROPRIETARY INFORMATION r Figure SRXB-107.5 ATRIUM-10 Upper Region Spacer Pressure Drop Using PHTF Loss Coefficient [ ] E2-32 ..J NON-PROPRIETARY INFORMATION r-Figure SRXB-107.6 Viewgraph From May 4,1995 Presentation to NRC Regarding ATRIUM-10 Fuel r-Figure SRXB-107.7 Viewgraph From May 4, 1995 Presentation to NRC Regarding ATRIUM-10 Fuel rn-i E2-33 r r NON-PROPRIETARY INFORMATION Figure SRXB-107.6 Viewgraph From May 4,1995 Presentation to NRC Regarding ATRIUM-10 Fuel Figure SRXB-107.7 Viewgraph From May 4,1995 Presentation to NRC Regarding ATRIUM-10 Fuel E2-33 ..J NON-PROPRIETARY INFORMATION NRC RAI SRXB-108 At EPU conditions there are a higher number of higher powered bundles. It is possible, and likely, for large axial sections of these bundles to be in an annular flow regime. Calculating pressure losses near bundle features such as fuel spacers can be important in the prediction of critical heat flux, which tends to occur below fuel spacers where the liquid film is typically thinnest.On page 25 of Exxon Nuclear Company's XN-NF-84-105(P)(A), XCOBRA-T: A Computer Code for BWR Transient Thermal-Hydraulic Core Analysis, it is stated that "[t]his [Martinelli-Nelson] formulation was developed for horizontal flow, but is reasonably accurate for vertical flow where both phasic flow rates are high enough to ensure turbulent co-current flow." Justify why the Martinelli-Nelson two phase friction multipliers are applicable in annular flow regimes.Supplemental Response to SRXB-108 When applying the XCOBRA-T two phase pressure drop models implemented in the 1-dimensional hydraulic model of the COTRANSA2 code, the local (spacer grid) pressure losses are automatically adjusted to preserve the core pressure drop predicted by the more detailed 3-dimensional hydraulic representation in MICROBURN-B2. The XCOBRA-T initial flow rate is defined by a hydraulic demand curve predicted by XCOBRA, which defines the relationship between assembly power and the initial flow rate and accounts for the lack of a core bypass model in XCOBRA-T.The orifice loss coefficient is automatically adjusted in XCOBRA-T to preserve the COTRANSA2 (and MICROBURN-B2) initial core pressure drop and the initial flow rate defined by the hydraulic demand curve. Therefore, adjustments made to the local (spacer grid) pressure losses in COTRANSA2 appear in the adjustments to the orifice loss coefficient in XCOBRA-T.The hydraulic channel nodalization of each code is discussed in the previous response to RAI SRXB-1 15 (ML082330187). NRC RAI SRXB-109 Section 3.3 of the Technical Evaluation Report attached to the NRC's safety evaluation approving XN-NF-84-105(P)(A) states that critical power calculations may be inaccurate if the inlet flow is negative or if the inlet quality is above zero. Verify that for the transient analyses that the bundle inlet flow is positive and that the inlet qualities are less than zero.Supplemental Response to SRXB-109 The transient code, XCOBRA-T, evaluates Reynolds number for each node for each step of the calculation. If the flow becomes negative at any node, the code stops the calculation. NRC RAI SRXB-112 Some models may have been updated to conservatively bound experimental data collected subsequent to the NRC review and approval of RODEX2. The staff notes that certain assumptions may be conservative in the assessment of linear heat generation rate limits that may not be conservative when evaluating transient heat flux during AOO simulation due to the E2-34 NON-PROPRIETARY INFORMATION NRC RAI SRXB-108 At EPU conditions there are a higher number of higher powered bundles. It is possible, and likely, for large axial sections of these bundles to be in an annular flow regime. Calculating pressure losses near bundle features such as fuel spacers can be important in the prediction of critical heat flux, which tends to occur below fuel spacers where the liquid film is typically thinnest. On page 25 of Exxon Nuclear Company's XN-NF-84-105(P)(A), XCOBRA-T: A Computer Code for BWR Transient Thermal-Hydraulic Core Analysis, it is stated that "[t]his [Martinelli-Nelson] formulation was developed for horizontal flow, but is reasonably accurate for vertical flow where both phasic flow rates are high enough to ensure turbulent co-current flow." Justify why the Martinelli-Nelson two phase friction multipliers are applicable in annular flow regimes. Supplemental Response to SRXB-108 When applying the XCOBRA-T two phase pressure drop models implemented in the 1-dimensional hydraulic model of the COTRANSA2 code, the local (spacer grid) pressure losses are automatically adjusted to preserve the core pressure drop predicted by the more detailed 3-dimensional hydraulic representation in MICROBURN-B2. The XCOBRA-T initial flow rate is defined by a hydraulic demand curve predicted by XCOBRA, which defines the relationship between assembly power and the initial flow rate and accounts for the lack of a core bypass model in XCOBRA-T. The orifice loss coefficient is automatically adjusted in XCOBRA-T to preserve the COTRANSA2 (and MICROBURN-B2) initial core pressure drop and the initial flow rate defined by the hydraulic demand curve. Therefore, adjustments made to the local (spacer grid) pressure losses in COTRANSA2 appear in the adjustments to the orifice loss coefficient in XCOBRA-T. The hydraulic channel nodalization of each code is discussed in the previous response to RAI SRXB-115 (ML082330187). NRC RAI SRXB-109 Section 3.3 of the Technical Evaluation Report attached to the NRC's safety evaluation approving XN-NF-84-1 05(P)(A) states that critical power calculations may be inaccurate if the inlet flow is negative or if the inlet quality is above zero. Verify that for the transient analyses that the bundle inlet flow is positive and that the inlet qualities are less than zero. Supplemental Response to SRXB-109 The transient code, XCOBRA-T, evaluates Reynolds number for each node for each step of the calculation. If the flow becomes negative at any node, the code stops the calculation. NRC RAI SRXB-112 Some models may have been updated to conservatively bound experimental data collected subsequent to the NRC review and approval of RODEX2. The staff notes that certain assumptions may be conservative in the assessment of linear heat generation rate limits that may not be conservative when evaluating transient heat flux during AOO simulation due to the E2-34 NON-PROPRIETARY INFORMATION competing effects of reactivity feedback and heat flux flow mismatch. If a model is"conservatively bounding" in RODEX2, and translated to XCOBRA-T, provide a discussion ofthe performance of the model for thermal margin transient calculations. Clarifications Provided by the NRC following a meeting on August 7, 2008 The draft response for SRXB-1 12 deals with changes to the RODEX2 code in its first part, but requests additional information regarding the use of conservative assumptions in the abnormal operating occurrence (AOO) transient response. The discussion regarding the conservatism of the gap properties should be addressed in the response to the second part of RAI 112. See the second and third sentences:The staff notes that certain assumptions may be conservative in the assessment of linear heat generation rate limits that may not be conservative when evaluating transient heat flux during AOO simulation due to the competing effects of reactivity feedback and heat flux/flow mismatch. If a model is "conservatively bounding" in RODEX2, and translated to XCOBRA-T, provide a discussion of the performance of the model for thermal margin transient calculations. Summary of staff concern: The NRC staff considered the coupling of the neutron flux and fluid conditions for AOO transient evaluations for both a reduced thermal time constant and an increased thermal time constant.When the time constant is over predicted, the fluid response to changing neutron power is lagged. A pressurization transient, therefore, would result in an increase in the reactor power that is not impeded by subsequent rapid void formation due to hold up of the heat flux in the pellet. An over prediction of the time constant will tend to increase the fission power for such a transient. However, the same effect of holding the heat up in the fuel pellet has the dual effect of reducing the cladding heat flux response; therefore, the ultimate effect on the transient critical power ratio (CPR) is a combination of the conservative prediction of peak neutron flux with the non-conservative prediction of the transient cladding heat flux.For the case where the time constant is under predicted the inverse is true, the gross reactor power increase due to pressurization is limited due to more rapid void formation in response tothe increasing neutron flux, but this is countered by a prediction of higher cladding surface heat flux relative to the pin power throughout the transient. The input assumptions regarding the gas gap may increase or decrease the thermal resistance, and similarly, an increase or decrease in the thermal resistance does not have a clear impact on the transient predicted CPR due to competing effects in the cladding heat flux and void reactivity. Supplemental Response to SRXB-112 A gap conductance sensitivity study was performed for the 100% power/1 05% flow BFN load rejection with no bypass (LRNB) transient event from Reference SRXB-1 12.1. The purpose of the sensitivity study was to show the ACPR trend for changes in gap conductance for COTRANSA2 versus XCOBRA-T. The gap conductance change considered was [ I The results are provided in Table SRXB-1 12.1. As seen from the results, an increase in COTRANSA2 core average gap conductance results in a decrease in ACPR; whereas an increase in XCOBRA-T gap hot channel conductance results in an increase in ACPR. A E2-35 NON-PROPRIETARY INFORMATION competing effects of reactivity feedback and heat flux flow mismatch. If a model is "conservatively bounding" in RODEX2, and translated to XCOBRA-T, provide a discussion of the performance of the model for thermal margin transient calculations. Clarifications provided bv the NRC following a meeting on August 7. 2008 The draft response for SRXB-112 deals with changes to the RODEX2 code in its first part, but requests additional information regarding the use of conservative assumptions in the abnormal operating occurrence (AOO) transient response. The discussion regarding the conservatism of the gap properties should be addressed in the response to the second part of RAI 112. See the second and third sentences: The staff notes that certain assumptions may be conservative in the assessment of linear heat generation rate limits that may not be conservative when evaluating transient heat flux during AOO simulation due to the competing effects of reactivity feedback and heat flux/flow mismatch. If a model is "conservatively bounding" in RODEX2, and translated to T, provide a discussion of the performance of the model for thermal margin transient calculations. Summary of staff concern: The NRC staff considered the coupling of the neutron flux and fluid conditions for AOO transient evaluations for both a reduced thermal time constant and an increased thermal time constant. When the time constant is over predicted, the fluid response to changing neutron power is lagged. A pressurization transient, therefore, would result in an increase in the reactor power that is not impeded by subsequent rapid void formation due to hold up of the heat flux in the pellet. An over prediction of the time constant will tend to increase the fission power for such a transient. However, the same effect of holding the heat up in the fuel pellet has the dual effect of reducing the cladding heat flux response; therefore, the ultimate effect on the transient critical power ratio (CPR) is a combination of the conservative prediction of peak neutron flux with the non-conservative prediction of the transient cladding heat flux. For the case where the time constant is under predicted the inverse is true, the gross reactor power increase due to pressurization is limited due to more rapid void formation in response to the increasing neutron flux, but this is countered by a prediction of higher cladding surface heat flux relative to the pin power throughout the transient. The input assumptions regarding the gas gap may increase or decrease the thermal resistance, and similarly, an increase or decrease in the thermal resistance does not have a clear impact on the transient predicted CPR due to competing effects in the cladding heat flux and void reactivity. Supplemental Response to SRXB-112 A gap conductance sensitivity study was performed for the 100% power/105% flow BFN load rejection with no bypass (LRNB) transient event from Reference SRXB-112.1. The purpose of the sensitivity study was to show the trend for changes in gap conductance for COTRANSA2 versus XCOBRA-T. The gap conductance change considered was [ ]. The results are provided in Table SRXB-112.1. As seen from the results, an increase in COTRANSA2 core average gap conductance results in a decrease in whereas an increase in XCOBRA-T gap hot channel conductance results in an increase in A E2-35 NON-PROPRIETARY INFORMATION decrease in gap conductance shows the opposite trend. The XCOBRA-T ATRIUM-10 hot channel model is slightly more sensitive to the change in gap conductance than the COTRANSA2 ATRIUM-1 0 average core model. When both COTRANSA2 and XCOBRA-T gap conductance are changed by an equivalent amount, the net impact is no significant change in ACPR.

Reference:

SRXB-1 12.1 EMF-2982(P) Revision 0, Browns Ferry Units 2 and 3 Safety Analysis Report for Extended Power Uprate A TRIUMTM-1O Fuel Supplement, Framatome ANP, June 2004.Table SRXB-112.1 Gap Conductance Study Increase in Gap Conductance Gap conductance condition A(ACPR)Core average[ ] -0.011 Hot channel [ ] +0.012 Core average and hot channel [ ] 0.000 Decrease in Gap Conductance Gap conductance condition A(ACPR)Core average[ ] +0.015 Hot channel [ ] -0.016 Core average and hot channel [ ] -0.001 NRC RAI SRXB-116 Address whether XCOBRA-T was used to demonstrate acceptable fuel rod thermal mechanical performance during transients. If XCOBRA-T is not used for this purpose, address how acceptable thermal mechanical performance is demonstrated during transients. If the method is not consistent with the models in RODEX2 or later NRC-approved thermal mechanical code, justify the approach. Clarifications Provided by the NRC following a meeting on August 7, 2008 Aside from describing the method for normalization of the transient LHGR to the initial LHGR, provide some additional minor clarifications: (1) The decay heat contribution will remain essentially static during the transient, addresswhether the normalization capture the varying rod decay heat sources;(2) Specify the source of the decay heat constants (i.e. ANS standard); E2-36 NON-PROPRIETARY INFORMATION decrease in gap conductance shows the opposite trend. The XCOBRA-T ATRIUM-10 hot channel model is slightly more sensitive to the change in gap conductance than the COTRANSA2 ATRIUM-10 average core model. When both COTRANSA2 and XCOBRA-T gap conductance are changed by an equivalent amount, the net impact is no significant change in flCPR.

Reference:

SRXB-112.1 EMF-2982(P) Revision 0, Browns Ferry Units 2 and 3 Safety Analysis Report for Extended Power Uprate ATRIUMTM-10 Fuel Supplement, Framatome ANP, June 2004. Table SRXB-112.1 Gap Conductance Study Increase in Gap Conductance Gap conductance condition fl(flCPR) Core average [ ] -0.011 Hot channel [ ] +0.012 Core average and hot channel [ ] 0.000 Decrease in Gap Conductance Gap conductance condition fl(flCPR) Core average [ ] +0.015 Hot channel [ ] -0.016 Core average and hot channel [ ] -0.001 NRC RAI SRXB-116 Address whether XCOBRA-T was used to demonstrate acceptable fuel rod thermal mechanical performance during transients. If XCOBRA-T is not used for this purpose, address how acceptable thermal mechanical performance is demonstrated during transients. If the method is not consistent with the models in RODEX2 or later NRC-approved thermal mechanical code, justify the approach. Clarifications provided bv the NRC following a meeting on August 7, 2008 Aside from describing the method for normalization of the transient LHGR to the initial LHGR, provide some additional minor clarifications: (1) The decay heat contribution will remain essentially static during the transient, address whether the normalization capture the varying rod decay heat sources; (2) Specify the source of the decay heat constants (i.e. ANS standard); E2-36 NON-PROPRIETARY INFORMATION (3) The rod power distribution is flattened due to gamma smearing of the thermal power, address how these gamma smeared power fractions are calculated; and (4) Address how the direct moderator heat is accounted for.The response should also provide a detailed description of the rod heat flux calculation for bundles with part length fuel rods, and address the code change as well as items 1-4 for each region (fully rodded, plena region, above plena region).Supplemental Response to SRXB-116 For bundles with part-length fuel rods (PLFRs), the rod heat flux calculation begins by computing the time-dependent heat flux generation rate at each axial section in the fuel rod.The updated equation corresponding to equation 2.130 of Reference SRXB-1 16.1 is: q"(t) = P(t) I (ff + fc )FriFli Fa 7FDrodj LNaNn where P(t) = transient reactor power ff = fraction of power produced in the fuel fý = fraction of power produced in the cladding Na = total number of assemblies in the core N,. = total number of heated rods for type i assembly at the axial plane F; = radial peaking factor of type i assembly F,, = local peaking factor of type i assembly Fa = axial peaking factor at the axial plane Drodj = fuel rod diameter of type i assembly L = axial heated length This equation differs from that in Reference SRXB-1 16.1 by replacing the initial reactor power in the denominator with Tr. In addition, the variable definitions have been modified to identify that the total number of heated rods is dependent on both the assembly type and axial elevation,and the definition of L has been corrected to the axial heated length of the assembly. This equation is substituted into equations 2.129a and 2.129b in Section 2.5.5 of Reference SRXB-1 16.1 to define the volumetric heat deposition rate for the fuel pellet and cladding, respectively. This volumetric heat deposition rate is used in the right hand side of equation 2.85 of Reference SRXB-1 16.1 to iteratively solve the transient heat conduction equation and the hydraulic conservation equations for the new time step temperatures and surface heat flux. The heat flux is introduced into the channel energy equation (2.2 of Reference SRXB-1 16.1) through the term q'. This linear heat deposition rate is a summation of the energy added by direct energy deposition and surface heat flux: J= P(t) fcoolFriFa+Hsurf"(TNodesT-Tfluid)'%'Drod,i"Nri Ni q'(t) 1 NaL I E2-37 NON-PROPRIETARY INFORMATION (3) The rod power distribution is flattened due to gamma smearing of the thermal power, address how these gamma smeared power fractions are calculated; and (4) Address how the direct moderator heat is accounted for. The response should also provide a detailed description of the rod heat flux calculation for bundles with part length fuel rods, and address the code change as well as items 1-4 for each region (fully rodded, plena region, above plena region). Supplemental Response to SRXB-116 For bundles with part-length fuel rods (PLFRs), the rod heat flux calculation begins by computing the time-dependent heat flux generation rate at each axial section in the fuel rod. The updated equation corresponding to equation 2.130 of Reference SRXB-116.1 is: where pet) f, fc Na Nri Fri F/i Fa Orad'; L = = = = = = = = = = transient reactor power fraction of power produced in the fuel fraction of power produced in the cladding total number of assemblies in the core total number of heated rods for type i assembly at the axial plane radial peaking factor of type i assembly local peaking factor of type i assembly axial peaking factor at the axial plane fuel rod diameter of type i assembly axial heated length This equation differs from that in Reference SRXB-116.1 by replacing the initial reactor power in the denominator with TT. In addition, the variable definitions have been modified to identify that the total number of heated rods is dependent on both the assembly type and axial elevation, and the definition of L has been corrected to the axial heated length of the assembly. This equation is substituted into equations 2.129a and 2.129b in Section 2.5.5 of Reference SRXB-116.1 to define the volumetric heat deposition rate for the fuel pellet and cladding, respectively. This volumetric heat deposition rate is used in the right hand side of equation 2.85 of Reference SRXB-116.1 to iteratively solve the transient heat conduction equation and the hydraulic conservation equations for the new time step temperatures and surface heat flux. The heat flux is introduced into the channel energy equation (2.2 of Reference SRXB-116.1 ) through the term q'. This linear heat deposition rate is a summation of the energy added by direct energy deposition and surface heat flux: , {P(t) } q (t) = --fcool FriFa + Hsurf . (TNodesT -Tf/uid)*ff* Drad i . Nri Ni NL ' a E2-37 NON-PROPRIETARY INFORMATION where f.0 1o = fraction of power produced in the coolant HSurd = film heat transfer coefficient at the axial plane TNodesT = cladding surface temperature at the axial plane Tr7,ud = fluid temperature at the axial plane N, = number of fuel assemblies in channel i In addition to axially varying number of heated rods, proper modeling of PLFRs also requires axial variations in the active flow area, the heated perimeter, and the wetted perimeter, and these parameters are now defined as axially dependent quantities in AREVA methods.Consequently, all references to these parameters or parameters derived from the basic geometry data in the approved topical reports should be interpreted as being axially dependent variables. The pressure drop due to the area expansion at the end of the PLFRs (or anywhere in the active flow path) is modeled using the specific volume for momentum as expressed in equations 2.78 and 2.79 of Reference SRXB-1 16.1. For current designs, area contractions occur in the single phase region, but the coding was generalized to address area contractions inthe two-phase region based on a solution of the two phase Bernoulli equation.An XCOBRA-T deposited power fraction sensitivity study was performed for the 100%power/1 05% flow BFN LRNB transient event from Reference SRXB-1 16.2. The purpose of the sensitivity study was to show the impact on ACPR from using generic ATRIUM-1 0 power fractions versus case-specific power fractions. The case-specific power fractions are used in COTRANSA2 and are obtained from CASMO-4/MICROBURN-B2. AREVA is in the process of automating the transfer of the case-specific power fractions into XCOBRA-T such that the generic values will no longer be used. [] The power that would have been deposited []. A review of an ATRIUM-10 power deposition study showed that the []. A study was performed by taking [ ]. The results are provided in Table SRXB-1 16.1. The study shows no significant change in ACPR. [] This study demonstrates that the ATRIUM-10 generic power fractions in XCOBRA-T are adequate.

References:

SRXB-116.1 XN-NF-84-105(P)(A) Volume 1 and Volume 1 Supplements 1 and 2, XCOBRA-T: A Computer Code for BWR Transient Thermal-Hydraulic Core Analysis, Exxon Nuclear Company, February 1987.SRXB-1 16.2 EMF-2982(P) Revision 0, Browns Ferry Units 2 and 3 Safety Analysis Report forExtended Power Uprate A TRIUMTM-1O Fuel Supplement, Framatome ANP, June 2004. (ML041840301) E2-38 NON-PROPRIETARY INFORMATION where fcool = fraction of power produced in the coolant Hsurf = film heat transfer coefficient at the axial plane TNodesT = cladding surface temperature at the axial plane Tf/uid = fluid temperature at the axial plane Ni = number of fuel assemblies in channel i In addition to axially varying number of heated rods, proper modeling of PLFRs also requires axial variations in the active flow area, the heated perimeter, and the wetted perimeter, and these parameters are now defined as axially dependent quantities in AREVA methods. Consequently, all references to these parameters or parameters derived from the basic geometry data in the approved topical reports should be interpreted as being axially dependent variables. The pressure drop due to the area expansion at the end of the PLFRs (or anywhere in the active flow path) is modeled using the specific volume for momentum as expressed in equations 2.78 and 2.79 of Reference SRXB-116.1. For current designs, area contractions occur in the single phase region, but the coding was generalized to address area contractions in the two-phase region based on a solution of the two phase Bernoulli equation. An XCOBRA-T deposited power fraction sensitivity study was performed for the 100% power/105% flow BFN LRNB transient event from Reference SRXB-116.2. The purpose of the sensitivity study was to show the impact on L\CPR from using generic ATRIUM-10 power fractions versus case-specific power fractions. The case-specific power fractions are used in COTRANSA2 and are obtained from CASM0-4/MICROBURN-B2. AREVA is in the process of automating the transfer of the case-specific power fractions into XCOBRA-T such that the generic values will no longer be used. [ ] The power that would have been deposited [ ]. A review of an ATRIUM-10 power deposition study showed that the [ ]. A study was performed by taking [ ]. The results are provided in Table SRXB-116.1. The study shows no significant change in L\CPR. [ ] This study demonstrates that the ATRIUM-10 generic power fractions in XCOBRA-T are adequate.

References:

SRXB-116.1 XN-NF-84-105(P)(A) Volume 1 and Volume 1 Supplements 1 and 2, XCOBRA-T: A Computer Code for BWR Transient Thermal-Hydraulic Core Analysis, Exxon Nuclear Company, February 1987. SRXB-116.2 EMF-2982(P) Revision 0, Browns Ferry Units 2 and 3 Safety Analysis Report for Extended Power Uprate ATRIUMTM-10 Fuel Supplement, Framatome ANP, June 2004. (ML041840301) E2-38 NON-PROPRIETARY INFORMATION Table SRXB-116.1 Deposited Heat Study Fuel Cladding Moderator Bypass Condition Heat Heat Heat Heat A(ACPR)Generic power fractions [ ] [ ] [ ] [ ] NA Case-specific power fractions [ ] [ ] [ ] [ ] -0.0004 Case-specific power fractions I ] ] [ ] [ ] [ ] +0.0008 E2-39 NON-PROPRIETARY INFORMATION Table SRXB-116.1 Deposited Heat Study Fuel Cladding Moderator Bypass Condition Heat Heat Heat Heat fl(flCPR) Generic power fractions [ ] [ ] [ ] [ ] NA Case-specific power fractions [ ] [ ] [ ] [ ] -0.0004 Case-specific power fractions [ ] [ ] [ ] [ ] [ ] +0.0008 E2-39 NON-PROPRIETARY INFORMATION NRC RAI SRXB-117 Enclosure 4 of the letter dated June 25, 2004, references NEDO-32047-A. In particular it is noted that operation at EPU conditions is generally achieved by flattening radial core power. As a result of this flattening the second harmonic eigenvalue separation is likely to be greatly reduced. Therefore, under non-isolation ATWS conditions it is expected that the core will be more susceptible to regional mode oscillations that at pre-EPU conditions. Given the information provided in the NRC's contractors' technical evaluation report attached to the safety evaluation approving NEDO-32047-A dated February 5, 1994, Appendix C: "Consequences of Out-of-Phase Instability Mode Not Proven More Favorable than In-Phase Mode." Provide an evaluation of the likelihood of a regional mode oscillation to develop under non-isolation ATWS conditions. It is acceptable to evaluate the regional and core wide mode decay ratios for these conditions for an equilibrium ATRIUM-10 Unit 2 core using STAIF to respond to this request for additional information (RAI). Based on the available analyses, determine if such an oscillation at BFN would result in a significant increase in the fuel damage relative to the results in NEDO-32047-A. The analyses in NEDO-32047-A were performed for General Electric (GE) fuel. The analyses are generally applicable for pre-EPU core designs since hydraulic stability of the fuel products has improved or at least remained the same. Provide a comparison of the channel stability characteristics of ATRIUM-10 to GE 8x8 fuel. If ATRIUM-10 is less stable than GE 8x8 fuel, consider any impact on the projected consequences of a non-isolation ATWS instability event.Response to SRXB-1 17 The pre-EPU stability analysis for BFN indicates that the global mode is dominant over the regional (out-of-phase) mode where relatively large subcritical reactivity values are calculated with STAIF. For EPU cores with flatter radial power distributions, the calculated subcritical reactivity values are noticeably lower in comparison. The resulting regional decay ratios calculated for the EPU core are larger than the corresponding global mode decay ratio in a minority of cases, which warrants the examination of the effect of regional mode oscillations dominating postulated ATWS instability events.The task of evaluating the impact of large regional versus global mode oscillations is first addressed below from an analytical point of view and calculations are presented using a reduced order model. The calculations will also address the effects of the parameters of interest, namely the subcritical reactivity due to core radial power flattening for EPU, increase in voidreactivity coefficient due to increasing the fresh fuel batch size, and fuel geometry effects(part-length rods and reduced pin conduction time constant for an ATRIUM-10 compared with an 8x8 fuel bundle). These effects will be demonstrated to result in equivalent consequences of a postulated ATWS event relative to the results in NEDO-32047-A (Reference SRXB-1 17.1).Furthermore, the mitigation of the ATWS instability by reducing the core inlet subcooling as a consequence of water level reduction by operator action (Reference SRXB-1 17.2) will be demonstrated to be as effective in suppressing regional mode oscillations as for global mode oscillations. Analytical Considerations Unstable global mode oscillations grow exponentially at a fixed rate (decay ratio) from a small perturbation. As the oscillation magnitude increases, nonlinear effects become important. The E2-40 NON-PROPRIETARY INFORMATION NRC RAI SRXB-117 Enclosure 4 of the letter dated June 25,2004, references NEDO-32047-A. In particular it is noted that operation at EPU conditions is generally achieved by flattening radial core power. As a result of this flattening the second harmonic eigenvalue separation is likely to be greatly reduced. Therefore, under non-isolation ATWS conditions it is expected that the core will be more susceptible to regional mode oscillations that at pre-EPU conditions. Given the information provided in the NRC's contractors' technical evaluation report attached to the safety evaluation approving NEDO-32047-A dated February 5, 1994, Appendix C: "Consequences of Out-of-Phase Instability Mode Not Proven More Favorable than In-Phase Mode." Provide an evaluation of the likelihood of a regional mode oscillation to develop under non-isolation ATWS conditions. It is acceptable to evaluate the regional and core wide mode decay ratios for these conditions for an equilibrium ATRIUM-10 Unit 2 core using STAIF to respond to this request for additional information (RAI). Based on the available analyses, determine if such an oscillation at BFN would result in a significant increase in the fuel damage relative to the results in NEDO-32047-A. The analyses in NEDO-32047-A were performed for General Electric (GE) fuel. The analyses are generally applicable for pre-EPU core designs since hydraulic stability of the fuel products has improved or at least remained the same. Provide a comparison of the channel stability characteristics of ATRIUM-10 to GE 8x8 fuel. If ATRIUM-10 is less stable than GE 8x8 fuel, consider any impact on the projected consequences of a non-isolation ATWS instability event. Response to SRXB-117 The pre-EPU stability analysis for BFN indicates that the global mode is dominant over the regional (out-of-phase) mode where relatively large subcritical reactivity values are calculated with STAIF. For EPU cores with flatter radial power distributions, the calculated subcritical reactivity values are noticeably lower in comparison. The resulting regional decay ratios calculated for the EPU core are larger than the corresponding global mode decay ratio in a minority of cases, which warrants the examination of the effect of regional mode oscillations dominating postulated ATWS instability events. The task of evaluating the impact of large regional versus global mode oscillations is first addressed below from an analytical point of view and calculations are presented using a reduced order model. The calculations will also address the effects of the parameters of interest, namely the subcritical reactivity due to core radial power flattening for EPU, increase in void reactivity coefficient due to increasing the fresh fuel batch size, and fuel geometry effects (part-length rods and reduced pin conduction time constant for an ATRIUM-10 compared with an 8x8 fuel bundle). These effects will be demonstrated to result in equivalent consequences of a postulated ATWS event relative to the results in NEDO-32047-A (Reference SRXB-117.1). Furthermore, the mitigation of the ATWS instability by reducing the core inlet subcooling as a consequence of water level reduction by operator action (Reference SRXB-117 .2) will be demonstrated to be as effective in suppressing regional mode oscillations as for global mode oscillations. Analytical Considerations Unstable global mode oscillations grow exponentially at a fixed rate (decay ratio) from a small perturbation. As the oscillation magnitude increases, nonlinear effects become important. The E2-40 NON-PROPRIETARY INFORMATION average power level drifts to higher values as a consequence of the nonlinearity of the neutron kinetics, which results in a negative reactivity feedback due to the increase of void fraction. The negative reactivity superimposed on the oscillating reactivity results in damping the neutron kinetics (References SRXB-117.3. SRXB-117.4, and SRXB-117.5). [The regional mode oscillations are well understood in the linear limit where the power oscillation is attributed to the excitation of the first azimuthal harmonic mode of the neutron flux.Compared with the fundamental flux mode excitation associated with the global oscillation, the subcritical reactivity of the first azimuthal eigenfunction contributes a damping effect on the neutron kinetics feedback. The hydraulic response is less damped compared to the global mode case due to bypassing the damping effects of the recirculation loop. The regional mode oscillations may become the preferred oscillation mode for large-orificed cores (hydraulic destabilization) and for small radial buckling (large core diameter and radial power distribution that is relatively flat or ring-of-fire with relatively low power in the center).E2-41[ NON-PROPRIETARY INFORMATION average power level drifts to higher values as a consequence of the nonlinearity of the neutron kinetics, which results in a negative reactivity feedback due to the increase of void fraction. The negative reactivity superimposed on the oscillating reactivity results in damping the neutron kinetics (References SRXB-117.3. SRXB-117.4, and SRXB-117.5). [ The regional mode oscillations are well understood in the linear limit where the power oscillation is attributed to the excitation of the first azimuthal harmonic mode of the neutron flux. Compared with the fundamental flux mode excitation associated with the global oscillation, the subcritical reactivity of the first azimuthal eigenfunction contributes a damping effect on the neutron kinetics feedback. The hydraulic response is less damped compared to the global mode case due to bypassing the damping effects of the recirculation loop. The regional mode oscillations may become the preferred oscillation mode for large-orificed cores (hydraulic destabilization) and for small radial buckling (large core diameter and radial power distribution that is relatively flat or ring-of-fire with relatively low power in the center). E2-41 NON-PROPRIETARY INFORMATION I Description of the Reduced Order Model The phenomenological description of large power oscillations in the global and regional modes is supported by the results of a reduced order model, which is used here to simulate large global and regional mode oscillations. [E2-42 NON-PROPRIETARY INFORMATION ] Description of the Reduced Order Model The phenomenological description of large power oscillations in the global and regional modes is supported by the results of a reduced order model, which is used here to simulate large global and regional mode oscillations. [ E2-42 NON-PROPRIETARY INFORMATION The reduced order model allows fast and robust simulation of both the global and regional modes and helps to resolve issues that were not apparent at the time NEDO-32047-A (Reference SRXB-117.1) was issued. Most importantly, it helps to explore and provide insight into the differences between the global and regional mode oscillations and their common ultimate limiting mechanism. Results The results of several cases performed with the reduced order model are presented. All of these calculations represent unstable oscillations growing to large magnitudes with parameter variations to address the issues of global versus regional and the effect of EPU core loading with fuel design differing from the fuel type used in NEDO-32047-A (Reference SRXB-1 17.1).These cases are: E2-43 NON-PROPRIETARY INFORMATION ] The reduced order model allows fast and robust simulation of both the global and regional modes and helps to resolve issues that were not apparent at the time NEDO-32047-A (Reference SRXB-117.1) was issued. Most importantly, it helps to explore and provide insight into the differences between the global and regional mode oscillations and their common ultimate limiting mechanism. Results The results of several cases performed with the reduced order model are presented. All of these calculations represent unstable oscillations growing to large magnitudes with parameter variations to address the issues of global versus regional and the effect of EPU core loading with fuel design differing from the fuel type used in NEDO-32047-A (Reference SRXB-117.1). These cases are: E2-43 NON-PROPRIETARY INFORMATION I Conclusions

• Large regional mode oscillations have [mode." ATRIUM-10 bundle design differences from an older 8x8 [I] effects compared with global EPU effects (lower subcritical reactivity and higher void reactivity coefficient)

[]* [I

References:

SRXB-1 17.1 SRXB-1 17.2SRXB-1 17.3SRXB-1 17.4SRXB-1 17.5 NEDO-32047-A, "ATWS Rule Issues Relative to BWR Core Thermal-Hydraulic Stability," June 1995.NEDO-32164 Revision 0, "Mitigation of BWR Core Thermal-Hydraulic Instabilities in ATWS," December 1992.Wulff, W., H. S. Cheng, A.N. Mallen, and U.S. Rohatgi, "BWR Stability Analysis with the BNL Engineering Plant Analyzer," NUREG/CR-5816, October 1992.March-Leuba, J., D.G. Cacuci, and R.B. Perez, "Nonlinear Dynamics andStability of Boiling Water Reactors: Part I -- Qualitative Analysis," Nuclear Science and Engineering: 93, 111-123 (1986).March-Leuba, J., "Density-Wave Instabilities in Boiling Water Reactors," NUREG/CR-6003, September 1992.E2-44 NON-PROPRIETARY INFORMATION ] Conclusions

• Large regional mode oscillations have [ mode.
• ATRIUM-10 bundle design differences from an older 8x8 [ ] ] effects compared with global
• EPU effects (lower subcritical reactivity and higher void reactivity coefficient)

[ ] *

References:

SRXB-117.1 NEDO-3204 7 -A, "A TWS Rule Issues Relative to BWR Core Thermal-Hydraulic Stability," June 1995. SRXB-117.2 NEDO-32164 Revision 0, "Mitigation of BWR Core Thermal-Hydraulic Instabilities in ATWS," December 1992. SRXB-117.3 Wulff, W., H. S. Cheng, A.N. Mallen, and U.S. Rohatgi, "BWR Stability Analysis with the BNL Engineering Plant Analyzer," NUREG/CR-5816, October 1992. SRXB-117.4 March-Leuba, J., D.G. Cacuci, and R.B. Perez, "Nonlinear Dynamics and Stability of Boiling Water Reactors: Part I --Qualitative Analysis," Nuclear Science and Engineering: 93,111-123 (1986). SRXB-117.5 March-Leuba, J., "Density-Wave Instabilities in Boiling Water Reactors," NUREG/CR-6003, September 1992. E2-44 NON-PROPRIETARY INFORMATION SRXB-1 17.6 SRXB-1 17.7 Farawila, Y.M., and D.W. Pruitt, "A Study of Nonlinear Oscillation and Limit Cycles in Boiling Water Reactors -I: The Global Mode," Nuclear Science and Engineering: 154, 302-315 (2006).Farawila, Y.M., and D.W. Pruitt, "A Study of Nonlinear Oscillation and Limit Cycles in Boiling Water Reactors -I1: The Regional Mode," Nuclear Science and Engineering: 154, 316-327 (2006).E2-45 NON-PROPRIETARY INFORMATION SRXB-117.6 Farawila, Y.M., and D.W. Pruitt, "A Study of Nonlinear Oscillation and Limit Cycles in Boiling Water Reactors -I: The Global Mode," Nuclear Science and Engineering: 154, 302-315 (2006). SRXB-117.7 Farawila, Y.M., and D.W. Pruitt, "A Study of Nonlinear Oscillation and Limit Cycles in Boiling Water Reactors -II: The Regional Mode," Nuclear Science and Engineering: 154,316-327 (2006). E2-45 NON-PROPRIETARY INFORMATION r-r..Figure SRXB-1 17.1.1 Relative Power for Case 1 Base Global Oscillation Figure SRXB-117.1.2 Relative Power for Case 2 Base Regional Oscillation E2-46 r r NON-PROPRIETARY INFORMATION Figure SRXB-117 .1.1 Relative Power for Case 1 Base Global Oscillation Figure SRXB-117 .1.2 Relative Power for Case 2 Base Regional Oscillation E2-46 ..J ..J NON-PROPRIETARY INFORMATION r-Figure SRXB-117.1.3 Relative Power for Case 3 Global Oscillation r-Figure SRXB-117.1.4 Relative Power for Case 4 Regional Oscillation E2-47 r r NON-PROPRIETARY INFORMATION Figure SRXB-117.1.3 Relative Power for Case 3 Global Oscillation Figure SRXB-117.1.4 Relative Power for Case 4 Regional Oscillation E2-47 .J NON-PROPRIETARY INFORMATION r-Figure SRXB-117.1.5 Relative Power for Case 5 Regional Oscillation With Decreased Subcriticality r-a-Figure SRXB-117.1.6 Relative Power for Case 6 Mitigated Global Oscillation E2-48 r r NON-PROPRIETARY INFORMATION Figure SRXB-117 .1.5 Relative Power for Case 5 Regional Oscillation With Decreased Subcriticality Figure SRXB-117 .1.6 Relative Power for Case 6 Mitigated Global Oscillation E2-48 .J .J NON-PROPRIETARY INFORMATION r-Figure SRXB-117.1.7 Relative Power for Case 7 Mitigated Regional Oscillation r" Figure SRXB-1 17.1.8 Relative Power for Case 8 Late-Mitigated Global Oscillation E2-49 r r NON-PROPRIETARY INFORMATION Figure SRXB-117 .1. 7 Relative Power for Case 7 Mitigated Regional Oscillation Figure SRXB-117 .1.8 Relative Power for Case 8 Late-Mitigated Global Oscillation E2-49 ..J NON-PROPRIETARY INFORMATION r-r-Figure SRXB-1 17.1.9 Relative Power for Case 9 Late-Mitigated Regional Oscillation Figure SRXB-117.2.1 Inlet Mass Flow Rate for Case I Base Global Oscillation E2-50 r r NON-PROPRIETARY INFORMATION Figure SRXB-117 .1.9 Relative Power for Case 9 Late-Mitigated Regional Oscillation Figure SRXB-117 .2.1 Inlet Mass Flow Rate for Case 1 Base Global Oscillation E2-50 .J ..J NON-PROPRIETARY INFORMATION r--U Figure SRXB-117.2.2 Inlet Mass Flow Rate for Case 2 Base Regional Oscillation r-J Figure SRXB-117.2.3 Inlet Mass Flow Rate for Case 3 Global Oscillation E2-51 r r NON-PROPRIETARY INFORMATION Figure SRXB-117.2.2 Inlet Mass Flow Rate for Case 2 Base Regional Oscillation Figure SRXB-117 .2.3 Inlet Mass Flow Rate for Case 3 Global Oscillation E2-51 r-NON-PROPRIETARY INFORMATION Figure SRXB-117.2.4 Inlet Mass Flow Rate for Case 4 Regional Oscillation r-Figure SRXB-117.2.5 Inlet Mass Flow Rate for Case 5 Regional Oscillation With Decreased Subcriticality E2-52 r r NON-PROPRIETARY INFORMATION Figure SRXB-117.2.4 Inlet Mass Flow Rate for Case 4 Regional Oscillation Figure SRXB-117.2.5 Inlet Mass Flow Rate for Case 5 Regional Oscillation With Decreased Subcriticality E2-52 ..J ..J NON-PROPRIETARY INFORMATION r-Figure SRXB-117.2.6 Inlet Mass Flow Rate for Case 6 Mitigated Global Oscillation r-Figure SRXB-117.2.7 Inlet Mass Flow Rate for Case 7 Mitigated Regional Oscillation E2-53 r r NON-PROPRIETARY INFORMATION Figure SRXB-117.2.6 Inlet Mass Flow Rate for Case 6 Mitigated Global Oscillation Figure SRXB-117.2.7 Inlet Mass Flow Rate for Case 7 Mitigated Regional Oscillation E2-53 ..J NON-PROPRIETARY INFORMATION r..rn-r Figure SRXB-117.2.8 Inlet Mass Flow Rate for Case 8 Late-Mitigated Global Oscillation Figure SRXB-117.2.9 Inlet Mass Flow Rate for Case 9 Late-Mitigated Regional Oscillation E2-54 r r NON-PROPRIETARY INFORMATION Figure SRXB-117.2.8 Inlet Mass Flow Rate for Case 8 Late-Mitigated Global Oscillation Figure SRXB-117.2.9 Inlet Mass Flow Rate for Case 9 Late-Mitigated Regional Oscillation E2-54 ..J NON-PROPRIETARY INFORMATION r.r..Figure SRXB-117.3.1 Exit Void Fraction for Case 1 Base Global Oscillation Figure SRXB-1 17.3.2 Exit Void Fraction for Case 2 Base Regional Oscillation E2-55 r r NON-PROPRIETARY INFORMATION Figure SRXB-117 .3.1 Exit Void Fraction for Case 1 Base Global Oscillation Figure SRXB-117.3.2 Exit Void Fraction for Case 2 Base Regional Oscillation E2-55 .J .J NON-PROPRIETARY INFORMATION r-r-Figure SRXB-1 17.3.3 Exit Void Fraction for Case 3 Global Oscillation Figure SRXB-117.3.4 Exit Void Fraction for Case 4 Regional Oscillation E2-56 r r NON-PROPRIETARY INFORMATION Figure SRXB-117 .3.3 Exit Void Fraction for Case 3 Global Oscillation Figure SRXB-117.3.4 Exit Void Fraction for Case 4 Regional Oscillation E2-56 ..J NON-PROPRIETARY INFORMATION r-rn-Figure SRXB-117.3.5 Exit Void Fraction for Case 5 Regional Oscillation With Decreased Subcriticality r" rn-Figure SRXB-1 17.3.6 Exit Void Fraction for Case 6 Mitigated Global Oscillation E2-57 r r NON-PROPRIETARY INFORMATION Figure SRXB-117.3.5 Exit Void Fraction for Case 5 Regional Oscillation With Decreased Subcriticality Figure SRXB-117.3.6 Exit Void Fraction for Case 6 Mitigated Global Oscillation E2-57 .J NON-PROPRIETARY INFORMATION r.r..Figure SRXB-1 17.3.7 Exit Void Fraction for Case 7 Mitigated Regional Oscillation Figure SRXB-1 17.3.8 Exit Void Fraction for Case 8 Late-Mitigated Regional Oscillation-d E2-58 r r NON-PROPRIETARY INFORMATION Figure SRXB-117.3.7 Exit Void Fraction for Case 7 Mitigated Regional Oscillation Figure SRXB-117.3.8 Exit Void Fraction for Case 8 Late-Mitigated Regional Oscillation E2-58 ..J NON-PROPRIETARY IN FORMATION r-r-Figure SRXB-117.3.9 Exit Void Fraction for Case 9 Late-Mitigated Regional Oscillation Figure SRXB-117.4.1 Void Fraction in Selected Nodes for Case 1 Base Global Oscillation E2-59 r r NON-PROPRIETARY INFORMATION Figure SRXB-117.3.9 Exit Void Fraction for Case 9 Late-Mitigated Regional Oscillation Figure SRXB-117.4.1 Void Fraction in Selected Nodes for Case 1 Base Global Oscillation E2-59 NON-PROPRIETARY INFORMATION r-Figure SRXB-117.4.2 Void Fraction in Selected Nodes for Case 2 Base Regional Oscillation r-Figure SRXB-117.4.3 Void Fraction in Selected Nodes for Case 3 Global Oscillation E2-60 r r NON-PROPRIETARY INFORMATION Figure SRXB-117.4.2 Void Fraction in Selected Nodes for Case 2 Base Regional Oscillation Figure SRXB-117.4.3 Void Fraction in Selected Nodes for Case 3 Global Oscillation E2-60 .J .J NON-PROPRIETARY INFORMATION r-Figure SRXB-117.4.4 Void Fraction in Selected Nodes for Case 4 Regional Oscillation r Figure SRXB-117.4.5 Void Fraction in Selected Nodes for Case 5 Regional Oscillation With Decreased Subcriticality E2-61 r NON-PROPRIETARY INFORMATION Figure SRXB-117.4.4 Void Fraction in Selected Nodes for Case 4 Regional Oscillation r Figure SRXB-117.4.5 Void Fraction in Selected Nodes for Case 5 Regional Oscillation With Decreased Subcriticality E2-61 ..J NON-PROPRIETARY INFORMATION NRC Introduction The following are related to the June 3, 2008 response to SRXB-88.NRC RAI SRXB-118 In the supplemental response to RAI SRXB-88, TVA provided the results of sensitivity analyses to evaluate the impact of void fraction uncertainty on the calculated delta-critical power ratio (DCPR) and the safety limit minimum critical power (SLMCPR). In the void fraction reduction case, the DCPR is apparently unaffected and is accompanied by an increase in SLMCPR.If the void fraction were reduced throughout the core by a fixed bias, the result would be to redistribute the reactor power according to the change in reactivity associated with the void perturbation. Since those bundles with the higher bundle average void fractions will have a greater reactivity response, a reduction in the void fraction will tend to increase, slightly, the power in those bundles with a higher bundle average void fraction relative to the bundles that had a lower void content prior to the perturbation. The bundles with a higher bundle average void fraction are the high powered bundles. Therefore, a fixed reduction in void fraction will increase the radial power peaking factor. The increased radial power peaking factor for a givensteady state power level would result in fewer rods entering boiling transition as a result of a transient initiated from this state.When this effect is considered, it is the equivalent of increasing the radial power peaking and reducing the SLMCPR since fewer rods are at the limiting end of the pin power statistical distribution. In effect, the span of pin powers to account for the 0.1 percent of highest powered pins increases. Results of the TVA sensitivity analysis demonstrate the opposite trend. It isexpected that the imposition of a fixed void fraction reduction would result in a lower SLMCPR.Explain this discrepancy. Response to SRXB-1 18 It should be noted that the sensitivity analyses presented in the SRXB-88 response were not based on "fixed" void fraction changes. Rather, the analyses were based on modifications to the void-quality correlation that resulted in a new nominal fit and offsets that were on average+0.05 void. The discussion above in the RAI question for SRXB-1 18 is based on a comparison of trends for an instantaneous change in void fraction. The RAI SRXB-88 response included the impact of the fuel depleted with the changes in the void-quality correlation. The difference in depletion changes the sensitivity of void friction modifications considerably due to the feedback of modified power distributions on exposure distribution. For the RAI SRXB-88 case, the change in the void-quality correlation was imposed over all fuel in the core from beginning-of-life. No changes were made to the fuel loading and rod patterns.The result of SRXB-88 was that a reduction in void resulted in more assemblies at higher power. The radial peaking factors of the high-powered assemblies that contributed to rods in boiling transition were slightly more "flat" and resulted in a slightly higher SLMCPR.Figure SRXB-1 18.1 shows the slight differences in radial distributions. The sensitivity analysis of SRXB-88 was repeated for an instantaneous change in voids. For an instantaneous change in voids, the SLMCPR trends were the same as SRXB-88; however, the change is small for both depleted and instantaneous void change, i.e., an SLMCPR change of-0.003 for +0.05 voids and +0.002 for -0.05 voids. The sensitivity can be explained by the E2-62 NON-PROPRIETARY INFORMATION NRC Introduction The following are related to the June 3, 2008 response to SRXB-88. NRC RAI SRXB-118 In the supplemental response to RAI SRXB-88, TVA provided the results of sensitivity analyses to evaluate the impact of void fraction uncertainty on the calculated delta-critical power ratio (DCPR) and the safety limit minimum critical power (SLMCPR). In the void fraction reduction case, the DCPR is apparently unaffected and is accompanied by an increase in SLMCPR. If the void fraction were reduced throughout the core by a fixed bias, the result would be to redistribute the reactor power according to the change in reactivity associated with the void perturbation. Since those bundles with the higher bundle average void fractions will have a greater reactivity response, a reduction in the void fraction will tend to increase, slightly, the power in those bundles with a higher bundle average void fraction relative to the bundles that had a lower void content prior to the perturbation. The bundles with a higher bundle average void fraction are the high powered bundles. Therefore, a fixed reduction in void fraction will increase the radial power peaking factor. The increased radial power peaking factor for a given steady state power level would result in fewer rods entering boiling transition as a result of a transient initiated from this state. When this effect is considered, it is the equivalent of increasing the radial power peaking and reducing the SLMCPR since fewer rods are at the limiting end of the pin power statistical distribution. In effect, the span of pin powers to account for the 0.1 percent of highest powered pins increases. Results of the TVA sensitivity analysis demonstrate the opposite trend. It is expected that the imposition of a fixed void fraction reduction would result in a lower SLMCPR. Explain this discrepancy. Response to SRXB-118 It should be noted that the sensitivity analyses presented in the SRXB-88 response were not based on "fixed" void fraction changes. Rather, the analyses were based on modifications to the void-quality correlation that resulted in a new nominal fit and offsets that were on average +/-0.05 void. The discussion above in the RAI question for SRXB-118 is based on a comparison of trends for an instantaneous change in void fraction. The RAI SRXB-88 response included the impact of the fuel depleted with the changes in the void-quality correlation. The difference in depletion changes the sensitivity of void friction modifications considerably due to the feedback of modified power distributions on exposure distribution. For the RAI SRXB-88 case, the change in the void-quality correlation was imposed over all fuel in the core from beginning-of-life. No changes were made to the fuel loading and rod patterns. The result of SRXB-88 was that a reduction in void resulted in more assemblies at higher power. The radial peaking factors of the high-powered assemblies that contributed to rods in boiling transition were slightly more "flat" and resulted in a slightly higher SLMCPR. Figure SRXB-118.1 shows the slight differences in radial distributions. The sensitivity analysis of SRXB-88 was repeated for an instantaneous change in voids. For an instantaneous change in voids, the SLMCPR trends were the same as SRXB-88; however, the change is small for both depleted and instantaneous void change, i.e., an SLMCPR change of -0.003 for +0.05 voids and +0.002 for -0.05 voids. The sensitivity can be explained by the E2-62 NON-PROPRIETARY INFORMATION small radial power distribution shifts in Figures SRXB-1 18.2 and SRXB-1 18.3. It is concludedthat the radial distribution is not significantly changed for the SLMCPR analysis; therefore, the impact of the prescribed void-quality correlation changes is insignificant on SLMCPR.It is very difficult to identify the expected direction of the radial power distribution change due to a modification of the void-quality correlation. In addition to the void coefficient dependency on void fraction, there is an even stronger dependency of the void coefficient on exposure. For the limiting case of SLMCPR the highest radial powers come from a range of assembly exposures. The importance of void changes in different assemblies of different exposures cannot be analyzed with simplified models and isolated trends.Independent of the trend, the analyses demonstrate insignificant impacts on SLMCPR.E2-63 NON-PROPRIETARY INFORMATION small radial power distribution shifts in Figures SRXB-118.2 and SRXB-118.3. It is concluded that the radial distribution is not significantly changed for the SLMCPR analysis; therefore, the impact of the prescribed void-quality correlation changes is insignificant on SLMCPR. It is very difficult to identify the expected direction of the radial power distribution change due to a modification of the void-quality correlation. In addition to the void coefficient dependency on void fraction, there is an even stronger dependency of the void coefficient on exposure. For the limiting case of SLMCPR the highest radial powers come from a range of assembly exposures. The importance of void changes in different assemblies of different exposures cannot be analyzed with simplified models and isolated trends. Independent of the trend, the analyses demonstrate insignificant impacts on SLMCPR. r E2-63 NON-PROPRIETARY INFORMATION 0.99° 0.98 0.97 0 z 0.96 0.95 0 100 Bundle Index 0.99 a. 0.980.97 0 z 0.96 0.95 Figure SRXB-118.1 SLMCPR Radial Power Distribution High-Powered Assemblies Depleted Voids'\. n\---Reference Modified...... Modified +0.05---Modified -0.05 0 100 Bundle Index Figure SRXB-118.2 SLMCPR Radial Power Distribution High-Powered Assemblies Instantaneous Voids E2-64 0.99 ; 0.98 OJ :;; ] OJ E 0.97 o z 0.96 E--Referen ce --Modified -----. Modifi e d +0.05 ----Modified -0.05 NON-PROPRIETARY INFORMATION 0.95 L-_____________________________ _____ 0.99 ; 0.98 OJ :;; " o 0.97 Z 0.96 \ \ Bundle Index Figure SRXB-118.1 SLMCPR Radial Power Distribution High-Powered Assemblies Depleted Voids

---

t-----Modified ------Modified +0.05 ----Modified -0.05 1---------------

---

100 0.95 I---------------------------------------l o Bundle Index Figure SRXB-118.2 SLMCPR Radial Power Distribution High-Powered Assemblies Instantaneous Voids E2-64 100 NON-PROPRIETARY INFORMATION 0.8 0.7 f 0.6 0 a.0.5 0.4 0 z 0.3 0.2 0.1---- Reference Modified------ Modified +0.05---- Modified -0.05 400 500 600 700 Bundle Index 800 Figure SRXB-118.3 SLMCPR Radial Power Distribution Low-Powered Assemblies Instantaneous Voids E2-65 NON-PROPRIETARY INFORMATION 0.8 ,---------------------------------------------------------------

--, 0.7 0.6 ! II. iii 0.5 '6 (i .., 04 .. . E o z 0.3 0.2 --Re f e r ence --Modi fi ed ------Modified +0.05 ----Modified -0.05 0.1 400 500 600 Bu ndl e Ind ex 700 Figure SRXB-118.3 SLMCPR Radial Power Distribution Low-Powered Assemblies Instantaneous Voids E2-65 800 NON-PROPRIETARY INFORMATION NRC RAI SRXB-119 Continuing with the void fraction reduction case, the decrease in void fraction would simultaneously result in a redistribution of the axial power. Since those higher void nodes would have a greater reactivity response than low void nodes, the axial power distribution would shift upwards in the core. The upward shift in the axial power distribution has the effect of increasing the reactor adjoint in the upper portions of the core. As pressurization transients are typically limiting, the impact of an upward shift in axial power on the transient power prediction should be considered. The upward shift in reactor adjoint directly affects the core void reactivity coefficient and tends to increase the sensitivity of the core reactivity to a pressure wave, since the back pressure wave is dissipated by void collapse in the upper parts of the core. Therefore, the core wide transient power would be increased relative to the base case, which appears to result in an increase in the DCPR.The results of the TVA sensitivity analysis do not demonstrate this trend. Address why imposing a fixed void fraction reduction does not result in a higher DCPR.Response to SRXB-1 19 The trend described above in the RAI question for SRXB-1 19 is for an instantaneous change in voids. As discussed in the previous response to SRXB-1 18, the results of the SRXB-88 sensitivity analyses were based on fuel depleted with the changes in the void-quality correlations. AREVA concurs with the general trend as described above for an instantaneous change in void. The analysis of SRXB-88 was repeated for an instantaneous change in voids.Relative to the Reference case, the change in ACPR was -0.002 for +0.05 voids and +0.01 for-0.05 voids, which is consistent with the staff's observations. NRC RAI SRXB-120 The void increase cases exhibited opposite trends relative to the void reduction cases. The staff found that the void reduction cases were not consistent with the staffs expectations. Provide information similar to the information requested in SRXB-1 18 and 119 for the fixed increase in void fraction sensitivity analyses.For each case in Study 1 provide:* The limiting bundle: core location, initial radial peaking factor and axial power shape" Plots of the perturbed axial and radial core power shape" Plots of transient limiting bundle peak rod heat flux and mass flow rate* Plots of transient critical CPR" A comparison of the predicted power pulse heights and widths.Response to SRXB-120 For an increase in void fraction, the responses to SRXB-1 18 and SRXB-1 19 provide the requested information. That is, the void trend was explained and the change did not result in a significant impact to the SLMCPR and the transient analyses are consistent with the staff's expectations when instantaneous voids are considered. E2-66 NON-PROPRIETARY INFORMATION NRC RAI SRX8-119 Continuing with the void fraction reduction case, the decrease in void fraction would simultaneously result in a redistribution of the axial power. Since those higher void nodes would have a greater reactivity response than low void nodes, the axial power distribution would shift upwards in the core. The upward shift in the axial power distribution has the effect of increasing the reactor adjoint in the upper portions of the core. As pressurization transients are typically limiting, the impact of an upward shift in axial power on the transient power prediction should be considered. The upward shift in reactor adjoint directly affects the core void reactivity coefficient and tends to increase the sensitivity of the core reactivity to a pressure wave, since the back pressure wave is dissipated by void collapse in the upper parts of the core. Therefore, the core wide transient power would be increased relative to the base case, which appears to result in an increase in the DCPR. The results of the TVA sensitivity analysis do not demonstrate this trend. Address why imposing a fixed void fraction reduction does not result in a higher DCPR. Response to SRX8-119 The trend described above in the RAI question for SRXB-119 is for an instantaneous change in voids. As discussed in the previous response to SRXB-118, the results of the SRXB-88 sensitivity analyses were based on fuel depleted with the changes in the void-quality correlations. AREVA concurs with the general trend as described above for an instantaneous change in void. The analysis of SRXB-88 was repeated for an instantaneous change in voids. Relative to the Reference case, the change in .'1CPR was -0.002 for +0.05 voids and +0.01 for -0.05 voids, which is consistent with the staff's observations. NRC RAI SRX8-120 The void increase cases exhibited opposite trends relative to the void reduction cases. The staff found that the void reduction cases were not consistent with the staff's expectations. Provide information similar to the information requested in SRXB-118 and 119 for the fixed increase in void fraction sensitivity analyses. For each case in Study 1 provide:

• The limiting bundle: core location, initial radial peaking factor and axial power shape
• Plots of the perturbed axial and radial core power shape
• Plots of transient limiting bundle peak rod heat flux and mass flow rate
• Plots of transient critical CPR
• A comparison of the predicted power pulse heights and widths. Response to SRX8-120 For an increase in void fraction, the responses to SRXB-118 and SRXB-119 provide the requested information.

That is, the void trend was explained and the change did not result in a significant impact to the SLMCPR and the transient analyses are consistent with the staff's expectations when instantaneous voids are considered. E2-66 NON-PROPRIETARY INFORMATION Below are the requested results for Study 1 and are based on depleting the fuel with the change in the void correlations (the results are not for an instantaneous change in the void correlations).

• The limiting bundle: core location and initial radial peaking factor (Table SXRB-1 20.1).* Initial axial power shape (Reference case) (Figure SXRB-1 20.1).* Plots of the perturbed axial power shapes (initial conditions) (Figure SXRB-1 20.1).* Plots of the perturbed radial core power shapes (initial conditions) (Figure SXRB-120.2).
• Plots of transient limiting bundle peak rod heat flux at axial heights (x/L) of 25%, 50%, and 75% (Figures SXRB-1 20.3 to -120.5).* Plots of transient limiting bundle mass flow rate at axial heights of 0%, 25%, 50%, 75%and 100% (Figures SXRB-120.6 to -120.10).* Plots of transient CPR (Figure SXRB-1 20.11).* A comparison of the predicted power pulse heights and widths (Table SXRB-120.2).

E2-67 NON-PROPRIETARY INFORMATION Below are the requested results for Study 1 and are based on depleting the fuel with the change in the void correlations (the results are not for an instantaneous change in the void correlations).

• The limiting bundle: core location and initial radial peaking factor (Table SXRB-120.1).
• Initial axial power shape (Reference case) (Figure SXRB-120.1).
• Plots of the perturbed axial power shapes (initial conditions) (Figure SXRB-120.1).
• Plots of the perturbed radial core power shapes (initial conditions) (Figure SXRB-120.2).
• Plots of transient limiting bundle peak rod heat flux at axial heights (x/L) of 25%, 50%, and 75% (Figures SXRB-120.3 to -120.5).
• Plots of transient limiting bundle mass flow rate at axial heights of 0%, 25%, 50%, 75% and 100% (Figures SXRB-120.6 to -120.10).
• Plots of transient CPR (Figure SXRB-120.11).
• A comparison of the predicted power pulse heights and widths (Table SXRB-120.2).

E2-67 NON-PROPRIETARY INFORMATION Table SXRB-120.1 Limiting Bundle Data Study 1 Modified V-Q (-0.05)Reference Calculation Study 1 Modified V-Q (0.0)Study 1 Modified V-Q (+0.05)Location in Core W,J 23,24 21,22 23,24 37,24 MB2 initial radialpeaking factor 1.314 1.295 1.329 1.346 XCT converged initial peaking factor radial 4[ ] [ ] [ ] [REGIONSM1 3579U13 15 JR: H 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 60 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 6 61 62 63 64 65 66 67 68 69 7071 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99100 101 102 103104 105 106107 108 109110 11132 113 114 115116117 118 119 120 121 122 123 124 125 126 371 28129 130 131 1 133 134 135 136 137 138 139 140 141 142 143144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 16W 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 206 209 210 211 212 213 214 215 216 217 218 219 220 221 m2 223 224 225 226 227 228 229 230 231 23M 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 26D 261 262 263 264 265 266 267 26B 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 2W 293 294 25 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 310 321 3 323 324 325 326 3V7 328 329 330 331 332 333 334 335 M6 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 3M2 353 354 355 356 357 358 359 36D 361 362 363 364 365 366 367 36B 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 M1 3W 394 3 396 397 398 399 400 40142 403 404 405 407 408 409 410 41142413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 43V 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 47 474 475 C76 477 478 479 480 481 482 483 484485 486 487 488 489 490 491.49Q 493 494 455 496 497 496 499 500 501 502 503 504 505 506 507 508 509 510 Sf11512 513 514 W15 516 517 518 519 520 521 =2 52 524 525 526 527 528 529 530 531 531 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 55 554 555556 557 558 559 56W 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 W7 608 6W9 610 611 612 613 614 615 616 617 618 619 60 621 622 623 624 625 626 6V7 628 629 630 631 632 633 634 635 636 67 638 639 640 641 642 643 644 645 646 647 648 649 650 61 652 653 604 655 656 657 658 659 66W 661 662 663 664 665 666 667 669 669 670 671 672673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711712 713 714715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 73V 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750751 752 753 754 755 756 757 758 759 760 761 762 763 764 IR: 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 The XCOBRA-T radial that results in a [] during the transient event.E2-68 4 NON-PROPRIETARY INFORMATION Table SXRB-120.1 Limiting Bundle Data Study 1 Study 1 Study 1 Modified Modified Modified Reference v-a v-a v-a Calculation (-0.05) (0.0) (+0.05) Location in Core I,J 23,24 21 , 22 23,24 37,24 MB2 initial radial peaking factor 1.314 1.295 1.329 1.346 XCT converged initial peaking factor radial 4 [ ] [ ] [ ] [ ] REGIONS lR: 1 3 5 7 9 11 13 15 17 19 21 23 25 'Z7 29 31. 33 35 37 39 41 43 45 47 49 51 53 55 57 59 WA ro 58 56 54 52 50 48 46 1 2 3 4 5 678 9 ro 11 U 13 M 15 16 17 18 19 20 21 22 23 24 25 26 'Z7 28 29 " 22 20 18 M U ro 8 6 4 2 lR: 1 3 5 7 9 11 13 15 17 19 21 23 25 'Z7 29 31. 33 35 37 39 41 43 45 47 49 51 53 55 57 59 The XCOBRA-T radial that results in a [ ] during the transient event. E2-68 NON-PROPRIETARY INFORMATION Table SXRB-120.2 Power Pulse Pulse Width Pulse Height Void (seconds) (% rated)Reference 0.53 318 Modified 0.56 330 Modified +0.05 0.58 328 Modified -0.05 0.54 330 E2-69 NON-PROPRIETARY INFORMATION Table SXRB-120.2 Power Pulse Pulse Width Pulse Height Void (seconds) (% rated) Reference 0.53 318 Modified 0.56 330 Modified +0.05 0.58 328 Modified -0.05 0.54 330 E2-69 NON-PROPRIETARY INFORMATION 2.00 1.80 1.60 1.40.1.20 o 1.00 a,.0.80 0.60 0.40 0.20 0.00-.-- Reference-Modified------ Modified +0.05---Modified -0.05 0 5 10 15 20 Axial Node 25 30 Figure SRXB-120.1 Initial Axial Power Shape-Depleted Voids 1.6 1.4 1.2 1 1 0.8 E 8 0.6 z 0.4 0.2 0 100 200 300 400 500 600 700 800 Bundle Index Figure SRXB-120.2 Initial Radial Power Distribution from Transient Analyses -Depleted Voids E2-70 NON-PROPRIETARY INFORMATION 2.00 r------------------------------------------------------------------, 1.80 .1---------------------- 1.60 +---------------------- 1.40 ---------------; :8 1.20 r:. .. ---Reference --Modified -----. Mod i fied +0.05 ----Modified -0.05 1.00 -I-------------- j'-J'-, n; 0_80 .1------O.BO 0.40 \--'--r-.. "-----------0.20 0.00 o 5 10 15 Axial Node 20 25 Figure SRXB-120.1 Initial Axial Power Shape -Depleted Voids 30 1.4 *1---------------- 1.2 _-,,-. \ 0.8 *1------------ O.B +---------0.4 1------------- ---Reference 0.2 --Modified -----. Modified +0.05 ----Modified -0.05 o 100 200 300 400 500 BOO Bundle Inde x Figure SRXB-120.2 Initial Radial Power Distribution from Transient Analyses -Depleted Voids E2-7 0 700 800 NON-PROPRIETARY INFORMATION 0.50 0.40 0.30 X 0.20 0.10 0.00-.-.-.--.-.-...... -..... .---- Reference-Modified...... Modified +0.05---. Modified -0.05 23 23.5 24 Time (Sec)24.5 25 Figure SRXB-120.3 Heat Flux vs. Time 25% x/L -Depleted Voids 0.50 0.40 0.30 K 0.20 0.10 0.00 23 23.5 24 24.5 25 Time (Sec)Figure SRXB-120.4 Heat Flux vs. Time 50% x/L -Depleted Voids E2-71 NON-PROPRIETARY INFORMATION OAO .s: 0.30 :; l:i !. >< " ,. /,," ---. .:.. . .:. . ..::.'...0-.., '-' .... _ ** ",._ --------------------------------- --.-.... ,. i 0.201:r 0.10 ---Referen ce --Modified ----...... M odified +0.05 -. -. Modified -0.05 0.00 __ -_ _,_-,__----_--_- __ .......J 23 23.5 24 Time (Sec) 24.5 Figure SRXB-120.3 Heat Flux vs. Time 25% x/L -Depleted Voids 25 OAO 1--------__________ , f .s: 0.30 :; l:i !. >< " : 0.20 --:r 0.10 ---Reference --Modified ..*... Modified +0.05 -. -. Modified -0.05 0.00 L--______________ .,--________________

---

____ __ --__ ___l 23 23.5 24 Time (Sec) 24.5 Figure SRXB-120.4 Heat Flux vs. Time 50% x/L -Depleted Voids E2-71 25 NON-PROPRIETARY INFORMATION 0.50 0.40.r 0.30.0 0.20 0.10 0.00.~ ~ ~ , .' .' ...............

---

---

Reference-Modified------ Modified +0.05.Modified -0.05 23 23.5 24 Time (Sec)24.5 25 Figure SRXB-120.5 Heat Flux vs. Time 75% x/L -Depleted Voids 50 40 0 0L oI 30 20 10 0 23 23.5 24 24.5 Time (Sec)25 Figure SRXB-120.6 Mass Flow vs. Time 0% x/L -Depleted Voids E2-72 NON-PROPRIETARY INFORMATION 0.50.,..-----------------------------------, 0.40

-0.10 .-= =:====;--- --Modified ------Modified +0.05 [---..... ----Modified -005 0.00 ____ -__ -__ __J 23 23.5 24 Time (Sec) 24.5 Figure SRXB-120.5 Heat Flux vs. Time 75% xlL -Depleted Voids 25 50 r--------------------------------, 40 ---Reference --Modified -----. Modified +0.05 ----Modified -0.05 ____ __ __ --____________ --__ 23 23.5 24 Time (Sec) 24.5 Figure SRXB-120.6 Mass Flow vs. Time 0% x/L -Depleted Voids E2-72 25 NON-PROPRIETARY INFORMATION 50 40 0 I0:-U=30 20---- Reference-Modified...... Modified +0.05.... Modified -0.05 10 0 23 23.5 24 24.5 25 50 40 Time (Sac)Figure SRXB-120.7 Mass Flow vs. Time 25% x/L -Depleted Voids.2 U.'A C-2 LU U=30 20 10 0 eeec-Modified....Modified +0.05-.Modified -0.05 23 23.5 24 24.5 25 Time (Sec)Figure SRXB-120.8 Mass Flow vs. Time 50% x/L -Depleted Voids E2-73 NON-PROPRIE T ARY IN F ORMA T ION 50 40 ./----------------------- .--.-10 1;====== -----. Mod i fied +0.05 _-_-_-----Modified -0.05 ________ ____ ______ 23 23.5 24 T ime (Sec) 24.5 Figure SRXB-120.7 Mass Flow vs. Time 25% x/L -Depleted Vo i ds 25 50 r---------------------------------------------------------------, 40-------_.-------g o u:: '" In co ::E 20 ----10 l r=======-, ---Refe r ence --M odified ------M odified +0.05 ----Modified -0.05 23 23.5 24 T i me (Sec) 24.5 Figure SRXB-120.8 Mass Flow vs. Time 50% x/L -Depleted Voids E2-73 25 NON-PROPRIETARY INFORMATION 50 40 0 U-,T 30 20 10 23 23.5 24 24.5 25 Time (Sec)Figure SRXB-120.9 Mass Flow Vs. Time 75% x/L -Depleted Voids 50 40 30 F LL'a a-a 2 20 10 23 23.5 24 24.5 25 Time (Sac)Figure SRXB-120.10 Mass Flow vs. Time 100% x/L -Depleted Voids E2-74.!!! g 0 u:: <II <II to :IE o u:: <II <II to NON-PROPRIETARY INFORMATION 4 0 3 0 --..... & 20 1 0 1 r=======-, ---R eference --Mo d ified ------Modi fi e d +0.05 ----M o difi ed -0.05 __ 2 3 23.5 24 Time (Sec) 24.5 Figure SRXB-120.9 Mass Flow Vs. Time 75% x/L -Depleted Voids 25 5 0.------------------------------------------------------------------. 40 ----:IE 20 +---------------------R efe r ence --M o d i fi ed -----. M o d i fi ed +0.05 ----M odi fi ed -0.0 5 ____ __ --______ 23 23.5 24 Time (Sec) 24.5 Figure SRXB-120.1 0 Mass Flow vs. Time 100% x1L -Depleted Voids E2-74 25 NON-PROPRIETARY INFORMATION f-'Figure SRXB-120.11 CPR vs. Time Depleted Voids E2-75 r NON-PROPRIETARY INFORMATION Figure SRX8-120.11 CPR VS. Time Depleted Voids E2-75 .J NON-PROPRIETARY INFORMATION NRC RAI SRXB-121 It should be noted that the increase in the operating limit CPR (OLMCPR) for the increased void fraction cases is substantial relative to the base case. The Study 1 increase in OLMCPR is 0.014 and the Study 2 increase in OLMCPR is 0.027.In response to RAI SRXB-88, TVA stated that COTRANSA2 includes a 110 percent multiplier on integral thermal power as a conservative assumption. However, XN-NF-80-19(P)(A) Exxon Nuclear Methodology for Boiling Water Reactors, THERMEX: Thermal Limits Methodology Summary Description Section 4.4 states: "In developing the methodology for the COTRANSA code Exxon Nuclear addressed uncertainties in the code through the integral power variable.The revised methodology uses a more conservative deterministic bounding value (+10 percent)for the integral power uncertainty." TVA's evaluation of the 110 percent conservatism found that the OLMCPR margin afforded by the conservatism is [[ ]].While analysis of pre-EPU reactor conditions, such as the Peach Bottom turbine trip tests, indicate that the 110 percent multiplier is adequate. The response to RAI SRXB-88 appears to indicate that at EPU conditions, that the integral thermal power response to a 5 percent uncertainty in void fraction may not be bounded by the conservatism afforded by the 110 percent multiplier. This is evidenced by an increase in the OLMCPR in Study 2 that exceeds the conservatism afforded by the total 110 percent multiplier. It should be noted that the intent of the 110 percent multiplier is to conservatively bound all uncertainties, including uncertainties in other important variables such as flow and friction factors.Given that the OLMCPR increase exceeds the 110 percent multiplier margin, provide a demonstration that the integrated effect of all conservatisms in COTRANSA2 for Unit 2 at EPU conditions is adequate. This demonstration may be provided by qualification against relevant operating plant transient data to ensure conservatism of the methodology for EPU or near-EPU conditions or by comparison against a rigorous statistical treatment of all uncertainties or by some alternative quantitative and applicable means.Response to SRXB-121 As discussed in the response to RAI SRXB-88 (Reference SRXB-121.1), the [ ]correlation in MICROBURN-B2 was modified to adjust the mean to match the measured ATRIUM-10 void fraction data for both high and low void fractions. The modified [ I correlation was then further modified to generate two bounding correlations for the ATRIUM-10 data of +/-0.05 mean void. The results of the modified correlations were shown in Figure SRXB-88.2 and were used in sensitivity Study 1 and Study 2.The sensitivity studies described in the response to RAI SRXB-88 are somewhat artificial and only capture the sensitivity of portions of the methodology to void correlation uncertainty. Study 2 is especially artificial in that the results of the study only reflect the increase in core void reactivity coefficient. Study 2 does not reflect that a different change in void fraction would occur for a given pressure change with the modified void correlation. Study 1 included this effect and resulted in a smaller effect on the OLMCPR. Other effects of using a different void correlation uncertainty are not incorporated into Study 1 (e.g., pressure drop correlation coefficients would be different). These sensitivity studies are not complete assessments of the impact of void correlation uncertainty on OLMCPR.E2-76 NON-PROPRIETARY INFORMATION NRC RAI SRXB-121 It should be noted that the increase in the operating limit CPR (OLMCPR) for the increased void fraction cases is substantial relative to the base case. The Study 1 increase in OLMCPR is 0.014 and the Study 2 increase in OLMCPR is 0.027. In response to RAI SRXB-88, TVA stated that COTRANSA2 includes a 110 percent multiplier on integral thermal power as a conservative assumption. However, XN-NF-80-19(P)(A) Exxon Nuclear Methodology for Boiling Water Reactors, THERMEX: Thermal Limits Methodology Summary Description Section 4.4 states: "In developing the methodology for the COTRANSA code Exxon Nuclear addressed uncertainties in the code through the integral power variable. The revised methodology uses a more conservative deterministic bounding value (+10 percent) for the integral power uncertainty." TVA's evaluation of the 110 percent conservatism found that the OLMCPR margin afforded by the conservatism is [[ ]]. While analysis of pre-EPU reactor conditions, such as the Peach Bottom turbine trip tests, indicate that the 110 percent multiplier is adequate. The response to RAI SRXB-88 appears to indicate that at EPU conditions, that the integral thermal power response to a 5 percent uncertainty in void fraction may not be bounded by the conservatism afforded by the 110 percent multiplier. This is evidenced by an increase in the OLMCPR in Study 2 that exceeds the conservatism afforded by the total 110 percent multiplier. It should be noted that the intent of the 110 percent multiplier is to conservatively bound all uncertainties, including uncertainties in other important variables such as flow and friction factors. Given that the OLMCPR increase exceeds the 110 percent multiplier margin, provide a demonstration that the integrated effect of all conservatisms in COTRANSA2 for Unit 2 at EPU conditions is adequate. This demonstration may be provided by qualification against relevant operating plant transient data to ensure conservatism of the methodology for EPU or near-EPU conditions or by comparison against a rigorous statistical treatment of all uncertainties or by some alternative quantitative and applicable means. Response to SRXB-121 As discussed in the response to RAI SRXB-88 (Reference SRXB-121.1), the [ correlation in MICROBURN-B2 was modified to adjust the mean to match the measured ATRIUM-10 void fraction data for both high and low void fractions. The modified [ ] correlation was then further modified to generate two bounding correlations for the ATRIUM-10 data of +/-0.05 mean void. The results of the modified correlations were shown in Figure SRXB-88.2 and were used in sensitivity Study 1 and Study 2. The sensitivity studies described in the response to RAI SRXB-88 are somewhat artificial and only capture the sensitivity of portions of the methodology to void correlation uncertainty. Study 2 is especially artificial in that the results of the study only reflect the increase in core void reactivity coefficient. Study 2 does not reflect that a different change in void fraction would occur for a given pressure change with the modified void correlation. Study 1 included this effect and resulted in a smaller effect on the OLMCPR. Other effects of using a different void correlation uncertainty are not incorporated into Study 1 (e.g., pressure drop correlation coefficients would be different). These sensitivity studies are not complete assessments of the impact of void correlation uncertainty on OLMCPR. E2-76 NON-PROPRIETARY INFORMATION It should be noted that a +/-0.05 perturbation of the void correlation used in the SRXB-88 sensitivity studies is substantial. For example, the +0.05 void scenario is equivalent to a[ ]. The measure of void correlation uncertainty used in the sensitivity analyses was somewhat arbitrarily defined as a value that would bound the ATRIUM-10 test data. In a BWR, the core power and power distribution are tightly coupled with the void fraction and a large error in predicted core void fraction would have a significant effect on the predicted power distribution measurements obtained from operating reactors. If the error in void fraction was as large as assumed in the SRXB-88 sensitivity studies, the effect would be observed in comparisons of predicted to measured power distributions obtained from operating reactors.Additional calculations were performed []. These results confirm the conclusion stated above that the increased void variation of +0.05 is not realistic. Integral power is a parameter obtainable from test measurements that is directly related to ACPR and provides a means to assess code uncertainty. The COTRANSA transient analysis methodology was a predecessor to the COTRANSA2 methodology. The integral power figureof merit was introduced with the COTRANSA methodology as a way to assess (not account for)code uncertainty impact on ACPR. From COTRANSA analyses of the Peach Bottom turbine trip tests, the mean of the predicted to measured integral power was 99.7% with a standard E2-77 NON-PROPRIETARY INFORMATION It should be noted that a +/-0.05 perturbation of the void correlation used in the SRXB-88 sensitivity studies is substantial. For example, the +0.05 void scenario is equivalent to a [ ]. The measure of void correlation uncertainty used in the sensitivity analyses was somewhat arbitrarily defined as a value that would bound the ATRIUM-10 test data. In a BWR, the core power and power distribution are tightly coupled with the void fraction and a large error in predicted core void fraction would have a significant effect on the predicted power distribution measurements obtained from operating reactors. If the error in void fraction was as large as assumed in the SRXB-88 sensitivity studies, the effect would be observed in comparisons of predicted to measured power distributions obtained from operating reactors. Additional calculations were performed [ ]. These results confirm the conclusion stated above that the increased void variation of +0.05 is not realistic. Integral power is a parameter obtainable from test measurements that is directly related to and provides a means to assess code uncertainty. The COTRANSA transient analysis methodology was a predecessor to the COTRANSA2 methodology. The integral power figure of merit was introduced with the COTRANSA methodology as a way to assess (not account for) code uncertainty impact on From COTRANSA analyses of the Peach Bottom turbine trip tests, the mean of the predicted to measured integral power was 99.7% with a standard E2-77 NON-PROPRIETARY INFORMATION deviation of 8.1%. AREVA (Exxon Nuclear at the time) initially proposed to treat integral power as a statistical parameter. However, following discussions with the NRC, it was agreed to apply a deterministic 110% integral power multiplier (penalty) on COTRANSA calculations for licensing analyses. That increase was sufficient to make the COTRANSA predicted to measured integral power conservative for all of the Peach Bottom turbine trip tests.COTRANSA2 (Reference SRXB-121.2) was developed and approved as a replacement for COTRANSA in the AREVA thermal limits methodology (Reference SRXB-121.3). Initially it was not planned to use the 110% integral power multiplier with the COTRANSA2 methodology. COTRANSA2 predictions of integral power were conservative for all Peach Bottom turbine trip tests. The minimum conservatism was [ ] and the mean of the predicted to measured integral power was [ ]. The comparisons to the Peach Bottom turbine trip tests demonstrated that the 110% integral power multiplier was not needed for COTRANSA2. However, because the thermal limits methodology that was approved independently of COTRANSA2 included discussion of the 110% integral power multiplier, the use of the multiplier was retained for COTRANSA2 licensing calculations. With the 110% multiplier, the COTRANSA2 predicted to measured mean integral power is [ ] for the Peach Bottom turbine trip tests. Applying a [ ] integral power multiplier provides an OLMCPR conservatism of [ ] versus the [ ] reported in the response to RAI SRXB-88 for the 110% multiplier alone.To summarize the above paragraphs, the sensitivity studies described in the response to RAI SRXB-88 overestimate the potential impact of uncertainty in the void correlation and the 110% integral power multiplier is just one part of the conservatism in the COTRANSA2 methodology and application process that covers methodology uncertainties. COTRANSA2 is not a statistical methodology and uncertainties are not directly input to the analyses. The methodology is a deterministic bounding approach that contains sufficient conservatism to offset uncertainties in individual phenomena. Conservatism is incorporated in the methodology in two ways: (1) computer code models are developed to produce conservative results on an integral basis relative to benchmark tests, and (2) important input parameters are biased in a conservative direction in licensing calculations. Justification that the integrated effect of all the conservatisms in COTRANSA2 licensing analyses is adequate for EPU operation is provided below.The COTRANSA2 methodology results in predicted power increases that are bounding] on average) relative to Peach Bottom benchmark tests. In addition, for licensing calculations a 110% multiplier is applied to the calculated integral power toprovide additional conservatism. This approach adds significant conservatism to the calculated OLMCPR as discussed previously. Biasing of important input parameters in licensing calculations provides additional conservatism in establishing the OLMCPR. The Peach Bottom turbine trips were performed assuming the measured performance of important input parameters such as control rod scram speed and turbine valve closing times. For licensing calculations, these (and other) parameters are biased in a conservative bounding direction. These conservative assumptions are not combined statistically; assuming all parameters are bounding at the same time produces very conservative results.Assessments such as the Peach Bottom tests indicate that the integrated effect of all the conservatism in COTRANSA2 is adequate for non-EPU reactor conditions (as stated in the RAI). To demonstrate that the impact of the change in void-quality correlations is E2-78 NON-PROPRIETARY INFORMATION deviation of 8.1 %. AREVA (Exxon Nuclear at the time) initially proposed to treat integral power as a statistical parameter. However, following discussions with the NRC, it was agreed to apply a deterministic 110% integral power multiplier (penalty) on COTRANSA calculations for licensing analyses. That increase was sufficient to make the COTRANSA predicted to measured integral power conservative for all of the Peach Bottom turbine trip tests. COTRANSA2 (Reference SRXB-121.2) was developed and approved as a replacement for COTRANSA in the AREVA thermal limits methodology (Reference SRXB-121.3). Initially it was not planned to use the 110% integral power multiplier with the COTRANSA2 methodology. COTRANSA2 predictions of integral power were conservative for all Peach Bottom turbine trip tests. The minimum conservatism was [ ] and the mean of the predicted to measured integral power was [ ]. The comparisons to the Peach Bottom turbine trip tests demonstrated that the 110% integral power multiplier was not needed for COTRANSA2. However, because the thermal limits methodology that was approved independently of COTRANSA2 included discussion of the 110% integral power multiplier, the use of the multiplier was retained for COTRANSA2 licensing calculations. With the 110% multiplier, the COTRANSA2 predicted to measured mean integral power is [ ] for the Peach Bottom turbine trip tests. Applying a [ ] integral power multiplier provides an OLMCPR conservatism of [ ] versus the [ ] reported in the response to RAI SRXB-88 for the 110% multiplier alone. To summarize the above paragraphs, the sensitivity studies described in the response to RAI SRXB-88 overestimate the potential impact of uncertainty in the void correlation and the 110% integral power multiplier is just one part of the conservatism in the COTRANSA2 methodology and application process that covers methodology uncertainties. COTRANSA2 is not a statistical methodology and uncertainties are not directly input to the analyses. The methodology is a deterministic bounding approach that contains sufficient conservatism to offset uncertainties in individual phenomena. Conservatism is incorporated in the methodology in two ways: (1) computer code models are developed to produce conservative results on an integral basis relative to benchmark tests, and (2) important input parameters are biased in a conservative direction in licensing calculations. Justification that the integrated effect of all the conservatisms in COTRANSA2 licensing analyses is adequate for EPU operation is provided below.

• The COTRANSA2 methodology results in predicted power increases that are bounding ( [ ] on average) relative to Peach Bottom benchmark tests. In addition, for licensing calculations a 110% multiplier is applied to the calculated integral power to provide additional conservatism.

This approach adds significant conservatism to the calculated OLMCPR as discussed previously.

• Biasing of important input parameters in licensing calculations provides additional conservatism in establishing the OLMCPR. The Peach Bottom turbine trips were performed assuming the measured performance of important input parameters such as control rod scram speed and turbine valve closing times. For licensing calculations, these (and other) parameters are biased in a conservative bounding direction.

These conservative assumptions are not combined statistically; assuming all parameters are bounding at the same time produces very conservative results.

• Assessments such as the Peach Bottom tests indicate that the integrated effect of all the conservatism in COTRANSA2 is adequate for non-EPU reactor conditions (as stated in the RAI). To demonstrate that the impact of the change in void-quality correlations is E2-78 NON-PROPRIETARY INFORMATION similar for EPU and non-EPU conditions, the RAI SRXB-88 sensitivity analyses (Study 1)were repeated for BFN Unit 3 Cycle 14 without EPU. The change in ACPR relative tothe reference cases for EPU and non-EPU are shown in the table below: A(ACPR)Case EPU Non-EPU+0.05 void +0.016 +0.024-0.05 void -0.001 -0.007 Based on these results for EPU and non-EPU conditions, it is concluded that EPU conditions do not increase the sensitivity to a change in the void correlation.

As discussed previously, the core axial power distribution is tightly coupled with the void fraction. A large error in predicted void fraction would have a significant effect on the predicted axial power distribution measurements obtained from operating reactors. The very good comparisons between predicted and measured axial power distributions obtained from operating reactors indicates that the void distribution within the core is being predicted well.Minimal plant transient data at EPU conditions is available to benchmark transient analysis methodologies. However, at the request of the NRC, a COTRANSA2 analysis was performed for a recent event that occurred at a BWR/4 approved for EPU operation. The event involved a reduction in pump speed in one of the recirculation loops followed by a sudden increase in the pump speed approximately 40 seconds later. The event did not pose a challenge to the fuel; however, the event did result in a significant change in core void fraction. Because of the tight coupling between core void fraction and corepower, a comparison of the predicted to measured core power response during the event is a good way to assess the accuracy of the void correlation. For this analysis, a best estimate approach was used and event specific licensing conservatisms were not applied (e.g., measured data used as boundary conditions, realistic control system parameters, best estimate core neutronics data). The recirculation pump speed versus time from the plant data was used as a boundary condition for the analysis (Figure SRXB-121.13). The COTRANSA2 analysis predicted the core power and reactor pressure response very well (Figures SRXB-121.14 and SRXB-121.15). The very good agreement for the predicted core power reached following the pump runback and the following pump runup indicates a good prediction of the core void fraction during the event.Based on the above discussions, the impact of void correlation uncertainty is inherently incorporated in the analytical methods used to determine the OLMCPR. No additional adjustments to the OLMCPR are required to address void correlation uncertainty.

References:

SRXB-121.1 June 3, 2008, TVA Letter to NRC, Browns Ferry Nuclear Plant (BFN) -Units 2 And 3 -Technical Specifications (TS) Change TT-418 -Extended Power Uprate(EPU) -Supplemental Response To Round 16 Request For Additional Information (RAI) -SRXB-88 (TAC Nos. MD5263 AND MD5264) (MI081640325). E2-79 NON-PROPRIETARY INFORMATION similar for EPU and non-EPU conditions, the RAI SRXB-88 sensitivity analyses (Study 1) were repeated for BFN Unit 3 Cycle 14 without EPU. The change in relative to the reference cases for EPU and non-EPU are shown in the table below: Case EPU Non-EPU +O.OS void +0.016 +0.024 -O.OS void -0.001 -0.007 Based on these results for EPU and non-EPU conditions, it is concluded that EPU conditions do not increase the sensitivity to a change in the void correlation.

• As discussed previously, the core axial power distribution is tightly coupled with the void fraction.

A large error in predicted void fraction would have a significant effect on the predicted axial power distribution measurements obtained from operating reactors. The very good comparisons between predicted and measured axial power distributions obtained from operating reactors indicates that the void distribution within the core is being predicted well.

• Minimal plant transient data at EPU conditions is available to benchmark transient analysis methodologies.

However, at the request of the NRC, a COTRANSA2 analysis was performed for a recent event that occurred at a BWR/4 approved for EPU operation. The event involved a reduction in pump speed in one of the recirculation loops followed by a sudden increase in the pump speed approximately 40 seconds later. The event did not pose a challenge to the fuel; however, the event did result in a significant change in core void fraction. Because of the tight coupling between core void fraction and core power, a comparison of the predicted to measured core power response during the event is a good way to assess the accuracy of the void correlation. For this analysis, a best estimate approach was used and event specific licensing conservatisms were not applied (e.g., measured data used as boundary conditions, realistic control system parameters, best estimate core neutronics data). The recirculation pump speed versus time from the plant data was used as a boundary condition for the analysis (Figure SRXB-121.13). The COTRANSA2 analysis predicted the core power and reactor pressure response very well (Figures SRXB-121.14 and SRXB-121.1S). The very good agreement for the predicted core power reached following the pump runback and the following pump runup indicates a good prediction of the core void fraction during the event. Based on the above discussions, the impact of void correlation uncertainty is inherently incorporated in the analytical methods used to determine the OLMCPR. No additional adjustments to the OLMCPR are required to address void correlation uncertainty.

References:

SRXB-121.1 June 3, 2008, TVA Letter to NRC, Browns Ferry Nuclear Plant (BFN) -Units 2 And 3 -Technical Specifications (TS) Change TT-418 -Extended Power Uprate (EPU) -Supplemental Response To Round 16 Request For Additional Information (RAI) -SRXB-88 (TAC Nos. MDS263 AND MDS264) (MI08164032S). E2-79 NON-PROPRIETARY INFORMATION SRXB-121.2 ANF-913(P)(A) Volume 1 Revision 1 and Volume 1 Supplements 2, 3 and 4, COTRANSA2: A Computer Program for Boiling Water Reactor Transient Analyses, Advanced Nuclear Fuels Corporation, August 1990.SRXB-121.3 XN-NF-80-19(P)(A) Volume 3 Revision 2, Exxon Nuclear Methodology for Boiling Water Reactors THERMEX: Thermal Limits Methodology Summary Description, Exxon Nuclear Company, January 1987.E2-80 NON-PROPRIETARY INFORMATION SRXB-121.2 ANF-913(P)(A) Volume 1 Revision 1 and Volume 1 Supplements 2,3 and 4, COTRANSA2: A Computer Program for Boiling Water Reactor Transient Analyses, Advanced Nuclear Fuels Corporation, August 1990. SRXB-121.3 XN-NF-80-19(P)(A) Volume 3 Revision 2, Exxon Nuclear Methodology for Boiling Water Reactors THERMEX: Thermal Limits Methodology Summary Description, Exxon Nuclear Company, January 1987. E2-80 NON-PROPRIETARY INFORMATION r-Figure SRXB-121.1 BFN 2D TIP Statistic Comparison for Variations of the Void Quality Correlation r" Figure SRXB-121.2 BFN 3D TIP Statistic Comparison for Variations of the Void Quality Correlation E2-81 r r NON-PROPRIETARY INFORMATION Figure SRXB-121.1 BFN 20 TIP Statistic Comparison for Variations of the Void Quality Correlation Figure SRXB-121.2 BFN 30 TIP Statistic Comparison for Variations of the Void Quality Correlation E2-81 NON-PROPRIETARY INFORMATION Figure SRXB-121.3 BFN Core Average Axial TIP Comparison at 9026 MWd/MTU for Variations of the Void Quality Correlation Figure SRXB-121.4 BFN Core Average Axial TIP Comparison at 1755 MWd/MTU for Variations of the Void Quality Correlation E2-82 NON-PROPRIETARY INFORMATION r r Figure SRXB-121.3 BFN Core Average Axial TIP Comparison at 9026 MWd/MTU for Variations of the Void Quality Correlation Figure SRXB-121.4 BFN Core Average Axial TIP Comparison at 1755 MWd/MTU for Variations of the Void Quality Correlation E2-82 NON-PROPRIETARY INFORMATION r-U-Figure SRXB-121.5 BFN Core Average Axial TIP Comparison at 9197 MWd/MTU for Variations of the Void Quality Correlation r U-Figure SRXB-121.6 BFN Core Average Axial TIP Comparison at 1340 MWd/MTU for Variations of the Void Quality Correlation E2-83 r NON-PROPRIETARY INFORMATION Figure SRXB-121.5 BFN Core Average Axial TIP Comparison at 9197 MWd/MTU for Variations of the Void Quality Correlation r Figure SRXB-121.6 BFN Core Average Axial TIP Comparison at 1340 MWd/MTU for Variations of the Void Quality Correlation E2-83 .J .J NON-PROPRIETARY INFORMATION Figure SRXB-121. 7 A BWR/4 at EPU 2D TIP Statistic Comparison for Variations of the Void Quality Correlation r Figure SRXB-121.8 A BWR/4 at EPU 3D TIP Statistic Comparison for Variations of the Void Quality Correlation E2-84 NON-PROPRIETARY INFORMATION r Figure SRXB-121. 7 A BWRl4 at EPU 20 TIP Statistic Comparison for Variations of the Void Quality Correlation r Figure SRXB-121.8 A BWRl4 at EPU 30 TIP Statistic Comparison for Variations of the Void Quality Correlation E2-84 .J .J NON-PROPRIETARY INFORMATION r.-J Figure SRXB-121.9 A BWR/4 at EPU Core Average Axial TIP Comparison at 2127 MWd/MTU for Variations of the Void Quality r Figure SRXB-121.10 A BWR/4 at EPU Core Average Axial TIP Comparison at 10621 MWd/MTU for Variations of the Void Quality Correlation E2-85 NON-PROPRIETARY INFORMATION r Figure SRXB-121.9 A BWRl4 at EPU Core Average Axial TIP Comparison at 2127 MWd/MTU for Variations of the Void Quality r Figure SRXB-121.10 A BWRl4 at EPU Core Average Axial TIP Comparison at 10621 MWd/MTU for Variations of the Void Quality Correlation E2-85 ..J ..J NON-PROPRIETARY INFORMATION Figure SRXB-121.11 A BWR/4 at EPU Core Average Axial TIP Comparison at 18459 MWd/MTU for Variations of the Void Quality Correlation r (.--Figure SRXB-121.12 A BWR/4 at EPU Core Average Axial TIP Comparison at 2054 MWd/MTU for Variations of the Void Quality Correlation E2-86 NON-PROPRIETARY INFORMATION r Figure SRXB-121.11 A BWR/4 at EPU Core Average Axial TIP Comparison at 18459 MWd/MTU for Variations of the Void Quality Correlation r Figure SRXB-121.12 A BWR/4 at EPU Core Average Axial TIP Comparison at 2054 MWd/MTU for Variations of the Void Quality Correlation E2-86 ..J NON-PROPRIETARY INFORMATION

---

Measured-4a- Analysis Input 0.80 A.0.I.0.40 0.20 0.00 0.0 20.0 40.0 60.0 80.0 Time (sac)Figure SRXB-121.13 Pump Speed 120-4---- M easured 100.0 120.0 100 80 a.LU S 60 0 40 20 0 -0.00 20.00 40.00 60.00 80.00 100.00 Time (sec)120.00 Figure SRXB-121.14 Core Power E2-87 NON-PROPRIETARY INFORMATION 1.20 ,----------------------------------, 1.00 Il-------------------------

r:rr--.;--:--- --i'------j -------Measured --<>-Ana l y si s I nput Q. 0.60 +-----\---------------- -+--------------1 E :l D. S-II. W 0 II. 0.40 4---------' 0.20 +----------------

---

1 0.00 -!------.,------,----------,------,---------,---------j 0.0 20.0 40.0 60.0 T i me (sec) 80.0 Figure SRXB-121.13 Pump Speed 100.0 120 r--------------------------------- ---+ --Measured --Ca l c ul a t e d 80 60* 40 0+------.,------,----------,------,---------r--------1 1 2 0.0 0.00 20.00 40.00 60.00 Time (sec) 80.00 100.00 1 2 0.00 Figure SRXB-121.14 Core Power E2-87 NON-PROPRIETARY INFORMATION 1100 -r-1050 1000 9.950+...- -. Reactor Pressure, Measured x Dome Pressure, Measured--Dome Pressure, Calculated 900 0 20 40 60 Time (sac)80 100 120 Figure SRXB-1 21.15 Reactor Pressure E2-88 Ci 'iii .S: NON-PROPRIETARY INFORMATION 1100 ,-------------------------------------------------------------

---

. I/) I/) e C1. 950 ---+ --Reactor Pressure, Measured

• Dome Pressure , Measured --Dome Pressure , Calculated o 20 40 60 Time (sec) 80 Figure SRXB-121.15 Reactor Pressure E2-88 100 120 NON-PROPRIETARY INFORMATION NRC RAI SRXB-122 The modified correlations are based on constant slip models. Provide a discussion regarding the treatment of subcooled boiling. This discussion should address void fraction continuity at the boiling boundary.

Describe any impact on the transient analyses arising from SCRAM reactivity worth if significant differences are expected based on treatment of subcooled boiling.Response to SRXB-122The thermal hydraulic methodology incorporates the effects of subcooled boiling through use of the Levy model. The Levy model predicts a critical subcooling that defines the onset of boiling.The critical subcooling is used with a profile fit model to determine the total flow quality that accounts for the presence of subcooled boiling. The total flow quality is used with the void-quality correlation to determine the void fraction. This void fraction explicitly includes the effects of subcooled boiling. Application of the Levy model results in a continuous void fraction distribution at the boiling boundary.The major influence that the void-quality models have on scram reactivity worth is through the predicted axial power shape. As discussed in previous responses (e.g., SRXB-121), the void-quality models used for ATRIUM-10 fuel result in a very good prediction of the axial power shape.Below are reponses to the five fuels related RAIs, SRXB-123 through SRXB-127, from NRC's September 16, 2008, Round 20 RAI.NRC Introduction to Round 20 RAI The following RAIs are based on proprietary draft responses provided during a public meeting held with the TVA regarding the BFN Units 2 and 3 EPU review on August 7, 2008. These questions focus on the proposed response to SRXB 106.The draft response states that the calculation terminates in the calculated pressure exceeds the correlation bounds ([[ fl). However, under anticipated transient without scram (ATWS)conditions the pressure is expected to exceed this value [[[ 1] pounds per square inch gage (psig)].NRC RAI SRXB-123 Discuss what allows the code to continue its evaluation of the ATWS transient without terminating. Response to SRXB-123 The response to SRXB-1 06 is relative to the XCOBRA-T computer code. The XCOBRA-T computer code is not used in the ATWS overpressurization analysis. The COTRANSA2computer code is the primary code used for the ATWS overpressurization analysis. The ATWSoverpressurization event is not used to establish operating limits for critical power; therefore, the SPCB critical power correlation pressure limit is not a factor in the analysis.E2-89 NON-PROPRIETARY INFORMATION NRC RAI SRXB-122 The modified correlations are based on constant slip models. Provide a discussion regarding the treatment of subcooled boiling. This discussion should address void fraction continuity at the boiling boundary. Describe any impact on the transient analyses arising from SCRAM reactivity worth if significant differences are expected based on treatment of subcooled boiling. Response to SRXB-122 The thermal hydraulic methodology incorporates the effects of subcooled boiling through use of the Levy model. The Levy model predicts a critical subcooling that defines the onset of boiling. The critical subcooling is used with a profile fit model to determine the total flow quality that accounts for the presence of subcooled boiling. The total flow quality is used with the void-quality correlation to determine the void fraction. This void fraction explicitly includes the effects of subcooled boiling. Application of the Levy model results in a continuous void fraction distribution at the boiling boundary. The major influence that the void-quality models have on scram reactivity worth is through the predicted axial power shape. As discussed in previous responses (e.g., SRXB-121), the void-quality models used for ATRIUM-10 fuel result in a very good prediction of the axial power shape. Below are reponses to the five fuels related RAls, SRXB-123 through SRXB-127, from NRC's September 16, 2008, Round 20 RAI. NRC Introduction to Round 20 RAI The following RAls are based on proprietary draft responses provided during a public meeting held with the TVA regarding the BFN Units 2 and 3 EPU review on August 7,2008. These questions focus on the proposed response to SRXB 106. The draft response states that the calculation terminates in the calculated pressure exceeds the correlation bounds ([[ ]]). However, under anticipated transient without scram (ATWS) conditions the pressure is expected to exceed this value [[[ ]] pounds per square inch gage (psig)]. NRC RAI SRXB-123 Discuss what allows the code to continue its evaluation of the ATWS transient without terminating. Response to SRXB-123 The response to SRXB-1 06 is relative to the XCOBRA-T computer code. The XCOBRA-T computer code is not used in the ATWS overpressurization analysis. The COTRANSA2 computer code is the primary code used for the ATWS overpressurization analysis. The ATWS overpressurization event is not used to establish operating limits for critical power; therefore, the SPCB critical power correlation pressure limit is not a factor in the analysis. E2-89 NON-PROPRIETARY INFORMATION NRC RAI SRXB-124 Discuss how the core coolability under 10 CFR 50.46 is evaluated for this event.Response to SRXB-124 The ATWS event is not limiting relative to acceptance criteria identified in 10 CFR 50.46. The core remains covered and adequately cooled during the event. Following the initial power increase during the pressurization phase, the core returns to natural circulation conditions after the recirculation pumps trip and fuel cladding temperatures are maintained at acceptably low levels. The ATWS event is significantly less limiting than the loss-of-coolant accident relative to 10 CFR 50.46 acceptance criteria.NRC RAI SRXB-125 Assuming that the pressure is out of bounds, address how does the code conservatively predicts the fuel temperature. Response to SRXB-125 As indicated in the response to SRXB-1 23, the pressure limit is for application of the SPCB critical power correlation. The SPCB correlation is not used in the ATWS overpressurization analysis.Dryout conditions are not expected to occur for the core average channel that is modeled in COTRANSA2. Dryout might occur in the limiting (high power) channels of the core during the ATWS event; however, these channels are not modeled in COTRANSA2 analyses. For the ATWS overpressurization analysis, ignoring dryout for the hot channels is conservative in that it maximizes the heat transferred to the coolant and results in a higher calculated pressure.NRC RAI SRXB-126 If a fuel rod is predicted in dryout, address how the heat transfer is modeled.Response to SRXB-126 Dryout conditions are not expected to occur for the core average channel that is modeled in COTRANSA2 for the ATWS overpressurization analysis. Dryout might occur in the limiting (high power) channels of the core during the ATWS event; however, these channels are not modeled in COTRANSA2 analyses. For the ATWS overpressurization analysis, ignoring dryout for the hot channels is conservative in that it maximizes the heat transferred to the coolant and results in a higher calculated pressure.E2-90 NON-PROPRIETARY INFORMATION NRC RAI SRXB-124 Discuss how the core coolability under 10 CFR 50.46 is evaluated for this event. Response to SRXB-124 The A TWS event is not limiting relative to acceptance criteria identified in 10 CFR 50.46. The core remains covered and adequately cooled during the event. Following the initial power increase during the pressurization phase, the core returns to natural circulation conditions after the recirculation pumps trip and fuel cladding temperatures are maintained at acceptably low levels. The ATWS event is significantly less limiting than the loss-of-coolant accident relative to 10 CFR 50.46 acceptance criteria. NRC RAI SRXB-12S Assuming that the pressure is out of bounds, address how does the code conservatively predicts the fuel temperature. Response to SRXB-12S As indicated in the response to SRXS-123, the pressure limit is for application of the SPCS critical power correlation. The SPCS correlation is not used in the ATWS overpressurization analysis. Dryout conditions are not expected to occur for the core average channel that is modeled in COTRANSA2. Dryout might occur in the limiting (high power) channels of the core during the ATWS event; however, these channels are not modeled in COTRANSA2 analyses. For the ATWS overpressurization analysis, ignoring dryout for the hot channels is conservative in that it maximizes the heat transferred to the coolant and results in a higher calculated pressure. NRC RAI SRXB-126 If a fuel rod is predicted in dryout, address how the heat transfer is modeled. Response to SRXB-126 Dryout conditions are not expeCted to occur for the core average channel that is modeled in COTRANSA2 for the ATWS overpressurization analysis. Dryout might occur in the limiting (high power) channels of the core during the ATWS event; however, these channels are not modeled in COTRANSA2 analyses. For the ATWS overpressurization analysis, ignoring dryout for the hot channels is conservative in that it maximizes the heat transferred to the coolant and results in a higher calculated pressure. E2-90 NON-PROPRIETARY INFORMATION NRC RAI SRXB-127Discuss whether the heat transfer modeling approach is conservative in terms of the figure of merit (vessel pressure). Response to SRXB-127 Dryout conditions are not expected to occur for the core average channel that is modeled in COTRANSA2. For the ATWS overpressurization analysis, ignoring dryout for the hot channels is conservative in that it maximizes the heat transferred to the coolant and results in a higher calculated pressure.For BWRs, the fluid heat transfer coefficients are high and the thermal resistance of the fluid film is much smaller than the thermal resistance of the cladding, the cladding-to-pellet gap, and the fuel pellet. Variations in the calculated heat transfer coefficients will have an insignificant effect on the calculated peak vessel pressure.E2-91 NON-PROPRIETARY INFORMATION NRC RAI SRXB-127 Discuss whether the heat transfer modeling approach is conservative in terms of the figure of merit (vessel pressure). Response to SRXB-127 Dryout conditions are not expected to occur for the core average channel that is modeled in COTRANSA2. For the ATWS overpressurization analysis, ignoring dryout for the hot channels is conservative in that it maximizes the heat transferred to the coolant and results in a higher calculated pressure. For BWRs, the fluid heat transfer coefficients are high and the thermal resistance of the fluid film is much smaller than the thermal resistance of the cladding, the cladding-to-pellet gap, and the fuel pellet. Variations in the calculated heat transfer coefficients will have an insignificant effect on the calculated peak vessel pressure. E2-91 ENCLOSURE 3 TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT (BFN)UNITS 2 AND 3 TECHNICAL SPECIFICATIONS (TS) CHANGE TS-418 EXTENDED POWER UPRATE (EPU)SUPPLEMENTAL RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION (RAI) ROUNDS 3 AND 18 AND RESPONSE TO ROUND 20 FUELS METHODS RAIs AREVA AFFIDAVIT This enclosure provides AREVA's affidavit for Enclosure 1.ENCLOSURE 3 TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT (BFN) UNITS2AND3 TECHNICAL SPECIFICATIONS (TS) CHANGE TS-418 EXTENDED POWER UPRATE (EPU) SUPPLEMENTAL RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION (RAI) ROUNDS 3 AND 18 AND RESPONSE TO ROUND 20 FUELS METHODS RAls AREVA AFFIDAVIT This enclosure provides AREVA's affidavit for Enclosure

1.

AFFIDAVIT COMMONWEALTH OF VIRGINIA )) ss.CITY OF LYNCHBURG )1. My name is Gayle F. Elliott. I am Manager, Product Licensing, for AREVA NP Inc. (AREVA NP) and as such I am authorized to execute this Affidavit.

2. I am familiar with the criteria applied by AREVA NP to determine whether certain AREVA NP information is proprietary.

I am familiar with the policies established by AREVA NP to ensure the proper application of these criteria.3. I am familiar with the AREVA NP information contained in the Responses to NRC RAI for Round 18 and Round 20 for Browns Ferry EPU, dated September 2008 and referred to herein as "Document." Information contained in this Document has been classified by AREVA NP as proprietary in accordance with the policies established by AREVA NP for the control and protection of proprietary and confidential information.

4. This Document contains information of a proprietary and confidential nature and is of the type customarily held in confidence by AREVA NP and not made available to the public. Based on my experience, I am aware that other companies regard information of the kind contained in this Document as proprietary and confidential.

5. This Document has been made available to the U.S. Nuclear Regulatory Commission in confidence with the request that the information contained in this Document be withheld from public disclosure.

The request for withholding of proprietary information is made in accordance with 10 CFR 2.390. The information for which withholding from disclosure is AFFIDAVIT COMMONWEALTH OF VIRGINIA ) ) ss. CITY OF LYNCHBURG ) 1. My name is Gayle F. Elliott. I am Manager, Product Licensing, for AREVA NP Inc. (AREVA NP) and as such I am authorized to execute this Affidavit.

2. I am familiar with the criteria applied by AREVA NP to determine whether certain AREVA NP information is proprietary.

I am familiar with the policies established by AREVA NP to ensure the proper application of these criteria.

3. I am familiar with the AREVA NP information contained in the Responses to NRC RAI for Round 18 and Round 20 for Browns Ferry EPU, dated September 2008 and referred to herein as "Document." Information contained in this Document has been classified by AREVA NP as proprietary in accordance with the policies established by AREVA NP for the control and protection of proprietary and confidential information.

4. This Document contains information of a proprietary and confidential nature and is of the type customarily held in confidence by AREVA NP and not made available to the public. Based on my experience, I am aware that other companies regard information of the kind contained in this Document as proprietary and confidential.

5. This Document has been made available to the U.S. Nuclear Regulatory Commission in confidence with the request that the information contained in this Document be withheld from public disclosure.

The request for withholding of proprietary information is made in accordance with 10 CFR 2.390. The information for which withholding from disclosure is requested qualifies under 10 CFR 2.390(a)(4) "Trade secrets and commercial or financial information." 6. The following criteria are customarily applied by AREVA NP to determine whether information should be classified as proprietary: (a) The information reveals details of AREVA NP's research and development plans and programs or their results.(b) Use of the information by a competitor would permit the competitor to significantly reduce its expenditures, in time or resources, to design, produce, or market a similar product or service.(c) The information includes test data or analytical techniques concerning a process, methodology, or component, the application of which results in a competitive advantage for AREVA NP.(d) The information reveals certain distinguishing aspects of a process, methodology, or component, the exclusive use of which provides a competitive advantage for AREVA NP in product optimization or marketability.(e) The information is vital to a competitive advantage held by AREVA NP, would be helpful to competitors to AREVA NP, and would likely cause substantial harm to the competitive position of AREVA NP.The information in the Document is considered proprietary for the reasons set forth in paragraphs 6(b) and 6(c) above.7. In accordance with AREVA NP's policies governing the protection and control of information, proprietary information contained in this Document have been made available, on a limited basis, to others outside AREVA NP only as required and under suitable agreement providing for nondisclosure and limited use of the information.

8. AREVA NP policy requires that proprietary information be kept in a secured file or area and distributed on a need-to-know basis.requested qualifies under 10 CFR 2.390(a)(4) "Trade secrets and commercial or financial information. " 6. The following criteria are customarily applied by AREVA NP to determine whether information should be classified as proprietary: (a) The information reveals details of AREVA NP's research and development plans and programs or their results. (b) Use of the information by a competitor would permit the competitor to significantly reduce its expenditures, in time or resources, to design, produce, or market a similar product or service. (c) The information includes test data or analytical techniques concerning a process, methodology, or component, the application of which results in a competitive advantage for AREVA NP. (d) The information reveals certain distinguishing aspects of a process, methodology, or component, the exclusive use of which provides a competitive advantage for AREVA NP in product optimization or marketability

.. (e) The information is vital to a competitive advantage held by AREVA NP, would be helpful to competitors to AREVA NP, and would likely cause substantial harm to the competitive position of AREVA NP. The information in the Document is considered proprietary for the reasons set forth in paragraphs 6(b) and 6(c) above. 7. In accordance with AREVA NP's policies governing the protection and control of information, proprietary information contained in this Document have been made available, on a limited basis, to others outside AREVA NP only as required and under suitable agreement providing for nondisclosure and limited use of the information.

8. AREVA NP policy requires that proprietary information be kept in a secured file or area and distributed on a need-to-know basis.

9. The foregoing statements are true and correct to the best of my knowledge, information, and belief.SUBSCRIBED before me this day of September 2008.Sherry L. McFaden NOTARY PUBLIC, COMMONWEALTH OF VIRGINIA MY COMMISSION EXPIRES: 10/31/10 Reg. # 7079129 NotIry Public Commonwealth of Vlrginla I 7079129 My Commission Expires Oct 31, 2010 1--- -- .----------

I 9. The foregoing statements are true and correct to the best of my knowledge, information, and belief. SUBSCRIBED before me this ----'-It'_"tb __ day of September 2008. Sherry L. McFaden NOTARY PUBLIC, COMMONWEALTH OF VIRGINIA MY COMMISSION EXPIRES: 10/31/10 Reg. # 7079129 SHERRY L. MCFADEN Noto"ry Public Commonwealth of V1rg.tnla 7079129 My Commission Expire. Oct 31. 2010 . --}}