ML11215A091

ANP-2975(NP), Revision 0, San Onofre Nuclear Generating Station, Unit 2 and Unit 3 - Realistic Large Break LOCA Report, Enclosure 6

ML11215A091
Person / Time
Site:	San Onofre
Issue date:	06/30/2011
From:	AREVA NP
To:	Office of Nuclear Reactor Regulation
References
ANP-2975(NP), Rev 0
Download: ML11215A091 (206)

Text

ENCLOSURE 6 ANP-2975(NP), Revision 0, "San Onofre Nuclear Generating Station Unit 2 and Unit 3 Realistic Large Break Report" June 2011 Non-Proprietary

ANP-2975(NP)

Revision 0 San Onofre Nuclear Generating Station Unit 2 and Unit 3 Realistic Large Break LOCA Report June 2011 A

AREVA NP Inc. AR EVA

A AiREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 2 Copyright © 2011 AREVA NP Inc.

All Rights Reserved AREVA NP Inc

AR San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 3 Nature of Changes Item Page Description and Justification 1 All Revision 0. This is a new document.

AREVA NP Inc

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 4 Table of Contents 1.0 Introduction ................................................................................................................... 11 2.0 Sum m ary ........................................................................................................................ 14 3.0 A nalysis ......................................................................................................................... 16 3.1 Description of the LBLO CA Event ................................................................... 17 3.2 Description of Analytical Models ..................................................................... 18 3.3 Plant Description and Summary of Analysis Parameters ................................ 20 3.4 SER Com pliance ............................................................................................ 22 3.5 Realistic Large Break LO CA Results .............................................................. 22 4.0 Generic Support for Transition Package ................................................................ 60 4.1 Reactor Power ................................................................................................. 60 4.2 Rod Q uench ................................................................................................... 60 4.3 Rod-to-Rod Therm al Radiation ....................................................................... 61 4.4 Film Boiling Heat Transfer Lim it ..................................................................... 66 4.5 Downcom er Boiling ......................................................................................... 67 4.6 Break Size ..................................................................................................... 83 4.7 Detailed inform ation for Containm ent Model (ICECO N) .................................. 94 4.8 Cross-References to North Anna .................................................................. 98 4.9 G DC 35 - LOO P and No-LOO P Case Sets .................................................. 99 4.10 Statem ent .......................................................................................................... 100 5.0 Conclusions ................................................................................................................. 101 6.0 Recent NRC Request for Additional Information (RAI) and AREVA NP Responses ................................................................................................................... 102 6.1 Therm al Conductivity Degradation - O nce-Burnt Fuel ...................................... 102 6.2 Decay Heat Treatm ent ...................................................................................... 115 6.3 Thermal Conductivity Degradation - Swelling, Rupture, and R e lo c a tio n ......................................................................................................... 12 2 6.4 Oxidation - Pre-transient and Single-Sided ...................................................... 126 6.5 Single Failure Assum ption ................................................................................. 127 6.6 Core Liquid Level .............................................................................................. 135 6.7 Plant Input Selection and Technical Specifications ........................................... 135 7.0 References ................................................................................................................... 137 This document contains a total of 138 pages AREVA NP Inc.

A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 5 List of Tables Table 2-1 Summary of Major Parameters for Limiting Transient ........................................... 15 Table 3-1 Sampled LBLOCA Parameters .............................................................................. 23 Table 3-2 Plant Operating Range Supported by the LOCA Analysis ..................................... 24 Table 3-3 Statistical Distributions Used for Process Parameters ......................................... 27 Table 3-4 SER Conditions and Limitations ............................................................................ 28 Table 3-5 Summary of Results for the Limiting PCT Case .................................................. 30 Table 3-6 Calculated Event Times for the Limiting PCT Case .............................................. 30 Table 3-7 Heat Transfer Parameters for the Limiting Case .................................................. 31 Table 3-8 Containment Initial and Boundary Conditions ....................................................... 33 Table 3-9 Passive Heat Sinks in Containment Geometry ..................................................... 34 Table 3-10 Material Properties for Passive Heat Sinks in Containment ................................ 35 Table 4-1 Typical Measurement Uncertainties and Local Peaking Factors .......................... 63 Table 4-2 FLECHT-SEASET & 17x17 FA Geometry Parameters ......................................... 63 Table 4-3 FLECHT-SEASET Test Parameters ..................................................................... 65 Table 4-4 Minimum Break Area for Large Break LOCA Spectrum ...................................... 85 Table 4-5 Minimum PCT Temperature Difference - True Large and Intermediate B re a ks ............................................................................................................................. 87 AREVA NP Inc.

A A: R EV.yA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 6 List of Figures Figure 3-1 Primary System Noding for SONGS ..................................................................... 36 Figure 3-2 Secondary System Noding .................................................................................. 37 Figure 3-3 Reactor Vessel Noding ....................................................................................... 38 Figure 3-4 C ore Noding D etail .............................................................................................. 39 Figure 3-5 Upper Plenum Noding Detail ................................................................................ 40 Figure 3-6 Scatter Plot of Operational Parameters ................................................................ 41 Figure 3-7 Scatter Plot of PCT versus PCT Time ................................................................ 43 Figure 3-8 Scatter Plot of PCT versus Break Size ................................................................ 44 Figure 3-9 Scatter Plot of Maximum Transient Oxidation versus PCT .................................. 45 Figure 3-10 Scatter Plot of Total Oxidation versus PCT ....................................................... 46 Figure 3-11 Peak Cladding Temperature (Independent of Elevation) for the Lim itin g C a s e .................................................................................................................. 47 Figure 3-12 Break Flow for the Limiting Case ....................................................................... 48 Figure 3-13 Core Inlet Mass Flux for the Limiting Case ....................................................... 49 Figure 3-14 Core Outlet Mass Flux for the Limiting Case ..................................................... 50 Figure 3-15 Void Fraction at RCS Pumps for the Limiting Case ........................................... 51 Figure 3-16 ECCS Flows (Includes SIT, HPSI and LPSI) for the Limiting Case .................... 52 Figure 3-17 Upper Plenum Pressure for the Limiting Case .................................................. 53 Figure 3-18 Collapsed Liquid Level in the Downcomer for the Limiting Case ....................... 54 Figure 3-19 Collapsed Liquid Level in the Lower Plenum for the Limiting Case ................... 55 Figure 3-20 Collapsed Liquid Level in the Core for the Limiting Case .................................. 56 Figure 3-21 Containment and Loop Pressures for the Limiting Case .................................... 57 Figure 3-22 Reactor Vessel Liquid Mass (Ibm) versus Time (sec) ........................................ 58 Figure 3-23 GDC 35 LOOP versus No-LOOP Cases ........................................................... 59 Figure 4-1 R2RRAD 5x5 Rod Segment ................................................................................ 64 Figure 4-2 Rod Thermal Radiation in FLECHT-SEASET Bundle and in a 17x17 FA .................................................................................................................................... 66 Figure 4-3 Reactor Vessel Downcomer Boiling Diagram ....................................................... 68 Figure 4-4 S-RELAP5 versus Closed Form Solution ........................................................... 71 Figure 4-5 Downcomer Wall Heat Release - Wall Mesh Point Sensitivity ............................. 72 Figure 4-6 PCT Independent of Elevation - Wall Mesh Point Sensitivity .............................. 73 AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 7 Figure 4-7 Downcomer Liquid Level - Wall Mesh Point Sensitivity ....................................... 74 Figure 4-8 Core Liquid Level - Wall Mesh Point Sensitivity .................................................. 75 Figure 4-9 Azim uthal Noding ................................................................................................ 77 Figure 4-10 Lower Compartment Pressure versus Time ....................................................... 78 Figure 4-11 Downcomer Wall Heat Release - Axial Noding Sensitivity Study ..................... 79 Figure 4-12 PCT Independent of Elevation - Axial Noding Sensitivity Study ....................... 80 Figure 4-13 Downcomer Liquid Level - Axial Noding Sensitivity Study ................................ 81 Figure 4-14 Core Liquid Level - Axial Noding Sensitivity Study ........................................... 82 Figure 4-15 Plant A - Westinghouse 3-Loop Design ........................................................... 88 Figure 4-16 Plant B - Westinghouse 3-Loop Design ........................................................... 89 Figure 4-17 Plant C - Westinghouse 3-Loop Design ............................................................ 90 Figure 4-18 Plant D - Combustion Engineering 2x4 Design ............................................... 91 Figure 4-19 Plant E - Combustion Engineering 2x4 Design ................................................ 92 Figure 4-20 Plant H - Westinghouse 4-Loop Design ............................................................ 93 Figure 4-21 PCT vs. Containment Volume ............................................................................ 95 Figure 4-22 PCT vs. Initial Containment Temperature ......................................................... 96 Figure 4-23 Containment Pressure for Limiting Case ........................................................... 97 Figure 6-1 Once-Burnt Fuel Power Ratios (2nd cycle) ............................................................ 108 Figure 6-2 Radial Temperature Profile for Hot Rod ................................................................ 109 Figure 6-3 Temperature versus Time for Fuel Centerline, Clad Surface, and Fuel A ve ra g e ......................................................................................................................... 11 0 Figure 6-4 Fresh versus Once-Burnt U0 2 Rod PCT Trace ...................................................... 111 Figure 6-5 Fractional Fuel Centerline Temperature Delta Between RODEX3A a nd Da ta ........................................................................................................................ 11 2 Figure 6-6 Fuel Centerline Temperature Delta of RODEX3A Calculations to Data (Original and Using the New Correlation) ..................................................................... 113 Figure 6-7 Correction Factor (as applied for temperatures in Kelvin) ...................................... 114 Figure 6-8 Decay Heat Comparisons, Infinite Operation U235, Finite Operation A ll Isoto pe s (0 .1 - 10 se c) ............................................................................................. 118 Figure 6-9 Decay Heat Comparisons, Infinite Operation U235, Finite Operation A ll Isotope s (10 - 1000 sec) .......................................................................................... 119 Figure 6-10 Decay Heat Ratios, Finite Operation over Infinite Operation for U235, A ll Isotopes (0 - 10 sec) ..................................................................................... 120 AREVA NP Inc.

A

,R' V-A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 8 Figure 6-11 Decay Heat Ratios, Finite Operation over Infinite Operation for U235, A ll Isotopes (0 - 600 sec) ................................................................................... 121 Figure 6-12 Clad Temperature Response from Single Failure Study ...................................... 129 Figure 6-13 Comparison of PCT Independent of Elevation for Maximum ECCS and Minim um E C C S (AO R ) .......................................................................................... 13 1 Figure 6-14 Comparison of Containment and System Pressure for Maximum E C C S and Minim um E C C S ........................................................................................... 132 Figure 6-15 Comparison of ECCS Flows for Maximum ECCS and Minimum E CC S ............................................................................................................................ 13 3 Figure 6-16 Comparison of Downcomer Level for Maximum ECCS and Minimum E C C S (A O R ) ................................................................................................................. 13 4 AREVA NP Inc.

A A RE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 9 Nomenclature AFD Axial Flux Difference ARO All Rods Out ASI Axial Shape Index CCTF Cylindrical Core Test Facility CE Combustion Engineering Inc.

CFR Code of Federal Regulations COLR Core Operating Limits Report CSAU Code Scaling, Applicability, and Uncertainty DC Downcomer DEGB Double-Ended Guillotine Break DNB Departure from Nucleate Boiling ECCS Emergency Core Cooling System EFPH Effective Full Power Hours EM Evaluation Model FQ Total Peaking Factor Fr Nuclear Enthalpy Rise Factor HFP Hot Full Power HPSI High Pressure Safety Injection LBLOCA Large Break Loss of Coolant Accident LANL Los Alamos National Laboratory LEFM Leading Edge Flow Meter LHGR Linear Heat Generation Rate LHR Linear Heat Rate LOCA Loss of Coolant Accident LOOP Loss of Offsite Power LPSI Low Pressure Safety Injection MSIV Main Steam Isolation Valve MTC Moderator Temperature Coefficient MWt Mega-Watt thermal NRC U. S. Nuclear Regulatory Commission NSSS Nuclear Steam Supply System PCT Peak Clad Temperature AREVA NP Inc.

A AR'EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 10 Nomenclature (Continued)

PDIL Power Dependent Insertion Limit PIRT Phenomena Identification and Ranking Table PLHGR Planar Linear Heat Generation Rate PWR Pressurized Water Reactor RAS Recirculation Actuation Signal RCP Reactor Coolant Pump RCS Reactor Coolant System RLBLOCA Realistic Large Break LOCA RV Reactor Vessel RWST Refueling Water Storage Tank SER Safety Evaluation Report SCE Southern California Edison Sl Safety Injection SIAS Safety Injection Actuation Signal SIT Safety Injection Tank SONGS San Onofre Nuclear Generating Station TS Technical Specifications VQP Vendor Qualification Package AREVA NP Inc.

A AR,EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 11 1.0 Introduction This report describes and provides results from a RLBLOCA analysis for the two-unit San Onofre Nuclear Generating Station (SONGS) (Unit 2 and Unit 3). The two-unit plant is a Combustion Engineering (CE) designed 3438 MWt plant with a large dry containment. The plant is a 2X4 loop design - two hot legs and four cold legs. The loops contain four reactor coolant pumps (RCPs), two U-tube steam generators and one pressurizer. The ECCS is provided by two independent safety injection trains and four safety injection tanks (SITs).

The analysis supports operation with AREVA NP's HTP 16X16 fuel design using standard U0 2 fuel with 2%, 4%, 6% and 8% Gd 20 3 and M5 cladding. The analysis was performed in compliance with the U.S. Nuclear Regulatory Commission (NRC) approved RLBLOCA EM (Reference 1) with exceptions noted below. Analysis results confirm the 10CFR50.46 (b) acceptance criteria presented in Section 3.0 are met and serve as the basis for operation of the SONGS Unit 2 and Unit 3 with AREVA NP fuel.

The non-parametric statistical methods inherent in the AREVA NP RLBLOCA methodology provide for the consideration of a full spectrum of break sizes, break configuration (guillotine or split break), axial shapes, and plant operational parameters. A conservative loss of an emergency diesel generator assumption is applied. The effect of this is the loss of one LPSI pump and one HPSI pump. The LPSI injects into the broken loop and one intact loop and HPSI injects into all four loops. Regardless of the single-failure assumption, all containment pressure-reducing systems are assumed fully functional. The effects of Gadolinia-bearing fuel rods and peak fuel rod exposures are considered.

The following are deviations from the approved RLBLOCA EM (Reference 1) that were requested by the NRC and are referred to as the "Transition Package." The "Transition Package" is fully described in Section 4.

The assumed reactor core power for the SONGS Unit 2 and Unit 3 realistic large break loss-of-coolant accident is 3458 MWt. This value represents the 100% primary power plus 20 MWt to account for the measurement uncertainty.

The RLBLOCA analysis was performed with a version of S-RELAP5 that requires both the void fraction to be less than 0.95 and the clad temperature to be less than 900OF before the rod is allowed to quench. This may result in a slight increase in peak clad temperature (PCT) results when compared to an analysis not subject to these constraints.

The RLBLOCA analysis was performed with a version of S-RELAP5 that limits the contribution of the Forslund-Rohsenow model to no more than 15% of the total heat transfer at and above a void fraction of 0.9. This may result in a slight increase in PCT results when compared to previous analyses for similar plants.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 12 The split versus double-ended guillotine break (DEGB) type is no longer related to break area.

In concurrence with Regulatory Guide 1.157, both the split and the double-ended guillotine break will range in area between the minimum break area (Amin) and an area of twice the cross-sectional area of the broken pipe. The determination of break configuration, split versus double-ended guillotine, will be made after the break area is selected based on a uniform probability for each occurrence. Amin was calculated to be 29% of the DEGB area (see Section 4.6 for further discussion). This is not expected to have an effect on PCT results.

In concurrence with the NRC's interpretation of GDC 35, a set of 59 cases was run with a loss of offsite power (LOOP) assumption and a second set with a No-LOOP assumption. The set of 59 cases that predicted the highest PCT is reported in Section 2 and Section 3, herein. The results from both case sets are shown in Figure 3-23. The effect on PCT results is expected to be minor.

During recent RLBLOCA EM modeling studies, it was noted that cold leg condensation efficiency may be under-predicted. Water entering the downcomer (DC) post-SIT injection remained sufficiently subcooled to absorb DC wall heat release without significant boiling.

However, tests (Reference 2) indicate that the steam and water entering the DC from the cold leg, subsequent to the end of SIT injection, reach near saturation resulting from the condensation efficiency ranging between 80 to 100 percent. To assure that cold leg condensation would not be under-predicted, a RLBLOCA EM update was made. Noting that saturated fluid entering the DC is the most conservative modeling scheme, steam and liquid multipliers were developed in order to approximately saturate the cold leg fluid before it enters the DC. The multipliers were developed through scoping studies using a number of plant configurations-Westinghouse-designed 3- and 4-loop plants, and CE-designed plants. The results of the scoping study indicated that [

] were appropriate to produce saturated fluid entering the DC. This RLBLOCA EM departure was discussed recently with the NRC and the NRC agreed that the approach described immediately above was satisfactory in the interim. The modification is implemented post-accumulation injection, 10 seconds after the vapor void fraction in the bottom of the SIT becomes greater than 90%. Thus, the SITs have injected all their water into the cold legs, and the nitrogen cover gas has entered the system and been mostly discharged through the break before [

]. Providing saturated fluid conditions at the DC entrance conservatively reduces both the DC driving head and the core flooding rate. Recall that test results indicate that fluid conditions entering the DC range from saturated to slightly subcooled. Hence, it is conservative to force an approximation of saturated conditions for fluid entering the DC.

The NRC raised the issue concerning fuel thermal conductivity degradation as a function of burnup in Information Notice 2009-23. In order to manage this issue, AREVA NP is modifying the way RODEX3A temperatures are compensated in the RLBLOCA Transition Package methodology. In the current process, the RLBLOCA computes PCTs at many different times during an operating cycle. For each specific time in cycle, the fuel conditions are computed AREVA NP Inc.

A AR.EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 13 using RODEX3A prior to starting the S-RELAP5 portion of the analysis. A steady-state condition for the given time in cycle using S-RELAP5 is established. A base fuel centerline temperature is established in this process. Then two-transformation adjustments to the base fuel centerline temperature are computed. The first transformation is a linear adjustment for an exposure of 10 GWd/MTU or higher. The second adjustment is performed in the S-RELAP5 initialization process for the transient case. In the new process, a polynomial transformation is used for the first transformation instead of a linear transformation. The rest of the RLBLOCA process for initializing the S-RELAP5 fuel rod temperature should not be altered and the rest of LOCA transient should also continue in the original fashion. Section 6 will provide additional information on the adjustment and adding once-burnt fuel to the analysis. Note that these changes are also deviations required by the NRC that are departures from the approved RLBLOCA EM.

Recent NRC concerns raised in the form of RAIs are responded to in Section 6.

AREVA NP Inc.

A, AE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 14 2.0 Summary The limiting PCT analysis is based on the parameter specification given in Table 2-1 for the limiting case. Gadolinia-bearing rods of 2%, 4%, 6% and 8% Gd 20 3 were analyzed for fresh and once-burnt fuel in all cases.

The RLBLOCA result is based on a case set of 59 individual transient cases for offsite power not available (LOOP) and 59 individual transient cases for offsite power available (No-LOOP) conditions. The core is composed of AREVA NP HTP 16x16 fuel, hydraulically compatible with 16x16 Westinghouse fuel. Only AREVA NP fuel is analyzed for PCT; a mixed core scenario is hydraulically modeled. The limiting PCT is 1605°F for a fresh fuel 6% Gad rod (Case 7) with offsite power not available (LOOP) conditions. From the same case, the limiting PCT for all once-burnt fuel is 1517°F for a U0 2 rod.

The analysis assumed full core power operation at 3458 MWt. The value represents the 100%

primary power plus 20 MWt uncertainty. The analysis assumed a steam generator tube plugging level of 8% in each steam generator, a LHR limit of 12.8 kW/ft per Technical Specification 3.2.1 and the COLR, which is equivalent to a total peaking factor (FQ) up to a value of 2.37, and a radial peaking factor (Fr) up to a value of 1.755 (including 6% uncertainty). This analysis bounds typical operational ranges or Technical Specifications limits (whichever is applicable) with regard to pressurizer pressure and level; SIT pressure, temperature, and level; core inlet temperature; core flow; containment pressure and temperature; and RWST.

The AREVA NP RLBLOCA Transition Package methodology (on a forward fit basis) explicitly analyzes fresh and once-burnt fuel assemblies to respond to recent NRC RAIs. The twice-burnt fuel assemblies are not considered in the analyses since burnups at this level or higher do not retain sufficient energy potential to achieve significant cladding temperatures or cladding oxidations during the transient. The analysis demonstrates that the 10 CFR 50.46(b) criteria listed in Section 3.0 are satisfied.

AREVA NP Inc.

A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 15 Table 2-1 Summary of Major Parameters for Limiting Transient Fresh 6% Gad Fuel Once-Burnt U0 2 Fuel Core Average Burnup (EFPH) 96641 96642 Core Power (MWt) 3458 Total Peaking (FQ) 2.37 Radial Peak (Fr) 1.633 1.694 Axial Offset 5 -0.1635 to +0.1635 -0.1728 to +0.1728 Break Type Split Break Size (ft2/side) 2.9358 Offsite Power Availability Not available Decay Heat Multiplier 1.0 1 This is - 18.2 GWd/MTU in burnup for the fresh fuel.

2 This is - 33.2 GWd/MTU in burnup for the once-burnt fuel.

3 [ ]

4 ]

5 A RLBLOCA sensitivity study was performed to demonstrate that the SONGS Technical Specification 3.2.1 and COLR analysis ASI limit of +/-0.3 is bounded by the results of the analysis documented herein.

AREVA NP Inc.

AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 16 3.0 Analysis The purpose of the analysis is to verify the adequacy of the ECCS and to demonstrate compliance to the 10CFR 50.46(b) criteria.

1. The calculated maximum fuel element cladding temperature shall not exceed 22000 F.

2. The calculated total oxidation of the cladding shall nowhere exceed 0.17 times the total cladding thickness before oxidation.

3. The calculated total amount of hydrogen generated from the chemical reaction of the cladding with water or steam shall not exceed 0.01 times the hypothetical amount that would be generated if all of the metal in the cladding cylinders surrounding the fuel excluding the cladding surrounding the plenum volume were to react.

4. The calculated changes in core geometry shall be such that the core remains amenable to cooling.

The analysis did not evaluate the 10CFR 50.46(b) long-term cooling criterion, as this is handled in a separate analysis.

The RLBLOCA analysis conservatively considers blockage effects due to clad swelling and rupture in the prediction of the hot fuel rod PCT. The effects of combined LOCA loads on the fuel assembly components have been evaluated by AREVA NP and the resulting loads are below the allowable stress limit for all the components. The combination of compliance with the 2200°F limit and the LOCA loads evaluation ensures no permanent deformation to the fuel assemblies; thereby demonstrating compliance with the criterion that the core remains amenable to cooling.

Section 3.1 of this report describes the postulated LBLOCA event. Section 3.2 describes the models used in the analysis. Section 3.3 describes the CE 2x4 PWR plant and summarizes the system parameters used in the analysis. Compliance to the SER is addressed in Section 3.4.

Section 3.5 summarizes the results of the RLBLOCA analysis. Section 4 discusses the additional information provided under the "Transition Package" on EMF-2103. Section 5 provides the conclusions. Section 6 addresses recent NRC RAIs on RLBLOCA submittals and Section 7 contains the reference list.

AREVA NP Inc.

A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 17 3.1 Description of the LBLOCA Event A LBLOCA is initiated by a postulated large rupture of the RCS primary piping. Based on deterministic studies, the worst break location is in the cold leg piping between the reactor coolant pump and the reactor vessel for the RCS loop containing the pressurizer. The break initiates a rapid depressurization of the RCS. A reactor trip signal is initiated when the low pressurizer pressure trip setpoint is reached; however, reactor trip is conservatively neglected in the analysis. The reactor is shut down by coolant voiding in the core.

The plant is assumed to be operating normally at full power prior to the accident. The cold leg break is assumed to open instantaneously. For this break, a rapid depressurization occurs, along with a core flow stagnation and reversal. This causes the fuel rods to experience DNB.

Subsequently, the limiting fuel rods are cooled by film convection to steam. The coolant voiding creates a strong negative reactivity effect and core criticality ends. As heat transfer from the fuel rods is reduced, the cladding temperature increases.

Coolant in all regions of the RCS begins to flash. At the break plane, the loss of subcooling in the coolant results in substantially reduced break flow. This reduces the depressurization rate, and leads to a period of positive core flow or reduced downflow as the RCPs in the intact loops continue to supply water to the RV (in No-LOOP conditions). Cladding temperatures may be reduced and some portions of the core may rewet during this period. The positive core flow or reduced downflow period ends as two-phase conditions occur in the RCPs, reducing their effectiveness. Once again, the core flow reverses as most of the vessel mass flows out through the broken cold leg.

Mitigation of the LBLOCA begins when the SIAS is issued. This signal is initiated by either high containment pressure or low pressurizer pressure. Regulations require that a worst single-failure be considered. The AREVA NP RLBLOCA methodology conservatively assumes an on-time start and normal lineups of the containment spray to conservatively reduce containment pressure and increase break flow, regardless of the single failure assumed. This single-failure has been determined to be the loss of one emergency diesel generator, which takes one train of ECCS pumped injection out. LPSI inject into the broken loop and one intact loop, HPSI inject into all four loops, and the containment spray system is fully functional with both trains operating.

When the RCS pressure falls below the SIT pressure, fluid from the SITs is injected into the cold legs. In the early delivery of SIT water, high pressure and high break flow will drive some of this fluid to bypass the core. During this bypass period, core heat transfer remains poor and fuel rod cladding temperatures increase. As RCS and containment pressures equilibrate, ECCS water begins to fill the lower plenum and eventually the lower portions of the core; thus, core heat transfer improves and cladding temperatures decrease.

Eventually, the relatively large volume of SIT water is exhausted and core recovery continues relying solely on pumped ECCS injection. As the SITs empty, the nitrogen gas used to AREVA NP Inc.

A, A REVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 18 pressurize the SITs exits through the break. This gas release may result in a short period of improved core heat transfer as the nitrogen gas displaces water in the downcomer. After the nitrogen gas has been expelled, the ECCS temporarily may not be able to sustain full core cooling because of the core decay heat and the higher steam temperatures created by quenching in the lower portions of the core. Peak fuel rod cladding temperatures may increase for a short period until more energy is removed from the core by the fluid delivered by the HPSI and the LPSI, while the decay heat continues to fall. Steam generated from fuel rod rewet will entrain liquid and pass through the core, vessel upper plenum, the hot legs, the steam generators, and the reactor coolant pumps before it is vented out the break. The resistance of this flow path to the steam flow is balanced by the driving force of water filling the downcomer.

This resistance may act to retard the progression of the core reflood and postpone core-wide cooling. Eventually (within a few minutes of the accident), the core reflood will progress sufficiently to ensure core-wide cooling. Full core quench occurs within a few minutes after core-wide cooling. Long-term cooling is then sustained with the LPSI pumped injection system.

3.2 Description of Analytical Models The RLBLOCA methodology is documented in EMF-2103 Realistic Large Break LOCA Methodology (Reference 1). The methodology follows the CSAU evaluation methodology (Reference 3). This method outlines an approach for defining and qualifying a best-estimate thermal-hydraulic code and quantifies the uncertainties in a LOCA analysis.

The RLBLOCA methodology consists of the following computer codes:

• RODEX3A for computation of the initial fuel stored energy, fission gas release, and fuel-cladding gap conductance.

• S-RELAP5 for the system calculation (includes ICECON for containment response).

• AUTORLBLOCA for generation of ranged parameter values, transient input, transient runs, and general output documentation.

The governing two-fluid (plus non-condensibles) model with conservation equations for mass, energy, and momentum transfer is used. The reactor core is modeled in S-RELAP5 with heat generation rates determined from reactor kinetics equations (point kinetics) with reactivity feedback, and with actinide and decay heating.

The two-fluid formulation uses a separate set of conservation equations and constitutive relations for each phase. The effects of one phase on the other are accounted for by interfacial friction, and heat and mass transfer interaction terms in the equations. The conservation equations have the same form for each phase; only the constitutive relations and physical properties differ.

The modeling of plant components is performed by following guidelines developed to ensure accurate accounting for physical dimensions and that the dominant phenomena expected during AREVA NP Inc.

A AIRE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 19 the LBLOCA event are captured. The basic building blocks for modeling are hydraulic volumes for fluid paths and heat structures for heat transfer. In addition, special purpose components exist to represent specific components such as the RCPs or the steam generator separators.

All geometries are modeled at the resolution necessary to best resolve the flow field and the phenomena being modeled within practical computational limitations.

System nodalization details are shown in Figures 3-1 through 3-5. A point of clarification: in Figure 3-1, break modeling uses two junctions regardless of break type-split or guillotine; for guillotine breaks, Junction 151 is deleted, it is retained fully open for split breaks. Hence, total break area is the sum of the areas of both break junctions.

A typical calculation using S-RELAP5 begins with the establishment of a steady-state initial condition with all loops intact. The input parameters and initial conditions for this steady-state calculation are chosen to reflect plant operational characteristics or to match measured data.

Additionally, the RODEX3A code provides initial conditions for the S-RELAP5 fuel models.

Specific parameters are discussed in Section 3.3.

Following the establishment of an acceptable steady-state condition, the transient calculation is initiated by introducing a break into one of the loops (specifically, the loop with the pressurizer).

The evolution of the transient through blowdown, refill and reflood is computed continuously using S-RELAP5. Containment pressure is also calculated by S-RELAP5 using containment models derived from ICECON (Reference 4), which is based on the CONTEMPT-LT code (Reference 5).

The methods used in the application of S-RELAP5 to the LBLOCA are described in Reference 1. A detailed assessment of this computer code was made through comparisons to experimental data, many benchmarks with cladding temperatures ranging from 1,700°F (or less) to above 2,200 0 F. These assessments were used to develop quantitative estimates of the ability of the code to predict key physical phenomena in a PWR LBLOCA. Various models-for example, the core heat transfer, the decay heat model and the fuel cladding oxidation correlation-are defined based on code-to-data comparisons and are, hence, plant independent.

The RV internals are modeled in detail (Figures 3-3 through 3-5) based on specific inputs supplied by SCE. Nodes and connectivity, flow areas, resistances and heat structures are all accurately modeled. The location of the hot assembly/hot pin(s) is unrestricted; however, the channel is always modeled to restrict appreciable upper plenum liquid fallback. The final step of the best-estimate methodology is to combine all the uncertainties related to the code and plant parameters, and estimate the PCT at a high probability level. The steps taken to derive the PCT uncertainty estimate are summarized below:

AREVA NP Inc.

A A:RiE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 20

1. Base Plant Input File Development First, base RODEX3A and S-RELAP5 input files for the plant (including the containment input file) are developed. Code input development guidelines are applied to ensure that model nodalization is consistent with the model nodalization used in the code validation.

2. Sampled Case Development The non-parametric statistical approach requires that many "sampled" cases be created and processed. For every set of input created, each "key LOCA parameter" is randomly sampled over a range established through code uncertainty assessment or expected operating limits (provided by plant Technical Specifications or data). Those parameters considered "key LOCA parameters" are listed in Table 3-1. This list includes both parameters related to LOCA phenomena (based on the PIRT provided in Reference 1) and to plant operating parameters.

3. Determination of Adequacy of ECCS The RLBLOCA methodology uses a non-parametric statistical approach to determine values of PCT at the 95% probability level. Total oxidation and total hydrogen are based on the limiting PCT case. The adequacy of the ECCS is demonstrated when these results satisfy the criteria set forth in Section 3.0.

3.3 Plant Descriptionand Summary of Analysis Parameters The plant analysis presented in this report is for a CE-designed PWR, which has 2X4-loop arrangement. There are two hot legs each with a U-tube steam generator and four cold legs each with an RCP 6 . The RCS includes one Pressurizer connected to a hot leg. The core contains 217 thermal-hydraulic compatible AREVA NP HTP 16X16 fuel assemblies with 2%,

4%, 6% and 8% Gadolinia pins. The break is modeled in the same loop as the pressurizer, as directed by the RLBLOCA methodology. The RLBLOCA transients are of sufficiently short duration that the switchover to sump cooling water (i.e., RAS) for ECCS pumped injection need not be considered The S-RELAP5 model explicitly describes the RCS, RV, Pressurizer, and ECCS. The ECCS includes one HPSI, one LPSI and one SIT injection path per RCS loop.. The HPSI and LPSI feed into a common header that connects to each cold leg pipe downstream of the RCP discharge. The ECCS pumped injection is modeled as a table of flow versus backpressure.

This model also describes the secondary-side steam generator that is instantaneously isolated (closed MSIV and feedwater trip) at the time of the break. A symmetric steam generator tube plugging level of 8% per steam generator was assumed.

6 The RCPs are Byron-Jackson Type DFSS pumps as specified by SCE. The homologous pump performance curves were input to the S-RELAP5 plant model; the built-in S-RELAP5 curves were not used.

AREVA NP Inc.

A A RE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 21 Plant input modeling parameters were provided by SCE specifically for the SONGS Unit 2 and Unit 3. By procedure, SCE maintains plant documentation current, and directly communicates with AREVA NP on plant design and operational issues regarding reload cores. SCE and AREVA NP will continue to interact in that fashion regarding the use of AREVA NP fuel in the SONGS Unit 2 and Unit 3. Both entities have ongoing processes that assure the ranges and values of input parameters for the SONGS Unit 2 and Unit 3 RLBLOCA analysis bound those of the as-operated plant.

As described in the AREVA NP RLBLOCA methodology, many parameters associated with LBLOCA phenomenological uncertainties and plant operation ranges are sampled. A summary of those parameters is given in Table 3-1. The LBLOCA phenomenological uncertainties are provided in Reference 1. Values for process or operational parameters, including ranges of sampled process parameters, and fuel design parameters used in the analysis are given in Table 3-2. Plant data are analyzed to develop uncertainties for the process parameters sampled in the analysis. Table 3-3 presents a summary of the uncertainties used in the analysis. Two parameters, RWST temperature for ECCS flows and diesel start time, are set at conservative bounding values for all calculations. Where applicable, the sampled parameter ranges are based on Technical Specifications limits or supporting plant calculations that provide more bounding values.

For the AREVA NP RLBLOCA EM, dominant containment parameters, as well as NSSS parameters, were established via a PIRT process. Other model inputs are generally taken as nominal or conservatively biased. The PIRT outcome yielded two important (relative to PCT) containment parameters-containment pressure and temperature. In many instances, the conservative guidance of CSB 6-2 (Reference 6) was used in setting the remainder of the containment model input parameters. As noted in Table 3-3, containment temperature is a sampled parameter. Containment pressure response is indirectly ranged by sampling the containment volume (Table 3-3). In accordance with Reference 1, the condensing heat transfer coefficient is intended to be closer to a best-estimate instead of a bounding high value. A [ ]

Uchida heat transfer coefficient multiplier was specifically validated for use in SONGS Unit 2 and Unit 3 through application of the process used in the RLBLOCA EM (Reference 1) sample problems.

The containment initial conditions and boundary conditions are given in Table 3-8. The building spray is modeled at maximum heat removal capacity. Passive heat sink parameters are listed in Table 3-9. Material Properties for passive heat sinks in the containment are listed in Table 3-10.

AREVA NP Inc.

A A-R: EV-A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 22 3.4 SER Compliance A number of requirements on the methodology are stipulated in the conclusions section of the SER for the RLBLOCA methodology (Reference 1). Assessment of the 13 analysis-related SER restrictions concludes that all requirements are met for the SONGS Unit 3 RLBLOCA analysis for Cycle 17 and beyond unless future changes mandate further evaluation. A summary discussion of the assessments and conclusions is provided in Table 3-4.

3.5 Realistic Large Break LOCA Results Two case sets of 59 transient calculations were performed sampling the parameters listed in Table 3-1 and the results of the limiting PCT case are reported. For each case set, PCT was calculated for a U0 2 rod and for Gadolinia-bearing rods with concentrations of 2, 4, 6 and 8 w/o Gd 2 O 3. The limiting case set, that contained the PCT, was the set with no offsite power available. The limiting PCT (1605 0 F) occurred in Case 7 for a fresh 6% Gad rod 7. From the same case, the limiting PCT for a once-burnt rod is 1517 0 F and occurred for a U0 2 rod. The fraction of total hydrogen generated was not directly calculated; however, it is conservatively bounded by the calculated total percent oxidation, which is well below the 1% limit. The best-estimate (median) PCT case is Case 11, which corresponded to the median case out of the 59-case set with no offsite power available. The nominal PCT was 1297 0 F for a 4% burnt Gd 2 0 3 rod. This result can be used to quantify the relative conservatism in the limiting case result. In this analysis, it was 3080 F.

The case results, event times and analysis plots for the limiting PCT case are shown in Table 3-6, and in Figures 3-12 through 3-23. The reference Level Zero for the liquid level in the reactor vessel is the bottom of the downcomer, which corresponds to the bottom of the lower support plate. Figure 3-6 shows linear scatter plots of the key parameters sampled for the 59 calculations. Parameter labels appear to the left of each individual plot. These figures show the parameter ranges used in the analysis. Figure 3-8 and Figure 3-9 show scatter plots for the time of occurrence of PCT and break size versus PCT for the 59 calculations, respectively.

Figure 3-10 and Figure 3-11 show the maximum oxidation and total oxidation versus PCT scatter plots for the 59 calculations, respectively. Key parameters for the limiting PCT case are shown in Figures 3-12 through 3-23. Figure 3-11 is the plot of PCT independent of elevation; this figure clearly indicates that the transient exhibits a sustained and stable quench. A comparison of PCT results from both case sets is shown in Figure 3-23.

7 The PCT for fresh fuel for the U0 2 only rod (Case 7) was 1541°F.

AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 23 Table 3-1 Sampled LBLOCA Parameters Phenomenological Time in cycle (peaking factors, axial shape, rod properties, burnup)

Break type (guillotine versus split)

Critical flow discharge coefficients (break)

Decay heat 8 Critical flow discharge coefficients (surgeline)

Initial upper head temperature Film boiling heat transfer Dispersed film boiling heat transfer Critical heat flux Tmin (intersection of film and transition boiling)

Initial stored energy Downcomer hot wall effects Steam generator inlet plenum interfacial effects 9 Condensation interphase heat transfer coefficient 9 Core Power 9 Metal-water reaction Plant 10 Offsite power availability11 Break size Pressurizer pressure Pressurizer liquid level SIT pressure SIT liquid level SIT temperature (based on containment temperature)

Containment temperature Containment volume Initial RCS flow rate Initial operating RCS temperature Diesel start (for loss of offsite power only) 8 Not sampled in analysis, multiplier set to 1.0.

9 Not sampled in analysis.

10 Uncertainties for plant parameters are based on typical plant-specific data.

11 Not sampled, see Section 4.9.

AREVA NP Inc.

A A;R.EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 24 Table 3-2 Plant Operating Range Supported by the LOCA Analysis Event Operating Range 1.0 Plant Physical Description 1.1 Fuel a) Cladding outside diameter 0.382 in.

b) Cladding inside diameter 0.332 in.

c) Cladding thickness 0.025 in.

d) Pellet outside diameter 0.3255 in.

e) Pellet density 96 percent of theoretical f) Active fuel length 150.0 in.

g) Resinter densification [ I h) Gd 2 0 3 concentrations 2%, 4%, 6%, 8%

1.2 RCS a) Flow resistance Analysis b) Pressurizer location Analysis assumes location giving most limiting PCT (broken loop) c) Hot assembly location Anywhere in core d) Hot assembly type 16X16 AREVA NP HTP fuel e) SG tube plugging 8% 12 2.0 Plant Initial Operating Conditions 2.1 Reactor Power a) Nominal reactor power 3458 MWt 13 b) LHR 12.8 kW/ft c) F0 2.37 d) F, 1.75514 2.2 Fluid Conditions a) Loop flow 376,200 gpm _ M _<443,520 gpm b) RCS Cold Leg temperature 533.0°F < T < 560.0°F c) Pressurizer pressure 2000 psia _ P _< 2300 psia d) Pressurizer level 22 percent < L < 61 percent e) SIT pressure 595 psia _<P < 675 psia t') SIT liquid volume 1650 ft 3 < V* 1825 ft3 50 0F*< T*< 130OF g) SIT temperature '(Coupled with containment temperature) 12 In the RLBLOCA analysis, only the maximum 8% tube plugging in each steam generator was analyzed. By independently sampling the break loss discharge coefficients, any flow differences attributed to asymmetry in the SG tube plugging is covered by use of the RLBLOCA methodology.

13 Includes 20 MWt uncertainties 14 The radial power peaking for the hot rod is including 6% measurement uncertainty and 3.5% allowance for control rod insertion affect.

Fr imit = Fr *(1+ uncertFr) * (1+uncert_cr_insertion) = 1.6*(1.0+0.06)*(1+0.035)=1.755 AREVA NP Inc.

A AR:E"I-V'A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 25 Event Operating Range h) SIT resistance fL/D As-built piping configuration i) Minimum ECCS boron _>2150 ppm 3.0 Accident Boundary Conditions a) Break location Cold leg pump discharge piping b) Break type Double-ended guillotine or split 0.29 < A < 1.0 full pipe area (split) c) Break size (each side, relative to cold leg pipe area) 0.29!*A:* 1.0 full pipe area (guit) 0.29:< A:!< 1.0 full pipe area (guillotine) d) Worst single-failure Loss of one emergency diesel generator e) Offsite power On or Off f) ECCS pumped injection temperature 110°F g) HPSI pump delay 31.0 (w/ offsite power) 46.0 (w/o offsite power) h) LPSI pump delay 31.2 (w/ offsite power) 46.2 (w/o offsite power) i) Containment pressure 14.7 psia, nominal value 15 j) Containment temperature 50°F < T < 130OF k) HPSI flow BROKENLOOP 1A AND INTACT LOOPS 1B, 2A & 2B RCS pressure HPSI flow (psia) (gpm) 0.0 200.0 150.0 200.0 200.0 195.0 400.0 175.1 600.0 151.8 700.0 139.6 800.0 127.0 900.0 114.0 1000.0 98.9 1100.0 80.8 1200.0 57.2 1276.0 23.8 1277.0 0.0 5000.0 0.0 1 Nominal containment pressure range is -0.9 to +2.1 psig. For RLBOCA, a reasonable value in this range is acceptable.

AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 26 Event I ODeratina Ranae I) LPSI flow BROKEN LOOP 1A & INTACT LOOP 2B RCS pressure LPSI flow (psia) (gpm) 0.0 2229.9 10.0 2167.3 20.0 2104.1 30.0 2038.5 40.0 1972.6 50.0 1902.6 60.0 1830.8 70.0 1748.2 80.0 1667.9 90.0 1582.9 100.0 1492.8 110.0 1397.5 120.0 1293.7 130.0 1173.4 140.0 1037.4 150.0 870.0 160.0 506.8 170.0 697.0 180.0 265.5 185.0 119.3 186.0 0.0 INTACT LOOPS 1B & 2A RCS pressure LPSI flow (psia) (gpm)

N/A 0.0 AREVA NP Inc.

A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 27 Table 3-3 Statistical Distributions Used for Process Parameters"6 Parameter Operational Uncertainty Parameter Range Distribution ParameterRange Pressurizer Pressure (psia) Uniform 2000 - 2300 Pressurizer Liquid Level (percent) Uniform 22 - 61 SIT Liquid Volume (ft 3) Uniform 1650 - 1825 SIT Pressure (psia) Uniform 595 - 675 Containment Temperature (OF) Uniform 50 - 130 Containment Volume (ft3) Uniform 2.305E+6 - 2.335E+6 Initial RCS Flow Rate (gpm) Uniform 376,200 - 443,520 Initial RCS Operating Temperature Uniform 533 - 560 (Tcold) (OF)

RWST Temperature for ECCS (OF) Point 110 Offsite Power Availability 17 Binary 0,1 Delay for Containment Spray (s) Point 0 31.2 (w/ offsite power)

LPSI Pump Delay (s) Point 46.2 (w/ offsite power) 46.2 (w/o offsite power)

HPSI Pump Delay (s) Point 46.0 (w/ offsite power) 46.0 (w/o offsite power) 16 Note that core power is not sampled, see Section 1.0 17 This is no longer a sampled parameter. One set of 59 cases is run with LOOP and one set of 59 cases is run with No-LOOP.

AREVA NP Inc.

'.A':

AWREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 28 Table 3-4 SER Conditions and Limitations SER Conditions and Limitations Response' A CCFL violation warning will be added to alert the analyst An occurrence-based scatter plot of downcomer velocities to CCFL violation in the downcomer should such occur. versus the applicable CCFL velocity (as Wallis parameters) was generated for all cases. These results were examined for gross CCFL violation (i.e., more than 50% of the data appearing above the Wallis model flooding line).

There was no significant occurrence of CCFL violation in the downcomer for this evaluation.

2. AREVA NP will not include the hot leg nozzle gaps in plant Hot leg nozzle gaps were not modeled.

models to which they apply the S-RELAP5 code.

3. If AREVA NP applies the RLBLOCA methodology to plants The PLHGR for SONGS Unit 2 and Unit 3 is lower than that using a higher planar linear heat generation rate (PLHGR) used in the development of the RLBLOCA EM than used in the current analysis, or if the methodology is to (Reference 1). An end-of-life calculation was not performed.

be applied to an end-of-life analysis for which the pin However, an assessment for SONGS against rupture criteria pressure is significantly higher, then the need for a was performed, which concluded that for the SONGS blowdown clad rupture model will be reevaluated. The RLBLOCA analysis, cladding rupture prior to the initiation of evaluation may be based on relevant engineering reflood does not occur for either the first or second cycle experience and should be documented in either the fuel.

RLBLOCA guideline or plant specific calculation file.

4. Slot breaks on the top of the pipe have not been evaluated. The AREVA NP PWR analysis guidelines provide detailed These breaks could cause the loop seals to refill during late discussion on the generic treatment of top slot breaks. For reflood and the core to uncover again. These break SONGS, the top of the cross-over piping is at a reference locations are an oxidation concern as opposed to a PCT elevation of -64.62 inches and the top of the active core is at concern since the top of the core can remain uncovered for a reference elevation of -77 inches. No additions to the extended periods of time. Should an analysis be performed calculation notebook or Design Report are required.

for a plant with spillunder (Top crossover pipe (ID) at the crossover pipes lowest elevation) that are below the top elevation of the core, AREVA NP will evaluate the effect of the deep loop seal on the slot breaks. The evaluation may be based on relevant engineering experience and should be documented in either the RLBLOCA guideline or plant-specific calculation file.

5. The model applies to 3 and 4 loop Westinghouse- and SONGS is a CE-designed 2X4 loop plant and the RLBLOCA CE-designed nuclear steam systems. EM applies to this type of plant.

6. The model applies to bottom reflood plants only (cold side SONGS is a bottom reflood plant and the RLBLOCA EM injection into the cold legs at the reactor coolant discharge applies to this type of plant.

piping).

7. The model is valid as long as blowdown quench does not The case set was examined and blowdown quench was not occur. If blowdown quench occurs, additional justification an issue in the SONGS Unit 2 and Unit 3 RLBLOCA for the blowdown heat transfer model and uncertainty are uncertainty analysis.

needed or the calculation is corrected. A blowdown quench is characterized by a temperature reduction of the peak cladding temperature (PCT) node to saturation temperature during the blowdown period.

8. The reflood model applies to bottom-up quench behavior. If The CCFL model is applied on all core exit junctions as a a top-down quench occurs, the model is to be justified or provision to prevent top-down quench.

corrected to remove top quench. A top-down quench is No top-down quench effects are observed in the SONGS characterized by the quench front moving from the top to Unit 2 and Unit 3 RLBLOCA uncertainty analysis.

the bottom of the hot assembly.

AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 29 SER Conditions and Limitations Response

9. The model does not determine whether Criterion 5 of 10 Long-term cooling was not evaluated in this analysis.

CFR 50.46, long term cooling, has been satisfied. This will be determined by each applicant or licensee as part of its application of this methodology.

10. Specific guidelines must be used to develop the The nodalization in the plant model is similar with the plant-specific nodalization. Deviations from the reference Westinghouse 3-loop sample calculations that were plant must be addressed. submitted to the NRC for review with a slight deviation in the upper plenum nodalization. This deviation should not impact the acceptability of the nodalization.

Figure 3-1 shows the loop noding used in this analysis.

(Note only Loop 1 is shown in the figure; Loop 2 is identical to loop 1, except that only Loop 1 contains the pressurizer and the break.) Figure 3-2 shows the steam generator model. Figures 3-3, 3-4, and 3-5 show the reactor vessel noding diagrams.

11. A table that contains the plant-specific parameters and the Simulation of clad temperature response is a function of range of the values considered for the selected parameter phenomenological correlations that have been derived either during the topical report approval process must be analytically or experimentally. The important correlations provided. When plant-specific parameters are outside the have been validated for the RLBLOCA methodology and a range used in demonstrating acceptable code performance, statement of the range of applicability has been the licensee or applicant will submit sensitivity studies to documented. The correlations of interest are the set of heat show the effects of that deviation, transfer correlations as described in Reference 1. Table 3-3 presents the summary of the full range of applicability for the important heat transfer correlations, as well as the ranges calculated in the limiting case of this analysis. Calculated values for other parameters of interest are also provided.

As is evident, the plant-specific parameters fall within the methodology's range of applicability.

12. The licensee or applicant using the approved methodology The design report presents the results of the calculations in must submit the results of the plant-specific analyses, accordance with the sample problem.

including the calculated worst break size, PCT, and local and total oxidation.

13. The licensee or applicant wishing to apply AREVA NP SONGS Unit 2 and Unit 3 plants have 16x16 HTP fuel with realistic large break loss-of-coolant accident (RLBLOCA) M5 cladding. SCE has temporary exemption approved for methodology to M5 clad fuel must request an exemption for Lead Fuel Assemblies (LFAs) with M5 cladding and is its use until the planned rulemaking to modify 10 CFR pursuing permanent exemption for unrestricted use of 50.46(a)(i) to include M5 cladding material has been AREVA fuel.

completed.

AREVA NP Inc.

A ARE:V:A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 30 Table 3-5 Summary of Results for the Limiting PCT Case Case #7 Fresh Fuel Once-Burnt (Offsite power unavailable) 6% Gad Rod U0 2 Rod PCT Temperature 1605°F 1517 0 F Time 8.1 s 8.Os Elevation 10.158 ft 10.158 ft Metal-Water Reaction Pre-transient Local Oxidation (%) 0.9085 1.9685 Transient Local Oxidation Maximum (%) 0.4142 0.3520 Total Local Oxidation Maximum (%) 1.3227 2.3205 Total Core-Wide Oxidation (%) < 0.01 <0.01 Table 3-6 Calculated Event Times for the Limiting PCT Case Event Time (s)

Break Opened 0.0 RCP Trip N/A SIAS Issued 0.5 Start of Broken Loop SIT Injection 12.2 Start of Intact Loop SIT Injection 13.0, 13.0, and 13.0 (Loops 2, 3 and 4 respectively)

Broken Loop LPSI Delivery Began 46.7 Intact Loop LPSI Delivery Began N/A, N/A and 46.7 (Loops 2, 3 and 4 respectively)

Broken Loop HPSI Delivery Began 46.5 Intact Loop HPSI Delivery Began 46.5, 46.5 and 46.5 (Loops 2, 3 and 4 respectively)

Beginning of Core Recovery (Beginning of Reflood) 25.6 Broken Loop SIT Emptied 79.3 Intact Loop SITs Emptied (Loops 2, 3 and 4 respectively) 73.9, 73.9 and 79.2 PCT Occurred 8.1 Transient Calculation Terminated 500.0 AREVA NP Inc.

A A RE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 31 Table 3-7 Heat Transfer Parameters for the Limiting Case AREVA NP Inc.

A A RiE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 32 Table 3-7 Heat Transfer Parameters for the Limiting Case (continued)

AREVA NP Inc.

A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 33 Table 3-8 Containment Initial and Boundary Conditions Containment Net Free Volume (ft3) 2.305E+6 - 2.335E+6 Initial Conditions Containment Pressure (nominal) 14.7 psia Containment Temperature 50°F - 130OF Outside Temperature 70°F Humidity 1.0 Containment Spray Number of Pumps operating 2 Spray Flow Rate (Total, both pumps) 4400 gpm Minimum Spray Temperature 35 0 F Fastest Post-LOCA initiation of spray 0s Containment Emergency Cooling Units Number of Emergency Cooling Units 4 Operating Minimum Post Accident Initiation Time of 0 Emergency Cooling Units (sec)

Emergency Cooling Units Capacity (1 Unit)

Containment Temperature (OF) Heat Removal Rate (BTU/hr) 105 0.0 150 18.9E+6 200 39.4E+6 250 60.7E+6 300 82.5E+6 AREVA NP Inc.

A A R E.VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 34 23 Table 3-9 Passive Heat Sinks in Containment Geometry Heat Sink Area, ft 2 Thickness, ft Material Containment Cylinder and Dome -Steel 76383 0.0470 Carbon Steel Containment Cylinder and Dome -Concrete 4.1826 Concrete Unlined Internal Concrete 75676 1.9489 Concrete Internal Concrete - CS lined (Steel) 30154 0.0227 Carbon Steel Internal Concrete - CS lined (Concrete) 0.9653 Concrete Internal Concrete - SS lined (Steel) 0.0156 Stainless Steel Internal Concrete - SS lined (Concrete) 2.0 Concrete Galvanized Steel 144498 0.0086 Galvanized Steel Stainless Steel 47653 0.0091 Stainless Steel Carbon Steel < 0.5 in 161899 0.0225 Carbon Steel Carbon Steel > 0.5 in 76417 0.0606 Carbon Steel Glass Part of Lighting 660 0.0104 Glass Copper Sheathed Cables 1192 0.0015 Copper Polyethylene Sheathed 14522 0.0075 Polyethylene Basemat 13207 10.9021 Concrete Miscellaneous SS 15000 0.016 Stainless Steel Miscellaneous CS 15000 0.0193 Carbon Steel 23 Passive heat sinks data listed in the table were used for RLBOCA analysis. Sensitivity studies were previously performed for the AREVA RLBLOCA Transition Package as applied to EMF-2103 to respond to the NRC's concerns. The results showed for a large dry containment, the PCT is not sensitive to change in containment back pressure. Hence, the heat sinks changes within +/-5% range will not change the presented RLBLOCA results.

AREVA NP Inc.

A AR E~vA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 35 Table 3-10 Material Properties for Passive Heat Sinks in Containment Thermal Volumetric Heat Heat Sink Conductivity Capacity (BTU/hr-ft-°F) (BTU/ft 3-OF)

Concrete 1.0 33.0 Carbon Steel 27.0 58.87 Stainless Steel 9.0 61.25 Galvanized Steel 27.0 58.87 Paint 0.3 33.27 Copper 225.0 51.15 Glass 0.625 25.9 Polyethylene 0.19 29.6 AREVA NP Inc.

A A.R.EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 36 Figure 3-1 Primary System Noding for SONGS AREVA NP Inc.

A AR:E VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 37 Figure 3-2 Secondary System Noding AREVA NP Inc.

A AR;EVA' San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 38 Figure 3-3 Reactor Vessel Noding AREVA NP Inc.

A, AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 39 Figure 3-4 Core Noding Detail AREVA NP Ind.

.A.

ARE:VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 40 Figure 3-5 Upper Plenum Noding Detail AREVA NP Inc.

A A REVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 41 One-Sided Break Area (ft2/side)

Burn Time (hours)

Core Power (MW) 3457.0 3457.5 3458.0 3458.5 3459.0 3459.5 3460.0 LHGR (KW/ft)

(Krf)I 10.0 10.5 m

I 11.0 O memne I

11.5 I

12.0 n ma 12.5

]

13.0 ASI r .m u 4O0 MIS

-0.2 -0.1 0.0 0.1 0.2 Pressurizer Pressure (psia) 2000.0 0

2100.0 0 . O 2200.0 0Me 0 H

6 2300.0 Pressurizer I I 2 Liquid Level 1 oooo 4OONM O (M1 210.0 30.0 40.0 50.0 60.0 70.0 RCS (Tcold)

Temperature

(°F) 530.0 r mNi 540.0 Inm m mummmeo 550.0 560.0 Figure 3-6 Scatter Plot of Operational Parameters AREVA NP Inc.

A AR-E-VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 42 Total Loop Flow (Mlb/hr)

SIT Liquid Volume

• emmm eeernemmemeeeaael (ft')

1650.0 1700.0 1750.0 1800.0 1850.0 SIT I.

Pressure em e Oemin me *seeme *

(psia) 580.0 600.0 620.0 640.0 660.0 680.0 Containment Volume 3 o

• e em mm (ft )

2.30e+06 2.31e+06 2.32e+06 2.33e+06 2.34e+06 Temperaturem i mom oo /mi ** o 40.0 60.0 80.0 100.0 120.0 140.0 Figure 3-6 Scatter Plot of Operational Parameters (Continued)

AREVA NP Inc.

A AR.E, V/A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 43 PCT vs Time of PCT 2000 1800 1600 1400 U-H 1200 1000 800 600 400 0 100 200 300 400 500 Time of PCT (s)

Figure 3-7 Scatter Plot of PCT versus PCT Time AREVA NP Inc.

A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 44 PCT vs One-sided Break Area 2000 1800 F 1600 k U E E

• m 1400 1 mms m E] D U []

E-

4. N DEm 0 LI-o 1200 F 1000 F 800 k 600 M Split Break El Guillotine Break 400 1.0 2.0 3.0 4.0 5.0 Break Area (ft2/side)

Figure 3-8 Scatter Plot of PCT versus Break Size AREVA NP Inc.

AiRE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 45 Maximum Oxidation vs PCT 1.0

• Split Break 0.9 [] Guillotine Break 0.8 h 0.7 0.6 I F

0 0.5 F Ox E

0.4 0.3 EU Non 0.2 0.1 LJnj 0.0 .

I

. m * .

E . . . . . . . . .

400 600 800 1000 1200 1400 1600 1800 2000 PCT (OF)

Figure 3-9 Scatter Plot of Maximum Transient Oxidation versus PCT AREVA NP Inc.

A AR'EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 46 Total Oxidation vs PCT 0.020 U Split Break 0.018 El Guillotine Break 0.016 0.014 0.012 C

0 0.010

-0 0

0.008 0.006 0.004 EP

• !m.

0.002 ED I.l ,,g

'u-inm 0.000 40'0 600 800 1000 1200 1400 1600 1800 2000 PCT (°F)

Figure 3-10 Scatter Plot of Total Oxidation versus PCT AREVA NP Inc.

A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 47 PCT Trace for Case #7 PCT = 1605.0 OF, at Time = 8.06 s, on 6% Gad Rod 2000 1500 C',

E 1000 I-C.C 500 0

0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-11 Peak Cladding Temperature (Independent of Elevation) for the Limiting Case AREVA NP Inc.

A A R'EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 48 Break Flow 100 80 60 CU) 0 U- 40 20 0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-12 Break Flow for the Limiting Case AREVA NP Inc.

A A IRE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 49 Core Inlet Mass Flux 1000 500 E

U- 0

-500

-1000 0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-13 Core Inlet Mass Flux for the Limiting Case AREVA NP Inc.

A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 50 Core Outlet Mass Flux 800 600 400 E

200 0

-200 0

100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-14 Core Outlet Mass Flux for the Limiting Case AREVA NP Inc.

A:

San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 51 Pump Void Fraction 1.0 0.8 0.6 C

0 C.U IL

-o 72 0.4 Broken Loop 1

---

Intact Loop 2

- - - Intact Loop 3 0.2 4 - -- Intact Loop 4 0.0 0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-15 Void Fraction at RCS Pumps for the Limiting Case AREVA NP Inc.

A ARIEVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 52 ECCS Flows 3000 2000 E

CO 0

1000 0

0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-16 ECCS Flows (Includes SIT, HPSI and LPSI) for the Limiting Case AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 53 Upper Plenum Pressure 3000 2000 (n)

(n 1000 0-0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-17 Upper Plenum Pressure for the Limiting Case AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 54 Downcomer Liquid Level 30 20 7J 10 0

0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-18 Collapsed Liquid Level in the Downcomer for the Limiting Case AREVA NP Inc.

A AR E'VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 55 Lower Vessel Liquid Level 10 8

6 75 Z) 4 2

0 0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-19 Collapsed Liquid Level in the Lower Plenum for the Limiting Case AREVA NP Inc.

ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 56 Core Liquid Level 15 10 0

-J 5

0 0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-20 Collapsed Liquid Level in the Core for the Limiting Case AREVA NP Inc.

A.

A:R!E V.A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 57 Containment and Loop Pressures 100 90 80 70 60 M,

S0-E, 50 (n

40 30 20 10 0

0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 3-21 Containment and Loop Pressures for the Limiting Case AREVA NP Inc.

.A A.REVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 58 300000 250000 E

- 200000 (n

D 150000 100000 Id Dx 50000 0

0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX:2 Figure 3-22 Reactor Vessel Liquid Mass (Ibm) versus Time (sec)

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 59 2000 2000 SLOOP ONo LOOP I 1800 1800 1600 1600

,ea 0 1400 1400 o 0 0 0 0 . .

0 1 00 0 III000

• 0 0

0 0 0 1200 - -0 1200

0. *.

0 0e 0 00

- 0 1000 0 1000 0

.0 0

800 j 800 600 600 10 20 30 40 50 60 Case Number Figure 3-23 GDC 35 LOOP versus No-LOOP Cases AREVA NP Inc.

A AREA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 60 4.0 Generic Support for Transition Package The following sections are responses to typical RAI questions posed by the NRC on EMF-2103 Revision 0 plant applications, these responses and changes are known as the "Transition Package." In some instances, these requests cross-reference documentation provided on dockets other than those for which the request is made. AREVA NP discussed these and similar questions from the NRC (draft SER for Revision 1 of EMF-2103) in a meeting with the NRC on December 12, 2007. AREVA NP agreed to provide the following additional information within new submittals of a Realistic Large Break LOCA report.

4.1 Reactor Power Question: Reactor Power - Table 3-2, Item 2.1, and its associated Footnote I indicate that the assumed reactor core power "includes uncertainties." The use of a reactor power assumption other than 102 %, regardless of BE or Appendix K methodology, is permitted by Title 10 of the Code of Federal Regulations (10 CFR), Part 50, Appendix K.I.A, "Required and Acceptable Features of The Evaluation Models, 'Sources of Heat During a LOCA." However, Appendix K.I.A also states: "... An assumed power level lower than the level specified in this paragraph

[1.02 times the licensed power level], (but not less than the licensed power level) may be used provided..."

Response: As indicated in Item 2.1 of Table 3-2 herein, the assumed reactor core power for the SONGS Unit and Unit 3 Realistic Large Break Loss-of-coolant Accident is 3458 MWt. This value represents the 100% primary power plus 20 MWt measurement uncertainty.

4.2 Rod Quench Question: Does the version of S-RELAP5 used to perform the computer runs assure that the void fraction is less than 95 % and the fuel cladding temperature is less than 900 *F before it allows rod quench?

Response: Yes, the version of S-RELAP5 employed for the SONGS Unit 2 and Unit 3 LAR requires that both the void fraction is less than 0.95 and the clad temperature is less than the minimum temperature for film boiling heat transfer (Tmin) before the rod is allowed to quench.

Tmin is a sampled parameter in the RLBLOCA methodology. For SONGS Unit 2 and Unit 3 case set, the mean value (lower bound) of Tmin is 626 K with standard deviation (upper bound) of 33.6 K, making it very unlikely that Tmin would exceed 755 K (900'F). Therefore, Tmin was never sampled above 696 K (793.7 0 F). This is a change to the approved RLBLOCA EM (Reference 1).

AREVA NP Inc.

A A-R'EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 61 4.3 Rod-to-Rod Thermal Radiation Question: Providejustification that the S-RELAP5 rod-to-rodthermal radiationmodel applies to the SONGS core.

Response: The Realistic LBLOCA methodology, (Reference 1), does not provide modeling of rod-to-rod radiation. The fuel rod surface heat transfer processes included in the solution at high temperatures are: film boiling, convection to steam, rod-to-liquid radiation and rod-to-vapor radiation. This heat transfer package was assessed against various experimental data sets involving both moderate (1600'F - 2000 0 F) and high (2000°F to over 2200 0 F) peak cladding temperatures and shown to be conservative when applied nominally. The normal distribution of the experimental data was then determined. During the execution of an RLBLOCA evaluation, the heat transferred from a fuel rod is determined by the application of a multiplier to the nominal heat transfer model. This multiplier is determined by a random sampling of the normal distribution of the experimental data benchmarked. Because the data include the effects of rod-to-rod radiation, it is reasonable to conclude that the modeling implicitly includes an allocation for rod-to-rod radiation effects. As will be demonstrated, the approach is reasonable because the conditions within actual limiting fuel assemblies assure that the actual rod-to-rod radiation is larger than the allocation provided through normalization to the experiments.

The Full-Length Emergency Core Heat Transfer Separate Effects and Systems Effects Tests (FLECHT-SEASET) tests evaluated covered a range of PCTs from 1,651 to 2,2390 F and the Thermal Hydraulic Test Facility (THTF) tests covered a range of PCTs from 1,000 to 2,2000 F.

Since the test bundle in either FLECHT-SEASET or THTF is surrounded by a test vessel, which is relatively cool compared to the heater rods, substantial radiation from the periphery rods to the vessel wall can occur. The rods selected for assessing the RLBLOCA reflood heat transfer package were chosen from the interior of the test assemblies to minimize the impact of radiation heat transfer to the test vessel. The result was that the assessment rods comprise a set which is primarily isolated from cold wall effects by being surrounded by powered rods at reasonably high temperatures.

As a final assessment, three benchmarks independent of THTF and FLECHT-SEASET were performed. These benchmarks were selected from the Cylindrical Core Test Facility (CCTF),

Loss of Fluid Test (LOFT), and the Semiscale facilities. Because these facilities are more integral tests and together cover a wide range of scale, they also serve to show that scale effects are accommodated within the code calculations.

The results of these calculations are provided in Section 4.3.4, Evaluation of Code Biases, page 4-100, of Reference 1. The CCTF results are shown in Figures 4.180 through 4.192, the LOFT results in Figures 4.193 through 4.201, and the Semiscale results in Figures 4.202 through 4.207 (Reference 1). As expected, these figures demonstrate that the comparison between the code calculations and data is improved with the application of the derived biases. The CCTF, LOFT, and Semiscale benchmarks further indicate that, whatever consideration of rod-to-rod radiation is implicit in the S-RELAP5 reflood heat transfer modeling, it does not significantly AREVA NP Inc.

A AAR .EVVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 62 effect code predictions under conditions where radiation is minimized. The measured PCTs in these assessments ranged from approximately 1,000 to 1,5400F. At these temperatures, there is little rod-to-rod radiation. Given the good agreement between the biased code calculations and the CCTF, LOFT, and Semiscale data, it can be concluded that there is no significant over prediction of the total heat transfer coefficient.

Notwithstanding any conservatism evidenced by experimental benchmarks, the application of the model to commercial nuclear power plants provides some additional margins due to limitations within the experiments. The benchmarked experiments, FLECHET-SEASET and ORNL Thermal Hydraulic Test Facility (THTF), used to assess the S-RELAP5 heat transfer model, Reference 1, incorporated constant rod powers across the experimental assembly.

Temperature differences that occurred were the result of guide tube, shroud or local heat transfer effects. In the operation of a pressurized water reactor (PWR) and in the RLBLOCA evaluation, a radial local peaking factor is present, creating power differences that tend to enhance the temperature differences between rods. In turn, these temperature differences lead to increases in net radiation heat transfer from the hotter rods. The expected rod-to-rod radiation will likely exceed that embodied within the experimental results.

4.3.1 Assessment of Rod-to-Rod Radiation Implicit in the RLBLOCA Methodology As discussed above, the FLECHT-SEASET and THTF tests were selected to assess and determine the S-RELAP5 code heat transfer bias and uncertainty. A uniform radial power distribution was used in these test bundles. Therefore, the rod-to-rod temperature variation in the rods away from the vessel wall is caused primarily by the variation in the sub-channel fluid conditions. In the real operating fuel bundle, on the other hand, there can be 5- to 10-percent rod-to-rod power variation. In addition, the methodology includes a provision to apply the uncertainty measurement to the hot pin. Table 4-1 provides the hot pin measurement uncertainty and a representative local pin peaking factor for several plants. These factors, however, relate the pin to the assembly average. To more properly assess the conditions under which rod-to-rod radiation heat transfer occurs, a more local peaking assessment is required.

Therefore, the plant rod-to-rod radiation assessments herein set the average pin power for those pins surrounding the hot pin at 96% of that of the peak pin. For pins further removed the average power is set to 94%.

AREVA NP Inc.

A AR.EiV-A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 63 Table 4-1 Typical Measurement Uncertainties and Local Peaking Factors FAH Measurement Local Pin Peaking Plant Uncertainty Factor (percent) 1 4.0 1.068 2 4.0 1.050 3 6.0 1.149 4 4.0 1.113 5 4.25 1.135 6 4.0 1.058 4.3.2 Quantification of the Impact of Thermal Radiation using R2RRAD Code The R2RRAD radiative heat transfer model was developed by Los Alamos National Laboratory (LANL) to be incorporated in the BWR version of the TRAC code. The theoretical basis for this code is given in References 7 and 8 is similar to that developed in the HUXY rod heatup code (Reference 9, Section 2.1.2) used by AREVA NP for BWR LOCA applications. The version of R2RRAD used herein was obtained from the NRC to examine the rod-to-rod radiation characteristics of a 5x5 rod segment of the 161 rod FLECHT-SEASET bundle. The output provided by the R2RRAD code includes an estimate of the net radiation heat transfer from each rod in the defined array. The code allows the input of different temperatures for each rod as well as for a boundary surrounding the pin array. No geometry differences between pin locations are allowed. Even though this limitation affects the view factor calculations for guide tubes, R2RRAD is a reasonable tool to estimate rod-to-rod radiation heat transfer.

The FLECHT-SEASET test series was intended to simulate a 17x17 fuel assembly and there is a close similarity, Table 4-2, between the test bundle and a modern 17x17 assembly.

Table 4-2 FLECHT-SEASET & 17x17 FA Geometry Parameters Design Parameter FLECHT-SEASET 17x17 Fuel Assembly Rod Pitch (in) 0.496 0.496 Fuel Rod Diameter (in) 0.374 0.374 Guide Tube Diameter (in) 0.474 0.482 Five FLECHT-SEASET tests (Reference 10) were selected for evaluation and comparison with expected plant behavior. Table 4-3 characterizes the results of each test. The 5x5 selected rod array comprises the hot rod, 4 guide tubes and 20 near adjacent rods. The simulated hot rod is rod 7J in the tests.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 64 0 0 000 Guide Tube -*

0 0 00 O C Hot Rod 0 0 Adjacent Rods 0 0 000 0 000 Figure 4-1 R2RRAD 5x5 Rod Segment Two sets of runs were made simulating each of the five experiments and one set of cases was run to simulate the RLBLOCA evaluation of a limiting fuel assembly in an operating plant. For the simulation of Tests 31805, 31504, 31021, and 30817, the thimble tube (guide tube) temperatures were set to the measured values. For Test 34420, the thimble tube temperature was set equal to the measured vapor temperature. For the first experimental simulation set, the temperature of all 21 rods and the exterior boundary was set to the measured PCT of the simulated test. For the second experimental set, the hot rod temperature was set to the PCT value and the remaining 20 rods and the boundary were set to a temperature 25°F cooler providing a reasonable measure of the variation in surrounding temperatures. To estimate the rod-to-rod radiation in a real fuel assembly at LOCA conditions and compare it to the experimental results, each of the above cases was rerun with the hot rod PCT set to the experimental result and the remaining rods conservatively set to temperatures expected within the bundle. Because peak rod powers frequently occur at fuel assembly corners away from either guide tubes or instrument tubes and for added conservatism, the guide tubes (thimble tubes) were replaced by fuel rods in the input model described above. The surrounding 24 rods were set to a temperature estimated for rods of 4% lower power. The boundary temperature was estimated based an average power 6% below the hot rod power. For both of these, the temperature estimates were achieved using a ratio of pin power to the difference in temperature between the saturation temperature and the PCT.

AREVA NP Inc.

A A R Ei VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 65 T24 rods = 0.96 * (PCT - Tsat) + Tsat and Trurrounding region = 0.94 * (PCT - Tsat) + Tsat.

Tsat was taken as 270 0 F.

Figure 4-2 shows the hot rod thermal radiation heat transfer for the two FLECHT-SEASET sets and for the plant set. The figure shows that for PCTs greater than about 1700 OF, the hot rod thermal radiation in the plant cases exceeds that of the same component within the experiments.

Table 4-3 FLECHT-SEASET Test Parameters Test Rod 7J PCT PCT HTC at PCT HCa C Steam Thimble Test aTime 2(Btu/hr- Temperature at 71 Temperature at 6-ft (OF) Time (s) ft -F) (6-ft) (OF) at 6-ft (OF) 34420 2205 34 10 1850 1850*

31805 2150 110 10 1800 1800 31504 2033 100 10 1750 1750 31021 1684 29 9 1400 1350 30817 1440 70 13 900 750

• set to steam temp AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 66 4.5-4-

o3.5-N r 3-j 2.5-0 Cr 0.5-0-

1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 PCT (TF)

Figure 4-2 Rod Thermal Radiation in FLECHT-SEASET Bundle and in a 17x17 FA 4.3.3 Rod-to-Rod Radiation Summary In summary, the conservatism of the heat transfer modeling established by benchmark can be reasonably extended to plant applications, and the plant local peaking provides a physical reason why rod-to-rod radiation should be more substantial within a plant environment than in the test environment. Therefore, the lack of an explicit rod-to-rod radiation model, in the version of S-RELAP5 applied for realistic LOCA calculations, does not invalidate the conclusion that the cladding temperature and local cladding oxidation have been demonstrated to meet the criteria of 10 CFR 50.46 with a high level of probability.

4.4 Film Boiling Heat Transfer Limit Question: In the SONGS calculations, is the Forslund-Rohsenow model contribution to the heat transfer coefficient limited to less than or equal to 15% when the void fraction is greater than or equal to 0.9?

Response: Yes, the version of S-RELAP5 employed for the SONGS RLBLOCA analysis limits the contribution of the Forslund-Rohsenow model to no more than 15% of the total heat transfer AREVA NP Inc.

A ARE V,A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 67 at and above a void fraction of 0.9. Because the limit is applied at a void fraction of 0.9, the contribution of Forslund-Rohsenow within the 0.7 to 0.9 interpolation range is limited to 15% or less. This is a change to the approved RLBLOCA EM (Reference 1).

4.5 Downcomer Boiling Question: If the PCT is greaterthan 1800°For the containment pressure is less than 30 psia, has the SONGS downcomer model been rebenchmarked by performing sensitivity studies, assuming adequate downcomer noding in the water volume, vessel wall and other heat structures?

Response: The downcomer model for the SONGS has been established generically as adequate for the computation of downcomer phenomena including the prediction of potential local boiling effects. The model was benchmarked against the UPTF tests and the LOFT facility in the RLBLOCA methodology, Revision 0 (Reference 1). Further, AREVA NP addressed the effects of boiling in the downcomer in a letter, from James Malay to U.S. NRC, April 4, 2003.

The letter cites the lack of direct experimental evidence but contains sensitivity studies on high and low pressure containments, the impact of additional azimuthal noding within the downcomer, and the influence of flow loss coefficients. Of these, the study on azimuthal noding is most germane to this question; indicating that additional azimuthal nodalization allows higher liquid buildup in portions of the downcomer away from the broken cold leg and increases the liquid driving head. Additionally, AREVA NP has conducted downcomer axial noding and wall heat release studies. Each of these studies supports the Revision 0 methodology and is documented later in this section.

This question is primarily concerned with the phenomena of downcomer boiling and the extension of the Revision 0 methodology and sensitivity studies to plants with low containment pressures and high cladding temperatures. Boiling, wherever it occurs, is a phenomenon that codes like S-RELAP5 have been developed to predict. Downcomer boiling is the result of the release of energy stored in vessel metal mass. Within S-RELAP5, downcomer boiling is simulated in the nucleate boiling regime with the Chen correlation. This modeling has been validated through the prediction of several assessments on boiling phenomenon provided in the S-RELAP5 Code Verification and Validation document (Reference 11).

AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 68

-Mhde Figure 4-3 Reactor Vessel Downcomer Boiling Diagram Hot downcomer walls penalize PCT by two mechanisms: by reducing subcooling of coolant entering the core and through the reduction in downcomer hydraulic head which is the driving force for core reflood. Although boiling in the downcomer occurs during blowdown, the biggest potential for impact on clad temperatures is during late reflood following the end of SIT injection.

At this time, there is a large step reduction in coolant flow from the ECC systems. As a result, coolant entering the downcomer may be less subcooled. When the downcomer coolant approaches saturation, boiling on the walls initiates, reducing the downcomer hydraulic static level.

With the reduction of the downcomer level, the core inlet flow rate is reduced which, depending on the existing core inventory, may result in a cladding temperature excursion or a slowing of the core cooldown rate.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 69 While downcomer boiling may impact clad temperatures, it is somewhat of a self-limiting process. If cladding temperatures increase, less energy is transferred in the core boiling process and the loop steam flows are reduced. This reduces the required driving head to support continued core reflood and reduces the steam available to heat the ECCS water within the cold legs resulting in greater subcooling of the water entering the downcomer.

The impact of downcomer boiling is primarily dependent on the wall heat release rate and on the ability to slip steam up the downcomer and out of the break. The higher the downcomer wall heat release, the more steam is generated within the downcomer and the larger the impact on core reflooding. Similarly, the quicker the passage of steam up the downcomer, the less resident volume within the downcomer is occupied by steam and the lower the impact on the downcomer average density. Therefore, the ability to properly simulate downcomer boiling depends on both the heat release (boiling) model and on the ability to track steam rising through the downcomer. Consideration of both of these is provided in the following text. The heat release modeling in S-RELAP5 is validated by a sensitivity study on wall mesh point spacing and through benchmarking against a closed form solution. Steam tracking is validated through both an axial and an azimuthal fluid control volume sensitivity study done at low pressures. The results indicate that the modeling accuracy within the RLBLOCA methodology is sufficient to resolve the effects of downcomer boiling and that, to the extent that boiling occurs, the methodology properly resolves the impact on the cladding temperature and cladding oxidation rates.

4.5.1 Wall Heat Release Rate The downcomer wall heat release rate during reflood is conduction limited and depends on the vessel wall mesh spacing used in the S-RELAP5 model. The following two approaches are used to evaluate the adequacy of the downcomer vessel wall mesh spacing used in the S-RELAP5 model.

AREVA NP Inc.

A.

San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 70 4.5.1.1 Exact Solution In this benchmark, the downcomer wall is considered as a semi-infinite plate. Because the benchmark uses a closed form solution to verify the wall mesh spacing used in S-RELAP5, it is assumed that the material has constant thermal properties, is initially at temperature Ti, and, at time zero, has one surface, the surface simulating contact with the downcomer fluid, set to a constant temperature, T0, representing the fluid temperature. Section 4.3 of Reference 12 gives the exact solution for the temperature profile as a function of time as (T(x,t) - T,) / (T, - T,) = erf {x / (2*(a t)05)}, (1) where, a is the thermal diffusivity of the material given by a = k/(p Cp),

k = thermal conductivity, p = density, Cp = specific heat, and erf{} is the Gauss error function (given in Table A-1 of Reference 9).

The conditions of the benchmark are Tj = 500'F and T. = 300TF. The mesh spacing in S-RELAP5 is the same as that used for the downcomer vessel wall in the RLBLOCA model.

Figure 4-4 shows the temperature distributions in the metal at 0.0, 100 and 300 seconds as calculated by using Equation 1 and S-RELAP5, respectively. The solutions are identical confirming the adequacy of the mesh spacing used in the downcomer wall.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 71 550 500 L. 450 CL I-- 400 E

(D

-- "dr-Closed Form, 0 s

• 350

-- Closed Form, 100 s

-4Closed Form, 300 s

- S-RELAP5, 0 s 300

-- O S-RELAP5, 100 s 0- S-RELAP5, 300 s 250 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Distance from Inner Wall, feet Figure 4-4 S-RELAP5 versus Closed Form Solution AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 72 4.5.1.2 Plant Model Sensitivity Study As additional verification, a typical 4-loop plant case was used to evaluate the adequacy of the mesh spacing within the downcomer wall heat structure. Each mesh interval in the base case downcomer vessel wall was divided into two equal intervals. Thus, a new input model was created by increasing the number of mesh intervals from 9 to 18. The following four figures show the total downcomer metal heat release rate, PCT independent of elevation, downcomer liquid level, and the core liquid level, respectively, for the base case and the modified case.

These results confirm the conclusion from the exact solution study that the mesh spacing used in the plant model for the downcomer vessel wall is adequate.

3C000.00 - --

"0 Base VSLýWall (9-meshl 18-Mesh VSL Wall 24DO000o - . . . .. . . ...... .. ... ..... . . . . . . .. .... _

12000.00

, r aA

0. 1000.DO --- - - _ _ _ _

WO 100 2320,0 4M00 Time (sec)

Figure 4-5 Downcomer Wall Heat Release - Wall Mesh Point Sensitivity AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Rev. 0 Realistic Large Break LOCA Report Page 73 2400 00 1800.00 U-0~

E I--

600,00 ln n 2400 400.0 Time (sec)

Figure 4-6 PCT Independent of Elevation - Wall Mesh Point Sensitivity AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 74 30M00

~6-

-J V

=

0*

400.0 Time (sec)

Figure 4-7 Downcomer Liquid Level - Wall Mesh Point Sensitivity AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 75 1200

-j

-j 400.0 Time (sec)

Figure 4-8 Core Liquid Level - Wall Mesh Point Sensitivity AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 76 4.5.2 Downcomer Fluid Distribution To justify the adequacy of the downcomer nodalization in calculating the fluid distribution in the downcomer, two studies varying separately the axial and the azimuthal resolution with which the downcomer is modeled have been conducted.

4.5.2.1 Azimuthal Nodalization In a letter to the NRC dated April 2003 (Reference 1), AREVA NP documented several studies on downcomer boiling. Of significance here is the study on further azimuthal break up of the downcomer noding. The study, based on a 3-loop plant with a containment pressure of approximately 30 psia during reflood, consisted of several calculations examining the affects on clad temperature and other parameters.

The base model, with 6 axial by 3 azimuthal regions, was expanded to 6 axial by 9 azimuthal regions (Figure 4-9). The base calculation simulated the limiting PCT calculation given in the EMF-2103 three-loop sample problem. This case was then repeated with the revised 6 x 9 downcomer noding.

The change resulted in an alteration of the blowdown evolution of the transient with little evidence of any affect during reflood. To isolate any possible reflood impact that might have an influence on downcomer boiling, the case was repeated with a slightly adjusted vessel-side break flow. Again, little evidence of impact on the reflood portion of the transient was observed.

The study concluded that blowdown or near blowdown events could be impacted by refining the azimuthal resolution in the downcomer but that reflood would not be impacted. Although the study was performed for a somewhat elevated system pressure, the flow regimes within the downcomer will not differ for pressures as low as atmospheric. Thus, the azimuthal downcomer modeling employed for the RLBLOCA methodology is reasonably converged in its ability to represent downcomer boiling phenomena.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 77 Base model C50 CH _D Cp D H()C Revised 9 Region Model I " CH L) " CHLQ( " (Hi)

Figure 4-9 Azimuthal Noding 4.5.2.2 Axial Nodalization The RLBLOCA methodology divides the downcomer into six nodes axially. In both 3-loop and 4-loop models, the downcomer segment at the active core elevation is represented by two equal length nodes. For most operating plants, the active core length is 12 feet and the downcomer segments at the active core elevation are each 6-feet high. (For a 14 foot core, these nodes would be 7-feet high.) The model for the sensitivity study presented here comprises a 4-loop plant with ice condenser containment and a 12 foot core. For the study, the two nodes spanning the active core height are divided in half, revising the model to include eight axial nodes. Further, the refined noding is located within the potential boiling region of the downcomer where, if there is an axial resolution influence, the sensitivity to that impact would be greatest.

AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 78 The results show that the axial noding used in the base methodology is sufficient for plants experiencing the very low system pressure characteristics of ice condenser containments.

Figure 4-10 provides the containment back pressure for the base modeling. Figures 4-11 through 4-14 show the total downcomer metal heat release rate, PCT independent of elevation, downcomer liquid level, and the core liquid level, respectively, for the base case and the modified case.

The results demonstrate that the axial resolution provided in the base case, 6 axial downcomer node divisions with 2 divisions spanning the core active region, are sufficient to accurately resolve void distributions within the downcomer. Thus, this modeling is sufficient for the prediction of downcomer driving head and the resolution of downcomer boiling effects.

3M00 2400 (D~

Cd, C1) 16 Co 8.00 0.00 L-0.0 80.0 160.0 240.0 320.0 400.0 Time (sec)

Figure 4-10 Lower Compartment Pressure versus Time AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 79 3C000.00 F-- Base 6x6 Casel w8

• x6Case ,

24000.C0 CO 1800000 2)

LUJ 12000.00 6000.00 Time (sec)

Figure 4-11 Downcomer Wall Heat Release - Axial Noding Sensitivity Study AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 80 240000 r- --- I 1S000 U-0

0. 1200,00 LE E

a, H-Time (sec)

Figure 4-12 PCT Independent of Elevation -Axial Noding Sensitivity Study AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 81 3000 20.00 10.00 0.00 1 0.0 400.0 Time (sec)

Figure 4-13 Downcomer Liquid Level - Axial Noding Sensitivity Study AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 82 12,00 10.00 a,

a,

-J V

0*

-J 4000 Time (sec)

Figure 4-14 Core Liquid Level - Axial Noding Sensitivity Study 4.5.3 Downcomer Boiling Conclusions To further justify the ability of the RLBLOCA methodology to predict the potential for and impact of downcomer boiling, studies were performed on the downcomer wall heat release modeling within the methodology and on the ability of S-RELAP5 to predict the migration of steam through the downcomer. Both azimuthal and axial noding sensitivity studies were performed. The axial noding study was based on an ice condenser plant that is near atmospheric pressure during reflood. These studies demonstrate that S-RELAP5 delivers energy to the downcomer liquid volumes at an appropriate rate and that the downcomer noding detail is sufficient to track the distribution of any steam formed. Thus, the required methodology for the prediction of downcomer boiling at system pressures approximating those achieved in plants with pressures as low as ice condenser containments has been demonstrated.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 83 4.6 Break Size Question: Were all break sizes assumed greater than or equal to 1.0 ft2 ?

Response: Yes.

The NRC has requested that the break spectrum for the realistic LOCA evaluations be limited to accidents that evolve through a range of phenomena similar to those encountered for the larger break area accidents. This is a change to the approved RLBLOCA EM (Reference 1). The larger break area LOCAs are typically characterized by the occurrence of dispersed flow film boiling at the hot spot, which sets them apart from smaller break LOCAs. This occurs generally in the vicinity of 0.2 DEGB (double-ended guillotine break) size (i.e., 0.2 times the total flow area of the pipe on both sides of the break). However, this transitional break size varies from plant to plant and is verified only after the break spectrum has been executed. AREVA NP has sought to develop sufficient criteria for defining the minimum large break flow area prior to performing the break spectrum. The purpose for doing so is to assure a valid break spectrum is performed.

4.6.1 Break / Transient Phenomena In determining the AREVA NP criteria, the characteristics of larger break area LOCAs are examined. These LOCA characteristics involve a rapid and chaotic depressurization of the reactor coolant system (RCS) during which the three historical approximate states of the system can be identified.

Blowdown The blowdown phase is defined as the time period from initiation of the break until flow from the SIT begins. This definition is somewhat different from the traditional definition of blowdown, which extends the blowdown until the RCS pressure approaches containment pressure. The blowdown phase typically lasts about 12- to 25-seconds, depending on the break size.

Refill is that period that starts with the end of blowdown, whichever definition is used, and ends when water is first forced upward into the core. During this phase the core experiences a near adiabatic heatup.

Reflood is that portion of the transient that starts with the end of refill, follows through the refilling of the core with water and ends with the achievement of complete core quench.

Implicit in this break-down is that the core liquid inventory has been completely, or nearly so, expelled from the primary system leaving the core in a state of near core-wide dispersed flow film boiling and subsequent adiabatic heatup prior to the reflood phase. Although this break down served as the basis for the original deterministic LOCA evaluation approaches and is valid for most LOCAs that would classically be termed large breaks, as the break area decreases the depressurization rate decreases such that these three phases overlap substantially. During these smaller break events, the core liquid inventory is not reduced as much as that found in AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 84 larger breaks. Also, the adiabatic core heatup is not as extensive as in the larger breaks which results in much lower cladding temperature excursions.

4.6.2 New Minimum Break Size Determination No determination of the lower limit can be exact. The values of critical phenomena that control the evolution of a LOCA transient will overlap and interplay. This is especially true in a statistical evaluation where parameter values are varied randomly with a strong expectation that the variations will affect results. In selecting the lower area of the RLBLOCA break spectrum, AREVA NP sought to preserve the generality of a complete or nearly complete core dry out accompanied by a substantially reduced lower plenum liquid inventory. It was reasoned that such conditions would be unlikely if the break flow rate was reduced to less than the reactor coolant pump flow. That is, if the reactor coolant pumps are capable of forcing more coolant toward the reactor vessel than the break can extract from the reactor vessel, the downcomer and core must maintain some degree of positive flow (positive in the normal operations sense).

The circumstance is, of course, transitory. Break flow is altered as the RCS blows down and the RC pump flow may decrease as the rotor and flywheel slow down if power is lost. However, if the core flow was reduced to zero or became negative immediately after the break initiation, then the event was quite likely to proceed with sufficient inertia to expel most of the reactor vessel liquid to the break. The criteria base, thus established, consists of comparing the break flow to the initial flow through all reactor coolant pumps and setting the minimum break area such that these flows match. This is done as follows:

Wbreak = Abreak

• Gbreak = Npump

• WRCP.

This gives Abreak = (Npump

• WRCP)/Gbreak.

The break mass flux is determined from critical flow. Because the RCS pressure in the broken cold leg will decrease rapidly during the first few seconds of the transient, the critical mass flux is averaged between that appropriate for the initial operating conditions and that appropriate for the initial cold leg enthalpy and the saturation pressure of coolant at that enthalpy.

Gbreak = (Gbreak(PO, HCLO) + Gbreak(PCLsat, HCLO))/ 2 .

Table 4-4 provides a listing of the plant type, initial condition, and the fractional minimum RLBLOCA break area, for all the plant types presented as generic representations in the next section.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 85 Table 4-4 Minimum Break Area for Large Break LOCA Spectrum The split versus double-ended guillotine break type is no longer related to break area. In concurrence with Regulatory Guide 1.157, both the split and the double-ended guillotine break will range in area between the minimum break area (Amin) and an area of twice the size of the broken pipe. The determination of break configuration, split versus double-ended guillotine, is made after the break area is selected based on a uniform probability for each occurrence.

4.6.3 Intermediate Break Size Disposition With the revision of the smaller break area for the RLBLOCA analysis, the break range for small breaks and large breaks are no longer contiguous. Typically the lower end of the large break spectrum occurs at between 0.2 to 0.3 times the total area of a 100% double-ended guillotine break (DEGB) and the upper end of the small break spectrum occurs at approximately 0.05 times the area of a 100% DEGB. This leaves a range of breaks that are not specifically analyzed during a LOCA licensing analysis. The premise for allowing this gap is that these breaks do not comprise accidents that develop high cladding temperature and thus do not comprise accidents that critically challenge the emergency core cooling systems (ECCS).

Breaks within this range remain large enough to blowdown to low pressures. Resolution is AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 86 provided by the large break ECC systems and the pressure-dependent injection limitations that determine critical small break performance are avoided.

A variety of plant types for which analysis within the intermediate range have been completed were surveyed. Although statistical determinations are extracted from the consideration of breaks with areas above the intermediate range, the AREVA NP best-estimate methodology remains suitable to characterize the ECCS performance of breaks within the intermediate range.

Table 4-4 provides a listing of the plant type, initial condition, and the fractional minimum RLBLOCA break area. Figures 4-15 through 4-20 provide the enlarged break spectrum results with the upper end of the small break spectrum and the lower end of the large break spectrum indicated by bars.

Table 4-5 provides differences between the true large break region and the intermediate break region (break areas between that of the largest SBLOCA and the smallest RLBLOCA). The minimum difference is 222 0 F; however, this case is not representative of the general trend shown by the other comparisons. Considering this point as an outlier, the table shows the minimum difference between the highest intermediate break spectrum PCT and large break spectrum PCT, for the eight plants, as at least 463 0 F, and including this point would provide an average difference of 640°F for the CE 2x4 design plants and a maximum difference of 840°F for the 4-loop W plant design.

Thus, by both measures, the peak cladding temperatures within the intermediate break range will be several hundred degrees below those in the true large break range. Therefore, these breaks will not provide a limit or a critical measure of the ECCS performance. Given that the large break spectrum bounds the intermediate spectrum, the use of only the large break spectrum meets the requirements of 10CFR50.46 for breaks within the intermediate break LOCA spectrum, and the method demonstrates that the ECCS for a plant meets the criteria of 10CFR50.46 with high probability.

AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 87 Table 4-5 Minimum PCT Temperature Difference - True Large and Intermediate Breaks Generic Maximum Maximum Plant Plant PCT (OF) PCT (OF) Delta PCT Average Delta Description Label Intermediate Large Size (OF) PCT (OF)

(Table 4-4) Size Break Break A 1206 193024 724 3-Loop W B 1273 1951 678 622 Design C 1326 1789 463 D 984 1751 767 E 1049 1740 691 2x4 CE 640 Design F 791 1670 879 G 1464 1686 22225 4-LoopW H 1127 1967 840 840 Design 24 The analysis for this W 3-Loop plant was performed with the Transition methodology and no break sizes fell into the intermediate break range. The PCT value of 1206'F is the closest point to the maximum end of the intermediate break spectrum.

25 The analysis for this 2x4 CE plant was performed with the Transition methodology and no break sizes fell into the intermediate break range. The PCT value of 1464'F is the closest point to the maximum end of the intermediate break spectrum. From the trends of the other 2x4 CE analyses, breaks falling within the intermediate break spectrum would be significantly lower.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 88 2000 r Upper End of Large Break SBLOCA

. Spectrum Break Size Minimum Spectrum 1800 - Break Area

.1 1600 4-- i* -

• 1

• 1 I.

I.

1400 a

• *1 1200 4 -4 0

1000 t 800 600 0.00000 01000 0.2000 0.3000 0,4000 0.5000 0.6000 0.7000 0.8000 0.9000 1.0000 Break Area Normalized to Double Ended Guillotine Figure 4-15 Plant A - Westinghouse 3-Loop Design AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 89 2000 -  !

I Upper End of Large Break

,SBLOCA Spectrum Break Size Minimumn Spectrum 1800 Break Area 4

.1 1600 4- -- - ----------

• A.---

• I*

• 4 4

4. 4 4;

1400 Z

o U

9-1200 1

.4 1000 800 600 0.0000 0.1000 0.2000 0.3000 0.4000 0.5000 0.6000 0.7000 0.8000 0.9000 1.0000 Break Area Normalized to Double Ended Guillotine Figure 4-16 Plant B -Westinghouse 3-Loop Design AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 90 2000 Upper End of Large Break SBLOCA Spectrum Break Size 7 - - - - F Mininum Spectrum 1800 Break Area 1600 -

V, 1400 +

U 1200 -

,-w "O

• 1000 +

800 00 600 0.0000 0. 1000 0.2000 0.3000 0.4000 0.5000 0.6000 0.7000 0.8000 0.9000 1.0000 Break Area Normalized to Double Ended Guillotih Figure 4-17 Plant C - Westinghouse 3-Loop Design AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 91 2000 Iv r r-Upper End of Large Break SBLOCA Spectrum Break Size h" Minimum 1800 - Spectrum Break Area 1600 +

1. 1

• 1  % P.

1400 +

V. *1 F.

4

'-I 1200 1000 +

800 OW)U 0.0000 0 1000 02000 03000 0.4000 0.5000 0.6000 0.7000 0.8000 0.9000 1.0000 Break Area Normalized to Double Ended Guillotine Figure 4-18 Plant D - Combustion Engineering 2x4 Design AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 92 2000 1 Upper End of Large Break SBLOCA Spectrum Break Size Minimum 1800 - Spectrum Break Area 0'

1600

• 1 4.

I.

1400 Z

I.- .

• 1 1200 1000 -- - - - - - -

800 +

-. ~ -4 600 I 0.0000 0.1000 0.2000 0.3000 0.4000 0.5000 0.6000 0.7000 0.8000 0.9000 1.0000 Break Area Normalized to Double Ended Guillotine Figure 4-19 Plant E - Combustion Engineering 2x4 Design AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 93 22000000 -

I Large Break Spectrum Upper End of Minimum 2000.0000 -

SBLOCA k,"Break Area

• 4-Break Size 4 Spectrum 44,'

1800.0000 + 4!

I.

I.

1600M0000 4'

4 441 o

1400.0000 -1 L 4'

• I

• 1 1200.0000 I.

1000.0000 o- - - - - - - - - - - - - -

4 800.0000 +

600.0000 1 0.0000 0.1000 0.2000 03000 0.4000 0.5000 0.6000 0.7000 0.8000 0.9000 1.0000 Break Area Normalized to Double Ended Guillotine Figure 4-20 Plant H - Westinghouse 4-Loop Design AREVA NP Inc.

A A`R'EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 94 4.7 Detailedinformation for Containment Model (ICECON)

Question: Verify that the ICECON model is that shown in Figure 5.1 of EMF-CC-39(P)

Revision 2, "ICECON:A Computer Program Used to Calculate Containment Back Pressurefor LOCA Analysis (Including Ice CondenserPlants)."

The AREVA NP RLBLOCA Report shows that the containment parameters treated statistically are: (1) upper compartment containment volume, (2) upper compartment containment temperature, and (3) lower compartment containment temperature. ANP-2903(P) states that "in many instances" the guidance of NRC Branch Technical Position CSB 6-2 was used in determining the other containment parameters.

a. How is the mixing of containmentsteam and ice melt modeled so as to minimize the containmentpressure?

b. Verify that all containment spray and fan coolers are assumed operating at maximum heat removal capacity.

c. Describe how the limits on the volume of the upper containment were determined.

d. How are the containment air return fans modeled and what is the effect of this modeling on the containmentpressure?

e. Describe how passive heat sink areas and heat capacities are modeled so as to minimize containmentpressure.

Response: See Section 3.3 for discussion of questions (a) through (e). Containment initial conditions and cooling system information are provided in Table 3-8 and Heat Sinks are provided in Table 3-9. For SONGS Unit 2 and Unit 3, the scatter plots of PCT versus the sampled containment volumes and initial atmospheric temperature are shown in Figure 4-21 and Figure 4-22. Containment pressure as a function of time for limiting case is shown in Figure 4-23.

AREVA NP Inc.

A A REVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 95 PCT vs Containment Volume 2000 1800 1600 0 U 0U LI LI Mm LI

• LI U 1400 U U LI U 0

EU ONO El 1200 []

C- LI m mI

[] mL M 1000 LI LI 0

800 600 0 Split Break ELGuillotine Break 400 2.300e+06 2.310e+06 2.320e+06 2.330e+06 2.340e+06 Containment Volume (ft3)

Figure 4-21 PCT vs. Containment Volume AREVA NP Inc.

A A RE:VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 96 2000 1800 S 4 1600 F M 0

mm-EP 0 Li 0 E -] 0 0 1400 0

MEIN El E, El 0 0 El mon 1200 MEE []

I-0 0EI' m El 1000 F D

M 800 F 600 F 0 Split Break EDGuillotne Break 400 40 50 60 70 80 90 100 110 120 130 140

'Temperature (OF)

Figure 4-22 PCT vs. Initial Containment Temperature AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 97 Containment and Loop Pressures 100 90 80 70 60 0~

a, 50 cn a,

0~

40 30 20 10 0

0 100 200 300 400 500 Time (s)

ID:55967 22Apr2011 22:38:43 R5DMX Figure 4-23 Containment Pressure for Limiting Case AREVA NP Inc.

.A.

AREV San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 98 4.8 Cross-References to North Anna Question: In order to conduct its review of the SONGS application of AREVA NP's realistic LBLOCA methods in an efficient manner, the NRC staff would like to make reference to the responses to NRC staff requests for additional information that were developed for the application of the AREVA NP methods to the North Anna Power Station, Units 1 and 2, and found acceptable during that review. The NRC Staff safety evaluation was issued on April 1, 2004 (Agency-wide Documentation and Management System (ADAMS) accession number ML040960040). The staff would like to make use of the information that was provided by the North Anna licensee that is not applicable only to North Anna or only to subatmospheric containments. This information is contained in letters to the NRC from the North Anna licensee dated September 26, 2003 (ADAMS accession number ML032790396) and November 10, 2003 (ADAMS accession number ML033240451). The specific responses that the staff would like to reference are:

September 26, 2003 letter: NRC Question I NRC Question 2 NRC Question 4 NRC Question 6 November 10, 2003 letter: NRC Question 1 Please verify that the information in these letters is applicable to the AREVA NP model applied to SONGS except for that information related specifically to North Anna and to sub-atmospheric containments.

Response: The responses provided to questions 1, 2, 4, and 6 are generic and related to the ability of ICECON to calculate containment pressures. They are applicable to the SONGS Unit2 and Unit 3 RLBLOCA submittal.

Question 1 - Completely Applicable Question 2 - Completely Applicable Question 4 - Completely Applicable (the reference to CSB 6-1 should now be to CSB Technical Position 6-2). The NRC altered the identification of this branch technical position in Revision 3 of NUREG-0800.

Question 6 - Completely applicable.

The supplemental request and response are applicable to SONGS.

AREVA NP Inc.

A AR:EVVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 99 4.9 GDC 35 - LOOP and No-LOOP Case Sets Question: 10CFR50, Appendix A, GDC [General Design Criterion] 35 [Emergency core cooling] states that, "Suitable redundancy in components and features and suitable interconnections, leak detection, isolation, and containment capabilities shall be provided to assure that for onsite electric power system operation (assuming offsite electric power is not available) and for offsite electric power operation (assuming onsite power is not available) the system function can be accomplished, assuming a single failure."

The Staff interpretation is that two cases (loss of offsite power with onsite power available, and loss of onsite power with offsite power available) must be run independently to satisfy GDC 35.

Each of these cases is separate from the other in that each case is represented by a different statistical response spectrum. To accomplish the task of identifying the worst case would require more runs. However, for LBLOCA analyses (only), the high likelihood of loss of onsite power being the most limiting is so small that only loss of offsite power cases need be run. (This is unless a particularplant design, e.g., CE [Combustion Engineering] plant design, is also vulnerable to a loss of onsite power, in which situation the NRC may require that both cases be analyzed separately. This would require more case runs to satisfy the statistical requirement than forjust loss of offsite power.)

What is your basis for assuming a 50% probability of loss of offsite power? Your statisticalruns need to assume that offsite power is lost (in an independent set of runs). If, as stated above, it has been determined that Palisades, being of CE design, is also vulnerable to a loss of onsite power, this also should be addressed (with an independent set of runs).

Response: In concurrence with the NRC's interpretation of GDC 35, a set of 59 cases each was run with a LOOP and No-LOOP assumption. The set of 59 cases that predicted the highest figure of merit, PCT, is reported in Section 2 and Section 3, herein. The results from both case sets are shown in Figure 3-23. This is a change to the approved RLBLOCA EM (Reference 1).

AREVA NP Inc.

A AýRE V*A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 100 4.10 Statement Question: Provide a statement confirming that SCE and its LBLOCA analyses vendor have ongoing processes that assure that the input variables and ranges of parameters for the SONGS LBLOCA analyses conservatively bound the values and ranges of those parameters for the as operated SONGS Unit 2 and Unit 3. This statement addresses certain programmatic requirementsof 10 CFR 50.46, Section (c).

Response: SCE and the LBLOCA Analysis Vendor have an ongoing process to ensure that all input variables and parameter ranges for the SONGS realistic large break loss-of-coolant accident are verified as conservative with respect to plant operating and design conditions. In accordance with SCE Quality Assurance program requirements, this process involves

1. Definition of the required input variables and parameter ranges by the Analysis Vendor.

2. Compilation of the specific values from existing plant design input and output documents by SCE and Vendor personnel in a formal analysis input summary document issued by the Analysis Vendor and

3. Formal review and approval of the input document by SCE. Formal SCE approval of the input document serves as the release for the Vendor to perform the analysis.

Continuing review of the input document is performed by SCE as part of the plant design change process and cycle-specific core design process. Changes to the input summary required to support plant modifications or cycle-specific core alternations are formally communicated to the Analysis Vendor by SCE. Revisions and updates to the analysis parameters are documented and approved in accordance with the process described above for the initial analysis.

AREVA NP Inc.

AE A

R VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 101 5.0 Conclusions A RLBLOCA analysis was performed for the SONGS Unit 2 and Unit 3 using NRC - approved AREVA NP RLBLOCA methods (Reference 1). Analysis results show that the limiting case has a PCT of 1605°F for a fresh fuel 6% Gad rod, and a maximum oxidation thickness and hydrogen generation that fall well within regulatory requirements.

The analysis supports operation at a nominal power level of 3458 MWt (including 20 MWt uncertainty), a steam generator tube plugging level of up to 8% in each steam generator, a total LHR of 12.8 kW/ft, which is equivalent to a total peaking factor (FQ) up to a value of 2.37, and a radial peaking factor (Fr) up to a value of 1.755 (including 6% uncertainty and 3.5% control rod insertion uncertainty) with no axial or burnup dependent power peaking limit and peak rod average exposures of up to 62 GWd/MTU. The twice-burnt fuel assemblies are not considered in the analyses since burnups at this level or higher do not retain sufficient energy potential to achieve significant cladding temperatures or cladding oxidations during the transient. For large break LOCA, the four 10GFR50.46 (b) criteria presented in Section 3.0 are met and operation of SONGS Unit 2 and Unit 3 with AREVA NP-supplied 16x16 M5clad fuel is justified.

AREVA NP Inc.

A A R'.E V.A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 102 6.0 Recent NRC Request for Additional Information (RAI) and AREVA NP Responses The NRC staff has found that strict adherence to currently referenced, or proposed for referencing AREVA NP methodologies are inconsistent with the NRC's requirements and review guidance without appropriate justification. This section addresses the NRC staffs concerns for the AREVA NP RLBLOCA methodology.

6.1 Thermal Conductivity Degradation- Once-Burnt Fuel Questions:

1. EMF-2103 considers only fresh fuel. Once-burnt fuel is more highly oxidized and has a lower thermal conductivity. The cladding of higher burnup fuel may heat differently than the analyzed fuel, and ffonce-burnt fuel may have a higher linearheat rate than fresh fuel)). This issue results in a potential non-conservatism for predicted peak cladding temperature and local oxidation.

a. 10 CFR 50.46 requires the ECCS cooling performance calculation to include a number of postulated loss of coolant accidents of different sizes, locations, and other properties sufficient to provide assurance that the most severe postulated loss of coolant accidents are calculated.

b. The methodology requires supplemental information, sensitivity studies, or revision to include analysis of the effects of once-burnt fuel, to demonstrate compliance with 10 CFR 50.46.

c. Please provide more information about the management of the fuel thermal conductivity degradation issue identified in NRC Information Notice 2009-23, "NuclearFuel Thermal Conductivity Degradation." Specifically:

Page 1-3, states, "Foreach specific time in cycle, the fuel conditions are computed using RODEX3A prior to starting the S-RELAP5 portion of the analysis. A steady-state condition for the given time in cycle using S-RELAP5 is established. A base fuel centerline temperature is established in this process. Then two-transformation adjustment to the base fuel centerline temperature is computed. The first transformationis a linear adjustment for an exposure of 10 GWd/MTU or higher. In the new process, a polynomial transformation is used in the first transformation instead of a linear transformation." Please clarify the following:

Explain how the fuel pellet radialtemperature profile is computed.

AREVA NP Inc.

A.

San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 103 ii. Explain which code is used to calculate this profile, both for initial conditions and through the postulatedaccident.

iii. Explain whether the polynomial transformation is applied merely to the centerline temperature, or to the entire pellet temperature.

d. Provide additional information to describe the polynomial transformation.

Summarize data used to develop the polynomial transformation and discuss considerationof applicable uncertainties.

e. For the PCT-limiting RLBLOCA case, please provide:

iv. Correctedand uncorrectedradial temperature profile of the hot rod at the time and location of peak cladding temperature.

v. Temperature vs. time for the limiting PCT case at the limiting location, including the fuel centerline, fuel average, and clad surface temperatures. Indicate the end of blowdown, start of refill, and startof reflood on this graph.

vi. Burnup for the limiting rod.

Response

The NRC concern covers a wide range of specific items but can be paraphrased as: "How does the AREVA NP RLBLOCA analysis for SONGS provide a licensing basis for fuel throughout its operational life with particular attention to the phenomena of thermal conductivity degradation with burnup?" In response, the following explanation of the methodology employed for SONGS is provided and followed by specific responses to each of the particular questions.

The AREVA transition package has been updated to specifically model once-burnt fuel rods.

This provides specific analytical results as to the compliance of once-burnt fuel rods with the criteria of 10CFR50.46. The highest burnup for which case calculations are performed is the maximum burnup anticipated for once-burnt fuel in the cycle design. For the fuel cycle design analyzed herein, the maximum anticipated average assembly burnup is 31.2 GWd/MTU for the fresh fuel and 53.3 GWd/MTU for the once-burnt fuel. Once-burnt fuel rods from assemblies with these burnup levels and higher do not retain sufficient energy potential to achieve significant cladding temperatures or cladding oxidations during the transient. Therefore, fuel at these burnups cannot challenge the criteria of 10CFR50.46 and are not specifically analyzed.

The approach provides bounding LOCA licensing for fuel up to the current AREVA PWR fuel licensed burnup limit of 62 GWd/MTU.

For individual cases within the LOCA evaluations, the first and second cycle rods are assigned burnups according to the sampled time in cycle. The time in cycle is sampled once and is the same for both the fresh (first cycle) and once-burnt (second cycle) fuel. Burnup for the fresh and once-burnt rods is different in accordance with the cycle management. Likewise, pin pressure and thermal conductivity differ.

AREVA NP Inc.

A A-R'EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 104 In addition to the thermal conductivity and fuel temperature adjustments for burnup, a burnup dependent reduction in allowed peaking is needed for the once-burnt fuel. For first cycle fuel, the RLBLOCA methodology increases the Fr to the Technical Specifications maximum (including uncertainty) for the first cycle hot rods in the model. Since SONGS does not have a Technical Specifications limit on Fr, a conservative value (without uncertainty) of 1.6 has been used in the RLBLOCA analysis. When used in the RLBLOCA uncertainty analysis, the Fr value includes measurement uncertainty and the Power Dependent Insertion Limit/All Rods Out (PDIL/ARO) uncertainty. Shortly into the cycle, once-burnt fuel, has insufficient energy potential to achieve this peaking. A burnup dependent reduction in allowed peaking is therefore applied through an adjustment in the second cycle Fr. For SONGS, the Fr for the once-burnt peak pin (6% Gad rod) is conservatively set to [ ] of the fresh peak pin at the beginning of the irradiation cycle. Then, it gets reduced to [ ] of the fresh peak pin by the end of the irradiation cycle. The modeling of the burnup dependent reduction in peaking is applied through an adjustment to the Fr based on the power ratio shown in Figure 6-1.

1.a, & 1.b. Paraphrased concern: Requirement to treat a wide range of conditions and sensitivity studies necessary to cover once-burnt fuel.

The inclusion of once-burnt fuel rods in each calculation of the case set provides the required range of parameters and sensitivity studies to satisfy the 10 CFR 50.46 requirements.

1.c Paraphrased concern: Provide corrected and uncorrected radial temperature results, temperatures in the pellet versus time, and the burnup for the limiting case.

Figure 6-2 shows the corrected radial temperature profile and the uncorrected centerline temperature for the limiting case hot rod at the initiation of the transient. Because the uncorrected radial profile is never used or recorded in the methodology, it cannot be provided.

However, the uncorrected centerline temperature can be calculated from the equation for adjusted temperature described below in the 'Thermal Conductivity Degradation Related Questions' section and is shown in Figure 6-2. As the pellet power is not adjusted, the radial temperature profile must follow the corrected profile closely and the two must converge at the surface of the pellet.

Figure 6-3 shows the centerline, surface, and average fuel temperatures of the limiting fresh hot rod (6% Gad) at the PCT elevation for the limiting PCT case. In this case, all of the fresh rods have higher PCTs than the once-burnt rods. The most limiting once-burnt rod is the U0 2 rod.

With a cycle burnup of approximately 9664 EFPH, the fresh 6% Gad rod has an assembly burnup of 18.2 GWd/MTU while the once-burnt U0 2 has an assembly burnup of 33.2 GWd/MTU. A plot comparing the PCT of the fresh and once-burnt U0 2 rods for this case is shown in Figure 6-4.

AREVA NP Inc.

.A.

ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 105 Thermal Conductivity Degradation Related Questions:

Paraphrasedconcern: Provide information on the treatment of thermal conductivity degradation.

Thermal conductivity degradation impacts the ability to transfer energy from within the pellet to the pellet surface and consequently through the cladding to the coolant. Both the initial pellet temperature and the transient release of energy from the pellet are affected. The impact of thermal conductivity changes with burnup are treated by applying a bias. This bias and a measure of the uncertainty in the data were determined by benchmarking the fuel performance code, RODEX3A, to a set of data that extends past the licensed burnup. The bias adjusts the initial fuel temperature to the mean of the benchmark results. The sampled uncertainty is used to provide for the variance of the benchmarks.

The database for the benchmarks is that used to qualify and approve the RODEX4 code (Reference 13). The data from three experimental rods (cases 432R2, 432R6, and 597R8) were not used in the benchmarks. Test 597R8 was not appropriate for this application. Cases 432R2 and 432R6 are rod studies that are not configured appropriately these types of comparisons. Essentially, these fuel rods are not representative of commercial PWR fuel. Part of the benchmark activity was to incorporate a fractional representation of difference between the RODEX3A calculated results and the data. The fractional adjustment provides a better adjustment over a range of initial temperatures. Therefore, for each benchmark case the Tfracaon was determined.

Tfraction iTrodex3A - Tdata Tl*"O Trodex 3A where:

Tfraction = Delta fractional temperature of computed to data (K),

Trodex3A = Temperature computed by RODEX3A (K) and Tdata = Temperature from the RODEX4 database (K)

Figure 6-5 shows the RODEX3A benchmark results along with a polynomial fitted to the results using the least squares method. The negative of this polynomial is the bias which is added to RODEX3A predictions to achieve agreement with the data.

Figure 6-6 shows the results of applying this bias in comparison to the results of applying the original RLBLOCA methodology Revision 0 bias. It is evident from Figure 6-6 that the bias makes the adjustment for burnup effects in accordance with the data.

The application of the bias within the methodology proceeds as follows: The burnup for the case hot rods, fresh and once burnt, is determined by sampling the time in cycle and a RODEX3A calculation of the initial fuel centerline temperature performed. From the fit in Figure 6-5 an adjusted temperature is determined as per the equation below:

AREVA NP Inc.

A A.R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 106 where:

Tnew = Adjusted fuel centerline temperature (K),

B = Burnup (GWd/MTU or MWd/KgU) and Toriginal = Unadjusted RODEX3A fuel centerline temperature (K).

T111 Figure 6-7 provides the bias adjustment new Toriginal

, as a function of burnup, using the above polynomial curve fit.

The uncertainty is determined from a Gaussian distribution characterized by a [ ]

standard deviation and added to Tew. The fuel temperature calculation is then repeated with a multiplier, fuel K, on the code calculated fuel thermal conductivity. The fuel centerline temperature is compared to 'Tnew + uncertainty' and the calculation is repeated with an adjusted fuel K as necessary. The process is continued until the calculated centerline fuel temperature matches 'Tnew + uncertainty'. Since the process applies an adjustment to the fuel thermal conductivity, the temperature throughout the pellet is adjusted appropriately. The final multiplier is applied to the thermal conductivity throughout the transient.

Because the data fitting covers the complete range of applicable burnup it is applied as such and the zero bias offset used in Revision 0 for the first 10 GWd/MTU burnup is eliminated.

Paraphrasedconcern: How is radialtemperature profile computed?

The RODEX3 topical report, ANF-90-145(P)(A), Appendix B (Reference 14) provides details of the calculation of the radial temperature distribution.

Paraphrasedconcern: Which codes are used?

A portion of the RODEX3A fuel model was incorporated into the S-RELAP5 code to calculate fuel response for transient analyses. This coding, referred to as the S-RELAP5IRODEX3A model, deals only with transient predictions and does not calculate the burnup response of the fuel. Instead, fuel conditions at the burnup of interest are transferred via a binary data file from RODEX3A to S-RELAP5/RODEX3A, establishing the initial state of the fuel prior to the transient. The data transferred from RODEX3A describes the fuel at zero power. A steady-state S-RELAP5/RODEX3A calculation is required to establish the fuel state at power.

AREVA NP Inc.

A ARE V.A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 107 The transient fuel pellet radial temperature profile is computed by solving the conduction equation in S-RELAP5. Material properties are calculated in S-RELAP5/RODEX3A.

Paraphrasedconcern: Is the adjustment made to the entire pellet?

The adjustment is applied to the entire fuel pellet. The polynomial transformation provides a bias adjustment to the fuel centerline temperature. A sampled parameter provides a random assessment and adjustment of the centerline temperature uncertainty. These are combined and the total adjustment is achieved by iterating a multiplicative adjustment to the fuel thermal conductivity until the desired fuel centerline temperature is reached.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 108 Figure 6-1 Once-Burnt Fuel Power Ratios (2nd cycle)

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 109 Figure 6-2 Radial Temperature Profile for Hot Rod AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 110 Figure 6-3 Temperature versus Time for Fuel Centerline, Clad Surface, and Fuel Average AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 111 Figure 6-4 Fresh versus Once-Burnt U0 2 Rod PCT Trace AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 112 Figure 6-5 Fractional Fuel Centerline Temperature Delta Between RODEX3A and Data AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 113 Figure 6-6 Fuel Centerline Temperature Delta of RODEX3A Calculations to Data (Original and Using the New Correlation)

AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 114 Figure 6-7 Correction Factor (as applied for temperatures in Kelvin)

AREVA NP Inc.

A A R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 115 6.2 Decay Heat Treatment Question:

1. Provide additionalinformation to justify the use of the selected analytic treatment for decay heat uncertainty in the RLBLOCA model.

a. The NRC needs to understand the sensitivity that PCT has with respect to the decay heat uncertainty, please re-execute the limiting case with a 1.03 decay heat multiplier and report the results.

Response

1 The RLBLOCA EM decay heat calculations are based on the 1979 ANSI/ANS standard (Reference 15). The standard is applicable to light water reactors containing low enriched uranium as the initial fissile material; all plants, to which the RLBLOCA EM is applicable, are such plants. The selected approach to simulate fission product decay assures a representative yet conservative treatment. The EM fission product decay heat simulation and the basis for the conservatism of the approach are outlined in the remainder of this response.

Non-SamDling Approach to Decay Heat The RLBLOCA methodology proposed herein utilizes the U235 decay curve from the 1979 ANSI/ANS standard for fully saturated decay chains as the decay for all fission products.

The fully saturated chains result from an assumption of infinite operation. The total energy per fission is assumed to be 200 MeV (Reference 15). No bias or uncertainty certainty is assigned to the fission product decay heat. Differing from the base EMF-2103 evaluation model approach, the uncertainty for the decay heat parameter is set to zero and no sampling is done on this parameter, resulting in the decay heat being used with a 1.0 multiplier. The decay heat in the analysis is always the 1979 ANS standard for decay heat from U235 with fully saturated decay chains, corresponding to infinite operation, assuming 200 MeV per fission.

Conservatism in the Approach In the approach used, the total energy per fission is assumed to be 200 MeV whereas a more accurate value for U235 would be greater than 202 MeV per fission. This imparts a direct 1% conservatism.

During irradiation, plutonium accumulates such that the ratio of plutonium-to-uranium fission-energy production rate is substantial and increasing. Because the decay energy resulting from plutonium fissions is less than that from U235, the decay energy is reduced from U235 fully saturated decay chains as the fuel is burnt. Thus, as burnup increases, the RLBLOCA decay heat modeling with U235 only, accrues conservatism. This AREVA NP Inc.

A,.

San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 116 conservatism applies to all regions of the core according to the mix of burnups represented within each region.

The fresh fuel, hot pin and hot assembly, begin operation with no plutonium. Therefore, the reduction in decay heat due to plutonium build-up is not applicable to the low burnup fuel in the initial period of the cycle. However, for fresh fuel, the concentrations of long decay term fission products will not have built up. The lack of long decay term sources comprises a reduction in decay heat rate of several percent over the first year of operation, making the infinite operation assumption conservative while the plutonium concentration is accumulating.

Calculations of these considerations based on the 1979 ANS standard have been performed to demonstrate the conservatism of the selected approach. Figure 6-8 and Figure 6-9 show the decay heat versus time for:

1) Infinite Operation of U235 (the AREVA NP decay heat model)

2) Finite Operation to 0.1 GWd/MTU of all fissionable isotopes with uncertainties added

3) Finite Operation to 1 GWd/MTU of all fissionable isotopes with uncertainties added

4) Finite Operation to 1 GWd/MTU of all fissionable isotopes without uncertainties

5) Finite Operation to 20 GWd/MTU of all fissionable isotopes with uncertainties added

6) Finite Operation to 40 GWd/MTU of all fissionable isotopes with uncertainties added

7) Finite Operation to 60 GWd/MTU of all fissionable isotopes with uncertainties added In order to treat the Plutonium buildup effect conservatively, the finite operations curves are based on cycle management and enrichment assumptions that minimize the build up of Plutonium. No uncertainty is included in the infinite operation curve. The uncertainties incorporated in the other curves are 2 sigma values for the individual isotopes as published in the 1979 ANS standard. This provides greater than a 95/95 confidence in each of the decay heat contributions. The contributions are added linearly according to the individual isotopes fractional occurrence of fission.

Because of the range of the decay heat parameter, the early comparison of the relationships is difficult to ascertain. Clearly the U235 infinite operation curve is conservative for all times after a few seconds (-2 seconds). To better demonstrate the relationships, Figure 6-10 and Figure 6-11 provide the ratios of the finite operation curves to the infinite operation curves. The curvature of the plotted ratios during the first 2 to 3 AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 117 seconds is due to the increased uncertainties during this time phase. The 1979 ANS standard is based on measured data and the difficulty of measuring decay heat within a few seconds of shutdown is reflected in these uncertainties. The highest combined finite operation decay heat curve with uncertainties exceeds the AREVA NP decay heat curve by only 2.5% at shutdown and falls below the AREVA NP curve in less than 2 seconds.

Thus, there is only a 5% probability that the infinite operation curve of decay heat will be exceeded by up to 2.5% and that possibility exists for the first 2 seconds of the transient.

The potential accumulated underprediction is of short duration and of no consequence to the LOCA evaluation. The decay heat curve selected is suitable while somewhat conservative for the realistic evaluation of LOCA.

In conclusion, the choice of infinite operation with pure U235 fission product decay heat provides a base model that is conservative relative to the decay heat for finite operation.

For RLBLOCA evaluation, the sampling of a decay heat multiplier has been removed such that the decay heat for all cases is 1.0 times the infinite operation U235 decay chain providing conservative treatment of the 1979 ANS standard with the assumption of 200 Mev/fission.

a. Not applicable to the SONGS analysis. The decay heat was not sampled.

AREVA NP Inc.

A..

AIR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 118 Figure 6-8 Decay Heat Comparisons, Infinite Operation U235, Finite Operation All Isotopes (0.1 - 10 sec)

AREVA NP Inc.

AE A RkE:t',VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 119 Figure 6-9 Decay Heat Comparisons, Infinite Operation U235, Finite Operation All Isotopes (10 - 1000 sec)

AREVA NP Inc.

A VA E.R San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 120 Figure 6-10 Decay Heat Ratios, Finite Operation over Infinite Operation for U235, All Isotopes (0 - 10 sec)

AREVA NP Inc.

A AR~VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 121 Figure 6-11 Decay Heat Ratios, Finite Operation over Infinite Operation for U235, All Isotopes (0 - 600 sec)

AREVA NP Inc.

A A:RE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 122 6.3 Thermal Conductivity Degradation- Swelling, Rupture, and Relocation Question:

1. The following questions are based on a July 14, 2009, letter from Gardner,AREVA NP, to the USNRC, re: Informational Transmittal Regarding Requested White Papers on the Treatment of Exposure Dependent Fuel Thermal Conductivity Degradationin Legacy Fuel Performance Codes and Methods.

a. AREVA NP postulates that clad swelling and rupture produces a benefit to PCT, and because of this, the realisticlarge break loss of coolant accident (RLBLOCA) model does not include a clad swelling and rupture model. Does this conjecture include considerationof test data, which has shown that following fuel rupture, the ballooned region fills with fuel fragments? What analytic studies support this conclusion? How are they applicable to SONGS Unit 2 and Unit 3? Please also address the potential for co-planarblockage with the fuel relocation evaluation.

b. Since blowdown ruptures can occur at end of life conditions, show that blowdown ruptures do not occur at the end of life for the postulated SONGS Unit 2 and Unit 3 large break LOCA.

AREVA NP Inc.

A.

ARE'VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 123

Response

Justification for this position is provided by consideration of the phenomena involved, analysis of the effects, and experimental results. These phenomena are applicable at all times in life. The impact of rupture and ballooning on clad cooling occurs through several rupture- or ballooning-induced cooling mechanisms and three detrimental heating effects:

Cooling effects:

Increased clad heat transfer surface area at ballooned or ruptured regions Increased velocities within the ballooned and ruptured regions Increased turbulence within the ballooned and ruptured regions Droplet shattering resulting in increased interphase heat transfer and steam de-superheating Decrease in gap heat transfer if the fuel does not strongly relocate Decrease in pellet thermal conductivity ifthe fuel relocates Formation of local quench fronts in ballooned and ruptured regions Heating effects:

Flow diversion around ballooned and ruptured regions Clad heating load increase at the ruptured elevation due to fuel relocation Cladding heat load increased due to interior oxidation AREVA NP Inc.

.A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 124 Experience with Appendix K methodologies has shown that the aggregate of these effects acts to decrease the cladding temperatures when no fuel relocation occurs. This was demonstrated in Appendix B, Section B.2 of RLBLOCA EM Topical (Reference 1) and the response to RAI 28 on the topical (page 79 of Amendment i to Reference 16) with sensitivity studies on both 3- and 4-loop PWRs with 15x1 5 and 17x1 7 fuel designs similar to the 16x1 6 fuel design used at SONGS Units 2 and 3. The studies included increased heat transfer surface area, increased local coolant velocities, a decrease in gap heat transfer, flow diversion, and interior cladding oxidation. The effects of increased turbulence, droplet shattering, and potential local quenching were not included within the modeling. Decrease in pellet thermal conductivity and a clad heating load increase also were not included since the studies were not meant to address fuel relocation. Even without half of the cooling mechanisms modeled, the cladding temperatures and local oxidations were reduced. This effect has also been observed experimentally in the FEBA (Reference 17) and FLECHT (Reference 18) test series.

Under a condition of fuel relocation, wherein the fuel above the ballooned region drops into the ballooned region, it has been postulated that increased decay heat generation will lead to an increase in cladding heat flux resulting in higher cladding temperatures. Various presentations (e.g., Reference 19, Articles 1 and 12) purport to show the effect. However, these studies have uniformly incorporated extreme assumptions on the conditions of relocation and the resultant heat transfer processes. Few include provisions for rupture-induced cooling mechanisms. Most assume that the cladding expands circularly without being encumbered by the surrounding pins in the fuel assembly. In fact, a free expansion of the fuel rod is only possible up to pin strains in the mid-30 percents. For higher strains the local gap volume no longer increases faster than the clad surface area. Finally, the packing factor of the rubble filling the ballooned region is over-predicted. If reasonable, yet conservative, assumptions are made, study results would lead to the expectation that fuel relocation, which is real, does not pose a condition by which the ruptured or ballooned regions will exceed the consequence of the non-ballooned regions of the hot pin.

The above conclusion was observed experimentally in the KfK experiments as reported in RAI 131 on the RLBLOCA EM topical report (page 120 of Amendment 1 to Reference 16). In the KfK in Pile Tests, fuel relocation into the ballooned area of the fuel rod occurred but did not adversely affect the subsequent clad temperature behavior. To determine when the fuel relocates two tests were performed with thermocouples located at the top of the pellet stack.

One test comprised low burnup fuel, which maintained its pellet geometry after rupture. The other test was of higher burnup fuel which relocated. Relocation, for the test that relocated, was demonstrated by temperatures from the upper thermocouples showing a significant drop, loss of energy source, at the time of fuel rod rupture. For this test, the heatup rate, at the rupture elevation, following the rupture was reduced relative to the heatup rate prior to rupture. This reduction in heatup rate indicates that the PCT at the time of turnover would be less than what would have be reached if rupture had not occurred, even with the increase in localized decay AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 125 heat from the pellet rubble residing at the ruptured region. Thus, the KfK experiments demonstrate that analyses which ignore the beneficial effects of swelling and rupture provide conservatively high clad temperature estimates for the ruptured region during reflood even when fuel relocation occurs.

In conclusion, the AREVA NP RLBLOCA EM does not incorporate a clad ballooning, rupture and fuel relocation model. To support this modeling, the cladding temperature and pin stress evolution 26 has been assessed against rupture criteria appropriate for the cladding being evaluated. No rupture occurred during blowdown or refill for fresh or once- and twice-burnt fuel.

For rupture during reflood, the cladding temperature for the most severe location on the un-ruptured rod has been demonstrated to conservatively bound the result for any possible rupture location.

26 The assessment was performed for the limiting PCT case. The limiting case is approximately 60% through the cycle. The PCT reduction between the limiting PCT case and the end-of-cycle cases is such that blowdown rupture would not be challenged.

AREVA NP Inc.

A.

ARýEVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 126 6.4 Oxidation - Pre-transientand Single-Sided Question:

EMF-2103 models unoxidized fuel rods, on the basis that more transient oxidation will occur on an unoxidized fuel rod, which will serve to generate additionalcladding heat loads and drive the oxidation rate higher. The NRC staff does not agree that this approach is conservative.

1. This issue may result in inaccurate cladding oxidation estimates and potential cladding embrittlement issues unaccounted for in the evaluation model. Provide information to illustrate the conservative nature of the single-side only oxidation model and its application to the SONGS RLBLOCA analysis.

2. Information Notice 98-29, "PredictedIncrease in Fuel Rod Cladding Oxidation," discusses the effect high-burnup phenomena have on fuel pellet to cladding heat transfer. The staff's position, that pre-transient oxidation be considered in ECCS evaluation models, is documented in letters to NEI dated 3/31 and 11/8/99

3. The NRC staff may consider a plant-specific disposition for this issue based on the unit-specific results of the ECCS evaluation for the large break LOCA.

Response

1. AREVA NP's NRC-approved RLBLOCA EM uses the maximum un-ruptured cladding oxidation as representative or bounding of the transient oxidation that would have been computed at a rupture location. The position is supported by three aspects of the performed oxidation calculation.

• The cladding is initialized with no initial corrosion layer. Because the oxidation rate is inversely proportional to the oxidation layer present, the use of clean cladding at the start of the accident leads to substantially higher reaction rates. For corrosions in the range of the first cycle, the difference in rate is a minimum of a 50 % increase and increases during the cycle. The increase applies to both exterior and post-rupture interior oxidation.

" The cladding temperature even in the presence of fuel relocation is reduced for the ruptured region of the cladding. In the KfK experiments (page 210 of NRC:02:062 Attachment 1 to Reference 16 and included in Reference 17) the temperature drop at rupture was between 50 and 75 K.

• For ruptured cladding either the cladding interior oxidation rate is reduced by attached pellet fragments, moderate to highly burnt fuel, or the cladding temperature decrease at rupture is much more than the 50 to 75 K explained above. In either case, an additional mechanism exists to reduce the local oxidation at the rupture location.

AREVA NP Inc.

A ARE:V A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 127 In conclusion, insights into the EM oxidation process and those that will evolve after rupture clearly identify differences that will reduce the transient oxidation at the rupture location to less than that which the EM calculates at un-ruptured locations. Thus, the RLBLOCA Revision 0 EM approach of determining local transient oxidation is clearly appropriate to demonstrate compliance with the local oxidation criterion of 10CFR50.46, when combined with the pre-transient oxidation.

2. The initial corrosion layer was calculated to be 0.9085% for the fresh 6% Gad rod (at 18.2 GWd/MTU) and 1.9685% for the once-burnt U0 2 rod (at 33.2 GWd/MTU). The initial corrosion layer was added to the transient calculated value and the total is in Table 3-5.

6.5 Single FailureAssumption Questions:

1. The current licensing basis, deterministic loss of coolant accident (LOCA) analysis concluded that the limiting condition did not involve a worst-case single failure, but rather that it depended on injected coolant delivered in such a condition that the resultant containment environment, specifically the lower containment pressure, contributed to the limiting peak cladding temperature (PCT). Please provide information describing how this potentially limiting scenario was evaluated using the proposed best-estimate methodology.

2. Please provide additionalinformation summarizing the single-failure evaluation performed to establish compliance with General Design Criterion (GDC) 35 requirements. Identify which single failures were considered, discuss whether each failure was evaluated or explicitly analyzed, and for those failures which were explicitly analyzed, explain whether they were analyzed in a reference case or explicitly as a part of the statistical methodology. Also discuss the basis for the single failure evaluation. For example, were single failures considered as a matter of experience with SONGS Unit 2 and Unit 3 specifically, or with a generic CE nuclearsteam supply system design?

a. The staff also needs to understand how the limiting single failure for the CE 2x4 NSSS was determined, since the basis for the RAI response defers to NRC-approved methodology. Poring through EMF-2103, the staff only located sensitivity results on 3-loop W systems. In some cases, the limiting failure would be a single LPSI and in others it was a diesel. The staff could not locate a clear, generic disposition for the single failure at any place in EMF-2103.

b. What was done under the auspices of EMF-2103 development to ensure that the containment analysis produced a sufficiently conservative prediction that a no failure, maximum SI spillage case, for a CE 2x4 NSSS, is bounded by the chosen single failure? The staff will need to see that work.

AREVA NP Inc.

A ARE.VA ,

San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 128

Response

1. AREVA NP EMF-2103(P)(A) Revision 0 (Reference 1) conservatively prescribes:

• The use of full containment sprays without a time delay at the minimum technical specification temperature;

• Pumped ECCS injection at the maximum technical specification temperature; and

• Sampling of the containment volume (indirectly sampling containment pressure) from its nominal volume to its empty volume.

Studies, comparing several failure assumptions, including a no-failure assumption (see Reference 1, EMF-2103(P)(A) Revision 0, RAI response Numbers 26 and 111) validate that the ECCS and containment modeling of the AREVA NP methodology trends to the conservative. The containment pressure response is indirectly ranged by sampling the containment volume. The possible range to be sampled from was 2.305E+6 ft3 to 2.335E+6 ft3 for SONGS Units 2 and 3 containment volume. Figure 4-21 shows that there is little sensitivity between containment volume (indirectly pressure) and PCT for a statistical application. Thus, the methodology is responsive to the goal of a realistic evaluation, yet slightly conservative.

2. Section 4.9 discusses GDC 35. The single failure prescribed by EMF-2103(P)(A) (AREVA NP's RLBLOCA EM) is a loss of one train of ECCS.

AREVA NP satisfies the GDC-35 criteria by running one set of 59 cases with offsite power available and one set of 59 cases with no offsite power available. The sampling seeds are held constant between these two case sets, with the only difference being the offsite power assumption. The case set that produces the most limiting PCT is reported, for SONGS, was offsite power not available. Figure 3-23 displays the results from the two case sets.

a. The definition for loss of a diesel scenario by itself would mean that in addition to loss of one LPSI and one HPSI pump, one train of containment spray would not be available. The current method models all containment pressure-reducing systems as fully functional. Containment fans and containment sprays start at time zero (Table 3-8).

The response to RAI #111 for EMF-2103 (Reference 1, Attachment 1 page 185 -

189) was based on sensitivities to 3-loop W plants. The Base Case, which produced the most limiting results, is described in the RAI #111 response as the loss of one diesel with full containment spray. Figure 6-12 (recreated from RAI #111, Figure 111.2) shows that for the sample plant analysis, W 3-loop, the base case, AREVA NP ECCS failure assumptions, is 35 0 F higher in PCT than a fully consistent loss of diesel and over 170°F greater than the loss of one LPSI case.

AREVA NP Inc.

ýA AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 129 Figure 6-12 Clad Temperature Response from Single Failure Study AREVA NP Inc.

A A!RiE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 130

b. A sensitivity study of the SONGS limiting case (Case 7) documented in Section 3 of this document as the Analysis of Record (AOR) was conducted with "maximum" ECCS flow conditions to demonstrate that the minimum ECCS single failure assumption is conservatively bounding.

Sensitivity studies were run for the limiting case (Case 7) in both offsite power configurations with a maximum ECCS delivery. The loss of offsite power (LOOP) case remained limiting for the maximum ECCS cases with a PCT value of 1605OF (6% Gad Rod) compared to the offsite power available (NOLOOP) case PCT of 1451°F (6% Gad Rod). The LOOP PCT for the AOR is 1605°F (6% Gad Rod) and the NOLOOP PCT is 14421F (6% Gad Rod).

Even though the PCTs for the LOOP cases are identical, the impact on quenching the transient is clearly seen in Figure 6-13. The time at temperature for the minimum ECCS case will increase the local maximum oxidation calculated. The PCT for the limiting AOR case was achieved at 8.1 seconds, which is well before the ECCS injection occurs. Results for both maximum and minimum ECCS remain identical until 13 seconds, which is when EGOS injection starts (Figure 6-15). This demonstrates that the AREVA NP single failure assumption produces conservative results.

Figure 6-13 through Figure 6-16 show the respective PCT trace, containment and system pressure, ECCS injection rates, and downcomer level for both the AOR and the maximum ECCS sensitivity.

Figure 6-14 demonstrates that the maximum EGGS flow does not have a significant impact on the containment pressure up to about 100 seconds (approximately the time that the SIT empties); the maximum ECCS containment pressure overlaps the AOR containment pressure.

Figure 6-16 gives the downcomer level for both the AOR and the maximum EGOS case. The reference Level Zero for the liquid level in the reactor vessel is the bottom of the downcomer, which corresponds to the bottom of the lower support plate. It can be seen that the downcomer level in the maximum ECCS case is higher than the AOR, consequently providing more driving head for the reflood of the core. The higher driving head in the maximum EGGS case is enough to compensate for small differences in containment pressure (Figure 6-14) resulting in a faster post peak cooldown.

The AREVA NP RLBLOCA application, regardless of the loss of diesel assumption, models all containment pressure-reducing systems and conservatively assumes them to be fully functional. The AOR conservatively assumes an on-time start and normal lineups of the containment spray and Emergency Cooling Units to conservatively reduce containment pressure and increase break flow. The results of AREVA NP Inc.

A.

AR.E- -A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 131 the study demonstrate that the AOR ECCS configuration is PCT-limiting and oxidation-limiting.

Figure 6-13 Comparison of PCT Independent of Elevation for Maximum ECCS and Minimum ECCS (AOR)

AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 132 Figure 6-14 Comparison of Containment and System Pressure for Maximum ECCS and Minimum ECCS AREVA NP Inc.

A ARE: V A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 133 Figure 6-15 Comparison of ECCS Flows for Maximum ECCS and Minimum ECCS AREVA NP Inc.

A:

ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 134 Figure 6-16 Comparison of Downcomer Level for Maximum ECCS and Minimum ECCS (AOR)

AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 135 6.6 Core Liquid Level Question:

1. Page 3-6 states, "the RLBLOCA transients are of sufficiently short duration that the switchover to sump cooling water (i.e., RAS) for ECCS pumped injection need not be considered." For the limiting transient, the collapsed core liquid level from 200-350 seconds appearsto trend downward (Figure 3-21). An indication of stable and increasing collapsed liquid level would substantiate the statement quoted above, but this is not the case for Figure 3-21. Is the SRELAP-5 model of the limiting case capable of generating credible results after 350s? If so, please provide results for a period of the transient sufficient to demonstrate that the core collapsed liquid levels are stable or increasing.

Response

The "Collapsed Liquid Level in the Core for the Limiting Case" in Figure 3-20 shows an increase in core collapsed liquid level from 100 seconds to about 150 seconds. After 150 seconds, it shows the core collapsed level decreasing slightly through transient termination. At this time the inner-vessel two-phase mixture extends into the upper plenum. The factors that govern the core liquid content are the steam generation rate, the steam rise velocity and the steam specific volume. After 150 seconds the steam generation rate and the steam velocity change only slightly. However, as the containment pressure falls and as the various steam generation sources outside of the core cool, reducing the differential pressure across the break, the RCS pressure decreases. This decrease in RCS pressure increases the specific volume of the steam in the core, allowing less room for water and thus, decreasing the core collapsed water level. A better measure of a stable cooling inventory is the vessel mass, which shows that water is being supplied to the vessel at the rate that boiling is occurring. The reactor vessel mass shown in Figure 3-22 confirms a stable reactor vessel liquid mass from about 150 seconds until the end of the transient.

6.7 PlantInput Selection and Technical Specifications Question:

1. Please provide information to enable comparison between Technical Specifications (TS) requirements and analytic input parameters for Pressurizer Level. The TS requirement is given in inches and the input parametersare specified in percent span.

2. Please provide discussion to confirm that the assumed 80°F upper containment temperature and 95 0F lower containment temperature are acceptable minimums without a TS requirement.

AREVA NP Inc.

A AR'EtV'A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 136

3. The TS minimum for the refueling water storage tank (RWST) temperature is 60°F.

Previous, deterministic analyses demonstratedthat minimum safety injection temperatures resulted in a limiting PCT. In light of this information, please explain why a minimum RWST temperature case was not evaluated, or if a minimum RWST temperature case was evaluated,please summarize the evaluation and discuss its conclusions.

Response

1. Technical Specification LCO 3/4.4.9 states "The pressurizer shall be OPERABLE with pressurizer water level of less than or equal to 57% and at least two groups of pressurizer heaters each having a capacity greater than or equal to 150 kW." The Technical Specifications for SONGS do not have the requirement in inches, just percent span.

2. The sampled range for the liquid level uncertainty in the pressurizer was 22 to 61 percent of span.

3. The containment air temperature Technical Specifications requirement for SONGS Units 2 and 3 is _<120'F (TS 3.6.5/LCO 3.6.5). The RLBLOCA analysis for SONGS covers containment/SIT temperatures ranging from 50°F to 130 0 F.

4. The RWST borated water temperature Technical Specifications requirement for SONGS Units 2 and 3 is Ž_ 40°F and < 100°F (SR 3.5.4.1). The NRC-approved RLBLOCA EM, EMF-2103(P)(A), prescribes use of the maximum temperature for the ECCS pumped injection and use of the minimum temperature for containment sprays. The temperatures for the SONGS analysis are 1 10°F for pumped injection and 35°F for the containment sprays. While inconsistent, the choice of the two temperatures is conservative.

AREVA NP Inc.

A, AR.EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 137 7.0 References 1 EMF-2103(P)(A) Revision 0, Realistic Large Break LOCA Methodology, Framatome ANP, Inc., April 2003.

2 G.P. Liley and L.E. Hochreiter, "Mixing of Emergency Core Cooling Water with Steam:

1/3 - Scale Test and Summary," EPRI Report EPRI-2, June 1975.

3 Technical Program Group, Quantifying Reactor Safety Margins, NUREG/CR-5249, EGG-2552, October 1989.

4 XN-CC-39 (A)Revision 1, "ICECON: A Computer Program to Calculate Containment Back Pressure for LOCA Analysis (Including Ice Condenser Plants)," Exxon Nuclear Company, October 1978.

5 Wheat, Larry L., "CONTEMPT-LT A Computer Program for Predicting Containment Pressure-Temperature Response to a Loss-Of-Coolant-Accident," Aerojet Nuclear Company, TID-4500, ANCR-1219, June 1975.

6 U. S. Nuclear Regulatory Commission, NUREG-0800, Revision 3, Standard Review Plan, March 2007.

7 NUREG/CR-0994, "A Radiative Heat Transfer Model for the TRAC Code,"

November 1979.

8 D. A. Mandell, "Geometric View Factors for Radiative Heat Transfer within Boiling Water Reactor Fuel Bundles," Nucl. Tech., Vol. 52, March 1981.

9 EMF-CC-130, "HUXY: A Generalized Multirod Heatup Code for BWR Appendix K LOCA Analysis Theory Manual," Framatome ANP, May 2001.

10 NUREG/CR-1532, EPRI NP-1459, WCAP-9699, "PWR FLECHT SEASET Unblocked Bundle, Forced and Gravity Reflood Task Data Report," June 1980.

11 EMF-2102(P)(A) Revision 0, S-RELAP5: Code Verification and Validation, Framatome ANP, Inc., August 2001.

12 J.P. Holman, Heat Transfer, 4th Edition, McGraw-Hill Book Company, 1976.

13 EMF-2994(P) Rev. 4, "RODEX4: Thermal-Mechanical Fuel Rod Performance Code Theory Manual," December 2009.

14 ANF-90-145(P)(A), "RODEX3 Fuel Rod Thermal-Mechanical Response Evaluation Model," April 1996.

15 ANSI/ANS-5.1-1979, American National Standard for Decay Heat Power in Light Water Reactors, approved August 29, 1979.

16 AREVA Letter NRC:02:062, December 20, 2002, Responses to a Request for Additional Information on EMF-2103(P) Revision 0, "Realistic Large Break LOCA Methodology for Pressurized Water Reactors," (TAC No. MB2865).

17 P. Ihle, Heat Transfer in Rod Bundles with Severe Clad Deformations, KfK 3607 B, April 1984.

AREVA NP Inc.

A A RE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2975(NP)

Realistic Large Break LOCA Report Rev. 0 Page 138 18 M.J. Loftus, et al., PWR FLECHT SEASET 163-Rod Bundle Flow Blockage: Task Data Report, No. 13, NUREG/CR-3314, October 1983.

19 NEA/CSNI/R(2004)19, SEGFSM Topical Meeting on LOCA Fuel Issues, Argonne National Laboratory, May 25-26 2004, Published by Organization for Economic Cooperation and Development Nuclear Energy Agency, Isy-les-Moulineaux, France, November 2004.

AREVA NP Inc.

ENCLOSURE 7 ANP-2974(NP), Revision 0, "San Onofre Nuclear Generating Station Unit 2 and Unit 3 Small Break Report" July 2011 Non-Proprietary

ANP-2974(NP)

Revision 0 San Onofre Nuclear Generating Station Unit 2 and Unit 3 Small Break LOCA Report July 2011 A

AREVA NP Inc. AR EVA

A A R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 2 Copyright © 2011 AREVA NP Inc.

All Rights Reserved AREVA NP Inc.

A A R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 3 Nature of Changes Item Page Description and Justification 1 All Revision 0. This is a new document.

AREVA NP Inc.

AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 4 Table of Contents T a b le o f C o nte nts .......................................................................................................................... 4 List of T a bles ................................................................................................................................. 5 List of Figure s ............................................................................................................................... 6 No m e nc la tu re ................................................................................................................................ 7 1.0 Intro d uctio n ....................................................................................................................... 8 2.0 Sum mary of Results ...................................................................................................... 9 3.0 Description of Analysis ................................................................................................. 10 3.1 Description of SBLOCA Event ........................................................................ 10 3.2 Analytical Models ............................................................................................ 11 3.3 Plant Description and Summary of Analysis Input Parameters ....................... 13 3.4 SER Com pliance ............................................................................................ 14 4.0 Analytical Results ........................................................................................................ 25 4 .1 R e sults ................................................................................................................ 25 4.2 Discussion of Transient for Lim iting Break ..................................................... 26 4.3 RCP Trip Sensitivity Study .............................................................................. 27 4.3.1 RCP Trip Sensitivity Study for Cold Leg Breaks .............................. 28 4.3.2 RCP Trip Sensitivity Study for Hot Leg Breaks ................................. 28 4.4 Attached Piping LOCA Cases ......................................................................... 29 4.5 Results Com parison against AOR ................................................................... 29 4.5.1 Evaluation Model Differences ............................................................ 29 4.5.2 Input Differences .............................................................................. 30 4.5.3 Qualitative PCT Impact ..................................................................... 30 5 .0 C o nc lu s ions ................................................................................... .................................. 65 6 .0 Refe re nce s ...................................................................................................................... 66 AREVA NP Inc.

This document contains a total of 66 pages

A A R E'VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 5 List of Tables Table 3-1 Small break LOCA development phases ............................................................................... 15 Table 3-2 System Parameters and Initial Conditions ............................................................................. 16 Table 3-3 HPSI Injection Flow Rate versus RCS Pressure ................................................................. 17 Table 3-4 LPSI Injection Flow Rate versus RCS Pressure .................................................................... 18 Table 3-5 Moderator Density Reactivity ................................................................................................ 19 T a b le 3-6 Do p ple r R e activity ...................................................................................................................... 19 Table 4-1 Summary of SBLOCA Break Spectrum Results - Cold Leg Break ....................................... 33 Table 4-2 Sequence of Events for the SBLOCA Break Spectrum (seconds) ....................................... 34 Table 4-3 Summary of Results for the Cold Leg Break RCP Trip Sensitivity ....................................... 36 Table 4-4 Summary of Results for the Hot Leg Break RCP Trip Sensitivity ......................................... 37 Table 4-5 Analysis Input C om parison .................................................................................................... 38 AREVA NP Inc.

A:R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 6 List of Figures Figure 3-1 S-RELAP5 SBLOCA Reactor Coolant System Nodalization .............................................. 20 Figure 3-2 S-RELAP5 SBLOCA Secondary System Nodalization ....................................................... 21 Figure 3-3 S-RELAP5 SBLOCA Reactor Vessel Nodalization ............................................................ 22 Figure 3-4 A xia l P ow e r S ha pe ................................................................................................................... 23 Figure 3-5 Axial Power Distribution Comparison ................................................................................. 24 Figure 4-1 PCT versus Break Size (SBLOCA Break Spectrum) ......................................................... 39 Figure 4-2 Reactor Power - 9.00-inch Break ....................................................................................... 40 Figure 4-3 Primary and Secondary System Pressures - 9.00-inch Break ............................................ 41 Figure 4-4 Liquid and Vapor Break Flow - 9.00-inch Break ................................................................. 42 Figure 4-5 Vapor Void Fraction at the Break - 9.00-inch Break .......................................................... 43 Figure 4-6 Loop Seal Void Fraction - 9.00-inch Break ........................................................................ 44 Figure 4-7 Total Core Inlet Mass Flow Rate - 9.00-inch Break ............................................................ 45 Figure 4-8 RCS Loop Flow Rate - 9.00-inch Break ............................................................................ 46 Figure 4-9 Steam Generator Main Feedwater Flow Rates - 9.00-inch Break ..................................... 47 Figure 4-10 Auxiliary Feedwater Flow - 9.00-inch Break ...................................................................... 48 Figure 4-11 Steam Generator Total Mass - 9.00-inch Break ............................................................... 49 Figure 4-12 Steam Generator Level - 9.00-inch Break ........................................................................ 50 Figure 4-13 Total HPSI Mass Flow - 9.00-inch Break ........................................................................... 51 Figure 4-14 Total SIT Flow - 9.00-inch Break ...................................................................................... 52 Figure 4-15 Integrated Break Flow and ECCS Flow - 9.00-inch Break .............................................. 53 Figure 4-16 Total Primary Mass - 9.00-inch Break ............................................................................... 54 Figure 4-17 Reactor Vessel Mass - 9.00-inch Break .......................................................................... 55 Figure 4-18 Hot Assembly Collapsed Liquid Level - 9.00-inch Break ....................................................... 56 Figure 4-19 Hot Rod Mixture Level - 9.00-inch Break ........................................................................... 57 Figure 4-20 Downcomer Collapsed Liquid Level - 9.00-inch Break .......................................................... 58 Figure 4-21 Volume Liquid Temperature in Hot Assembly - 9.00-inch Break ........................................... 59 Figure 4-22 Liquid Void Fraction Distribution in the Hot Channel 0.75 - 7.75 ft Elevation

- 9 .0 0 -inc h Bre a k ........................................................................................................................... 60 Figure 4-23 Liquid Void Fraction Distribution in the Hot Channel 8.25 - 12.25 ft Elevation

- 9 .0 0 -inc h Bre a k ........................................................................................................................... 61 Figure 4-24 Heat Transfer Coefficient at PCT Location - 9.00-inch Break .......................................... 62 Figure 4-25 Peak Cladding Temperature at PCT Location - 9.00-inch Break ..................................... 63 Figure 4-26 Decay Pow er Fraction ...................................................................................................... 64 AREVA NP Inc.

A R'E VVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 7 Nomenclature ADV Atmospheric dump valve AFW Auxiliary feedwater AOR Analysis of Record ASI Axial Shape Index B&W Bobcock & Wilcox BOO Beginning-of-cycle CE Combustion Engineering CEA Control Element Assembly CFR Code of Federal Regulations DC Downcomer DEG Double-ended guillotine ECC(S) Emergency core cooling (system)

EDG Emergency diesel generator EFAS Emergency feedwater actuation signal EM Evaluation Model EOC End-of-cycle EOI Emergency Operating Instruction FSAR Final Safety Analysis Report HPSI High pressure safety injection LAR License Amendment Request LHR Linear heat rate LOCA Loss of coolant accident LOOP Loss of offsite power LPSI Low pressure safety injection MFW Main feedwater MSIV Main steam isolation valve MSSV Main steam safety valve MWt Mega-Watt thermal NPSH Net Positive Suction Head NRC Nuclear Regulatory Commission PCT Peak cladding temperature PD Pipe Diameter PWR Pressurized water reactor RAI Request for Additional Information RCP Reactor coolant pump RCS Reactor coolant system RWST Refueling Water Storage Tank SBLOCA Small Break Loss-of-Coolant-Accident SCE Southern California Edison SER Safety Evaluation Report SG Steam generator SGTP Steam generator tube plugging SI Safety injection SIAS Safety injection actuation signal SIT Safety injection tank SONGS San Onofre Nuclear Generating Station W Westinghouse AREVA NP Inc.

A IREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 8 1.0 Introduction This report documents the SBLOCA analysis for the San Onofre Nuclear Generating Station (SONGS) Unit 2 and Unit 3 for transition to AREVA fuel. The plant is a Combustion Engineering (CE) designed 3438 MWt plant with a large dry containment. The plant is a 2X4 loop design with two hot legs and four cold legs. The loops contain four reactor coolant pumps (RCPs), two U-tube steam generators and one pressurizer. The Emergency Core Cooling System (ECCS) is provided by two independent safety injection trains and four safety injection tanks (SITs).

The analysis supports operation with AREVA NP's HTP 16X16 fuel design with M5 cladding.

This analysis also supports plant operation at a power level of 3458 MWt (including 20 MWt uncertainty) and 8% steam generator tube plugging. The core peaking supported by this analysis is for a peak LHR technical specification limit of 12.8 kW/ft1 and a -0.3 _:ASI _<+0.3 band. The analysis was performed with the S-RELAP5 methodology (References [1] and [2]).

A spectrum of cold leg break sizes from 0.0055 ft2 (1-inch diameter) to 0.491 ft2 (9.49-inch diameter, 10% of cold leg pipe area) was analyzed. This is the only SER restriction on EMF-2328(P)(A) (Reference [1]). Additional study includes the reactor coolant pump trip sensitivity study, axial shape index sensitivity study and the attached piping (SIT line) break study 2 to support the operation of SONGS Unit 2 and Unit 3.

1 This is equivalent to a total peaking factor (FQ) up to a value of 2.37, and a radial peaking factor (Fr) up to a value of 1.755 (including 6% uncertainty and 3.5% control rod insertion uncertainty).

2 Although the SIT line break is greater than 10% of the cold leg pipe area, it was a specific request by the NRC.

AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 9 2.0 Summary of Results A SBLOCA break spectrum analysis was performed for SONGS Units 2 and 3. The limiting PCT is 1509°F for a break of 9.00-inch diameter (0.4418 ft 2) pump discharge cold leg break.

This analysis was performed to demonstrate that the following acceptance criteria for Emergency Core Cooling Systems, as stated in 10 CFR 50.46(b)(1-4), have been met.

(1) Peak cladding temperature:The calculated limiting fuel element cladding temperature is 1509°F, which is less than the 2200°F limit criterion.

(2) Maximum local cladding oxidation: The calculated maximum local oxidation of the cladding is 0.17%, which is less than the 17% limit of the criterion.

(3) Maximum core-wide oxidation: The calculated core-wide total oxidation is 0.0033%,

which is less than the 1% limit of the criterion.

(4) Coolable geometry: The cladding remains amenable to cooling. None of the cases analyzed predicted hot rod rupture, hence no blockage is predicted to occur, which would degrade core cooling. Further, an M5 mechanical deformation analysis presented in Appendix F of Reference [3] is not changed by the fuel design changes.

Both thermal and mechanical deformations of the fuel assemblies in the core have been assessed and the resultant deformations have been shown to maintain coolable core configurations.

The break spectrum calculations conservatively assumed RCP trip at reactor trip and loss of offsite power. An evaluation of delayed RCP trip was performed since a delayed RCP trip can potentially produce more limiting results. Section 4.3 discusses the RCP trip sensitivity study.

The results of the delayed pump trip evaluation for SONGS indicate that all four RCPs should be tripped no later than 7 minutes once the RCP NPSH criteria are met in order to assure that the 10 CFR 50.46(b)(1-4) criteria remain satisfied. The analysis, which included hot leg breaks, also demonstrates that the delayed RCP trip cases do not produce a limiting PCT.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 10 3.0 Description of Analysis Section 3.1 below provides a brief description of the postulated SBLOCA event. Section 3.2 describes the analytical models used in the analysis. Section 3.3 presents a description of SONGS Units 2 and 3 plant parameters and outlines the system parameters used in the SBLOCA analysis.

3.1 Description of SBLOCA Event The postulated SBLOCA is defined as a break in the RCS pressure boundary which has an area of up to approximately 10% of the cold leg pipe area. For SONGS, 10% of the cold leg pipe area is 0.491 ft2 with equivalent diameter of 9.49 inches. The base case break spectrum limiting break location is in the cold leg (i.e., on the discharge side of the RCP) assuming LOOP conditions. A break in the cold leg of a reactor system is considered to be the most limiting because it results in the maximum loss of ECCS fluid through the break during the course of the event, and adds severity to the transient by separating the core exit from the break via "loop seals." This produces the greatest degree of core uncovery, the longest fuel rod heatup time and consequently, the greatest challenge to the 10CFR 50.46(b) criteria.

A small break LOCA is assumed to occur while operating at 100.58 percent of full power. The transient can be generally characterized as developing in the following distinct phases: (1) subcooled depressurization, (2) loop saturation and loop flow coastdown, (3) loss of loop circulation and reflux mode cooling, (4) loop seal clearing and core refill and (5) long-term cooling provided by high head safety pump and SIT injections. Small break LOCA development phases are outlined in Table 3-1.

Following the break, the reactor coolant system (RCS) rapidly depressurizes to the saturation pressure of the hot leg fluid. During the initial depressurization phase, a reactor trip is generated on low pressurizer pressure; the turbine is tripped on the reactor trip. The assumption of a loss-of-offsite-power concurrent with the reactor trip results in reactor coolant pump trip 1 .

In the second phase of the transient, the reactor coolant pumps coastdown. In this phase natural circulation flows are sufficient to provide continuous core heat removal via the steam generators. Mass continues to be lost to the break during this period, however.

Tripping the reactor coolant pumps at the time of SCRAM instead of time zero is -

• A small delay relative to the time of loop seal uncovery for the limiting cases

" Considered to be representative of actual plant configuration, and

" Expected to be slightly conservative, due to the additional loss of primary system inventory through the break.

AREVA NP Inc.

AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 11 The third phase in the transient is characterized as a period of loop draining that results in the loss of RCS flows. During this period, the core decay heat removal is provided via reflux boiling.

The RCS stabilizes at an equilibrium pressure above the steam generator secondary side pressure. The system reaches a quiescent state in which the core decay heat, leak flow, steam generator heat removal, and system hydrostatic head balance combine to control the core inventory.

The fourth phase in the transient is characterized by loop seal clearing and core recovery. The RCS inventory continues to decrease. During this phase the loop seal, liquid trapped in the reactor coolant pump suction piping, can prevent steam from venting via the break. For a small break, the transient develops slowly, and liquid level in the reactor coolant system may descend to the loop seal level prior to establishing a steam vent. The core can become temporarily uncovered in this loop seal clearing process.

Once the loop seal clears, venting of steam through the break causes a rapid RCS depressurization below the secondary pressure. Boiling in the core increases. The depressurization also promotes an increase in ECCS flows and the mass loss through the break decreases substantially as a result of phase change. These occurrences combine to either cause the core to uncover and heat up or to cause an increase in RCS liquid inventory, preventing core uncovery.

The last phase of the transient is characterized as a long-term cooling period during which the RCS inventory control is provided by the Emergency Core Cooling System. Pumped injection will continue and the passive SIT injection will occur when the RCS pressure decreases below the SIT tank pressure. Long term RCS inventory and decay heat removal will be successfully controlled in this manner.

3.2 Analytical Models The AREVA S-RELAP5 SBLOCA evaluation model for event response of the primary and secondary systems and the hot fuel rod used in this analysis (Reference [1]) consists of two computer codes. The appropriate conservatisms, as prescribed by Appendix K of 10 CFR 50, are incorporated. This methodology has been reviewed and approved by the NRC to perform SBLOCA analyses. The two AREVA computer codes used in this analysis are:

1. The RODEX2-2A code was used to determine the burnup-dependent initial fuel rod conditions for the system calculations.

2. The S-RELAP5 code was used to predict the thermal-hydraulic response of the primary and secondary sides of the reactor system and the hot rod response.

The gap conditions used to initialize S-RELAP5 are taken at EOC, consistent with an EOC top-peaked axial power distribution. The use of EOC fuel rod conditions along with an EOC power shape is bounding of BOC because (1) the gap conductance is higher at EOC, and (2) the AREVA NP Inc.

A A R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 12 power shape is more top-skewed at EOC. The initial stored energy, although higher at BOC, has a negligible impact on SBLOCA results since the stored energy is dissipated long before core uncovery.

System nodalization details are shown in Figure 3-1 (RCS), Figure 3-2 (Secondary System),

and Figure 3-3 (Reactor Vessel). The RV internals are modeled in detail based on specific inputs supplied by SCE, including nodes and connectivity, flow areas, resistances and heat structures.

AREVA NP Inc.

A A R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 13 3.3 Plant Description and Summary of Analysis Input Parameters SONGS Unit 2 and Unit 3 are CE designed 2X4 loop PWRs with two hot legs, four cold legs, and two vertical U-tube steam generators. The reactor has a rated core power of 3458 MWt (including 20 MWt uncertainty). The reactor vessel contains a downcomer, upper and lower plenums, and a reactor core containing 217 fuel assemblies. The hot legs connect the reactor vessel with the vertical U-tube SGs. Main feedwater (MFW) is injected into the downcomer of each steam generator (SG). There are three auxiliary feedwater (AFW) pumps: two motor-driven pumps and one turbine-driven pump. The ECCS includes one HPSI, one low pressure safety injection (LPSI) and one SIT injection path per RCS loop.

The RCS was nodalized in the S-RELAP5 model into control volumes interconnected by flow paths or "junctions" (Figure 3-1). The model includes four SITs, a pressurizer, and two SGs with both primary and secondary sides modeled. All of the loops were modeled explicitly to provide an accurate representation of the plant. A steam generator tube plugging level of 8% in each steam generator was assumed. The HPSI system was modeled to deliver the total SI flow asymmetrically to the broken loop and three intact loops in the S-RELAP5 model. The LPSI system is included in the model.

The heat generation rate in the S-RELAP5 reactor core model was determined from reactor kinetics equations with actinide and decay heating as prescribed by 10CFR50 Appendix K.

The analysis assumed LOOP concurrent with reactor scram on low pressurizer pressure. The single failure criterion required by 10CFR50 Appendix K was satisfied by assuming the loss of one emergency diesel generator (EDG), which resulted in the disabling of one HPSI pump, one LPSI pump and the motor-driven AFW pump. Thus, a single HPSI pump was assumed to be operable. Charging pump flow was not credited in the analysis. Initiation of the HPSI system was delayed by 46 seconds beyond the time of SIAS. The 46-second delay represents the time required for EDG startup and switching. The disabling of the motor-driven AFW pump leaves one motor-driven AFW pump available. The turbine-driven AFW pump was not considered in the analysis.

The input model included details of both main steam lines from the SGs to the turbine control valve, including the MSSV inlet piping connected to the main steam lines. The main steam safety valves (MSSVs) were set to open at their nominal setpoints plus 2% tolerance.

Important system parameters and initial conditions used in the analysis are given in Table 3-2.

The degraded HPSI flow (Loss of 1 HPSI Pump with and without loss of EDG) used in the analysis is shown in Table 3-3. The LPSI injection flow is shown in Table 3-4.

The axial power shapes for this analysis is shown in Figure 3-4. Figure 3-5 compares the axial power shape at mid-node elevation for hot rod, hot assembly, inner and outer core used in the analysis.

AREVA NP Inc.

A.

San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 14 3.4 SER Compliance A spectrum of cold leg break sizes from 0.0055 ft2 (1-inch diameter) to 0.491 ft2 (9.49-inch diameter, 10% of cold leg pipe area) was analyzed, including the reactor coolant pump trip sensitivity study, axial shape index sensitivity study and the attached piping break study to support the operation of SONGS Unit 2 and Unit 3. This is the only SER restriction on EMF-2328 (Reference [1]).

AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 15 Table 3-1 Small break LOCA development phases (1) (2) (3) (4) (5)

SBLOCA Phase Subcooled Loop Loss of loop Loop seal Long-term

>>>> depressurization saturation and circulation and clearing and cooling loop flow reflux mode core refill coastdown cooling RCS Primary depressurization, hot leg RCS boils down loop seal clears, core covered Description reactor trip, saturates, RCS and flow is steam is vented and cooled by Decito ubn ' flow coasts interrupted by to break ECCS injection turbine trip, loss down void formation of offsite power Characteristics discharge subcooled / subcooled I saturated / subcooled I saturated saturated liquid superheated saturated liquid discharge discharge vapor discharge discharge RCS Flow forced flow and coastdown coastdown to natural stagnant steam flow to break pool boiling circulation forced reflux RCS Heat Removeal forced via convection steam convection va stneamio via condensation in steam boiling and boiling and Removal vasemtam steam break flow break flow generators generators generators, break flow pressure rapid rapid rapidaplateaus just depressurization slow depressurization depressurization above below depressurization secondary pressure secondary pressure dpesrzto continuous, continuous, ECCS Injection none none initiates potential SIT potential SIT injection injection Core Level covered covered covered uncovered or1 covered or2 uncovered core recovery 1 Depending on the loop seal elevation with respect to the top of the active core.

2 Depending on the break size.

AREVA NP Inc.

A ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 16 Table 3-2 System Parameters and Initial Conditions Reactor Power, MWt 34581 Radial Peaking Factor (includes uncertainty) 1.7552 Total Power Peaking Factor (includes uncertainty) 2.372 Minimum RCS Flow Rate, gpm 376200 Average Coolant Temperature, OF 590 Pressurizer Pressure, psia 2250 Core Inlet Coolant Temperature, OF 5603 SIT Pressure, psia 595 Maximum SIT Fluid Temperature, IF 130 Minimum SIT Water Volume, ft3 1650 SG Tube Plugging Level, % 8 SG Secondary Pressure, psia 878 MFW Temperature, OF 440.2 AFW Temperature, OF 110 Low SG Level EFAS Setpoint, %NR 0 HPSI Fluid Temperature, OF 1104 Pressurizer Pressure - Low Reactor Trip Setpoint (RPS), psia 1560.0 Reactor Scram Delay Time on Low Pressurizer Pressure, sec 0.9 Scram CEA Holding Coil Release Delay Time, sec 1.01 SIAS Activation Setpoint Pressure, psia 1560.0 HPSI Pump Delay Time on SIAS (LOOP), sec 46.0 MSSV lift pressures, psia 1129, 1136,1143,1151, 1158, 1165, 1172, 1178 1 Includes 20 MWt uncertainty.

2 For a peak LHR technical specification limit of 12.8 kW/ft and a -0.3:5 ASI < +0.3 band.

3 Includes 2°F uncertainty.

4 This is the analysis value for the maximum RWST temperature. The value includes measurement uncertainty.

The lowest set MSSV (1122 psia) was disabled. Pressures shown include 2% tolerance.

AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 17 Table 3-3 HPSI Injection Flow Rate versus RCS Pressure RCS Pressure Broken Loop Intact Loop (psia) (gpm) (gpm) 0.0 200.0 200.0 150.0 200.0 200.0 200.0 195.0 195.0 400.0 175.1 175.1 600.0 151.8 151.8 700.0 139.6 139.6 800.0 127.0 127.0 900.0 114.0 114.0 1000.0 98.9 98.9 1100.0 80.8 80.8 1200.0 57.2 57.2 1276.0 23.8 23.8 1277.0 0.0 0.0 AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 18 Table 3-4 LPSI Injection Flow Rate versus RCS Pressure RCS Broken Intact RCS Broken Intact Pressure Loop Loops Pressure Loop Loops (psia) (gpm) (gpm) (psia) (gpm) (gpm) 0.0 2229.9 2229.9 140.0 1037.4 1037.4 10.0 2167.3 2167.3 150.0 870 870 20.0 2104.1 2104.1 160.0 697 697 30.0 2038.5 2038.5 170.0 506.8 506.8 40.0 1972.6 1972.6 180.0 265.5 265.5 50.0 1902.6 1902.6 185 119.3 119.3 60.0 1830.8 1830.6 186 0 0 70.0 1748.2 1748.0 80.0 1667.9 1667.9 90.0 1582.9 1582.9 100.0 1492.8 1492.8 110.0 1397.5 1397.5 120.0 1293.7 1293.7 130.0 1173.4 1173.4 AREVA NP Inc.

A A R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 19 Table 3-5 Moderator Density Reactivity1 Table 3-6 Doppler Reactivity1 1

AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 20 Figure 3-1 S-RELAP5 SBLOCA Reactor Coolant System Nodalization AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 21 Figure 3-2 S-RELAP5 SBLOCA Secondary System Nodalization AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 22 Figure 3-3 S-RELAP5 SBLOCA Reactor Vessel Nodalization AREVA NP Inc.

Ar ARE EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 23 Figure 3-4 Axial Power Shape AREVA NP Inc.

A A R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 24 Figure 3-5 Axial Power Distribution Comparison AREVA NP Inc.

A A R E VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 25 4.0 Analytical Results The analysis results demonstrate the adequacy of the ECCS to support the criteria given in 10 CFR 50.46(b)(1-4) for SONGS Unit 2 and Unit 3 operating with AREVA supplied 16x16 HTP fuel with M5 cladding; as follows:

(1) Peak cladding temperature:The calculated limiting fuel element cladding temperature is 1509 0 F, which is less than the 22000 F limit criterion.

(2) Maximum local cladding oxidation: The calculated maximum local oxidation of the cladding is 0.17%, which is less than the 17% limit of the criterion.

(3) Maximum core-wide oxidation: The calculated core-wide total oxidation is 0.0033%,

which is less than the 1% limit of the criterion.

(4) Coolable geometry: The cladding remains amenable to cooling. None of the cases analyzed predicted hot rod rupture; hence no blockage is predicted to occur, which would degrade core cooling. Both thermal and mechanical deformations of the fuel assemblies in the core have been assessed and the resultant deformations have been shown to maintain coolable core configurations. Therefore, the coolable geometry requirements are met.

Section 4.1 describes the SBLOCA break spectrum for the cold leg break. Section 4.2 discusses for the limiting break transient. Section 4.3 describes the event for the limiting break.

Section 4.4 describes the impact of delayed RCP trip on the PCT for SBLOCA. The key parameters for the limiting case are shown in Figure 4-2 through Figure 4-25.

4.1 Results The base case break spectrum analysis for SBLOCA includes breaks of varying diameter up to 10% of the flow area for the cold leg. Once the preliminary limiting break size diameter (DLIM) was identified, additional calculations were performed in varying steps to a minimum of DLIM+/-0.01 in., in order to:

1. Cover the entire 0-10% small break area range; and

2. Demonstrate that the actual limiting break size has been identified within 0.01-inch diameter.

The break spectrum calculations were executed for breaks of 1.00, 2.00, 2.25, 2.40, 2.42, 2.46, 2.50, 2.70, 3.00, 3.25, 3.50, 3.75, 4.00, 4.50, 5.00, 5.50, 6.00, 6.50, 7.00, 7.50, 8.00, 8.50, 8.75, 8.97, 8.98, 8.99, 9.00, 9.01, 9.02, 9.03, 9.25, and 9.49-inch diameter (the 9.49-inch diameter break corresponds to an area equal to 10% of the cold leg flow area). The results for the break spectrum calculations are presented in Table 4-1. Predicted event times are summarized in Table 4-2.

AREVA NP Inc.

A ARF EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 26 Figure 4-1 shows the calculated PCTs for these breaks.

The limiting PCT is 1509°F for a break of 9.00-inch diameter (0.4418 ft2).

The break spectrum calculations assumed RCP trip at reactor trip due to an assumed LOOP at reactor trip.

4.2 Discussion of Transient for Limiting Break The Low Pressurizer Pressure Trip setpoint is reached at 16 seconds and within 2 seconds the reactor is tripped, the offsite power is lost, coincident with the turbine trip, RCP' trip, Main Feedwater Pump trip, and MSIV closure. The Safety Injection Actuation Signal (SIAS) is issued on Low Pressurizer Pressure.

The primary side depressurizes rapidly for the first approximately 50 seconds, after which it decreases more slowly up to approximately 175 seconds. Similar behavior is also seen in the loop flow rates found in Figure 4-8. It then remains relatively steady for the remainder of the transient. From Figure 4-3, the secondary side pressure remains constant until MSIV closure at approximately 18 seconds. The pressure increases until approximately 34 seconds. Then the MSSV inlet reaches the lowest opening pressure setpoint, allowing one MSSV in each steam line to open. This rapidly reduces the SG pressure and mass inventory (Figure 4-11) until the closing pressure setpoint is reached at approximately 85 seconds. The SG pressure slowly decreases through the rest of the transient. At approximately 85 seconds, the primary side pressure drops below that of the secondary side, ending SG secondary side inventory reduction and producing SG secondary side condensation (Figure 4-12). The single motor driven AFW pump available for the transient is not actuated (Figure 4-10).

The continued depressurization of the primary system causes the flashing of the liquid and the generation of steam bubbles that tend to accumulate to the high elevation points, e.g. the bend of the Steam Generator U-tubes. The primary system mass decreases until 150 seconds and then begins to recover (Figure 4-16). The liquid in the pressurizer drains rapidly and liquid in the core starts to boil as the saturation pressure is reached and progressively the nodes in the hot assembly start voiding. As a result, the collapsed liquid level in the hot assembly decreases rapidly at the beginning. Because the heat source decreases rapidly following the reactor trip, some of the steam bubbles that had formed initially collapse and the level briefly recovers around 30 seconds into the transient. The level then continues to decrease until the entire hot assembly is completely void of liquid around approximately 75 seconds. The level recovers temporarily from approximately 90 to 125 seconds, then begins full recovery at approximately 150 seconds. (Figure 4-18).

The hot rod mixture level remains at the top of the core up to approximately 50 seconds and then drops abruptly to the bottom over the next 10 seconds. The level then recovers to 5 ft. at approximately 125 seconds, drops briefly and then recovers to about 11 ft. at approximately 200 seconds and remains relatively steady thereafter (Figure 4-19). The liquid temperature in AREVA NP Inc.

A AR EtVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 27 the hot assembly (Figure 4-21) has inflection points at similar timing as the changes seen in the hot rod mixture level. The void fraction for the hot assembly is displayed at all elevations in the core on Figure 4-22 and Figure 4-23.

The loop seals for the three intact loops clear at approximately 70 seconds, and 96 seconds for the broken loop (Figure 4-6), allowing for a slightly higher depressurization rate after clearance as well as a small increase in inventory to the vessel at 70 seconds (Figure 4-20). Break uncovery occurs at 82 seconds (Figure 4-5). When the break uncovers, the amount of liquid lost out the break decreases sharply, and steam is released from the break (Figure 4-4 and Figure 4-5) releasing energy from the primary system.

Adiabatic heatup occurs between approximately 50 seconds and 180 seconds (Figure 4-24).

Meanwhile, the HPSI pumps start delivering fluid to the cold legs at 54 seconds (Figure 4-13),

but the liquid injected cannot make up for the large amount of fluid lost to the break up to this point (Figure 4-15) and the clad temperature continues to rise (Figure 4-25). The SIT injection begins at 132 seconds into the event (Figure 4 -14) and the large amount of liquid added to the primary system turns the transient around. The LPSI flow begins at 194 seconds. The downcomer level and the reactor vessel inventory start increasing at about 175 seconds (Figure 4-17 and Figure 4-20).

Cladding temperature reaches its peak of 1509°F at 160 seconds, and then drops as core cooling improves due to the increased inventory.

Figure 4-26 shows the decay power fraction used in the analysis.

4.3 RCP Trip Sensitivity Study A delayed RCP trip sensitivity study has been performed. For plants that do not have an automatic RCP trip, a delayed RCP trip can potentially result in a more limiting condition than tripping the RCPs at reactor trip. Continued operation of the RCPs can result in earlier loop seal clearing, with associated two-phase flow out the break, which would result in less inventory loss out the break early in the transient but more overall inventory loss out the break in the longer term. It is possible that tripping the pumps when the minimum RCS inventory occurs could potentially cause a collapse of core voids, thus depressing the core level and provoking a deeper core uncovery and resulting in a potentially higher PCT. Delayed RCP trip is required to be analyzed to support the Emergency Operating Instruction (EOI).

The SONGS EOI, "Standard Post Trip Actions" states:

• If pressurizer pressure is less than 1430 psia, then ensure one RCP in each loop is stopped.

AREVA NP Inc.

A, San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 28 If RCP NPSH requirements of Attachment 3 (Post-Accident Pressure/Temperature Limits) are not satisfied, then ensure all RCPs are stopped.

Observing the behavior of several of the break spectrum runs, the cold leg temperature tends to remain fairly constant for a period after reactor trip, while pressure decreases rapidly. At the nominal full-power TCoId temperature, the NPSH curve shows that the pressure limit is approximately 1600 psia. Pressure is falling rapidly in this range, and the difference between reaching the NPSH curve and the 1430 psia value is only a few seconds.

For SONGS, a 7 minute delay is imposed after pressure reaches 1430 psia before the RCPs are tripped. The 1430 psia is adjusted downward to 1339 psia to account for a harsh condition uncertainty of 91 psi.

Both cold leg and hot leg breaks are considered for the RCP trip sensitivity study.

4.3.1 RCP Trip Sensitivity Study for Cold Leg Breaks For the cold leg break RCP trip sensitivity study, the entire spectrum of break sizes from 3.0 inch to 9.49 inch diameter was re-analyzed with a delayed RCP trip.

If the pressurizer pressure less than 1339 psia, all 4 RCPs are tripped after a 420 second (7 minute) delay. Otherwise, the input is identical to the corresponding break spectrum run with the RCPs tripped at time of reactor trip, with the exception that for breaks larger than 4.0 inches, the turbine control valve was inadvertently not set to a constant position.

A summary of results of the cold leg break RCP trip sensitivity runs are presented in Table 4-3.

The results of the cold leg breaks RCP trip sensitivity study do not impact the limiting PCT of 1509'F from the SBLOCA break spectrum study.

4.3.2 RCP Trip Sensitivity Study for Hot Leg Breaks For the hot leg breaks RCP trip sensitivity study, the same selected breaks as for the cold leg were analyzed with a delayed RCP trip. For these cases the RCP trip time based on EOls is the same. The transient input file has been modified to change the RCP trip time for all four pumps.

Otherwise, the input is identical to the corresponding break spectrum run with the RCPs tripped at time of reactor trip, with the exception that for breaks larger than 4.0 inches, the turbine control valve was inadvertently not set to a constant position. This allows the valve to close slightly as the RCS heats up prior to reactor trip. This tends to conservatively hold up the RCS pressure. These break sizes are large enough that the period of time before the reactor trip is short, so this effect is minimized. The loop seal biasing for each run was the same as in the corresponding break spectrum run. A summary of results of the hot leg break RCP trip sensitivity runs are presented in Table 4-4.

AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 29 The maximum PCT from the sensitivity study is 1470°F for the 4.00-inch diameter break. The results of the hot leg breaks RCP trip sensitivity study do not affect the limiting PCT of 1509OF from the SBLOCA break spectrum study.

The results are similar to the cold leg break behavior. A comparison of the PCTs for the limiting cold leg and hot leg break sensitivity study cases shows that the hot leg break results in the higher PCT, but still lower than the limiting break spectrum results. The conclusion of the RCP trip study is that the operator has 7 minutes to manually trip all RCPs once the RCP NPSH criteria are met.

4.4 Attached Piping LOCA Cases Sensitivity calculations for breaks in the attached piping have also been performed. These consist of a double ended guillotine (DEG) break of the SIT line.

The calculated PCT was 1461°F, which is less limiting than the maximum PCT of the break spectrum cases documented in the body of this report. The minimal HPSI and LPSI flow in this analysis are sufficient to prevent a subsequent heatup after the initial quench from the SIT discharge.

4.5 Results Comparison against AOR 4.5.1 Evaluation Model Differences The current Analysis of Record (AOR) is based on the W SBLOCA evaluation Supplement 2 Model (FSAR Section 15.6.4). It uses a decoupled approach in which the CEFLASH-4AS computer code is used to perform the hydraulic analysis of the RCS until the safety injection tanks begin to inject. After injection from the safety injection tanks begins, the COMPERC-1I computer code is used to perform the hydraulic analysis in conjunction with CEFLASH-4AS.

The hot rod cladding temperature and maximum cladding oxidation are calculated by the STRIKIN-II computer code during the initial period of forced convection heat transfer and by the PARCH computer code during the subsequent period of pool boiling heat transfer. The initial steady state fuel rod conditions used in the analysis are determined using the FATES3B computer code.

The AREVA analysis documented herein uses an integral approach in which a single computer code, S-RELAP5, is used to calculate the thermal-hydraulic response of the system throughout the entire transient. The initial steady state fuel rod conditions used in the analysis are determined using the RODEX2 computer code.

This difference in evaluation models is one of the main reason for the difference in results,. The decoupled approach is much more conservative because it lacks the ability to model the complex interaction between various phenomena that is captured in the integral approach. One AREVA NP Inc.

0

A ARF: EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 30 such phenomenon is the cross flow between core channels due to the "hot chimney" effect in the hot channel, which may also be amplified by the differences in radial peaking factors.

In addition, based on the discussion in Section 6.3.3.3 of the UFSAR, the AOR assumed all safety injection flow delivered to the broken leg to be spilled out the break. The present analysis does assume, per the approved EM, that ECC flow will be injected into the broken loop and allows the code to determine how this flow will be directed to the break. This allows for liquid from the ECC to be diverted to the break, leaving the corresponding fluid in the primary and available later for core cooling, which otherwise would have been expelled to the break.

Additionally, it has as secondary effect some condensation of the steam in the broken cold leg due to the contact with colder ECC water injected, providing pressure and energy relief for the primary. This difference can also be a cause for the dissimilar results.

4.5.2 Input Differences Some differences were identified between the inputs used in the AREVA analysis model and the AOR in Table 4-5. The most important factors affecting the transient behavior in SBLOCA are the very significant differences in Steam Generator tube plugging, the decrease in Linear Heat Rate limit, the increase in ASI value and the decrease in axial peaking factor.

Based on the AREVA analysis model, the difference in tube volume due to the reduced tube plugging accounts for more than 580 ft3 of the primary (for both steam generators). At an average density of 45 Ibrm/ft3 , this means an excessive liquid mass of over 26,000 Ibm that would have to be drained from the tubes and expelled to the break in comparison to the AOR. This may also contribute to the loop seal clearing delay.

The present analysis uses a LHR Technical Specification limit of 12.8 kW/ft, which is 0.7 kW/ft lower than the AOR and it contributes significantly to the difference. A lower LHR limit would have as an effect a lower power generated in the hot rod and this first order effect would tend to reduce the PCT.

Another contributing factor is the difference in axial shape index. The current analysis uses an ASI of -0.172, compared to an ASI of -0.3. The ASI (-0.172) used in the current analysis will have as an effect an increase in the power generated in the lower portion of the core (compared to the AOR), and thus to a higher decay heat contribution in that region. This contribution will result in a higher steam production, therefore a higher mixture level. The mixture level will decrease the temperature of the steam in the upper portion of the core, improve heat transfer and thus a lower PCT is expected. This effect may also be amplified by the increased radial peaking factor due to the reduced axial peaking factor.

4.5.3 Qualitative PCT Impact The following section provides a qualitative estimation of PCT impact. The qualitative estimation is based on AREVA's extensive SBLOCA methodology experience with S-RELAP5 AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 31 (EMF-2328(P)(A) Rev. 0) and RELAP5/MOD2-B&W (BAW-10168(P)(A) Rev. 3). Where possible, a quantitative estimation of PCT impact will be provided.

Without access to the complete results and knowledge of the SONGS FSAR AOR Evaluation Model, it is difficult to do more than a qualitative evaluation of the differences in results. It is expected that SCE will provide a quantitative estimation for the SONGS SBLOCA AOR differences in the LAR application.

The values given below are a first-order approximation that can be used to obtain a qualitative evaluation of the expected differences in results between the AOR case and the current AREVA methodology based case results. Items of major impact on PCT are discussed below. All the discussions are based on a direct comparison between the corresponding 2.70 in PD (-0.04 ft 2) break cases in the current AREVA analysis and the FSAR AOR case.

1, Cladding rupture:

The FSAR AOR limiting case (0.04 ft2 break) showed cladding rupture at cladding temperature of about 1900°F and the temperature increased to 2070°F most likely due to the inside metal-water reaction. In the AREVA SBLOCA case there is no cladding rupture and the PCT is below 18000F, the effect of inside and outside metal-water reaction will be minimal and therefore the net impact on PCT is estimated to be around 2000F.

2. De-coupled vs. Integral Methodology.

As stated earlier, a combination of several computer codes were used in obtaining the PCT in the AOR case. Based on the AREVA's understanding of the decoupled approach, the hot rod thermal-hydraulic evaluation was done using the average core steaming rate above the core mixture level. In reality, there will be substantial cross-flow of steam from surrounding bundles into the hot channel will occur due to the "chimney effect" resulting from higher Fr in the hot channel. On the other hand, this "chimney effect" will be properly calculated in the S-RELAP5-based integral approach. Based on AREVA's SBLOCA analysis experience for 4-loop W-PWR, 3-loop W-PWR, and B&W plants, the impact of this "chimney effect" is a significant PCT reduction.

3. ASI difference:

The ASI of about -0.3 was used in the AOR case and about -0.172 in the AREVA analysis. This results in the use of Fr about 1.69 which occurs near the top of the hot rod in the AOR case versus an Fr of 1.35 in the S-RELAP5-based case. The higher axial peak in the upper region of the core in the AOR case will results in the lower mixture level and the lower steam production at the mixture level compared to the S-RELAP5-based case. Its effect is amplified in the AOR case due to the use of de-coupled methodology. The lower ASI has less impact on the S-RELAP5-based integral approach due to cross flow between different core channels. The impact on the AOR case is estimated to be a reduction of PCT.

AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 32

4. Peak LHR The peak LHR of 13.5 kW/ft was used in the AOR case and 12.8kW/ft in the S-RELAP5-base case. Assuming a normalization factor based on PCT above saturation temperature, the impact of reducing LHR by 0.7 kW/ft is estimated to be about 100 0 F.

5. Broken Cold Leg ECC Injection In the AOR methodology, the ECC injected in the broken cold leg is assumed to be spilled directly into containment. In addition, the ECC injected into the intact cold leg is assumed to be injected directly into the downcomer, thus potentially avoiding the steam-ECC interaction. It is noted that in the AOR case the mixture level falls below the top of the core at around 1050 seconds and this occurs around 2000 seconds in the S-RELAP5 case. The loop seal clearing occurred at about the same time in both cases. A sensitivity study using S-RELAP5 was performed by not injecting about 25% of total HPSI flow and the SIT connected to the broken cold leg. This resulted in PCT increase of about 250°F and the core uncovery occurred at about 200 seconds earlier than in the base case. The ECC injection directly into the downcomer in the AOR case can possibly result in highly subcooled water entering the core.

This can result in lower steam production and lower mixture level.

6. Core Uncovery Timing Several items could contribute to the approximately 1000 second delay in core uncovery time in the AREVA analysis versus the SONGS FSAR AOR. One potential item observed in the S-RELAP5 case is the draining of liquid in the steam generator during the transient. Another item is the power distribution in the core. The power distribution differences would change the transition of the core to a so-called "boiling pot" of liquid. Another item is differences in system mass and modeling nodalization. This is most likely the key difference in the two cases.

However, without knowledge of the AOR methodology, it is difficult to quantify this impact on PCT. The later core uncovery timing results in a reasonable reduction in PCT. The AOR case used 30% tube plugging whereas the AREVA case used 8%, which would delay the loop seal clearing time in the AREVA case by about 100 seconds. As stated earlier, ECC is not injected in the AOR case that will delay the core uncovery time by about 200 seconds. The mixture level calculation will be affected by the slip models used in the code. The flow distribution following the loop seal clearing will affect the time of core uncovery.

7. Net PCT Impact The net impact on AOR PCT due to the major items that can have substantial PCT impact is roughly equivalent to the differences in the PCT results between the AOR and the AREVA analysis for the 2.7-inch break. It is expected that SCE will provide a quantitative estimation for the SONGS SBLOCA AOR differences in the LAR application.

AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 33 Table 4-1 Summary of SBLOCA Break Spectrum Results - Cold Leg Break Break diameter (in) 1.00 2.00 2.25 2.40 2.42 2.46 2.50 2.70 Peak Clad Temperature (*F) 688.0 688.0 989.5 942.5 1166.5 1128.0 1102.2 987.0 Time of PCT (sec) 20 2.0 4150.0 3482 3027.0 3057.0 3160.0 2684.0 Time of Rupture (sec) N/A N/A N/A N/A N/A N/A N/A N/A Local Maximum Oxidation (%) 0.0005 0.0003 0.0207 0.0095 0.0821 0.0633 0.0517 0.0171 Core Wide Oxidation (%) 0.0003 0.0002 0M0006 0.0003 0.0023 0.0017 0.0013 0.0004 PCT Elevation (ft) 11.13 11.13 12.13 11.63 12.13 12.13 12.13 11.63 Break diameter (in) 3.00 3.25 3.50 3.75 4.00 4.50 5.00 5.50 Peak Clad Temperature ('F) 688.0 694.6 842.9 846.2 907.9 850.2 913.0 988.0 Time of PCT (sec) 2.0 411.0 91.0 81.0 72.0 58.0 48.0 188.0 Time of Rupture (sec) N/A N/A N/A N/A N/A N/A N/A N/A Local Maximum Oxidation (%) 0.0002 0.0002 0.0004 0.0004 0.0009 0.0007 0.0018 0.0047 Core Wide Oxidation (%) 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0002 PCT Elevation (ft) 11.13 11.63 11.63 11.63 11.63 11.63 11.63 11.63 Break diameter (in) 6.00 6.50 7.00 7.50 8.00 8.50 8.75 8.97 Peak Clad Temperature ('F) 1069.5 1183.5 1255.3 1306.6 1357.9 1426.5 1444.4 1460.8 Time of PCT (sec) 177.0 177.0 171.0 169.0 164.0 169.0 158.0 156.0 Time of Rupture (sec) N/A N/A N/A N/A N/A N/A N/A N/A Local Maximum Oxidation (%) 0.0110 0.0316 0.0549 0.0826 0.1138 0.1488 0.1348 0.1575 Core Wide Oxidation (%) 0.0003 0.0007 0.0010 0.0014 0.0019 0.0026 0.0025 0.0031 PCT Elevation (f) 11.63 11.63 11.63 11.63 11.63 11.63 11.63 11.63 Break diameter (in) 8.98 8.99 9.00 9.01 9.02 9.03 9.25 9.49 Peak Clad Temperature (°F) 1507.1 1493.7 1509.0 1473.6 1496.4 1451.1 1439.6 1433.3 Time of PCT (sec) 159.0 157.0 160.0 159.0 159.0 154.0 146.0 136.0 Time of Rupture (sec) N/A N/A N/A N/A N/A N/A N/A N/A Local Maximum Oxidation (%) 0.1611 0.1709 0.1661 0.1561 0.1564 0.1548 0.1318 0.1100 Core Wide Oxidation (%) 0.0033 0.0033 0.0032 0.0030 0.0030 0.0030 0.0026 0.0023 PCT Elevation (ft) 11.63 11.63 11.13 11.63 11.63 11.63 11.63 11.63 AREVA NP Inc.

A REVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 34 Table 4-2 Sequence of Events for the SBLOCA Break Spectrum (seconds) 0CF 0 0 00 a 0 0 C i~~~~ u uC C~ J2uC 2.46 74 4-212 0 29678 2E 3 30 02 S~l Z 75-C.

Fa o0~ o) 302

. 4 "a

01 C

~ CL.. 0 DI 0~ I..

0I 02.

9: -

u 03L

0. U.

0E 0

a.)u 0 CuC 0 oo.

Cu CUt V5 ) 4 0 0 0 0 M . 1-ý 0 8 0.0 08 020 030 0 0 90 - 335 - 342 22 0) 2.20 17 8 21 0 302770 78

• 2818. -- 32 2 02 1.00 0 1048 1050 1086 - 1174 - - - - 2184 - 2 2.5 00a 71 26 0)9 29071 0 72 66 288 - 16 30 2.00 0 266 268 304 - 390 - -- 1118 1128 - 5500 - 2 - -

27 0 144 14 18 -- 26 2U60 370 25U- 64 018 -

2.25 0 208 210 246 - 332 - -- 862 872 - 3860 - 4150 3210-2.40 0 183 185 220 - 306 - -- 780 790 - 3354 - 3482 2828-2.42 0 179 181 216 - 302 - -- 770 780 - 2818 - 3027 2302-2.46 0 174 176 212 - 296 - -- 738 748 4926 2964 - 3057 2380 -

2.50 0 169 171 206 - 290 - - 718 728 4616 2888 -3160 2340 -

2.70 0 144 146 182 - 266 - -- 620 630 3570 2658 - 2684 2178 -

3. 00' 0 117 119 154 - 238 525 525 *486 544 2752 2758 - 2 302 -

3.25 0 100 102 138 - 220 * *

• 412 454 2100 2108 - 411 274 -

3.50 0 86 88 124 - 206 * *

• 368 410 1598 1606 - 91 274 -

3.75 0 76 78 114 - 196 360 *

• 328 372 1264 1272 - 81 242 - I The loop seal clears after the time of SIT injection

+ Multiple loop seals clear for this break size. This result was validated by performing a case where only one loop seal clears.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 35 Table 4-2 (Continued)

• 0 C a) cc C---- 0 0 A! CL c .. wO 0 CO U UE 04 M V 0(

0C)

E a V 0 N 5V 0

a. CL M 0a).E. 2 2 =1 C*

58 0

54 00 C EE 1

• W 0

• C 6.50 0 0 0 0 0 CCC 100 0 0 2 6 020 L3 8-

-J M*JE 61 .0 a. 0. -0 -J -J -J CO () ~ 0 x (0Z.

4.00 5*5 0 0 67 36 68 38 104 74 -- 186 15

• 07 320 17

• 328 18 1020 4212 40 - 72 206-4 U-18 12 4850 0 53 55 90 - 170 238 252 264 712 718 - 58 164 -

5.00 0 43 45 80 - 162 202 208 220 522 528 - 48 138 -

5.50 0 36 38 74 - 152 174 178 186 404 428 - 188 120 -

6.00 0 31 33 68 - 146 166 156 190 164 326 348 - 177 102 -

6.50 0 27 29 64 - 150 144 140 132 142 276 294 - 177 90 -

7.00 0 24 25 60 280 146 110 150 114

• 126 238 256 - 171 80 -

7.50 0 21 23 58 246 148 96 122 98 122 1112 202 222 - 169 72 -

8.00 0 19 21 56 842 144 90 90 148 86 100 174 192 - 164 64 -

8.50 0 17 19 54 186 # 80 80 80 92 90 150 166 - 169 60 968 8.75 0 17 19 54 668 # 74 74 74 104 86 140 152 - 158 58 882 8.97 0 16 18 54 190 # 70 72 72 98 82 132 146 - 156 56 -

8.98 0 16 18 54 194 - 70 70 72 96 82 132 148 - 159 56 -

8.99 0 16 18 54 196 - 70 70 70 96 82 132 148 - 157 56 -

9.00 0 16 18 54 194 - 70 70 70 96 82 132 144 - 160 54 266 9.01 0 16 18 54 190 - 70 70 70 96 82 132 146 - 159 54 -

902 0 16 18 54 194 - 70 70 70 96 80 132 146 - 159 54 270 9.03 0 16 18 54 166 # 70 70 70 96 80 132 144 - 154 56 -

9.25 0 16 18 52 182 - 68 68 68 80 78 124 126 - 146 54 246 9.49 0 15 17 52 168 - 64 64 64 76 74 118 120 - 136 54 238

• The loop seal clears after the time of SIT injection

1. AFW injection may occur after the time of PCT.

AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 36 Table 4-3 Summary of Results for the Cold Leg Break RCP Trip Sensitivity Core Local Break PCT Time of Wide Maximum diameter PCT Elevation Rupture Oxidation Oxidation (in) (F) (ft) (s)  !% - (%)

3.00 688 11.13 1..17E-04 1.69E-04 3.25 838 12.13 1.18E-04 1.29E-03 3.50 934 11.63 1.35E-04 3.41E-03 3.75 1046 11.63 - 2.27E-04 9.62E-03 4.00 1108 11.63 3.29E-04 1.50E-02 4.25 1215 11.63 7.32E-04 2.99E-02 4.50 1080 11.63 2.86E-04 1.06E-02 4.75 952 11.63 1.27E-04 3.03E-03 5.00 888 11.13 8.38E-05 1.04E-03 5.50 726 11.13 6.43E-05 1.51 E-04 6.00 688 11.13 5.94E-05 8.77E-05 6.50 688 11.13 5.55E-05 8.32E-05 7.00 869 11.63 5.22E-05 6.73E-04 7.50 872 11.63 4.94E-05 7.02E-04 8.00 962 11.63 4.73E-05 1.77E-03 8.50 974 11.63 4.51E-05 2.14E-03 8.75 975 11.63 4.45E-05 2.23E-03 9.00 979 11.63 4.36E-05 2.41E-03 9.25 980 11.63 4.46E-05 2.49E-03 9.40 1060 11.63 4.47E-05 4.13E-03 9.49 1072 11.63 1 4.51E-05 4.71E-03 AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 37 Table 4-4 Summary of Results for the Hot Leg Break RCP Trip Sensitivity Core Local Break PCT Time of Wide Maximum diameter Elevation Rupture Oxidation Oxidation (in) PCT (F) (ft) J (%) (%)

3.00 688 11.13 1.34E-04 2.06E-04 3.25 804 12.13 1.30E-04 1.88E-04 3.50 1128 12.13 7.71E-04 4.43E-02 3.75 1216 12.13 - 9.93E-04 5.16E-02 4.00 1470 11.63 - 5.27E-03 1.63E-01 4.25 1189 11.63 - 7.02E-04 2.42E-02 4.50 995 11.63 - 1.73E-04 4.23E-03 4.75 870 11.13 - 943E-05 9.29E-04 5.00 801 11.13 - 7.67E-05 3.86E-04 5.50 823 11.63 - 6.88E-05 7.11E-04 6.00 854 11.13 - 6.33E-05 7.93E-04 6.50 833 11.13 - 5.85E-05 5.01E-04 7.00 869 11.63 - 5.22E-05 8,62E-05 7.50 840 11.13 - 5.14E-05 5.05E-04 8.00 836 11.13 - 4.86E-05 4.52E-04 8.50 807 11.13 - 4.55E-05 3.14E-04 8.75 777 11.13 - 4.41E-05 2.05E-04 9.00 827 11.13 - 4.41E-05 4.12E-04 9.25 801 11.13 - 4.26 E-05 2.90E-04 9.40 776 11.13 - 4.20E-05 2.07E-04 9.49 788 11.13 - 4.13E-05 2.38E-04 AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 38 Table 4-5 Analysis Input Comparison AOR Analysis Units Parameter Value Value Reactor Power (including measurement uncertainty) 3458 3458 MWt Peak Linear Heat Rate (PLHR) of the Hot Rod 13.5 12.8 kW/ft Axial Shape Index -0.3 -0.172 -

4 Moderator Temperature Coefficient at Initial Density 0.0x10 0.0 Ap/°F RCS Flow Rate 139.4x10 6 139.48x10 6 Ibm/hr Core Flow Rate 135.2x10 6 135.28x10 6 Ibm/hr RCS Pressure 2250 2250 psia Cold Leg Temperature 560 560 OF Hot Leg Temperature 621 621 OF Number of Plugged Tubes per Steam Generator 2805 779 -

Main Steam Safety Valve First Bank Opening Pressure 1122 1129 psia Low Pressurizer Pressure Reactor Trip Setpoint 1560 1560 psia Low Pressurizer Pressure SIAS Setpoint 1560 1560 psia Time Delay for Actuation of HPSI Flow (with loss of 36.2 46.0 Sec offsite power) 36.2_46.0 Sec No difference in flow per intact loop, however, the AOR assumes all safety HPSI Pump Flow Rate Versus RCS Pressure injection flow delivered to the broken leg to be spilled out the break, while the current analysis allows ECC to be delivered to the broken loop.

Safety Injection Tank Pressure 595 595 psia AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 39 1600 1400- Aý6i A~n A

1200- A 1000-A A IL AA A I- 800 U

a-A A AA 600 400 200 0.00 1.00 2.00 3.00 4M00 5.00 6.00 7.00 8.00 9.00 1000 Break Size (in)

Figure 4-1 PCT versus Break Size (SBLOCA Break Spectrum)

AREVA NP Inc.

A AR E -V:A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 40 Reactor Power 4000.0 7

3000.0

-- Power 2000.0 1000.0 0.0 0.0 200.0 400.0 600.0 D 1*1 13:1721 DEMUX Time (s)

Figure 4-2 Reactor Power - 9.00-inch Break AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 41 System Pressures 3000.0 2000.0 0.

1000.0 400.0 600.0 Time (s)

Figure 4-3 Primary and Secondary System Pressures - 9.00-inch Break AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 42 Liquid and Vapor Break Flow 8000.0 6000.0 4000.0 20000 Time (s)

Figure 4-4 Liquid and Vapor Break Flow - 9.00-inch Break AREVA NP Inc.

A A RAEVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 43 Break Vapor Void Fraction g

• 5 12_

600.0 Time (s)

Figure 4-5 Vapor Void Fraction at the Break - 9.00-inch Break AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 44 Loop Seal Void Fraction 0.6 0

0.

600.0 Time (s)

Figure 4-6 Loop Seal Void Fraction - 9.00-inch Break AREVA NP Inc.

ARE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 45 Total Core Inlet Mass Flow Rate 40000.0 30000.0 20000.0

¢,-

o*

E 10000.0 0.0

-10000,0 Time (s)

Figure 4-7 Total Core Inlet Mass Flow Rate - 9.00-inch Break AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 46 RCS Loop Flow Rates 15000.0 5000.0 0t

-50000 0.0 200.0 4(4015JUo,,'1( 2* oo, 0o .ux Time (s)

Figure 4-8 RCS Loop Flow Rate - 9.00-inch Break AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 47 SG MFW Flow Rates 3000O

---

a Loop 1 2000.0 -

• Loop 2 0¢ 1000.0 -

0.0 0.0 200.0 400.0 6000 Time (s)

Figure 4-9 Steam Generator Main Feedwater Flow Rates - 9.00-inch Break AREVA NP Inc

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 48 Auxiliary Feedwater Flow 1,0 0.5

-- Loop 1

-4 Loop 2 R

ýE 0:

0. 0 1~U

-0 5

-1{* -C i, 0.0 200,0 400.0 600.0 Time (s)

Figure 4-10 Auxiliary Feedwater Flow - 9.00-inch Break AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 49 SG Total Mass 180000.0 175000.0 k j -- a Loop 1

--- Loop 2 170000.0 165000.0 1 0.0 200.0 400.0 600.0 Time (s)

D-1ý011 131721-Figure 4-11 Steam Generator Total Mass - 9.00-inch Break AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 50 SG Level 80.0 60.0 40.0 20.0 0.0 ,

0.0 0.0 Time (s)

Figure 4-12 Steam Generator Level - 9.00-inch Break AREVA NP Inc.

AR A

REV.A.

San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 51 Total HPSI Flow 150.0 100.0 02 D47-8 1-84,01 1317.21 -00 Time (s)

Figure 4-13 Total HPSI Mass Flow - 9.00-inch Break AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 52 Total SIT Flow 6000.0 2000.0 o

600.0 Time (s)

Figure 4-14 Total SIT Flow - 9.00-inch Break AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 53 Integrated Break Flow and Total ECC Flow 600.0 Coos MUTime 5.0)C~ 70 (s)

Figure 4-15 Integrated Break Flow and ECCS Flow - 9.00-inch Break AREVA NP Inc.

A ARE'VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 54 Total Primary Mass 600000.0 600.0 Time (s)

Figure 4-16 Total Primary Mass - 9.00-inch Break AREVA NP Inc.

A

-AR E.V"A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 55 RV Mass 250000.0 1D-7S 1317

-0hQ11 21DEMUX Time (s)

Figure 4-17 Reactor Vessel Mass - 9.00-inch Break AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 56 Hot Assembly Collapsed Level 0

Time (s)

Figure 4-18 Hot Assembly Collapsed Liquid Level - 9.00-inch Break AREVA NP Inc.

A A RE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 57 Hot Rod Mixture Level 15.0 10.0

.5 5.0 0.0 1 I 0.0 200.0 400.0 600.0 18J.-01 131721 -X 047484 Time (s)

Figure 4-19 Hot Rod Mixture Level - 9.00-inch Break AREVA NP Inc.

A A RE VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 58 Downcomer Level 40.0 30.0 20.0 10 0 00.0 200.0 400.0 600.0 Time (s) 1D- ia-1 DEMX 1 317.21 Figure 4-20 Downcomer Collapsed Liquid Level - 9.00-inch Break AREVA NP Inc.

A San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 59 Volume Liquid Temperature in Hot Assembly 700.0 600.0 E 500.0 400.0 300.0 0.0 200.0 400.0 600.0 Time (s)

Figure 4-21 Volume Liquid Temperature in Hot Assembly - 9.00-inch Break AREVA NP Inc.

A AR EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 60 Liquid Void Fraction Distribution in the Hot Assembly

.5 600.0 CMUXTime1&,QO~13j2~

Q4~4 (s)

Figure 4-22 Liquid Void Fraction Distribution in the Hot Channel 0.75 - 7.75 ft Elevation -

9.00-inch Break AREVA NP Inc.

A AREVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 61 Liquid Void Fraction Distribution in the Hot Assembly 0.8 0.6

• 8 0.4 0.2 Time (s)

Figure 4-23 Liquid Void Fraction Distribution in the Hot Channel 8.25 - 12.25 ft Elevation

- 9.00-inch Break AREVA NP Inc.

.A A R E.VA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 62 Heat Transfer Coefficient at PCT Location 4.0 3.0 2.0 1.0 0 .0 i - I '

0.0 200.0 Time (s)

Figure 4-24 Heat Transfer Coefficient at PCT Location - 9.00-inch Break AREVA NP Inc.

A

.A E.VIA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 63 Peak Cladding Temperature 2000.0 1500.0 E2 1000.0 E

500.0 0.0 0.0 Time (s)

Figure 4-25 Peak Cladding Temperature at PCT Location - 9.00-inch Break AREVA NP Inc.

ARV San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 64 Decay Heat Power Fraction 0.100 0.080 C

0 0.040 0.020 0.0 Time (s)

Figure 4-26 Decay Power Fraction AREVA NP Inc.

AA A R EVA San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 65 5.0 Conclusions An SBLOCA analysis was performed for the SONGS Unit 2 and Unit 3 using NRC-approved AREVA NP methods (References [1] and [2]). The results of the analysis demonstrate the adequacy of the ECCS to support the criteria given in 10 CFR 50.46(b) for SONGS Unit 2 and Unit 3 operating with AREVA supplied 16x16 HTP fuel with M5 cladding.

The PCT for the limiting case is 15090 F for a 9.00-inch diameter (0.4418 ft 2) cold leg break. The maximum oxidation thickness and hydrogen generation fall within regulatory requirements.

In addition, a break spectrum analysis of the hot leg and a DEG break in the SIT injection line were performed. The PCTs and maximum oxidation thickness were shown to be bounded by the PCT and the maximum oxidation thickness for the cold leg break.

An evaluation of delayed RCP trip was performed and showed that if all RCPs are tripped no later than 7 minutes once the RCP NPSH criteria are met, the limiting PCT is not challenged and the 10 CFR 50.46(b)(1-4) criteria remain satisfied.

The analysis supports operation at a nominal power level of 3458 MWt (including 20 MWt uncertainty) at the following conditions:

Steam generator tube plugging level of up to 8% in all steam generators; For a peak LHR technical specification limit of 12.8 kW/ft1 and a -0.3 < ASI < +0.3 band; and

• Peak rod average exposures of up to 62 GWd/MTU.

For SBLOCA, the 10CFR50.46 (b) criteria presented in Section 3.0 are met and operation of SONGS Unit 2 and Unit 3 with AREVA NP-supplied 16x16 HTP fuel with M50 clad fuel is justified.

1 This is equivalent to a total peaking factor (FQ) up to a value of 2.37, and a radial peaking factor (Fr) up to a value of 1.755 (including 6% uncertainty and 3.5% control rod insertion uncertainty).

AREVA NP Inc.

ARE San Onofre Nuclear Generating Station Unit 2 and Unit 3 ANP-2974(NP)

Small Break LOCA Report Rev. 0 Page 66 6.0 References

1. AREVA NP Document EMF-2328(P)(A) Revision 0, PWR Small Break LOCA Evaluation Model, S-RELAP5 Based, March 2001.

2. AREVA NP Document BAW-10240(P)(A) Revision 0, Incorporation of M5 Properties in Framatome ANP Approved Methods, May 2004.

3. Topical Report BAW-10227P-A, Revision 1, "Evaluation of Advanced Cladding and Structural Material (M5) in PWR Reactor Fuel," AREVA NP, June 2003.

AREVA NP Inc.