ML19039A092

Enclosure 1 Non-LOCA Audit Summary Public Version

ML19039A092
Person / Time
Site:	PROJ0769
Issue date:	02/25/2019
From:	Rani Franovich NRC/NRO/DLSE/LB1
To:	Samson Lee NRC/NRO/DLSE/LB1
Franovich R L/nro/7334
Shared Package
ML19039A090	List: ML19039A092 ML19039A093
References
TR-0516-49416-P
Download: ML19039A092 (22)

Text

AUDIT

SUMMARY

FOR THE REGULATORY AUDIT OF NUSCALE TOPICAL REPORT TR-0516-49416-P, NON-LOSS-OF-COOLANT ACCIDENT ANALYSIS METHODOLOGY 1

BACKGROUND By letter dated January 10, 2017 (Reference 1), NuScale Power, LLC (NuScale, or the applicant) submitted topical report (TR) TR-0516-49416-P, Non-Loss-of-Coolant-Accident Analysis Methodology, Revision 0, to the U.S. Nuclear Regulatory Commission (NRC) for review and approval. The applicant submitted the TR in support of the design certification application for the NuScale Power Small Modular Reactor, which the NRC accepted for review on March 23, 2017 (Reference 2). On April 27, 2017, the NRC issued a letter accepting TR-0516-49416-P, Revision 0, for review (Reference 3). Revision 0 of TR-0516-49416-P was superseded by the August 10, 2017, submittal of TR-0516-49416-P, Revision 1 (Reference 4),

which incorporated editorial changes but no changes in content.

TR-0516-49416-P seeks approval for the application of the proposed evaluation model (EM) for the analysis of system transient response to non-loss-of-coolant accident (non-LOCA) initiating events for the NuScale Power Module (NPM). The staff initiated a regulatory audit related to the review of the TR-0516-49416-P on May 31, 2017 (Reference 5). The audit was conducted according to NRC Office Instruction NRO-REG-108, Regulatory Audits (Reference 6). The audit was performed primarily via the NuScale electronic reading room (ERR) with limited in-person audits at NuScales Rockville, MD, office and several telephone audit discussions with the applicant.

The purposes of this audit were to: (1) gain a better understanding of the applicants proposed methodology for non-LOCA analysis; (2) better determine whether the methodology meets NRC regulations and conforms to regulatory guidance; (3) develop requests for additional information (RAI) in areas not adequately covered in the topical report documentation; and (4) identify supplemental information that should be added to the topical report to allow the staff to make its safety finding. Additional background is available in the audit plan associated with this audit summary (Reference 4).

2 REGULATORY AUDIT BASIS 10 CFR 52.47(a)(4) states that a final safety analysis report (FSAR) submitted as part of a standard design certification must include:

[a]n analysis and evaluation of the design and performance of structures, systems, and components with the objective of assessing the risk to public health and safety resulting from operation of the facility and including determination of the margins of safety during normal operations and transient conditions anticipated during the life of the facility, and the adequacy of structures, systems, and components provided for the prevention of accidents and the mitigation of the consequences of accidents.

Based on its review of TR-0516-49416-P, the staff determined the need for an audit to better understand the non-LOCA analysis methodology. The non-LOCA analysis methodology is incorporated by reference into Chapter 15 of the NuScale FSAR and is used to develop safety conclusions.

The following general design criteria (GDC) from Title 10 of the Code of Federal Regulations (10 CFR), Part 50, Appendix A are applicable to the non-LOCA analysis:

GDC 5, Sharing of structures, systems, and components GDC 10, Reactor design GDC 13, Instrumentation and control GDC 15, Reactor coolant system design GDC 17, Electric power systems GDC 20, Protection system functions GDC 25, Protection system requirements for reactivity control malfunctions GDC 26, Reactivity control system redundancy and capability GDC 27, Combined reactivity control systems capability GDC 28, Reactivity limits GDC 31, Fracture prevention of reactor coolant pressure boundary GDC 34, Residual heat removal In addition, relevant regulatory guidance includes:

Regulatory Guide 1.203, Transient and Accident Analysis Methods, December 2005.

NuScale Design-Specific Review Standard (DSRS) 15.0, Introduction - Transient and Accident Analyses, Revision 0, June 2016.

Standard Review Plan (SRP) 15.0.2, Review of Transient and Accident Analysis Method, Revision 0, March 2007.

DSRS 15.1.1-15.1.4, Decrease in Feedwater Temperature, Increase in Feedwater Flow, Increase in Steam Flow, and Inadvertent Opening of the Turbine Bypass System or Inadvertent Operation of the Decay Heat Removal System, Revision 0, June 2016.

DSRS 15.1.5, Steam System Piping Failures Inside and Outside of Containment, Revision 0, June 2016.

DSRS 15.1.6, Loss of Containment Vacuum, Revision 0, June 2016.

DSRS 15.2.1-15.2.5, Loss of External Load; Turbine Trip; Loss of Condenser Vacuum; Closure of Main Steam Isolation Valve; and Steam Pressure Regulator Failure (Closed),

Revision 0, June 2016.

DSRS 15.2.6, Loss of Nonemergency AC Power to the Station Auxiliaries, Revision 0, June 2016.

DSRS 15.2.7, Loss of Normal Feedwater Flow, Revision 0, June 2016.

DSRS 15.2.8, Feedwater System Pipe Break Inside and Outside Containment, Revision 0, June 2016.

SRP 15.4.1, Uncontrolled Control Rod Assembly Withdrawal from a Subcritical or Low Power Startup Condition, Revision 3, March 2007.

SRP 15.4.2 Revision 3, Uncontrolled Control Rod Assembly Withdrawal at Power, Revision 3, March 2007.

SRP 15.4.3, Control Rod Misoperation (System Malfunction or Operator Error),

Revision 3, March 2007.

SRP 15.4.6, Inadvertent Decrease in Boron Concentration in the Reactor Coolant System (PWR), Revision 2, March 2007.

SRP 15.4.7, Inadvertent Loading and Operation of a Fuel Assembly in an Improper Position, Revision 2, March 2007.

DSRS 15.5.1 - 15.5.2, Chemical and Volume Control System Malfunction that Increases Reactor Coolant Inventory, Revision 0, June 2016.

3 AUDIT LOCATION AND DATES The audit was conducted from the NRC headquarters via NuScales electronic reading room and telephone and at NuScales Rockville, MD, office.

Dates:

Phase 1: May 31, 2017, through August 29, 2017 Phase 2: August 30, 2017, through January 25, 2017 Locations:

NRC Headquarters Two White Flint North 11545 Rockville Pike Rockville, MD 20852-2738 NuScale 11333 Woodglen Dr.

Suite 205 Rockville, Maryland 20852

4 AUDIT TEAM MEMBERS Alexandra Burja, NRO, Audit Team Lead Ray Skarda, NRO Jason Thompson, NRO Jeff Schmidt, NRO, SRSB NuScale Team Lead Rebecca Karas, NRO, SRSB Branch Chief Matt Thomas, NRO Andrew Proffitt, NRO Andrew Bielen, RES Ronald Harrington, RES Peter Lien, RES Shawn Marshall, RES Peter Yarsky, RES Mohsen Khatib-Rahbar, Energy Research Incorporated (ERI)

Alfred Krall, ERI Morgan Libby, ERI Walter Tauche, ERI subcontractor Zhe Yuan, ERI Michael Zavisca, ERI 5

APPLICANT PARTICIPANTS Phil Boylan Adam Brigantic Ben Bristol Patrick Byfield Morris Byram Cristhian Galvez-Velit Darrell Gardner Shen Hengliang Nathan Hottle Paul Infanger Peter Jensen Jeff Magedanz Meghan McCloskey Adam Brigantic Lance Larsen Jeff Luitjens John Marking Ken Rooks Pravin Sawant Dean Throckmorton Brian Wolf Chong Zhou 6

AUDIT DOCUMENTS The staff audited the following documents that were part of the original audit plan:

TI-0415-13067, Revision 0, SIET Helical Coil Steam Generator Test Program: Fluid Heated Test Facility Test Section Installation and As-Built Report, October 20, 2015 EC-T050-3638, Revision 0, Assessment of NRELAP5 Using SIET Fluid Heated Test Facility (TF-2) Data, November 6, 2015 EC-0000-3155, Rev. 0, Assessment of NRELAP5 with KAIST Condensation Experiments, October 27, 2015 ER-0000-3171, Rev. 0, Qualification of KAIST High Pressure Condensation Data, April 15, 2015 EE-T080-13757, Rev. 2, NuScale Integral System Test (NIST-1) Facility Scaling Analysis, July 8, 2015 EC-0000-3853, Rev. 1, Calculations to Support NIST-1 Distortion Analysis and Modeling of Containment and Pool Heat Transfer

SDR-0615-15509, Rev. 4, OSU NIST-1 Facility Description Report, February 27, 2017 ER-A010-4066, Rev. 1, HP-03 DHRS Full Length Characterization Final Test Report and Certificate of Conformance ER-A010-4063, Rev. 1, HP-04 Cooling Pool Characterization Final Test Report EC-T080-4703, Rev. 0, NLT-02 and NLT-02b Loss of Feedwater Flow w/DHRS Final Test Report and Certificate of Conformance ED-A011-2689, Rev. 0, Steam Generator, May 21, 2015 ED-B030-2770, Rev. 0, DHRS Passive Condenser, June 15, 2015 ED-A023-1958, Rev. 2, Reactor Vessel Internals, May 11, 2016 ED-A023-2304, Rev. 1, Reactor Vessel Internals - Lower Riser, April 22, 2016 ED-A023-2303, Rev. 1, Reactor Vessel Internals - Upper Riser, April 21, 2016 ED-A014-3905, Rev. 0, Steam Generator System Piping Layout, April 29, 2016 ER-0000-2934, Rev. 0, Non-Loss-of-Coolant Accident Phenomena Identification and Ranking Table for NuScale Power Plant, May 12, 2015 SwUM-0304-15495, Rev. 3, NRELAP5 Version 1.3 Input Data Requirements, November 30, 2016 SwUM-0304-17111, Rev. 3, NRELAP5 Version 1.3 Developers Manual, November 30, 2016 Excerpt from EC-0000-2714, Rev. 0, "Steam System Piping Failure Analysis, November 30, 2016 Pressurizer Modeling: Using Different Thermodynamic Models and Comparing Results with RELAP Code Results, Applied Mechanics and Materials (Volumes 423-426), pp.

1444-1448, 2013 "Heat transfer in tube coils with laminar and turbulent flow," by Seban, R.A., and E.F.

McLaughlin, International Journal of Heat and Mass Transfer, vol. 6, pp. 387-395, 1963 Pressure drop, heat transfer and performance of a helically coiled tubular exchanger, by Prasad, B.V.S.S.S, D.H. Das, and A.K. Prabhakar, Heat Recovery Systems and CHP, 9: pp. 249-256, 1989 Turbulent film condensation of high pressure steam in a vertical tube of passive secondary condensation system, by Kim, Sang Jae (Korea Advanced Institute of Science and Technology PhD thesis), 2000 Evaluation of Helical Geometry Effects for SG Heat Transfer and Pressure Drop, by H.

Noyes, ER-A014-2268, Revision 1, October 4, 2016 In addition, as the staff identified additional information needs throughout the audit, the applicant referred the staff to information in the following documents that were not part of the original audit plan:

EC-6649-R1.i.pdf, "SIET Heat Coil Simulation Using NRELAP5 Post-Test Model"

[NRELAP5 SIET input deck for the TF-1 test]

SIET-TF2_base.i.pdf, "SIET TF2 Base R5 Model" [NRELAP5 SIET input deck for the TF-2 test]

EC-A014-6649, Rev. 1, "NRELAP5 Model for the SIET Heated Tube test Facility" EC-T050-3234, Rev. 0, "NRELAP5 Model for the SIET Fluid Heated Test Facility," June 30, 2015 EC-T050-3498, Rev. 1, SIET Heated Tube Test Facility (TF-1) Adiabatic and Diabatic NRELAP5 Assessment, June 24, 2016 Piping and instrumentation diagrams (P&IDs) SDR-0516-49255 through -49263 and -

49265, all Revision 1, June 7, 2016 [P&IDs for NIST-1 facility]

SDR-0815-16916, Data Acquisition and Control System Configuration Report, Rev. 4, February 16, 2017 Multiple SIET Facility Drawings:

o 078.00.02, Revision 6, March 25, 2015 o 078.00.23, Revision 5, March 20, 2015 o 078.02.09, Rev. 7, February 19, 2015 o 078.02.10, Rev. 6, March 20, 2015 o 078.02.13, Rev. 1, February 6, 2014 o 078.02.14, Rev. 4, March 13, 2015 o 078.02.15, Rev. 3, February 13, 2015 o 078.02.16, Rev. 2, March 16, 2015 o 078.02.17, Rev. 2, March 25, 2015 o 078.02.18, Rev. 7, January 7, 2015 o 078.02.19, Rev. 4, March 20, 2015 o 078.03.00, Rev. 2, March 17, 2015 o 078.03.01, Rev. 3, March 16, 2015 NuScale-generated document Responses to NRC Request 8-3-17 EC-0000-2908, Rev. 1, Loss of Non-Emergency AC Power to the Station Auxiliaries Analysis, December 15, 2016 EC-0000-2910, Rev. 1, Uncontrolled Control Rod Assembly Withdrawal From A Subcritical or Low Power Startup Condition, December 20, 2016 EC-0000-4617, Rev. 0, Comparison of MCHFR predicted by NR5 and VIPRE Codes for Non-LOCA Events, October 20, 2016 EC-B030-2121, Revision 0, DHRS Thermal Hydraulic Analysis EC-A030-2359, Rev. 4, RCS Flow Form Loss Calculation, June 7, 2017 EC-A030-2713, Rev. 2, Primary and Secondary Steady State Parameters, September 7, 2016 ER-0000-3221, Rev. 0, Identification and Classification of Deterministic Design Basis Events for the NuScale Power Small Module Reactor, April 15, 2016 ER-0000-4738, Rev. 1, NRELAP5 Version 1.3 Assessment Report, February 20, 2017 7

DESCRIPTION OF AUDIT ACTIVITIES AND

SUMMARY

OF OBSERVATIONS The staff audited information related to multiple aspects of the NuScale non-LOCA analysis methodology. Many of the audited documents helped to inform the staffs confirmatory models and were associated with the facility layouts, descriptions and models of the test facilities, and the NPM. The remainder of the staffs audit focused on the ability of the non-LOCA EM to predict key phenomena, especially phenomena unique to the NPM. In particular, the staff examined the non-LOCA models; the phenomena the applicant identified and the corresponding importance rankings; the applicants validation of the non-LOCA EM against test data; the non-LOCA EM analysis process; and the applicants example transient calculations. The staff also obtained clarification of certain statements in the non-LOCA TR and examined other documents referenced in the non-LOCA TR.

Furthermore, the staff held multiple audit discussions with the applicant via telephone. A detailed summary of these discussions is provided in Appendix A.

7.1 Facility Layouts, Descriptions, and Models

The staff audited documents describing the NPM and the test facilities used to validate the non-LOCA EM capability. These included several drawings and piping and instrumentation diagrams of the NIST-1 facility, the SIET facility, and the NPM, which the staff used to inform its confirmatory models. For example, the staff reviewed the applicants documentation to understand the specific geometries and instrumentation locations in the facilities to build confirmatory models reflecting the actual facility layout. The staff also used the facility descriptions to inform other modeling choices, such as materials used, and discussed some of these descriptions with the applicant to ensure a thorough understanding. In addition, the staff audited the applicant's input decks and related calculation notes for the SIET model for additional understanding of the model.

The staff noted several potential concerns involving the facilities and models, and discussion of these concerns with the applicant and/or auditing related documentation led to the need for additional information. Therefore, the staff issued the following RAIs:

RAI 9351, Question 15.00.02-33, to obtain information regarding the NRELAP5 model of the SIET facility; particularly, it appeared that a (( ))

RAI 9374, Question 15.00.02-24, to obtain justification that the potential for film boiling and dryout on DHRS tubes, which would reduce DHRS performance, does not exist or is negligible RAI 9466, Question 15.00.02-7, to obtain additional information about the decay heat model used in non-LOCA transient analyses and justification of its conservatism RAI 9466, Question 15.00.02-10, to obtain justification for the applicability and adequacy of the heat transfer correlation used for the DHRS tubes RAI 9466, Question 15.00.02-11, to obtain clarification of, and justification for the applicability and adequacy of, the heat transfer correlation used to calculate the heat transfer from the DHRS to the cooling pool RAI 9466, Question 15.00.02-12, to obtain information about the nodalization used for the test facility models and the NPM model A summary of audit discussions related to this topic is provided in Appendix A.1.

7.2 Non-LOCA Phenomena TR-0516-49416-P discusses the highly-ranked phenomena in the non-LOCA phenomena identification and ranking table (PIRT) but does not list or discuss phenomena of moderate or low ranking. The staff audited document ER-0000-2934, Revision 0, Non-Loss-of-Coolant Accident Phenomena Identification and Ranking Table [PIRT] for NuScale Power Plant, to understand the basis for the highly-ranked phenomena presented in TR-0516-49416-P, including rationale for the rankings of all phenomena. In general, the staff determined that document ER-0000-2934 supported the docketed information. However, the staff and applicant discussed additional details in the document including the rankings of certain phenomena, any changes in the NPM design since completion of the PIRT and how those changes are reflected in the PIRT, and recommendations made in ER-0000-2934 for addressing highly-ranked phenomena. These discussions are summarized in Appendix A.2. The staff issued RAI 9351, Question 15.00.02-31, in part, to address the recommendations in ER-0000-2934, regarding highly ranked phenomena with large uncertainties.

7.3 Non-LOCA EM Validation and Assessments

The staff audited several documents related to separate and integral effects tests and assessments of the non-LOCA EM capability against those tests. The documents covered the NIST-1, SIET, and KAIST facilities.

7.3.1 NIST-1 The staff audited several NIST-1 facility test reports related to the separate effects tests HP-03 and HP-04, which focused on the full-length DHRS and cooling pool, and the integral effects tests NLT-02a and NLT-02b. The staff sought additional background on the behaviors observed in some of the NIST tests, including rationale for discrepancies between NRELAP5 predictions and the data. Although the test assessments provided an expanded discussion, the staff determined that additional information was needed on the docket to justify the capability of NRELAP5 to accurately predict key phenomena. In addition, the staff noted that it appeared that the tuning of some parameters in the NRELAP5 models to match the tests results was not fully supported by the audited information. As a result, the staff issued the following RAIs:

RAI 9158, Question 15.00.02-2, to obtain an explanation for the use of different (( ))

in the NRELAP5 simulation of the NLT-02a test than those used in the test RAI 9374, Question 15.00.02-23, to obtain justification that the acceptability of the predicted total heat removal and heat transfer for the NIST tests is not due to compensating errors in the code models and correlations RAI 9466, Question 15.00.02-9, to obtain information about assessments when external adjustments are made to improve simulation results.

A summary of audit discussions related to NIST-1 is provided in Appendix A.3.1.

7.3.2 SIET The staff sought a better understanding of the validation of the helical coil steam generator model, which the applicant based primarily on data from the SIET facility. The staff audited document EC-T050-3638, Revision 0, Assessment of NRELAP5 Using SIET Fluid Heated Test Facility (TF-2) Data, and made several observations related to assumptions and modeling options used in the NRELAP5 model of the SIET facility as well as the assessment results and conclusions. The staff discussed these observations further with the applicant, as listed in Appendix A.3.2 of this audit report. The staff issued RAI 9158, Question 15.00.02-5, to obtain an explanation of how the recommendations discussed in EC-T050-3638 have resulted in any modeling changes, or if no changes have resulted, why the existing model is adequate.

7.3.3 KAIST The staff audited two documents related to the KAIST facility, as listed in Section 6 of this audit summary, to gain additional understanding of the brief discussion of the KAIST assessments in the non-LOCA TR. While the documents supported the docketed material, the staff noted that additional comparisons of NRELAP5 predictions to KAIST data would better demonstrate the qualification of the non-LOCA EM to predict DHRS phenomena. Therefore, the staff issued RAI 9513, Question 15.00.02-15, requesting the applicant to provide those comparisons.

A summary of audit discussions related to KAIST is provided in Appendix A.3.3.

7.4 Non-LOCA Analysis Process and Example Calculations Based on the information in TR-0516-49416-P, it was unclear to the staff how an analyst would specify an appropriate set of initial conditions and how the analyst would determine that a unique steady-state solution was obtained for an analysis. After discussing these topics with the applicant, the staff issued RAI 9374, Questions 15.00.02-25 and 15.00.02-26, requesting the applicant to update its documentation or provide new documentation that would make the topics unambiguous to an analyst.

The staff also discussed with the applicant the behavior observed in several example calculations in TR-0516-49416-P. The applicants clarifications were largely adequate, but the applicant stated that an error had been discovered in the loss of condenser vacuum analysis.

Therefore, the staff issued RAI 9374, Question 15.00.02-30, to obtain the results of the revised analysis.

A summary of audit discussions related to this topic is provided in Appendix A.4.

7.5 Other Supporting Documents The staff requested the applicant to make available for audit the references listed in the non-LOCA TR. The staff audited the input data requirements and developers manual for NRELAP5 Version 1.3 to confirm that the documents contained adequate instruction for a user of the methodology. The staff further noted that the input data requirements document is largely consistent with the RELAP5-3D input data requirements manual with a few modifications corresponding to the differences between NRELAP5 and RELAP5-3D.

In addition, the staff reviewed multiple journal articles and a thesis for further background on statements in the non-LOCA TR and noted that the articles adequately supported the docketed statements.

The staff also reviewed several documents the applicant referenced during audit discussions of various topics. These documents are discussed under the corresponding topics in Appendix A.

8 EXIT BRIEFING The staff conducted an audit closing meeting at NuScales Rockville, MD, office on January 25, 2018. During the meeting, the staff reiterated the purpose of the audit and discussed the audit activities and outcome. The staff also indicated that the open items remaining at the conclusion of the audit would be transmitted as questions in RAIs. References to the detailed questions are provided in Section 9 of this audit summary.

9 REQUESTS FOR ADDITIONAL INFORMATION RESULTING FROM AUDIT The staff issued 15 questions in five different RAIs that either arose from or were informed by observations made during the audit. These RAIs are available in Agencywide Documents Access and Management System (ADAMS). ADAMS accession numbers are provided in Table

1. The questions in the RAIs that are not explicitly listed below resulted from the staffs review independent of the audit.

Table 1. RAIs Resulting from Audit RAI Number Question Numbers ADAMS Accession Number 9158 15.00.02-2, 15.00.02-5 ML18156A171 9351 15.00.02-31, 15.00.02-33 ML18156A177 9374 15.00.02-23, 15.00.02-24, 15.00.02-25, 15.00.02-26, 15.00.02-30 ML18129A407 9466 15.00.02-7, 15.00.02-9, 15.00.02-10, 15.00.02-11, 15.00.02-12 ML18128A341 9513 15.00.02-15 ML18128A389 10 OPEN ITEMS AND PROPOSED CLOSURE PATHS Not applicable.

11 DEVIATIONS FROM THE AUDIT PLAN The audit was originally scheduled to exit following the original audit end date of November 17, 2017, but was extended to January 25, 2018, to accommodate the examination and discussion of additional documentation requested by NRC staff.

12 REFERENCES

1. Letter from Thomas A. Bergman (NuScale Power, LLC) to NRC, LO-0117-52647, NuScale Power, LLC Submittal of Non-Loss-of-Coolant Accident Analysis Methodology, TR-0516-49416 (NRC Project No. 0769), January 10, 2017 (ADAMS Accession No. ML17010A427 [Proprietary information; not publicly available]).

2. Letter to NuScale Power, LLC, NuScale Power, LLC. - Acceptance of an Application for Standard Design Certification of a Small Modular Reactor, March 23, 2017 (ADAMS Accession No. ML17074A087).

3. Letter from Rani Franovich (NRC) to Samuel Lee (NRC), Acceptance Letter for the Review of NuScale Power, LLC Topical Reports TR-0516-49422, Loss-of-Coolant Accident Evaluation Model, Revision 0, and TR-0516-49416, Non-Loss-of-Coolant Accident Analysis," Revision 0 (CAC RN6303, RN6305), April 27, 2017 (ADAMS Accession No ML17116A063).

4. Letter from Zackary W. Rad (NuScale Power, LLC) to NRC, LO-0817-55242, NuScale Power, LLC Submittal of Topical Report Non-Loss-of-Coolant Accident Analysis Methodology, TR-0516-49416 Revision 1, August 10, 2017 (ADAMS Accession No. ML17222A827).

5. Letter to Samuel Lee Audit Plan for the Regulatory Audit Of NuScale Power, LLC Topical Report TR-0516-49416-P, Non-Loss-of-Coolant Accident Analysis Methodology, May 30, 2017 (ADAMS Accession Number ML17138A112).

6. NRO-REG-108, Regulatory Audits, April 2, 2009, ADAMS Accession Number ML081910260.

A-1 Appendix A - Audit Discussions A.1 Facility Layouts, Descriptions, and Models In response to staff inquiries about aspects of the integral pressurizer operation, the applicant stated that the degree of hermal stratification across the baffle plate is at most (( )) from the integral pressurizer to the top of the riser at 1850 psia. A small amount of normal spray flow passes across the baffle plate (~0.1 lb/sec), which has a negligible effect on the forces driving natural circulation considering that the reactor coolant system (RCS) flow rate is approximately 1290 lb/s. There are eight 4-inch diameter holes in the baffle plate, which roughly correspond to the flow capability of a 12-inch diameter surge line. Since the NPM has a larger RCS volume-to-core-power ratio than a conventional pressurized water reactor (PWR), the in-surge and out-surge rates would be lower than a conventional PWR.

To expand upon the interface of the non-LOCA methodology with fuel rod performance design, the applicant described the process used to determine the representative time-in-cycle core average burnup and the correlation used to calculate the burnup dependence of thermal conductivity by stating that the burnup corresponds to an average value for a typical UO2 core ranging from ~12 gigawatt days per metric ton (GWd/MT) (beginning of cycle [BOC]) to ~24 GWd/MT (end of cycle [EOC]). In addition, the thermal conductivity is consistent with the burnup-dependent value calculated by the fuel performance code.

With regard to the non-LOCA methodology interface with the core design analysis, the applicant explained that the 1973 decay heat model was selected because it was subjected to greater scrutiny in the QA process and is believed to have a superior pedigree to the other models. The applicant performed a calculation that examined the various decay heat models and determined that an appropriate multiplier for a maximum decay heat or minimum decay heat level was appropriate. The staff issued RAI 9466, Question 15.00.02-7, in part, to request information and justification for using this decay heat methodology.

TR Section 4.3.1.1.2, Axial Power Shape, discusses the use of a nominal center-peaked average axial power shape for non-LOCA analyses. The staff questioned the effect of different axial power shapes. The applicant stated that a nominal center-peaked shape is used in the non-LOCA EM for consistency with the development of reactivity coefficients in core design analyses. A loss of nonemergency AC power calculation examined the sensitivity to a top-and bottom-peaked shape and showed that the axial power shape has a negligible effect on primary and secondary pressures and system flow. This justifies a nominal center-peaked power shape for evaluating the system transient response. In addition, (( )).

With the average core, there is no radial peaking factor.

The applicant further stated that (( )).

The applicant stated that (( )) is important to the calculation of the MCHFR and is therefore of high importance in the PIRT, but in the non-LOCA transient analysis it may only be important for (( )).

In response to staff inquiries pertaining the critical heat flux (CHF) correlations, the applicant stated that he new NSP4 CHF correlation is based on a revised data set showing the same trends as the NSP2 correlation, and therefore no impact is expected on the non-LOCA methodology. Figure 5-1 in TR-0116-21012-P shows a (( )) trend in CHF as a function of pressure (( )). However, pressures during non-LOCA events remain above that pressure. (( )).

The staff asked the applicant to describe the magnitude of (( )) and whether any sensitivity studies have been performed to assess the effect of (( ))

A-2 on the stall and chug flow behavior noted in the TR. The applicant stated that the NPM evaluation model (( )).

In response to staff questions about how the NRELAP5 (( )). If a hot channel were modeled in NRELAP5, there would be preferential flow to the hot channel from the chimney effect.

)).

The staff asked the applicant to explain the purpose for the slotted riser columns, how the flow behaves in the lower riser region, and (( )), to which the applicant stated the slotted structures are control rod assembly guide tubes. (( )).

The staff and applicant also discussed (( )).

The applicant stated that the amount of heat loss through the riser under steady-state conditions would not generate the temperature gradient necessary for countercurrent flow. In a transient, after reactor trip, there is a density flip-flop that results in temporary flow stagnation and flow oscillations. Once the decay heat removal starts, a more steady flow pattern is established.

(( )).

The staff notes that 3-D effects could cause recirculating or convective regions to develop with the potential to reduce overall natural circulation and buoyancy. In the absence of any studies or calculations to demonstrate otherwise, the staff issued RAI 9351, Question 15.00.02-31.

The staff also asked for further details regarding DHRS modeling. The applicant described how

(( )).

TR Section 6.1.5 discusses that (( )).

Based upon the dominant bulk boiling heat transfer mode, the conclusion was that the performance would not be sensitive to temperature gradients. The cooler liquid near the bottom of the DHRS should enhance performance, and (( )).

)).

applicant stated that

)).

The staff asked the applicant to further explain whether the (( )) is applicable to the NPM model of the DHRS. The applicant explained that (( )).

NIST-1 tests HP-03 and HP-04 assessed (( )).

The NIST-1 test geometry, as well as the DHRS geometry, is consistent with the development of (( )). The NRELAP5 results gave a reasonably good prediction of the heat transfer coefficient. In the KAIST assessments, the applicant confirmed that (( )).

The Reynolds number range for condensation in the NPM DHRS is consistent with the range of

(( )). KAIST and NIST-1 HP-03 and HP-04 cover most of the Reynolds number range of the NPM DHRS operation. In addition, the applicant felt that the Reynolds number is an adequate indicator of whether the flow and the (( )) correlations are applicable.

In NRELAP5, all of the heat transfer correlations use (( )).

Because the staff needs additional docketed information to support the use of the (( )),

the staff issued RAI 9466, Questions 15.00.02-10 and 15.00.02-11.

In response to staff questions regarding documentation of user guidelines for the non-LOCA methodology, the applicant stated that the TR provides some specific guidance, such as

(( )) and heat transfer model options in each specific section. However, there is no separate, stand-alone modeling guideline document. However, NuScale has only

A-3 one plant design to model. Any changes to the model options would need to be documented and justified in the engineering calculation.

A.2 Non-LOCA Phenomena In response to staff inquiries regarding the NPM event frequency categorization, the applicant stated that the event classification was based upon historical precedent. Events like the feedwater line break, steam line break, and steam generator (SG) tube failure are common to the NPM and conventional plants and should have the same event frequency. The small line failure event is unique to the NPM and is classified as an infrequent event based on dose consequences. The PRA was not used for classification of the non-LOCA events. The report ER-0000-3221, Identification and Classification of Deterministic Design Basis Events for the NuScale Power Small Module Reactor, provides the description and justification for the event classifications.

The staff asked the applicant to explain how the NRELAP5 core and lower plenum representation of the NPM address certain highly-ranked phenomena and how some phenomena are addressed by the subchannel analysis. In response, the applicant explained that the phenomena identification and ranking table (PIRT) development process came well before specific model considerations and holistically looked at all the phenomena, including the subchannel effects. The important parameters for the subchannel analysis are not necessarily important to the non-LOCA transient response. Local effects do not affect the non-LOCA system transient response because there is no evidence of substantial two-phase conditions during non-LOCA events. (( )) are covered in the subchannel analysis.

The staff asked the applicant to further explain PIRT rankings for (( ))

since the audit document (( )).

The applicant explained that (( )).

In response to staff inquiries about how any of the changes resulting in the current design of the NPM and the current EM have been reflected in the PIRT, the applicant stated that there have been only a few changes since the PIRT was completed, such as the riser hanger, feedwater plenum, control rod assembly (CRA) guide tubes, and replacement of the riser gap with a bellows design.

have only a minor impact with regard to non-LOCA phenomena, although the riser gap change would have some impacts for long-term cooling. There is no formal procedure for dispositioning the effects of a design change on the PIRT, but all high-ranked PIRT items are addressed as part of the final EM development.

In addition, the applicant made some changes to NRELAP5 after the PIRT was prepared, but these changes were deemed insignificant. The analysis cases the applicant performed to support PIRT development were not re-done with the updated code. The PIRT was not changed due to changes in the plant design or NRELAP5 since it was determined that the changes were not meaningful to the PIRT phenomena rankings.

In ER-0000-2934, the staff noted that certain rankings do not seem to be consistent with their importance to primary natural circulation. The applicant stated that natural circulation is important to the plant behavior and response, but specific phenomena for each phase of the plant response are considered. For example, phenomena related to the (( )).

The staff noted that in ER-0000-2934, Table 5-3 identifies the certain H/2 phenomena (( )).

The applicant stated that CFD was not used specifically for the non-LOCA development. CFD is used at NuScale in design engineering, but not in this area.

A-4 The staff issued RAI 9351, Question 15.00.02-31, to obtain additional information about how the non-LOCA EM accounts for multidimensional effects.

A.3 Non-LOCA EM Validation and Assessments A.3.1 NIST-1 Validation and Assessments The staff asked about differences between the NIST-1 test facility and NPM models, particularly, nodalization differences. The applicant discussed the differences between the NRELAP5 NIST-1 test facility nodalization and the NRELAP5 NPM nodalization and stated that there is no comprehensive document that captures the modeling differences. Some parts of the NPM model have distinct bases from the test facilities, such as the core modeling, which is influenced by the interface to the core design codes. However, there is a general consistency between the length-to-diameter ratios.

In response to staff inquiries about the data assessment of NIST test HP-03, the applicant stated that he data agreement criteria in the TR is the same as in Regulatory Guide (RG) 1.203.

had to do with how the test was set up.

Many of the test conditions were input as boundary conditions for the simulation. Under-prediction of the heat transfer would lead to under-prediction of the level but excellent agreement in flow enthalpy. Other test assessments at higher pressure were in better agreement with heat transfer and level. Compensating effects do exist and may account for reasonable agreement between the test data and calculated parameters.

The staff questioned the difference in the calculated tube water level between the NIST low pressure test (HP-03-01) and the medium pressure test (HP-03-02c). The applicant responded that, consistent with the definition for reasonable agreement in RG 1.203, the major trends were well predicted even though some values were outside the range of uncertainty.

)).

there was no longer a significant difference between the medium pressure results compared to the low pressure results.

)).

In addition, the applicant speculated that the NRELAP5 over-prediction at higher elevations could be due to (( ))

could also account for some of the differences.

The staff noted a lack of agreement between the NRELAP5 simulation results and the measured test data for test HP-03-02c. The applicant stated that )).

The issues of how lessons learned from NIST tests were translated into NPM model development and re-examination of benchmarking studies are addressed in RAI 9466, Question 15.00.02-9.

In response to staff inquiries regarding differences in DHRS collapsed water level and pool water level in the RELAP5 simulation of test HP-03-03, the applicant stated that the figures in the TR were zoomed in, which accentuates the small differences. Given the uncertainties in the measurements, the predicted results are reasonable, even though the calculated trends are not quantitatively the same as those of the measurements.

The staff sought additional explanation of the (( )) observed in NIST test HP-04-02 results. The applicant stated that sensitivity studies were performed in Section 4.2 of document EC-T080-4163, NRELAP5 Assessment Against NIST-1 HP-04 NuScale Cooling Pool Characterization Test. (( )).

The staff asked the applicant about the lack of impact of the thermal stratification layer that is formed near the top of the DHRS and progressively moves downward, as discussed in

A-5 supporting document EC-T080-4163. The applicant explained that the DHRS power is driven by condensation inside the DHRS, which is in turn mostly driven by the primary temperature.

The specific pool temperature distribution has a much lower effect on the amount of power the DHRS can remove, and therefore, the thermal stratification does not have a significant effect.

(( )).

The staff questioned why (( )).

The applicant stated that the test procedure specifies (( )).

In the NPM, a loss of feedwater results in plant pressurization, resulting in a reactor trip and activation of the DHRS for decay heat removal. (( )).

The staff asked for further explanation of the NRELAP5 predictions of NIST tests NLT-02a and NLT-02b provided in supporting document EC-T080-4330, NRELAP5 Assessment Against NuScale Loss of Feedwater Flow with DHRS Test NIST-NLT-02 and NIST-NLT-02b. The applicant stated for test NLT-02a, )).

)).

When asked about differences between NLT-02a and NLT-02b results, the applicant stated that the initial conditions for NLT-02a and NLT-02b were very different: (( )).

In addition, for NLT-02b, (( )).

To explain other NLT-02b behavior, the applicant stated that

)).

)).

((

)).

It remains unclear how the lessons learned from the tests were translated into the plant model development. The staff issued RAI 9466, Question 15.00.02-9, in part, to obtain information about how the lessons learned are incorporated in the NPM model and to obtain the results of any revised analyses.

The staff sought an explanation for the NRELAP5 modeling of the NLT-02a test, (( )).

The staff inquired about the difference between the calculated riser flow rates and data for test NLT-02a and why (( )) after termination of power to the heater rods. The applicant stated that he (( ))

are about the same as those of the measurement. (( )).

The applicant had not planned to update the TR.

The applicant further stated that the (( )) is bounded by adjusting the fuel rod gap thermal conductivity in the NPM calculations. Flow rates are also biased high and low for the ranges of flow and the limiting condition for the case being analyzed. The approach used for the NPM analyses is not affected by (( )).

The applicant explained that

)).

RAI 9466, Question 15.00.02-9, requests that the results of reanalysis be provided and the TR updated to reflect the revised results as appropriate. RAI 9374, Question 15.00.02-23, notes several discrepancies between measured and calculated parameters in the NIST tests and seeks to obtain confirmation that the acceptable overall DHRS heat removal is not due to compensating errors.

A.3.2 SIET Validation and Assessments

A-6 The staff asked the applicant to further explain certain aspects of primary and secondary side modeling of the helical coil SG (HCSG).discussed in supporting document EC-T050-3638. The applicant stated:

(( )).

The plant model applies (( )).

The applicant developed the plant model and performed the validation tests in parallel but attempted to maintain consistency between the two. The TR discusses that (( )).

In addition, the staff discussed several aspects of the SIET facility and tests with the applicant.

The applicant clarified (( )).

The applicant acknowledged that the some of the information about materials was misleading or inconsistent with what is now known about the facility. (( )).

The staff noticed that (( )).

The staff expressed concern for the potential for (( )).

The staff noted large differences in (( )) readings for SIET TF-2 tests (( )).

The applicants explanation accounts for (( )) but led to the question of whether (( )) was included in the assessments.

(( )).

Therefore, the staff issued RAI 9351, Question 15.00.02-33.

A.3.3 KAIST Validation and Assessments The staff asked the applicant for further clarification regarding a supporting document, EC-0000-3155, Assessment of NRELAP5 with KAIST Condensation Experiments, Rev. 0. The applicant stated that

)).

The heat transfer coefficient has a 28 percent uncertainty band compared to the 6 percent uncertainty band for inner wall temperature. Falling outside the inner wall temperature uncertainty band does not mean that the heat transfer coefficient is outside its uncertainty band.

In response to the staffs observation of (( )),

the applicant stated that the maximum heat transfer location changes as the pressure changes, and there were several tests at different pressures. The heat transfer regime also changes for different elevations depending upon the heat and mass flow rates in the test, and the laminar-to-turbulent transition accounts for (( )).

A.4 Analysis Process and Example Calculations Regarding the statement in the TR that For non-LOCA initiating events that actuate the DHRS

[decay heat removal system], the evaluation model (EM) is applicable for the short-term transient progression; during this time frame the mixture level remains above the top of the riser and primary side natural circulation is maintained, the applicant stated that the TR addresses short-term transients and not necessarily the long term when the DHRS is fully effective. The methodology predicts periods of flow stagnation as the DHRS is ramping up and therefore does not require maintaining positive flow. During extended periods of DHRS cooling, fluid shrinkage may result in the fluid level falling below the top of the riser, which is not intended to be analyzed using the non-LOCA methodology. No short-term events that result in the mixture level falling below the top of the riser.

The staff requested further explanation about the use of the graded approach to the development of the non-LOCA EM and the difference between the LOCA EM and non-LOCA EM. The applicant stated that here is a large overlap between the LOCA and non-LOCA code

A-7 requirements and in the initial conditions of the NPM for LOCA and non-LOCA. The non-LOCA TR references the LOCA TR to preclude the need to duplicate information in the non-LOCA TR, and the applicant intended for the LOCA EM to be approved so the non-LOCA EM could use a graded approach based upon the LOCA EM. The RELAP5 base code version was the same for LOCA and non-LOCA use and is described in more detail in the LOCA TR.

The non-LOCA EM examines the core average transient response, and a separate subchannel analysis is performed for the SAFDL evaluation, while the LOCA EM incorporates a hot channel and other Appendix K requirements. The module control system is better represented in the non-LOCA EM to allow better understanding of the typical module response to a transient and to enable conservative event analysis. The nodalization is fairly similar between the LOCA and non-LOCA EMs but not identical, and differences are based on phenomena of importance. For example, the DHRS nodalization is more detailed for the non-LOCA EM, whereas the emergency core cooling system nodalization is more detailed for the LOCA EM. Generally, the NPM nodalization is comparable, except for the hot channel in the LOCA EM and the reactor pool. Due to the initialization process, the riser and downcomer modeling is generally the same.

In response to staff inquiries about the adequacy of the (( )).

In response to staff questions about the purpose of TR Table 7-1, the applicant stated that the table was intended to list all parameters that could be set for an analysis to provide the analyst a list to prioritize which parameters should be covered for a specific event. Because of the natural circulation flow, not all the parameters will lead to a given set of cogent conditions. This is the list that one starts with before eliminating parameters from consideration.

The applicant explained the terms null transient and loop transit that are used in the TR by stating that the base model steady-state calculation is performed at nominal full-power conditions. Depending upon the transient, some variables and initial conditions are biased to add the necessary conservatism to the transient calculation, which is accomplished through a restart. A null transient is a restart from the re-initialization case to demonstrate that steady-state conditions are achieved before starting the transient calculation. Part of the null transient evaluation is to examine the primary and secondary conditions to ensure that they are correct for the biased conditions. Nominal operating conditions are calculated in RCS flow calculations for a given power level as described in document EC-A030-2713, Primary and Secondary Steady State Parameters. (( ))

A loop transit is a full cycle around the RCS that corresponds to a fluid particle circulating from one point in the RCS around the fluid circuit back to the same point. At full power, the loop time is (( )).

The staff asked the applicant how dependent parameters can be biased without violating conservation principles. The applicant stated that biased-low RCS flow is achieved by

(( )).

The staff questioned how the applicant determined the rate of decrease in feedwater temperature for the decrease in feedwater temperature representative calculation and requested further explanation of the transient response from 500 to 750 seconds, which showed an inflection point. The applicant stated that it determined the range of feedwater temperature decrease rates by examining the a feedwater temperature change from a maximum feedwater temperature of 307.5°F with uncertainties to the minimum feedwater temperature of 100°F over a period ranging from 0.1 seconds (the minimum time) to 30 minutes. The limiting point occurred for coincident trips on high power and high RCS temperature trip signals.

00 seconds, and hot fluid is being generated in the core with a hot slug of water at the top of the riser and a cold slug of

A-8 water at the core exit. There is a very small RCS flow rate. From 600 seconds to 700 seconds, the hot slug of water is sufficient to cause buoyant flow up the riser and causes a spike in RCS flow. Around 750 seconds, colder fluid from the downcomer flows into the core. Some radiant heat transfer carries heat away. Those heat transfer mechanisms may be causing the inflection around this time.

The staff asked the applicant to explain why the increase in steam flow event was more limiting than an increase in feedwater flow event since the latter has a higher reactivity insertion. The applicant stated that a fundamental difference exists in the secondary side behavior between the two events. In the increase in feedwater flow event, there is a constant steam flow at the turbine. As feedwater flow increases, secondary pressure increases and reduces the secondary side heat transfer coefficient, which mitigates the increase in primary-to-secondary heat transfer that might be expected. The low superheat trip always occurs first for the increase in feedwater flow event since the RCS cannot slowly heat up to reach a coincident high power and high RCS temperature trip. In the increase in steam flow event, the secondary side depressurizes, enhancing the overall heat transfer, and the slow heat-up results in reaching coincident high power and high RCS temperature trip conditions. This is generally limiting for cooldown events. The limiting (( )) increase in steam flow event trips on high RCS temperature, nearly coincident with the time that a high power trip would occur. Higher ranges of increases in steam flow will cause a trip on low steam pressure.

For the increase in steam flow representative calculation, the staff asked the applicant to explain the multiplier of 0.7 applied to the heat transfer coefficient in the steady-state initialization, which reduces RCS flow at the beginning of and during the transient. The applicant stated that the multiplier is applied (( )).

Multiple adjustments can be made to achieve the same conditions. When the heat transfer was reduced, to maintain the same RCS flow rate, (( )). During the transient, reduced SG heat transfer resulted in a lower ramp-up in flow. So, at the time of MCHFR, there was (( )). Since the biased-low initial fuel temperature (( )), it was carried forward into the demonstration analysis.

However, the staff noted that some changes to maintain identical initial conditions, such as

(( )), may also affect transient response. RAI 9374, Question 15.00.02-26, related to the establishment of unique steady state initial conditions, relates to the topic of changes made to implement the bounding bias conditions for transient analyses.

To clarify the modeling of and response to a main steam line break, the applicant stated that each SG has two steam plena, each with two steam lines that join to result in just two steam lines exiting containment. In the model, (( )).

The staff also asked about the fuel temperature bias in the main steam line break representative calculation. The applicant clarified that gap conductance is changed to achieve the target fuel temperatures, which reduces the negative Doppler feedback and maximizes the power response. The temperatures are based on fuel performance code data that examine many different factors, including burnup and operating conditions. Conservative fuel temperature ranges at BOC and EOC are used in the transient calculations.

the failed SG blowdown in the main steam line break event, which causes a colder slug of water to build up. The resulting lower core inlet temperature causes an increase in the peak flow amplitudes in the oscillations.

A-9 To further explain the behavior of a turbine trip/loss of external load assuming a feedwater isolation valve (FWIV) fails to close, the applicant clarified that n the model, (( )).

Once the DHRS valves open, it takes (( )) for condensation to start driving flow in the DHRS. During this period, the loss of the secondary heat sink and secondary isolation results in secondary side pressurization. This (modeling) process mitigates variations in peak SG pressure since the peak pressure occurs after the DHRS valves are fully open.

(( )) since the DHRS operates at saturation. (( )), and RCS temperature oscillations result in different secondary peak pressure times. A loss of AC power results in an immediate loss of feedwater, while other cases result in a feedwater runback rather than an immediate feedwater loss. The reduced secondary supply, and therefore cooling capability, for a loss of AC power causes a (( )).

To clarify staff inquiries into the difference between the response of the MSIV closure event and the loss of condenser vacuum (LOCV) event, the applicant explained that the turbine and condenser are farther away from the module than the MSIV, which is at the top of the NPM.

Consequently, an MSIV closure would result in faster steam flow isolation. Additional bias sensitivity cases were performed for some heat-up events. (( )).

not show the questioned sensitivity dependence, and the staff requested to see the results of that analysis in RAI 9374, Question 15.00.02-30.

To address staff inquiries about the loss of normal AC power representative calculation, the applicant clarified that it had not performed sensitivity studies to examine the effect of biasing the pool water temperature for this event; the biasing was based on engineering judgement.

The difference in peak primary temperatures between the core outlet and average is a result of the stall and chug response. After reactor trip, the reduction in RCS flow results in higher density water build-up in the riser, resulting in a stall in riser flow. While the core is continuing to heat fluid, the flow inlet to the core is also providing hotter fluid to the core as a result of the loss of the heat sink. This combination results in a peak in the core average temperature at an earlier time than the core exit temperature.

The staff requested further explanation of the sensitivity studies for the loss of normal feedwater event. The applicant explained that the range of feedwater flow reduction tends to dominate the other initial conditions. Therefore, the feedwater sensitivities focus more on the range of feedwater flow reduction cases. The limiting bias conditions were determined mainly through experience with other transients and engineering judgment.

As the amount of the feedwater flow reduction is reduced, the peak SG pressure increases asymptotically, approaching the limiting case of a (( )) reduction. As the feedwater reduction is decreased further, the reduction no longer approaches the trip setpoint, but instead settles into a new steady state at a lower power level. Values less than (( )) were not examined because any smaller reduction in flow that does result in a reactor trip will only result in a very small increase in SG pressure above the value calculated for the (( )) reduction.

A reduction in the feedwater flow of (( )) causes a slow increase in RCS temperature, which slowly reduces core power due to Doppler feedback. As the RCS temperature increases to the trip setpoint, the riser temperature increases to a quasi-steady state, maximizing primary temperature and the potential to pressurize the secondary. With a full reduction in feedwater flow, there is a rapid increase in temperature that does not reflect a larger overall energy content in the primary.

A-10 The staff requested further explanation of the low-power CRA bank withdrawal methodology, including associated sensitivity studies. The applicant stated that the eactor trip is generated

(( )). The rods begin to drop two seconds later due to a signal processing delay. The power is very low, so the reactivity feedback effects are small. Because the power is so low and reactivity feedback is very small, the effect of a (( )).

As power is increased above (( )).

The staff also asked the applicant to explain the limiting condition and range of the sensitivity studies for the uncontrolled CRA bank withdrawal at power. The applicant clarified that ((

)). ower reactivity insertion rates result in a simultaneous increases in power, temperature, and pressure. The limiting cases result from simultaneous trips on two or more of these conditions. Higher reactivity insertion rates result in an earlier reactor trip on high power rate.

By targeting the highest reactivity insertion rate that resulted in simultaneous trips on high power, temperature, and/or pressure, without tripping on power rate, it was determined that the limiting case could be identified without need for further sensitivity cases between (( )).

The staff requested additional explanation of the oscillatory behavior noted in the single rod withdrawal representative calculation. (( )).

))

explains why it takes 187 seconds to increase from the minimum to maximum post trip core outlet temperature.

The post-reactor trip flow peak occurs at approximately 538 seconds into the transient. During the oscillatory stall-and-chug response, the flow is a function of the stable core heat-driven mass flow rate and the monometric balance of buoyant forces. The difference between the core inlet specific volume and the approximate core exit specific volume and riser exit specific volume is approximately (( )), which explains why the flow spike reaches a peak of approximately 40 percent of the initial steady state flow rate. In addition, the applicant acknowledged that multi-dimensional effects would likely mitigate the amplitude of the flow and temperature oscillations.

The staff also asked for an explanation of some of the biases for the CRA misoperation methodology. For the single CRA withdrawal, the applicant stated that the limiting case for MCHFR results when the over-power and over-temperature reactor trips occur simultaneously.

Biasing the RCS temperature LOW delays the high temperature trip as power is increasing, creating a more limiting end state condition. The increase in power drives the increase in temperature since the turbine load is constant.

Pressurizer level has an indirect effect on pressure control, which determines when the high pressure trip occurs. (( )).

In general, the strategy to find the limiting condition is to seek the simultaneous combination of high power, high pressure, and high temperature trips. Experience and knowledge of the plant response allows the analyst to know the conditions that lead to the simultaneous trips. There is a cliff edge effect in the reactivity rate such that a transition from one reactivity rate to another causes a reactor trip that will make a case non-limiting.

(( )),

which is less limiting. Cases between (( )) were not analyzed since the existing cases identified the bounding conditions, and the difference in reactivity insertion rates between these cases was not significant for the cases identified with the limiting power.

A-11 Rod drop cases result in (( )).

The events that result in an immediate reactor trip are largely driven by the initial conditions rather than moderator feedback.

In response to staff inquiries regarding the system response to the failure of small lines outside of containment, the applicant stated that the core temperature at both the inlet and outlet increase linearly proportional to the increase in the RCS average temperature. When the RCS average temperature is biased high, an increase in the RCS flow rate due to the break causes the temperature at the core inlet to increase and the temperature at the core outlet to decrease because the core differential temperature decreases to maintain the same energy transport from the core. The temperature change at the core is relatively small, (( )). Since these analyses assume a loss of offsite power at the beginning of the transient, the SGs isolate at the beginning of the event while the core continues at full power until a reactor trip is generated.

This expands the fluid in the RCS and increases pressurizer level. Since the break isolates at 1600 psia, the higher level and pressure results in a longer period of break flow.

When the RCS average temperature is biased low, the density difference is greater. The lower RCS average temperature case has a higher break flow but results in a more rapid isolation

(( )) than the higher RCS average temperature. In the low RCS average temperature case, reactor trip results from (( )).

In the high RCS average temperature case, reactor trip results from (( )).

The time difference results in more energy deposition into the RCS primary, increasing the level and pressure, which results in a longer time to isolate the break.

The staff requested clarification of the biases and sensitivity studies associated with the SG tube failure (SGTF) methodology. The applicant explained that the intent of the bias sensitivity analyses was to identify which parameters need to be varied for each analysis and for which parameters the conservative bias direction was known.

The applicant also stated that the mass release during an SGTF is primarily driven by (( )).

Kinetics conditions could lead to a slightly different steady state initialization condition but would not affect the transient progression since there is very little change in RCS temperature. The RCS initialization has a larger influence than the kinetics influence.