ML20010D133: Difference between revisions

From kanterella
Jump to navigation Jump to search
(StriderTol Bot insert)
 
(StriderTol Bot change)
Line 65: Line 65:


8
8
{{                                         }} that are not standard for use inside nuclear reactors. The NRC staff that are reviewing DCD Chapter 7 have issued RAIs which may have implications for this LOCA PIRT.
((                                         }} that are not standard for use inside nuclear reactors. The NRC staff that are reviewing DCD Chapter 7 have issued RAIs which may have implications for this LOCA PIRT.
1.2    NRELAP5 Code The NuScale NRELAP5 systems code is the Idaho National Laboratory (INL) RELAP5-3D (Reactor Excursion and Leak Analysis Program-Three Dimensional) computer code version v.4.1.3 with commercial grade dedication and name change to NRELAP5 v0.0. NuScale subsequently made many modifications and additions to develop the baseline version for the LOCA TR, i.e., NRELAP5 v1.3. Each coding change is documented in an NCI (NRELAP5 change implementation) that is documented under a SwRN (software release notice). The NCIs review in detail, the coding changes made and contain specific benchmarks and regression testing needed to confirm that new code modeling is performing as intended. The scope of this audit did not encompass detailed examination of all coding NCIs but a subset directly affecting LOCA phenomena and FOMs. The validation and verification of the NRELAP5 code should be driven by the phenomena PIRT process as described by RG 1.203, which identifies the important and key phenomena to be modeled by this evaluation model (EM) for the LOCA event. Since the NuScale Power Module (NPM) is designed to reduce the consequences of LOCA events, as compared to traditional pressurized water reactors, many of the phenomena that are the subject of 10 CFR Part 50, Appendix K requirements are not reached in the NPM LOCAs. They are essentially considered to be designed out or satisfied by design rather than by analysis. A list of these designed out features are described in Section 2.2.2 and Table 2-2 of the LOCA TR (Reference 2).
1.2    NRELAP5 Code The NuScale NRELAP5 systems code is the Idaho National Laboratory (INL) RELAP5-3D (Reactor Excursion and Leak Analysis Program-Three Dimensional) computer code version v.4.1.3 with commercial grade dedication and name change to NRELAP5 v0.0. NuScale subsequently made many modifications and additions to develop the baseline version for the LOCA TR, i.e., NRELAP5 v1.3. Each coding change is documented in an NCI (NRELAP5 change implementation) that is documented under a SwRN (software release notice). The NCIs review in detail, the coding changes made and contain specific benchmarks and regression testing needed to confirm that new code modeling is performing as intended. The scope of this audit did not encompass detailed examination of all coding NCIs but a subset directly affecting LOCA phenomena and FOMs. The validation and verification of the NRELAP5 code should be driven by the phenomena PIRT process as described by RG 1.203, which identifies the important and key phenomena to be modeled by this evaluation model (EM) for the LOCA event. Since the NuScale Power Module (NPM) is designed to reduce the consequences of LOCA events, as compared to traditional pressurized water reactors, many of the phenomena that are the subject of 10 CFR Part 50, Appendix K requirements are not reached in the NPM LOCAs. They are essentially considered to be designed out or satisfied by design rather than by analysis. A list of these designed out features are described in Section 2.2.2 and Table 2-2 of the LOCA TR (Reference 2).
1.2.1  Numerics During the audit of the NuScale LOCA EM, the NRC staff examined the NRELAP5 theory manual given in SwUM-0304-17023, Revision 4, regarding code model and correlation modifications made to NRELAP5 for application of the code assessments to validation predictions of NPM ECCS performance. This audit addressed only NuScale core changes starting from the INL RELAP5-3D code version v.4.1.3 as described in INEEL-EXT-98-00834, dated September 2013. The NRC staff was made aware that a developmental assessment of the RELAP5-3D computer code was performed by INL (INL/MIS-15-36723, RELAP5-3D Code Manual Volume III: Developmental Assessment, October 2015). This assessment used a combination of phenomenological, separate effects, and integral effects test cases to validate the codes performance. This recent review of the code by INL (based on RELAP5-3D version 4.3, released October 2015) adds pedigree to the version obtained by NuScale, although the NRELAP5 code is taken from Version 4.1.3 (2012). Because the NRC staff has identified generic code performance issues with past versions of RELAP5, NuScale was requested to respond to several issues during the audit. Below is a summary of the issues, interactions with NuScale and the NRC staffs observations during the audit.
1.2.1  Numerics During the audit of the NuScale LOCA EM, the NRC staff examined the NRELAP5 theory manual given in SwUM-0304-17023, Revision 4, regarding code model and correlation modifications made to NRELAP5 for application of the code assessments to validation predictions of NPM ECCS performance. This audit addressed only NuScale core changes starting from the INL RELAP5-3D code version v.4.1.3 as described in INEEL-EXT-98-00834, dated September 2013. The NRC staff was made aware that a developmental assessment of the RELAP5-3D computer code was performed by INL (INL/MIS-15-36723, RELAP5-3D Code Manual Volume III: Developmental Assessment, October 2015). This assessment used a combination of phenomenological, separate effects, and integral effects test cases to validate the codes performance. This recent review of the code by INL (based on RELAP5-3D version 4.3, released October 2015) adds pedigree to the version obtained by NuScale, although the NRELAP5 code is taken from Version 4.1.3 (2012). Because the NRC staff has identified generic code performance issues with past versions of RELAP5, NuScale was requested to respond to several issues during the audit. Below is a summary of the issues, interactions with NuScale and the NRC staffs observations during the audit.
Line 78: Line 78:
The NRC staff also noted that Table 7-1 in TR-0516-49416-P appears to describe a list of parameters an analyst would use to specify the appropriate set of initial conditions and how the analyst would determine that a unique steady-state solution was obtained for an analysis. The NRC staff issued RAI 9374, Question 15.00.02-25, requesting NuScale to clearly state the purpose of Table 7-1 in TR-0516-49416-P and to revise Table 7-1 and the associated text in TR Section 7.1.1.2, accordingly.
The NRC staff also noted that Table 7-1 in TR-0516-49416-P appears to describe a list of parameters an analyst would use to specify the appropriate set of initial conditions and how the analyst would determine that a unique steady-state solution was obtained for an analysis. The NRC staff issued RAI 9374, Question 15.00.02-25, requesting NuScale to clearly state the purpose of Table 7-1 in TR-0516-49416-P and to revise Table 7-1 and the associated text in TR Section 7.1.1.2, accordingly.


10 1.2.3    Containment Condensation Model During the audit, the NRC staff noted that NuScale included additional correlations and capabilities in the NRELAP5 code to handle condensation heat transfer to the inside CNV wall with a CNV design pressure up to 1000 psia. NuScale added {{
10 1.2.3    Containment Condensation Model During the audit, the NRC staff noted that NuScale included additional correlations and capabilities in the NRELAP5 code to handle condensation heat transfer to the inside CNV wall with a CNV design pressure up to 1000 psia. NuScale added ((
                 }} condensation heat transfer correlations to the NRELAP5 code. However, the use of {{                         }} correlations for NuScale CNV condensation are outside the applicability range of this empirical correlation as it was developed for vertical pipes with small diameters and low-pressure conditions. In addition, the NRC staff noted that {{
                 }} condensation heat transfer correlations to the NRELAP5 code. However, the use of ((                         }} correlations for NuScale CNV condensation are outside the applicability range of this empirical correlation as it was developed for vertical pipes with small diameters and low-pressure conditions. In addition, the NRC staff noted that ((
                                                                                 }} and it was not clear whether the overall condensation heat transfer was being predicted conservatively. Because the steam condensation in CNV is of major importance to the prediction of ECCS performance for the NuScale design and the descriptions of condensation heat transfer correlations in the NuScale LOCA EM was unclear, the NRC staff requested additional information regarding these modifications. The NRC staff issued RAI 8990, Question 15.06.05-7, requesting NuScale to provide information related to this issue.
                                                                                 }} and it was not clear whether the overall condensation heat transfer was being predicted conservatively. Because the steam condensation in CNV is of major importance to the prediction of ECCS performance for the NuScale design and the descriptions of condensation heat transfer correlations in the NuScale LOCA EM was unclear, the NRC staff requested additional information regarding these modifications. The NRC staff issued RAI 8990, Question 15.06.05-7, requesting NuScale to provide information related to this issue.
(i)      {{                                   }}, designed for flow in pipes, is being used beyond the stated range of applicability of the hydraulic diameter. The NRC staff also questioned
(i)      ((                                   }}, designed for flow in pipes, is being used beyond the stated range of applicability of the hydraulic diameter. The NRC staff also questioned
{{                                                                   }} to this application for the NuScale design, which is not a pipe.
((                                                                   }} to this application for the NuScale design, which is not a pipe.
(ii)    The NRC staff observed through audit of the NRELAP5 theory manual SwUM-0304-17023 and the audit discussions, that there is a lack of clarity in NuScales understanding of the performance of the wall condensation model as well as a lack of key sensitivity studies showing the effects of differing condensation coefficients and/or variations in the heat removal rate by the CNV metal walls and how the correlations were implemented in the code subroutines. Sensitivity studies are necessary to establish a conservative LOCA EM analysis of ECCS function including CNV pressure response and its impact of RRV activation and the IAB differential pressure set point.
(ii)    The NRC staff observed through audit of the NRELAP5 theory manual SwUM-0304-17023 and the audit discussions, that there is a lack of clarity in NuScales understanding of the performance of the wall condensation model as well as a lack of key sensitivity studies showing the effects of differing condensation coefficients and/or variations in the heat removal rate by the CNV metal walls and how the correlations were implemented in the code subroutines. Sensitivity studies are necessary to establish a conservative LOCA EM analysis of ECCS function including CNV pressure response and its impact of RRV activation and the IAB differential pressure set point.
In RAI 8990, NuScale was asked to provide detailed numerical calculation procedures showing how {{                                   }} correlations are actually used in the NRELAP5 code.
In RAI 8990, NuScale was asked to provide detailed numerical calculation procedures showing how ((                                   }} correlations are actually used in the NRELAP5 code.
1.2.4    NRELAP5 Input Manual The NRC 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 to develop input models for application of the methodology. The NRC 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.
1.2.4    NRELAP5 Input Manual The NRC 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 to develop input models for application of the methodology. The NRC 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.
1.3    LOCA Evaluation Model The NRC staff observed that the NPM LOCA model is developed from the NRELAP5 base deck that is designed for general Chapter 15 safety analysis applications. The LOCA nodalization is downsized considerably from that in the base deck except for the core modeling, where a hot channel is added. Also {{                                             }}. Most other changes were related to eliminating details in the secondary and feedwater systems that NuScale asserted as not necessary for LOCA applications.
1.3    LOCA Evaluation Model The NRC staff observed that the NPM LOCA model is developed from the NRELAP5 base deck that is designed for general Chapter 15 safety analysis applications. The LOCA nodalization is downsized considerably from that in the base deck except for the core modeling, where a hot channel is added. Also ((                                             }}. Most other changes were related to eliminating details in the secondary and feedwater systems that NuScale asserted as not necessary for LOCA applications.


11 The NRC staff confirmed that the model addresses the key components of the NPM for LOCA transients, including: RPV internals, containment, steam generator primary and secondary sides, the reactor pool and the ECCS system. The reactor core is modeled with three parallel channels: (1) the average channel representing the remaining 36 fuel assemblies, (2) the hot channel representing a single NPM fuel assembly and (3) the core bypass that includes the reactor reflector channel bypass and the fuel assembly instrument and guide tube bypass.
11 The NRC staff confirmed that the model addresses the key components of the NPM for LOCA transients, including: RPV internals, containment, steam generator primary and secondary sides, the reactor pool and the ECCS system. The reactor core is modeled with three parallel channels: (1) the average channel representing the remaining 36 fuel assemblies, (2) the hot channel representing a single NPM fuel assembly and (3) the core bypass that includes the reactor reflector channel bypass and the fuel assembly instrument and guide tube bypass.
1.3.1    NPM Core Model The reactor core is modeled with three parallel channels as indicated above. The loss coefficients for {{
1.3.1    NPM Core Model The reactor core is modeled with three parallel channels as indicated above. The loss coefficients for ((
                                                           }}. The NRC staff audited the detailed calculation procedures for these coefficients and had no significant issues related to the initial blowdown and establishment of the initial ECCS recirculation through the reactor vent valves (RVVs) and reactor recirculation valves (RRVs). The longer-term concerns of this modeling were previously discussed in Section 7.2.1 regarding flow anomalies in parallel piping channels.
                                                           }}. The NRC staff audited the detailed calculation procedures for these coefficients and had no significant issues related to the initial blowdown and establishment of the initial ECCS recirculation through the reactor vent valves (RVVs) and reactor recirculation valves (RRVs). The longer-term concerns of this modeling were previously discussed in Section 7.2.1 regarding flow anomalies in parallel piping channels.
The NRC staff observed that the core fuel rods use {{                                       }}, which invokes a NuScale-specific CHF correlation package that uses: {{(
The NRC staff observed that the core fuel rods use ((                                       }}, which invokes a NuScale-specific CHF correlation package that uses: (((
               }}. The applicant used benchmarks to the subchannel topical report (Reference 7) to show that this critical heat flux (CHF) package yields conservative results. The decay heat model used is ANS 73 with a 1.2 multiplier that is shown to be in compliance of 10 CFR 50.46 Appendix K requirement. NuScale is requesting an exemption to use ANS 73 rather than Appendix K ANS 71.
               }}. The applicant used benchmarks to the subchannel topical report (Reference 7) to show that this critical heat flux (CHF) package yields conservative results. The decay heat model used is ANS 73 with a 1.2 multiplier that is shown to be in compliance of 10 CFR 50.46 Appendix K requirement. NuScale is requesting an exemption to use ANS 73 rather than Appendix K ANS 71.
The NRC staff audited the implementation of these models and the supporting calculations. No significant issues were identified in these models. However, the NRC staff noted that in FSAR, Section 15.6.6., which also describes {{                                                         }},
The NRC staff audited the implementation of these models and the supporting calculations. No significant issues were identified in these models. However, the NRC staff noted that in FSAR, Section 15.6.6., which also describes ((                                                         }},
that the CHF ratio (CHFR) limit was not being consistently stated. As such, the NRC staff issued RAI 9536 to request the applicant to provide documentation that consistently reflects the final CHFR limits.
that the CHF ratio (CHFR) limit was not being consistently stated. As such, the NRC staff issued RAI 9536 to request the applicant to provide documentation that consistently reflects the final CHFR limits.
1.3.2    Pressurizer Model The pressurizer baffle plate and the fuel assembly upper tie plate are considered to a have potential for counter current flow limitation (CCFL) and therefore the junctions are modeled with
1.3.2    Pressurizer Model The pressurizer baffle plate and the fuel assembly upper tie plate are considered to a have potential for counter current flow limitation (CCFL) and therefore the junctions are modeled with
{{                         }} activated based on the plate geometry. The NRC staff raised the issue about how {{                           }} is being validated and used at the baffle plate location. The detailed supporting documents showed that the use of this correlation has insignificant impact on the two-phase water level above the core. Therefore, the NRC staff determined that the issue was not significant and doesnt need to be addressed.
((                         }} activated based on the plate geometry. The NRC staff raised the issue about how ((                           }} is being validated and used at the baffle plate location. The detailed supporting documents showed that the use of this correlation has insignificant impact on the two-phase water level above the core. Therefore, the NRC staff determined that the issue was not significant and doesnt need to be addressed.
1.3.3    Steam Generator Model
1.3.3    Steam Generator Model


12 The two helical coil SGs are modeled using a NuScale-specific helical coil SG component that uniquely addresses the flow geometry and heat transfer effects. This new hydrodynamic component and heat transfer package were added to NRELAP5. The modeling makes specific heat transfer and wall friction changes based on the SIET test data. The NRC staff found no significant issues with the actual SG model implemented in the NRELAP5 code. However, the NRC staff noted inconsistencies regarding specific SIET test configuration and results, which may call in question the heat transfer modeling changes developed.
12 The two helical coil SGs are modeled using a NuScale-specific helical coil SG component that uniquely addresses the flow geometry and heat transfer effects. This new hydrodynamic component and heat transfer package were added to NRELAP5. The modeling makes specific heat transfer and wall friction changes based on the SIET test data. The NRC staff found no significant issues with the actual SG model implemented in the NRELAP5 code. However, the NRC staff noted inconsistencies regarding specific SIET test configuration and results, which may call in question the heat transfer modeling changes developed.
1.3.4  Containment Model The CNV surrounds the RPV and the NRC staff noted that it is being modeled as a vertical PIPE located in an annulus between the outer RPV wall and the CNV inner wall. The CNV retains steam and liquid in the event of a LOCA, such that after ECCS activation, discharged inventory can be returned to the core via the RRVs. The volume of the CNV is just over 2.4 times the volume of the RPV. Each NPM is mounted with a segmented bay within a vast reactor pool. The pool area relevant for each NPM was assumed to be {{
1.3.4  Containment Model The CNV surrounds the RPV and the NRC staff noted that it is being modeled as a vertical PIPE located in an annulus between the outer RPV wall and the CNV inner wall. The CNV retains steam and liquid in the event of a LOCA, such that after ECCS activation, discharged inventory can be returned to the core via the RRVs. The volume of the CNV is just over 2.4 times the volume of the RPV. Each NPM is mounted with a segmented bay within a vast reactor pool. The pool area relevant for each NPM was assumed to be ((
                                         }}. Although the numerical nodalization of the CNV and the reactor pool appeared to be reasonable, the NRC staff identified issues in the following areas:
                                         }}. Although the numerical nodalization of the CNV and the reactor pool appeared to be reasonable, the NRC staff identified issues in the following areas:
* NRELAP5 condensation modeling applied.
* NRELAP5 condensation modeling applied.
Line 112: Line 112:


13 pressure, however finer noding was not addressed. The NRC staff believes that finer noding in the liquid region could produce a greater inventory of subcooled water leading to less condensation and a higher computed peak containment pressure. The NRC staff issued RAI 9380, Question 06.02.01.01.A-5, requesting that NuScale provide additional information to enable the NRC staff to assess the impact of liquid thermal stratification and nodalization in the CNV liquid region.
13 pressure, however finer noding was not addressed. The NRC staff believes that finer noding in the liquid region could produce a greater inventory of subcooled water leading to less condensation and a higher computed peak containment pressure. The NRC staff issued RAI 9380, Question 06.02.01.01.A-5, requesting that NuScale provide additional information to enable the NRC staff to assess the impact of liquid thermal stratification and nodalization in the CNV liquid region.
Modeling Containment and Pool During the audit, the NRC staff requested that NuScale justify its NRELAP5 one dimensional model of the containment, effects of non-condensibles, and ultimate heat sink pool. The NuScale NPM model employs one-dimensional PIPE components to model both containment and {{                 }} the cooling pool outside of each containment module. As a result, the pool water within each NRELAP5 PIPE node is numerically predicted to sense a uniform bulk temperature increase during a LOCA. The NRC staff questioned the impact on the LOCA analysis of this simplified approach. The NRC staff and NuScale agreed in audit discussions that layers of convection currents would develop near the outer CNV wall that cannot be modeled by NRELAP5, which would result in warmer temperatures being in direct contact with the CNV wall.
Modeling Containment and Pool During the audit, the NRC staff requested that NuScale justify its NRELAP5 one dimensional model of the containment, effects of non-condensibles, and ultimate heat sink pool. The NuScale NPM model employs one-dimensional PIPE components to model both containment and ((                 }} the cooling pool outside of each containment module. As a result, the pool water within each NRELAP5 PIPE node is numerically predicted to sense a uniform bulk temperature increase during a LOCA. The NRC staff questioned the impact on the LOCA analysis of this simplified approach. The NRC staff and NuScale agreed in audit discussions that layers of convection currents would develop near the outer CNV wall that cannot be modeled by NRELAP5, which would result in warmer temperatures being in direct contact with the CNV wall.
The NRC staff issued RAI 8776, Question 15.06.05-2, requesting that NuScale provide additional justification for the modeling used and to evaluate the impact on the FOMs, i.e., on peak containment pressure and collapsed water level above top of active fuel (TAF) for the limiting LOCA cases.
The NRC staff issued RAI 8776, Question 15.06.05-2, requesting that NuScale provide additional justification for the modeling used and to evaluate the impact on the FOMs, i.e., on peak containment pressure and collapsed water level above top of active fuel (TAF) for the limiting LOCA cases.
1.3.5    LOCA Initial Conditions The assumptions for conservatively biasing the initial plant conditions were based on Table 5-6 of the LTR and the NRC staff determined that the initial plant conditions were not consistent with the uncertainty values given in Table 15.0-6 of the FSAR. The NRC staff audited the LOCA calculation supporting the EM, EC-0000-4888 NuScale LOCA Evaluation Model Supporting Calculations, and noted that Tave used was lower than that in the table. The following two scenarios for loss of power are considered: (1) complete loss of normal alternating current (AC) and DC power and (2) loss of only AC power with DC power availability. The NRC staff issued RAI 9475, Question 15.06.05-17, requesting that NuScale describe the method and bias for selecting the maximum value and a sample upper bound value for RCS average temperature, riser temperature, pressurizer pressure, main steam pressure and feedwater temperature, and reactor pool temperature.
1.3.5    LOCA Initial Conditions The assumptions for conservatively biasing the initial plant conditions were based on Table 5-6 of the LTR and the NRC staff determined that the initial plant conditions were not consistent with the uncertainty values given in Table 15.0-6 of the FSAR. The NRC staff audited the LOCA calculation supporting the EM, EC-0000-4888 NuScale LOCA Evaluation Model Supporting Calculations, and noted that Tave used was lower than that in the table. The following two scenarios for loss of power are considered: (1) complete loss of normal alternating current (AC) and DC power and (2) loss of only AC power with DC power availability. The NRC staff issued RAI 9475, Question 15.06.05-17, requesting that NuScale describe the method and bias for selecting the maximum value and a sample upper bound value for RCS average temperature, riser temperature, pressurizer pressure, main steam pressure and feedwater temperature, and reactor pool temperature.
Line 126: Line 126:
15 interfere with insertion. NuScale should provide a complete NPM analysis to confirm insertability of rods under LOCA plus seismic conditions.
15 interfere with insertion. NuScale should provide a complete NPM analysis to confirm insertability of rods under LOCA plus seismic conditions.
The NRC staff determined that this issue is beyond the scope of the LOCA topical report, however, it should be addressed by the NRC staff as a part of the LOCA 15.6.5 reviews, which will be supported by the NRC staff reviewers of DCD Sections 3.9.4 and 4.2.
The NRC staff determined that this issue is beyond the scope of the LOCA topical report, however, it should be addressed by the NRC staff as a part of the LOCA 15.6.5 reviews, which will be supported by the NRC staff reviewers of DCD Sections 3.9.4 and 4.2.
1.4    NIST Scaling and Distortion Analysis 1.4.1    Containment Vessel Scaling Section 8.3.2 of the LOCA LTR (TR-0516-49422-P) describes and provides conclusions on important phenomena and scaling distortions, based on NuScale report EE-T080-13757, Revision 2. The NRC staff identified several significant distortions related to NIST-1 containment vessel (CNV) scaling including: {{(
1.4    NIST Scaling and Distortion Analysis 1.4.1    Containment Vessel Scaling Section 8.3.2 of the LOCA LTR (TR-0516-49422-P) describes and provides conclusions on important phenomena and scaling distortions, based on NuScale report EE-T080-13757, Revision 2. The NRC staff identified several significant distortions related to NIST-1 containment vessel (CNV) scaling including: (((
                           }}.
                           }}.
It appeared to the NRC staff that the applicant only partially addressed these distortions and did not address the inter-relationship among the significant distortions clearly. The NRC staff was concerned that the existing distortion reports, which the NRC staff audited, may not adequately capture all the high-ranking important phenomena in the PIRT, nor correctly quantify the distortions to determine the uncertainty in the figure of merit (FOM) prediction. The NRC staff issued RAI 9208, Questions 15.06.05-14 through 15.06.05-16, requesting NuScale provide information on similarity groups (PI groups), the H2TS scaling methodology referenced in Section 8.3.2 of the LOCA LTR, heat transfer resistance sensitivity study conclusions, and potential NIST-1 test facility distortions which the NRC staff believes could cause an over prediction of NPM condensation.
It appeared to the NRC staff that the applicant only partially addressed these distortions and did not address the inter-relationship among the significant distortions clearly. The NRC staff was concerned that the existing distortion reports, which the NRC staff audited, may not adequately capture all the high-ranking important phenomena in the PIRT, nor correctly quantify the distortions to determine the uncertainty in the figure of merit (FOM) prediction. The NRC staff issued RAI 9208, Questions 15.06.05-14 through 15.06.05-16, requesting NuScale provide information on similarity groups (PI groups), the H2TS scaling methodology referenced in Section 8.3.2 of the LOCA LTR, heat transfer resistance sensitivity study conclusions, and potential NIST-1 test facility distortions which the NRC staff believes could cause an over prediction of NPM condensation.
1.4.2    Power Level Scaling The NRC staff noted that NIST-1 scaling, per EE-T080-13757, Revision 2, is based {{
1.4.2    Power Level Scaling The NRC staff noted that NIST-1 scaling, per EE-T080-13757, Revision 2, is based ((
                                             }}. The NRC staff also noted that the testing results for various power levels for the HP-05 test did not compare well with the NRELAP coolant temperature predictions for those power levels. NuScale is requested to address this distortion regarding its implication for peak containment pressure estimation. The NRC staff issued RAI 9208, Question 15.06.05-15, requesting NuScale address distortion regarding its implication for important figures of merit.
                                             }}. The NRC staff also noted that the testing results for various power levels for the HP-05 test did not compare well with the NRELAP coolant temperature predictions for those power levels. NuScale is requested to address this distortion regarding its implication for peak containment pressure estimation. The NRC staff issued RAI 9208, Question 15.06.05-15, requesting NuScale address distortion regarding its implication for important figures of merit.
1.4.3    NIST-1 Riser Scaling Also, in the scaling document, EE-T080-13757, Revision 2, staff noticed that the NIST-1 riser is scaled to {{                                                               }}. No basis is documented in the scaling report for choosing this {{             }}. This value {{
1.4.3    NIST-1 Riser Scaling Also, in the scaling document, EE-T080-13757, Revision 2, staff noticed that the NIST-1 riser is scaled to ((                                                               }}. No basis is documented in the scaling report for choosing this ((             }}. This value ((
                                                                                   }}. This could also affect the flow regime transition and therefore potentially distort the RPV level and RPV depressurization. Considering little Reactor Pressure Vessel (RPV) level margin following a small break LOCA, the distortion of Jg introduced in the current scaling method may adversely impact the RPV level prediction directly and containment pressure indirectly. With the current
                                                                                   }}. This could also affect the flow regime transition and therefore potentially distort the RPV level and RPV depressurization. Considering little Reactor Pressure Vessel (RPV) level margin following a small break LOCA, the distortion of Jg introduced in the current scaling method may adversely impact the RPV level prediction directly and containment pressure indirectly. With the current


Line 138: Line 138:
1.4.4    Initial Steady State Scaling The NRC staff noted significant distortions in initial temperature and pressure conditions due to design limitations of NIST-1. In the scaling distortion report (EC-0000_3853_R1), the applicant attempts to quantify the effect of the scaling distortions by performing sensitivity calculations for NPM and NIST-1 configuration. In the course of its audit, the NRC staff identified several discrepancies in the scaling methodology used. These issues are addressed in RAI 9208 Question 15.06.05-16.
1.4.4    Initial Steady State Scaling The NRC staff noted significant distortions in initial temperature and pressure conditions due to design limitations of NIST-1. In the scaling distortion report (EC-0000_3853_R1), the applicant attempts to quantify the effect of the scaling distortions by performing sensitivity calculations for NPM and NIST-1 configuration. In the course of its audit, the NRC staff identified several discrepancies in the scaling methodology used. These issues are addressed in RAI 9208 Question 15.06.05-16.
1.4.5    Integral Estimate of Scaling Uncertainty Due to Distortion The NRC staff noted that the purpose of the NIST-1 facility is to provide realistic data for evaluation model validation. The NRC staff noted scaling overall and discussion of scaling distortions were not adequately addressed within the LOCA EM, and the NRC staff needs to better presentation of how these discrepancies and distortions contribute to the LOCA figures of merit (e.g. reactor level, containment pressure) quantitatively. These NRC staff requests and questions are addressed in RAI 9390, Question 15.06.05-19.
1.4.5    Integral Estimate of Scaling Uncertainty Due to Distortion The NRC staff noted that the purpose of the NIST-1 facility is to provide realistic data for evaluation model validation. The NRC staff noted scaling overall and discussion of scaling distortions were not adequately addressed within the LOCA EM, and the NRC staff needs to better presentation of how these discrepancies and distortions contribute to the LOCA figures of merit (e.g. reactor level, containment pressure) quantitatively. These NRC staff requests and questions are addressed in RAI 9390, Question 15.06.05-19.
1.4.6    Potential Distortion Due to {{                 }}
1.4.6    Potential Distortion Due to ((                 }}
The NRC staff noted that several of the NIST-1 tests required {{
The NRC staff noted that several of the NIST-1 tests required ((
                       }} and questioned the potential scaling distortion introduced by {{
                       }} and questioned the potential scaling distortion introduced by ((
                                 }}. The NRC staff needs to understand the basis for {{
                                 }}. The NRC staff needs to understand the basis for ((
               }}, the criteria used to determine this temperature, and the effect it has on the test results, including additional discussion regarding the associated scaling distortion. The NRC staff issued RAI 8777, Question 15.06.05-1 requesting NuScale describe the criteria and summarize the supporting analyses that form the basis for selecting the {{
               }}, the criteria used to determine this temperature, and the effect it has on the test results, including additional discussion regarding the associated scaling distortion. The NRC staff issued RAI 8777, Question 15.06.05-1 requesting NuScale describe the criteria and summarize the supporting analyses that form the basis for selecting the ((
               }} during the NIST-1 testing.
               }} during the NIST-1 testing.
1.4.7 Scaling Distortion Analysis Related to Containment Peak Pressure Measurement In the scaling distortion report (Calculations to Support NIST-1 Distortion Analysis and Modeling of Containment and Pool Heat Transfer, EC-0000-3853_R1), the applicant quantified the effect of the scaling distortions by performing sensitivity calculations for NPM and NIST-1 configuration. The NRC staff also noticed that Section 4.1 of the Containment Response Analysis Methodology Technical Report (CRAM TeR) (TR-0516-49084-P Revision
1.4.7 Scaling Distortion Analysis Related to Containment Peak Pressure Measurement In the scaling distortion report (Calculations to Support NIST-1 Distortion Analysis and Modeling of Containment and Pool Heat Transfer, EC-0000-3853_R1), the applicant quantified the effect of the scaling distortions by performing sensitivity calculations for NPM and NIST-1 configuration. The NRC staff also noticed that Section 4.1 of the Containment Response Analysis Methodology Technical Report (CRAM TeR) (TR-0516-49084-P Revision
: 0) addresses some scaling distortions for the primary and secondary system releases. In the course of its audit, the NRC staff identified several discrepancies that may affect peak containment pressure, and the NRC staff determined that the scaling and distortions also were not adequately addressed in the audited documents. These issues are captured in RAI 9494, Question 06.02.01.01.A-16.
: 0) addresses some scaling distortions for the primary and secondary system releases. In the course of its audit, the NRC staff identified several discrepancies that may affect peak containment pressure, and the NRC staff determined that the scaling and distortions also were not adequately addressed in the audited documents. These issues are captured in RAI 9494, Question 06.02.01.01.A-16.


17 1.5    NIST Test Facility and Testing Procedures 1.5.1    Heater Rods {{                     }}.
17 1.5    NIST Test Facility and Testing Procedures 1.5.1    Heater Rods ((                     }}.
On Page 217 of this TR, NuScale stated that the heater rods {{
On Page 217 of this TR, NuScale stated that the heater rods ((
                                               }}. During the Phase II Corvallis on-site audit, NuScale explained that {{
                                               }}. During the Phase II Corvallis on-site audit, NuScale explained that ((
                                                                                                   }}.
                                                                                                   }}.
However, the NRC staff noted that for the follow-on NIST-1 Test HP-49 assessment, NuScale revised the NRELAP5 heater rod modeling that significantly increased the thermal resistance and thereby the stored energy in each rod. This issue remains unresolved and is being addressed as a part of the next phase of the LOCA audit.
However, the NRC staff noted that for the follow-on NIST-1 Test HP-49 assessment, NuScale revised the NRELAP5 heater rod modeling that significantly increased the thermal resistance and thereby the stored energy in each rod. This issue remains unresolved and is being addressed as a part of the next phase of the LOCA audit.
1.5.2    {{                         }} for LOCA Tests
1.5.2    ((                         }} for LOCA Tests
{{
((
                                                 }}. RAI 8777, Question 15.06.05-1, {{(
                                                 }}. RAI 8777, Question 15.06.05-1, (((
                               )}} has been identified to investigate the issue of {{
                               )}} has been identified to investigate the issue of ((
                         }}.
                         }}.
1.5.3    Pressure Over-prediction of Condensation Test HP-02 The NRC staff noted in EC-T080-3822, Revision 1, HP-02 Assessment, that the computed results for this larger scale condensation test increasingly over-predicts pressure as the absolute pressure increases. This is the only larger separate effects test available to validate
1.5.3    Pressure Over-prediction of Condensation Test HP-02 The NRC staff noted in EC-T080-3822, Revision 1, HP-02 Assessment, that the computed results for this larger scale condensation test increasingly over-predicts pressure as the absolute pressure increases. This is the only larger separate effects test available to validate
{{                       }} condensation modeling in NRELAP5. NuScale reported in audit calls that the NRELAP5 code modeling did not consider heat loss. The NRC staff questioned the validity of the test for the condensation rate if heat losses from the CNV shell were not addressed. The NRC staff issued RAI 9317, Question 06.02.01.01.A-8, requesting NuScale provide additional NRELAP5 analyses that include heat losses and other phenomena that may have been neglected in the original analyses (Runs 1, 2, and 3) and to include plots that clearly show the impact of modeling changes.
((                       }} condensation modeling in NRELAP5. NuScale reported in audit calls that the NRELAP5 code modeling did not consider heat loss. The NRC staff questioned the validity of the test for the condensation rate if heat losses from the CNV shell were not addressed. The NRC staff issued RAI 9317, Question 06.02.01.01.A-8, requesting NuScale provide additional NRELAP5 analyses that include heat losses and other phenomena that may have been neglected in the original analyses (Runs 1, 2, and 3) and to include plots that clearly show the impact of modeling changes.
Natural Circulation Test HP-05 The NRC staff noted that NIST-1 Test HP-05 was a calibration test to refine hydraulic loss coefficients to improve the NRELAP5 computation of natural circulation flow rates observed in NIST-1. Loss coefficients for the NPM are to be estimated and refined from hydraulic tests on an actual plant after the NPM is built. No hydraulic mockup is anticipated. NuScale relied on the hydraulic analysis of the NPM to produce the expected natural circulation performance characteristics. Technical Report (EC-A030-2359) appears to indicate the RCS form losses, in the base deck (EC-A010-1782), are only based on theoretical formulations from Idelchik. The NRC staff could not locate any documented changes to the RCS NPM form losses that incorporated NIST-1 HP-05 flow data or observations. The NRC staff then noted some minor changes in form losses in the HCSG regions, but the changes are not documented in the LOCA calculations EC-0000-4888.
Natural Circulation Test HP-05 The NRC staff noted that NIST-1 Test HP-05 was a calibration test to refine hydraulic loss coefficients to improve the NRELAP5 computation of natural circulation flow rates observed in NIST-1. Loss coefficients for the NPM are to be estimated and refined from hydraulic tests on an actual plant after the NPM is built. No hydraulic mockup is anticipated. NuScale relied on the hydraulic analysis of the NPM to produce the expected natural circulation performance characteristics. Technical Report (EC-A030-2359) appears to indicate the RCS form losses, in the base deck (EC-A010-1782), are only based on theoretical formulations from Idelchik. The NRC staff could not locate any documented changes to the RCS NPM form losses that incorporated NIST-1 HP-05 flow data or observations. The NRC staff then noted some minor changes in form losses in the HCSG regions, but the changes are not documented in the LOCA calculations EC-0000-4888.


18 During the Phase 2 on-site audit, NuScale explained that the adjustments to the hydraulic loss coefficients were very small. They also stated that the major pressure loss is in the steam generator. The NRC staff agreed that the changes would have a very small effect on flow.
18 During the Phase 2 on-site audit, NuScale explained that the adjustments to the hydraulic loss coefficients were very small. They also stated that the major pressure loss is in the steam generator. The NRC staff agreed that the changes would have a very small effect on flow.
CVCS Discharge Pipe Break, NIST-1 Tests HP-06, HP-06b Decay Power after Trip It was stated in the NuScale distortion analysis report that HP-06 was conducted with {{
CVCS Discharge Pipe Break, NIST-1 Tests HP-06, HP-06b Decay Power after Trip It was stated in the NuScale distortion analysis report that HP-06 was conducted with ((
         }} of the scaled decay heat power. Because the decay heat is the only reason that ECCS needs to be designed to protect the core fuel integrity and the two-phase water level above the core, NuScale performed a new CVCS pipe break test using the full-scale decay heat power. This new test and HP-06 are both addressed in the LTR and are used for the validation.
         }} of the scaled decay heat power. Because the decay heat is the only reason that ECCS needs to be designed to protect the core fuel integrity and the two-phase water level above the core, NuScale performed a new CVCS pipe break test using the full-scale decay heat power. This new test and HP-06 are both addressed in the LTR and are used for the validation.
1.6    NRELAP5 Codes Assessment Cases 1.6.1    Edward Pipe Blowdown and LOFT test L2-5 The NRC staff noted during the audit, that the NuScale LOCA EM TR refers to Edward pipe and LOFT L2-5. However, the NRC staff noted that these assessments are not described in any part of the topical report and that they were not modeled to support the NRELAP5. NuScale personnel agreed to remove this sentence from a future revision of the NuScale LOCA EM topical report. The NRC staff will confirm that this modification is made to the TR.
1.6    NRELAP5 Codes Assessment Cases 1.6.1    Edward Pipe Blowdown and LOFT test L2-5 The NRC staff noted during the audit, that the NuScale LOCA EM TR refers to Edward pipe and LOFT L2-5. However, the NRC staff noted that these assessments are not described in any part of the topical report and that they were not modeled to support the NRELAP5. NuScale personnel agreed to remove this sentence from a future revision of the NuScale LOCA EM topical report. The NRC staff will confirm that this modification is made to the TR.
Line 179: Line 179:
: 2.      Helical coil SG component model with specific heat transfer and pressure drop models
: 2.      Helical coil SG component model with specific heat transfer and pressure drop models
: 3.      Appendix K ANS 1971 decay heat models
: 3.      Appendix K ANS 1971 decay heat models
: 4.      Appendix K {{                       }} CHF models
: 4.      Appendix K ((                       }} CHF models
: 5.      {{               }} Wall condensation models The NPM NRELAP5 design evaluation employs a combination of proven RELAP5 features, models, and components as well as new, untested, advanced operating features and components requiring new and unique correlations and models to simulate the needed conditions and behavior. Of particular importance is the use of the containment vessel as an integral part of the ECCS and the modeling of wall condensation under high pressure conditions. However, for LOCA, since the NPM should maintain adequate reverse flow in the RRVs such that collapsed liquid remains above the core throughout the transient, the specific condensation and CHF modeling should not be a significant issue. The review of NRELAP changes focused on the items 1 through 5 but included many other changes made to the code.
: 5.      ((               }} Wall condensation models The NPM NRELAP5 design evaluation employs a combination of proven RELAP5 features, models, and components as well as new, untested, advanced operating features and components requiring new and unique correlations and models to simulate the needed conditions and behavior. Of particular importance is the use of the containment vessel as an integral part of the ECCS and the modeling of wall condensation under high pressure conditions. However, for LOCA, since the NPM should maintain adequate reverse flow in the RRVs such that collapsed liquid remains above the core throughout the transient, the specific condensation and CHF modeling should not be a significant issue. The review of NRELAP changes focused on the items 1 through 5 but included many other changes made to the code.
Additionally, since the INLRELAP5-3D code is also currently under development, NuScale implements periodic updates and bug fixes related to the generic INL RELAP5-3D code as well as NuScale specific changes in accordance with its Appendix B quality control standards.
Additionally, since the INLRELAP5-3D code is also currently under development, NuScale implements periodic updates and bug fixes related to the generic INL RELAP5-3D code as well as NuScale specific changes in accordance with its Appendix B quality control standards.
The NRC staff audited NuScales NRELAP5 Version 1.3 Software Release Notes document SwRN-0304-51578 and the NRC staff decided to examine only the changes relevant for phenomena that are important for LOCA, including namely, (1), (3), and (4). The audit included the applicable NCIs (NRELAP5 Change Implementation) and NRELAP5 source coding was provided on a workstation in the NuScale Rockville office.
The NRC staff audited NuScales NRELAP5 Version 1.3 Software Release Notes document SwRN-0304-51578 and the NRC staff decided to examine only the changes relevant for phenomena that are important for LOCA, including namely, (1), (3), and (4). The audit included the applicable NCIs (NRELAP5 Change Implementation) and NRELAP5 source coding was provided on a workstation in the NuScale Rockville office.
Version                      Description                  NCI Beta 00.01                    Appendix K Moody            NCI-1214-9943 Based on the NRC staffs examination of SwRN-0304-51578, Revision 0, many of the early changes to the code are implemented by the INL staff, the original code developers. The NRC staff notes that INL has vast experience and knowledge of the code. The changes for Moody implementation are based on previous INL work, document LTD-14-3183. The changes included adding two additional options for choking models, i.e., c=3 {{(
Version                      Description                  NCI Beta 00.01                    Appendix K Moody            NCI-1214-9943 Based on the NRC staffs examination of SwRN-0304-51578, Revision 0, many of the early changes to the code are implemented by the INL staff, the original code developers. The NRC staff notes that INL has vast experience and knowledge of the code. The changes for Moody implementation are based on previous INL work, document LTD-14-3183. The changes included adding two additional options for choking models, i.e., c=3 (((
                                                                                           }}, and c=4 {{(
                                                                                           }}, and c=4 (((
                 )}}. The subroutine jchoke.F contains the coding for the implementation of the choking criterion. This implementation numerically imposes the choking criterion on the junctions determined to be in a choked state based on user input. The NRC staff examined changes made and the results of regression test cases. The preferred option is c=3 with the interpolation, which was used by NuScale for the LOCA analyses.
                 )}}. The subroutine jchoke.F contains the coding for the implementation of the choking criterion. This implementation numerically imposes the choking criterion on the junctions determined to be in a choked state based on user input. The NRC staff examined changes made and the results of regression test cases. The preferred option is c=3 with the interpolation, which was used by NuScale for the LOCA analyses.
Version                      Description                  NCI Beta 00.02                    ANS 1971 Decay Heat          NCI-0115-10047
Version                      Description                  NCI Beta 00.02                    ANS 1971 Decay Heat          NCI-0115-10047
Line 190: Line 190:
21 Beta 00.06                      ANS 1971 Decay Heat Fix      NCI-0115-10047 In the base RELAP5-3D code, a conservative approximation to the 1973 ANS Standard fission product data is used as the default. Appendix K to 10 CFR Part 50, indicates that heat generation rates from radioactive decay of fission products shall be assumed to be equal to 1.2 times the values for infinite operating time in the ANS Standard (Proposed American Nuclear Society Standards Decay Energy Release Rates Following Shutdown of Uranium-Fueled Thermal Reactors. Approved by Subcommittee ANS-5, ANS Standards Committee, issued October 1971). The subroutine dcyans71.F was added for the implementation of the ANS-1971 standard. The NRC staff examined the code changes against the referenced standard formulations and noted that NuScale has submitted an exemption request to instead use ANS-73 with a multiplier of 1.2.
21 Beta 00.06                      ANS 1971 Decay Heat Fix      NCI-0115-10047 In the base RELAP5-3D code, a conservative approximation to the 1973 ANS Standard fission product data is used as the default. Appendix K to 10 CFR Part 50, indicates that heat generation rates from radioactive decay of fission products shall be assumed to be equal to 1.2 times the values for infinite operating time in the ANS Standard (Proposed American Nuclear Society Standards Decay Energy Release Rates Following Shutdown of Uranium-Fueled Thermal Reactors. Approved by Subcommittee ANS-5, ANS Standards Committee, issued October 1971). The subroutine dcyans71.F was added for the implementation of the ANS-1971 standard. The NRC staff examined the code changes against the referenced standard formulations and noted that NuScale has submitted an exemption request to instead use ANS-73 with a multiplier of 1.2.
The specific code modification reviewed, focused on the following versions and NCIs:
The specific code modification reviewed, focused on the following versions and NCIs:
Version                Description                  NCI Beta 00.10                Critical Heat Flux (CHF)    NCI-0415-13361 v1.0RC2                  CHF final Revision 1        NCI-0415-13361 v1.1RC15                  {{             }} CHF        NCI-0316-48153 Correlation v1.1                      Version Release with CHF    NCI-0416-48924 Fix v1.2RC2                  CHF Update                  NCI-0616-49518 v1.3                      Implementation of CHF        NCI-0916-51506 NCRs The NuScale LOCA EM limits core heat transfer modeling to pre-CHF phenomena only. If a scenario can be postulated where CHF limits are exceeded during a LOCA, this EM would be invalidated. Therefore, the applicant must demonstrate that fuel does not closely approach or experience CHF condition, i.e., collapsed water levels must always remain above the top of active fuel, the containment remains intact, and pressure and temperature in the CNV remain below design limits. This allows the applicant to apply for exemptions to many Appendix K criteria that would normally be required for PWR LOCA analysis methods where significant heat-up generally always occur during postulated LOCAs. As noted for NPM, inventory to prevent heat-up is provided by return flow from the CNV via the RRVs. The NuScale design does not have large piping typical to PWRs so the range of breaks considered for LOCA is limited to small piping that eliminates the gross loss of RCS inventory and the need for ECCS injection pumps.
Version                Description                  NCI Beta 00.10                Critical Heat Flux (CHF)    NCI-0415-13361 v1.0RC2                  CHF final Revision 1        NCI-0415-13361 v1.1RC15                  ((             }} CHF        NCI-0316-48153 Correlation v1.1                      Version Release with CHF    NCI-0416-48924 Fix v1.2RC2                  CHF Update                  NCI-0616-49518 v1.3                      Implementation of CHF        NCI-0916-51506 NCRs The NuScale LOCA EM limits core heat transfer modeling to pre-CHF phenomena only. If a scenario can be postulated where CHF limits are exceeded during a LOCA, this EM would be invalidated. Therefore, the applicant must demonstrate that fuel does not closely approach or experience CHF condition, i.e., collapsed water levels must always remain above the top of active fuel, the containment remains intact, and pressure and temperature in the CNV remain below design limits. This allows the applicant to apply for exemptions to many Appendix K criteria that would normally be required for PWR LOCA analysis methods where significant heat-up generally always occur during postulated LOCAs. As noted for NPM, inventory to prevent heat-up is provided by return flow from the CNV via the RRVs. The NuScale design does not have large piping typical to PWRs so the range of breaks considered for LOCA is limited to small piping that eliminates the gross loss of RCS inventory and the need for ECCS injection pumps.
Based on the NRC staffs audit of the NCI-0316-48153 and NCI-0616-49518 change assessments, the applicant initially installed the {{                     }} CHF correlations for application the NuScale by adding heat transfer geometry options 170-176. The CHF correlations were interfaced with other correlations, e.g., {{
Based on the NRC staffs audit of the NCI-0316-48153 and NCI-0616-49518 change assessments, the applicant initially installed the ((                     }} CHF correlations for application the NuScale by adding heat transfer geometry options 170-176. The CHF correlations were interfaced with other correlations, e.g., ((
       }}, based on local mass flow rates. These changes were implemented by adding subroutines {{                                                   }} in 2015.
       }}, based on local mass flow rates. These changes were implemented by adding subroutines ((                                                   }} in 2015.
In early 2016, the applicant elected to replace the {{
In early 2016, the applicant elected to replace the ((
                 }}. The motivation being that {{             }} is endorsed by Appendix K and would allow for easier and expedited reviews by the NRC staff, even though the correlation was
                 }}. The motivation being that ((             }} is endorsed by Appendix K and would allow for easier and expedited reviews by the NRC staff, even though the correlation was


22 cited by Appendix K for the BWR application and is being applied by the applicant to the PWR, at a significantly higher pressure. Consequently in 2016, the applicant added {{
22 cited by Appendix K for the BWR application and is being applied by the applicant to the PWR, at a significantly higher pressure. Consequently in 2016, the applicant added ((
                                                                           }}.
                                                                           }}.
{{
((
                                                                                               }}
                                                                                               }}
{{
((
                                                         }} The NRC staff audited these changes in NCIs and in ER-0000-2413, CHF Modeling from Sterns Data.
                                                         }} The NRC staff audited these changes in NCIs and in ER-0000-2413, CHF Modeling from Sterns Data.
The NRC staff understands, based on the Chapter 6.2 containment public audit discussions (ML18141A735) that NuScale is planning to change the reference version of NRELAP5 v1.3 to version 1.4. The NRC staff are still in process of reviewing these latest code changes.
The NRC staff understands, based on the Chapter 6.2 containment public audit discussions (ML18141A735) that NuScale is planning to change the reference version of NRELAP5 v1.3 to version 1.4. The NRC staff are still in process of reviewing these latest code changes.

Revision as of 02:01, 27 February 2020

Enclosure 1 - LOCA Audit Summary Public Version
ML20010D133
Person / Time
Site: NuScale
Issue date: 01/28/2020
From: Bavol B
NRC/NRR/DNRL/NRLB
To: Michael Dudek
NRC/NRR/DNRL/NRLB
Bavol B, 415-6715
Shared Package
ML20010D112 List:
References
TR-0516-49422-P
Download: ML20010D133 (23)
v • d • e


Text

AUDIT REPORT FOR THE REGULATORY AUDIT OF NUSCALE POWER, LLC TOPICAL REPORT TR-0516-49422-P, LOSS-OF-COOLANT ACCIDENT EVALUATION MODEL

1.0 BACKGROUND

By letter dated December 30, 2016 (ADAMS Accession No ML17004A202), NuScale Power, LLC (NuScale) submitted to the U.S. Nuclear Regulatory Commission (NRC) staff for review topical report (TR) TR-0516-49422, Loss-of-Coolant Accident Evaluation Model, Revision 0, in support of the NuScale design certification application, which the NRC accepted for review on March 23, 2017 (ADAMS Accession No. ML17074A087). On April 27, 2017, the NRC issued a letter accepting TR-0516-49422-P, Revision 0 for review (ADAMS Accession No. ML17116A063). The audit entrance meeting was held on May 11, 2017, at NuScales office located in Rockville, Maryland.

Phase 1 of the audit included the NRC staffs review of docketed and non-docketed information via the NuScale electronic reading room (ERR), and several weekly audit teleconferences. The Phase 2 audit involved an on-site audit in the NuScale Corvallis offices from July 24, 2017, to July 28, 2017. The audit continued with Phase 3 that included examination of documents provided by NuScale in the ERR and weekly audit teleconferences to discuss NRC staff questions through March 15, 2018.

During this audit, the NRC staff, assisted by NUMARK contractors, examined referenced documents and analyses that support the analyses, evaluations, and conclusions in the topical report. The NRC staff provided questions and requests for clarification to NuScale throughout the audit as issues arose. Many of the NRC staffs questions and requests, for clarification, were resolved during the audit. Where issues were identified that could not be resolved and closed during the audit teleconferences, the NRC staff prepared and issued requests for additional information (RAIs) to NuScale. The resolution of the RAIs will be documented in the staffs safety evaluation report (SER) for the LOCA topical report.

2.0 REGULATORY AUDIT BASIS Title 10 of the Code of Federal Regulations (CFR), Section 50.46 and Appendix K state the acceptance criteria for emergency core cooling systems for light-water nuclear power reactors.

Regulatory Guide 1.203, Transient and Accident Analysis Methods, December 2005, provides guidance for preparing and reviewing computer codes and analysis methods for safety analyses to support licensing applications including Design Certification (DC) applications.

An audit was conducted to review the loss-of-coolant-accident (LOCA) evaluation methodology and associated supporting documents described in topical report (TR) TR-0516-49422, Loss-of-Coolant Accident Evaluation Model. The NRC staff conducted this audit to better understand how the evaluation methodology was applied by NuScale to demonstrate its small modular reactor (SMR) design can be certified with reasonable assurance of adequate public health safety.

Enclosure 1

3 Relevant regulatory guidance includes:

  • NuScale Design-Specific Review Standard (DSRS) 15.0, Introduction - Transient and Accident Analyses, Revision 0.
  • Standard Review Plan (SRP) 15.0.2, Review of Transient and Accident Analysis Method, Revision 0.
  • DSRS 15.6.5, Loss-Of-Coolant Accidents Resulting from Spectrum of Postulated Piping Breaks within the Reactor Coolant Pressure Boundary, Revision 0.

3.0 AUDIT LOCATION AND DATES The audit was conducted from the NRC headquarters via NuScales ERR, via telephone and at NuScales Rockville, Maryland, office.

Dates: Phase 1: May 11, 2017, through July 21, 2017 Phase 2: July 24, 2017, through July 28, 2017 Phase 3: July 31, 2017, through March 15, 2018 Locations: NRC Headquarters Two White Flint North 11545 Rockville Pike Rockville, MD 20852-2738 NuScale 1100 NE Circle Blvd., Suite 200 Corvallis, OR 97330 4.0 AUDIT TEAM MEMBERS NRC staff Shanlai Lu, Sr. Reactor Systems Engineer, Lead, NRO/DSRA/SRSB Carl Thurston, Reactor Systems Engineer, NRO/DSRA/SRSB Peter Lien, Sr. Reactor Systems Engineer, RES/DSA/CRAB Christopher Boyd, Senior Level Staff, RES/DSA Brookhaven National Laboratory Upendra Singh (Kumar) Rohatgi, Senior Scientist Numark Associate LLC Marvin Smith, Project manager Bert Dunn, Consultant Technical expert Charles Solbrig, Consultant Technical expert Donald Rowe, Consultant Technical expert Liliane Schor, Consultant Technical expert

4 Leonard Ward, Consultant Technical expert 5.0 APPLICANT PARTICIPANTS Mark Chitty, Licensing specialist Brian Wolfe, Code Development Supervisor Meghan McCloskey, Nuclear Methods Safety Analyst Drew Peebles, Licensing Engineer Eric Coryell, Transient Analysis Supervisor Gary Becker, Reg Affairs Counselor Claudio Delfino, Nuclear Methods Supervisor Daniel Diefendorf, Engineer NSSS Analysis Jackson Kappen, Licensing Intern Morris Byram, Licensing Engineer Jeffrey Luitjens, Code Development Zackary Rad, Director, Reg Affairs Jennie Wike, Licensing Manager Selim Kuran, Safety Analysis Engineer Ben Bristol, Safety Analysis Engineer Patrick Byfield, Code Development Pravin Sawant, Nuclear Methods Robert Houser, Manager of Testing and Code Development Kent Welter, Manager of Nuclear Safety Engineering Jose Reyes, Chief Technology Officer Qiao Wu, OSU Professor and NIST-1 Program Manager Cyrus Afshar, Licensing Supervisor 6.0 AUDIT DOCUMENTS The staff audited the following documents provided by NuScale:

LOCA Electronic Reading Room Document List Document Number Document Title EC-T050-3234, Revision 0 NRELAP5 Model for the SIET Fluid Heated Test Facility EC-0000-3853, Revision 1 Calculations to Support NIST-1 Distortion Analysis and Modeling of Containment and Pool Heat Transfer EC-0000-4684, Revision 0 Spurious Opening of an RPV Valve EC-0000-4888, Revision 0 NuScale LOCA Evaluation Model Supporting Calculations EC-A013-2341, Revision 1 Containment Pressure and Temperature Response to Design Basis Events Analysis EC-A013-2341, Revision 2 Containment Pressure and Temperature Response to Design Basis Events Analysis EC-T050-3498, Revision 1 SIET Heated Tube Test facility (TF-1) Adiabatic and Diabatic NRELAP5 Assessment EC-T080-3822, Revision 1 NRELAP5 Assessment Against NuScale Separate Effects High Pressure Condensation Test Series NIST-1 HP-02 EC-T080-4872, Revision 0 NRELAP5 Assessment Against NuScale integral System CVCS Discharge Line Break Test HP-06b (NIST-HP 0630)

5 EC-T080-5045, Revision 0 NRELAP5 Assessment of Spurious RVV Opening Test with 3 RVVs NIST-1 HP-43 EE-T080-13757, Revision 2 NuScale Integral System Test (NIST-1) facility Scaling Analysis ER-0000-3095, Revision 0 NuScale Model Loss-of-Coolant Accident Phenomena Identification and Ranking Table ER-A010-4065, Revision 3 HP-2 High Pressure Condensation Test Procedure Final Test Report ER-A010-4599, Revision 1 HO-06 CVCS Discharge Pipe Break (w/o DHRS) Repeat Run Final Test Report ER-T080-5201, Revision 0 Summary of Completed and Planned NIST-1 Tests NP-TR-0610-289-P, Revision NuScale Preliminary Loss-of-Coolant Accident (LOCA) 1 Thermal-Hydraulic and Neutronics Phenomena Identification and Ranking Table (PIRT)

NP-TR-0610-289-P, Revision NuScale Module Small-Break Loss-of-Coolant Accident 2 Phenomena Identification Ranking Table SDR-0615-15509, Revision 4 OSU NIST-1 Facility Description Report TSD-T080-10675, Revision 2 NIST-1 HP-06 CVCS Discharge Pipe Break Test Specification (w/o DHRS)

EC-A014-6649, Revision 1 NRELAP5 Model for the SIET Heated Tube Test Facility EC-T080-3820, Revision 0 NRELAP5 Assessment Against NuScale Integral System Pressurizer Spray Line Break test NIST-HP-07-1002 EC-0000-3155, Revision 0 Assessment of NRELAP5 with KAIST Condensation Experiments EC-T080-3828, Revision 0 NRELAP5 Assessment Against NuScale Integral System CVCS Discharge Line Break Test NIST-HP-06-0926 EC-T080-4180, Revision 0 NRELAP5 Assessment Against NuScale Integral system ECCS RVV Spurious Opening (w/o DHRS) Test NIST-HP-09 EC-T080-5389, Revision 0 Preheating of NIST-1 Containment Wall During LOCA Tests ER-0000-3171, Revision 0 Qualification of KAIST High Pressure Condensation Data ER-0000-4738, Revision 1 NRELAP5 Version 1.3 Assessment Report EC-A010-1782, Revision 0 NuScale NRELAP5 Module Basemodel EC-A010-1507, Revision 3 System Transient Model Input Parameters Calculation EC-A021-1627, Revision 1 Core Design Parameters Analysis EC-T080-3468, Revision 2 NIST-1 NRELAP5 Base Input Model EC-0000-4153, Revision 0 Assessment of NRELAP5 with Marviken Critical Flow Tests EC-0000-4162, Revision 0 NRELAP5 Assessment Against Bankoff CCFL 15-hole Experiment EC-A030-2353, Revision 0 High-Energy Line Break Benchmark Analysis EC-T0000-3107, Revision 0 Development Assessment of NRELAP5 with Wilson Bubble Rise Experiments EC-0000-3180, Revision 0 NRELAP5 Assessment Against Semiscale Natural Circulation Test S-NC-2 and S-NC-10 EC-0000-3141, Revision 0 NRELAP5 Comparison to FLECHT SEASET Core Boiloff Testing ER-0000-3174, Revision 0 Qualification of FRIGG Void Fraction Data Validation of NRELAP5 EC-0000-3183, Revision 0 Assessment of NRELAP5 with GE Level Swell Experiments (Large Vessel)

6 EC-0000-3192, Revision 0 Assessment of NRELAP5 with GE Level Swell Experiments (Small Vessel)

EC-0000-3190, Revision 0 Assessment of NRELAP5 with Ferrell-McGee Experiments EC-0000-3216, Revision 1 NRELAP5 Comparison EC-A021-2404, Revision 0 Axial Offset and Axial Power Shape Analysis ER-0000-2486, Revision 4 Safety Analysis Analytical Limits Report ER-A025-3276, Revision 1 Design Inputs for NuScale Neutronics/Safety EC-T050-3638, Revision 0 Assessment of NRELAP5 using SIET Fluid heated test Facility (TF-2) Data NCI-0216-21648, Revision 0 NCR-1215-19748 notes that the implementation of the extended Shah correlation in NRELAP5 is not in accordance with Shahs 2009 paper presenting correlation.

NCI-0315-12869, Revision 0 NRELAP5 Change Implementation - Add a helical coil component that will be used in HCSG modeling NCI-0916-51421, Revision 0 The NCI make a correction to the laminar heat transfer coefficient for the inside of an NRELAP5 helical coil steam generator (HCSG) component.

NCI-1214-9804, Revision 0 NRELAP5 Extended Shah Condensation Heat Transfer Correlation Implementation NCI-1214-9943, Revision 1 Appendix K moody critical flow option change performed by INL (report attached), and reviewed/implemented by Glenn Roth EC-A030-2713, Revision 2 Primary and Secondary Steady State Parameters SDR-0615-15509, Revision 4 OSU NIST-1 Facility Description Report SDR-0815-16915, Revision 4 Data Acquisition and Control System Configuration Report TSD-A010-51877, Revision 0 NIST-1 HP-43 ECCS RVV Spurious Opening with 3 RVVs Test Specification (w/o DHRS)

EC-0000-3803, Revision 1 Predicting Critical Heat Flux Margin during LOCA in NuScale Power Module NP12-01-A013-M-GA-2601, Lower CNV Section Revision 1 NP12-01-A013-M-GA-2602, CNV Upper Section Revision 1 ER-A025-4728, Revision 0 Control Rod Assembly Drop Simulation Data Applicability analysis TR-1015-18177-P, Revision 0 Pressure and Temperature Limits Methodology NP12-01-A011-M-SA-2689, Steam Generator Revision 0 ER-A014-2268, Revision 0 Evaluation of Helical Geometry Effects for SG Heat Transfer and Pressure Drop SDR-0516-49257, Revision 1 Piping & Instrumentation Diagram - RPV SDR-0516-49259, Revision 1 Piping & Instrumentation Diagram - CNV Configuration 2 (Discharge, Scaled DHR)

NP-12-01-A010-M-GA-1957, Reactor Module Assembly Revision 1 NP12-01-A010-M-SA-2761, Internal RXM Platform and Ladder Revision 0 NP12-01-A011-M-SA-2639, Lower RPV Section Revision 2

7 NP12-01-A023-M-GA-2304, Reactor Vessel Internals-Lower Riser Revision 1 NP12-01-A023-M-GA-2305, Reactor Vessel Internals-Core Support Revision 1 NP12-01-A013-M-GA-1933, Containment Vessel Assembly Revision 3 NP12-01-A013-M-GA-2603, CNV Top Head Assembly Revision 1 NP12-01-A013-M-KR-2702, CNV Internal Electrical Conduit Revision A TR-0915-17564-P, Revision 1 Subchannel Analysis Methodology NP12-01-B020-M-PD-1027, Piping & Instrumentation: Emergency Core Cooling System Revision 2 NP12-01-B010-M-PD-1021, Piping & Instrumentation: Chemical and Volume Control Revision 0 System NP12-01-A030-M-PD-1504 Reactor Coolant System P&ID

7.0 DESCRIPTION

OF AUDIT ACTIVITIES AND

SUMMARY

OF OBSERVATIONS The NRC staff audited information related to NuScales LOCA analysis methodology through three phases of the audit. During the first phase, the audit team became familiar with the initial set of information provided in the NuScale ERR. After the NRC staff determined the need to examine the actual NuScale Integral System Test-1 (NIST-1) facility, the audit team initiated the second phase of the audit at the applicants site. Following the week-long on-site audit, the audit team continued its examination of documents provided in the ERR, by NuScale, as the Phase 3 audit. The NRC staff focused its audit on the NRELAP-5 code models and correlations, NuScale reactor LOCA phenomenon identification ranking table (PIRT), LOCA evaluation model, separate effects and developmental test assessments, NIST scaling and distortion analysis, the integral test assessment cases, LOCA sample calculations and NRELAP5 source code updates. The specific observations are documented below.

1.1 Nuclear Power Module LOCA PIRT The NRC staff observed that the initial NuScale LOCA PIRT was developed in 2008. The PIRT was then subsequently updated in 2013, and 2015, to address the design changes. The applicant noted that within the selected LOCA scenario, the evolution of the fluid volume distribution between and within the RPV and the CNV is such that the core is always covered with a mixture of steam and water that would converge to approximately 10 feet, in which case, the figure of merit that the core fuel rods do not experience critical heat flux would be appropriate. However, the NRC staff observed that this will set a limitation on the validity of the PIRT and the LOCA methodology that no core uncovery occurs during any LOCA transient, i.e.,

reactor vessel mixture level does not fall below the top of the core heated region.

In addition, the NRC staff learned that dry out is not specifically identified as a figure of merit for PIRT considerations, although it is certainly implied by the CHF consideration. Therefore, the NRC staff observes another limitation on the applicability of the PIRT is that the event to which the PIRT is applied must not evolve to dry out of any of the fuel rods at any location along their heated length.

The audit team noted that ECCS activation trips (for core and containment liquid level) use

8

(( that are not standard for use inside nuclear reactors. The NRC staff that are reviewing DCD Chapter 7 have issued RAIs which may have implications for this LOCA PIRT. 1.2 NRELAP5 Code The NuScale NRELAP5 systems code is the Idaho National Laboratory (INL) RELAP5-3D (Reactor Excursion and Leak Analysis Program-Three Dimensional) computer code version v.4.1.3 with commercial grade dedication and name change to NRELAP5 v0.0. NuScale subsequently made many modifications and additions to develop the baseline version for the LOCA TR, i.e., NRELAP5 v1.3. Each coding change is documented in an NCI (NRELAP5 change implementation) that is documented under a SwRN (software release notice). The NCIs review in detail, the coding changes made and contain specific benchmarks and regression testing needed to confirm that new code modeling is performing as intended. The scope of this audit did not encompass detailed examination of all coding NCIs but a subset directly affecting LOCA phenomena and FOMs. The validation and verification of the NRELAP5 code should be driven by the phenomena PIRT process as described by RG 1.203, which identifies the important and key phenomena to be modeled by this evaluation model (EM) for the LOCA event. Since the NuScale Power Module (NPM) is designed to reduce the consequences of LOCA events, as compared to traditional pressurized water reactors, many of the phenomena that are the subject of 10 CFR Part 50, Appendix K requirements are not reached in the NPM LOCAs. They are essentially considered to be designed out or satisfied by design rather than by analysis. A list of these designed out features are described in Section 2.2.2 and Table 2-2 of the LOCA TR (Reference 2). 1.2.1 Numerics During the audit of the NuScale LOCA EM, the NRC staff examined the NRELAP5 theory manual given in SwUM-0304-17023, Revision 4, regarding code model and correlation modifications made to NRELAP5 for application of the code assessments to validation predictions of NPM ECCS performance. This audit addressed only NuScale core changes starting from the INL RELAP5-3D code version v.4.1.3 as described in INEEL-EXT-98-00834, dated September 2013. The NRC staff was made aware that a developmental assessment of the RELAP5-3D computer code was performed by INL (INL/MIS-15-36723, RELAP5-3D Code Manual Volume III: Developmental Assessment, October 2015). This assessment used a combination of phenomenological, separate effects, and integral effects test cases to validate the codes performance. This recent review of the code by INL (based on RELAP5-3D version 4.3, released October 2015) adds pedigree to the version obtained by NuScale, although the NRELAP5 code is taken from Version 4.1.3 (2012). Because the NRC staff has identified generic code performance issues with past versions of RELAP5, NuScale was requested to respond to several issues during the audit. Below is a summary of the issues, interactions with NuScale and the NRC staffs observations during the audit. (1) Information Notice No. 92-02, Supplement 1: RELAP5/MOD3 Computer Code Error Associated with the Conservation of Energy Equation IN 92-02, Supplement 1 describes a deficiency that whenever the code is applied to situations in which the pressure drops significantly between cells, the energy in the downstream volume may be underestimated. INL proposed an improvement to include the PV (pressure and volume) term in the energy equation to correct the pressure via a junction flag (e=1). This fix is carried

9 over into the base version of NRELAP5 and appears to be used properly in the applicants NRELAP5 LOCA input models. (2) Formulation of the momentum equations The NRC staff has determined that for previous versions of RELAP5, when compared against a simple frictionless manometer, the amplitude as well as the period of the oscillations calculated by RELAP5 began to deviate substantially after a few seconds from an exact solution computed with an alternative method. The NRC staff has observed that calculations with previous versions of RELAP5 have shown that the code predicts an incorrect period that becomes completely out of phase with the correct solution after about 20 seconds. Because the geometry modeled in NRELAP fixes the length of the fluid column in the code, the inertia is determined by the NRELAP geometry input - not the actual fluid column length in the cell. Thus, the inertia of the transient fluid column in NRELAP can be inconsistent with the constant fixed geometry. In view of these considerations, the NRC staff indicated that it is a common practice for RELAP developers to verify the code numerical scheme using a simple manometer problem with and without friction. Because hydrostatic forces dominate NPM long-term LOCA fluid behavior, the NRC staff needed to determine that the NuScale LOCA EM adequately addresses fluid inertia and momentum in the NRELAP5 modeling. (3) Flow anomaly in RELAP series system code The NRC staff has identified flow anomalies in thermal hydraulic codes employed for previous LWR applicants and in some integral test experiments. These flow anomalies appeared as non-physical recirculating two-phase or single-phase flows in multidimensional models and models with single multiple parallel fluid pipes (flow paths). So that the NRC staff could determine whether this anomalous flow behavior has been adequately addressed in the NuScale LOCA EM NRELAP5 code, the NRC staff requested NuScale to make available for the audit the results of a study showing whether NRELAP5 correctly predicted the liquid flow rates in a simple system containing two parallel pipes and another with three parallel pipes, similar to the core modeling feature contained in the NuScale LOCA EM. The NRC staff noted that the core NuScale LOCA EM uses the 1-D modeling in NRELAP5 with multiple parallel pipes in the core. The NRC staff issued RAI 9476 and RAI 9517 to examine this issue. 1.2.2 Initial Condition Determination The NRC staff noted that initial conditions were not optimized for heat transfer performance of the steam generators during a LOCA. The NRC staff issued RAI 9085 to request NuScale to address evaluation of steam generator heat transfer conditions that would result in higher stored energy in the RPV. The NRC staff also noted that Table 7-1 in TR-0516-49416-P appears to describe a list of parameters an analyst would use to specify the appropriate set of initial conditions and how the analyst would determine that a unique steady-state solution was obtained for an analysis. The NRC staff issued RAI 9374, Question 15.00.02-25, requesting NuScale to clearly state the purpose of Table 7-1 in TR-0516-49416-P and to revise Table 7-1 and the associated text in TR Section 7.1.1.2, accordingly.

10 1.2.3 Containment Condensation Model During the audit, the NRC staff noted that NuScale included additional correlations and capabilities in the NRELAP5 code to handle condensation heat transfer to the inside CNV wall with a CNV design pressure up to 1000 psia. NuScale added (( 

                }} condensation heat transfer correlations to the NRELAP5 code. However, the use of ((                         }} correlations for NuScale CNV condensation are outside the applicability range of this empirical correlation as it was developed for vertical pipes with small diameters and low-pressure conditions. In addition, the NRC staff noted that ((
                                                                               }} and it was not clear whether the overall condensation heat transfer was being predicted conservatively. Because the steam condensation in CNV is of major importance to the prediction of ECCS performance for the NuScale design and the descriptions of condensation heat transfer correlations in the NuScale LOCA EM was unclear, the NRC staff requested additional information regarding these modifications. The NRC staff issued RAI 8990, Question 15.06.05-7, requesting NuScale to provide information related to this issue.

(i) (( }}, designed for flow in pipes, is being used beyond the stated range of applicability of the hydraulic diameter. The NRC staff also questioned (( }} to this application for the NuScale design, which is not a pipe. (ii) The NRC staff observed through audit of the NRELAP5 theory manual SwUM-0304-17023 and the audit discussions, that there is a lack of clarity in NuScales understanding of the performance of the wall condensation model as well as a lack of key sensitivity studies showing the effects of differing condensation coefficients and/or variations in the heat removal rate by the CNV metal walls and how the correlations were implemented in the code subroutines. Sensitivity studies are necessary to establish a conservative LOCA EM analysis of ECCS function including CNV pressure response and its impact of RRV activation and the IAB differential pressure set point. In RAI 8990, NuScale was asked to provide detailed numerical calculation procedures showing how (( }} correlations are actually used in the NRELAP5 code. 1.2.4 NRELAP5 Input Manual The NRC 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 to develop input models for application of the methodology. The NRC 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. 1.3 LOCA Evaluation Model The NRC staff observed that the NPM LOCA model is developed from the NRELAP5 base deck that is designed for general Chapter 15 safety analysis applications. The LOCA nodalization is downsized considerably from that in the base deck except for the core modeling, where a hot channel is added. Also (( }}. Most other changes were related to eliminating details in the secondary and feedwater systems that NuScale asserted as not necessary for LOCA applications.

11 The NRC staff confirmed that the model addresses the key components of the NPM for LOCA transients, including: RPV internals, containment, steam generator primary and secondary sides, the reactor pool and the ECCS system. The reactor core is modeled with three parallel channels: (1) the average channel representing the remaining 36 fuel assemblies, (2) the hot channel representing a single NPM fuel assembly and (3) the core bypass that includes the reactor reflector channel bypass and the fuel assembly instrument and guide tube bypass. 1.3.1 NPM Core Model The reactor core is modeled with three parallel channels as indicated above. The loss coefficients for (( 

                                                         }}. The NRC staff audited the detailed calculation procedures for these coefficients and had no significant issues related to the initial blowdown and establishment of the initial ECCS recirculation through the reactor vent valves (RVVs) and reactor recirculation valves (RRVs). The longer-term concerns of this modeling were previously discussed in Section 7.2.1 regarding flow anomalies in parallel piping channels.

The NRC staff observed that the core fuel rods use (( }}, which invokes a NuScale-specific CHF correlation package that uses: ((( 

              }}. The applicant used benchmarks to the subchannel topical report (Reference 7) to show that this critical heat flux (CHF) package yields conservative results. The decay heat model used is ANS 73 with a 1.2 multiplier that is shown to be in compliance of 10 CFR 50.46 Appendix K requirement. NuScale is requesting an exemption to use ANS 73 rather than Appendix K ANS 71.

The NRC staff audited the implementation of these models and the supporting calculations. No significant issues were identified in these models. However, the NRC staff noted that in FSAR, Section 15.6.6., which also describes (( }}, that the CHF ratio (CHFR) limit was not being consistently stated. As such, the NRC staff issued RAI 9536 to request the applicant to provide documentation that consistently reflects the final CHFR limits. 1.3.2 Pressurizer Model The pressurizer baffle plate and the fuel assembly upper tie plate are considered to a have potential for counter current flow limitation (CCFL) and therefore the junctions are modeled with (( }} activated based on the plate geometry. The NRC staff raised the issue about how (( }} is being validated and used at the baffle plate location. The detailed supporting documents showed that the use of this correlation has insignificant impact on the two-phase water level above the core. Therefore, the NRC staff determined that the issue was not significant and doesnt need to be addressed. 1.3.3 Steam Generator Model

12 The two helical coil SGs are modeled using a NuScale-specific helical coil SG component that uniquely addresses the flow geometry and heat transfer effects. This new hydrodynamic component and heat transfer package were added to NRELAP5. The modeling makes specific heat transfer and wall friction changes based on the SIET test data. The NRC staff found no significant issues with the actual SG model implemented in the NRELAP5 code. However, the NRC staff noted inconsistencies regarding specific SIET test configuration and results, which may call in question the heat transfer modeling changes developed. 1.3.4 Containment Model The CNV surrounds the RPV and the NRC staff noted that it is being modeled as a vertical PIPE located in an annulus between the outer RPV wall and the CNV inner wall. The CNV retains steam and liquid in the event of a LOCA, such that after ECCS activation, discharged inventory can be returned to the core via the RRVs. The volume of the CNV is just over 2.4 times the volume of the RPV. Each NPM is mounted with a segmented bay within a vast reactor pool. The pool area relevant for each NPM was assumed to be (( 

                                        }}. Although the numerical nodalization of the CNV and the reactor pool appeared to be reasonable, the NRC staff identified issues in the following areas:
  • NRELAP5 condensation modeling applied.
  • Heat structure modeling of the CNV wall.
  • Thermal stratification within lower portion of the CNV and its potential impact on peak containment pressure.
  • Coarse noding of the reactor pool.

The first three items above are addressed via existing RAI 8990 and RAI 9380. The NRC staff reviewed the reactor pool modeling and determined that it was adequate for the period of the LOCA transient addressed by this EM. Heat Structure Modeling the CNV Wall The NRC staff observed that heat diffusion and conduction into and through the CNV wall is the major factor controlling the rate at which heat is absorbed from the RPV following a LOCA. Thus, the NRC staff documented its request that NuScale demonstrate that CNV heat conduction is conservatively calculated in the NuScale LOCA EM in RAI 8990, Question 15.06.05-7. In this RAI, the NRC staff requested that NuScale describe how NRELAP5 calculates heat convection and conduction to the CNV wall surfaces below the liquid water level and above the water level inside the CNV and to describe the initial boundary conditions applied. Thermal Stratification in Containment Vessel During LOCA The NRC staff reviewed the experimental data for HP-02 and observed that stratified subcooled water below the saturated water liquid level with saturated steam above within the containment vessel throughout this transient test. The presence and importance of subcooled water in the containment is not acknowledged in the topical report. The NRC staff observed, through its own independent analysis, that subcooled water in lower containment is important to the computation of peak containment pressure. In the EM, the applicant evaluated coarser nodalization in the CNV and it had no significant impact on the computed peak containment

13 pressure, however finer noding was not addressed. The NRC staff believes that finer noding in the liquid region could produce a greater inventory of subcooled water leading to less condensation and a higher computed peak containment pressure. The NRC staff issued RAI 9380, Question 06.02.01.01.A-5, requesting that NuScale provide additional information to enable the NRC staff to assess the impact of liquid thermal stratification and nodalization in the CNV liquid region. Modeling Containment and Pool During the audit, the NRC staff requested that NuScale justify its NRELAP5 one dimensional model of the containment, effects of non-condensibles, and ultimate heat sink pool. The NuScale NPM model employs one-dimensional PIPE components to model both containment and (( }} the cooling pool outside of each containment module. As a result, the pool water within each NRELAP5 PIPE node is numerically predicted to sense a uniform bulk temperature increase during a LOCA. The NRC staff questioned the impact on the LOCA analysis of this simplified approach. The NRC staff and NuScale agreed in audit discussions that layers of convection currents would develop near the outer CNV wall that cannot be modeled by NRELAP5, which would result in warmer temperatures being in direct contact with the CNV wall. The NRC staff issued RAI 8776, Question 15.06.05-2, requesting that NuScale provide additional justification for the modeling used and to evaluate the impact on the FOMs, i.e., on peak containment pressure and collapsed water level above top of active fuel (TAF) for the limiting LOCA cases. 1.3.5 LOCA Initial Conditions The assumptions for conservatively biasing the initial plant conditions were based on Table 5-6 of the LTR and the NRC staff determined that the initial plant conditions were not consistent with the uncertainty values given in Table 15.0-6 of the FSAR. The NRC staff audited the LOCA calculation supporting the EM, EC-0000-4888 NuScale LOCA Evaluation Model Supporting Calculations, and noted that Tave used was lower than that in the table. The following two scenarios for loss of power are considered: (1) complete loss of normal alternating current (AC) and DC power and (2) loss of only AC power with DC power availability. The NRC staff issued RAI 9475, Question 15.06.05-17, requesting that NuScale describe the method and bias for selecting the maximum value and a sample upper bound value for RCS average temperature, riser temperature, pressurizer pressure, main steam pressure and feedwater temperature, and reactor pool temperature. 1.3.6 Single Failure Assumptions Single failures considerations included: (1) failure of a single RVV to open, (2) failure of a single RRV to open, (3) failure of one ECCS division (i.e., one RVV and one RRV) to open, and (4) no single failure. For the NPM, the limiting conditions noted by NuScale do not include a single failure since no single failure is limiting. The NRC staff noted that NuScale included single failure of the ECCS valves to open but did not address failure to close, meaning one valve would open as the initiating event and another valve would simultaneously open as the single failure. The NRC staff reviewed this information in NuScale calculations EC-0000-4888 and EC-0000-4684.

14 The ECCS valves each have an inadvertent actuation block (IAB) set point feature that NuScale has asserted is passive that would prevent the opening of these valves until the differential pressure between RCS and CNV falls below the IAB release pressure (see RAI 8815). If DC power is not available, all ECCS valves will open at the IAB release pressure with the IAB functioning properly. The NRC staff questioned NuScale if it was also possible for the valves to fail partially open or closed. The NRC staff issued RAI 8985, Question 15.06.06-1, requesting that NuScale provide sufficient evidence to justify that a sufficient break spectrum has been considered such that the limiting break size has been identified and that it meets the applicable acceptance criteria. 1.3.7 Break Spectrum and Locations The NRC staff noted that also in EC-0000-4888 and Section 9.0 of the LTR, there is are no sensitivity studies presented to demonstrate the effects of varying the magnitude of the wall condensation heat transfer on CNV pressure and the resulting activation of the ECCS. The sensitivities did show that the 10 percent RCS injection line break was the limiting lowest RPV riser liquid level for the spectrum of breaks and locations. This analysis shows the liquid level in the riser receding to an elevation close to the top of the core. The NRC staff noted that a key design feature for the NuScale EM per the LTR Section 2.2 is the ability of the NPM ECCS to ensure no break size or location would result in the collapsed liquid level in the riser decreasing below the top elevation of the core. Based on GE level swell benchmark results, the NRC staff suspected NRELAP5 may over predict two-phase level swell, so the NRC staff issued RAI 8785, Question 15.06.05-1, requesting that NuScale provide sufficient evidence to justify that a sufficient break spectrum has been considered such that the limiting break size has been identified and that it meets the applicable acceptance criteria. The NRC staff noted that the break location evaluated in LTR and EC-0000-4888 only addressed locations for the NPM design in the CVCS injection and discharge lines, the pressurizer spray supply line, and high point vent lines. The NRC staff also noted that NuScales high energy line break (HELB) criteria in DCD Chapter 3, eliminates consideration of breaks in the RRV and RVV nozzles and the inadvertent opening of these valves from the LOCA break spectrum. The NRC staff also questioned whether a partial break or partial inadvertent opening of a RRV with a single failure of the other RRV to open is a credible LOCA scenario and issued RAI 8776, Question 15.06.05-5 and RAI 8985, Question 15.06.06-1. The NRC staff subsequently determined that these RAIs that relate to break locations would be referenced to RAI 9358 which will determine if these locations meet the break exclusion criteria per Branch Technical Position (BTP) 3-4. 1.3.8 Sensitivity Studies Model nodalization only addressed coarser noding and did not considered if the existing noding scheme contained adequate detail. The NRC staff identified that the Counter Current Flow Limitation behavior study should have used the limiting 10 percent CVCS line break instead of the full area CVCS line break case. 1.3.9 Control Rod Insertion for LOCA The EM takes credit for the control rods entering the core and shutting down the nuclear reaction. However, it was not clear to the NRC staff, if an analysis to support rod insertion had been completed. Analysis should include support structure beyond the core that can potentially

15 interfere with insertion. NuScale should provide a complete NPM analysis to confirm insertability of rods under LOCA plus seismic conditions. The NRC staff determined that this issue is beyond the scope of the LOCA topical report, however, it should be addressed by the NRC staff as a part of the LOCA 15.6.5 reviews, which will be supported by the NRC staff reviewers of DCD Sections 3.9.4 and 4.2. 1.4 NIST Scaling and Distortion Analysis 1.4.1 Containment Vessel Scaling Section 8.3.2 of the LOCA LTR (TR-0516-49422-P) describes and provides conclusions on important phenomena and scaling distortions, based on NuScale report EE-T080-13757, Revision 2. The NRC staff identified several significant distortions related to NIST-1 containment vessel (CNV) scaling including: ((( 

                         }}.

It appeared to the NRC staff that the applicant only partially addressed these distortions and did not address the inter-relationship among the significant distortions clearly. The NRC staff was concerned that the existing distortion reports, which the NRC staff audited, may not adequately capture all the high-ranking important phenomena in the PIRT, nor correctly quantify the distortions to determine the uncertainty in the figure of merit (FOM) prediction. The NRC staff issued RAI 9208, Questions 15.06.05-14 through 15.06.05-16, requesting NuScale provide information on similarity groups (PI groups), the H2TS scaling methodology referenced in Section 8.3.2 of the LOCA LTR, heat transfer resistance sensitivity study conclusions, and potential NIST-1 test facility distortions which the NRC staff believes could cause an over prediction of NPM condensation. 1.4.2 Power Level Scaling The NRC staff noted that NIST-1 scaling, per EE-T080-13757, Revision 2, is based (( 

                                            }}. The NRC staff also noted that the testing results for various power levels for the HP-05 test did not compare well with the NRELAP coolant temperature predictions for those power levels. NuScale is requested to address this distortion regarding its implication for peak containment pressure estimation. The NRC staff issued RAI 9208, Question 15.06.05-15, requesting NuScale address distortion regarding its implication for important figures of merit.

1.4.3 NIST-1 Riser Scaling Also, in the scaling document, EE-T080-13757, Revision 2, staff noticed that the NIST-1 riser is scaled to (( }}. No basis is documented in the scaling report for choosing this (( }}. This value (( 

                                                                                  }}. This could also affect the flow regime transition and therefore potentially distort the RPV level and RPV depressurization. Considering little Reactor Pressure Vessel (RPV) level margin following a small break LOCA, the distortion of Jg introduced in the current scaling method may adversely impact the RPV level prediction directly and containment pressure indirectly. With the current

16 riser flow area scaling, the downcomer flow area outside riser scaling may be affected accordingly. The NRC staff is concerned about the potential impact on the level swell in the riser section and the scaled NPM two-phase water level above the core. NuScale was asked to provide an evaluation of the uncertainty of RPV level response and peak containment pressure due to current riser area scaling distortion. The NRC staff issued RAI 9208, Question 15.06.05-16, requesting NuScale provide an evaluation of the uncertainty of RPV level response and peak containment pressure and quantify the scaling similarity parameter. 1.4.4 Initial Steady State Scaling The NRC staff noted significant distortions in initial temperature and pressure conditions due to design limitations of NIST-1. In the scaling distortion report (EC-0000_3853_R1), the applicant attempts to quantify the effect of the scaling distortions by performing sensitivity calculations for NPM and NIST-1 configuration. In the course of its audit, the NRC staff identified several discrepancies in the scaling methodology used. These issues are addressed in RAI 9208 Question 15.06.05-16. 1.4.5 Integral Estimate of Scaling Uncertainty Due to Distortion The NRC staff noted that the purpose of the NIST-1 facility is to provide realistic data for evaluation model validation. The NRC staff noted scaling overall and discussion of scaling distortions were not adequately addressed within the LOCA EM, and the NRC staff needs to better presentation of how these discrepancies and distortions contribute to the LOCA figures of merit (e.g. reactor level, containment pressure) quantitatively. These NRC staff requests and questions are addressed in RAI 9390, Question 15.06.05-19. 1.4.6 Potential Distortion Due to (( }} The NRC staff noted that several of the NIST-1 tests required (( 

                     }} and questioned the potential scaling distortion introduced by ((
                               }}. The NRC staff needs to understand the basis for ((
             }}, the criteria used to determine this temperature, and the effect it has on the test results, including additional discussion regarding the associated scaling distortion. The NRC staff issued RAI 8777, Question 15.06.05-1 requesting NuScale describe the criteria and summarize the supporting analyses that form the basis for selecting the ((
              }} during the NIST-1 testing.

1.4.7 Scaling Distortion Analysis Related to Containment Peak Pressure Measurement In the scaling distortion report (Calculations to Support NIST-1 Distortion Analysis and Modeling of Containment and Pool Heat Transfer, EC-0000-3853_R1), the applicant quantified the effect of the scaling distortions by performing sensitivity calculations for NPM and NIST-1 configuration. The NRC staff also noticed that Section 4.1 of the Containment Response Analysis Methodology Technical Report (CRAM TeR) (TR-0516-49084-P Revision

0) addresses some scaling distortions for the primary and secondary system releases. In the course of its audit, the NRC staff identified several discrepancies that may affect peak containment pressure, and the NRC staff determined that the scaling and distortions also were not adequately addressed in the audited documents. These issues are captured in RAI 9494, Question 06.02.01.01.A-16.

17 1.5 NIST Test Facility and Testing Procedures 1.5.1 Heater Rods (( }}. On Page 217 of this TR, NuScale stated that the heater rods (( 

                                             }}. During the Phase II Corvallis on-site audit, NuScale explained that ((
                                                                                                 }}.

However, the NRC staff noted that for the follow-on NIST-1 Test HP-49 assessment, NuScale revised the NRELAP5 heater rod modeling that significantly increased the thermal resistance and thereby the stored energy in each rod. This issue remains unresolved and is being addressed as a part of the next phase of the LOCA audit. 1.5.2 (( }} for LOCA Tests (( 

                                                }}. RAI 8777, Question 15.06.05-1, (((
                              )}} has been identified to investigate the issue of ((
                        }}.

1.5.3 Pressure Over-prediction of Condensation Test HP-02 The NRC staff noted in EC-T080-3822, Revision 1, HP-02 Assessment, that the computed results for this larger scale condensation test increasingly over-predicts pressure as the absolute pressure increases. This is the only larger separate effects test available to validate (( }} condensation modeling in NRELAP5. NuScale reported in audit calls that the NRELAP5 code modeling did not consider heat loss. The NRC staff questioned the validity of the test for the condensation rate if heat losses from the CNV shell were not addressed. The NRC staff issued RAI 9317, Question 06.02.01.01.A-8, requesting NuScale provide additional NRELAP5 analyses that include heat losses and other phenomena that may have been neglected in the original analyses (Runs 1, 2, and 3) and to include plots that clearly show the impact of modeling changes. Natural Circulation Test HP-05 The NRC staff noted that NIST-1 Test HP-05 was a calibration test to refine hydraulic loss coefficients to improve the NRELAP5 computation of natural circulation flow rates observed in NIST-1. Loss coefficients for the NPM are to be estimated and refined from hydraulic tests on an actual plant after the NPM is built. No hydraulic mockup is anticipated. NuScale relied on the hydraulic analysis of the NPM to produce the expected natural circulation performance characteristics. Technical Report (EC-A030-2359) appears to indicate the RCS form losses, in the base deck (EC-A010-1782), are only based on theoretical formulations from Idelchik. The NRC staff could not locate any documented changes to the RCS NPM form losses that incorporated NIST-1 HP-05 flow data or observations. The NRC staff then noted some minor changes in form losses in the HCSG regions, but the changes are not documented in the LOCA calculations EC-0000-4888.

18 During the Phase 2 on-site audit, NuScale explained that the adjustments to the hydraulic loss coefficients were very small. They also stated that the major pressure loss is in the steam generator. The NRC staff agreed that the changes would have a very small effect on flow. CVCS Discharge Pipe Break, NIST-1 Tests HP-06, HP-06b Decay Power after Trip It was stated in the NuScale distortion analysis report that HP-06 was conducted with (( 

        }} of the scaled decay heat power. Because the decay heat is the only reason that ECCS needs to be designed to protect the core fuel integrity and the two-phase water level above the core, NuScale performed a new CVCS pipe break test using the full-scale decay heat power. This new test and HP-06 are both addressed in the LTR and are used for the validation.

1.6 NRELAP5 Codes Assessment Cases 1.6.1 Edward Pipe Blowdown and LOFT test L2-5 The NRC staff noted during the audit, that the NuScale LOCA EM TR refers to Edward pipe and LOFT L2-5. However, the NRC staff noted that these assessments are not described in any part of the topical report and that they were not modeled to support the NRELAP5. NuScale personnel agreed to remove this sentence from a future revision of the NuScale LOCA EM topical report. The NRC staff will confirm that this modification is made to the TR. 1.6.2 SIET Test Issues The NRC staff audited the supporting documents relevant to the SIET tests and post test data processing and identified several significant issues. First, the NRC staff noted that the test loop by-pass flow rates were not zero at least for several TF-2 test cases. However, the assessment calculations assumed zero flow through the by-pass valve. Second, the NRC staff questioned the SIET post-test benchmark pitch/diameter ratio and issued RAI 9519, Question 15.06.05-20 to request NuScale clarify the test results. The NRC staff also noted that TF-2 series of tests had a small secondary side flow rate that may not sufficiently cover the full range the NPM steam generator operating conditions. This issue is captured in the following two RAIs, RAI 9190, Question 15.06.05-11 requesting that NuScale provide a comparison between the heat transfer coefficients for the HCSG calculated with the incorrect equations 14 and 15 of NCI-0315-12869 versus the corrected HTC equations as shown in NCI-0916-51421, and RAI 9351, Question 15.00.02-33 requesting that NuScale confirm whether or not the bypass flow path was active during any of the tests, and if active, either: (a) revise the assessment to account for the correct flow rates if it does not already, or (b) justify the acceptability of the current assessment, and update relevant topical reports and any other documents as appropriate, based on the responses. 1.6.3 Decay Heat modeling The NRC staff noted in the LOCA TR and TR-0516-49416 that actinide contributions were not being treated consistently to ensure conservative results. Therefore, the NRC staff issued RAI 9466, Question 15.00.02-7 requesting NuScale provide the actual values of and used, and to justify that the use of these values, the specified multipliers, and inclusion (or lack thereof) of actinides in combination with the 1973 decay heat model leads to a conservative result for decay heat contribution for all event types and update TR-0516-49416-P and any other affected documentation as appropriate.

19 1.7 Loss- of-Coolant Accident Sample Calculations 1.7.1 Containment Peak Pressure vs Break Size During the audit, the NRC staff questioned why the time to peak containment pressure decreases with smaller breaks down to about 35 percent breaks, as it was expected that smaller breaks would increase the time to the IAB release pressure as depicted in Figure 9-12 of this TR. NuScale explained that the x axis in Figure 9-12 is a time scaled variable that is time multiplied by the break area ratio. 1.7.2 Secondary Side Isolation During the audit, the NRC staff noted that the ratio of NuScale Helical Coil Steam Generator tube flow area to primary system coolant inventory is much greater than that of traditional PWRs. In addition, the NRC staff noted that the NuScale design does not have any safety grade make-up capacity. Therefore, if too much inventory is lost through the secondary system, there will be no coolant for the Decay Heat Removal System and the ECCS may have to be actuated during this event. If it does happen, and primary coolant continues leaking to the secondary system, ECCS actuation may not be able to mitigate the event. NuScale responded that the secondary side isolation will be activated once a LOCA event is detected by the module protection system. Although this issue is relevant to the LOCA topical report review, this issue is within the review scope of the NuScale Design Certification application for Section 15.6.3. 1.7.3 Fuel Pellet Thermal Conductivity Degradation During the audit, the NRC staff questioned the fuel behavior calculation and the approach of using the COPERNIC code to calculate the initial core stored energy. Since insufficient information was provided, the NRC staff issued RAI 9065, Question 15.06.05-8, requesting NuScale provide additional information which justifies that the hand calculations are conservative for the various fuel inputs provided by EC-0000-4888. 1.8 NRELAP5 Source Code Update The RELAP5-3D computer code is the later of a series of computer programs developed at INL for modeling nuclear power plants for the Department of Energy (DOE) and the NRC. It was originally designed to model small-break loss-of-coolant accidents (SBLOCAs) for pressurized water reactors. The modeling capability and fidelity of the RELAP codes have grown immensely with each successive release of the software due the large user base. The NuScale NRELAP5 code is developed from RELAP5-3D, which is an extension of the one-dimensional RELAP5/MOD3 code developed at INL. The NRELAP5 code is taken from RELAP5-3D Version 4.1.3, released October 8, 2012, and has been specifically modified for the NuScale NPM. RELAP5-3D Version 4.1.3 was procured from INL through via commercial grade dedication and was renamed as the baseline version, NRELAP5 Version 0.0. Further code development resulted in NRELAP5 Version 1.0, NRELAP5 Version 1.1, and NRELAP Version 1.2. The current version release is NRELAP5 Version 1.3, however the NRC staff understands that after completion of this audit, Version 1.4 has been developed. The NRC staff examined documentation (NRELAP5 Change Notices) related to the modifications and features that were added to NRELAP5 through Version 1.3 to address the unique aspects of the NPM

20 design and licensing methodology. Those changes occurred from December 2014, to December 2016, and include primarily:

1. Appendix K Moody critical flow models
2. Helical coil SG component model with specific heat transfer and pressure drop models
3. Appendix K ANS 1971 decay heat models
4. Appendix K (( }} CHF models
5. (( }} Wall condensation models The NPM NRELAP5 design evaluation employs a combination of proven RELAP5 features, models, and components as well as new, untested, advanced operating features and components requiring new and unique correlations and models to simulate the needed conditions and behavior. Of particular importance is the use of the containment vessel as an integral part of the ECCS and the modeling of wall condensation under high pressure conditions. However, for LOCA, since the NPM should maintain adequate reverse flow in the RRVs such that collapsed liquid remains above the core throughout the transient, the specific condensation and CHF modeling should not be a significant issue. The review of NRELAP changes focused on the items 1 through 5 but included many other changes made to the code.

Additionally, since the INLRELAP5-3D code is also currently under development, NuScale implements periodic updates and bug fixes related to the generic INL RELAP5-3D code as well as NuScale specific changes in accordance with its Appendix B quality control standards. The NRC staff audited NuScales NRELAP5 Version 1.3 Software Release Notes document SwRN-0304-51578 and the NRC staff decided to examine only the changes relevant for phenomena that are important for LOCA, including namely, (1), (3), and (4). The audit included the applicable NCIs (NRELAP5 Change Implementation) and NRELAP5 source coding was provided on a workstation in the NuScale Rockville office. Version Description NCI Beta 00.01 Appendix K Moody NCI-1214-9943 Based on the NRC staffs examination of SwRN-0304-51578, Revision 0, many of the early changes to the code are implemented by the INL staff, the original code developers. The NRC staff notes that INL has vast experience and knowledge of the code. The changes for Moody implementation are based on previous INL work, document LTD-14-3183. The changes included adding two additional options for choking models, i.e., c=3 ((( 

                                                                                         }}, and c=4 (((
               )}}. The subroutine jchoke.F contains the coding for the implementation of the choking criterion. This implementation numerically imposes the choking criterion on the junctions determined to be in a choked state based on user input. The NRC staff examined changes made and the results of regression test cases. The preferred option is c=3 with the interpolation, which was used by NuScale for the LOCA analyses.

Version Description NCI Beta 00.02 ANS 1971 Decay Heat NCI-0115-10047

21 Beta 00.06 ANS 1971 Decay Heat Fix NCI-0115-10047 In the base RELAP5-3D code, a conservative approximation to the 1973 ANS Standard fission product data is used as the default. Appendix K to 10 CFR Part 50, indicates that heat generation rates from radioactive decay of fission products shall be assumed to be equal to 1.2 times the values for infinite operating time in the ANS Standard (Proposed American Nuclear Society Standards Decay Energy Release Rates Following Shutdown of Uranium-Fueled Thermal Reactors. Approved by Subcommittee ANS-5, ANS Standards Committee, issued October 1971). The subroutine dcyans71.F was added for the implementation of the ANS-1971 standard. The NRC staff examined the code changes against the referenced standard formulations and noted that NuScale has submitted an exemption request to instead use ANS-73 with a multiplier of 1.2. The specific code modification reviewed, focused on the following versions and NCIs: Version Description NCI Beta 00.10 Critical Heat Flux (CHF) NCI-0415-13361 v1.0RC2 CHF final Revision 1 NCI-0415-13361 v1.1RC15 (( }} CHF NCI-0316-48153 Correlation v1.1 Version Release with CHF NCI-0416-48924 Fix v1.2RC2 CHF Update NCI-0616-49518 v1.3 Implementation of CHF NCI-0916-51506 NCRs The NuScale LOCA EM limits core heat transfer modeling to pre-CHF phenomena only. If a scenario can be postulated where CHF limits are exceeded during a LOCA, this EM would be invalidated. Therefore, the applicant must demonstrate that fuel does not closely approach or experience CHF condition, i.e., collapsed water levels must always remain above the top of active fuel, the containment remains intact, and pressure and temperature in the CNV remain below design limits. This allows the applicant to apply for exemptions to many Appendix K criteria that would normally be required for PWR LOCA analysis methods where significant heat-up generally always occur during postulated LOCAs. As noted for NPM, inventory to prevent heat-up is provided by return flow from the CNV via the RRVs. The NuScale design does not have large piping typical to PWRs so the range of breaks considered for LOCA is limited to small piping that eliminates the gross loss of RCS inventory and the need for ECCS injection pumps. Based on the NRC staffs audit of the NCI-0316-48153 and NCI-0616-49518 change assessments, the applicant initially installed the (( }} CHF correlations for application the NuScale by adding heat transfer geometry options 170-176. The CHF correlations were interfaced with other correlations, e.g., (( 

      }}, based on local mass flow rates. These changes were implemented by adding subroutines ((                                                    }} in 2015.

In early 2016, the applicant elected to replace the (( 

                }}. The motivation being that ((              }} is endorsed by Appendix K and would allow for easier and expedited reviews by the NRC staff, even though the correlation was

22 cited by Appendix K for the BWR application and is being applied by the applicant to the PWR, at a significantly higher pressure. Consequently in 2016, the applicant added (( 

                                                                          }}.

(( 

                                                                                             }}

(( 

                                                       }} The NRC staff audited these changes in NCIs and in ER-0000-2413, CHF Modeling from Sterns Data.

The NRC staff understands, based on the Chapter 6.2 containment public audit discussions (ML18141A735) that NuScale is planning to change the reference version of NRELAP5 v1.3 to version 1.4. The NRC staff are still in process of reviewing these latest code changes. 8.0 EXIT BRIEFING The NRC staff conducted an audit closing meeting at NuScales Rockville, Maryland, office on March 15, 2018. During the meeting, the NRC staff reiterated the purpose of the audit and discussed the audit activities and expected outcome. The NRC staff also indicated that the safety significant issues at the end of the audit would be transmitted as questions in RAIs. References to the detailed questions are provided in Section 9 of this audit summary. In summary, the NRC staff examined more than one hundred technical reports, calculation notes and supporting documents in paper and in ERR, performed an onsite audit in July 2017, and held 29 weekly teleconferences to discuss various topics and issues as they emerged and developed during the audit. In general, the NRC staff gained tremendous insights regarding the NuScale design and its LOCA methodology. With NuScales strong support and corporation during the audit, the NRC staff was able to resolve many technical issues, optimizing the number of RAIs that needed to be issued. 9.0 REQUESTS FOR ADDITIONAL INFORMATION RESULTING FROM AUDIT The NRC staff issued 15 questions in 5 different RAIs that either arose from or were informed by observations made during the audit. These RAIs are available in ADAMS and the accession numbers are provided in Table 1, below. The questions in the RAIs that are not explicitly listed below resulted from the NRC staffs review, independent of the audit. Table 1. RAIs Resulting from Audit

23 ADAMS RAI Requests Summary Accession Number Number 8776 Question 15.06.05-2 ML17291B322 Question 15.06.05-3 Question 15.06.05-4 Question 15.06.05-5 Question 15.06.05-6 8777 Question 15.06.05-1 ML17121A404 8785 Question 15.06.05-1 ML17146B301 8786 Question 15.06.05-2 ML17146B303 8985 Question 15.06.06-1 ML17310B504 8990 Question 15.06.05-7 ML17324B391 9065 Question 15.06.05 ML17353A950 9085 Question 15.06.05-1 ML18271A157 9126 Question 15.06.05-9 ML18031B318 9149 Question 15.06.05-10 ML18030B253 9190 Question 15.06.05-11 ML18038B602 9208 Question 15.06.05-14 ML18134A250 Question 15.06.05-15 Question 15.06.05-16 

 *9317 Question 06.02.01.01.A-8               ML18256A360
 *9351 Question 15.00.02-33                   ML18194A747
 *9380 Question 06.02.01.01.A-5               ML18061A073 9390 Question 15.06.05-18.                  ML18166A349 Question 15.06.05-18
 *9466 Question 15.00.02-7                    ML18128A341 9475 Question 15.06.05-17                   ML18148A003 9476 Question 15.06.05-13                   ML18128A388
 *9482 Question 06.02.01.01.A-20.             ML18125A003

24 ADAMS RAI Requests Summary Accession Number Number 9492 Question 15.06.05-21 ML18199A193 

   *9494 Question 06.02.01.01A-16                                                 ML18125A002 9519 Question 15.06.05-20                                                     ML18167A012
      *These RAIs are not directly related to the LOCA TR review but resulted from this audit.

10.0 OPEN ITEMS AND PROPOSED CLOSURE PATHS Not applicable. 11.0 DEVIATIONS FROM THE AUDIT PLAN The audit was originally scheduled to exit following the original audit end date of January 15, 2018, but was extended to March 15, 2018, to accommodate the examination and discussion of additional documentation requested by the NRC staff regarding the SIET tests.

12.0 REFERENCES

1. Audit Plan for the Regulatory Audit of NuScale Topical Report TR-0516-49422-P, LOSS-OF-COOLANT ACCIDENT EVALUATION MODEL, issued May 30, 2017 (ADAMS Accession No. ML171436A141).
2. Loss-of-Coolant Accident Evaluation Model, NuScale Topical Report TR-0516-49422-P, Revision 0, issued December 2016.
3. Shah, M.M., 2009. An Improved and Extended General Correlation for Heat Transfer during Condensation in Plain Tube, HVAC&R Research, Vol. 15, No. 5, pp. 889-913.
4. NRO-REG-108, Regulatory Audits, issued April 2, 2009 (ADAMS Accession No. ML081910260).
5. Audit Summary Report for NuScale Power, LLC Pre-Application Activities Associated with NuScale Emergency Core Cooling System/Containment Performance Testing at NIST-1 Test Facility (PROJ0769), issued June 30, 2016 (ADAMS Accession No. ML16272A136).
6. Wolverine Tubes engineering data book, Wieland-Werke AG, 2016.
7. Subchannel Analysis Methodology, NuScale Topical Report TR-0915-17564-P, Revision 0, issued October 2016.
8. Containment Response Analysis Methodology Technical Report, NuScale Technical Report TR-0516-49084-P, Revision 0.}}
Retrieved from "https://kanterella.com/w/index.php?title=ML20010D133&oldid=2514544"