ML19275D438

Redacted - Issuance of Amendment No. 256 for Use of the Tranflow Code for Determining Pressure Drops Across the Steam Generator Secondary Side Internal Components

ML19275D438
Person / Time
Site:	Waterford
Issue date:	10/24/2019
From:	Jennifer Dixon-Herrity Plant Licensing Branch IV
To:	Entergy Operations
Pulvirenti A
References
EPID L-2018-LLA-0112
Download: ML19275D438 (29)

Text

OFFICIAL USE ONLY - PROPRIETARY INFORMATION October 24, 2019 Site Vice President Entergy Operations, Inc.

Waterford Steam Electric Station, Unit 3 17265 River Road Killona, LA 70057-3093

SUBJECT:

WATERFORD STEAM ELECTRIC STATION, UNIT 3 - ISSUANCE OF AMENDMENT NO. 256 FOR USE OF THE TRANFLOW CODE FOR DETERMINING PRESSURE DROPS ACROSS THE STEAM GENERATOR SECONDARY SIDE INTERNAL COMPONENTS (EPID L-2018-LLA-0112)

Dear Sir or Madam:

The U.S. Nuclear Regulatory Commission (the Commission) has issued the enclosed Amendment No. 256 to Renewed Facility Operating License No. NPF-38 for the Waterford Steam Electric Station, Unit 3 (Waterford 3). This amendment consists of changes to the Waterford 3 Updated Final Safety Analysis Report (UFSAR) Section 3.9, Mechanical Systems and Components, to incorporate the TRANFLOW computer code, in response to your application dated April 12, 2018 (Agencywide Documents Access and Management System (ADAMS) Accession No. ML18106A074), as supplemented by letters dated June 13, 2018, January 19, 2019, and July 11, 2019 (ADAMS Accession Nos. ML18169A275, ML19019A025, and ML19192A332, respectively).

Specifically, the amendment deletes Subsection 3.9.1.2.2.1.28 of the Waterford 3 UFSAR, which states that the computer code CEFLASH-4A is used to calculate internal loadings following a postulated main steam line break. The deletion of this subsection would clarify that the pressure drops across the steam generator secondary side due to a steam line break accident are calculated by the TRANFLOW code.

Enclosure 2 to this letter contains Proprietary information. When separated from Enclosure 2, this document is DECONTROLLED.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION The NRC staff has determined that the related safety evaluation contains proprietary information pursuant to Title 10 of the Code of Federal Regulations Section 2.390. The proprietary version of the safety evaluation is provided in Enclosure 2. Accordingly, the NRC staff has also prepared a non-proprietary version of the safety evaluation, which is provided in Enclosure 3.

The Notice of Issuance will be included in the Commissions biweekly Federal Register notice.

Sincerely,

/RA/

April L. Pulvirenti, Project Manager Plant Licensing Branch IV Division of Operating Reactor Licensing Office of Nuclear Reactor Regulation Docket No. 50-382

Enclosures:

1. Amendment No. 256 to NPF-38

2. Safety Evaluation (Proprietary)

3. Safety Evaluation (Non-Proprietary) cc w/o Enclosure 2: Listserv OFFICIAL USE ONLY - PROPRIETARY INFORMATION

ML19275C580 Nonpublic (Proprietary);

ML19275D438-Public (Non-proprietary) *by email dated ** by memo dated OFFICE NRR/DORL/LPL4/PM NRR/DORL/LPL4/LA NRR/DSS/SRXB/BC(A)* NRR/DE/EMIB/BC**

NAME APulvirenti PBlechman JBorromeo SBailey DATE 10/7/2019 10/4/2019 8/22/2019 4/16/2019 OFFICE OGC NLO NRR/DORL/LPL4/BC NRR/DORL/LPL4/PM NAME STurk JDixon-Herrity (BSingal for) APulvirenti DATE 10/14/2019 10/24/2019 10/24/2019 ENTERGY OPERATIONS, INC.

DOCKET NO. 50-382 WATERFORD STEAM ELECTRIC STATION, UNIT 3 AMENDMENT TO RENEWED FACILITY OPERATING LICENSE Amendment No. 256 Renewed License No. NPF-38

1. The Nuclear Regulatory Commission (the Commission) has found that:

A. The application for amendment by Entergy Operations, Inc. (EOI), dated April 12, 2018, as supplemented by letters dated June 13, 2018, January 19, 2019, and July 11, 2019, complies with the standards and requirements of the Atomic Energy Act of 1954, as amended (the Act), and the Commissions rules and regulations set forth in 10 CFR Chapter I; B. The facility will operate in conformity with the application, the provisions of the Act, and the rules and regulations of the Commission; C. There is reasonable assurance (i) that the activities authorized by this amendment can be conducted without endangering the health and safety of the public, and (ii) that such activities will be conducted in compliance with the Commissions regulations; D. The issuance of this amendment will not be inimical to the common defense and security or to the health and safety of the public; and E. The issuance of this amendment is in accordance with 10 CFR Part 51 of the Commissions regulations and all applicable requirements have been satisfied.

Enclosure 1

2. Accordingly, by Amendment No. 256, Renewed Facility Operating License No. NPF-38 is hereby amended to authorize revision to the Waterford Steam Electric Station, Unit 3, Updated Final Safety Analysis Report as set forth in the licensees application dated April 12, 2018, as supplemented by letters dated June 13, 2018, January 19, 2019, and July 11, 2019, and evaluated in the NRC staffs safety evaluation enclosed with this amendment.

3. This license amendment is effective as of its date of issuance and shall be implemented within 30 days from the date of issuance.

FOR THE NUCLEAR REGULATORY COMMISSION

/RA Balwant K. Singal for/

Jennifer Dixon-Herrity, Chief Plant Licensing Branch IV Division of Operating Reactor Licensing Office of Nuclear Reactor Regulation Date of Issuance: October 24, 2019

ENCLOSURE 3 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO AMENDMENT NO. 256 TO RENEWED FACILITY OPERATING LICENSE NO. NPF-38 ENTERGY OPERATIONS, INC.

WATERFORD STEAM ELECTRIC STATION, UNIT 3 DOCKET NO. 50-382 (NON-PROPRIETARY)

Proprietary information pursuant to Section 2.390 of Title 10 of the Code of Federal Regulations has been redacted from this document.

Redacted information is identified by blank space enclosed within double brackets

OFFICIAL USE ONLY - PROPRIETARY INFORMATION SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO AMENDMENT NO. 256 TO RENEWED FACILITY OPERATING LICENSE NO. NPF-38 ENTERGY OPERATIONS, INC.

WATERFORD STEAM ELECTRIC STATION, UNIT 3 DOCKET NO. 50-382

1.0 INTRODUCTION

By application dated April 12, 2018 (Reference 1), as supplemented by letters dated June 13, 2018, January 19, 2019, and July 11, 2019 (References 2, 3, and 4, respectively), Entergy Operations, Inc. (the licensee) submitted a license amendment request (LAR) for use of the TRANFLOW computer code for determining the pressure drops across steam generator (SG) secondary side internal components during a postulated main steam line break (MSLB).

Specifically, the amendment deletes Subsection 3.9.1.2.2.1.28, CEFLASH-4A, of the Waterford Steam Electric Station, Unit 3 (Waterford 3) Updated Final Safety Analysis Report (UFSAR) (Reference 5), which states that the computer code CEFLASH-4A is used to calculate internal loadings following a postulated MSLB. The deletion of this subsection would clarify that the pressure drops across the SG secondary side due to a steam line break accident are calculated by the TRANFLOW code.

This LAR, to incorporate the use of TRANFLOW, is the licensees intended resolution of a Severity Level IV non-cited violation, as documented in Problem Identification and Resolution Inspection Report 05000382/2016008 (Reference 6). The licensee stated in its application that TRANFLOW was used during the evaluation of the Waterford 3 replacement SGs, which were placed in service in 2013. However, the licensees UFSAR incorrectly stated that CEFLASH-4A, rather than TRANFLOW, was used to determine the pressure drops across the replacement SG secondary side internal components.

The supplemental letters dated January 19, 2019, and July 11, 2019, provided additional information that clarified the application, did not expand the scope of the application as originally noticed, and did not change the U.S. Nuclear Regulatory Commission (NRC, the Commission) staffs original proposed no significant hazards consideration determination as published in the Federal Register on September 4, 2018 (83 FR 44919).

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

2.0 REGULATORY EVALUATION

2.1 Background TRANFLO (Reference 7) is a Westinghouse Electric Company, LLC (Westinghouse) thermal-hydraulic code that has been approved by the NRC for use in blowdown calculations for SGs. Specifically, the use approved by the NRC was in conjunction with the MARVEL code, a general-purpose thermal-hydraulic systems analysis code used to analyze the reactor coolant system response to various transients, to determine the quality of the flow at a break in the main steam line. TRANFLO solves the mass, energy, and momentum conservation equations for a system of control volumes and connecting junctions. The approved version of the code uses a homogeneous equilibrium model for fluid conditions.

The TRANFLOW computer code, as stated by the licensee in its LAR, is the workstation version of the TRANFLO code.1 TRANFLOW was used during the evaluation of the Waterford 3 replacement SGs, which were placed in service in 2013. As discussed later in this safety evaluation (SE), the version of TRANFLOW that was used for SG design and evaluation at Waterford 3 included several variations from the approved version of TRANFLO, including the implementation of a drift-flux model.

Waterford 3 was issued a Severity Level IV non-cited violation for incorrectly stating in its UFSAR that pressure drops across the replacement SG secondary side internal components had been determined by the CEFLASH-4A code, rather than by TRANFLOW. The issue was documented in Problem Identification and Resolution Inspection Report 05000382/2016008 and was entered into the licensees corrective action program. The approval of TRANFLOWs use in the licensees UFSAR for this application is the licensees intended resolution of this matter.

The NRC staff reviewed the licensees initial application and determined that further information was necessary to accept the application for review. This determination was communicated to the licensee by letter dated June 1, 2018 (Reference 8). The licensee responded with the requested supplemental information by letter dated June 13, 2018 (Reference 2), and the NRC staff accepted the licensees application by letter dated June 28, 2018 (Reference 9). Two rounds of requests for additional information (RAIs) were sent to the licensee by letters dated November 26, 2018, and June 4, 2019 (References 10 and 11, respectively), and responses were received by letters dated January 19, 2019, and July 11, 2019 (References 3 and 4, respectively).

2.2 Description of Changes In Section 2.4 of the LAR, Description of the Proposed Change, the licensee proposed to incorporate TRANFLOW into the Waterford 3 licensing basis for the purpose of calculating the 1 From this point forward, TRANFLO is used when discussing the version of the code approved by the NRC in the 1980s, TRANFLOW is used when discussing to the version of the code used at Waterford 3, and TRANFLO(W) is used when discussing concepts that are applicable to both codes.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION pressure drops in the secondary side of the SGs during a postulated steam line break event. To make this change, the licensee will delete UFSAR Subsection 3.9.1.2.2.1.28, CEFLASH-4A, which reads:

This program is used to calculate transient conditions resulting from a flow line rupture in a water/steam flow system. The program is used to calculate steam generator internal loadings following a postulated main steam line break.

This program is used in a steam line break accident structural analysis.

The Waterford 3 UFSAR already contains a reference to TRANFLOW for replacement SG pressure drop analysis in Subsection 3.9.1.2.2.1.35, which states:

TRANFLOW is a code that solves the mass, energy and momentum conservation equations for transient thermal-hydraulic phenomena using an implicit backward differencing technique.

The code was used for Waterford 3 replacement steam generator analyses for determining the detailed distribution of fluid temperatures and heat transfer coefficients for SG secondary side stress analyses and the steam line break accident, determining the pressure drops across the SG secondary side internal components.

In its LAR, the licensee proposed to keep this reference to TRANFLOW in the UFSAR for the purpose of determining the pressure drops in the secondary side of the replacement SGs during a postulated steam line break.

2.3 Regulatory Requirements Title 10 of the Code of Federal Regulations (10 CFR) Section 50.34(b), Final safety analysis report, provides requirements for licensees to submit a final safety analysis report that describes the facility, presents the design bases and limits on operation, and presents a safety analysis of the structures, systems, and components and of the facility as a whole. The proposed deletion of the reference to the CEFLASH-4A code and implicit adoption of TRANFLOW into the plant licensing basis relates to the required plant safety analysis.

The requirement for a licensee to provide design control is contained in Criterion III, Design Control, of Appendix B to 10 CFR Part 50, which requires, in part, that The design control measures shall provide for verifying or checking the adequacy of design, such as by the performance of design reviews, by the use of alternate or simplified calculational methods, or by the performance of a suitable testing program.

The requirement for a licensee to provide facility principal design criteria to the NRC is contained in 10 CFR 50.34(a)(3)(i), which states, in part:

Appendix A [to 10 CFR Part 50], General Design Criteria for Nuclear Power Plants, establishes minimum requirements for the principal design criteria for water-cooled nuclear power plants.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION Waterford 3s UFSAR, Section 3.1, Conformance with NRC General Design Criteria, indicates that the facility complies with the Appendix A General Design Criteria (GDC). In particular, GDC 14, Reactor coolant pressure boundary, requires that:

The reactor coolant pressure boundary shall be designed, fabricated, erected, and tested so as to have an extremely low probability of abnormal leakage, of rapidly propagating failure, and of gross rupture.

According to the licensees submittal (and confirmed by the NRC staffs review of the plant UFSAR), the Waterford 3 reactor coolant system components are designed in accordance with the American Society of Mechanical Engineers Boiler and Pressure Vessel Code (ASME Code),

Section III, Division 1.

The use of TRANFLOW to provide input to the ASME Code,Section III, stress analysis represents a departure from an approved method. Therefore, in reviewing the LAR, the NRC staff considered the following regulatory requirements in Appendix A to 10 CFR Part 50:

GDC 1, Quality Standards and records, requires, in part, that Structures, systems and components important to safety to be designed, fabricated, erected, and tested to quality standards commensurate with the importance of the safety functions to be performed.

GDC 4, Environmental and dynamic effects design bases, requires structures, systems and components important to safety to be protected against dynamic effects associated with flow instabilities and loads such as those resulting from water hammer.

GDC 14, Reactor coolant pressure boundary, requires that The reactor coolant boundary shall be designed, fabricated, erected and tested so as to have an extremely low probability of abnormal leakage, of rapidly propagating failure, and of gross rupture.

At Waterford 3, the reactor coolant system components are designed in accordance with ASME Code,Section III, Division 1.

2.4 Review Guidance The licensees request relates to the computer codes used in the analysis of secondary system pipe ruptures. NUREG-0800, Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants: LWR [Light-Water Reactor] Edition (hereafter, referred to as the SRP), Section 6.2.1.4, Mass and Energy Release Analysis for Postulated Secondary System Pipe Ruptures (Reference 12),2 provides guidance for the NRC staff review of analyses of mass and energy release for steam line breaks. In particular, the SRP acceptance criteria provide guidance on the sources of energy that should be considered in a mass and energy release analysis, and the degree of conservatism that should be included in the models and inputs.

2 The licensees submittal identified that data from TRANFLOW are used to evaluate loss-of-coolant accidents (LOCAs), and that guidance for mass and energy release for postulated LOCAs is provided in SRP Section 6.2.1.3, Mass and Energy Release Analysis for Postulated Loss-of-Coolant Accidents (LOCAs). However, the NRC staffs review of the proposed amendment and the documentation associated with TRANFLOW did not identify that TRANFLOW is used in LOCA mass and energy release analysis; rather, it is used for mass and energy release associated with secondary system pipe ruptures (i.e., steam line breaks), for which review guidance is documented in SRP Section 6.2.1.4.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION SRP Section 15.0.2, Review of Transient and Accident Analysis Methods (Reference 13),

provides guidance for the NRC staffs review of methods for performing transient and accident analysis. This guidance is applicable to a wide variety of transient analysis methods, including general-purpose thermal-hydraulic codes such as TRANFLOW. It specifies that the NRC staff should review various aspects of the method, including the documentation, the evaluation model (including the code and its associated inputs), and the code assessment. Subsection III.5 of SRP Section 15.0.2 notes that application of the full review process described in the SRP may not be needed when the new evaluation model is a change to or extension of an existing evaluation model and provides attributes that should be considered when determining the extent to which the review process should be reduced for a specific application.

3.0 TECHNICAL EVALUATION

Based on the guidance provided in SRP Section 15.0.2, the NRC staff focused its review efforts on the aspects of TRANFLOW that had been modified in the time since the codes original NRC approval as TRANFLO in the 1980s. The staffs evaluation is structured roughly as follows:

an overview of the models in the approved version of TRANFLO; a discussion of the models and features of TRANFLO that were updated since its original approval and incorporated into the version of TRANFLOW used at Waterford 3; a generic review (to the extent that one is practical with the documentation provided by the licensee and its contractor, Westinghouse) of the modified version of TRANFLOW for the purpose of determining SG secondary side component pressure drops; a review of the specific TRANFLOW models used at Waterford 3 and their calculational results; and evaluation of TRANFLOW as it applies to ASME Code Section III stress analysis.

3.1 TRANFLOW 3.1.1 Approved Version of TRANFLO As discussed in Section 2.1 of this SE, TRANFLO, which is the predecessor of the TRANFLOW code used in the Waterford 3 analysis, has been previously reviewed and approved for use by the NRC. In particular, TRANFLO was approved to calculate flow quality at the break as part of a methodology to analyze the mass and energy release of a steam line break.

TRANFLO is based on a one-dimensional control volume approach, solving mass, energy, and momentum conservation equations. Homogeneous equilibrium conditions are assumed in each control volume. As in other similar codes, scalar quantities (such as mass, pressure, and energy) are conserved within each control volume, while vector quantities (such as velocities) are conserved across the junctions between volumes. The solution is based on an implicit backward differencing technique, like that used in the FLASH series of codes. The TRANFLO code was originally used in conjunction with the MARVEL code to determine the mass and energy release caused by (and secondary system response to) an MSLB.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION The following sections of this SE discuss the models included in the approved version of TRANFLO as a basis for discussing the modifications made to the code before it was applied as TRANFLOW to the pressure drop calculations at Waterford 3.

3.1.1.1 TRANFLO Field Equations As discussed above, TRANFLO conserves mass and energy within control volumes and momentum on the connectors between volumes. Heat may be added directly to a control volume. Leak paths can also be defined to add energy and mass to control volumes.

3.1.1.2 Constitutive Relations The following pressure forces are considered in each connector:

thermodynamic pressure differential between adjoining volumes; skin friction based on the inertial length of the node; form losses based on user-specified resistance or flow coefficients; expansion losses based on area changes between volumes; and elevation head losses.

Homogeneous equilibrium conditions are assumed in each fluid volume. However, the effects of phase slip on the pressure drop are modeled using empirical correlations. The Armand correlation is applied on a control volume to determine the flowing quality in any downstream connectors; this is then used to determine the effective specific volume applied in the form and skin friction pressure drop equations. The flowing quality is also modified to account for entrainment based on the Davis correlation. Swirl vanes and steam separators are also included as special connectors with proprietary models.

Energy addition can be modeled in any control volume. As discussed in the TRANFLO topical report (Reference 7), heat transfer to a control volume is modeled as a single-pass tube bundle in the control volume with hot pressurized water flowing through it to represent primary system piping. The water inside the tubes can be specified either as constant pressure and flow (with temperature calculated based on heat transfer) or with flow, temperature, and pressure from input tables. Heat transfer between the tube bundle and the control volume is modeled using the following correlations:

Single-phase forced convection to subcooled water: Dittus-Boelter Single-phase forced convection to superheated steam: Heineman Subcooled boiling: Thom Forced convection vaporization: Schrock-Grossman Departure from nucleate boiling: MacBeth Transition boiling: Westinghouse correlation Stable film boiling to a two-phase mixture: Dougall-Rohsenow Stable film boiling to a subcooled liquid: Sandberg Leaks can be defined as constant flow, table input, or critical flow (with liquid water or steam).

Saturated critical flow is calculated using the Moody correlation, while subcooled critical flow is modeled with the Zaloudek correlation. Discharge coefficients are applied as a user input.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION 3.1.1.3 NRC Approval Conditions The NRC staffs SE on TRANFLO, included in WCAP-8859 (Reference 7), states, in part, that the review of the code was confined to the acceptability of the model to provide input for long term containment analysis and did not include considerations for use of the code for other purposes.

3.1.2 Differences between TRANFLO and TRANFLOW as applied at Waterford 3 The licensee noted in its LAR that the TRANFLOW code used to evaluate Waterford 3s replacement SGs was the workstation version of TRANFLO, a computer code originally developed for Westinghouse in the early 1970s to determine thermal-hydraulic conditions in SGs. In its request for supplemental information dated June 1, 2018 (Reference 8), NRC staff asked what other modifications had been made to TRANFLO to arrive at the version of TRANFLOW used at Waterford 3. The licensees response by letter dated June 13, 2018 (Reference 2), detailed the following changes:

A reformulation from a homogeneous equilibrium model to a drift flux model, which required modification of the conservation equations; The ability to handle separate feedwater flow inlets for main feedwater and auxiliary feedwater; Improvements to input and output handling; Additional material properties; and Computing platform changes to HP-UNIX and GNU/Linux.

Of these changes, the NRC staff finds the implementation of the drift flux model to be the most significant. In response to the staffs RAIs, the licensee provided further details on the most important changes made to TRANFLOW, which is discussed in the following sections.

3.1.2.1 Field Equations When compared to the approved version of TRANFLO, there were no significant modifications to the field equations in TRANFLOW to accommodate the drift flux model implementation. The mass conservation equation was unchanged. The enthalpy transport terms in the energy conservation equation were modified to a form that fits the drift flux model, but the remainder was essentially unchanged. The momentum conservation equation was also mostly unchanged, except for the skin friction losses in the volume, which were recast in a form that fits the drift flux model.

3.1.2.2 Drift-Flux Model Implementation The primary difference between the homogeneous model in the original version of the code and the drift flux model in TRANFLOW is the phase slip relationship. The model in the approved version of TRANFLO was based on the Armand correlation, which defines a flowing quality for a connector adjacent to a node based on the homogeneous void fraction in the upstream node, and the properties of the liquid and vapor. The drift flux formulation employed in TRANFLOW OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION defines separate mass flow rates for liquid and vapor phases in terms of the void fraction, fluid density, and a relative velocity term.

Additional detail on the drift flux model was provided in the licensees responses to the first-round RAI Question 1 in the letter dated January 19, 2019 (Reference 3), and second-round RAI Questions 1 and 2 in the letter dated July 11, 2019 (Reference 4). The response to first-round RAI Question 1 referenced report MPR-663, TRANFLO: A Computer Program for Transient Thermal Hydraulic Analysis with Drift Flux (Reference 14), by MPR Associates, who contracted with Westinghouse to develop TRANFLO(W). This document in turn indicated that the source for the (( void fraction and relative velocity )) models used in the drift flux formulation was (( JAERI-M 7490, Preliminary Analysis of Downcomer Effective Water Head During Reflood Phase in PWR LOCA (Reference 15). This report was in Japanese, but the NRC staff found an English-language journal article by one of the authors of JAERI-M 7490 (Reference 16) that contained the version of the correlation used in TRANFLOW. )) The (( void )) correlation will be hereafter referred to as the (( Sudo ))

correlation.

References 15 and 16 indicated that the (( Sudo )) correlation was developed based on air-water experiments in an (( open tank )) with (( low or zero liquid velocities, designed to simulate the effect of voiding on the downcomer head during a LOCA )). In the response to the second-round RAI Question 1, Westinghouse explained that the reasoning behind choosing the (( Sudo )) correlation could not be verified. However, to confirm that the (( Sudo ))

correlation is adequate for the drift flux model in TRANFLOW, Westinghouse also provided comparisons to the well-known and extensively validated (( Lellouche-Zolotar )) correlation.

Figure 1 of the second-round RAI response demonstrates that the (( Sudo )) correlation produces higher relative velocities than the (( Lellouche-Zolotar )) correlation, except at

(( lower void fractions/lower velocities )). To demonstrate that the (( Sudo )) correlation as implemented in TRANFLOW, produces conservative results, Westinghouse applied a multiplier to the void correlation so that the calculated (( relative velocity )) from the (( Sudo )) correlation would exceed that of the (( Lellouche-Zolotar )) correlation across the range of possible (( void fraction values )). Westinghouse then computed the peak component pressure drop using the code with this multiplier applied. The study, provided in Table 2 of the RAI response, demonstrated that the original correlation without the multiplier produces bounding values for the component pressure drop.

Reference 16 states that the correlation chosen for use in TRANFLOW is valid for pressures ranging from (( 6 to 111 atmospheres (approximately 88 to 1631 pounds per square inch (psi)) for steam-water systems )), for superficial gas velocities ranging from (( 0.02 to 2 meters per second (m/s) (approximately 0.07 to 6.56 feet per second (ft/s)) )), and for upward superficial water velocities ranging from (( -0.04 to 0.3 m/s (approximately -0.13 to 0.98 ft/s) )). The NRC staff was concerned that the superficial steam velocities would exceed the upper end of the correlations range. However, Table 1 of Westinghouses response to second-round RAI Question 1 provided results demonstrating that the peak superficial steam velocity in the SG was (( 1.5 m/s (4.95 ft/s) )), well within the correlation range. Additionally, the components provided in Table 1 that experience the highest pressure drops are even further from the maximum allowable superficial steam velocity for the correlation, in a region of the correlation that provides good predictions based on the data reviewed by the NRC staff.

The NRC staff was also concerned about one of the assumptions made in the derivation of the

(( relative velocity term )) in MPR-663. The staff was unable to identify why the (( void fraction )) was (( taken to the 2/3rds power )), as discussed in second-round RAI Question 2.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION In the response to this RAI, Westinghouse calculated the peak pressure loads assuming (( the exponent on the void fraction was 1, rather than 2/3 )). The original pressure load calculation remained bounding.

3.1.2.3 Heat Transfer Correlations As indicated by the licensee in the RAI responses, most of the heat transfer correlations remain the same between the approved version of the TRANFLO code and the TRANFLOW code employed at Waterford 3. One apparent difference is in the correlation used for departure from nucleate boiling: an (( upper mass flux )) limit of (( 250,000 pounds per hour square foot (lb/hr-ft2 )) is placed on the Macbeth correlation, and (( a modified Konkov correlation )) is used for (( mass fluxes greater )) than the limit.

Limited detail was available on (( the modified Konkov correlation )). However, the NRC staff was able to find more detail in the Fortran code listing provided with MPR-663. The two critical heat flux correlations are used to determine the critical quality for a given heat flux, according to the following formulae. (( The Macbeth correlation is provided as:

and the modified Konkov correlation is provided as where For both correlations, Xcrit is the critical quality (dimensionless), Q is the heat flux in British Thermal Unit per hour square foot (BTU/hr-ft2), G is the mass flux in lbm/hr-ft2, P is the pressure in psi, D is the hydraulic diameter in ft, and hfg is the specific heat of vaporization in British Thermal Units/pound mass (BTU/lbm). )) The quality in the node is checked against the critical quality, and if the critical quality is exceeded the heat flux is set to be the greater of the transition boiling or film boiling heat flux.

3.1.2.4 Pressure Drop Correlations The NRC staff reviewed the pressure drop correlations provided in the response to first-round RAI Question 1 and found that they were consistent with those provided in the approved version of TRANFLO.

3.1.2.5 Steam Separators As discussed in the responses to first-round RAI Questions 2 and 8 in the letter dated January 19, 2019 (Reference 3), the implementation of the steam separators in TRANFLOW OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION also changed because of the drift flux model. Before the drift flux model was included, the steam separators would only pass liquid water mass into the steam dome and steam nozzle when the quality in the separators reached zero (i.e., they were full of liquid water). With the drift flux model, the separator volumes and steam dome volumes can all have a two-phase quality. Westinghouse stated in the RAI response that this allows two-phase fluid to reach the break earlier in the transient and is more conservative for the break mass flow calculation.

3.1.2.6 Other Models In the approved version of TRANFLO, heat connectors were used to model the transfer of energy from the primary fluid to the secondary fluid. As discussed in the response to first-round RAI Question 3 in the letter dated January 19, 2019, the version of TRANFLOW that implemented the drift-flux model also enabled the use of metal volumes between heat connectors. Metal volumes conserve energy, and can transfer heat to other metal volumes by conduction or to the fluid using the fluid heat transfer equations discussed above. The RAI response indicated that metal volumes do not contribute significantly to fast transients, but become a significant factor in slower transients or transients that do not initiate from a steady-state condition.

Along with the addition of metal volumes, the NRC staff asked in first-round RAI Question 2 whether the drift flux implementation of TRANFLOW also included condensation models.

Westinghouse responded that applying film condensation heat transfer on the metal volumes in the SG (( (particularly in the steam dome, where condensation would be most likely to occur) produces lower heat transfer coefficients than using the forced convection heat transfer correlations included in the set of models discussed above. Because it is conservative to apply the larger heat transfer coefficients for both the thermal-hydraulic and structural analysis, Westinghouse did not model film condensation in the steam generators. ))

Similarly, the NRC staff questioned in first-round RAI Question 2 by letter dated November 26, 2018 (Reference 10), whether countercurrent flow limits were incorporated into TRANFLOW along with the addition of the drift flux model, which allows countercurrent flow to exist.

Westinghouse responded that explicit countercurrent flow limit models were not included, and that the modeling in TRANFLOW allows two-phase fluid to reach the break earlier and is conservative with respect to the calculation of break mass flow.

3.1.3 Review of TRANFLOW for Steam Generator Component Pressure Drop Calculations The NRC staff considered the changes between TRANFLO and TRANFLOW, as described above, in the context of using TRANFLOW to calculate SG secondary-side component pressure drops during an MSLB. Despite the NRC staffs conclusion from the TRANFLO safety evaluation that the review of the TRANFLO code has been confined to the acceptability of the model to provide input for long term containment analysis and did not include considerations for use of the code for other purposes, as discussed in Section 3.1.1.3 of this SE, there is no inherent limitation on TRANFLO(W) that would prevent it from being used for purposes other than mass and energy release calculations.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION 3.1.3.1 Accident Scenario For the component pressure drop calculations, the analysis places greater emphasis on the instantaneous values of parameters than on the integrated parameters typically considered in mass and energy release calculations. Additionally, the component pressure drop calculations reach their peak values within the first second of the transient, so this presents an area of focus.

However, the key models are expected to be the same between the original application of the code and its use at Waterford 3 for determining component pressure drops. These models are discussed in more detail in the following sections.

3.1.3.2 Field Equations As discussed in Section 3.1.2.1 above, the field equations received minimal modifications except for the incorporation of the drift-flux model. The NRC staff determined that TRANFLOW includes sufficient detail in the mass, energy, and momentum conservation equations to calculate fluid conditions in an SG during blowdown (commensurate with other NRC-approved codes used for the same purpose) and, is thus, acceptable.

3.1.3.3 Constitutive Relations 3.1.3.3.1 Drift Flux Model The primary change to the constitutive relations is the implementation of the drift-flux model.

Based on the evaluation performed by the licensee and discussed in Section 3.1.2.2 of this SE, the NRC staff finds the selection of the (( Sudo )) correlation to be appropriate for the drift-flux model in TRANFLOW. The staff also determined overall that the implementation of the drift flux model in TRANFLOW was done in such a way that it provides conservative predictions of the SG secondary-side component pressure drops. The drift-flux model is expected to provide improved predictive capability relative to the original homogeneous model in the approved version of TRANFLO, particularly during the period immediately following the initiation of the steam line break, because the model is more capable of handling phase separation.

3.1.3.3.2 Heat Transfer Models For the heat transfer models, the only change from TRANFLO to TRANFLOW is the implementation of (( a modified Konkov correlation )) to evaluate (( the critical quality at mass fluxes greater than 250,000 lb/hr-ft2 )). The NRC staff evaluated (( the Macbeth and modified Konkov correlations )) across a range of conditions potentially relevant to SG transients and found that the (( modified Konkov correlation tended to predict similar, but slightly lower critical qualities for a given heat flux than the Macbeth correlation in the range of interest.)) Because the predicted (( critical qualities )) are comparable, the NRC staff finds the use of the (( modified Konkov correlation )) to be acceptable.

Despite the fact that the above new model is the only change, it is also worth considering that in TRANFLOW the (( Dougall-Rohsenow correlation )) is still used to calculate the heat transfer coefficient in (( the film boiling regime )). Since the approval of TRANFLO in the early 1980s, additional research has been conducted on loss-of-coolant accident (LOCA) heat transfer mechanisms, including (( film boiling )). This research indicates that (( Dougall-Rohsenow ))

may over-predict the heat transfer, which is non-conservative in (( LOCA conditions, as discussed in 10 CFR Part 50 Appendix K, Section I.C.5.c )). However, since over-prediction of the heat transfer from the primary system to the secondary system will provide additional heat to OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION the SG, it will likely be conservative for the SG evaluation. Considering also that the (( film boiling regime )) is not active in most SG transients (as discussed in the response to first-round RAI Question 1 (Reference 3)), the NRC staff finds the continued use of (( Dougall-Rohsenow ))

acceptable.

The inclusion of metal volumes in TRANFLOW, as discussed in Section 3.1.2.6 of this SE, should provide improved predictive capability, particularly for longer transients. For fast transients, such as the steam line break, metal volumes are expected to result in at most minimal impacts to the response given that peak component pressure drops occur within the first second of the transient. The NRC staff therefore considers the metal volume modeling to be acceptable.

3.1.3.3.3 Steam Separators The steam separator modeling, discussed in Section 3.1.2.5 of this SE, was included as part of the drift flux model implementation. Given that it replaces (( the artificial assumption that the separators are 100 percent efficient )) in the previously-approved version of TRANFLO and still results in a conservative break mass flow calculation, the NRC staff finds the steam separator modeling to be acceptable.

3.1.3.3.4 Flow Models As discussed in Section 3.1.1.2 of this SE, the approved version of TRANFLO included the Moody critical flow model for two-phase mixtures and the Zaloudek critical flow model for subcooled liquid. These remain unchanged in TRANFLOW, and thus the NRC staff considered whether they would be appropriate for determining the break flow during an SG component pressure drop evaluation. The Moody critical flow model has been used in numerous NRC applications, and is in fact required as a discharge model in 10 CFR Part 50, Appendix K, ECCS Evaluation Models, for LOCA evaluation models. Thus, it is acceptable to the staff in this application. The Zaloudek correlation is a semi-empirical non-equilibrium correlation that has shortcomings compared to newer non-equilibrium critical flow correlations (e.g.,

Ransom-Trapp and Henry-Fauske, which are widely in use in industry). However, the NRC staff does not expect that the use of a Zaloudek correlation during a steam line break will significantly impact the results, since the break flow is expected to be saturated from the initiation of the transient past the point of peak loading on the SG internals, which occurs within the second following the break initiation. Thus, the use of Zaloudek for subcooled critical flow is not expected to affect this particular application and is therefore acceptable.

3.1.3.4 Input, Nodalization, and Boundary Condition Assumptions Because the NRC staff considered TRANFLOW in the context of calculations performed for Waterford 3, the NRC staff did not review (and did not receive from the licensee or Westinghouse) generic plant modeling specifications or assumptions that would be applicable to the use of TRANFLOW for SG secondary side component pressure drop calculations at any pressurized water reactor (PWR) generically. The staff considered these modeling assumptions only in the context of the Waterford 3 analysis, as is discussed later in this SE.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION 3.1.3.5 Validation and Benchmarks 3.1.3.5.1 Validation of TRANFLOW against Test Data In the response to first-round RAI Question 9 by letter dated January 19, 2019 (Reference 3),

Westinghouse provided several plots intended to demonstrate the TRANFLOW code qualification relative to test data. Westinghouses response stated that the code was compared to Frankfurt/Main 7, 12, and 14 tests as well as the Battelle Northwest B53B test, all of which were included in the original TRANFLO topical report for validation. Only Frankfurt/Main 7 and Battelle B35B tests were discussed in detail in the RAI response. Additionally, the original TRANFLO report included several tests from Centro Informazioni Studi Experienze (CISE) that were not reproduced for the TRANFLOW validation. As discussed in the response to RAI Question 9, validation studies were performed with the 1980 version of TRANFLOW but are expected to be applicable to the modern version, given that the only changes since the drift flux model was implemented are new material properties, new steam tables, changes to input/output handling, and computing platform changes.

For the Frankfurt/Main testing, the results with TRANFLOW compare better to measurements overall than the approved version of TRANFLO. Differences in the time trace of pressure, including the peak pressure and depressurization rate during the blowdown, are in the conservative direction. The code continued to have difficulties calculating the pressure as a function of the integrated mass flow, but as discussed in the original TRANFLO topical report this is likely a result of structures in the vessel acting as flow separators that were not described and therefore not able to be appropriately modeled.

For the Battelle B53B test, the code performed nearly identically to the original calculation, and very close to the measurements.

The NRC staff concluded that the validation presented demonstrates that the performance of the TRANFLOW code is as good as or better than the NRC-approved TRANFLO code at predicting SG blowdown transients. Particularly during the initial blowdown phase, TRANFLOW is conservative relative to the experiments presented.

3.1.3.5.2 Benchmarking of TRANFLOW to other Codes As discussed in the application and in response to Sufficiency Item No. 1 (Reference 2) requested in the Request for Supplement dated June 1, 2018 (Reference 8), code-to-code benchmarks of TRANFLOW were performed with respect to the RELAP5, CEFLASH-4A/4B, NOTRUMP, and CATHARE 2 codes.

RELAP5 RELAP5 is a general-purpose thermal-hydraulic systems analysis code originally developed by Idaho National Engineering Laboratory (now Idaho National Laboratory, or INL) for the NRC.

The code uses a two-fluid model and is generally applicable to a wide variety of thermal-hydraulic system transients.

The licensees original application stated, in part, that RELAP5 has been approved for calculating mass and energy release. However, RELAP5 may not strictly apply to Combustion Engineering PWRs such as Waterford 3. Section 6.2.1.3 of the SRP, Mass and Energy Release Analysis for Postulated Loss-of-Coolant Accidents (LOCAs) (Reference 12)

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION specifies that RELAP4 is approved for containment sub-compartment analysis and RELAP5/MOD2-B&W, which specifically refers to Babcock and Wilcox (B&W) nuclear steam supply system designs, is approved for mass and energy release. Nonetheless, RELAP5 is a well-validated thermal-hydraulic code, and it is reasonable to apply RELAP5 to SG depressurization transients in Combustion Engineering PWRs. Comparisons to RELAP5 provide an adequate benchmark of TRANFLOWs predictive capabilities.

Section 3.1 of Appendix A to LTR-SGMP-17-107 (provided as a proprietary attachment to the letter dated June 13, 2018 (Reference 2)), discusses the benchmarking of TRANFLOW to RELAP5. Table 3.1-1, in particular, compares predicted TRANFLOW and RELAP5 tube support plate loading during a steam line break. Though the pressure drop on each tube support plate is comparable, TRANFLOW predicted a (( higher )) pressure drop for the

(( lowest and highest )) tube support plates and a (( lower )) pressure drop for the other tube support plates. Overall, the total pressure drop across the tube support plates is within a few percent between the two codes.

Given that TRANFLOW tended to predict (( lower )) pressure drops for many of the tube support plates than RELAP5, the NRC staff asked for further discussion on the comparison between the two codes in first-round RAI Question 10 (Reference 10). In its response to first-round RAI Question 10 (Reference 3), Westinghouse indicated that the results reported by RELAP5 contained several terms that contribute to the reported pressure loads that should not be included for a comparison to TRANFLOW. Removing these terms would bring the RELAP5 results more in line with TRANFLOW. Additionally, the areas where TRANFLOW tended to predict higher pressure drop tend to be those that are the limiting stress locations for the Waterford 3 replacement SG analysis. Based on the generally good comparison, Westinghouse considers the codes to be technically equivalent.

While the NRC staff does not necessarily agree that TRANFLOW and RELAP5 are technically equivalent, the NRC staff finds that, given the discussion above, TRANFLOW predictions are expected to be more conservative than (or approximately equivalent to) RELAP5 for the portions of the SG of interest in the Waterford 3 SG analysis, and therefore, is acceptable.

CEFLASH-4A/4B CEFLASH-4A is a code for analyzing the reactor coolant system response during the blowdown phase of a LOCA. To specifically analyze blowdown loads, CEFLASH-4A was modified to produce CEFLASH-4B, which was approved by the NRC (Reference 17). The CEFLASH-4B conservation equations assume homogeneous equilibrium conditions (like the original version of TRANFLO). A proprietary Combustion Engineering critical flow model is applied at the break; this model was compared to the Moody and homogeneous equilibrium models and found to result in 10 percent higher blowdown loads. Based on the use of these models, the NRC staff found CEFLASH-4B acceptable for analyzing PWR primary system decompression. As discussed in Section 3.2 of Appendix A to LTR-SGMP-17-107, CEFLASH-4B has also been used for SG analysis, including determination of blowdown loads for steam line break and feedwater line break transients. Because the thermal-hydraulic conditions of a primary system decompression are fundamentally similar to those of a secondary system decompression, the NRC staff finds the comparison of TRANFLOW to CEFLASH-4B to be appropriate.

Section 3.2 of Appendix A to LTR-SGMP-17-107 P discusses the CEFLASH-4B and TRANFLOW calculations. The two codes were compared for their predicted response to a feedwater line break, which is very similar to the steam line break for the initial blowdown OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION portion of the transient. The SG modeled in the letter was a preheat-type SG, which differs in several key characteristics from the feedring-style SG used at Waterford 3. Nonetheless, the NRC staff determined that the comparison provided a good demonstration of the relative predictive capability of the two codes.

Table 3.2-1 in Section 3.2 of Appendix A to LTR-SGMP-17-107 P, shows that the TRANFLOW and CEFLASH-4B calculated maximum pressure drops across segments of components are generally in very good agreement, with TRANFLOW, generally tending to predict higher pressure drops than CEFLASH-4B. The plots of the transient response provided in Figures 3.2-2 through 3.2-5 of Section 3.2 show that the transient differential pressures calculated by TRANFLOW and CEFLASH-4B are in good agreement, especially at the time of the peak differential pressure response within the first few hundredths of a second. Again, TRANFLOW generally tends to predict higher initial peak differential pressure, though pressure oscillations seen in both codes tend to damp out faster in the TRANFLOW calculations than in the CEFLASH-4B calculations.

The NRC staff determined that the benchmarking of TRANFLOW to CEFLASH-4B demonstrates that TRANFLOW provides reasonable to conservative predictions of SG component pressure drops following a steam or feedwater line rupture.

NOTRUMP NOTRUMP is a general-purpose thermal-hydraulics code that has been approved by the NRC staff for small-break LOCA calculations (Reference 18). In its supplement dated June 13, 2018, Westinghouse also stated that the NRC staff has accepted plant-specific applications of NOTRUMP for feedwater line breaks, steam line breaks, and other non-LOCA transients, including anticipated transients without scram. NOTRUMP is similar in concept and execution to TRANFLOW, but includes various improvements, including capabilities to model a mixture level within a volume (with separate mass and energy balances for each region), flooding models, bubble rise models, and horizontal stratified flow models.

TRANFLOW was compared to NOTRUMP for a loss of normal feedwater transient. Because, in comparison to a steam or feedwater line break, this transient is relatively slow, it is not directly applicable to the scenario considered in this SE. The NRC staff therefore did not review the benchmarking to TRANFLOW in detail. Nonetheless, the results, presented in Section 3.5 of Appendix A to LTR-SGMP-17-107 P, demonstrate that TRANFLOW provides similar results to NOTRUMP. The rate at which the SG level decreases during the transient is similar between the two codes, and the pressure increase is comparable (though NOTRUMPs peak pressure is slightly higher).

CATHARE2 CATHARE2 is a general-purpose thermal-hydraulics code used by Frances Institut de Radioprotection et de Sûreté Nucléaire for PWR safety analysis. It uses a two-fluid model (separate conservation equations for the liquid and vapor phases). The code has not been reviewed or approved by the NRC but is regularly used for analysis in France.

TRANFLOW was qualitatively compared to CATHARE2 for dynamic stability analyses of the SGs at the Cruas Nuclear Power Station in southern France. As discussed in Section 3.4 of Appendix A to LTR-SGMP-17-107 P, results from TRANFLOW were qualitatively compared to OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION CATHARE2 results from a paper presented at the 16th International Conference on Nuclear Engineering.

As with the NOTRUMP comparison, the results are not directly applicable to the transient considered in this SE but do provide some demonstration of TRANFLOWs capabilities. The results show reasonably good comparison of the trends and timing of certain transient effects, though the magnitudes of the pressure and level responses are different. Overall, the comparison qualitatively demonstrates that TRANFLOW can predict dynamic stability behavior in an SG.

3.1.3.6 Conclusion Regarding the use of TRANFLOW for Steam Generator Component Pressure Drop Calculations Based on the NRC staffs review of the models that have changed in the TRANFLOW code since TRANFLO was approved by the NRC staff, the adequate validation of the code to several blowdown tests, and the good benchmarking of the code to several NRC-approved and/or generally well-known and well-validated thermal-hydraulic systems codes, the NRC staff finds that TRANFLOW is technically capable of providing a conservative analysis of hydraulic loads during a rapid SG depressurization transient at Waterford 3.

3.2 Blowdown Load Calculations for Waterford 3 TRANFLOW was used at Waterford 3 for two major purposes, as described in the licensees application. First, it was used to determine the distribution of fluid temperatures and heat transfer coefficients for the SG secondary side stress analyses. The methods used to determine the parameters discussed in this first item were not discussed in the licensees Final Safety Analysis Report (FSAR), and therefore, the licensee determined that it was not necessary to seek NRC approval for this particular application. Second, TRANFLOW was used to calculate the maximum dynamic loading caused by a main steam line break on various SG secondary side internal components, including the tube support plates, the wrapper surrounding the tube bundle, the lower and mid deck plates, and the primary and secondary steam separators. It is this second case that was reviewed by the NRC staff in the present application.

3.2.1 Accident Scenario Limited detail on the accident scenario was provided in the initial LAR; however, additional detail was provided in response to first-round RAI Questions 6 and 7 (Reference 3). The licensee analyzed an open-ended guillotine break of the main steam line at the SG outlet nozzle to steam pipe weld, with a break size of (( 2.78 ft2 )). Though this break location places the main steam line flow restrictors downstream of the break, the steam outlet nozzle integral flow limiter (composed of several venturi nozzles) was upstream of the break and was considered in the model. This is consistent with the UFSAR steam line break analysis. A nozzle discharge coefficient of (( 1.0 )) was used.

The SG initial conditions are (( hot-standby zero power conditions )) with the liquid and vapor mass in the SG in equilibrium at saturated conditions equal to 970 pounds per square inch absolute (psia) and 541 degrees Fahrenheit. In its response, Westinghouse stated that this

(( hot-standby )) condition results in the highest secondary pressure and the highest secondary fluid mass. The NRC staff reviewed the transient initial conditions in Chapter 15 of the licensees UFSAR and agrees that (( hot-standby )) conditions maximize the SG pressure and thus, will provide limiting results for the SG depressurization.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION Containment is assumed to be full of saturated steam at (( atmospheric pressure ))

throughout the duration of the transient. The NRC staff considers this to be conservative because it would maximize the pressure differential between the inside and outside of the SG.

As discussed in response to Sufficiency Item No. 2 (Reference 2), (( three )) separate cases were analyzed under the conditions discussed above, each with a different initial water level

(( (the top tube support plate, the lower deck plate, and the nominal operating setpoint).

The tube support plate (TSP) pressure drops )) were evaluated and the limiting initial condition was selected to be the one that produced the most bounding pressure drops. The initial water level was selected to be at the (( top TSP )) based on this evaluation, which found that the results were limiting for all of the TSPs except (( TSP B, which is only non-limiting by 0.02 psi, a negligible amount )). As discussed in the licensees response to first-round RAI Question 6 (Reference 3), the water level considered also bounds the level assumed in the main steam line break analysis. The NRC staff finds the initial water level assumption to be acceptable, because it results in limiting pressure drops for determining the component loading.

The NRC staff reviewed the accident scenario and the initial conditions for the Waterford 3 SG analysis and concluded that they provide a set of bounding conditions for evaluating the SG secondary side pressure drop in response to a main steam line break transient. The staff therefore determined that the accident scenario and initial conditions were acceptable.

3.2.2 Plant Model/Nodalization 3.2.2.1 Nodalization The licensee provided text and diagrams explaining the SG nodalization in response to Sufficiency Item No. 2 (Reference 2). The NRC staff reviewed the nodalization and found that, generally, it was a similar level of detail to prior SG analyses approved by the staff. Important details of the SG structures are captured in the nodalization in a way that their effects will be included appropriately in the analysis.

Enough metal nodes were included in the model to adequately capture the heat transfer between the primary and secondary system and between the SG shell and the secondary fluid.

Steam generator U-tubes were represented by two metal nodes per fluid volume in the tube bundle region, one for each side (hot and cold) of the U-tube. In the upper portions of the SG, there is one metal node representing the SG shell for each major fluid node. In the lower portions of the SG, the shell is broken into a series of thin metal nodes on the surface of the shell directly adjacent to the fluid, with thicker nodes connected by heat connectors to the surface metal nodes. This allows for better representation of the heat transfer in the downcomer region when cold auxiliary feedwater is being injected to the SG.

Because the thermal and hydraulic details of the SG were adequately captured with the nodalization, the NRC staff concluded that it was acceptable.

3.2.2.2 Primary System Modeling The CENTS computer code is Waterford 3s licensing basis code for modeling the nuclear steam supply system response to an MSLB (see UFSAR Section 15.1.3, Limiting Faults). As discussed in the licensees response to first-round RAI Question 5 by letter dated January 19, 2019 (Reference 3), the time-dependent primary side flow, pressure, and hot leg temperatures OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION were taken from the CENTS MSLB transient analysis and used as inputs to TRANFLOW.

Though cold leg temperatures were calculated in the CENTS analysis, they were not used in the TRANFLOW modelthe cold leg temperature was allowed to fluctuate based on the TRANFLOW calculation of heat transfer from the primary to secondary system through the SG U-tubes.

Even if the scenario modeled in CENTS is not identical to the TRANFLOW scenario, the NRC staff expects it to be sufficiently close in the short time period in which the peak pressure occurs that significant effects are not anticipated. Therefore, the NRC staff considers the use of the CENTS MSLB primary system response to be an adequate representation of the primary system for the purposes of determining SG secondary side component blowdown loads, since it faithfully represents the primary system response to an MSLB.

3.2.2.3 Form Loss Coefficients The licensees approach for determining form loss coefficients for input to TRANFLOW was discussed in the response to Sufficiency Item No. 2 in the letter dated June 13, 2019 (Reference 2). Methods from the Idelchik Handbook of Hydraulic Resistance (Reference 19),

for different geometries were used along with design information for the SG components and sub-components. This is a standard industry approach and was correctly implemented based on the example calculation provided. The NRC staff therefore considers the licensees method for determining the form loss coefficients to be acceptable.

3.2.2.4 Conclusion for Plant Model/Nodalization Based on the considerations discussed above, the NRC staff considers the representation of the Waterford 3 SGs presented by the licensee to be adequate for use in determining the SG secondary-side response to an MSLB under the conditions discussed above.

3.2.3 Comparison to Original Steam Generator Calculations Table 1 in the application provided a comparison of the steady-state pressure drop, the original SG design basis pressure drop, and the maximum pressure drop result from the TRANFLOW calculations for a number of components.

As discussed in the application (and confirmed with additional detail shown in Page 4 of LTR-SGMP-17-107 P (Reference 2)), the original SG analysis was completed by multiplying the component steady-state pressure drop by 16, based on an assumption that the velocity in the tube bundle from a steam line break would be four times the 100 percent power velocity. This is an oversimplification of the complex hydrodynamic phenomena at play during a rapid depressurization event and provides very conservative results, particularly in the tube bundle region. Consequently, the TRANFLOW pressure drop results are an order of magnitude lower for the tube support plates than the original SG design basis calculations. Since the original SG design calculations are based on very conservative assumptions with limited physical meaning, the NRC staff found it reasonable to essentially disregard them, particularly in the tube bundle region.

Comparison to the steady-state pressure drops is warranted, however, and for much of the tube bundle region, pressure drops calculated by TRANFLOW are lower than the steady state pressure drop. This is because TRANFLOW calculates flow reversal in the lower portion of the tube bundle region, with a stagnation point occurring somewhere between TSP B and C. Only OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION at the top of the tube bundle region are TRANFLOW-calculated pressure drops greater than or equal to the steady-state pressure drops. The NRC staff finds it reasonable that such a flow reversal would occur due to the design of the SG, which may allow water to reverse and flow outward around the bottom of the tube wrapper and up the downcomer, rather than flow out through the steam separators and dryers. This finding is also consistent with the RELAP5 comparison calculations.

Elsewhere, the TRANFLOW-calculated pressure drops are generally reasonable. Pressure drop across the lower deck plate is comparable to, but slightly less than, the original design calculations. Pressure drop across the steam dryers is much higher than the original design calculations. This is physically reasonable to the NRC staff. Given that the design of the steam dryers provides a tortuous path to collect water droplets out of the steam flow, a value of 16 times the steady-state pressure drop may not have adequately accounted for how the pressure drop increases as a function of the flow velocity. On the other hand, TRANFLOW predicted a much lower pressure drop for the primary separators. Again, this makes physical sense to the NRC staff, since the primary separators use swirl vanes to centrifugally separate droplets from the steam while providing relatively little diversion to the steam flow path.

3.2.4 Conclusion Regarding TRANFLOW Thermal-Hydraulic Calculations Given the considerations discussed in the preceding sections, the NRC staff determined that the TRANFLOW calculations of SG secondary side component pressure drop resulting from an MSLB were acceptable. The licensee used a code that, as discussed in Section 3.1.3 is technically capable of providing conservative results, along with a conservative accident scenario and an acceptable plant model. The results, while not comparable to the very conservative original SG calculations, are reasonable in light of the physics of the transient.

3.3 Stress Analysis The NRC staff reviewed how TRANFLOW results are used in the licensees overall method to perform the ASME Code Section III stress analysis. The Waterford 3 TRANFLOW model is composed of a network of nodes and connectors that represent the secondary side fluid, tube metal heat transfer, SG shell metal heat transfer, tube plate metal heat transfer, and primary coolant. The staff notes that nodalization used in the licensees TRANFLOW model is coarse when compared to the finite element analyses models that are used to analyze metal stresses.

In first-round RAI Question 11 by letter dated November 26, 2019 (Reference 10), NRC staff asked the licensee to describe how TRANFLOW pressure, temperature and heat transfer coefficients are used as input to the complex 3-dimensional finite element analyses stress analyses, with specific focus on the tube sheets, the tube to tube sheet welds, the shell (lower, transition zone, and upper), and the primary separator assembly. In its response to first-round RAI Question 11 in the letter dated January 19, 2019 (Reference 3), the licensee described how a combination of TRANFLOWs output and design inputs was used to achieve conservative results in the overall stress analysis for Normal, Upset, Emergency and Faulted conditions.

Prior to using TRANFLOW, the licensee used manual calculations to determine the internal loads for the ASME Code Section III analysis. The licensee considered this approach to be overly conservative in general for the replacement SGs, and therefore, employed TRANFLOW in its UFSAR, as described in Section 2.1 of this SE. TRANFLOW was benchmarked against the NRC-approved codes CEFLASH-4A, CEFLASH-4B, RELAP5, NOTRUMP, and CATHARE2, and produced similar results. Thus, the licensees verification of TRANFLOW has superseded manual calculations and independent computer code predictions.

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION 3.4 Technical Evaluation Conclusion Based on the considerations discussed above, the NRC staff finds that the TRANFLOW code has been acceptably used to calculate secondary side internal loads in the Waterford 3 replacement SGs during a postulated steam line break. The analytical results presented by the licensee are acceptable to include in the licensees UFSAR pursuant to 10 CFR 50.34, and the TRANFLOW code is acceptable for use in evaluating portions of the Waterford 3 reactor coolant pressure boundary as required by GDC 14. The NRC staff notes that this safety evaluation does not constitute generic NRC approval for the TRANFLOW code.

In addition, the NRC staff reviewed the licensees use of TRANFLOW in the overall evaluation of stresses on the replacement SGs for Normal, Upset, Emergency and Faulted conditions in accordance with ASME Code,Section III. The staff concludes that the licensees use of TRANFLOW will ensure that the structural integrity is maintained, and that 10 CFR Part 50, Appendix A, GDC 1, 4, and 14, and 10 CFR Part 50, Appendix B are met. Therefore, the staff finds the proposed change acceptable.

4.0 STATE CONSULTATION

In accordance with the Commissions regulations, the Louisiana State official was notified of the proposed issuance of the amendment on October 8, 2019. The Louisiana State official had no comments.

5.0 ENVIRONMENTAL CONSIDERATION

The amendment changes a requirement with respect to installation or use of a facility component located within the restricted area as defined in 10 CFR Part 20. The NRC staff has determined that the amendment involves no significant increase in the amounts, and no significant change in the types, of any effluents that may be released offsite, and that there is no significant increase in individual or cumulative occupational radiation exposure. The Commission has previously issued a proposed finding that the amendment involves no significant hazards consideration, published in the Federal Register on September 4, 2018 (83 FR 44919), and there has been no public comment on such finding. Accordingly, the amendment meets the eligibility criteria for categorical exclusion set forth in 10 CFR 51.22(c)(9).

Pursuant to 10 CFR 51.22(b) no environmental impact statement or environmental assessment need be prepared in connection with the issuance of the amendment.

6.0 CONCLUSION

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

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

7.0 REFERENCES

1. Dinelli, J. C., Entergy Operations, Inc., letter to U.S. Nuclear Regulatory Commission, License Amendment Request for Use of the TRANFLOW Code for Determining the Pressure Drops Across the Steam Generator Secondary Side Internal Components, Waterford Steam Electric Station, Unit 3 (Waterford 3) Docket No. 50-382, License No. NPF-38, dated April 12, 2018 (Agencywide Documents Access and Management System (ADAMS) Accession No. ML18106A074).

2. Dinelli, J. C., Entergy Operations, Inc., letter to U.S. Nuclear Regulatory Commission, Supplemental Information Supporting the License Amendment Request Regarding Use of the TRANFLOW Code for Determining the Pressure Drops Across the Steam Generator Secondary Side Internal Components, Waterford Steam Electric Station, Unit 3 (Waterford 3) Docket No. 50-382, License No. NPF-38, dated June 13, 2018 (ADAMS Accession No. ML18169A275).

3. Dinelli, J. C., Entergy Operations, Inc., letter to U.S. Nuclear Regulatory Commission, Response to U.S. Nuclear Regulatory Commission Request Regarding Use of the TRANFLOW Code for Determining the Pressure Drops Across the Steam Generator Secondary Side Internal Components, Waterford Steam Electric Station, Unit 3 (Waterford 3) Docket No. 50-382, Renewed Facility Operating License No. NPF-38, dated January 19, 2019 (ADAMS Accession No. ML19019A025).

4. Dinelli, J. C., Entergy Operations, Inc., letter to U.S. Nuclear Regulatory Commission, Response to U.S. Nuclear Regulatory Commission Second Round Request for Additional Information Regarding License Amendment Request for Use of the TRANFLOW Code for Determining the Pressure Drops Across the Steam Generator Secondary Side Internal Components, Waterford Steam Electric Station, Unit 3 (Waterford 3) Docket No. 50-382, Renewed Facility Operating License No. NPF-38, dated July 11, 2019 (ADAMS Accession No. ML19192A332).

5. Waterford Steam Electric Station, Unit 3, Revision 309 to Final Safety Analysis Report, Chapter 3, Design of Structures, Components Equipment and Systems, Section 3.9, Mechanical Systems and Components (ADAMS Accession No. ML16256A188).

6. Hipschman, T. R., U.S Nuclear Regulatory Commission, letter to Mr. Michael R. Chisum, Entergy Operations, Inc., Waterford Steam Electric Station, Unit 3 - Nuclear Regulatory Commission Problem Identification and Resolution Inspection Report 05000382/20160008, dated January 26, 2017 (ADAMS Accession No. ML17026A338).

7. Westinghouse Electric Company LLC, TRANFLO Steam Generator Code Description, WCAP-8859-A, dated June 2001 (ADAMS Accession No. ML011970414).

8. Pulvirenti, A. L., U.S. Nuclear Regulatory Commission, letter to Site Vice President, Entergy Operation, Inc., Waterford Steam Electric Station, Unit 3 - Supplemental Information Needed for Acceptance of Requested Licensing Action Re: Use of TRANFLOW Code for Determining Pressure Drops Across Steam Generator Secondary Side Internal Components (EPID L-2018-LLA-0112), dated June 1, 2018 (ADAMS Accession No. ML18145A265).

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

9. Pulvirenti, A. L., U.S. Nuclear Regulatory Commission, letter to Site Vice President, Entergy Operation, Inc., Waterford Steam Electric Station, Unit 3 - Acceptance of Requested Licensing Action Re: Use of TRANFLOW Code for Determining Pressure Drops Across Steam Generator Secondary Side Internal Components (EPID L 2018-LLA-0112), dated June 28, 2018 (ADAMS Accession No. ML18178A293).

10. Pulvirenti, A. L., U.S. Nuclear Regulatory Commission, letter to Site Vice President, Entergy Operation, Inc., Waterford Steam Electric Station, Unit 3 - Request for Additional Information Regarding License Amendment Request for Use of the TRANFLOW Code for Determining Pressure Drops Across Steam Generator Secondary Side Internal Components (EPID L-2018-LLA-0112), dated November 26, 2018 (ADAMS Accession No. ML18320A090).

11. Pulvirenti, A. L., U.S. Nuclear Regulatory Commission, letter to Site Vice President, Entergy Operation, Inc., Waterford Steam Electric Station, Unit 3 - Second Round Request for Additional Information Regarding License Amendment Request for Use of the TRANFLOW Code for Determining Pressure Drops Across Steam Generator Secondary Side Internal Components (EPID L-2018-LLA-0112), dated June 4, 2019 (ADAMS Accession No. ML19151A610).

12. U.S. Nuclear Regulatory Commission, Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants: LWR Edition, NUREG-0800, Section 6.2.1.3, Mass and Energy Release Analysis for Postulated Loss-of-Coolant Accidents (LOCAs), Revision 3, dated March 2007 (ADAMS Accession No. ML053560191) and Section 6.2.1.4, Mass and Energy Release Analysis for Postulated Secondary System Pipe Ruptures, Revision 2, dated March 2007 (ADAMS Accession No. ML070620010).

13. U.S. Nuclear Regulatory Commission, NUREG-0800, Section 15.0.2, Review of Transient and Accident Analysis Methods, dated March 2007 (ADAMS Accession No. ML070820123).

14. MPR Associates, TRANFLO: A computer Program for Transient thermal Hydraulic Analysis with Drift Flux, MPR-663, dated November 1980.

15. (( Y. Sudo and Y. Murao, Preliminary Analysis of Downcomer Effective Water Head during Reflood Phase in PWR LOCA, Japan Atomic Energy Research Institute, JAERI-M 7490, 1978. ))

16. (( Y. Sudo, Estimation of Average Void Fraction in Vertical Two-Phase Flow Channel under Low Liquid Velocity, Journal of Nuclear Science and Technology, vol. 1, no. 17, pp. 1015, 1980. ))

17. R. L. Baer, U.S. Nuclear Regulatory Commission, letter to A.E. Scherer, Combustion Engineering, Staff Evaluation of Topical Report CENPD-252-P, dated February 12, 1979.

18. Wiesemann, R. A., Westinghouse Electric Company, letter to Dr. Cecil O. Thomas, U.S. Nuclear Regulatory Commission, NOTRUMP Code and Small Break ECCS Evaluation Model, dated August 19, 1985 (ADAMS Accession No. ML100060364; not publicly available, proprietary information).

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

OFFICIAL USE ONLY - PROPRIETARY INFORMATION

19. I.E. Idelchik, Handbook of Hydraulic Resistance, Coefficients of Local Resistance and of Friction, dated 1960 (ADAMS Accession No. ML12209A041).

Principal Contributors: Reed Anzalone, NRR Michael Breach, NRR Date: October 24, 2019 OFFICIAL USE ONLY - PROPRIETARY INFORMATION