ML23310A166

NRC Advanced SER for NuScale Topical Report TR-0716-50350-P, Revision 3 Rod Ejection Accident Methodology NON- Proprietary

ML23310A166
Person / Time
Site:	99902078
Issue date:	01/04/2024
From:	NRC/NRR/DNRL/NRLB
To:
References
TR-0716-50350-P, Rev. 3
Download: ML23310A166 (20)

Text

1 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION TOPICAL REPORT TR-0716-50350-P, REVISION 3 ROD EJECTION ACCIDENT METHODOLOGY NUSCALE POWER, LLC Proprietary information pursuant to Title 10 of the Code of Federal Regulations (10 CFR) section 2.390, Public inspections, exemptions, requests for withholding, has been redacted from this document. Redacted information is identified by blank space enclosed within bolded double brackets, as shown here: ((

)).

1 INTRODUCTION 1.1 Summary By letter dated December 17, 2021 (Agencywide Documents Access and Management System (ADAMS) Accession No. ML21351A399), NuScale Power, LLC (NuScale, the applicant),

submitted, for U.S. Nuclear Regulatory Commission (NRC) staff review and approval, Topical Report (TR) TR-0716-50350-P, Revision 2, Rod Ejection Accident Methodology (Reference 1).

NuScale supplemented its submittal by letter dated September 14, 2022 (Reference 2) in response to requests for additional information (RAI), RAI No. 9936, from the NRC staff. The NRC staff conducted a limited scope audit for TR-0716-50350, Revision 2, starting on April 19, 2023 (Reference 3). On October 20, 2023, NuScale submitted Revision 3 of TR-0716-50350-P (Reference 4), hereafter referred to as the TR.

In the TR, the applicant described a method for analyzing the consequences of a control rod ejection accident (REA) for a NuScale Power Module (NPM) design. The staff performed a review of the methodology presented in the TR, information made available as part of the audit (Reference 3), as well as the applicants responses to RAI No. 9936, questions NTR-01 and NTR-02 (Reference 2).

The NRC staff review utilized the guidance in Regulatory Guide (RG) 1.236, Pressurized-Water Reactor Control Rod Ejection and Boiling-Water Reactor Control Rod Drop Accidents, (Reference 5). Based on its review, as provided below, the staff determined that the TR provides a methodology for analyses of REAs with the limitations and conditions as listed in Section 6 of this SER.

1.2 Description of a Generic Rod Ejection Accident Transient Event REAs are a class of postulated reactivity accidents that pressurized-water reactor (PWR) vendors are required to analyze to demonstrate compliance with the requirements of Title 10 of the Code of Federal Regulations (10 CFR) Part 50, Domestic Licensing of Production and Utilization Facilities, Appendix A, General Design Criteria for Nuclear Power Plants, General

2 Design Criteria (GDC) 28, Reactivity Limits (as described in Standard Review Plan (SRP) for the Review of Safety Analysis Reports for Nuclear Power Plants SRP Section 15.4.8 (Reference 6)), to obtain an NRC license for a particular reactor design. Additionally, REAs must be considered (among other postulated accidents) in dose consequence analysis required by 10 CFR 100.11, Determination of exclusion area, low population zone, and population center distance, or equivalent, such as Subsection (a)(2)(iv) of 10 CFR 52.47, Contents of applications; technical information.

The postulated REA is initiated by the sudden ejection of a control rod assembly (CRA) from the core of a reactor that is critical. The reactor can be at any state from hot zero power to hot full power, and the core could be at any stage of the reactor operation, from the beginning of cycle to the end of cycle. Partial power situations should be considered to explore bounding conditions. In general, a large number of initial conditions can affect the transient response and its ultimate termination and therefore must be examined to assess the safety of the reactor with fuel damage in focus.

In a typical rod ejection event, a CRA is rapidly ejected and accelerated by the system pressure, resulting in a near step insertion of positive reactivity to the core. The sudden addition of positive reactivity results in a corresponding rapid increase in local power and local fuel temperature. The only feedback mechanism that can counter this power increase is the Doppler effect (Doppler) of the fuel, which adds negative reactivity as the fuel temperature increases.

The Doppler feedback accumulates until it reverses the power increase, resulting in a typical power pulse. Finally, the ex-core power detectors trip the scram system, and the transient is terminated after non-ejected control rods are inserted and stable cooling is established. The duration of the rod ejection is approximately the scram delay time, which is short enough to ignore all system-related changes to the coolant temperature and pressure as the control system will trip the reactor with either a high-rate power increase or high-power level.

A second type of transient may occur when the worth of ejected CRA is relatively small. In this scenario, the rate of power increase is relatively small and slow. Consequently, the ex-core detectors do not reach the setpoint that trip the reactor because both the integrated flux increase and the rate of flux increase are small. In this case, a system-level response occurs, since the activation of reactor trips associated with the system response will terminate the transient.

1.3 Scope of the NRC Staffs Review and Approval The purpose of the TR is to describe the methodology that NuScale intends to use for REA analysis, as stated Section 1.1 of the TR. Accordingly, the NRC staff reviewed the REA methodology presented in the TR.

Because the supporting calculations provided in Sections 5 and 6 of the TR use NPM-20 as the target reactor design that is not finalized and the applicant requested approval of the TR as a generic methodology for REA analysis, the NRC staff placed Limitation and Condition 1 (as listed in Section 6 of this SER) on the TR identifying the need for an applicability review when the TR is referenced by an applicant or licensee. Limitation and Condition 1 applies to the NPM-20 design, once finalized, as well as to other NPM designs. Applicability should be addressed in specific licensing applications referencing this TR.

3 2

REGULATORY REQUIREMENTS, RELEVANT REGULATORY GUIDANCE, AND ACCEPTANCE CRITERIA 2.1 Regulatory Requirements The REA methodology presented in the TR was developed to support compliance with the regulatory requirements in GDC 28 of 10 CFR Part 50, Appendix A, which states:

Criterion 28Reactivity limits. The reactivity control systems shall be designed with appropriate limits on the potential amount and rate of reactivity increase to assure that the effects of postulated reactivity accidents can neither (1) result in damage to the reactor coolant pressure boundary greater than limited local yielding nor (2) sufficiently disturb the core, its support structures or other reactor pressure vessel internals to impair significantly the capability to cool the core. These postulated reactivity accidents shall include consideration of rod ejection (unless prevented by positive means), rod dropout, steam line rupture, changes in reactor coolant temperature and pressure, and cold water addition.

In addition, the REA methodology is predicated on compliance of the instrumentation and control system for a specific application with GDC 13:

Criterion 13Instrumentation and control. Instrumentation shall be provided to monitor variables and systems over their anticipated ranges for normal operation, for anticipated operational occurrences, and for accident conditions as appropriate to assure adequate safety, including those variables and systems that can affect the fission process, the integrity of the reactor core, the reactor coolant pressure boundary, and the containment and its associated systems. Appropriate controls shall be provided to maintain these variables and systems within prescribed operating ranges.

2.2 Relevant Guidance The TR references the acceptance criteria and guidance outlined in SRP Section 4.2 (Reference 7), SRP Section 15.4.8 (Reference 6), and RG 1.236 (Reference 5) for reactivity-initiated accidents. By following the provided guidance, described as follows, an applicant can demonstrate compliance with GDC 28. The following is a summary of this guidance and the associated acceptance criteria:

(1)

Cladding Failure: RG 1.236 describes cladding failure phenomena and fuel rod cladding failure thresholds that are acceptable to the NRC staff. The pellet cladding mechanical interaction (PCMI) caused by the sudden rise in power during the pulse phase of a REA requires a limit on the total energy (calories per gram (cal/g)) deposited as a function of cladding hydrogen content. Cladding failure can occur also because of high-temperature post-DNB (departure from nucleate boiling) oxygen-induced embrittlement and fragmentation, or high-temperature cladding creep. RG 1.236 combines these two cladding failure mechanisms into a composite high-temperature failure threshold.

Specifically, fuel cladding failure is presumed if predicted fuel temperature anywhere in the pellet exceeds incipient fuel melting conditions. Finally, fuel cladding failure is presumed if heat flux exceeds the critical heat flux criterion.

4 (2)

Coolability: Per RG 1.236, pin cooling is assumed failed for all pins with a total enthalpy of 230 cal/g or greater. In addition, pin cooling is assumed to fail if there is incipient fuel melting in the outer 90 percent of the fuel volume.

(3)

Radiological Impact: RG 1.236 provides guidance related to the calculation of fission product inventory that would be available after an event. These limits are used at the decision points for fuel temperature and enthalpy determinations, as well as for the number of failed rods that may lead to unacceptable radiological release.

2.3 Rod Ejection Accident Analysis Method Acceptance Criteria The methodology defined in the TR identifies a set of acceptance criteria for the performance of a reactor under a REA. Section 3.2.4 of the TR states that the methodology requires that no fuel failure occurs, and therefore no radiological release will occur as the result of a REA. NuScale defines the acceptance criteria in Section 2.2.2 of the TR to assure that no fuel failure will occur, as summarized below.

Fuel cladding failure is presumed if local heat flux exceeds the critical heat flux (CHF) thermal design limit.

The increase in radial average fuel enthalpy is limited to less than 33 cal/g.

The peak radial average fuel enthalpy is below 100 cal/g.

Fuel cladding failure is presumed if fuel temperature anywhere in the pellet exceeds incipient melting conditions.

Section 2.2.3 of the TR provides additional criteria to ensure core coolability in order to meet the requirements of GDC 28. Integrity of the reactor coolant pressure boundary is assured by ensuring that the maximum reactor coolant system (RCS) pressure remains below 120 percent of design pressure, as described in Section 2.2.1 of the TR. The staffs evaluation of these acceptance criteria is described in Section 4.2 of this SER.

3

SUMMARY

OF TECHNICAL INFORMATION This section summarizes the applicants methodology and briefly describes the codes used by the applicant, including their input, output, and analytic modeling methods, and assumptions.

3.1 General Information The methodology presented in the TR consists of the following three major parts:

(1) General Rod Ejection Progression, Phenomena Identification and Ranking Analysis.

(2) Computer codes and cross-section library used for the REA methodology.

(3) REA Methodology.

5 The REA methodology uses CASMO5 (Reference 8), SIMULATE5 (Reference 9),

SIMULATE-3K (Reference 10), NRELAP5 (Reference 11), and VIPRE-01 (References 12 and 13) computer codes to perform analyses of system response to a REA. ENDF/B-VII cross-section library is used in the nuclear analyses that are performed with the computer codes CASMO5, SIMULATE5, and SIMULATE-3K. The applicant provides a flow diagram in Figure 3-1 of the TR to show the interfaces between these different computer codes. Table 3-1 of the TR shows the parameters that are passed along between these computer codes.

The applicant provides a sample analysis to illustrate how the methodology can be used for REA analysis in Section 6 of the TR.

3.2 Outline of Rod Ejection Accident Physical Phenomena, Modeling, and Overall Methodology A REA is postulated to be caused by a failure of a control rod drive mechanism housing, which allows a control rod to be rapidly ejected from the reactor by the RCS pressure. The rate of reactivity insertion is dependent on the speed of the ejected rod and the differential rod worth (which is dependent on the location of the CRA).

Section 3.2 of the TR describes the computer codes used in the NuScale methodology and the evaluation flow-path is shown in Figure 3-1. The analysis starts with the steady-state neutronics calculations performed with the methods using CASMO5/SIMULATE5 computer codes that the NRC staff approved in TR-0616-48793-P-A, Revision 1 (Reference 14). CASMO5 is used to generate a cross-section data library for use by the 3-D steady-state nodal code SIMULATE5 and 3-D transient analysis code SIMULATE-3K. SIMULATE5 initializes the cycle-specific model and reactor conditions that are used as input into the SIMULATE-3K evaluation.

SIMULATE5 is used to determine the steady-state initial condition portion of the REA calculations for the core response analysis. This steady-state assessment involves two calculations: determining the worst rod stuck out and development of the initial conditions to SIMULATE-3K model in a format of SIMULATE5 restart file.

Then, SIMULATE-3K solves the transient 3-D, two-group neutron diffusion equations, using the SIMULATE5 restart files. The transient simulation involves two calculations: simulation of the transient and determination of the parameter uncertainties used to bias inputs to the transient simulation. SIMULATE-3K analyzes the transient neutronic behavior under a REA at various times in the reactor cycle, power levels, control rod positions, and initial core conditions and provides the total core power, 3-D power distributions, and peaking factors during the REA transient.

The transients that result in a power spike with a lower magnitude, but longer duration necessitate a system-level code to determine the coolant temperature and pressure and to identify any phase change in cases where the pressure is dropping. The TR employs the NRELAP5 code to calculate the system response based on input from SIMULATE-3K. Results from the system response analysis determine whether the reactor coolant system pressure limit is exceeded. ((

))

6

((

)). Section 4.4.3.2 of this SER provides further discussion on the screening method and criteria.

The applicant used the VIPRE-01 subchannel analysis computer code to examine the fluid dynamics and heat transfer behaviors of the hot channel identified by SIMULATE-3K. The applicants fluid dynamics and heat transfer calculations cover the most highly challenged fuel assemblies, recognizing every fuel rod and allowing for both axial and transverse flow.

VIPRE-01 uses radial and axial power distribution input from SIMULATE-3K and core exit pressure, system flow, and core inlet temperature forcing functions from NRELAP5. The primary output from this analysis is detection of rod failure based on the fuel failure criteria, as identified in Section 2.3 of this SER.

The subchannel methodology also implicitly relies on the fuel performance code COPERNIC (Reference 15) to supply application-specific fuel thermal properties, as described in Section 4.4 of TR-0915-17564-P-A, Revision 2 (Reference 12). In addition, the NuScale REA methodology uses the statistical subchannel analysis methodology as defined in TR-108601-P, Revision 4 (Reference 13), to evaluate fuel failure criteria and uncertainties associated with the input parameters for subchannel analyses.

In Section 4 of the TR, NuScale presented an overview of the phenomena important for the REA, which is used to develop conservative assumptions for the analysis. There is no change in this revision of the TR in phenomenon identification and ranking table (PIRT) results compared to that of the previous NRC approved Revision 1 of this TR (Reference 16).

Section 5 of the TR describes the REA methodology. The methodology requires a REA analysis for each core reload design. Section 5.1 discusses the general assumptions and considerations used in the REA analysis. These assumptions and considerations include reload cycle-specific core design features, the time in the cycle when a REA occurs, core power level, single active failure, automatic system response of non-safety systems, and loss of alternating current power.

Section 5.2.1 of the TR describes calculation procedures for the rod ejection analysis methodology. These procedures include initial condition calculations and transient calculations, determination of the worst rod stuck out and treatments of key parameters such as Doppler temperature coefficient (DTC) uncertainty, moderator temperature coefficient uncertainty, travel time of the ejected rod, and ex-core detector modeling.

Section 5.2.2 of the TR provides the assumptions and treatment of uncertainties associated with the key system parameters that must be used in rod ejection analyses. These assumptions include the worst rod is stuck out and the regulating groups of control rod assemblies are inserted at the power dependent insertion limit (PDIL).

Section 5.2.2 of the TR discusses key input parameters. These parameters include rod ejection time, rod ejection location, and modeling of the reactor trip signals. The values for the reactor trip setpoints are input parameters that are specified based on the NPM-20 design. The TR provides high power trip and high power rate trip setpoints used in the methodology.

7 In addition, Section 5.2.2 of the TR discusses treatments of reactivity coefficients DTC and moderator temperature coefficient (MTC) and the effective delayed neutron fraction variation as a function of core burnup and time in the cycle at which a REA takes place.

Table 5-1 of the TR provides example uncertainties for rod ejection calculations. The TR requires that the uncertainties associated with the key parameters listed in Table 5-1 must be reexamined for each rod ejection analysis for each reload design.

TR Section 5.3 provides calculation procedures for the system response of the rod ejection analysis accident of the NuScale NPM-20 reactor. The objective is to determine the system responses to a sudden injection of a large reactivity resulting from a REA. The two most paramount system parameters are the peak reactor coolant system pressure and MCHFR.

The purposes of the system response calculations are to determine the peak RCS pressure analysis and to provide inputs to the subchannel analysis for CHF determination. The critical heat flux scoping cases are performed to determine the general trend and to select the cases to be evaluated in the VIPRE-01 subchannel analysis for final confirmation that no MCHFR fuel failures occur.

The TR considers competing scenario evaluations between the peak pressure and the MCHFR calculations. To examine the MCHFR, the system responses to maximum ejected CRA worth scenarios for different initial power levels are calculated. To examine the system pressurization, a reduced reactivity insertion rate assumption is used to allow the reactor power to rise less rapidly without triggering the high power rate setpoint that will trip the reactor.

Section 5.3.2 of the TR discusses assumptions and parameter treatment used in the system response calculation. No pressure reduction associated with the postulated failure of the control rod drive mechanism is assumed. Parameter treatments include initial power, inlet temperature, coolant flow, and model options governing direct moderator and cladding heating.

In addition, the axial power distribution is used as the assembly-average axial power at time of maximum core neutron power for the assembly containing the highest peak FH rod that is calculated using the SIMULATE-3K computer code.

Section 5.4 of the TR provides detailed procedures for performing thermal-hydraulic and fuel response analyses. The VIPRE-01 (References 12, 13, and 17) code is used to perform subchannel analyses. However, several deviations ((

)).

8 The TR states that ((

)). For these reasons, cycle-specific sensitivity studies must be performed for determining the adequacy of these values.

Power distributions within the fuel assembly in which the ejected rod is inserted and across the core are of a concern for rod ejection analyses because power shape could potentially affect the worth of the control rod. Figure 6-9 provides an example core radial power distribution, while Figure 6-10 provides an example hot assembly radial power distribution from the limiting statepoint at time of peak power. In the default radial nodalization, SIMULATE-3K power distribution inputs would be used to represent all fuel rods in the core. ((

)).

As discussed in Section 3.2.4 of the TR, the REA methodology does not include assessment of inventory or dispersion of radiological materials because the acceptance criterion of the methodology requires that no fuel failure occurs and, therefore, there is no need to include such analyses.

The staff reviewed Sections 1, 2, 3, 4, and 5 of the TR. The staff finds that Section 1 of the TR provides an overview of the methodology for REA analysis; Section 2 provides regulatory considerations for a REA; and Sections 3, 4, and 5 provide a detailed description of the methodology for performing analysis of a REA associated with the NuScale NPM-20 design.

The TR also specifies the assumptions and limitations of the REA methodology.

Section 6 of the TR includes results of selected sample calculations and sensitivity analyses.

Table 6-1 provides a list of these sample calculations and sensitivity analyses. Notably, however, some of the calculations are used as justifications of the adequacy of the REA methodology and referenced in Section 5 of the TR; the results are for illustration purpose only.

Therefore, unless being referenced in Section 5 of the TR, these sample calculations are not part of the methodology for which NuScale is seeking NRC staffs approval.

4 TECHNICAL EVALUATION 4.1 General Information The NRC staff reviewed Section 1 of the TR and finds it to be consistent with the purpose and scope of the TR, subject to an applicability determination, reviewed and approved by the NRC, when applied to a specific design.

Section 2 of the TR includes discussion of applicable regulatory requirements and acceptance criteria, including limits to evaluate RCS integrity, fuel cladding failure, and core coolability.

Section 4.2 of this SER provides detailed evaluations of the acceptance criteria with respect to the regulatory requirements.

9 Section 4 of the TR includes a PIRT. The staff compared this PIRT to the previously approved PIRT in TR-0716-50350-P-A, Revision 1 (Reference 16) and noted there were no changes.

However, the staff noted that Revision 3 of the TR is intended for higher power levels and different thermal-hydraulic conditions than Revision 1, and includes changes that may alter the magnitudes of parameters such as power peaking, enthalpy increase, or RCS pressure; however, the NRC staff determined that these changes do not introduce new phenomena or alter the relative importance of phenomena for the rod ejection accident. Based on the above discussion, the NRC staff determined that use of the PIRT in this TR is acceptable.

Revision 4 of the SIMULATE5 and Revision 7 of the CASMO5 code and the ENDF/B-VII cross-section library are used to generate core neutronic characteristic parameters for the SIMULATE5 code to calculate the reactor power distribution and control rod reactivity worth.

These codes and library have been previously reviewed and approved by the NRC staff (Reference 14) and the staff finds them to be acceptable for REA analysis.

The applicant updated important core neutronic characteristics, such as the reactivity, effective delayed neutron fraction (eff), Doppler coefficient, moderator reactivity coefficient, and other parameters, as functions of time in the cycle, with the increased power rate using the same methodology as approved by the staff (Reference 16). As discussed in Section 4.3 of this SER, the staff reviewed these revised parameters and the supporting calculations and finds the new values to be acceptable for use in demonstrating the REA analyses methodology. However, because the values of these parameters are design specific, they will be determined when this REA methodology is applied.

The staff reviewed the general descriptions of the rod ejection methodology as presented in the TR and finds it to be acceptable because the TR provides an overall description of the methodology with sufficient details to evaluate consistency with the guidance provided in RG 1.236 (Reference 5).

4.2 Acceptance Criteria The NRC staff compared fuel failure criteria identified in TR Section 2.2 to those in RG 1.236.

The rod ejection analysis methodology ensures that PCMI fuel failure does not occur by setting a maximum radial average energy deposition of 33 cal/g. The applicant determined this value by examining the RG 1.236 PCMI failure threshold for the fuel with unlined recrystallization annealed (RXA) cladding that is used in NuScale designs, as stated in Section 5.4.3 of the TR.

The methodology also requires that peak radial average fuel enthalpy remain below 100 cal/g.

This limit corresponds to the cladding-pressure-dependent fuel failure criterion for brittle and ductile fuel failure modes at high temperature. The NRC staff compared the peak radial average fuel enthalpy criterion in the TR to that in Figure 1 of RG 1.236 and finds that the TR limit is below the RG 1.236 threshold at all cladding pressures. On this basis, the NRC staff finds this criterion is conservative and acceptable.

The NRC staff finds that NuScales approach of applying steady-state threshold that is below the RG 1.236 hydrogen-concentration-dependent PCMI failure threshold and cladding-pressure-

10 dependent high-temperature failure threshold is consistent with the regulatory position in 2.2.1.5 of RG 1.236.

NuScale evaluates MCHFR fuel failure for all analyzed power levels and exposure points, as discussed in Sections 5.1.2 and 5.1.3 of the TR. The NRC staff finds this to be consistent with the RG 1.236 guidance that fuel cladding failure should be presumed if local heat flux exceeds thermal design limits in prompt critical scenarios which experience a prolonged power level following the prompt pulse and non-prompt critical excursions, regardless of initial power level.

In addition, the methodology assumes cladding failure if fuel temperature anywhere in the pellet exceeds incipient melting conditions, which is consistent with the guidance in RG 1.236.

The staff notes that the TR fuel failure acceptance criterion is established based on the PCMI fuel failure criterion of RXA cladding. Therefore, cladding type must be evaluated when this TR is referenced in licensing applications, consistent with Limitation and Condition 1.

The TR provides criteria to ensure core coolability in Section 2.2.3. The NRC staff reviewed these criteria and finds them to be consistent with guidance provided in RG 1.236 and therefore, acceptable.

4.3 Software Applicability Section 3.2 of the TR presents the computer codes used in the NuScale REA methodology and states that as part of the applicability review these computer codes have been previously reviewed and approved by the staff for the applicants and licensees to use in TR-0616-48793-P-A, Revision 1 (Reference 14), with the exception of SIMULATE-3K.

Applicability needs to be demonstrated by the applicant or licensee and reviewed by NRC staff when the TR is referenced. However, NRC staff reviewed the rod ejection methodology with the codes and methods approved in TR-0616-48793-P-A, Revision 1 (Reference 14) as part of the methodology. It is important to note that use of different nuclear analysis method may represent a significant change to an element of the methodology and would likely require NRC staff review, as identified in Limitation and Condition 3 in Section 6 of this SER.

Section 3.2.1.3 of the TR includes the code description and Section 3.2.1.4 describes the validation of SIMULATE-3K. In TR Section 3.2.1.4, NuScale used data from the SPERT-III tests and a Nuclear Energy Agency Committee on Reactor Physics control rod ejection benchmark problem to validate SIMULATE-3K for use in analyzing a REA. Details for the validation are provided in Appendix A of the TR.

The applicant has demonstrated that SIMULATE5 and SIMULATE-3K computer codes used in the REA methodology can capture the impacts of power tilt in calculating the reactivity worth of the ejected control rod, total reactivity insertion, and the power excursion response inside the fuel assembly in which the ejected rods were inserted. Based on the staffs review of the information provided by the applicant and the staffs confirmatory analyses, the staff determined that the power tilt inside the fuel assemblies and its impacts on rod ejection accident is adequately captured by the codes and calculation procedures. The staff also performed confirmatory analyses to confirm the SIMULATE5 and SIMULATE-3K computer codes are capable of reliably predicting the power tilt within a fuel assembly and its impact on the reactivity

11 worth of the control rod. The NRC staff performed confirmatory analysis using the POLARIS lattice physics module of SCALE, and the Purdue Advanced Reactor Core Simulator (PARCS).

The results of the confirmatory analyses provide evidence that SIMULATE5 and SIMULATE-3K are capable of reliably predicting assembly-wise power distribution and control rod worth, and that ejected control rod worth is not sensitive to the detailed flux distribution within individual assemblies.

4.4 Methodology Section 3 of the TR provides a general description of the computer codes that are used in the rod ejection analysis methodology and a flowchart to show the interfaces between different codes.

4.4.1 Steady-State Initialization Section 3.2.1.2 of the TR describes how SIMULATE5 initializes the cycle-specific model and reactor conditions, which SIMULATE-3K then uses to simulate the REA. Section 5.2.1.1 describes the steady-state calculations methodology. The steady-state analysis consists of an assessment of the worst rod stuck out and the development of the restart file for initial conditions for SIMULATE-3K. The NRC staff-approved TR-0616-48793-P-A, Revision 1 (Reference 14) describes the use of SIMULATE5 for non-LOCA analyses. The initial conditions of reactor power, inlet temperature, coolant mass flux, fission product material, identification of the CRA groups, positions of the CRAs, and information about the spacer grids are passed as input to SIMULATE-3K for use in the REA simulation. In Section 5.2.1.1, NuScale stated that the core flow, and thus the coolant mass flux, for a given initial power is held constant through a modeling option. The staff finds that this is consistent with previously approved TR-0716-50350-P-A, Revision 1 (Reference 16). The NRC staff considered whether other changes (that is, changes in the methodology, changes in the expected application, or changes in NRC staff guidance) would require detailed review of these initial conditions. The NRC staff did not identify any such changes and determined that there is no need for detailed review for this TR.

The staff reviewed Section 5.2 of the TR and finds that the method for developing steady-state conditions is consistent with the non-LOCA accident methodology, as presented in TR-0516-49416-P, Revision 4 (Reference 11), using the nuclear analysis codes and methods in TR-0616-48793-P-A, Revision 1 (Reference 14) and is, therefore, acceptable.

The TR does not describe steady-state initialization methodologies for non-baseload operation.

Because some non-baseload operation schemes may involve operation with regulating rods being inserted, the axial and radial power may be significantly skewed compared with baseload operation. The insertion of the regulating rods will suppress power density (and therefore suppress fissile material depletion) in the upper region of the core, which exacerbates the power excursion under a rod ejection. Further, the staff considered the fact that the worth of an ejected rod is dependent on the magnitude of the axial offset, and that the axial offset assumptions drive the level of conservatism in the analysis. To estimate the significance of the impact, the staff conducted a confirmatory analysis. The codes used in the analysis are described in Section 4.3 of this safety evaluation. The staff examined sensitivity of the control rod worth to axial offset by varying the axial xenon distribution present during a control rod ejection. The analysis confirmed that there is a significant sensitivity between the ejected rod worth and the magnitude of the

12 axial offset. Since changes to axial offset from non-baseload operation are not accounted for in NuScale's REA methodology, the methodology cannot be applied to analysis of REA events where core operation includes control rod insertion resulting in significant skewed axial fuel depletion. Additionally, depending on the non-baseload operation scheme and attendant frequency, magnitude, and rate of power changes, assumptions regarding axial power shape and xenon distribution may require modification or additional justification or clarification. This limitation is reflected in Limitation and Condition 2.

4.4.2 Core Response Section 5.2.1.2 of this TR describes the transient core response calculations performed using SIMULATE-3K for the NuScale REA methodology. The methodology first determines conservative parameter uncertainties and then simulates the transient based on conservatively applying the uncertainties. The staff reviewed the spectrum of input values used in the dynamic core response analysis, the initial conditions considered, the ability to capture the most limiting case, and the analytical methods.

NuScale applies numerical multipliers to conservatively bias the Doppler temperature coefficient, moderator temperature coefficient, effective delayed neutron fraction, and CRA worth according to the assessed uncertainty. The magnitude of the uncertainty is based on Nuclear Reliability Factors assessed in TR-0616-48793-P-A, Revision 1 (Reference 14) for SIMULATE-5, and NuScale stated in TR Section 3.2.1.4 that the benchmark comparison between SIMULATE-5 and SIMULATE-3K establishes applicability of these uncertainties to SIMULATE-3K. ((

)). As discussed in Section 5.2.2.3.4 of the TR, MTC and DTC are biased to be as least negative as possible, and the effective delayed neutron fraction is biased to be as small as possible. The worth of the ejected CRA is increased, and the worth of the CRAs inserted with reactor trip is decreased. The NRC staff finds these biases to be appropriately conservative as they will increase the severity of the analyzed accident. Additionally, because the methodology for determining the maximum rod worth accounts for calculation uncertainties in neutronic parameters, the NRC staff finds that this treatment is consistent with position 2.2.1.4 of RG 1.236.

The input core geometry and material compositions, core operating conditions, and core configuration come from a SIMULATE5 restart file according to the methodology described in TR-0616-48793-P-A (Reference 14). The VIPRE-01 thermal-hydraulic conditions are based on conservative NRELAP5 runs and include VIPRE-01-specific conservatisms consistent with the methodology presented in TR-108601-P, Statistical Subchannel Analysis Methodology, Revision 4 (Reference 13). Additionally, the NuScale methodology includes turning off the point kinetics in NRELAP5 while performing MCHFR analyses (as power pulses from SIMULATE-3K are used directly) but continues to use them for the overpressure analyses. The staff reviewed the information provided in the TR and TR-108601-P (Reference 13) and determined that NuScales methodology ensures SIMULATE-3K conservatively calculates potential fuel failures by choosing conservative input values and following the NRC staff-approved methodology described in TR-0616-48793-P-A (Reference 14). This supports the statements provided in the

13 TR. The staff finds that NuScale has conservatively chosen input values to ensure that the consequences of a reactivity-initiated accident are not underpredicted and is, therefore, acceptable.

The TR describes the process for performing the transient calculations once the uncertainties have been applied to the nuclear parameters. The staff finds that the information is the same as presented in the previously approved TR-0716-50350-P-A, Revision 1 (Reference 16). The NRC staff considered whether other changes (that is, changes in the methodology, changes in the expected application, or changes in NRC staff guidance) would require detailed review. The NRC staff did not identify any such changes and therefore, did not perform a detailed review.

Section 5.2.2.3.1 of the TR describes the calculation for time to eject the CRA from the core.

The ejection time is calculated using the appropriate PDIL depth at the initial power and the CRA acceleration equation in Equation 5-1. Section 5.2.2.3.1 of the TR states that the acceleration is based on the CRA cross-sectional area and weight of the CRA and control rod driveshaft. The NRC staff audited the example calculations and confirmed that no pressure barrier restriction is assumed (Reference 3). Because the method of calculating the rate of ejection, including the assumption of no pressure barrier restriction, is consistent with position 2.2.1.7 of RG 1.236, the NRC staff finds it acceptable.

Section 4.3(B) of the TR states that the limiting rod worth for the REA occurs when the rods are at the PDIL and that is used as the starting point for the calculations. The staff notes that plant operation only allows the rods at or above the Power Dependent Insertion Limit (PDIL) (see response to NRC Question No. 15.04.08-8 in RAI No. 9306 in Reference 16). The staff finds this conclusion acceptable because, with all else equal, a rod ejection event starting with CRAs above the PDIL will result in a smaller reactivity insertion and hence a smaller power pulse. The NRC staff considers the assumption that control rods are initially at PDIL, factoring in uncertainty for the CRA position, to be consistent with position 2.2.1.4 of RG 1.236 and, therefore, acceptable.

Because the reactivity worth of an ejected CRA is dependent on its location in the core and the three-dimensional core power distribution, and rod ejections from different locations in the core may produce different sizes (widths and heights) of pulses and levels of power peaking, the TR requires examining multiple CRA ejections. Explicit evaluation is performed for each regulating rod unless the core design is quarter-core or eighth-core symmetric. In this case, regulating rod ejections in each unique quadrant-symmetric location are evaluated. The NRC staff considers this an acceptable method of identifying the location of the limiting control rod consistent with position 2.2.1.4 of RG 1.236 because it evaluates ejection from each control rod location that will produce a different result.

Examples of trip setpoints are provided in Section 5.2.2.3.3 of the TR. NRC staff expects that fuel temperature, enthalpy, and heat flux will be insensitive to these trip setpoints in prompt critical scenarios, as limiting transients for the expected application will involve a rapid increase in power to well above the initial condition. The outcome of rod ejections with smaller initial power increases will be more sensitive to trip setpoints. Reactor trip delays are assumed, as discussed in Section 5.1.4 and Section 5.3.1.2 of the TR.

As discussed in Section 5.1.4 of the TR, the conservative single active failure for a REA is a failure of the neutron flux detector in the high-flux region. The staff verified that this is identical to

14 the previously approved TR-0716-50350-P-A, Revision 1 (Reference 16). The NRC staff considered whether other changes (that is, changes in the methodology, changes in the expected application, or changes in NRC staff guidance) would require additional review of this assumption. The NRC staff did not identify any such changes, and, therefore, did not perform additional review.

4.4.3 Dynamic System Response Section 5.3 of the TR presents the system response for the REA analysis. These system response calculations determine the peak RCS pressure and provide thermal-hydraulic response inputs to the subchannel analysis for CHF determination. The NuScale REA methodology follows the non-LOCA evaluation methodology in TR-0516-49416-P, Revision 4 Non-Loss-of-Coolant Accident Analysis Methodology (Reference 11) but with modifications to ensure conservative results when modeling reactivity-initiated accidents. The following sections discuss NRC staff evaluation of the peak pressure and MCHFR portions of the system response calculations.

4.4.3.1 Peak Pressure Calculations The calculation procedure in Sections 5.3.1 and 5.3.1.2 of the TR details the methods used to calculate the peak pressure resulting from a REA. To conservatively perform the peak pressure analysis, the methodology uses an ejected CRA worth, which results in a power increase just below the high power and high power rate trip setpoints for the reactor. This maximizes the length of the transient, which is then terminated by high RCS pressure. These cases do not require an upstream SIMULATE-3K calculation. The staff reviewed the methodology and input assumptions in Section 5.3.1.2 of the TR and finds that the methodology as described would conservatively calculate the maximum RCS pressure because the calculation uses bounding assumptions in the transient analysis and is, therefore, acceptable.

The peak pressure calculation methodology is unchanged from the previously approved TR-0716-50350-P-A, Revision 1 (Reference 16). The NRC staff considered whether other changes (that is, changes in the methodology, changes in the expected application, or changes in NRC staff guidance) would require additional review of this assumption. The NRC staff did not identify any such changes and therefore, did not perform a detailed review of the peak pressure calculation methodology.

4.4.3.2 Minimum Critical Heat Flux Ratio The calculation procedure detailed in Section 5.3.1 of the TR states that NRELAP5 scoping cases determine the general trend for selecting the cases to be evaluated in the VIPRE-01 subchannel analysis for final confirmation that no MCHFR fuel failures occur. Section 5.3.3 of the TR states that the scoping methodology described in Section 4.3.5 of TR-0516-49416-P, Revision 4 (Reference 11), is used to determine generally limiting scenarios for final MCHFR calculation with the subchannel analysis methodology. The screening criteria are defined in Section 4.3.5 of TR-0516-49416-P, Revision 4 (Reference 11). The staff reviewed the MCHFR calculation with the subchannel analysis, the initial conditions considered, the ability to capture the most limiting case, and the analytical methods. Because of the interdependence between the scoping methodology for the rod ejection methodology and the screening criteria of TR-

15 0516-49416-P, Revision 4 (Reference 11), NRC staff assessed this TR and TR-0516-49416-P, Revision 4 (Reference 11) together for consistency.

Section 5.3.1.1 of the TR provides the conservatisms included in the methodology for the MCHFR analyses. The staff concludes that the system condition assumptions used in the MCHFR analysis methodology are conservative, in that they result in increased fuel and coolant temperatures, which is conservative for MCHFR. Additionally, Section 5.3.1.1 states that high and low pressure conditions are investigated due to the unique nature of REA and the potential impact on core flow. The staff finds that the method for determining MCHFR is consistent with the methodology outlined in TR-108601-P, Revision 4 (Reference 13) and is, therefore, acceptable.

4.4.4 Detailed Thermal-Hydraulic and Fuel Response Section 5.4 of the TR presents the subchannel response methodology, which calculates the MCHFR, fuel enthalpy and fuel temperature and compares them against the relevant fuel failure criteria to verify that no fuel failure occurs.

As described in Section 3.2.3 of the TR, VIPRE-01 contains a fuel rod model with radial profile, theoretical density, and gap conductance supplied from a fuel design-specific calibration to COPERNIC. Additional detail of this process is described in Section 4.4 of TR-0915-17564-P-A, Revision 2 (Reference 12), which clarifies that the entire range of possible time-in-cycle parameters are evaluated, and the VIPRE-01 model is calibrated to ensure that it produces conservative fuel temperatures for each fuel design. The NRC staff previously approved the use of this fuel rod model in evaluation of rod ejection accidents in the previous revision (Revision 1) of this TR (Reference 16). The current revision (Revision 3) of the rod ejection methodology requires cycle-specific sensitivity studies on fuel heat transfer parameters to ensure conservative evaluation of all fuel failure quantities of interest, such as critical heat flux and radial average fuel enthalpy increase. Because these studies cover the full range of the fuel rods lifetime, and cycle-specific sensitivity studies are used to identify limiting biases for each fuel failure criterion, NRC staff finds that this approach is consistent with position 2.2.1.9 of RG 1.236 and is, therefore, acceptable.

The radial power distribution used in each case-specific VIPRE-01 case comes from the SIMULATE-3K pin power reconstruction model. In order to account for uncertainty in radial peaking, ((

)). Section 5.4.2.1 of the TR describes two options for evaluating fuel failure criteria since the MCHFR limit implicitly includes FH engineering uncertainty and non-CHF fuel failure limits do not. The first option ((

)). The NRC staff finds both options acceptable, as they both account for radial power distribution uncertainties Commented [SJ1]: The staff will confirm the Reference 11 screening method is consistent with this LTR through its review of Reference 11. The final SER for this LTR will be issued at the same time as issuance of the Reference 11 SER.

16 identified in TR-0915-17564-P-A (Reference 12), which the NRC staff previously found to be acceptable, when evaluating relevant fuel failure criteria.

Section 5.4.5 of the TR describes the sensitivity studies necessary for applying the REA analysis method. Table 5-3 lists the sensitivity studies. Optional sensitivity studies are performed when any non-default VIPRE-01 calculation control parameters or options are used.

When a non-default parameter is selected, the user will perform additional calculations with greater and lesser values than the selected parameter value. To ensure that the calculation is converged, NuScale will ensure that excellent agreement is achieved as defined in RG 1.203, Transient and Accident Analysis Methods (Reference 18), as stated in Section 5.4.5 of the TR.

These sensitivity studies will demonstrate that convergence is achieved by ensuring that changes in the VIPRE-01 parameters do not change the results or the progression of the transient. TR-108601-P, Revision 4 (Reference 13) Section 7.2.1, bullet 3 describes procedures for ensuring convergence is appropriately achieved, as stated in Section 5.4.1.1 of the TR (i.e.,

TR-0716-50350, Revision 3).

Section 5.4.1.1 of the TR describes several considerations taken in VIPRE-01 calculations in order to capture the rapid power increase associated with the REA transient. ((

)) and is, therefore, acceptable.

Cycle-specific sensitivity studies on axial and radial nodalization are only performed when a deviation from the default nodalizations described in the statistical subchannel report are taken.

Based on NRC staff review and approval of these nodalizations in the statistical subchannel report TR-108601-P, Revision 4 (Reference 13), as well as the sensitivity results presented in Section 6 of the TR, and the REA methodology requirement to perform sensitivity studies when non-default nodalizations are used, the NRC staff finds these default nodalizations acceptable.

The REA methodology requires that applicants and licensees perform other mandatory sensitivity studies to address the VIPRE-01 timestep and fuel heat transfer inputs. Timestep selection affects the Courant number, which was discussed above. The VIPRE-01 timestep must also be sufficiently small to resolve the time-dependent power as calculated by SIMULATE-3K. The REA methodology requires the user to plot time-dependent power with SIMULATE-3K and VIPRE-01 timesteps and evaluate whether excellent agreement is achieved. Based on this, and on the preceding Courant limit discussion, the NRC staff finds treatment of VIPRE-01 timestep within the rod ejection methodology acceptable.

17 Cycle-specific sensitivity studies on fuel heat transfer inputs (such as gap conductance) are performed to ensure that different fuel failure criteria are evaluated conservatively. NuScale provided sensitivity study results in Section 6.3.6 of the TR which demonstrate that high fuel heat transfer cases can be bounding for MCHFR. However, maximizing heat transfer from the fuel may not be bounding for other fuel failure criteria, such as radial average fuel enthalpy rise.

Because fuel heat transfer inputs are varied to ensure that limiting biases are identified for each fuel failure criterion, the NRC staff finds this sensitivity study acceptable.

Table 5-1 of the TR provides a list of the parameters that need to account for uncertainties for each time a rod ejection analysis is performed. As described in Section 5.2.2.3.4 of the TR, the uncertainties defined in Table 5-1 must be verified to ensure they are current and consistent with References 8.2.6, Nuclear Analysis Codes and Methods Qualification, TR-0616-48793-P-A, Revision 1, and 8.2.10, Statistical Subchannel Analysis Methodology, TR-108601-P, Revision 4. This is consistent with Limitation and Condition 3 identified in Section 6 of this SER.

5 CONCLUSIONS Based upon the NRC staffs review, as discussed above, the NRC staff concludes that TR-0716-50350-P, Revision 3 provides a systematic methodology for performing REA analysis.

This conclusion is based on the following, as summarized below:

The applicant uses GDC 28 requirements for prevention of postulated reactivity accidents that could result in damage to the reactor coolant pressure boundary greater than limited local yielding or result in sufficient damage to impair the core cooling capability significantly. The REA methodology is developed with the acceptance criteria provided in the regulatory guidance of RG 1.236. The staff has evaluated the applicants methodology for analyzing of the assumed control REA and finds the assumptions, calculation techniques, and consequences acceptable.

The staff finds that acceptance criteria for fuel enthalpy and fuel enthalpy rise under a REA to be conservative for preventing fuel damage. The staff determined that the calculational procedures are clear and the acceptance criteria are sufficiently conservative, both in initial assumptions and analytical models, to maintain primary system integrity subject to the limitations and conditions listed in Section 6 of this SER. On these bases, the staff determined that the methodology presented in the TR is acceptable for REA analyses.

6 LIMITATIONS AND CONDITIONS The staffs approval is limited to the application of this methodology to the NuScale reactor design with the following limitations and conditions:

1. An applicant or licensee referencing this report is required to demonstrate the applicability of the REA methodology to the specific NPM design. The use of this methodology for a specific NPM design requires the NRC staff review and approval of the applicant or licensee determination of applicability.

2. The REA methodology is limited to evaluation of REAs for fuel that has not experienced significant depletion with control rods inserted, such as from non-baseload operation.

18

3. The staffs approval is limited to the use of the REA methodology with TR-0616-48793-P-A, Revision 1 (Reference 14), Nuclear Analysis Codes and Methods Qualification, and TR-108601-P, Revision 4 (Reference 13), Statistical Subchannel Analysis Methodology, Supplement 1 to TR-0915-17564-P-A, Revision 2, Subchannel Analysis Methodology.

19 7

REFERENCES

1.

NuScale Power, LLC, TR-0716-50350-P, Rod Ejection Accident Methodology, Revision 2, December 17, 2022, ADAMS Accession Nos. ML21351A399 (public) and ML21351A400 (non-public).

2.

NuScale Power, LLC, Response to RAI No. 9936, September 14, 2022, ADAMS Accession Nos. ML22257A187 (package) and ML22257A188 (public version).

3.

U.S. Nuclear Regulatory Commission, Memorandum from Joseph, S.K., NRC, to Jardaneh, M., NRC Audit Summary for the Regulatory Audit of NuScale Power Topical Report Supplement Entitled Statistical Analysis Subchannel Methodology, TR-108601, Revision 1 Incorporating a Limited Scope Audit for Rod Ejection Accident Methodology, TR-0716-50350, Revision 2, November 1, 2023 (ML23295A001).

4.

NuScale Power, LLC, TR-0716-50350-P, Rod Ejection Accident Methodology, Revision 3, October 20, 2023, ADAMS Accession No. ML23293A291 (package) and ML23293A292 (public version).

5.

U.S. Nuclear Regulatory Commission, Regulatory Guide 1.236, Pressurized-Water Reactor Control Rod Ejection and Boiling-Water Reactor Control Rod Drop Accidents, Revision 0, June 2020, ADAMS Accession No. ML20055F490.

6.

U.S. Nuclear Regulatory Commission, NUREG-0800, NUREG-0800 - Chapter 15, Section 15.4.8, Revision 3, Spectrum of Rod Ejection Accidents (PWR), March 2007, ADAMS Accession No. ML070550014.

7.

U.S. Nuclear Regulatory Commission, NUREG-0800, NUREG-0800 - Chapter 4, Section 4.2, Revision 3, Fuel System Design, March 2007, ADAMS Accession No. ML070740002.

8.

Studsvik Scandpower, SSP-07/431, CASMO5: A Fuel Assembly Burnup Program Users Manual, Revision 7, December 2013.

9.

Studsvik Scandpower, SSP-10/438, SIMULATE5 Advanced Three-Dimensional Multigroup Reactor Analysis Code, Revision 4, December 2013.

10.

Studsvik Scandpower, SSP-98/13, SIMULATE-3K Models and Methodology, Revision 9, September 2013.

11.

NuScale Power, LLC, TR-0516-49416-P, Non-Loss-of-Coolant Accident Analysis Methodology, Revision 4, December 2022, ADAMS Accession Nos. ML23005A304 (package) and ML23005A305 (public version).

12.

NuScale Power, LLC, TR-0915-17564-P-A, Subchannel Analysis Methodology, Revision 2, March 2019, ADAMS Accession Nos. ML19067A255 (package) and ML19067A256 (public version).

Commented [SJ2]: The NRC review of TR051649416-P is not yet complete. Staff will withhold final approval of this TR until approval of the LOCA and Non-LOCA topical reports currently under review.

20

13.

NuScale Power, LLC, TR-108601-P Statistical Subchannel Analysis Methodology, Supplement 1 to TR-0915-17564-P-A, Revision 2, Subchannel Analysis Methodology, Revision 4, November 6, 2023, ADAMS Accession Nos. ML23310A121 (package) and ML23310A122 (public version).

14.

NuScale Power, LLC, TR-0616-48793-P-A, Nuclear Analysis Codes and Methods Qualification, Revision 1, December 14, 2018, ADAMS Accession Nos. ML18348B035 (package) and ML18348B036 (public version).

15.

AREVA Inc., BAW-10231-P-A, COPERNIC Fuel Rod Design Computer Code, Revision 1, January 2004, ADAMS Accession Nos. ML042930233 (package),

ML042930236, ML042930240, and ML042930247 (public version).

16.

NuScale Power, LLC, TR-0716-50350-P-A, Rod Ejection Accident Methodology, Revision 1, June 2020, ADAMS Accession Nos. ML20168B202 (package) and ML20168B203 (public version).

17.

U.S. Nuclear Regulatory Commission, Safety Evaluation by the Office of Nuclear Reactor Regulation Relating to VIPRE-01 Mod 02 for PWR and BWR Applications, EPRI-NP-2511-CCMA, Revision 3, October 30, 1993, ADAMS Accession No. ML102100004 (non-public).

18.

U.S. Nuclear Regulatory Commission, Regulatory Guide 1.203, Transient and Accident Analysis Methods, Revision 0, December 2005, ADAMS Accession No. ML053500170.

19.

U.S. Nuclear Regulatory Commission, Safety Evaluation Report on EPRI NP-2511CCM VIPRE-01, May 1986, ADAMS Accession No. ML18033A075 (public version).