W3F1-2017-0065, License Amendment Request for Use of RAPTOR-M3G Code for Neutron Fluence Calculations
ML17332A898 | |
Person / Time | |
---|---|
Site: | Waterford ![]() |
Issue date: | 11/28/2017 |
From: | Dinelli J Entergy Operations |
To: | Document Control Desk, Office of Nuclear Reactor Regulation |
References | |
W3F1-2017-0065 | |
Download: ML17332A898 (102) | |
Text
17265 River Road Killona, LA 70057-3093 Tel 504-739-6660 Fax 504-739-6678 jdinelli@entergy.com Site Vice President Waterford 3 10 CFR 50.90 W3F1-2017-0065 November 28, 2017 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555
Subject:
LICENSE AMENDMENT REQUEST FOR USE OF RAPTOR-M3G CODE FOR NEUTRON FLUENCE CALCULATIONS Waterford Steam Electric Station, Unit 3 (Waterford 3)
Docket No. 50-382 License No. NPF-38
Dear Sir or Madam:
In accordance with the provisions of Section 50.90 of Title 10 of the Code of Federal Regulations (10 CFR), Entergy Operations, Inc. (Entergy) is submitting a request for approval for use of the RAPTOR-M3G code for performing reactor vessel neutron fluence calculations for Waterford Steam Electric Station, Unit 3 (Waterford 3). RAPTOR-M3G is a 3 dimensional (3D),
parallel-processing discrete ordinates radiation transport code developed by Westinghouse.
The proposed change revises Section 4.3.3 of the Final Updated Safety Analysis Report (UFSAR) to indicate that the RAPTOR-M3G code is used for the reactor vessel fluence calculations. Attachment 1 contains markups of all of the affected UFSAR sections impacted by this change and Attachment 2 contains the evaluation of the proposed change.
Additionally, Waterford 3 is providing information in Attachment 3 which is not part of the license amendment request but is relevant information pertinent to the review of the Waterford 3 license renewal application submitted by Entergy on March 23, 2016. The inclusion of this information was discussed at the presubmittal meeting held between the NRC and Entergy on October 19, 2017.
Waterford 3 requests approval of the request to incorporate RAPTOR-M3G into the licensing basis by May 1, 2018.
concerning the issue of no significant hazards consideration using the standards in 10 CFR 50.92 is included in Attachment 2.
W3F1 -2017-0065 Page 2 of 2 In accordance with 10 CFR 50.91(b)(1), "Notice for Public Comment; State Consultation," a copy of this application and its reasoned analysis about no significant hazards considerations is being provided to the designated Louisiana Official.
This letter contains no new commitments.
If you have any questions or require additional information, please contact John Jarrell, Regulatory Assurance Manager, at 504-739-6685.
I declare under penalty of perjury that the foregoing is true and correct. Executed on November 28, 2017.
Proposed Changes to UFSAR Sections Evaluation of Proposed Change 10 CFR Part 54 License Renewal Supplemental Discussion cc: Mr. Kriss Kennedy, Regional Administrator U.S. NRC, Region IV RidsRgn4MailCenter@nrc.gov U.S. NRC Project Manager for Waterford 3 April. Pulvirenti@nrc.gov U.S. NRC, Division of License Renewal Phyllis.Clark@nrc.gov Elaine. Keegan@nrc.gov U.S. NRC Senior Resident Inspector for Waterford 3 Frances.Ramirez@nrc.gov Chris.Speer@nrc.gov Louisiana Department of Environmental Quality Office of Environmental Compliance Surveillance Division Ji.Wiley@LA.gov
to W3F1-2017-0065 Page 1 of 3 TABLE 4.1-1 (Sheet 3 of 3)
(LBDCR 17-020, R310)
ANALYSIS TECHNIQUES Internal Components Description Analysis Technique Reactor Core Static and time dependent Diffusion code using one dimensional (axial) up to 140 distinct studies, with control regions with variable rod motion mesh intervals. Feed-back, Eigenvalue searches and power shaping searches Reactor Vessel and Fast neutron flux and Combination of discrete Vessel Internals fluence ordinate transport and point kernel codes using core power distributions from PDQ-X Reactor Core Xenon stability analysis Linear modal analysis employing the funda-mental and first harmonic modes
- ANISN-SHADRAC was used for initial vessel radiation calculations. The most recent radiation calculations were performed when surveillance capsule W the DORT RAPTOR-M3G(15) code (LBDCR 17-020, R310) to W3F1-2017-0065 Page 2 of 3 4.3.2.8 Vessel Irradiation The design of the reactor internals and of the water annulus between the active core and vessel wall is such that for reactor operation at the full power rating and an 80 percent capacity factor, the vessel fluence greater than 1 MeV at the vessel wall will not exceed 3.68 x 1019 n/cm2 over the 40 year design life of the vessel. The calculated exposure includes a 10 percent uncertainty factor.
(LBDCR 16-060, R310)
The initially-calculated maximum fast neutron fluxes greater than one MeV incident on the vessel ID and shroud ID are as shown in Table 4.3-10. The fluxes are based on a time averaged equilibrium cycle radial power distribution and an axial power distribution with a peak to average of 1.20. The calculation assumed a thermal power of 3560 MWt. The models used in these calculations, ANISN and SHADRAC, are discussed in Subsection 4.3.3.3.
(LBDCR 16-060, R310)
(LBDCR 17-020, R310)
Updated reactor vessel fluence analysis was performed in conjunction with the testing of surveillance capsule W -263 83 at the end of cycle 11 19. This analysis used the DORT RAPTOR-M3G code to predict a peak vessel wall fluence of 2.48 2.57 x 1019 n/cm2 (E > 1 MeV) at the end of the 40-year vessel design life. Thermal power assumptions were 3390 MWt for the first 11 cycles, 3441 MWt for the 12th and 13th cycles, and 3716 MWt from cycle 14 through the end of the design life. Fuel enrichment, core power distributions, and fuel burnup were based on cycle-specific fuel designs and used to develop spatial- and energy-dependent core source distributions, which were averaged over each fuel cycle. The DORT RAPTOR-M3G model used in this calculation is discussed in Subsection 4.3.3.3.
(LBDCR 17-020, R310) to W3F1-2017-0065 Page 3 of 3 4.3.3.3 Reactor Vessel Fluence Calculation Model (LBDCR 16-060, R310)
The initially calculated vessel fluence was obtained by combining the results of ANISN (41) and SHADRAC(19) in the following manner:
(LBDCR 16-060, R310)
(6) where:
(E) is the neutron energy flux at the inner surface of the vessel, (ANISN) is the neutron energy flux obtained from ANISN, A (SHADRAC) is the neutron energy flux as calculated by SHADRAC in which the exact source geometry and a three-dimensional time averaged power distribution are used.
B (SHADRAC) is the neutron energy flux as calculated by SHADRAC using a cylindrical source geometry and the power distribution obtained from ANISN.
The neutron flux as calculated by the above method has uncertainty limits of +10 percent, -40 percent. The total uncertainty is composed of 0 percent, -30 percent in the calculational method and 10 percent uncertainty in the combined radial and axial power distribution. The calculational uncertainty factors are obtained by comparing the ANISN-SHADRAC results with measurements from various operational reactors (42).
(LBDCR 17-020, R310)
Calculation of vessel fluence in conjunction with analysis of surveillance capsule W -263 83 was performed using the (47) (48)
DORT RAPTOR-M3G code and BUGLE-96 cross-section library . DORT RAPTOR-M3G is a three-dimensional discrete ordinates transport code that takes advantage of parallel processing to solve complex geometric problems efficiently. two-dimensional discrete ordinates transport code. Synthesis of a solution in 3 dimensions is accomplished by combining cycle-specific forward transport calculations using the following method:
Where is the synthesized three dimensional flux distribution, is the transport solution in geometry, is the two-dimensional solution for a cylindrical reactor model using the actual axial core power distribution, and is the one-dimensional solution for a cylindrical reactor model using the same source per unit height as that used in the calculation.
(49,50)
The DORT RAPTOR-M3G method meets the requirements of Reg. Guide 1.190 and was approved by the NRC for industry use. The calculation method uncertainties were calculated based on the requirements of the Reg. Guide and were determined to be 14%; comparisons of measured and calculated fluence for the WF3 surveillance capsules (51) show an average M/C ratio of 1.11 and standard deviation of 7% 12% for the surveillance capsule fluence and (51) 13% for the WF3 reactor vessel fluence . No systematic bias was applied to the analysis results. WF3-specific dosimetry measurements from surveillance capsules W-97 and W-263 validate the uncertainties.
(LBDCR 17-020, R310)
Attachment 2 to W3F1-2017-0065 Evaluation of Proposed Change (63 Pages) to W3F1-2017-0065 Page 1 of 63 Summary Description This evaluation supports a request to amend the Operating License for Waterford 3 (WF3) located in Killona, LA.
The proposed change would revise the Operating License to reflect the use of the RAPTOR-M3G code for reactor vessel beltline neutron fluence calculations.
Detailed Description 1.0 WF3 Reactor Vessel The Waterford 3 reactor vessel is constructed of nine SA533 Grade (Gr.) B, Class (Cl.) 1 rolled plates welded together in lower, intermediate, and upper cylindrical shells. The lower shell section is welded to an SA-533 Gr. B, Cl. 1 bottom head torus and bottom head dome. The upper shell section has six SA508 Cl. 2 nozzles welded to its circumference and the SA508 Cl. 2 vessel flange welded to its top. The reactor vessel closure head (RVCH) is a single forging made of SA508 Gr. 3, Cl. 1 steel and bolted to the vessel flange. The interior surfaces of the reactor vessel are clad with austenitic stainless steel containing >5% delta ferrite 35 .
The reactor vessel functions to maintain the reactor coolant system pressure boundary and inventory and to support the weight of the reactor core under normal and design basis conditions.
2.0 Reactor Vessel Surveillance Program Material Testing Ferritic materials, such as SA533 and SA516, experience degradation of fracture toughness due to high energy {E>1 MeV) neutron irradiation. Additionally, steels are known to experience large ductile-to-brittle transitions due to service temperature changes. The WF3 Reactor Vessel Fracture Toughness and Surveillance Material Testing (commonly referred to as reactor vessel integrity) program 36 was developed to track and predict fracture toughness degradation in accordance with 10CFR50 App. H37 and to facilitate the development of pressure-temperature limit curves for RCS operation. The reactor vessel integrity program compares tensile and Charpy V-notch material test results from irradiated material housed in the reactor to the properties of the unirradiated material of the same origin. The program is governed by 10CFR50 App. H, which references ASTM E185-82 38 for specific instructions for execution of the program. ASTM E185-82 requires that three surveillance capsules be withdrawn from the reactor at specified times throughout the initial license period for the WF3 reactor conditions.
Appendix H references Appendix G of 10CFR50 39 for fracture toughness requirements and acceptance criteria.
3.0 Reactor Vessel Surveillance Program Fluence Calculation Methods The properties discussed in Section 2.0 are projected for future years of operation with respect to neutron fluence, and thus the neutron irradiation to which the to W3F1-2017-0065 Page 2 of 63 reactor vessel will be subjected must be calculated. Simulation methods are used to accurately model the complex geometry and interactions of nuclear particles.
The computer codes DORT and TORT are industry-standard methods for calculating the neutron fluence, and they were accepted for general use by the NRC in the SER for WCAP-14040 Rev. 340 . DORT is a two-dimensional discrete ordinates transport code, which calculates fluence in the (r, 8) plane and (r, z) plane separately and combines the two results. TORT is a three-dimensional discrete ordinates transport code, which calculates the fluence over the entire (r, 8, z) domain simultaneously. The second surveillance capsule withdrawn from the WF3 reactor was analyzed using the DORT code30 *31 . The third surveillance capsule (83°) was analyzed using the RAPTOR-M3G code, which is a 3-dimensional code like TORT; however, it uses parallel processing for enhanced computational efficiency23 *24 .
A review of the change from the DORT method to the RAPTOR-M3G method has determined that a license amendment is required.
4.0 Description of Proposed Amendment The proposed license amendment will incorporate into the WF3 current licensing basis the use of the RAPTOR-M3G method for the purpose of calculating neutron fluence in the reactor vessel beltline. Upon approval of this license amendment, the RAPTOR-M3G code will be referenced as the neutron fluence calculation method in the WF3 UFSAR. The neutron fluence values calculated using RAPTOR-M3G for 32 effective full power years (EFPY) of plant operation (approximate end of the WF3 current license period) will also be incorporated into the UFSAR. This includes both direct references to the 32 EFPY fluence and material properties for the end of the current license period calculated using the 32 EFPY fluence. The new calculated material properties do not affect any 10 CFR 50.61 or 10 CFR 50 Appendix G screening criteria for Pressurized Thermal Shock (RT PTs) or Upper Shelf Energy (USE), and the current WF3 pressure-temperature (P-T) curves are still valid until 32 EFPY.
Technical Evaluation 1.0 Waterford 3 Current Licensing Basis The current reactor vessel neutron fluence calculation method listed in the Waterford-3 UFSAR is DORT3 . The DORT code was used to perform radiation analysis of surveillance capsule 263° 30 , which was removed from the WF3 reactor vessel at 13.83 EFPY of operation. The outputs of the capsule 263° analysis were used to calculated the current WF3 heatup and cooldown (P-T) curves 31 approved in license amendment 19632 . The basis for acceptance of neutron fluence calculation methods is Regulatory Guide 1.190, "Calculational and Dosimetry Methods for Determining Pressure Vessel Neutron Fluence"4 .
to W3F1-2017-0065 Page 3 of 63 2.0 Regulatory Guide 1.190: Neutron Fluence Calculation Methods Reactor vessel neutron fluence has traditionally been quantified using discrete ordinates radiation transport calculations. Codes used to perform early calculations include TWOTRAN 1 and DOT2 . With the limitations on computing power at the time, both TWOTRAN and DOT were only capable of analyzing one-dimensional and two-dimensional models. In the 1980s, Oak Ridge National Laboratory developed the DORT (two-dimensional) and TORT (three-dimensional) codes 3 , and these codes remain in widespread use today.
The methodology employed by RAPTOR-M3G is essentially the same as the methodology employed by the TORT code, with solution enhancements resulting from the last two decades of research. RAPTOR-M3G has been designed from its inception as a parallel-processing code, and adheres to best practices of software development. It has been rigorously tested against the TORT code and benchmarked on an extensive set of academic and real-world problems.
The methodology used to provide neutron exposure evaluations for the reactor pressure vessel (RPV) follows the guidance provided in Regulatory Guide 1.1904.
The following discussion will show that the use of RAPTOR-M3G at WF3 satisfies all regulatory positions in Table 1 of Regulatory Guide 1.190 pertinent to deterministic neutron fluence calculation methods. Table 1 shows a summary of the Regulatory Positions in Table 1 of Regulatory Guide 1.190.
to W3F1-2017-0065 Page 4 of 63 Table 1: Summary of Regulatory Positions relevant to fluence calculations in Regulatory Guide 1.190 Reg.
Topic Description Position 1.3 Fluence Determination Use fluence calculation, not extrapolated measurements 1.1.1 Modeling Data Modeling based on plant specific data 1.1.2 Nuclear Data Use latest Evaluated Nuclear Data File {ENDF/B) 1.1.2 X-Section Angular Use minimum P3 angular decomposition Representation 1.1.2 X-Section Group Collapse Demonstrate adequacy of collapsed job library
{BUGLE-96) 1.2 Neutron Sources Neutron source model attributes requirements 1.2 End-of-Life Predictions Use best estimate or conservative power distribution 1.3.1 Spatial Representation Discrete ordinate neutron transport calculation minimum requirements 1.3.1 Multiple Transport N/A- Bootstrapping not performed Calculations 1.3.2 Point Estimates N/A- only applies to Monte Carlo analysis, which was Statistical Tests not used Variance Reduction 1.3.3 Capsule Modeling Capsule Geometry modeling requirements 1.3.3 Spectral Effects on RTNDT Use of spectral lead factor required 1.3.5 Cavity Calculations N/A - No dosimeters in Reactor Cavity 1.4.1 Analytic Uncertainty Analysis Sensitivity analysis of components comprising uncertainty 1.4.2 Comparisons with Benchmark Validation of method against plant model, vessel Measurements and simulator model, calculational benchmark model Calculations 1.4.3 Estimate of Fluence Requirement to meet 20% {maximum) vessel fluence Calculational Bias and uncertainty for RTPTS and RTNDT Uncertainty to W3F1-2017-0065 Page 5 of 63 2.1 Fluence Determination Regulatory Position 1.3 of Regulatory Guide 1.190 can be summarized as follows: Absolute fluence calculations, rather than extrapolated fluence measurements, must be used for the f/uence determination.
RAPTOR-M3G solves the time-independent Linear Boltzmann Equation (LBE) in the absence of fission in three dimensions via the discrete ordinates approximation. The method of discrete ordinates is described in detail in Reference 3.
RAPTOR-M3G calculations are performed with an Sa (or higher) level-symmetric angular quadrature set. Neutron fluence values are determined directly from the results of the radiation transport calculations performed with RAPTOR-M3G. For Waterford 3, the calculations were performed in directional theta weighted (DTW) mode with an S16 level-symmetric angular quadrature set17 .
The RAPTOR-M3G calculations described in this section comply with Regulatory Position 1.3 of Regulatory Guide 1.190.
2.2 Modeling Data Regulatory Position 1.1.1 of Regulatory Guide 1.190 is summarized as follows: To the extent possible, plant-specific as-built dimensions and materials should be used as model inputs.
The modeling data for the Waterford 3 reactor vessel is described in WCAP-17969-NP Rev. 1, "Analysis of Capsule 83° from the Enter~¥ Operations, Inc.
Waterford Unit 3 Reactor Vessel Surveillance Program" . Six irradiation capsules attached to the pressure vessel inside wall are included in the reactor design that constitutes the reactor vessel surveillance program. The capsules are located at azimuthal angles of 83°, 97°, 104°, 263°, 277°, and 284 °. These full-core positions correspond to the following octant symmetric locations represented in Figure 1: 7° from the core cardinal axes (for the 83°,
97°, 263° and 277° capsules) and 14° from the core cardinal axes (for the 104° and 284° capsules). The stainless steel specimen containers are 1.402-inch by 0.652-inch and are approximately 98 inches in height. The containers are positioned axially such that the test specimens are centered 6.25 inches above the core midplane, thus spanning the approximate central eight feet of the 12.5-foot-high reactor core 17 .
In performing the fast neutron exposure evaluations for the Waterford 3 reactor vessel and surveillance capsules, a series of fuel-cycle-specific forward transport calculations were carried out using a three-dimensional geometrical reactor model. For the Waterford 3 transport calculations, the r,8 ,z models depicted (given as r,8 plan view) in Figure 1 and Figure 2 were to W3F1-2017-0065 Page 6 of 63 utilized since, with the exception of the capsules, the reactor is octant symmetric. The r,z section view depicted in Figure 3 shows the model having an axial span from an elevation 5.5 feet below the bottom of the active fuel to 5 feet above the top of the active fuel. These r,e ,z models include the core, the reactor internals, the surveillance capsules, the pressure vessel cladding and vessel wall, the insulation external to the pressure vessel, and the primary biological shield wall. These models formed the basis for the calculated results and enabled making comparisons to the surveillance capsule dosimetry evaluations. In developing these analytical models, nominal design dimensions were employed for the various structural components with a few exceptions 17
- The radius to the center of the surveillance capsule holder and the radius to the pressure vessel were taken from as-built drawings for the Waterford 3 reactor for key differences between the nominal and as-built dimensions. For the reactor pressure vessel, the minimum vessel thickness was used. Water temperatures, and hence, coolant densities in the reactor core and downcomer regions of the reactor were taken to be representative of full-power operating conditions.
The coolant densities were treated on a fuel-cycle-specific basis. The reactor core itself was treated as a homogeneous mixture of fuel, cladding, water, and miscellaneous core structures such as fuel assembly grids, guide tubes, et cetera. The stainless steel former plates located between the core baffle and core barrel regions are also included in the model. Sensitivities of the analytical results to tolerances in the internals dimensions and fluctuations in water temperature are discussed and quantified later in this document in section 2.16.1.
The geometric modeling described in this section and applied for Waterford 3 complies with Regulatory Position 1.1.1 of Regulatory Guide 1.19018 .
to W3F1 -2017-0065 Page 7 of 63
- Core - Stainless Stl - Bypass - Stainless Stl Downcomer
- Stainless SIi Carton Steel - Insulation Air - Concrete N
0 ir 0
(")
E
(.)
N 0
+
L1J 0
C!
(")
0 0
+
L1J 0
0 0
O.OOE+OO 3.00E+02 Figure 1: Waterford 3 r,9,z Reactor Geometry Plan View at the Core Midplane without Surveillance Capsules 17 to W3F1-2017-0065 Page 8 of 63
- Core - Stainle~ SU - Bypass - Stain!~ Stl Downcomer
- S:tainless Stl Carbon Steel - Insulation Air - Concrete N
~r 0
0
~
E
(.)
N 0
+
LU 0
0
(")
0 0
+
LU 0
0 ci 3.00E+02 cm O.OOE+OO 3.00E+02 Figure 2: Waterford 3 r,9 ,z Reactor Geometry Plan View at the Core Midplane with 7° and 14" Surveillance Capsules 17 to W3F1-2017-0065 Page 9 of 63 Lower Plenum - Bypass - Stainless St~I Oawncomer - Stainless Steel Carbon Steel - Insulation Concrete - Stainless S1ael
- Stair.less Steel - Core - Upper Plenum N
0
+
LU 0
'<:I:
C")
N 0
+
LU 0
(!)
C")
3.00E+02 cm _J R O.OOE+OO 3.00E+02 Figure 3: Waterford 3 r, 9, z Reactor Geometry Section View at 7deg Azimuthal Angle 17 to W3F1-2017-0065 Page 10 of 63 2.2.1 Coolant Temperatures In the analytical models used for the Waterford 3 transport calculations, water densities in the core, bypass, and downcomer regions were determined on a fuel-cycle-specific basis consistent with the average temperature rise in the fuel assemblies located on the periphery of the reactor core. Since the neutron fluence at the pressure vessel is dominated by leakage from these peripheral assemblies, the use of the peripheral water density in the analytical models is justified.
The core average and downcomer temperatures for each cycle, along with the radial power distribution data, were used to calculate the average relative power level for the peripheral fuel assemblies along with the core average, bypass, downcomer, and outlet temperatures used in the transport calculations. These temperatures and the average relative power are calculated using the following relationships:
Toutlet = Tavg + (Tavg - Tinlet)
Where: Tautlet = core outlet temperature Tavg = core average temperature Tinlet = core and vessel inlet temperature= Tdowncomer Trise = flT
- Pperipheral Where: Trise = temperature rise of peripheral assemblies Pperipheraz = relative power of peripheral assemblies Trise Tperipheral = Tinlet + Where: Tperipheral = average peripheral assembly temperature Trise Tbypass = (Tinlet +-2 -)
Where: Tbypass = bypass temperature The density of the reactor coolant in the region below the core is based on Tinlet* The density of the reactor coolant in the core region is based on Tperipheraz because neutron fluence at the pressure vessel is dominated by leakage from peripheral assemblies. The density of reactor coolant in the region above the core is based on Tautzet* The density of reactor coolant in the downcomer region and bypass region are based on Tdowncomer and Tbypass, respectively.
The temperature modeling described in this section is representative of Waterford 3 operating conditions and satisfies Regulatory Position 1.1.1 of Regulatory Guide 1.190.
to W3F1-2017-0065 Page 11 of 63 2.3 Nuclear Data Regulatory Position 1.1.2 (Nuclear Data) of Regulatory Guide 1.190 can be summarized as follows: The latest version of the ENDF/B (nuclear data file) should be used for determining nuclear cross-sections. Cross-section sets based on earlier or equivalent nuclear data sets that have been thoroughly benchmarked are also acceptable.
The transport calculations are carried out using the BUGLE-96 cross-section library5*17 . The BUGLE-96 library provides a 67-group coupled neutron-gamma ray cross-section data set produced specifically for light water reactor (LWR) application. Anisotropic scattering is treated with a minimum Pa Legendre expansion. For the Waterford 3 calculations, a (more detailed)
P5 Legendre expansion was used 17 . Number densities used in the representation of material mixtures for LWR applications are taken from the documentation provided with BUGLE-96 for carbon steel, stainless steel, concrete, zircalloy-4, and uranium dioxide. Other mixtures, such as lnconel-718, come from internal Westinghouse data sources 18 .
The BUGLE-87 cross-section library6 is an update to the BUGLE-96 library.
The primary difference between the two libraries is that the BUGLE-87 library is derived from ENDF-BNII.O nuclear data, whereas BUGLE-96 is derived from older ENDF/8-VI nuclear data. The energy group boundaries and techniques used to construct both libraries are the same.
In the documentation released with BUGLE-87, Oak Ridge National Laboratory (ORNL) analyzed the H. B. Robinson Unit 2, Pool Critical Assembly (PCA), and VENUS-3 benchmarks using BUGLE-96 and BUGLE-876. Calculations with both libraries were compared to measurements from reactions that cover high-energy portions of the neutron spectrum that are of greatest concern for reactor vessel integrity evaluations. The ORNL report demonstrates that the differences between BUGLE-96 and BUGLE-87 for high-energy neutron applications are minor. This finding is corroborated by comparisons that Westinghouse has performed by reviewing the data from the H.B. Robinson and VENUS-3 benchmarks performed by ORNL6. Thus, for applications that concern Regulatory Guide 1.190, the differences between BUGLE-96 and BUGLE-87 are not significant.
Westinghouse has completed an additional comparative study that revealed differences between BUGLE-87 and BUGLE-96 for low energy neutrons that resulted from discrepancies in the upscatter-removed BUGLE-87 library in the lower energy range. This discrepancy does not affect the validity of the BUGLE-87 library for reactor vessel integrity evaluations, but Westinghouse has chosen to continue using BUGLE-96 until the discrepancy with BUGLE-87 is resolved. The use of the BUGLE-96 cross-section library with the P5 Legendre expansion mode for Waterford 3 and comparisons to the newer to W3F1-2017-0065 Page 12 of 63 BUGLE-87 library described in this section complies with Regulatory Position 1.1.2 of Regulatory Guide 1.19018 .
2.4 Cross-Section Angular Representation Regulatory Position 1.1.2 (Cross-Section Angular Representation) of Regulatory Guide 1.190 can be summarized as follows: In discrete ordinates transport calculations, a P3 angular decomposition of the scattering cross section (at a minimum) must be employed.
As discussed in Section 2.3, the RAPTOR-M3G transport calculations for WF3 utilized a P5 Legendre expansion of the scattering cross-section 17 . This complies with the applicable provision of Regulatory Position 1.1.2 in Regulatory Guide 1.19018 .
2.5 Cross-Section Group Collapsing Regulatory Position 1.1.2 (Cross-Section Group Collapsing) of Regulatory Guide 1.190 can be summarized as follows: The adequacy of the collapsed job library must be demonstrated by comparing calculations for a representative configuration performed with both the master and job library.
Work by Oak Rid~e National Laboratory has established the adequacy of the BUGLE-96 library . Westinghouse has reviewed the calculations performed by ORNL for the VENUS-3 and H.B. Robinson benchmarks and verified that the differences between the BUGLE-96 and newer BUGLE-87 library are minor for high-energy applications. However, Westinghouse has chosen to continue with the use of the BUGLE-96 library because of discrepancies identified by Westinghouse between the two libraries in the lower energy range 18 . Therefore, the applicable provision of Regulatory Position 1.1.2 is met18 .
2.6 Neutron Source Regulatory Position 1.2 (Core Neutron Source) of Regulatory Guide 1.190 can be summarized as follows: The core neutron source should account for local fuel isotopes and moderator density. Neutron source normalization and energy dependence must account for the fuel exposure dependence of the fission spectra, number of neutrons produced per fission, and energy released per fission.
The spatial variation of the neutron source is obtained from a burnup weighted average of the respective power distributions from individual fuel cycles 18 . These spatial distributions include pinwise gradients for all fuel assemblies located at the periphery of the core and include a uniform or flat distribution for fuel assemblies interior to the core. The spatial component of the neutron source is transposed from Cartesian to cylindrical geometry by to W3F1-2017-0065 Page 13 of 63 overlaying the mesh schematic to be used in the transport calculation on the pin by pin array and then computing the appropriate relative source applicable to each cylindrical mesh interval.
The Cartesian-to-cylindrical transposition is accomplished by first defining a fine cylindrical mesh working array. The fl.rand fl.fJ mesh are chosen so that there is typically a 10 x 10 array of fine mesh over the area of each fuel pin at the core periphery. The coordinates of the center of each fine mesh interval and its associated relative source strength are assigned to the fine mesh based on the pin that is coincident with the center of the fine mesh. In the limit as fl.r and fl.fJ approach zero, this technique becomes an exact transformation.
Each space mesh in the cylindrical transport geometry is checked to determine if it lies totally within the area of a particular fine working mesh. If it does, the relative source of that fine mesh is assigned to the transport space mesh. If, otherwise, the transport space mesh covers a part of one or more fine mesh, then the relative source assigned to the transport mesh is determined by an area weighting process as follows:
n _ r-A*P*
i i i
- m-riAi Pm = the relative source assigned to transport mesh m A = the area of fine working mesh i within transport mesh m A= the relative source within fine working mesh i The energy distribution of the source is determined by selecting a fuel burnup representative of conditions averaged over the irradiation period under consideration and an initial fuel assembly enrichment characteristic of the core designs used over the applicable period. From the assembly burnup and initial U-235 enrichment, a fission split by isotope including U-235, U-238, Pu-239, Pu-240, Pu-241, and Pu-242 is derived; and, from that fission split, composite values of energy release per fission, neutron yield per fission, and fission spectrum are determined. These composite values are then combined with the spatial distribution to produce the overall absolute neutron source for use in the transport calculations.
The core neutron source definition methodology described in this section complies with Regulatory Position 1.2 in Regulatory Guide 1.19018 .
to W3F1-2017-0065 Page 14 of 63
- 2. 7 End-of-Life Predictions Regulatory Position 1.2 (End of Life Predictions) of Regulatory Guide 1.190 can be summarized as follows: Predictions of end-of-life fluence should be made with a best-estimate or conservative generic power distribution.
Future neutron fluence projections for the WF3 reactor vessel have been calculated at 32, 36, 40, 48, 55, and 60 EFPY (corresponding to the end of the Period of Extended Operation). These data include plant- and fuel-cycle-specific neutron exposures at the end of the 19th fuel cycle (corresponding to 24.66 EFPY), including the power uprates from 3390 MWt to 3441 MWt prior to cycle 12 and from 3441 MWt to 3716 MWt prior to cycle 14. The fluence projections are based on the assumption that the core power distributions and associated plant operating characteristics from cycles 17, 18, and 19 are representative of future plant operation. The current power level of 3716 MWt plus an additional 5% positive bias applied to power generated by the peripheral fuel assemblies is assumed for conservatism 17 .
Therefore, the end-of-life vessel fluence prediction complies with Regulatory Position 1.2 in Regulatory Guide 1.19019 .
2.8 Spatial Representation Regulatory Position 1.3.1 (Spatial Representation) of Regulatory Guide 1.190 can be summarized as follows: Discrete ordinates calculations should incorporate mesh of 2 and 3 intervals/in at core periphery and water, respectively. Steel mesh should include at least 1. 5 intervals/in in steel.
Axial mesh of 0.5 intervals/in is acceptable, except where material and source interfaces cause high flux gradients. Calculations must employ at least an S8 quadrature and at least 40 intervals per octant.
The geometric mesh description of the reactor model is normally accomplished using from 150 to 250 radial, 80 to 150 azimuthal, and 100 to 200 axial intervals depending on the overall size of the reactor and on the complexity required to model the core periphery, the in-vessel surveillance capsules, and the details of the reactor cavity. Mesh sizes are chosen to assure that proper convergence of the inner iterations is achieved on a pointwise basis. The pointwise inner iteration convergence criterion utilized in the transport calculations is set at a value of 0.001.
The Waterford 3 model uses 160 radial mesh intervals, 121 azimuthal mesh intervals, and 247 axial mesh intervals in an azimuthal domain of one octant.
The peripheral assemblies and water regions were modeled with approximately 4 intervals per inch, and the steel regions were modeled with approximately 1. 7 intervals per inch. Axial meshes were more than 0.5 to W3F1-2017-0065 Page 15 of 63 intervals per inch, with the exception of regions at 282.38 cm and above from the core midplane that contained 0.4 intervals per inch. The axial distance of 282.38 cm is above the bottom of the vessel support pad of the outlet nozzle.
The support pad location was conservatively used when reporting the fluence at the nozzle weld; therefore, the axial mesh does not coarsen until above the limiting axial location for reporting nozzle area fluence. Although the use of 0.4 intervals per inch is slightly less detailed than the suggested 0.5 interval/in, it is acceptable because the model achieved geometrical convergence, i.e., the calculated fast neutron fluence results at the regions of interest do not change significantly with finer mesh structure20 . Angular discretization was modeled with an S16 order of angular quadrature. Nominal design dimensions were used to model most steel components, as pertinent as-built information was not readily available for Waterford 3. The exception to this is that the as-built vessel inner radius and in-vessel surveillance capsule positions were used. The mesh selection process results in a smaller spatial mesh in regions exhibiting steep gradients, in material zones of high cross section (rt), and at material interfaces. In modeling the stainless steel baffle region at the periphery of the core, a relatively fine spatial mesh is required to adequately describe this rectilinear component in cylindrical geometry. In performing this Cartesian to cylindrical transition, care is taken to preserve both the thickness and volume of the steel region in order to accurately address the shielding effect of the steel 18 .
The spatial representation of the of the transport problem ~eometry complies with Regulatory Position 1.3.1 in Regulatory Guide 1.190 1 . An exception is the use of 0.4 intervals/in at axial locations >282.38 cm above the core midplane; however, this is acceptable because the model achieved geometric convergence.
2.9 Multiple Transport Calculations Regulatory Position 1.3.1 (Multiple Transport Calculations) of Regulatory Guide 1.190 can be summarized as follows: If the calculation is performed using two or more "bootstrap" calculations, the adequacy of the overlap regions must be demonstrated.
The transport calculation was not performed using bootstrapping; therefore, the part of Regulatory Position 1.3.1 of Regulatory Guide 1.190 which addresses bootstrapping does not apply19 .
2.1 O Point Estimates Regulatory Position 1.3.2 (Point Estimates) of Regulatory Guide 1.190 can be summarized as follows: If the dimensions of the tally region or the definition of the average-flux region introduce a bias in the tally edit, the Monte Carlo prediction should be adjusted to eliminate the bias.
to W3F1-2017-0065 Page 16 of 63 RAPTOR-M3G is a deterministic fluence calculation method. No Monte Carlo techniques are used, and therefore, Regulatory Position 1.3.2 does not apply1s_
2.11 Statistical Tests Regulatory Position 1.3.2 (Statistical Tests) of Regulatory Guide 1.190 can be summarized as follows: The Monte Carlo estimated mean and relative error should be tested and satisfy all statistical criteria.
RAPTOR-M3G is a deterministic fluence calculation method. No Monte Carlo techniques are used, and therefore, Regulatory Position 1.3.2 does not apply1s_
2.12 Variance Reduction Regulatory Position 1.3.2 (Variance Reduction) of Regulatory Guide 1.190 can be summarized as follows: All variance reduction methods should be qualified by comparison with calculations performed without variance reduction.
RAPTOR-M3G is a deterministic fluence calculation method. No Monte Carlo techniques are used, and therefore, Regulatory Position 1.3.2 does not apply1s_
2.13 Capsule Modeling Regulatory Position 1.3.3 (Capsule Modeling) of Regulatory Guide 1.190 can be summarized as follows: The adequacy of the capsule representation and mesh must be demonstrated.
In the modeling of in-vessel surveillance capsules, a sufficiently fine mesh grid is employed within the test specimen array to assure that accurate information is produced for use in the assessment of fluence gradients within the materials test specimens, as well as in the determination of gradient corrections for neutron sensors. Additional radial and azimuthal mesh are employed to model the capsule structure surrounding the materials test specimen array.
The adequacy of the surveillance capsule modeling has been verified by the agreement between measured and calculated neutron exposures for the surveillance capsule library (which includes WF3 data) presented in section 2.16.4 18 . The surveillance capsule model complies with the applicable provision of Regulatory Position 1.3.3 in Regulatory Guide 1.19019 .
to W3F1-2017-0065 Page 17 of 63 2.14 Spectral Effects on RT NDT Regulatory Position 1.3.3 (Spectral Effects) of Regulatory Guide 1.190 can be summarized as follows: When fluence is extrapolated from inside surface of vessel to T/4 and 3T/4 locations using E > 1MeV f/uence, a spectral lead factor must be applied to the fluence for the calculation of L!RTNoT.
For WF3 fluence calculations, the neutron fluence is extrapolated from the inside surface of the reactor vessel to the 1/4T and 3/4T locations. Since Equation 3 in Regulatory Guide 1.99, Rev. 221 is used to determine the fluence at 1/4T and 3/4T 19 , the spectral lead factor is already incorporated into the extrapolated fluence. Therefore, the fluence values extrapolated from the inside vessel surface comply with Regulatory Position 1.3.3 of Regulatory Guide 1.19019 .
2.15 Cavity Fluence Calculations Regulatory Position 1.3.5 of Regulatory Guide 1.190 can be summarized as follows: In discrete ordinates transport calculations, the adequacy of the S8 angular quadrature used in cavity transport calculations must be demonstrated.
The RAPTOR-M3G method uses, at a minimum, an S8 level-symmetric angular quadrature for discrete ordinates transport calculations. An S15 level-symmetric angular quadrature was used for WF3 fluence calculations 17 .
However, Waterford 3 does not utilize ex-vessel dosimetry in its reactor vessel fluence measurement program. Thus, reactor cavity fluence was not determined using the RAPTOR-M3G method for WF3, and Regulatory Position 1.3.5 is not applicable to WF3 19 .
2.16 Methods Qualification The validation of the transport methodology in RAPTOR-M3G follows the guidance provided in Regulatory Guide 1.1904 . In particular, the validation consists of the following stages:
- 1. Simulator Benchmark Calculations: comparisons of calculations with measurements from simulator benchmarks, including the PCA simulator8 at ORNL and the VENUS-1 experiment10 .
- 2. Operating Reactor and Calculational Benchmarks: comparisons of calculations with surveillance capsule and reactor cavity measurements from the H. B. Robinson Unit 2 power reactor benchmark experiment12 and comparisons of calculations performed with RAPTOR-M3G to results published in the NRC's fluence calculation benchmark13 .
to W3F1-2017-0065 Page 18 of 63
- 3. Analytic Uncertainty Analysis: an analytic sensitivity study addressing the uncertainty components resulting from important input parameters applicable to the plant-specific transport calculations used in the exposure assessments.
At each subsequent application of the methodology, comparisons are made with plant-specific dosimetry results to demonstrate that the plant-specific transport calculations are consistent with the uncertainties derived from the methods qualification.
The first stage of the methods validation addresses the adequacy of basic transport calculation and dosimetry evaluation techniques and associated cross sections. This phase, however, does not test the accuracy of commercial core neutron source calculations, nor does it address uncertainties in operational or geometric variables that affect power reactor calculations. The second stage of the validation addresses uncertainties that are primarily methods related and would tend to apply generically to all fast neutron exposure evaluations. The third stage of the validation identifies the potential uncertainties introduced into the overall evaluation due to calculational methods approximations as well as to a lack of knowledge relative to various plant-specific parameters. The overall calculational uncertainty is established from the results of these three stages of the validation process. Final qualification of the calculation method was completed by comparing the measured and calculated (using RAPTOR-M3G) neutron fluences from a large database of pressurized water reactors.
2.16.1 Analytic Uncertainty Analysis Regulatory Position 1.4.1 of Regulatory Guide 1.190 can be summarized as follows: The calculation methodology must be qualified by an analytic uncertainty analysis to determine the sensitivity of the calculated flux to variations in model inputs.
Operating reactors are subject to several uncertainties that may influence the validity of the calculated neutron fluence results. The most significant among these are:
- Uncertainties in the core neutron source
- Uncertainties in the as-built thicknesses and locations of the reactor vessel and internal components
- Uncertainties in the full-power coolant temperatures (water density)
This listing of parameters is consistent with the findings of other neutron fluence uncertainty studies 14*15*16 . This section presents the results of a sensitivity study performed using RAPTOR-M3G that evaluate the impacts of to W3F1-2017-0065 Page 19 of 63 variations in the parameters listed above on calculated neutron fluence values. The sensitivity study is not specific to the WF3 reactor geometry or operation; however, it has been shown that the uncertainty analysis inputs (i.e., dimensional and operational variances) are similar when applied to WF3 20-this is discussed in greater detail in section 3.0. Although there are no cavity surveillance capsules at WF3, the results of the generic sensitivity study for the associated reactor cavity locations are included for completeness.
Note that the uncertainty analysis was performed for both the theta-weighted (TW) and DTW differencing schemes. In general, the analytic uncertainty values are consistent between the two differencing schemes. Since the WF3 analysis used the DTW differencing scheme, the results obtained using only DTW are incorporated into the net uncertainty calculation.
Analytic Uncertainty Analysis - Core Neutron Source Uncertainties To assess the impact of uncertainties in the core neutron source on calculated neutron fluence results, changes in the following parameters were evaluated 18 :
- Absolute source st rength of peripheral fuel assemblies - Studies have shown that the neutron fluence rate in regions external to the core is dominated by the neutron source from fuel assemblies on the core periphery. In-core measurements indicate that a source magnitude uncertainty of 5% is bounding 18 .
- Pin-by-pin spatial distributions of n eutron so urce at the core periphery - Core management studies indicate that uncertainties in the relative pin powers in peripheral fuel assemblies can be on the order of 10%.
- Burnup of the peripheral fuel assemblies - Perturbations in fuel assembly burnup impact the fission spectrum, neutron yield per fission, and energy released per fission for each peripheral fuel assembly. A 5000 MWD/MTU uncertainty in the peripheral fuel assembly burnups is considered conservative. The sensitivity study is performed using a series of calculations starting with midcycle burnup at 3000 MWD/MTU, and 5000 MWD/MTU to 50,000 MWD/MTU with 5000 MWD/MTU delta mid-cycle burnup between each run.
- Axial power distribution - Based on variations in axial peaking factors over the course of a fuel cycle, a 10% uncertainty in the shape of the axial power distribution is considered conservative.
Each case evaluated as part of the sensitivity study is described in Table 2.
The base case consisted of a low-leakage power distribution cycle from a to W3F1-2017-0065 Page 20 of 63 Westinghouse 4-Loop reactor. Table 3 through Table 5 provides the differences between calculated fast neutron (E > 1.0 MeV) fluence rate results at several locations for each permutation case, each normalized to the corresponding base case result. The overall uncertainty estimates are summarized in Table 6 through Table 8. Note that the results are rounded to the nearest whole percent.
Table 2: Summary of Core Neutron Source Sensitivity Study18 Case Description Number 1 Peripheral source strenQth biased by a factor of 0.95 2 Peripheral source strength biased by a factor of 1.05 3 Pin power distribution gradient diminished according to:
Pminus = [(P - 1.0) X 0.9] + 1.0 4 Pin power distribution gradient intensified according to:
Pminus = [(P - 1.0) X 1.1] + 1.0 511' Mid-cycle burnup at 3,000 MWD/MTU 6# Mid-cycle burnup at 50,000 MWD/MTU 7 Axial power distribution intensified according to :
Axialvzus = [(Axial - 1.0) X 1.1] + 1.0 8 Axial power distribution intensified according to :
Axialvzus = [(Axial - 1.0) X 0.9] + 1.0
- Cases 5 and 6 span a mid-cycle burnup range of 47000 MWD/MTU. The uncertainty in the neutron fluence attributable to a 5000 MWD/MTU uncertainty in burnup is obtained by scaling the difference between Cases 5 and 6 accordingly by F = (5000 / 47000). For example, Case 5 (mid-cycle burnup of 3000 MWD/MTU) results in a 7%
fluence rate decrease relative to the base case. Case 6 (mid-cycle burnup of 50000 MWD/MTU) results in a 1%
fluence rate increase relative to the base case. Therefore the fluence rate difference associated with a 47000 MWD/MTU difference is 8%. Scaled by 5000 / 47000, the fluence rate difference associated with a 5000 MWD/MTU difference is 1%.
to W3F1-2017-0065 Page 21 of 63 Table 3: Source Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Surveillance Capsule Locations 22 Case Surveillance Number Capsule Location 1 -4%
2 4%
3 1%
4 -1%
5 -7%
6 1%
7 1%
8 -1%
Table 4: Source Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Pressure Vessel Locations22 Case RPVInside Number Radius +12 cm Relative to Middle-of-Core Elevation*
1 -5%
2 5%
3 1%
4 -1%
5 -7%
6 2%
7 1%
8 -1%
- The selected location on the inner radius of the reactor pressure vessel is typical of circumferential welds that join base metal forgings.
to W3F1-2017-0065 Page 22 of 63 Table 5: Source Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Reactor Cavity Locations22 Case Reactor Cavity Reactor Cavity Reactor Cavity Number Top-of-Core Middle-of-Core Bottom-of-Core Elevation Elevation Elevation 1 -4% -5% -5%
2 4% 5% 5%
3 1% 1% 1%
4 -1% -1% -1%
5 -7% -7% -7%
6 2% 2% 2%
7 -3% 1% -2%
8 3% -1% 2%
Table 6: Summary of Neutron Fluence Rate Uncertainties at Surveillance Capsule Locations Resulting from Core Neutron Source Uncertainties22 Uncertainty Component Surveillance Capsule Location Peripheral Assembly Source +/-4%
Strength Pin Power Distribution +/-1%
Peripheral Assembly Burnup +/-1%
(+5000 MWD/MTU)
Axial Power Distribution +1%
to W3F1-2017-0065 Page 23 of 63 Table 7: Summary of Neutron Fluence Rate Uncertainties at Pressure Vessel Locations Resulting from Core Neutron Source Uncertainties22 Uncertainty RPVInner Component Radius +12cm Relative to Middle-of-Core Elevation Peripheral +/-5%
Assembly Source Strength Pin Power +/-1%
Distribution Peripheral +/-1%
Assembly Burnup
(+/-5000 MWD/MTU)
Axial Power +/-1%
Distribution Table 8: Summary of Neutron Fluence Rate Uncertainties at Reactor Cavity Locations Resulting from Core Neutron Source Uncertainties22 Uncertainty Reactor Cavity Top- Reactor Cavity Reactor Cavity Component of-Core Elevation Middle-of-Core Bottom-of-Core Elevation Elevation Peripheral Assembly +/-4% +/-5% +/-5%
Source Strength Pin Power +/-1% +/-1% +/-1%
Distribution Peripheral Assembly +/-1% +/-1% +/-1%
Burnup (+/-5000 MWD/MTU)
Axial Power +/-3% +/-1% +/-2%
Distribution Geometry and Temperature Uncertainties To assess the impact of uncertainties in the location and thickness of reactor components, as well as uncertainties in reactor coolant temperature, on calculated neutron fluence results, changes in the following parameters were evaluated 18 :
to W3F1-2017-0065 Page 24 of 63
- Reactor internals dimensions - Thickness tolerances on stainless steel reactor internals components (e.g., core baffle, core barrel, thermal shield/neutron pad) are typically specified as 1/16 inch or tighter.
- Reactor vessel inner radius - Reactor vessels typically specify an inner radius with tolerance bounds of -0.00 inches and +1/32 inches. A tolerance of+/- 1/8 inch is considered.
- Reactor vessel thickness - Some techniques for fabricating reactor vessels result in larger-than-nominal reactor vessel base metal plate thicknesses. A tolerance of+/- 1/16 inch is considered.
- Dosimetry Positioning - Surveillance capsules have a tolerance of +/-
1/16 inch associated with the positioning of the dosimetry in radial, azimuthal, and axial directions. A larger positioning uncertainty of +/- 2 inches is associated with ex-vessel neutron dosimetry in radial azimuthal, and axial directions.
- Coolant Temperature - Variations in water temperature over the course of a fuel cycle are expected to be less than+/- 10 °F.
- Core Peripheral Modeling - The modeling of the rectilinear core baffle in cylindrical geometry represents another potential source of uncertainty in the geometric modeling of the reactor. The sensitivity of the solution to the modeling approach is determined by a direct comparison of the results of a cylindrical geometry calculation with those of a Cartesian geometry calculation in which the baffle region and core periphery were modeled explicitly. The comparisons of interest were taken at various locations external to the core baffle, but inside the core barrel.
Each case evaluated as part of the sensitivity study is described in Table 9.
The base case consisted of a low-leakage power distribution from a Westinghouse 4-Loop reactor. Table 10 through Table 12 provide the differences between calculated fast neutron (E > 1.0 MeV) fluence rate results at several locations for each permutation case, each normalized to the corresponding base case result. The overall uncertainty estimates are summarized in Table 13 through Table 15. Note that the results are rounded to the nearest whole percent.
to W3F1-2017-0065 Page 25 of 63 Table 9: Summary of Geometry and Temperature Sensitivity Study18 Case Description umber I Baffle plates oore barre], and neutron pad th:ickuess decreased by 1/ 16 inch Baffle plates oore barre] and neutron pad thickness increased by 1/ 16 *nch 3 Reactor coolant temperatures decrea ed by IO °F 4 Reactor coolant temperatures increased by 0 op
- Reactor vessel radirus d.e creased by l/8 inch 6 Reactor ves el radirus increased by U8 inch 7 Reactor ves el thickness decreased by 1/ 16 inch 8 Reactor essel thickness incr-eased by 1/ 16 inch Surveinance capsule position adj usted by I / 16 inch e -vessel dosimetry 9
position adjusted by 2 inches lO Cartesian versus cylindrical geometry 1nodeling difference in core periphery Table 10: Geometry and Temperature Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Surveillance Capsule Locations 22 Case Surveillance Number Capsule Location 1 1%
2 -1%
3 -4%
4 5%
5 0%
6 0%
7 0%
8 0%
9 2%(a) 10 5%(b)
(a) Surveillance capsule positioning uncertainty includes radial, azimuthal, and axial position variations (b) Core periphery modeling uncertainty determined from direct comparison between cylindrical and Cartesian results in bypass region to W3F1-2017-0065 Page 26 of 63 Table 11: Geometry and Temperature Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Pressure Vessel Locations22 Case RPVInside Number Radius +12 cm Relative to Middle-of-Core Elevation 1 -1%
2 -3%
3 -6%
4 6%
5 4%
6 -4%
7 -1%
8 -1%
9 NIA 10 5%:i;
$Core periphery modeling uncertainty determined from direct comparison between cylindrical and Cartesian results in bypass region Table 12: Geometry and Temperature Permutation-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Reactor Cavity Locations 22 Case Reactor Cavity Reactor Cavity Reactor Cavity Number Top-of-Core Middle-of-Core Bottom-of-Core Elevation Elevation Elevation 1 0% 0% 0%
2 -3% -3% -3%
3 -6% -6% -6%
4 7% 6% 6%
5 1% 1% 1%
6 -4% -4% -4%
7 2% 2% 2%
8 -3% -3% -3%
9 12% 4% 14%
10 5% 5% 5%
to W3F1-2017-0065 Page 27 of 63 Table 13: Summary of Neutron Fluence Rate Uncertainties at Surveillance Capsule Locations Resulting from Geometry and Temperature Uncertainties22 Surveillance Uncertainty Component Capsule Location Internals Dimensions +/-1%
Vessel IR +0%
Vessel Thickness +0%
Dosimetry Position +2%
Coolant Temperature +5%
Core Periphery Modeling +5%
Table 14: Summary of Neutron Fluence Rate Uncertainties at Pressure Vessel Locations Resulting from Geometry and Temperature Uncertainties22 RPVInner Radius +12cm Uncertainty Relative to Component Middle-of-Core Elevation Internals Dimensions +/-1%
Vessel IR +/-4%
Vessel Thickness +0%
Coolant Temperature +6%
Core Periphery +/-5%
Modeling Table 15: Summary of Neutron Fluence Rate Uncertainties at Reactor Cavity Locations Resulting from Geometry and Temperature Uncertainties22 Uncertainty Component Reactor Cavity Reactor Cavity Reactor Cavity Top-of-Core Middle-of-Core Bottom-of-Core Elevation Elevation Elevation Internals Dimensions +/-1% +/-1% +/-1%
Vessel IR +/-3% +/-3% +/-3%
Vessel Thickness +2% +2% +2%
Dosimetry Position +12% +4% +14%
Coolant Temperature +/-6% +/-6% +/-6%
Core Periphery Modeling +/-5% +/-5% +/-5%
to W3F1-2017-0065 Page 28 of 63 Table 16 through Table 18 summarize the analytic uncertainties determined from the reference Westinghouse 4-Loop reactor model with calculations performed with RAPTOR-M3G. The total analytic uncertainty is derived by combining the individual uncertainty components in quadrature using the "rootsum-of-the-squares" method.
This analytic uncertainty analysis meets Regulatory Position 1.4.1 of Regulatory Guide 1.19019 .
Table 16: Summary of Neutron Fluence Rate Uncertainties at Surveillance Capsule Locations22 Uncertainty Component Surveillance Capsule Location Peripheral Assembly Source Strength 4%
Pin Power Distribution 1%
Peripheral Assembly Burnup (+/-5000 1%
MWD/MTU)
Axial Power Distribution 1%
Internals Dimensions 1%
Vessel IR 0%
Vessel Thickness 0%
Dosimetrv Position 2%
Coolant Temperature 5%
Core Periphery Modeling 5%
Total Analytical Uncertainty 8%
to W3F1-2017-0065 Page 29 of 63 Table 17: Summary of Neutron Fluence Rate Uncertainties at Pressure Vessel Inner Radius Locations22 Uncertainty RPVInner Component Radius
+12cm Relative to Middle-of-Core Elevation Peripheral 5%
Assembly Source Strength Pin Power 1%
Distribution Peripheral 1%
Assembly Burnup
(+/-5000 MWD/MTU)
Axial Power 1%
Distribution Internals 1%
Dimensions Vessel IR 4%
Vessel Thickness 0%
Dosimetry NIA Position Coolant 6%
Temperature Core Periphery 5%
Modeling Total Analytical 10%
Uncertainty to W3F1-2017-0065 Page 30 of 63 Table 18: Summary of Neutron Fluence Rate Uncertainties at Reactor Cavity Locations22 Uncertainty Reactor Cavity Reactor Cavity Reactor Cavity Component Top-of-Core Middle-of-Core Bottom-of-Core Elevation Elevation Elevation Peripheral Assembly 4% 5% 5%
Source Strength Pin Power Distribution 1% 1% 1%
Peripheral Assembly 1% 1% 1%
Burnup (+/-5000 MWD/MTU)
Axial Power 3% 1% 2%
Distribution Internals Dimensions 1% 1% 1%
Vessel IR 3% 3% 3%
Vessel Thickness 2% 2% 2%
Dosimetry Position 12% 4% 14%
Coolant Temperature 6% 6% 6%
Core Periphery 5% 5% 5%
Modeling Total Analytical 16% 11% 17%
Uncertainty 2.16.2 Comparisons with Benchmark Measurements and Calculations Pressure Vessel Simulator Measurements Regulatory Position 1.4.2.2 of Regulatory Guide 1.190 can be summarized as follows: Pressure vessel simulator benchmarks are available and should be used for methods qualification, which provide experimental results with well-known and documented uncertainties.
Several simulator benchmark experiments have been performed for the purpose of providing a qualification basis for neutron fluence analysis methods. This section provides the results of comparisons of simulator benchmark measurement results with calculations performed with RAPTOR-M3G. Calculations were performed using both the theta-weighted and directional theta-weighted differencing methods. However, since the WF3 calculations used the DTW method, only the DTW results were used to develop the RAPTOR-M3G uncertainty for WF3.
Pool Critical Assembly Benchmark The Pool Critical Assembly (PCA) Pressure Vessel Facility Benchmark8 is an industry-standard benchmark that can be used to partially qualify a fluence determination methodology according to Regulatory Guide 1.190. The PCA to W3F1-2017-0065 Page 31 of 63 facility provides a small-scale simulation of the configuration of a Pressurized Water Reactor (PWR). The geometry, material compositions, and neutron source for this experiment were all well-characterized, and accurate dosimetry measurements were collected at several locations of interest. A complete description of the benchmark is available in Reference 8. Table 19 shows the distribution of measurement locations.
The RAPTOR-M3G analysis of the PCA problem with the 12/13 configuration was modeled on a 67 x 139 x 102 Cartesian mesh grid. Angular quadrature was modeled with an Sa level-symmetric quadrature set, and anisotropic scattering was treated with a P3 Legendre expansion. The transport cross-section set was constructed from the BUGLE-96 library, and dosimetry reaction rate cross-sections were taken from the SNLRML library9
- The PCA problem was analyzed by RAPTOR-M3G using both the TW and DTW differencing schemes.
Results of the benchmark comparisons using RAPTOR-M3G are presented in Table 20 for DTW differencing. Note that the only the DTW results were used in the net uncertainty calculation.
Table 19: PCA Experimental Measurement Locations18 Position V 1(cm) Location Description A] 1.2.0 Therma] Shield (Fr-ont)
A2 23 .8 Thermal Shield (Back A3 29.7 P11essure essel (Front)
A4 39.5 P11essure Vessel ( 1/4 T)
AS 44.7 P11essure Vessel ( 1/2 T)
A6 50. J P11essure Vessel (3/4 T)
A7 59. J oid Box to W3F1-2017-0065 Page 32 of 63 Table 20: MIC Reaction Rate Comparisons for the PCA 12/13 Blind Test Experiment18 MIC Ratio for Dosimetry Position Noted Reaction I AJ A2 A3 A4 AS A6 A7 0.98 0.99 0.95 0.97 I 0.99 LOO 1.02 1.02 0.99 LO l I l.02 0.98 (n,n') l l.'*rn1n (Cd) 1.02 I l.00 1l~[n l.02 J 0.9o 0.97 1.00 1.06 ic'Rh (n,n') 1s*;mRh (Cd) - 1*.00 *- 0.97 o. 98 0.98 l . 04 l.07 1.08
~" U (n.f) FP (Cd)
...,....,...,. ..J -~-; Ni--0.92 , .,_ .... .,~..l.O*I -~-"<' .,.
1.04 1.06 1.07
- "Np (n.f) FP(Cd) lI li l.0".* ,, LO I J 00 1.03 0.99 Average 1.01 1.00 0.98 0.99 t.02 1.02 J.05 f ........................................................................................................................................................, .....................................,;....................................;......................................... ............................
% std dev 1.9 2.4 4.9 2.1 2.1 3.5 3.9 VENUS-1 Benchmark The VENUS-1 experiment10 is another commonly-used qualification benchmark. As with the PCA benchmark, the critical variables affecting the measurements were carefully measured and recorded . The VENUS-1 benchmark correctly represents the heterogeneities in a PWR, and includes a stainless steel core baffle, core barrel, and neutron pad . The benchmark experiment was performed at room temperature (300 K). Forty-one measurement locations exist in the benchmark, which are given in Table 21.
The RAPTOR-M3G analysis of the VENUS-1 problem was modeled on a 192 x 123 x 65 cylindrical mesh grid . Angular quadrature was modeled with an Sa level-symmetric quadrature set, and anisotropic scattering was treated with a P3 Legendre expansion. The transport cross-section set was constructed from the BUGLE-96 library, and dosimetry reaction rate cross sections were taken from the SNLRML library.
Results of the benchmark comparisons using RAPTOR-M3G are presented in Table 22 for DTW differencing . Note that only the DTW results were used in the net uncertainty calculation .
to W3F1-2017-0065 Page 33 of 63 Table 21: VENUS-1 Measurement Locations 18 Point Description
- 1 Central Water Hole 2-3 hmer Baffle 4- 10 Outer Baffle U - 14 Along 45° in 3/0 fuel region 15- ]9 Along 45° in \\'ater Gap [
20-27 Core Barnel (0-45 °
- 2.&-3 At a radius of 55.255 cm in Water Gap U 3741 eutrou Pad to W3F1-2017-0065 Page 34 of 63 Table 22: M/C Reaction Rate Comparisons for the VENUS-1 Experiment18 M/C Ratio Point 58 Ni (n,p) 1151n (n,n') 103 Rh (n,n') 238 0 (n,t) 231Np (n,t) 1 1.00 0.98 0.95 1.10 -
2 0.98 0.98 0.94 1.04 0.99 3 0.97 0.98 0.93 1.03 0.97 4 0.98 0.98 - 1.05 1.06 5 0.98 0.97 0.95 1.03 -
6 1.00 0.97 0.94 1.05 1.03 7 1.03 0.98 0.95 - 1.04 8 1.03 0.98 0.95 1.04 1.04 9 0.96 0.97 0.93 1.04 1.01 10 0.97 0.97 0.95 1.01 1.01 11 - - - - 0.98 12 - - - - 1.00 13 - - - - 0.99 14 - - - - 1.08 15 - 0.98 0.97 1.04 -
16 - 0.98 0.97 - 0.99 17 - 0.99 1.00 1.07 0.99 18 - 1.01 0.94 - 1.03 19 - 1.00 1.02 1.09 -
20 0.99 1.00 - 1.08 0.99 21 1.01 1.00 - 1.08 1.01 22 1.01 1.00 0.95 1.10 1.09 23 1.05 1.01 - 1.08 1.09 24 1.07 1.00 - 1.12 1.07 25 1.08 0.99 - 1.13 1.01 26 1.09 1.02 - 1.07 1.08 27 1.11 1.01 1.00 1.09 1.08 28 - 1.04 - 1.14 1.08 29 - 1.02 - 1.10 1.11 30 - 1.03 - 1.17 -
31 - 1.03 - 1.09 1.07 32 - 1.05 - 1.12 1.12 33 - 1.02 - 1.11 1.06 34 - - - 1.14 1.12 35 - 1.09 - - -
36 - 1.01 - 1.11 1.14 37 - - - - -
38 - 1.15 - - -
39 - - - - -
40 - - - - -
41 - 1.11 - - -
Results of the PCA and VENUS-1 simulator benchmarks, grouped by reaction, are summarized in Table 23. Also included in the table are the energy response ranges between which 90% of activity is produced in a U-235 fission spectrum, taken from ASTM E844 11 . The U-238 and Np-237 to W3F1-2017-0065 Page 35 of 63 measurements exhibit slightly worse agreement with the calculations; however, the U-238 and Np-237 reactions are subject to higher uncertainties in the measurement process.
The simulator benchmarks test the adequacy of the transport and dosimetry evaluation techniques, and the underlying nuclear data. The simulator benchmark comparison results demonstrate that, when the configuration of the system is well-known, the level of agreement between RAPTOR-M3G calculations and measurements is within the uncertainties associated with the measurements, themselves. The uncertainty assigned to the calculational methodology from simulator benchmarks is 5%.
Table 23: Summary of Simulator Benchmark M/C Reaction Rate Comparisons 18 Neutron Energy Number of Reaction :verage MIC % stddev Response Observations 2 A1 n,o.) 1"Na Cd) 6.45 - JJ.9 MeV 6 0.98 1.8
] n,p 5 Co Cd ) 1.9'8 - 7.5 1 MeV 24 LOl 4.0 usln ' 11 n1 I 15m]n (Cd), I.. l2 - 5.86 MeV 40 LOl 3.9 101 Rh , 11 n1 IO}mR h ,(Cd) 0.73 1. - 5.73 MeV 23 0.98 4.4 m u {n,f) FP Cd , 1.44 - 6.69 MeV 33 ].07 4.5 2l p (n,f) FP (Cd) 0.684 - 5.6 ] MeV 35 ].04 4.5 Total 161 1.02 5.2 H.B. Robinson Benchmark In addition to measurements from laboratory-scale simulator benchmark experiments, Regulatory Guide 1.190 recommends that methods qualification should be based on comparisons with measurement data from pressure vessel cavities. This section provides the comparisons of reactor cavity measurements to calculations performed with RAPTOR-M3G.
H. B. Robinson Unit 2 is a Westinghouse 3-loop PWR. As part of the NRC-sponsored LWR Pressure Vessel Surveillance Dosimetry Improvement Program, a comprehensive set of surveillance capsule and ex-vessel neutron dosimetry measurements were performed during Cycle 912 .
For the Cycle 9 benchmark, a replacement surveillance capsule was installed in a vacant surveillance capsule holder at the 20° azimuth with respect to the nearest cardinal axis. The dosimetry sets were placed at the geometric to W3F1-2017-0065 Page 36 of 63 center of the surveillance capsule such that all measurements were taken within 30 cm of the core mid plane.
The ex-vessel neutron dosimetry was installed in the reactor cavity, between the concrete biological shield and the reactor vessel insulation. Multiple foil sensor sets were placed in capsules that were attached to gradient wires.
The gradient wires were installed in the reactor cavity and axially spanned the length of the reactor core. Only the midplane capsule from the Cycle 9 ex-vessel neutron dosimetry set was analyzed.
H. B. Robinson Unit 2 was modeled in cylindrical geometry with 158x136x172 mesh using RAPTOR-M3G. Angular quadrature was modeled with an Sa level-symmetric quadrature set, and anisotropic scattering was treated with a P3 Legendre expansion. Detailed geometry data for H. B.
Robinson Unit 2 can be found in Reference 12.
Neutron spectra obtained from the transport calculations were combined with the dosimetry cross sections from the SNLRML library9 in order to obtain calculated reaction rates for comparison to the measurements. These comparisons are presented in Table 24 for transport results obtained with the DTW differencing scheme.
The H. B. Robinson Unit 2 benchmark represents an experimental configuration that is broadly reflective of most operating reactors: data was collected during full-power operation at a commercial LWR; the power distribution and power history data supporting the analysis were derived using methods similar to those employed by most operating LWRs; geometric dimensions specified are nominal dimensions, and not necessarily identical to their as-built configuration. These characteristics make the H. B.
Robinson Unit 2 benchmark a compelling data set.
For the H. B. Robinsion Unit 2 benchmark, the maximum average of the in-vessel M/C ratios is 1.08, and the maximum average of the ex-vessel M/C ratios is 1.02. The uncertainty assigned to the calculational methodology from H.B. Robinson Unit 2 benchmark is 7%.
The pressure vessel simulator measurement comparisons described above satisfy Regulatory Position 1.4.2.2 of Regulatory Guide 1.19018 .
to W3F1-2017-0065 Page 37 of 63 Table 24: Results for the H.B. Robinson Unit 2 Cycle 9 Dosimetry Benchmark Experiment18 ln-V ssel Reaction Rate Irrps/atoml M JC Rat!lo, R eactlo111 M,e asured. aku.l ted Cu:-63 (n,a) Co-60 3.86E- I 3.68E- I I.OS Ti-46 (n,p Sc-46 6.9 E-16 5.84E- 16 1.18
!Fe-54 (n p) Mn-54 3.8.1 E- J 5 3.68E- l.5 1.04 Ni.-58 (11,p C-0-58 5.34E- J5 4.99E- 15 J.07
-238 (n,f) FP L74E- J4 1.65E- 14 J.05 Np-237 (n.f) FP 1.19E- 13 U6E- 13 1.03 Yerage 1.!07 Ex-'\i\essel R eact:loo Rate lrrps/atoml J\UC Rat!lo, R ,eactlo111 ;\*I, sured. Cah:u.L ted Cu:-63 (n,a) Co-60 3.86E- 19 4. ]0E- 19 0.94 Ti-46 (n,p Sc-46 6.55E- 18 5.87E- 18. 1.12
!Fe-54 (n p Mn-54 3.55E- 17 3.78E- 17 0.94 Ni.-58 (n,p C-0-58 5.85E- 17 5.71E- 17 1.02
-238 (n,f) FP 2.74E- 16 2.66E- 16 1.03
, *erage 1.:01 Calculational Benchmark Regulatory Position 1.4.2.3 of Regulatory Guide 1.190 can be summarized as follows: The vessel fluence benchmark problems provided by the NRG in NUREG/CR-6115 should be used for methods qualification.
The NRC's fluence benchmark13 is a calculational exercise, developed by Brookhaven National Laboratory at the request of the NRC, which provides reference solutions for typical PWR and Boiling Water Reactor (BWR) pressure vessel fluence calculations. Regulatory Guide 1.190 recommends the use of the calculational benchmark exercise as part of qualifying a fluence calculation method 4 to provide a detailed assessment and verification of the numerical procedures, code implementation, and modeling approximations.
to W3F1-2017-0065 Page 38 of 63 The PWR problem models a typical 204 fuel assembly PWR core including the core baffle and barrel, thermal shield, and a pressure vessel. Three types of fuel loadings are defined, including a standard out-in core loading, a low-leakage core loading, and a core that includes partial length shield assemblies (PLSAs). Reference solutions derived from multigroup synthesis and Monte Carlo simulations are provided in Reference 13.
The PWR problem was analyzed with RAPTOR-M3G. The PWR model analysis presented herein considers the standard core loading configuration.
Angular quadrature was modeled with an S8 level-symmetric quadrature set, and anisotropic scattering was treated with a P3 Legendre expansion.
Comparisons relative to the fast (E > 1.0 MeV) fluence rate synthesis solutions published in Reference 13 are provided in Table 25.
The calculational benchmark does not provide real measurement data, and the methods and data used in the reference results are somewhat dated by contemporary standards. Therefore, results of these evaluations are not used as a direct input to the overall bias and uncertainty assessment for the fluence determination methodology. Nonetheless, the consistency of the RAPTOR-M3G results, both with the reference calculations provided by Brookhaven and the self-consistency demonstrated by the two differencing schemes in RAPTOR-M3G, provides additional confidence that RAPTOR-M3G is correctly applying the discrete ordinates method.
The comparisons of RAPTOR-M3G fluence calculations to the calculational benchmark address Regulatory Position 1.4.2.3 of Regulatory Guide 1.19018 .
to W3F1-2017-0065 Page 39 of 63 Table 25: NRC Fluence Benchmark Comparisons with RAPTOR-M3G PWR Problem with Standard Core Loading at Pressure Vessel Inner-Wall Lower Weld Location (R=219.393 cm, 2=67.1048 cm)
' * : *
- Fast (E > 1.0 McV) Neutron Flucncc Rate RAPTO R-M3G ! RAPTOR-M3G with DT\V I with TI\'
- Theta
""J .D rgr.~~
. 1.125 3.375 2.76E+IO
-4.48 % i -4.96%
5.625 2.87E+IO e--------*-------..~---------------*-----,-------------------------- ------------
2 77[+!0 _________:i_:~~:.:~:~:=.;.1_-------
7.875 3.04£+10 2.9oE+10 2.94E+IO *********** . . -2..78.'~1u ...................,1.....................................
,................!. *o......_. *1***2--*- s* _- -* **-********-----*--3*-
- _* *3* * - l*- - E- * -+****-1c**-1** ............. - .........3. . . _
....i*-s--E*****---:*--**w**--------------- 1*** 3.1 oE+ 1o
! -l.03%i l2 . .n5 J.49F10 3.39£+!0 '" ,.,,....,""" ,....,
J.38E+l0 .. -3.0[%
14.625 3.62E+JO J.5JE +l0 3.52E+ JO -244'}(,
Hi.875 3.55[+ lO J.5t,E+ 10 -2.29'~*i, *.....*...*...... , ...........................................................
-2.07,:, .
............,,.,,125
- 19. _____ ,,,,_
3.45E+
..,.... -----~-~""""_, _____
JO 3.35f>-rn
,, _____ ~. """ ---- -""""'"'"~"*"'-"""'"""-*-*
J.32[ + J0 -2.79"(.
21 RE; 10 3. l 5E* l0 J.lJE+lO -0. 97':- o 2.95E+JO 2.90E+ !O 2.87E+ IO .75}h -'.!.94°.o 25 .875 2.79[+10 2.73[+ !0 2.o7E+IO
-.,.,~~ .... M,.s.s .... ,a,,,,, .... hM.s.s.,.
-2.48%
28 . 125 2.74E+l(} 2.68E+JO 2.64E+!O 30.-1,75 2.77E+JO 2..72[+!0 2.o8E+JO -l.95%
32 ,625 2.77E+ 10 2.7JE+J.O '.!.72E+l0 -2.01 % -l.68%
34.875 2.67E+JO 256E+ l0 2.57E+ l0 -I.Om(. -3.88')-o 37,125 2.3DE+!O 2.35[+l0 '.!.34E+l0 -1.18" ',i 39.375 2.08£'-10 207E"l0 '.!.OoE+ lO -1.10%
41.625 --- * .1.88E+JO ' __..!:.'...;;_OE+ lO ____ l._T_ 1'E_:+_l(_) __+---..--.-1 __ .3_l_'! o________ -5_.8_: *s_*"t_
') - -;
4.1, .875 l.72E+10 I.ME* HI l.62E+ 10 -J.70":,) -5.93%,
- 2. 16.3 Overall Calculational Bias and Uncertainty Regulatory Position 1.4.3 of Regulatory Guide 1.190 can be summarized as follows : The overall f/uence calculational bias and uncertainty must be determined by an appropriate combination of the (1) analytic uncertainty analysis results of Reg. Position 1. 4. 1 and (2) the results of the uncertainty analysis based on the comparisons to the operating reactor and simulator benchmark measurements of Reg. Position 1. 4. 2. Vessel f/uence uncertainty of 20% is acceptable for RTprs and RTNor determination.
The individual uncertainties from the H.B. Robinson, PCA, and VENUS-1 benchmarks combined with the analytic sensitivity study uncertainty forms the basis for the net calculational uncertainty. The simulator benchmark comparison results demonstrate that, when the configuration of the system is well-known, the level of agreement between RAPTOR-M3G calculations and measurements is within the uncertainties associated with the measurements, themselves. Therefore no systematic bias is assigned to the calculational methodology.
to W3F1-2017-0065 Page 40 of 63 The following Table 26 summarizes the uncertainties applicable to pressure vessel beltline (including the surveillance capsules) locations, determined from the results of the methodology qualification process, rounded to the nearest whole percent:
Table 26: Calculational Uncertainties for Reactor Vessel Inner Radius Neutron Fluence Rate 22 RPVInner Radius +12cm Relative to Uncertainty Component Middle-of-Core Elevation Simulator Benchmark +/-5%
Comparisons H.B. Robinson Benchmark +/-7%
Comparisons Analytic Sensitivity Studies +10%
Peripheral Assembly +/-5%
Source Strength Pin Power Distribution +/-1%
Peripheral Assembly +/-1%
Burnup Axial Power Distribution +1%
Internals Dimensions +1%
Vessel IR +/-4%
Vessel Thickness +/-0%
Coolant Temperature +/-6%
Core Periphery Modeling +/-5%
Other Factors +5%
Net Uncertainty +14%
The category designated "Other Factors" is intended to attribute an additional uncertainty to other geometrical or operational variables that individually have an insignificant effect on the overall uncertainty, but collectively should be accounted for in the assessment.
The uncertainty components tabulated above represent percent uncertainty at the 1a level. In the tabulation, the net uncertainty from the analytic sensitivity studies has been broken down into its individual components.
When the four uncertainty values listed above (5%, 7%, 10%, and 5%) are combined in quadrature, the resultant overall 1a calculational uncertainty is to W3F1-2017-0065 Page 41 of 63 estimated to be bounded by 15% for pressure vessel inner radius and surveillance capsules within the core-adjacent beltline region.
This uncertainty quantification addresses Regulatory Position 1.4.3 of Regulatory Guide 1.19018 .
2.16.4 Validation by Comparison to Additional Operating Power Reactors Regulatory Position 1.4.2.1 of Regulatory Guide 1.190 can be summarized as follows: Well-documented f/uence dosimetry measurements for operating power reactors may be used for methods and data qualification.
Comparisons must be performed for the specific reactor being analyzed or for reactors of similar design.
Operating Power Reactor Database In addition to the uncertainty qualification comparisons described above, the radiation transport methodology in RAPTOR-M3G has been extensively compared with data from operating power reactors. These comparisons are intended to provide support for the validation of the transport calculation itself as well as validation for the uncertainties assigned to the results of those calculations.
There are 69 in-vessel surveillance capsules with 295 threshold foil measurements from 18 nuclear power plants that have been analyzed with RAPTOR-M3G. In addition to the in-vessel surveillance capsules, 87 ex-vessel neutron dosimetry (EVND) capsules with 454 threshold foil measurements from locations opposite the core have been analyzed with RAPTOR-M3G.
The average M/C reaction rate ratio over all the fast neutron sensors from each reactor is listed in Table 27. This tabulation provides a direct comparison, on an absolute basis, of measurement and calculation. For surveillance capsules, these comparisons show an average M/C ratio of 1.03 with a standard deviation of 5% at the 1o level. For ex-vessel dosimeters irradiated opposite the core midplane, these comparisons show an average M/C ratio of 0.92 with a standard deviation of 6% at the 1o level.
These results show that the M/C reaction rate ratios for the in-vessel measurements are essentially unbiased and well within the +/- 20%
acceptance criterion given in Regulatory Guide 1.190. The M/C reaction rate ratios for the ex-vessel measurements are within the +/- 30% criterion given in Regulatory Guide 1.190 for the cavity capsules.
The operating power reactor comparisons described in this section address Regulatory Position 1.4.2.1 of Regulatory Guide 1.19018 .
to W3F1-2017-0065 Page 42 of 63 Table 27: In-Vessel and Ex-Vessel Capsules Threshold Reactions M/C Reaction Rate Ratios 18 Plant Number In-Vessel MIC EVND Midplane MIC Domestic Plant # 1 1.05 0.96 Domestic Plant #2 0.99 0.97 International Plant # 1 1.13 1.03 International Plant #2 1.06 1.00 International Plant #3 NIA 0.97 International Plant #4 0.99 0.89 International Plant #5 1.09 0.88 International Plant #6 0.95 0.87 International Plant #7 0.95 0.86 Domestic Plant #3 1.02 0.89 Domestic Plant #4 1.01 0.89 Domestic Plant #5 1.00 0.93 International Plant #8 0.96 0.87 International Plant #9 1.08 0.83 Domestic Plant #6 1.01 0.90 Domestic Plant #7 1.07 NIA Domestic Plant #8 1.11 NIA Domestic Plant #9 1.07 NIA Avera2e 1.03 0.92 Std. Dev.% 5% 6%
Total Number of Capsules 69 87 Total Number of Threshold Foils 295 454 Measurement Data from Waterford 3 The latest analysis of Waterford 3 surveillance capsule dosimetry is described in detail in Reference 17. This analysis was performed with the RAPTOR-M3G code. Anisotropic scattering was treated with a Ps Legendre expansion, and angular discretization was modeled with an S1 6 order of angular quadrature.
The comparison of the calculated results with the available plant-specific dosimetry results was used solely to demonstrate the adequacy of the radiation transport calculations and to confirm the uncertainty estimates associated with the analytical results. The comparison was used only as a check and was not used to bias the final calculated neutron fluence results in anyway.
to W3F1-2017-0065 Page 43 of 63 Results of the evaluations of the dosimetry from the Waterford 3 surveillance capsules withdrawn to date are provided in Table 28. Calculations of individual threshold sensor reaction rates are compared directly with the corresponding measurements. These threshold reaction rate comparisons provide a good evaluation of the accuracy of the fast neutron portion of the calculated energy spectra. For the individual threshold foils, the average M/C comparisons for fast neutron reactions range from 1.07 to 1.17. The overall average M/C ratio for the entire set of Waterford 3 data is 1.11 with an associated standard deviation of 7.0%. These data comparisons show that the measurements and calculations agree within the 20% criterion specified in Regulatory Guide 1.190 Regulatory Position 1.4.318 .
Table 28: Comparison of M/C Sensor Reaction Rate Ratios for Fast Neutron Threshold Reactions from Waterford 317' 18 M/C Capsule 63 Cu(n,a) 4 1>Ti(n,p) 04 Fe(n,p) :,11 Ni(n,p) "'"
0 U(n,f)
+
97° 1.25 - 1.07 1.061.06 -+
263° 0.98 1.15 1.05 1.11 -+
83° 1.05 1.19 1.10 1.17 -+k Average 1.09 1.17 1.07 1.11 -
% Standard 12.8 2.4 2.3 4.9 -
Deviation Average 1.11
% Standard 7.0 Deviation
- As part of the dosimetry analysis process, reaction rates for each foil were normalized to the measures Fe-54(n,p) reaction rate from each capsule. When normalized reaction rates departed by more than 3u from database averages for similar plants, they were discarded.
2.17 Fluence Calculational Uncertainty Core-adjacent Beltline Region For WF3, the RAPTOR-M3G neutron fluence calculations were determined to have a net uncertainty bounded by 15% in the core-adjacent beltline region at the reactor vessel inner radius. The average reaction rate M/C ratio in the inner radius midplane region from the operating reactor database was 1.03 with a standard deviation of 5%. The average reaction rate M/C ratio in the WF3 reactor vessel was 1.11 with a standard deviation of 7%. The M/C ratios fall within the 15% net uncertainty, and all results meet the 20% uncertainty requirement for RT PTs and RT NDT determination in Regulatory Position 1.4.3 of Regulatory Guide 1.19018 .
3.0 Justification for use of generic analytic uncertainty analysis for WF3 application of RAPTOR-M3G The analytic uncertainty analysis discussed in Section 2.16.1 and applied to WF3 is based on the base case of a Westinghouse 4-loop reactor. The to W3F1-2017-0065 Page 44 of 63 following discussion provides a comparison of the configurational and operational differences between WF3 and Catawba Unit 1, a Westinghouse 4-loop designed plant. There is also discussion of the computational parameters used in the Catawba analytic uncertainty analysis compared to the parameters used for the WF3 vessel fluence calculations.
3.1 Plant Configuration Differences between Catawba Unit 1 and Waterford 3 Both Catawba Unit 1 and Waterford 3 contain square fuel assemblies, baffle plates (or core shroud in CE plants), former plates (girth ribs in CE plants),
core barrel, pressure vessel clad, pressure vessel, reflective insulation, and a concrete biological shield. These components were modeled in both References 23 and 17, where Reference 17 dosimetry evaluations supported the Waterford 3 time-limited aging analysis (TLAA) on reactor vessel integ rity24 .
In Catawba Unit 1, neutron pads exist that also have surveillance capsule attachments. In Waterford 3, neutron pads are not present (neither is a thermal shield) and capsules are attached to the inner surface of the pressure vessel. Neutron pads were modeled in Reference 23 and surveillance capsules were modeled in both References 23 and 17. From a source generation standpoint, the analyses for Catawba Unit 123 and Waterford 3 17 were performed the same. The source was generated on a cycle-by-cycle basis using a square array of fuel pins inside a square fuel assembly. Each fuel assembly was assigned a relative power based on the cycle burnup and initial assembly enrichment. Using the cycle burnups and initial assembly enrichments, the recoverable energy per fission and fission fractions were determined for U-235, U-238, Pu-239, Pu-240, Pu-241, and Pu-242. This data was combined with particles per fission for each nuclide, relative energy distribution for particles emitted in fission for each nuclide, assembly relative powers, and power density to obtain an energy-dependent source for each assembly. The product of the energy-dependent source per assembly and relative pin-by-pin power distributions produced the source for each fuel pin, which was then transformed into a cylindrical geometry and input to the RAPTOR-M3G code.
From a radiation anarsis standpoint, the RAPTOR-M3G code was used for the Catawba Unit 12 and Waterford 317 with the same configurations. In each case, the RAPTOR-M3G code was used with the forward transport solution and an r,8,z geometry.
Table 29 gives the geometrical uncertainty parameters in the Catawba Unit 1 uncertainty analysis as well as corresponding data for Waterford 3. For the reactor internals, the tolerances on the Waterford 3 core shroud and core barrel thicknesses are different than those used in the Catawba Unit 3 uncertainty analysis. Additionally, Waterford 3 does not contain a neutron to W3F1-2017-0065 Page 45 of 63 pad; therefore, the reactor internals uncertainty for Catawba Unit 1 contains an additional component compared to Waterford 3. Furthermore, the contribution of the uncertainty from the stainless steel reactor internals components on the total uncertainty at the +12cm from core midplane location is less than 1.5% for Catawba Unit 1 with the RAPTOR-M3G directional theta weighted (dtw) differencing scheme, which is the differencing scheme used in the Waterford 3 analysis 17
- For example, if the uncertainty from the reactor internals thicknesses was doubled to 3%, and the total uncertainty was calculated for the intermediate shell to upper shell weld location, the resulting total uncertainty would be 10.37% rather than 10.02%, which are essentially the same. Therefore, differences in uncertainty from the stainless steel reactor internals components between the Catawba Unit 1 and Waterford 3 analyses is not expected to significantly change the total uncertainty calculated for Catawba Unit 1.
For the pressure vessel inner radius and in-vessel surveillance capsule positioning, tolerances were specified for Catawba Unit 1, whereas measurement data were available for Waterford 3. For Waterford 3, the average value of the measurement data with a +/- 1CJ standard deviation was calculated for both the pressure vessel inner radius and in-vessel surveillance capsule position. The +/- 1CJ standard deviation on the Waterford 3 pressure vessel inner radius and in-vessel surveillance capsule position were compared with the tolerances conservatively chosen for the Catawba Unit 1 uncertainty analysis. The comparison indicated that Catawba Unit 1 and Waterford 3 uncertainty parameter data for the pressure vessel inner radius and in-vessel surveillance capsule positions are the same20 .
The tolerance on the pressure vessel thickness impacts:
(1) Calculations beyond the outside surface of the pressure vessel, which is not relevant to Waterford 3.
(2) The pressure vessel lower shell to lower head circumferential weld maximum fluences that are taken from the pressure vessel outer radius; however, the fluence on this location at 55 effective full-power years (EFPY) (end of license extension) is well below 1.0E+17 n/cm 2 (i.e.,
2.00E+16 n/cm2 24 ) and is not considered in reactor vessel embrittlement evaluations of Waterford 3.
For the modeling of the rectilinear geometry of the core periphery/baffle plates with an r,8,z geometry, an x,y,z model of the Waterford 3 reactor was not available to make a comparison. However, a comparison of r,8,z and x,y,z geometric models of Catawba Unit 1 was performed in the Catawba Unit 1 uncertainty analysis. For Catawba Unit 1, fluence rates were compared at various radii at 1° azimuthal increments. Since the mesh refinement of Catawba Unit 1 and Waterford 3 were performed consistent to W3F1-2017-0065 Page 46 of 63 with NRC Regulatory Guide 1.190, Section 1.3.14, the uncertainty from modeling of the rectilinear core periphery/baffle plates in the Catawba Unit 1 analysis is applicable to Waterford 3. Additionally, the average measurement-to-calculation (M/C) reaction rate ratio from surveillance capsules 97°, 263°, and 83° being 1.11 with a standard deviation of 7% 17 also indicates the adequacy of the modeling of the rectilinear geometry of the core periphery and baffle plates in Waterford 3.
Item numbers 1 and 5 in Table 29 represent the reactor vessel internals parameters that impact the fluence uncertainty calculated at the pressure vessel. The maximum contributions from reactor vessel internals parameters to the total pressure vessel fluence uncertainty are -1.6% and -4.6% for the internals dimensions and core periphery modeling, respectively.
Table 29: Geometrical Uncertainty Evaluation Parameters used in the Catawba Unit 1 Uncertainty Analysis and Relevant Data for Waterford 320 Item No. Geometric Uncertainty Catawba Unit 1 Waterford 3 Parameter 1 Thickness tolerance on stainless +O and -1 /4 for core steel reactor internals components +/-1/16(a) shroud (inches) +9/64 and -0.01 for core support barrel 2 Water annuli thickness between the core barrel and the reactor pressure vessel determined by the +/-1/a<a) +/-1/8Cbl tolerance of the inner radius of the reactor pressure vessel (inches) 3 In-vessel surveillance capsule +/-0.061(b)
+/-1/16 (0.0625)<8 )
positioning tolerance (inches) 4 Reactor pressure vessel thickness +/-1/16(a) Minimum vessel tolerance (inches) thickness was used 5 Modeling of the rectilinear core Fluence rate Mesh refinement periphery/ baffle plates in r,fJ,z compared (r,fJ,z vs.
similar to Catawba geometry: comparison between x,y,z) at 1° azimuthal models x,y,z and r,fJ,z models intervals (a) Tolerances were conservatively chosen for uncertainty analysis in Catawba Unit 1 (b) Corresponds to 1u standard deviation of the measurement data 3.2 Operational Differences between Catawba Unit 1 and Waterford 3 If it is assumed that low-leakage loading patterns (LLLP) are utilized when core loading patterns do not contain highly-enriched fresh fuel assemblies on the core periphery, both Catawba Unit 1 and Waterford 3 transitioned to LLLP early in plant life: at 8.17 and 8.30 EFPY at Catawba Unit 1 and Waterford 3, respectively. Both plants operate at the same system pressure.
to W3F1-2017-0065 Page 47 of 63 Catawba Unit 1 and Waterford 3 have different water temperatures. The plants also have slightly different peripheral assembly average burnup ranges and cycle-average peak axial relative powers. However, these operational differences are inputs to the RAPTOR-M3G code and do not affect validity of comparison between the two reactors, as discussed below.
Table 30 provides the mean values of the parameters used in the uncertainty analysis of Catawba Unit 1 as well as corresponding data for Waterford 3.
The system pressure in the uncertainty analysis is the same as the data in Waterford 3. The core average moderator temperature and downcomer temperature in the uncertainty analysis are higher and the same, respectively, than the corresponding data in Waterford 3 Cycles 1-5, and higher than the corresponding data in Waterford 3 Cycles 6-19. The peripheral assembly average burnup range in the uncertainty analysis is different compared to the data in Waterford 3. The fluence change at the lower and upper burnup range limit (3,000 MWD/MTU and 50,000 MWD/MTU) relative to a reference case were used in determining the uncertainty due to the peripheral fuel assembly burnup change in the Catawba Unit 1 analysis. The burnup range of 3,000 MWD/MTU to 50,000 MWD/MTU does not encompass the Waterford 3 peripheral assembly burnup range. However, since the differences are small (3,000 - 2,548 = 452 MWD/MTU for the lower and 51,356 - 50,000 = 1,356 MWD/MTU for the upper limit) the impact from the different burnup range of Waterford 3 does not cause a significant change on the results of the Catawba Unit 1 MUR uncertainty analysis. To demonstrate this, the uncertainty analysis was re-executed using the burnup range specific to Waterford 3, and the uncertainty associated with peripheral assembly average burnup changed less than 0.1 %20 . Therefore, this difference can be deemed negligible. The cycle-average axial peak relative power in the uncertainty analysis is slightly different compared to the data in Waterford 3 for Cycle 1. However, this parameter does not directly impact the outcome of the uncertainty analysis.
Rather, it is the gradient shape change of the average axial relative power that impacts the uncertainty analysis.
to W3F1-2017-0065 Page 48 of 63 Table 30: Summary of Operational Parameters used in the Catawba Unit 1 Uncertainty Analysis and in Waterford 320 Catawba Unit I Parameter \ Yaterford Unit 3 Uncertainty Anah'sis
- ~******** ............................ ,11(********* ~.......................... .
System pressure (psia) 2250 2250 Core avernge moderator 583 (Cycles 1-5) temperature ( .,F) 585 574-575 (Cycles 6-19)
-Core inlet (dow-n_co_n-1e_r_)_"""_,_ __
553 553 (Cycles l-5) I
. .. ___ temp-0raturcJ.,F:1... . _____ . ...... 5.~3 ~?1?.J~Y~!-~~ ~'-~ I_?) .. j Peripheral assembly average bumup range 3.000 lo 50,000 2.5-tS ro 51 .356 (MWD!MTC)
Cycle-average axial peak I.~ (Cycle l )
LI rdative pmvcr l . l (Cycles ~ -19)
Variations in the operational parameters given in Table 30, taken from the Catawba Unit 1 uncertainty analysis was assumed to be applicable to Catawba Unit 1 and is also assumed to be applicable to Waterford 3. The variations presented in Table 31 were determined to be conservative estimates for Pressurized Water Reactors. Note that the density change occurring from the +/- 10°F temperature change in the Catawba Unit 1 uncertainty analysis bounds the density change occurring from the +/- 10°F temperature change applied to Waterford 3 temperatures, as demonstrated in Table 32 and Table 33. The temperature distribution shape used in the Catawba Unit 1 model applies to Waterford 3 as well. Most crucially, water enters the bottom of the core at the inlet temperature and exits the top of the core at the outlet temperature. Bypass and downcomer regions are also similar between the two plants.
Table 31 : Operational Uncertainty Evaluation Parameters for Catawba Unit 120 Operational Uncertainty Parameter Data Reactor coolant system water
+/- 10 tempcrnturc variation eF)_
Peripheral assembly source magnitude
+/-5 change {%i) I Peripheral fuel assembly relative pin
+/- 10
- power cha.:1~~ (': o) .
Peripheral foci assembly burnup
+/- 5000 change (1'v1\\'Di!vfTU) 1
_ Axial powc~distribution chan~('~o) .... .*l. . . . . . . . . . . . . =
.= * =:.. ._.. . l.__.O. . . . ._.. . .._:::_:::=_=_=:
to W3F1-2017-0065 Page 49 of 63 Table 32: Core Average Moderator Temperature and Density Variations used in the Catawba Unit 1 Uncertainty Analysis and Relevant Data for Waterford 320 Waterford Unit 3 Cata*wba Unit 1 Parameter C . -cles l-5 Cycles 6-19 Base Variation Base Variation Base Variation Temperature 585 595 575 5 3 593 5 3 574 584 564
(<>F)
Deusity 0.7 1 11 0.69 3 0.7232 0 .7136 0.7009 0.7256 0 .7244 0.7124 0.7358 (g/oc)
Density Cha11ge with
-0.0 29 0.0 121 -0.012 0.01 19 -0.0 120 0.0113 R,espect to Base (gfoc)
Table 33: Core Inlet Temperature and Density Variations used in the Catawba Unit 1 Uncertainty Analysis and Relevant Data for Waterford 320 Wate rford Unit 3 Catawba Unit 1 Parameter C,,cles 1-5 Cycles 6-19 Base Va riation Base Variation Base Varia tion Temperature 553 563 543 553 563 543 543 553 533
(<>F)
Density 0.74 6 0.7369 0.757 0.7476 0.7369 0.75 0. 57 0.74 6 0.7676 (g/c,c)
Density Change with
-0.0 10 O.OI02 -0.0 10 0.0 102 -0.0 102 0.0098 Respect o Base (g/oc) 3.3 Key Computational Parameters Table 34 provides a comparison of the radiation transport code and cross-section library used in the Catawba Unit 1 measurement uncertainty recapture (MUR) power uprate 23 and Waterford 3 surveillance Capsule 83° dosimetry and fluence analysis 17 , where results from Reference 17 were used to support the Waterford 3 TLAA on reactor vessel integrity24 . Both Catawba Unit 1 and Waterford 3 analyses used the same radiation transport code and cross section library.
Table 34: Radiation Transport Co de and Cross-Section Library Used in the Catawba Un it 1 Measu rement Uncertainty Recapture Power Uprate and Waterford 3 Surveillance Capsule 83° Dosimetry Evaluation20 Plant Description Catawba. Unit 1 \Vaterford
- nit 3 (Ref. 6) (Ref. 9)
Radiation trn:11sp011 code RAPTOR-MJG RAPTOR-M3G Cross-section library BUGLE-96 (Ref. I 0) BUGILE-96 to W3F1-2017-0065 Page 50 of 63 Table 35 compares the RAPTOR-M3G problem control parameters used in the Catawba Unit 1 and Waterford 3 analyses. The differences are summarized as follows:
- 1. Angular Quadrature
- 2. Domain Decomposition -Spatial
- 3. Order of Legendre expansion used in calculation
- 4. Flag to control printing cross-sections to the problem output
- 5. Number of materials in the cross-section file
- 6. Maximum number of outer iterations Among the listed six items above, the important RAPTOR-M3G problem control parameters differing between the Catawba Unit 1 and Waterford 3 analyses are angular quadrature and order of Legendre expansion used in the calculation. The choice to print cross sections to the problem output file has no impact on the results and the number of materials in the cross-section file is a problem-specific entry. The difference in the maximum number of outer iterations has no impact on the calculations as these iterations are used for cases where upscattering cross sections are used in the thermal energy range. Both Catawba Unit 1 and Waterford 3 analyses do not contain upscattering cross sections. The difference in spatial domain decomposition number indicates that more processors were used in the problem space domain in the z-direction for Waterford 3 compared to Catawba Unit 1. Differences in domain decomposition schemes may lead to different convergence behaviors but has negligible impact on the results.
Regarding the angular quadrature, Section 1.3.1 from the NRC Regulatory Guide 1.1904 states:
An S8 fully symmetric angular quadrature must be used as a minimum for determining the fluence at the vessel.
Regarding the order of Legendre expansion, Regulatory Position 1.1.2 from the NRC Regulatory Guide 1.190 states:
Cross-Section Ang ular Repr esentation. In discrete ordinates transport calculations, a P3 angular decomposition of the scattering cross-sections (at a minimum) must be employed.
The analysis for Waterford 3 uses an angular quadrature and order of Legendre expansion higher than those used in Catawba Unit 1 and higher than the minimum values required in Regulatory Guide 1.190. Therefore, the Waterford 3 analysis provides a more detailed analysis with respect to the to W3F1-2017-0065 Page 51 of 63 angular quadrature and order of Legendre expansion. Furthermore, a comparison of using lower order of quadrature (Sa and S12) in the Catawba Unit 1 MUR analysis22 for three circumferential welds (below, adjacent, and above the fuel stack, opposite to the core on the pressure vessel) showed that relative differences between the fast neutron fluence rates (E > 1.0 MeV) between Sa and S12 quadrature set results are less than 1%.
In addition, a sensitivity case was run in which the Catawba Unit 1 base case was modeled using the Ps, S16 decomposition and quadrature, rather than the P3 , Sa used in the analytic uncertainty analysis. The differences in fluence rate at the circumferential welds opposite the core were less than 0.4% 20 . Therefore, the conclusions of the uncertainty analysis for Catawba Unit 1 using P3, Sa would not be impacted by a change to Ps, S16, and the applicability of the Catawba Unit 1 uncertainty analysis to Waterford 3 is not impacted by the difference in angular quadrature and Legendre expansion.
to W3F1-2017-0065 Page 52 of 63 Table 35: RAPTOR-M3G Problem Control Parameters used in the Catawba Unit 1 Measurement Uncertainty Recapture Power Uprate and Waterford 3 Surveillance Capsule 83° Dosimetry and Fluence Evaluation 20 (Continued on next page)
RA.PTOR-M3G Plant Description VariabJe Name Catawba Unit 1 Waterford Unit 3 GEOM Geometry Type r,61,z r,8',z SOLUTlON Solution Type Forward transpolit Fornrani transpolit ISN Angular Quadrature Ss S16 Domain Decomposition -
PROCS_A_S (I) 4 4 A:mrular Domain Decomposition -
PROCS_A_S (2)
Spatial 16 rn
[GM Energy Groups 67 67 The h *ghest 1(fastest) energy GRPS_ RSTRT ( I) g1-10up number over which l I the c:aJcufati.ou is to proceed The lm..'est (slowest) energy GRPS- RSTRT 2) gmup number ove,r which 67 67 the c::akuJation is to J.)moeed Flag to co11tm] the v citing of flux moment data to GRPS- RSTRT 3) 0 0 "fl.LJx_moments.hS' (11estart flag)
]HT ]HF IHM I) Position of ,cr1 in the cmss-3 3 section file Position of vor in the cross-
]HT IHF [HM 2) 2 2 section file Table length in cross-
]HT IHF [HM 3) 70 70 section file Highest group included in PS_GRPS (1) 0 0 outer iterations Lowest gmup included in I PS_GRPS (2) 0 0 outer iterations to W3F1-2017-0065 Page 53 of 63 R*l\1JG Plant Description -~, ,,., -*,, V ,.,.,,
~amc Catawba Unit l \\iaterford Unit 3 Order of Legendre IPNXS_IPN (l) expansion ofthe cross- P, ! p, section set Order of Legendre I
. . ~~:.:~=-I-~~-~~--------**::r::;~~1~.'.'~*-**--JI. ******----~:- .... ... *1 IPNXS_IPN (3)
Flag to control printing cross-sections to the j
J-----***~-----~
0 1
. roh1em out )Ut
!':umber of materials in the l 49 l cross-section fik 1 i Directional theta-""'Tr5Jrectional theta-:-1 I IDS Differencing scheme wci.-,hted dtw) ,veighted (dtw) j
!-****** ""'"'" "'""'"""'"""""l'""""'""'"'" *' """""'""'"""""""""'""'""""'"""'"""""""""'""1 I Theta weight (ignored if THETA\:V ms,, dt,v) 0 *9 0 *9 BNDCON . -i boundary condilion reflective reflective BNDCON 12) +i boundary condition vacuum BNDCON(J) reflective CON (4 )
BNDCONt5.l _ -k boundary condition vacuum vacuum BNDCON l vacuum MXIN 500 r--- ------ t Maximum number of outer !
100
~ X O U I... iterations 1 l EPSFLX Inner iteration convergence criteria LOE-03 lOE-03 I EPSSCT Outer iteration convergence criteria I.OE-03 lOE-03 Sll'v1PLE SRC Simple source constmction 0 0 Generali zed coarse mesh GCl\:IR ALPHA 0 0 rebalance acceleration Table 36 compares the RAPTOR-M3G input/ output parameters used in the Catawba Unit 1 and Waterford 3 analyses. Table 36 shows that a boundary source input file format was specified for Waterford 3, and no boundary source input file format was specified for Catawba Unit 1. Since a boundary source file was not actually used in Waterford 3, there is no difference due to this variable. In addition , the Catawba Unit 1 analysis requested that all FIDO input data are written to a file called "fido_echo" whereas this was not requested in Waterford 3. This difference has no impact on the results of the analysis as it is an option to write data in an output file.
to W3F1-2017-0065 Page 54 of 63 Table 36: RAPTOR-M3G Input/ Output Parameters used in the Catawba Unit 1 Measurement Uncertaintr Recapture Power Uprate and Waterford 3 Surveillance Capsule 83° Dosimetry and Fluence Evaluation 2 RAP-TOR-MJG Plant Description Variiable Na01e* Cata\\'ba Unit l Waterford Unit .J EXTERNAL GEOM External geomelry input fi1e None None Fix.ed distributed ou11ce FIX- DIST- :SRC tort small tort smaU input file fonnat Boundary source input file*
BOUND SRC one varbnd format Scalar flux output file SCL FLUX- OUT varsd
- arsc.
format Directional flux output fi]e DLR FLUX- OUT . one one format Write a)) FIDO input data to FIDO ECHO Yes 0 a fi]e called ' fido echo '
Maximum number of 0 {a)) energy 0 ,(aU energy energy gmups to maintain grnupswill be gmu,ps win be 1AX- MEM- GRPS m memory maintained in maintained in throughout the calculation memory), memory)
Table 37 compares the RAPTOR-M3G geometry parameters used in the Catawba Unit 1 and Waterford 3 analyses. Table 37 shows that different number of i-mesh, j-mesh, k-mesh, zone bodies, and zones were used. Both Catawba Unit 1 and Waterford 3 did not request any key flux positions to report. A different key flux data format was specified in the two analyses, but since key flux reporting was not requested this is not considered as a difference. Even if key flux reporting was requested, the existence of this data has no impact on results. In order to evaluate the geometrical meshes in more detail, Table 38 was generated.
Table 37 RAPTOR-M3G Geometry Parameters used in the Catawba Unit 1 Measurement Uncertainty Recapture Power Uprate and Waterford 3 Surveillance Capsule 83° Dosimetry and Fluence Evaluation20 RAPTOR-1\1.JG Plaut Description Variable N a01e Catawba Unit l Waterford Unit J umber of i-mesh in IM 209 160 pr-oblem umber of j -mesh in JM 195 121 problem Number of k-mesh in KM .179 247 problem ZBOmES Number of zone bodies 3069 1.603 lZM Number of zones 48 l6 Number of key flux NKEYFX 0 0 positions to report KEYFX FORMAT Key flux data format nonconv calar to W3F1-2017-0065 Page 55 of 63 Table 38: RAPTOR-M3G Geometry Data used in the Catawba Unit 1 Measurement Uncertainty Recapture Power Uprate and Waterford 3 Surveillance Capsule 83° Dosimetry and Fluence Evaluation 20 Plant Description -
Catawba Unit 1 1aterford Unit 3 Radi'.:! Dimensions (cr-n) 0 to 4 l l .48 00
. . . . . . . . . . . Azimuthal Span ("1} ......_. J . . . . . . . . . . . .':o'._~t~o._~4*.: :5 ........................!. __. . . . .. .~~.l!.. '.:!:::'i ............. l
-363.296 to -360.000 to Axial Extent {cm) 343 .4 60 340.000 Table 38 compares geometrical data used in the Catawba Unit 1 and Waterford 3 analyses. Both Catawba Unit 1 and Waterford 3 modeled octants in r,S,z geometry. In both analyses, the models extended radially from the centerline of the reactor core out to a location interior to the primary biological shield.
NRC Regulatory Guide 1.190 Section 1.3.1 states that:
An azimuthal (S) mesh using at least 40 intervals over an octant in (r,S) geometry in the horizontal plane should provide an accurate representation of the spatial distribution of the material compositions and source described in Regulatory Position 1.2. The radial mesh in the core region should be -2 intervals per inch for peripheral assemblies and may be much more coarse for assemblies more than approximately two assembly pitches removed from the core-reflector interface. In excore regions, a spatial mesh that ensures the flux in any group changes by less than a factor of -2 between adjacent intervals should be applied, and a radial mesh of at least -3 intervals per inch in water and -1 .5 intervals per inch in steel should be used .
Because of the relatively weak axial variation of the fluence, a coarse axial mesh of -0.5 interval per inch may be used except near material and source interfaces, where flux gradients can be large.
The Waterford 3 and Catawba Unit 1 models contained mesh intervals consistent with NRC Regulatory Guide 1.190 Section 1.3.120 . In the Waterford 3 reactor model, the peripheral assembly region, bypass region, and core barrel were modeled more than 3 intervals per inch. The downcomer region was modeled -3 intervals per inch or more than 3 intervals per inch. The pressure vessel was modeled more than 1.5 intervals per inch. Finer meshes were used near material interfaces. Axial meshes were more than 0.5 intervals per inch, with the exception of regions at 282.38 cm and above from the core midplane that contained 0.4 intervals per inch . The axial distance of 282.38 cm is above the bottom of the vessel support pad of the outlet nozzle, where the bottom of the vessel support pad of the outlet nozzle elevation was conservatively used instead of the nozzle to upper shell weld elevation in reporting the fast neutron fluence. The most to W3F1-2017-0065 Page 56 of 63 important commonality between the Waterford 3 and Catawba Unit 1 RAPTOR-M3G transport models is that both models have achieved geometrical convergence, i.e., the calculated fast neutron fluence results at the regions of interest do not change significantly with finermesh structure.
The axial extent was modeled approximately between -360 cm to 340 cm in both the Catawba Unit 1 and Waterford 3 analyses. Azimuthally, octants of the reactor were modeled (from 0° to 45°) in both analyses. The number of azimuthal meshes was greater than 40 intervals per octant in both analyses, in accordance with NRC Regulatory Guide 1.190 Section 1.3.1.
Note that the Catawba Unit 1 and Waterford 3 reactor models have different number of meshes; however, it is the mesh size and meeting the relevant guidance in the NRC Regulatory Guide 1.190 that is important and not the number of meshes used.
3.4 Conclusion This discussion has reviewed the plant configuration and operational parameters between Catawba Unit 1 and Waterford 3, as well as the computational and uncertainty evaluation parameters in calculating the neutron fluences for the Catawba Unit 1 MUR power uprate reactor vessel integrity evaluations and Waterford 3 surveillance Capsule 83° dosimetry evaluation. The evaluation indicates that Catawba Unit 1 is a 4-loop plant and Waterford 3 is a 2-loop plant, but both plants have the same configuration (square fuel assemblies in the core and rectilinear core shroud/baffle plates surrounded by cylindrical components of the core barrel and pressure vessel), with the exception of dimensional changes, presence of neutron pads in Catawba Unit 1, and different capsule attachment locations. Both plants have different operational parameters (temperatures, peripheral assembly burnup ranges, and axial peak relative powers). These differences do not invalidate the application of RAPTOR-M3G, which was shown to be acceptable for Catawba Unit 1 to Waterford 3. An evaluation of the uncertainty evaluation parameters also indicate that the fluence uncertainties developed for the Catawba Unit 1 analysis using RAPTOR-M3G are applicable to Waterford 3. Computational parameters between the Catawba Unit 1 and Waterford 3 analyses are consistent, where Waterford 3 used a more detailed angular quadrature set and Legendre expansion 20 .
The methodology used in the RAPTOR-M3G code has been qualified and benchmarked for Light Water Reactors (LWR). The detailed RAPTOR-M3G analyses performed for the Catawba Unit 1 MUR are applicable to Waterford 3 given that computational and input parameters are consistent with the Catawba Unit 1 MUR analysis.
to W3F1-2017-0065 Page 57 of 63 4.0 Impact on UFSAR Accident Analyses The RAPTOR-M3G method is not used to perform any accident analyses. The results of neutron fluence calculations are used to develop heatup and cooldown limit curves and project USE and RTPTs values for comparison with screening criteria in future years of operation. Based on a review of ECS10-001, Rev. 5, none of these fluence calculation outputs or material properties affect the safety analyses 33 .
5.0 Regulatory Precedent The RAPTOR-M3G code was approved for neutron fluence calculations at Catawba Nuclear Station Units 1 and 2 as part of their measurement uncertainty recapture power uprates in April 201641 . Justifications for approval which were related to the method-specific (rather than plant-specific) Regulatory Positions of Regulatory Guide 1.190 (i.e. cross-section library, simulator benchmarks, operating reactor benchmarking, etc.) are similar to those made in this document for Waterford 3.
Regulatory Evaluation 1.0 Factors Allowing This Change The change in reactor vessel fluence calculation methodology from DORT to RAPTOR-M3G is allowed because the RAPTOR-M3G has been shown to comply with the requirements of Regulatory Guide 1.190 in the Technical Discussion above. The RAPTOR-M3G has been benchmarked against simulator and operational reactor dosimetry, and when combined with analytic sensitivity studies, the uncertainty in calculated fluences are bounded by 20%. Since Regulatory Guide 1.190 provides a basis for accepting neutron fluence calculation methods, compliance with the Guide shows the acceptability of RAPTOR-M3G at WF3.
The fluence calculational methodology outputs are not used as inputs to accident analyses, and therefore have no impact on them.
2.0 No Significant Hazards Consideration Determination Entergy Operations, Inc. (Entergy) is proposing that the Waterford 3 Steam Electric Station, RAPTOR-M3G be the neutron fluence calculation method documented in the Final Safety Evaluation Report (FSAR) section 4.3. This license application demonstrates that RAPTOR-M3G meets the criteria presented in RG 1.190 (Calculational and Dosimetry Methods for Determining Pressure Vessel Neutron Fluence), dated March 2001.
The proposed change does not change any plant system, structure or component, or change any plant operating parameters.
Entergy has evaluated whether or not a significant hazards consideration is involved with the proposed amendment(s) by focusing on the three standards set forth in 10 CFR 50.92, "Issuance of Amendment," as discussed below:
to W3F1-2017-0065 Page 58 of 63
- 1. Does the proposed change involve a significant increase in the probability or consequences of an accident previously evaluated?
Response: No.
The probability of occurrence of an accident previously evaluated for Waterford 3 is not altered by the proposed license amendment. The accidents currently analyzed in the Waterford 3 Final Safety Analysis Report (FSAR) remain the same. The proposed change does not impact the integrity of the reactor coolant pressure boundary (RCPB) (i.e., there is no change to the operating pressure, materials, loadings, etc.). The proposed change does not affect the probability nor consequences of any design basis accident (DBA). The proposed neutron fluence calculational methodology meets the criteria in RG 1.190 and will be used to ensure that the PIT limit curves, maximum heatup and cooldown rates, and LTOP enable temperature remain acceptable to maintain reactor pressure vessel integrity.
Fracture toughness test data are obtained from material specimens contained in capsules that are periodically withdrawn from the reactor vessel. These data, combined with the neutron fluence calculations, permit determination of the conditions under which the vessel can be operated with adequate safety margins against brittle fracture throughout its service life. For each analyzed transient and steady state condition, the allowable pressure is determined as a function of reactor coolant temperature considering postulated flaws in the reactor vessel beltline, inlet nozzle, outlet nozzle, and closure head.
The predicted radiation induced i1RTNDT is calculated using the respective reactor vessel beltline materials' copper and nickel contents and the neutron fluence determination. The RT NDT and, in turn, the operating limits for Waterford 3 are adjusted, if necessary, to account for the effects of irradiation on the fracture toughness of the reactor vessel materials and maintain reactor vessel integrity within design assumptions.
Therefore, the proposed change does not involve a significant increase in the probability or consequences of an accident previously evaluated.
- 2. Does the proposed change create the possibility of a new or different kind of accident from any accident previously evaluated?
Response: No.
The proposed change to the neutron fluence calculational method will not create a new accident scenario. The requirements to have PIT limits and LTOP protection are part of the licensing basis for Waterford 3. The neutron fluence calculation method will validate, and when necessary, provide input to the development of new operating limits. The data analysis for the vessel surveillance specimens are used to confirm that the vessel materials are responding as predicted based on previous neutron fluence projections.
Therefore, the proposed change does not create the possibility of a new or to W3F1-2017-0065 Page 59 of 63 different kind of accident from any previously evaluated.
- 3. Does the proposed change involve a significant reduction in a margin of safety?
Response: No.
The proposed change to the neutron fluence calculational method conforms to the criteria presented in RG 1.190 and will ensure that Waterford 3 continues to operate within the operating margins allowed by 10 CFR 50.60 and the ASME Code.
Therefore, the proposed change does not involve a significant reduction in a margin of safety.
Based on the above, Entergy concludes that the proposed amendment(s) present no significant hazards consideration under the standards set forth in 10 CFR 50.92(c), and, accordingly, a finding of "no significant hazards consideration" is justified.
Environmental Consideration The proposed amendment does not involve (i) a significant hazards consideration, (ii) a significant change in the types or significant increase in the amounts of any effluent that may be released offsite, or (iii) a significant increase in individual or cumulative occupational radiation exposure. Accordingly, the proposed amendment meets the eligibility criterion for categorical exclusion set forth in 10 CFR 51.22(c)(9). Therefore, pursuant to 10 CFR 1.22(b), no environmental impact statement or environmental assessment need be prepared in connection with the proposed amendment.
to W3F1-2017-0065 Page 60 of 63 References
- 1. Gulf General Atomic Report GA-8747, "TWOTRAN, a FORTRAN Program for Two Dimensional Transport," July 1968.
- 2. WANL-PR-(LL)-034, "Nuclear Rocket Shielding Methods, Modification, Updating and Input Data Preparation. Vol. 5 - Two Dimensional Discrete Ordinates Transport Technique," August 1970.
- 3. RSICC Computer Code Collection CCC-650, "DOORS 3.2a, One, Two-, and Three-Dimensional Discrete Ordinates Neutron/Photon Transport Code System,"
Radiation Safety Information Computational Center, Oak Ridge National Laboratory (ORNL), May 2007.
- 4. Regulatory Guide 1.190, "Calculational and Dosimetry Methods for Determining Pressure Vessel Neutron Fluence," U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research, March 2001.
- 5. RSICC Data Library Collection DLC-185, "BUGLE-96, Couple 47 Neutron, 20 Gamma-ray Group Cross Section Library Derived from ENDF/8-VI for LWR Shielding and Pressure Vessel Dosimetry Applications," Radiation Safety Information Computational Center, Oak Ridge National Laboratory (ORNL), July 1999.
- 6. RSICC Data Library Collection DLC-246, "VITAMIN-87/BUGLE-87, Broad-Group and Fine-Group Coupled Neutron /Gamma Cross-Section Libraries Derived from ENDF/8-VII.O Nuclear Data," Radiation Safety Information Computational Center, Oak Ridge National Laboratory (ORNL), October 2011.
- 7. Westinghouse Report WCAP 17993-NP, Rev. 0-8, "Justification for the Use of RAPTOR-M3G for Catawba Unit 1 Measurement Uncertainty Recapture (MUR)
Power Uprate Fluence Evaluations," April 2015. (Available as ADAMS Accession Number ML15117A012.)
- 8. ORNL Report ORNL/TM-13205, "Pool Critical Assembly Pressure Vessel Facility Benchmark," (NUREG/CR-6454), July 1997.
- 9. RSICC Data Library Collection DLC-178, "SNLRML Recommended Dosimetry Cross Section Compendium," Radiation Shielding Information Computational Center, Oak Ridge National Laboratory, July 1994.
- 10. Nuclear Energy Agency (Organization for Economic Co-Operation and Development), "Prediction of Neutron Embrittlement in the Reactor Pressure Vessel: VENUS-1 and VENUS-3 Benchmarks," 2000.
11.ASTM Designation E844, 2009 (2014), "Standard Guide for Sensor Set Design and Irradiation for Reactor Surveillance," ASTM International, West Conshohocken, PA, 2014, DOI: 10.1520/E844-09R14E01, www.astm.org to W3F1-2017-0065 Page 61 of 63 12.0RNL Report ORNL/TM-13204, "H.B. Robinson-2 Pressure Vessel Benchmark,"
(NUREG/CR-6453), February 1998.
- 13. BNL Report BNL-NUREG-52395, "PWR and BWR Pressure Vessel Fluence Calculation Benchmark Problems and Solutions," (NUREG/CR-6115),
September, 2001.
- 14. Westinghouse Report WCAP-13348, Rev. 0, "Consumers Power Company Palisades Nuclear Plant Reactor Vessel Fluence Analysis," May 1992.
- 15. Westinghouse Report WCAP-13362, Rev. 0, "Westinghouse Fast Neutron Exposure Methodology for Pressure Vessel Fluence Determination and Dosimetry Evaluation," May 1992.
- 16. R.E. Maerker, "Application of LEPRICON Methodology to LWR Pressure Vessel Surveillance Dosimmetry," Reactor Dosimetry, Proc. 5th ASTM-Euratom Symposium, Jackson Hole, WY, May 31-June 5, 1987, American Society of Testing and Materials (1989).
- 17. Westinghouse Report WCAP-17969-NP, Rev. 2, "Analysis of Capsule 83° from the Entergy Operations, Inc. Waterford Unit 3 Reactor Vessel Surveillance Program," November, 2017.
18.Westinghouse Letter LTR-REA-16-117, Rev. 4, "Response to the NRC Request for Additional Information Regarding RAPTOR-M3G on the Waterford Unit 3 License Renewal Application." November 2017.
- 19. Westinghouse Letter LTR-REA-17-56, Rev. 2, "Documentation Demonstrating Adherence to Regulatory Guide 1.190 and Response to the NRC Request for Additional Information Regarding Neutron Fluence Determination in the Extended Beltline Region on the Waterford Unit 3 License Renewal Application." November 2017.
- 20. Westinghouse Letter LTR-REA-17-75, Rev. 2, "Evaluation per NEI 96-07 Section 4.3.8.2 for Changing from Use of DORT code to the RAPTOR-M3G Code for Waterford Unit 3." November 2017.
- 21. Regulatory Guide 1.99, Rev. 2, "Radiation Embrittlement of Reactor Vessel Materials," U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research. May 1988.
- 22. Westinghouse Report WCAP-18060-NP, "Response to RAls Concerning the Use of RAPTOR-M3G for the Catawba Unit 1 Measurement Uncertainty Recapture (MUR) Power Uprate Fluence Evaluations," November 2015.
- 23. Westinghouse Report WCAP-17669-NP, Rev. 1, "Catawba Unit 1 Measurement Uncertainty Recapture (MUR) Power Uprate: Reactor Vessel Integrity and Neutron Fluence Evaluations," October 2015.
to W3F1-2017-0065 Page 62 of 63
- 24. Westinghouse Report WCAP-18002-NP, Rev. 0, "Waterford Unit 3 Time-Limited Aging Analysis on Reactor Vessel Integrity," July 2015.
- 25. Waterford 3 Updated Final Safety Analysis Report, Rev. 309. Section 5.3, "Reactor Vessel."
26.ASTM Designation E1018, 2013, "Standard Guide for Application of ASTM Evaluated Cross Section Data File, Matrix E706 (118)," ASTM International, West Conshohocken, PA, 2013, DOI: 10.1520/E1018-09R13, www.astm.org.
27.ASTM Designation E261, 2015, "Standard Practice for Determining Neutron Fluence, Fluence Rate, and Spectra by Radioactivation Techniques," ASTM International, West Conshohocken, PA, 2015, DOI: 10.1520/E0261-15, www.astm.org.
28.Schmittroth, A, FERRET Data Analysis Core, HEDL-TME 79-40, Hanford Engineering Development Laboratory, Richland, WA, September 1979.
29.ASTM Standard E944-13, Standard Guide for Application of Neutron Spectrum Adjustment Methods in Reactor Surveillance, E 706 (IIA), 2013.
- 30. WCAP-16002-NP, "Analysis of Capsule 263° from the Entergy Operations Waterford Unit 3 Reactor Vessel Radiation Surveillance Program." March 2003.
31.WCAP-16088-NP, Rev. 2, "Waterford Unit 3 Reactor Vessel Heatup and Cooldown Limit Curves for Normal Operation." June 2012.
- 32. Letter from Mr. N Kalyanam to Mr. Joseph E. Venable, "Waterford 3 Steam Electric Station, Unit 3-lssuance of Amendment Re: Pressure Temperature Limit Curves to 32 Effective Full Power Years with Power Uprate." 6-16-2004.
ADAMS Accession Number ML041700466.
33.ECS10-001, Rev. 5, "Waterford 3 Cycle 22 Reload Analysis Groundrules." June 2016.
- 34. NUREG 1801, "Generic Aging Lessons Learned (GALL) Report", Rev. 2
- 36. CEP-FTP-W3, "Program Section for Reactor Vessel Fracture Toughness and Surveillance Material Testing at Waterford 3." July 2007.
- 37. U.S. 10CFR50, Appendix H, "Reactor Vessel Material Surveillance Program Requirements" 38.ASTM E185-82, "Conducting Surveillance Tests for Light-Water Cooled Nuclear Power Reactor Vessels."
- 39. U.S. 10CFR50, Appendix G, "Fracture Toughness Requirements" to W3F1-2017-0065 Page 63 of 63 40.SER for WCAP-14040, Rev. 3, "Methodology Used to Develop Cold Overpressure Mitigating System Setpoints and RCS Heatup and Cooldown Limit Curves." ADAMS Accession No. ML050120209.
- 41. Letter from Mr. Jeffrey Whited to Mr. Kelvin Henderson, "Catawba Nuclear Station Units 1 and 2-lssuance of Amendments Regarding Measurement Uncertainty Recapture Power Uprate (CAC Nos. MF4526 and MF4527)." April 29, 2016. ADAMS Accession Number ML 16081A33
Attachment 3 to W3F1-2017-0065 10 CFR Part 54 License Renewal Supplemental Discussion (31 Pages)
Note: Atta chment 3 is not part of the Li cense Amendment Request to incorporate RAPTOR-M3G code into Waterford's curre nt licensing basis, but is included to provide relevant information pertinent to t he revie w of the Ii cense renew al application submitted March 23, 2016.
to W3F1-2017-0065 Page 1 of 31 10 CFR Part 54 License Renewal Supplemental Discussion 1.0 Regulatory Positions to Address for Period of Extended Operation Section C.1. of Regulatory Guide 1.190 states that the Regulatory Guide "describes the application and quantification of a methodology acceptable to the NRC staff for determining the best-estimate neutron fluence experienced by materials in the beltline region of light water reactor (LWR) pressure vessels, as well as for determining the overall uncertainty associated with those best-estimate values." The Regulatory Positions of Reg. Guide 1.190 will be reviewed during the Staff review of the License Amendment Request to incorporate the RAPTOR-M3G method in the Waterford 3 current license basis (CLB). However, for the Period of Extended Operation, the beltline region is considered to be extended to all reactor vessel components which experience >10 17 n/cm 2 fast neutron fluence. For calculations related to those components, most Regulatory Positions of Reg.
Guide 1.190 are still applicable because the calculation method is unchanged by increasing the calculational domain. However, several Regulatory Positions require demonstration that they are met for the new extended beltline calculations.
They are listed and briefly discussed below.
1.1 Analytic Uncertainty Analysis (Reg. Position 1.4.1)
The analytic uncertainty analysis presented in the Waterford 3 License Amendment Request (LAR) only accounts for locations within the traditional vessel beltline region. It is expected that the sensitivity of the calculated fluence to variations in calculation inputs will be different in the extended beltline region; therefore, the sensitivity study was extended to include the limiting material (with respect to projected Adjusted Reference Temperature) of the extended beltline region in Section 2.0.
1.2 Operating Reactor Measurements (Reg. Position 1.4.2.1)
The benchmark dosimetry measurements in the Waterford 3 license amendment request only include data from the core midplane region of operating power reactors, including Waterford 3. In Section 3.1, the operating power reactor database locations were expanded to include ex-vessel dosimetry data at off-midplane locations.
1.3 Estimate of Fluence Calculational Bias and Uncertainty (Reg. Position 1.4.3)
Since analytic uncertainty analyses and operating reactor measurements are inputs to the overall calculational bias and uncertainty determination, it is necessary to consider a new calculational bias and uncertainty for the extended beltline region for the Period of Extended Operation. This is discussed in Sections 2.0 and 3.0.
2.0 RAPTOR-M3G Analytic Uncertainty Analysis Including Extended Beltline Region Operating reactors are subject to several uncertainties that may influence the to W3F1-2017-0065 Page 2 of 31 validity of the calculated neutron fluence results. The most significant among these are:
- Uncertainties in the core neutron source
- Uncertainties in the as-built thicknesses and locations of the reactor vessel and internal components
- Uncertainties in the full-power coolant temperatures (water density)
This listing of parameters is consistent with the findings of other neutron fluence uncertainty studies4*5*6 . This section presents the results of a sensitivity study performed using RAPTOR-M3G that evaluate the impacts of variations in the parameters listed above on calculated neutron fluence values. The sensitivity study is not specific to the WF3 reactor geometry or operation; however, it has been shown that the uncertainty analysis inputs (i.e., dimensional and operational variances) are similar when applied to WF37 . Although there are no cavity surveillance capsules at WF3, the results of the generic sensitivity study for the associated reactor cavity locations are included for completeness.
Note that the uncertainty analysis was performed for both the theta-weighted (TW) and directional-theta-weighted (DTW) differencing schemes. In general, the analytic uncertainty values are consistent between the two differencing schemes.
Since the WF3 analysis used the DTW differencing scheme, the results obtained using only DTW are incorporated into the net uncertainty calculation.
2.1 Analytic Uncertainty Analysis - Core Neutron Source Uncertainties To assess the impact of uncertainties in the core neutron source on calculated neutron fluence results, changes in the following parameters were evaluated 2 :
- Absolute source st rength of peripheral fuel assemblies - Studies have shown that the neutron fluence rate in regions external to the core is dominated by the neutron source from fuel assemblies on the core periphery. In-core measurements indicate that a source magnitude uncertainty of 5% is bounding 2 .
- Pin-by-pin spatial distributions of n eutron so urce at the core periphery - Core management studies indicate that uncertainties in the relative pin powers in peripheral fuel assemblies can be on the order of 10%2 .
- Burnup of the peripheral fuel assemblies - Perturbations in fuel assembly burnup impact the fission spectrum, neutron yield per fission, and energy released per fission for each peripheral fuel assembly. A 5000 MWD/MTU uncertainty in the peripheral fuel assembly burnups is considered conservative. The sensitivity study is performed using a to W3F1-2017-0065 Page 3 of 31 series of calculations starting with midcycle burnup at 3000 MWD/MTU, and 5000 MWD/MTU to 50,000 MWD/MTU with 5000 MWD/MTU delta mid-cycle burnup between each run 2 .
- Axial power distribution - Based on variations in axial peaking factors over the course of a fuel cycle, a 10% uncertain~ in the shape of the axial power distribution is considered conservative .
Each case evaluated as part of the sensitivity study is described in Table 1.
The base case consisted of a low-leakage power distribution cycle from a Westinghouse 4-Loop reactor. Table 2 through Table 4 provide the differences between calculated fast neutron (E > 1.0 MeV) fluence rate results at several locations for each permutation case, each normalized to the corresponding base case result. The overall uncertainty estimates are summarized in Table 5 through Table 7. Note that all results are rounded to the nearest whole percent.
Table 1: Summary of Core Neutron Source Sensitivity Study2 Case Description Number 1 Peripheral source strength biased by a factor of 0.95 2 Peripheral source strength biased by a factor of 1.05 3 Pin power distribution gradient diminished according to:
P"";"" ., = [(P - 1.0) X 0.91 + 1.0 4 Pin power distribution gradient intensified according to:
Pplus = [(P - 1.0) X 1.1] + 1.0 5"11- Mid-cycle burnup at 3,000 MWD/MTU 611' Mid-cycle burnup at 50,000 MWD/MTU 7 Axial power distribution intensified according to :
Axialvzus = [(Axial - 1.0) X 1.1] + 1.0 8 Axial power distribution intensified according to :
Axial""';.,,,,., = r(Axial - 1.0) X 0.91 + 1.0
- Cases 5 and 6 span a mid-cycle burnup range of 47000 MWD/MTU. The uncertainty in the neutron fluence attributable to a 5000 MWD/MTU uncertainty in burnup is obtained by scaling the difference between Cases 5 and 6 accordingly by F = (5000 / 47000). For example, Case 5 (mid-cycle burnup of 3000 MWD/MTU) results in a 7%
fluence rate decrease relative to the base case. Case 6 (mid-cycle burnup of 50000 MWD/MTU) results in a 1%
fluence rate increase relative to the base case. Therefore the fluence rate difference associated with a 47000 MWD/MTU difference is 8%. Scaled by 5000 / 47000, the fluence rate difference associated with a 5000 MWD/MTU difference is 1%.
to W3F1-2017-0065 Page 4 of 31 Table 2: Source Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Surveillance Capsule Locations 2 Case Surveillance Number Capsule Location 1 -4%
2 4%
3 1%
4 -1%
5 -7%
6 1%
7 1%
8 -1%
Table 3: Source Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Pressure Vessel Locations2 Case RPVInside RPVInside Number Radius +52 cm Radius +12 cm Relative to Top- Relative to of-Core Middle-of-Core Elevation* Elevation*
1 -3% -5%
2 3% 5%
3 0% 1%
4 0% -1%
5 -7% -7%
6 4% 2%
7 -10% 1%
8 10% -1%
- The selected locations on the inner radius of the reactor pressure vessel are typical of circumferential welds that join base metal forgings.
to W3F1-2017-0065 Page 5 of 31 Table 4: Source Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Reactor Cavity Locations2 Case Reactor Cavity Reactor Cavity Reactor Cavity Number Top-of-Core Middle-of-Core Bottom-of-Core Elevation Elevation Elevation 1 -4% -5% -5%
2 4% 5% 5%
3 1% 1% 1%
4 -1% -1% -1%
5 -7% -7% -7%
6 2% 2% 2%
7 -3% 1% -2%
8 3% -1% 2%
Table 5: Summary of Neutron Fluence Rate Uncertainties at Surveillance Capsule Locations Resulting from Core Neutron Source Uncertainties2 Uncertainty Component Surveillance Capsule Location Peripheral Assembly Source +/-4%
Strength Pin Power Distribution +/-1%
Peripheral Assembly Burnup +/-1%
(+5000 MWD/MTU)
Axial Power Distribution +1%
to W3F1-2017-0065 Page 6 of 31 Table 6: Summary of Neutron Fluence Rate Uncertainties at Pressure Vessel Locations Resulting from Core Neutron Source Uncertainties2 Uncertainty RPVInner RPVInner Component Radius +52cm Radius +12cm Relative to Relative to Top-of-Core Middle-of-Core Elevation Elevation Peripheral +/-3% +/-5%
Assembly Source Strength Pin Power +/-0% +/-1%
Distribution Peripheral +/-1% +/-1%
Assembly Burnup
(+/-5000 MWD/MTU)
Axial Power +/-10% +/-1%
Distribution Table 7: Summary of Neutron Fluence Rate Uncertainties at Reactor Cavity Locations Resulting from Core Neutron Source Uncertainties2 Uncertainty Reactor Cavity Top- Reactor Cavity Reactor Cavity Component of-Core Elevation Middle-of-Core Bottom-of-Core Elevation Elevation Peripheral Assembly +/-4% +/-5% +/-5%
Source Strength Pin Power +/-1% +/-1% +/-1%
Distribution Peripheral Assembly +/-1% +/-1% +/-1%
Burnup (+/-5000 MWD/MTU)
Axial Power +/-3% +/-1% +/-2%
Distribution 2.2 Geometry and Temperature Uncertainties To assess the impact of uncertainties in the location and thickness of reactor components, as well as uncertainties in reactor coolant temperature, on calculated neutron fluence results, changes in the following parameters were evaluated 2 :
to W3F1-2017-0065 Page 7 of 31
- Reactor internals dimensions - Thickness tolerances on stainless steel reactor internals components (e.g., core baffle, core barrel, thermal shield/neutron pad) are typically specified as 1/16 inch or tighter.
- Reactor vessel inner radius - Reactor vessels typically specify an inner radius with tolerance bounds of -0.00 inches and +1/32 inches. A tolerance of+/- 1/8 inch is considered.
- Reactor vessel thickness - Some techniques for fabricating reactor vessels result in larger-than-nominal reactor vessel base metal plate thicknesses. A tolerance of+/- 1/16 inch is considered.
- Dosimetry Positioning - Surveillance capsules have a tolerance of +/-
1/16 inch associated with the positioning of the dosimetry in radial, azimuthal, and axial directions. A larger positioning uncertainty of +/- 2 inches is associated with ex-vessel neutron dosimetry in radial azimuthal, and axial directions.
- Coolant Temperature - Variations in water temperature over the course of a fuel cycle are expected to be less than+/- 10 °F.
- Core Peripheral Modeling - The modeling of the rectilinear core baffle in cylindrical geometry represents another potential source of uncertainty in the geometric modeling of the reactor. The sensitivity of the solution to the modeling approach is determined by a direct comparison of the results of a cylindrical geometry calculation with those of a Cartesian geometry calculation in which the baffle region and core periphery were modeled explicitly. The comparisons of interest were taken at various locations external to the core baffle, but inside the core barrel.
Each case evaluated as part of the sensitivity study is described in Table 8.
The base case consisted of a low-leakage power distribution from a Westinghouse 4-Loop reactor. Table 9 through Table 11 provide the differences between calculated fast neutron (E > 1.0 MeV) fluence rate results at several locations for each permutation case, each normalized to the corresponding base case result. The overall uncertainty estimates are summarized in Table 12 through Table 14. Note that all results are rounded to the nearest whole percent.
to W3F1-2017-0065 Page 8 of 31 Table 8: Summary of Geometry and Temperature Sensitivity Study2 I Case :
I Numbt*r i Description j I l_;-1-::::;.;::.:~::::::::~::~::::~~;J,':.i::- -:
t r
- 3 t -- -- -
Reactor coolant temperatures decreased by IO ,;F
- 1 I
4 Reactor coolant temperatures increase<l hy lO °F 5 I Reactor vessel radius decreased by 1/8 inch
-******-************************************************************************-****************************************************************************--1 6 ! Reactor vessel radius increased by 1/8 inch l 7 .J Reactor vessel thickness decreased hv l.! 16 inch
[--------8----------!--R~;;;~;:-~:~~~~-i;h*i~k,~~;~--,~~c-;;;~d-b; 1/ 16 inch I,
9 1 S~t~~*eillaiic_c ~a,rsu~~
1 plhltton ad.1usk:d b., ~
r)s~tii1:
mi.:ht:!>
ad_justt*d by I/ I() inch, ex-vessel dosimetry lO i Cartesian versus cylmdncal geometry modclmg difforence in core periphery Table 9: Geometry and Temperature Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Surveillance Capsule Locations2 Case Surveillance Number Capsule Location I 1%
2 -1%
3 -4%
4 5%
5 0%
6 0%
7 0%
8 0%
9 2%(a) 10 5%(b)
(a) Surveillance capsule positioning uncertainty includes radial, azimuthal, and axial position variations (b) Core periphery modeling uncertainty determined from direct comparison between cylindrical and Cartesian results in bypass region to W3F1-2017-0065 Page 9 of 31 Table 1O: Geometry and Temperature Permutation-to-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Pressure Vessel Locations2 Case RPVInside RPVInside Number Radius +52 cm Radius + 12 cm Relative to Top- Relative to of-Core Elevation Middle-of-Core Elevation 1 0% -1%
2 -3% -3%
3 -9% -6%
4 10% 6%
5 3% 4%
6 -2% -4%
7 1% -1%
8 0% -1%
$Core periphery modeling uncertainty determined from direct comparison between cylindrical and Cartesian results in bypass region Table 11: Geometry and Temperature Permutation-Nominal Fast Neutron (E > 1.0 MeV) Fluence Rate Difference at Reactor Cavity Locations 2 Case Reactor Cavity Reactor Cavity Reactor Cavity Number Top-of-Core Middle-of-Core Bottom-of-Core Elevation Elevation Elevation 1 0% 0% 0%
2 -3% -3% -3%
3 -6% -6% -6%
4 7% 6% 6%
5 1% 1% 1%
6 -4% -4% -4%
7 2% 2% 2%
8 -3% -3% -3%
9 12% 4% 14%
10 5% 5% 5%
to W3F1-2017-0065 Page 10 of 31 Table 12: Summary of Neutron Fluence Rate Uncertainties at Surveillance Capsule Locations Resulting from Geometry and Temperature Uncertainties2 Surveillance Uncertainty Component Capsule Location Internals Dimensions +/-1%
Vessel IR +0%
Vessel Thickness +0%
Dosimetry Position +2%
Coolant Temperature +5%
Core Periphery Modeling +5%
Table 13: Summary of Neutron Fluence Rate Uncertainties at Pressure Vessel Locations Resulting from Geometry and Temperature Uncertainties2 RPVInner RPVInner Radius +52cm Radius +12cm Uncertainty Relative to Relative to Component Top-of-Core Middle-of-Elevation Core Elevation Internals Dimensions +/-2% +/-1%
Vessel IR +/-3% +/-4%
Vessel Thickness +0% +0%
Coolant Temperature +10% +6%
Core Periphery +/-5% +/-5%
Modeling Table 14: Summary of Neutron Fluence Rate Uncertainties at Reactor Cavity Locations Resulting from Geometry and Temperature Uncertainties2 Uncertainty Component Reactor Cavity Reactor Cavity Reactor Cavity Top-of-Core Middle-of-Core Bottom-of-Core Elevation Elevation Elevation Internals Dimensions +/-1% +/-1% +/-1%
Vessel IR +/-3% +/-3% +/-3%
Vessel Thickness +2% +2% +2%
Dosimetry Position +12% +4% +14%
Coolant Temperature +/-6% +/-6% +/-6%
Core Periphery Modeling +/-5% +/-5% +/-5%
to W3F1-2017-0065 Page 11 of 31 Table 15 through Table 17 summarize the analytic uncertainties determined from the reference Westinghouse 4-Loop reactor model with calculations performed with RAPTOR-M3G. The total analytic uncertainty is derived by combining the individual uncertainty components in quadrature using the "rootsum-of-the-squares" method.
This analytic uncertainty analysis meets Regulatory Position 1.4.1 of Regulatory Guide 1.1908 .
Table 15: Summary of Neutron Fluence Rate Uncertainties at Surveillance Capsule Locations2 Uncertainty Component Surveillance Capsule Location Peripheral Assembly Source Strength 4%
Pin Power Distribution 1%
Peripheral Assembly Burnup (+/-5000 1%
MWD/MTU)
Axial Power Distribution 1%
Internals Dimensions 1%
Vessel IR 0%
Vessel Thickness 0%
Dosimetrv Position 2%
Coolant Temperature 5%
Core Periphery Modeling 5%
Total Analytical Uncertainty 8%
to W3F1-2017-0065 Page 12 of 31 Table 16: Summary of Neutron Fluence Rate Uncertainties at Pressure Vessel Inner Radius Locations2 Uncertainty RPVInner RPVInner Component Radius Radius
+52cm +12cm Relative to Relative to Top-of- Middle-of-Core Core Elevation Elevation Peripheral 3% 5%
Assembly Source Strength Pin Power 0% 1%
Distribution Peripheral 1% 1%
Assembly Burnup
(+/-5000 MWD/MTU)
Axial Power 10% 1%
Distribution Internals 2% 1%
Dimensions Vessel IR 3% 4%
Vessel Thickness 0% 0%
Dosimetry NIA NIA Position Coolant 10% 6%
Temperature Core Periphery 5% 5%
Modeling Total Analytical 15% 10,0 Uncertainty to W3F1-2017-0065 Page 13 of 31 Table 17: Summary of Neutron Fluence Rate Uncertainties at Reactor Cavity Locations 2 Uncertainty Reactor Cavity Reactor Cavity Reactor Cavity Component Top-of-Core Middle-of-Core Bottom-of-Core Elevation Elevation Elevation Peripheral Assembly 4% 5% 5%
Source Strength Pin Power Distribution 1% 1% 1%
Peripheral Assembly 1% 1% 1%
Burnup (+/-5000 MWD/MTU)
Axial Power 3% 1% 2%
Distribution Internals Dimensions 1% 1% 1%
Vessel IR 3% 3% 3%
Vessel Thickness 2% 2% 2%
Dosimetry Position 12% 4% 14%
Coolant Temperature 6% 6% 6%
Core Periphery 5% 5% 5%
Modeling Total Analytical 16% 11% 17%
Uncertainty 2.3 Justification for Use of generic uncertainty analysis for WF3 application of RAPTOR-M3G Section 2.0 of the Waterford 3 License Amendment Request for changing neutron fluence calculation methods from DORT to RAPTOR-M3G provided justification for use of the generic Westinghouse analytic uncertainty analysis for Waterford 3 11 . The same justification is applicable to the period of extended operation, with some amendments to account for the additional materials and extended time period to be considered.
2.3.1 Reactor Vessel Internals Dimensions Section 3.1, Plant Configuration Differences, of the Waterford 3 License Amendment Request for changing neutron fluence calculation methods, describes a difference in the reactor vessel internals thickness tolerances between Catawba Unit 1 (a Westinghouse 4-loop design) and Waterford 3, where the tolerance for Catawba Unit 1 does not bound Waterford 3. The acceptability of this difference is justified by doubling the analytic uncertainty contribution due to reactor vessel internals and showing that the total analytic uncertainty is essentially the same. However, the analytic uncertainty was determined at a point in the traditional reactor beltline, which is not bounding for the period of extended operation.
to W3F1-2017-0065 Page 14 of 31 The same justification is acceptable if the analytic uncertainty at the Waterford 3 upper circumferential weld is examined. Due to the difference in tolerances of the reactor vessel internals, the uncertainty contribution would be bounded by 3.4% (compared to 1.7% for Catawba Unit 1). This would result in a total analytic uncertainty of 15.53% compared to 15.25%, which is essentially the same. Therefore, differences in the uncertainty from the stainless steel reactor internals components between the Catawba Unit 1 and Waterford 3 analyses is not expected to significantly change the total uncertainty calculated for Catawba Unit 1 using the generic uncertainty analysis.
2.3.2 Angular Quadrature and Cross Section Representation Section 3.3, Key Computational Parameters, of the Waterford 3 License Amendment Request for changing neutron fluence calculation methods, discusses the difference in order of angular quadrature and Legendre expansion of the scattering cross section used by the Catawba Unit 1 uncertainty analysis and the Waterford 3 fluence calculations. The study consisting of the Catawba Unit 1 base case modeled using the Ps, S15 decomposition and quadrature, rather than P3, Sa, determined that at the upper circumferential weld, the differences in fluence rate were less than 0.4% between the Ps, S1a, and P3, Sa cases. Therefore, for the period of extended operation, the applicability of the Catawba Unit 1 uncertainty analysis to Waterford 3 is not impacted by the difference in angular quadrature and Legendre expansion.
In conclusion, the use of the generic Westinghouse 4-loop uncertainty analysis for Waterford 3 calculations remains valid for the period of extended operation in the extended beltline region.
3.0 Overall Calculational Bias and Uncertainty The individual uncertainties from the H.B. Robinson, PCA, and VENUS-1 benchmarks, which do not change for the extended beltline, combined with the analytic sensitivity study uncertainty, form the basis for the net calculational uncertainty. The simulator benchmark comparison results demonstrate that, when the configuration of the system is well-known, the level of agreement between RAPTOR-M3G calculations and measurements is within the uncertainties associated with the measurements, themselves. Therefore, no systematic bias is assigned to the calculational methodology2 .
Table 18 summarizes the uncertainties applicable to pressure vessel beltline (including the surveillance capsules) and extended beltline locations, determined from the results of the methodology qualification process:
to W3F1-2017-0065 Page 15 of 31 Table 18: Calculational Uncertainties for Reactor Vessel Inner Radius Neutron Fluence Rate 2 RPVInner RPVInner Radius Radius +12cm
+52cm Relative to Uncertainty Component Relative to Middle-of-Top-of-Core Core Elevation Elevation Simulator Benchmark +/-5% +/-5%
Comparisons H.B. Robinson Benchmark +/-7% +/-7%
Comparisons Analytic Sensitivity Studies +/-15% +/-10%
Peripheral Assembly +/-3% +/-5%
Source Strength Pin Power Distribution +/-0% +/-1%
Peripheral Assembly +/-1% +/-1%
Burnup Axial Power Distribution +10% +1%
Internals Dimensions +/-2% +/-1%
Vessel IR +/-3% +/-4%
Vessel Thickness +/-0% +/-0%
Coolant Temperature +10% +6%
Core Periphery Modeling +5% +5%
Other Factors 5% +/-5%
Net Uncertainty +/-18% +/-14%
The category designated "Other Factors" is intended to attribute an additional uncertainty to geometrical or operational variables that individually have an insignificant effect on the overall uncertainty, but collectively should be accounted for in the assessment.
The uncertainty components tabulated above represent percent uncertainty at the 1cr level. In the tabulation, the net uncertainty from the analytic sensitivity studies has been broken down into its individual components. When the four uncertainty values listed above (5%, 7%, 10%, and 5%) are combined in quadrature, the resultant overall 1cr calculational uncertainty is estimated to be bounded by 15% for pressure vessel inner radius and surveillance capsules within the core-adjacent beltline region, and 18% at the upper-to-middle plate circumferential weld.
This uncertainty quantification addresses Regulatory Position 1.4.3 of Regulatory Guide 1.1902 .
Operating Power Reactor Database Measurement Data from Waterford 3 Core-adjacent Beltline Region Extended Beltline Region to W3F1-2017-0065 Page 19 of 31 sensitivity study. Although WF3 does not have ex-vessel dosimetry, the average M/C of 0.93 with 7% standard deviation for off-midplane EVND capsules falls within the 30% 1cr calculational uncertainty for ex-vessel locations2 . This gives credibility to the validity of the 18% calculated uncertainty for the vessel inner radius.
4.0 Axial Fluence Distribution at Waterford 3 The reactor pressure vessel fluence analyses described in the Waterford 3 Capsule 83 analysis9 and TLAA10 have been performed for the pressure vessel inner radius (PVIR), the radial location at which the maximum exposure occurs at axial locations opposite the reactor core. However, due to scattering up the reactor cavity, there is an axial location at which the exposure at the pressure vessel outer radius (PVOR) exceeds that of the PVIR. To quantify this effect and to ensure the analyses at the PVIR remain limiting, axial fluence profiles at the PVIR and PVOR were extracted from the Waterford 3 fluence models. The results are shown in Figure 1 for projected fluence at 32 EFPY and in Figure 2 for projected fluence at 60 EFPY.
Based on these figures, the PVIR exposure is limiting for the upper circumferential weld and remains limiting for the lowest extent of the nozzle weld and well into the nozzle shell. Figure 2 also shows that at 60 EFPY, 5 EFPY after the projected end of the period of extended operations 10, the PVOR at the upper circumferential weld will not receive the threshold 1017 n/cm 2 fluence for inclusion in the reactor vessel extended beltline. Also at 60 EFPY, the PVIR at the nozzle weld will not receive the threshold 1017 fluence for inclusion in the reactor vessel extended beltline.
The low fluence at the nozzles, particularly at the PVOR, relative to other plant configurations is attributed to the meter step change in reactor cavity gap at approximately 3 feet above the core midplane, as shown in Figure 3. This decreases the amount of neutron streaming in the reactor cavity compared to a cavity which maintains a small {<1ft) cavity gap throughout the cavity.
to W3F1-2017-0065 Page 20 of 31 t.OO E+19 1.00E+l8 I;"
E l
i:ii - - PVIR C
~ 1.00[+17 - - PVOR e Top of Ac.tlve rl!lel
.,fj Lipper Circ Weld l...
,::; Nozzle we [Low5t Cxte nl]
l.OOE+l6 1.00E+1S.
0 50 100 150 200 2.50 300 3SO Dirta:ne* 'l rom ,cor. Mldpl -* [cml Figure 1: Waterford 3 Reactor Pressure Vessel Projected Fast Neutron (E > 1.0 MeV) Fluence at 32 EFPY2 to W3F1-2017-0065 Page 21 of 31 U)OE+19
.OOE+18
- - P\IIR
- - PVOll TOI) of A tili Fu I lJl)tJer Circ Weld Nozzle Weld {Lowest Exte nt]
UN)E+16 UN)E+1!;
0 so 100 1SO 200 2SO 300 350 Dls.t:aooe from Core MJdplane (om!
Figure 2: Waterford 3 Reactor Pressure Vessel Projected Fast Neutron (E > 1.0 MeV) Fluence at 60 EFPY2 to W3F1-2017-0065 Page 22 of 31 a...
1-Za:,
~-
8u n-.a.-...?:
-- 1~----,~ ~
..._ ,~
II' ,.,
q'
,. 1 I ...-.
1
. ."'~ & ..
~I
~
l
., __ J L---
Figure 3: Reactor cavity gap increase shown in general arrangement drawing 16 to W3F1-2017-0065 Page 23 of 31 5.0 Considerations from NRC Public Meeting on October 19, 2017 5.1 Bias and Uncertainty Estimates 5.1.1 Use of RAPTOR-M3G for Reactor Vessel Internals (RVI) Fluence Calculations Waterford 3 currently uses MRP-227A to manage RVI aging effects 12 ,
and the Waterford 3 LRA states that it will also be used in the period of extended operation 13 . Therefore, Waterford 3 does not intend to calculate the fluence on RVls for the period of extended operation.
There is no request to approve the use of RAPTOR-M3G for RVI fluence calculations.
5.1.2 Use of Generic Westinghouse 4-loop Model for Analytic Uncertainty Analysis of Combustion Engineering Plant In its LAR for the approval of RAPTOR-M3G as the fluence calculation method of record 11, Waterford 3 provided justification for the use of a generic Westinghouse 4-loop plant model for analytic uncertainty analysis to qualify the RAPTOR-M3G method. The variations in model inputs used in the generic analysis were shown to be either bounding or essentially the same when compared to the Waterford 3 plant geometry and operating conditions 11 . Amendments to the discussion were included in section 2.3 of this document, and the conclusions remain the same. Therefore, the generic Westinghouse 4-loop uncertainty analysis remains valid when applied to Waterford 3 in the period of extended operation.
5.1.3 Overall Method Uncertainty Provided for Extended Beltline Document LTR-REA-16-117, Rev. 2, submitted as part of the response to RAI 4.2.1-1 a, did not include overall method uncertainty for RAPTOR-M3G in the extended beltline region. Both the analytical uncertainty and overall method uncertainty have been provided in sections 2.0 and 3.0 of this document, respectively, for the limiting material in the extended beltline region, weld 106-121 (upper-to-middle shell circumferential weld).
5.2 Sensitivity Calculations in the Extended Beltline At Waterford 3, the vessel nozzles and nozzle welds are not projected to experience fluence >10 17 n/cm 2 at 60 EFPY of operation based on Figure 2.
Additionally, the inside vessel radius experiences higher fluence than the outside radius until an axial location corresponding to the nozzle shell 2 . The vessel outer radius experiences at a fluence of approximately 5 x 1016 n/cm 2 at the upper circumferential weld. Therefore, the nozzle shells and welds are not part of the extended beltline region for the Waterford 3 reactor vessel.
The NRC staff expressed specific interest in how the Waterford 3 fluence to W3F1-2017-0065 Page 24 of 31 calculation uncertainty analysis incorporated the effects from variation in the following inputs, which have been shown to have a significant effect on the nozzle weld and nozzle fluence values 1 .
5.2.1 Axial distribution and isotopic content of the U-235 and Pu-239 in the fuel when calculating the fission source The analysis considers spatial and spectral variations in the neutron source, derived from detailed assembly burnup distributions from individual fuel cycles. The source is spatially shaped by radial pin gradients for fuel assemblies located on the core periphery and an axial power distribution representative of mid-cycle operating conditions.
The energy distribution of the source is determined by selecting a fuel burnup representative of conditions averaged over the irradiation period under consideration and an initial fuel assembly enrichment characteristic of the core designs used. From the assembly burnup and initial U-235 enrichment, a fission split by isotope including U-235, U-238, Pu-239, Pu-240, Pu-241, and Pu-242 is derived; and, from that fission split, composite values of energy release per fission, neutron yield per fission, and fission neutron energy spectrum are determined.
Both the spatial shape of the axial power distribution and the energy spectrum of the source were treated in the analytic uncertainty analysis2 . The following methods were employed to assess the effects of uncertainties in these inputs:
- Burn up of the peri pheral fue I assemblies - Perturbations in fuel assembly burnup impact the fission neutron energy spectrum, neutron yield per fission, and energy released per fission for each peripheral fuel assembly. A 5000 MWD/MTU uncertainty in the peripheral fuel assembly burnups is considered conservative. The sensitivity study is performed using a series of calculations starting with mid-cycle burnup at 3000 MWD/MTU, and 5000 MWD/MTU to 50,000 MWD/MTU with 5000 MWD/MTU delta mid-cycle burnup between each run.
- Axial power distri bution - Based on variations in axial peaking factors over the course of a fuel cycle, a 10% uncertainty in the shape of the axial power distribution is considered conservative.
This estimate was derived from a review of numerous axial distributions from a wide variety of pressurized water reactors employing both low leakage and non-low leakage fuel management.
The Waterford 3 upper-middle girth weld (106-121) is located at the reactor vessel inner radius at an elevation 52 cm relative to the top of the active fuel. The analytic uncertainty analysis was performed generically for a Westinghouse 4-loop design, and it considered a to W3F1-2017-0065 Page 25 of 31 location 52 cm above the top of the active fuel. The following uncertainty components were attributed as a result of the analytic uncertainty analysis2 :
Table 21: Contributions to calculational uncertainty from peripheral assembly burnup and axial power distribution2 Uncertainty Component RPV Inside Radius +52 cm Relative to Top-of-Core Elevation Peripheral Assembly Burnup (+/-5000 MWD/MTU) 1%
Axial Power Distribution 10%
5.2.2 Biological Shield Concrete Composition In the modeling of the reactor cavity region a standard, "generic" concrete composition was assumed.
Biological shield concrete composition was not considered in the analytic uncertainty analysis described in Reference 2 because it was deemed unlikely to significantly affect the calculated values of fast neutron (E > 1.0 MeV) fluence for materials near or above the 1.00E+17 n/cm 2 threshold. Note that the analytic uncertainty estimate includes an "other factors" uncertainty component (5%), intended to address geometrical or operational variables that individually have an insignificant effect on the overall uncertainty, but collectively should be accounted for in the assessment.
A concrete hydrogen number density of 7.8E-03 atoms/b-cm is assumed in the concrete mixture. This is near the low end of the hydrogen content range considered by Oak Ridge National Laboratory (ORNL) in their recent study of extended beltline neutron fluence calculations 1 . The ORNL results suggest that lower hydrogen content would tend to provide conservative results relative to assumed concrete compositions with higher hydrogen content.
5.2.3 Cavity gap between the reactor vessel and biological shield Plant drawings showing as-designed/as-built reactor cavity dimensions were used in the construction of the radiation transport models.
Reactor cavity air ~ap dimension was not considered in the analytic uncertainty analysis . The effect of this air gap is accounted for in the radiation transport calculations, and any uncertainty in this value was not deemed a significant contributor to the calculated values of fast neutron (E > 1.0 MeV) fluence for materials near or above the 1.00E+17 n/cm 2 threshold. Note that the analytic uncertainty estimate includes an "other factors" uncertainty component (5% ), intended to address to W3F1-2017-0065 Page 26 of 31 geometrical or operational variables that individually have an insignificant effect on the overall uncertainty, but collectively should be accounted for in the assessment.
5.2.4 Homogenized materials above and below the active core region The radiation transport model used an approximate mixture of stainless steel and borated water for regions directly above and below the core.
The effect of varying assumptions in the balance between water and steel was not considered in the analytic uncertainty analysis described in Reference 2. The model that formed the basis of the analytic uncertainty analysis used material mixtures derived from drawings of the reactor internals, and any uncertainties in these values were not deemed significant contributors to the calculated values of fast neutron (E > 1.0 MeV) fluence for materials near or above the 1.00E+17 n/cm 2 threshold. Note that the analytic uncertainty estimate includes an "other factors" uncertainty component (5% ), intended to address geometrical or operational variables that individually have an insignificant effect on the overall uncertainty, but collectively should be accounted for in the assessment.
5.2.5 Discretization effects on deterministic calculations The radiation transport model used a cylindrical mesh grid with 160 radial, 121 azimuthal, and 247 axial intervals. Anisotropic scattering was treated with a P5 Legendre expansion, and the angular discretization was modeled with an S 16 order of angular quadrature.
Appendix F of WCAP-18060-NP, Rev. 13 compares calculations performed with Sa quadrature and S12 quadrature sets for three locations on the inner radius of the Catawba reactor vessel. Reactor vessel Weld W06 is located 42 cm above the elevation of the top of the active core, Weld W05 is located 12 cm above the middle of the core, and Weld W04 is located 26 cm below the bottom of the core. Appendix F of Reference 3 shows no significant difference between calculated values of fast neutron (E > 1.0 MeV) fluence rate for any of the materials evaluated. The additional study discussed in section 2.3.2 which compared fluences calculated using the P3 , Sa and P5 , S16 Legendre expansion and order of quadrature showed that the difference in flux at the upper circumferential weld location was bounded by 0.4%. Therefore, any changes in quadrature were deemed unlikely to significantly change the calculated fluence.
The analytic uncertainty analysis documented in Reference 2 did not specifically evaluate discretization effects because they were considered unlikely to significantly affect the calculated values of fast neutron (E > 1.0 MeV) fluence for materials near or above the to W3F1-2017-0065 Page 27 of 31 1.00E+17 n/cm2 threshold. Note that the analytic uncertainty estimate includes an "other factors" uncertainty component (5%), intended to address geometrical or operational variables that individually have an insignificant effect on the overall uncertainty, but collectively should be accounted for in the assessment.
The judgment that discretization effects are unlikely to significantly affect calculated neutron fluence results is supported by the results in Reference 1. The Waterford 3 reactor vessel materials likely to be near or above fast neutron (E > 1.0 MeV) fluence levels of 1.00E+17 n/cm 2 in the extended beltline region are below the bottom extents of the inlet and outlet nozzle welds. The Reference 1 analysis (slide 53) indicates that the difference between using QR16 and S16 quadrature in these regions is negligible; therefore, a 5% uncertainty for "other factors" sufficiently encompasses any effect that could be imparted by different quadrature settings.
5.2.6 Conclusion The variation of axial power distribution and isotopic content of U-235 and Pu-239 in the source were considered in the RAPTOR-M3G analytic uncertainty analysis for Waterford 3. The effects of variation in biological shield concrete composition, reactor cavity gap, homogenized material approximation above the core, and discretization effects were not explicitly considered in the analytic uncertainty analysis. However, the effect of those inputs were deemed to be insignificant, in part because the area in which they have the largest effect, the nozzles/nozzle welds, is not included in the Waterford 3 extended beltline region. Since these and other inputs may have a small impact on the calculated fluence, an "other factors" contribution of 5% was included in the overall calculational uncertainty to envelope them.
5.3 Measurement Benchmarks for the Extended Beltline Regulatory Guide 1.190 Reg. Position 1.4.2 recommends the use of pressure vessel neutron dosimetry to benchmark neutron fluence calculation methods 15 . Neither the Waterford 3 fluence calculations, nor any RAPTOR-M3G fluence calculations, have been benchmarked with dosimetry in the vessel nozzle region because such dosimetry does not exist.
Of the four dosimetry and measurement benchmarks provided as possible references in Item 3 of the NRC Discussion Topics 14, two were calculated using the DORT method, one was not relevant to ex-core regions, and the last was not subjected to a rigorous Westinghouse quality assurance process.
Therefore, these measurement benchmarks are not considered as adequate justification for the accuracy of the RAPTOR-M3G calculations in the extended beltline region at Waterford 3.
Figure 2 in Section 4.0 shows the axial fluence distribution at the pressure to W3F1-2017-0065 Page 28 of 31 vessel inner radius and outer radius. As discussed in section 5.2, the ends of the active core region experience an increasing axial flux gradient at the vessel outer radius, as shown in Figure 2. The flux gradient at the vessel outer radius is approximately the same as that experienced at the upper circumferential weld, which is the limiting material in the extended beltline region. Table 16 and Table 17 also show the increased sensitivity of the upper circumferential weld (vessel inner radius) and top-of-core (reactor cavity) regions to the axial power distribution. The analytic uncertainty increases from 1% to 10% and 3% at the upper circumferential weld (at the vessel inner radius) and top-of-core (in the reactor cavity), respectively. The M/C ratio of 0.93 in Table 19 for EVND located at the upper and lower core extents shows that the RAPTOR-M3G method accurately calculates the neutron fluence in regions which experience high axial flux gradients, similar in magnitude to the upper circumferential weld, and which are subject to high sensitivity of the calculated flux to axial power distribution. This 0.93 M/C ratio supports the 18% overall method uncertainty for RAPTOR-M3G at the limiting extended beltline material, circumferential weld 106-121 (at the vessel inner radius), which is less than the recommended 20% uncertainty by Regulatory Guide 1.19015 . Since the reactor vessel nozzles and nozzle welds are not part of the Waterford 3 extended beltline region due to their projected fluence being well below the 1017 n/cm 2 threshold, and since there is high confidence in the fluence calculation in the axially-adjacent plate/circumferential weld area which exhibits a fluence decrease below 10 17 n/cm 2 , qualification of RAPTOR-M3G calculations in the nozzle/nozzle weld region was not performed.
In conclusion, the existing off-midplane EVND provide sufficient confidence that the RAPTOR-M3G fluence calculations are accurate within 20% in the Waterford 3 extended beltline region.
5.4 Calculation Benchmark for Extended Beltline Regulatory Guide 1.190 Reg. Position 1.4.2 recommends the use of a calculational benchmark for new neutron fluence calculation methods 15 . The purpose of the benchmark is to provide "a detailed assessment and verification of the numerical procedures, code implementation, and the various modeling approximations relative to state-of-the-art solutions ... " The fluence calculation benchmark does not provide solutions for an axial domain beyond the active core region, and therefore, it cannot be used to qualify calculation methods in the extended beltline region.
RAPTOR-M3G was used to calculate the NRC fluence calculation benchmarks recommended in Regulatory Guide 1.19015 . Results are reported in the License Amendment Request for using RAPTOR-M3G for fluence calculations at Waterford 311 . They show that the benchmark/RAPTOR-M3G difference is bounded by 5% at all azimuthal locations reported, for z =
67.1048cm. The difference influence calculated by the two methods shows that RAPTOR-M3G is appropriately applying the discrete ordinates method.
to W3F1-2017-0065 Page 29 of 31 The analytic uncertainty analysis has demonstrated an overall calculational uncertainty of 18% at the limiting extended beltline material, circumferential weld 106-121. This is lower than the 20% uncertainty maximum for calculation of changes in RT NDT and RTPTS* Additionally, EVND benchmarking in off-midplane regions discussed in sections 3.1 and 5.3 validates the fluence calculations in regions of increased axial power distribution sensitivity, and which also exhibit similar vessel outer radius flux gradients as the limiting extended beltline material. The reactor vessel nozzles and their associated welds will experience a 60-EFPY fluence below 1017 n/cm 2 , and therefore they are not considered part of the extended beltline.
In conclusion, based on the agreement of RAPTOR-M3G with the NRC fluence calculation benchmark and the availability of sufficient dosimetry benchmarking, the suggested action to perform site-specific calculational benchmarking would not significantly contribute to increased confidence of adequate margin in the Waterford 3 fluence calculations and pressure-temperature limit curves.
to W3F1-2017-0065 Page 30 of 31 References
- 1. NRC Letter, "Summary U.S. Nuclear Regulatory Commission Computation of Neutron Fluence Information Exchange Public Meeting," February 7, 2017.
(Available as ADAMS Accession Number ML17038A134. The meeting presentation slides are available as ADAMS Accession Numbers ML17038A135 and ML17038A136.)
- 2. Westinghouse Letter LTR-REA-16-117, Rev. 4, "Response to the NRC Request for Additional Information Regarding RAPTOR-M3G on the Waterford Unit 3 License Renewal Application." November 2017.
- 3. Westinghouse Report WCAP-18060-NP, "Response to RAls Concerning the Use of RAPTOR-M3G for the Catawba Unit 1 Measurement Uncertainty Recapture (MUR) Power Uprate Fluence Evaluations," November 2015.
- 4. Westinghouse Report WCAP-13348, Rev. 0, "Consumers Power Company Palisades Nuclear Plant Reactor Vessel Fluence Analysis," May 1992.
- 5. Westinghouse Report WCAP-13362, Rev. 0, "Westinghouse Fast Neutron Exposure Methodology for Pressure Vessel Fluence Determination and Dosimetry Evaluation," May 1992.
- 6. RE. Maerker, "Application of LEPRICON Methodology to LWR Pressure Vessel Surveillance Dosimmetry," Reactor Dosimetry, Proc. 6th ASTM-Euratom Symposium, Jackson Hole, WY, May 31-June 5, 1987, American Society of Testing and Materials (1989).
- 7. Westinghouse Letter LTR-REA-17-75, Rev. 2, "Evaluation per NEI 96-07 Section 4.3.8.2 for Changing from Use of DORT code to the RAPTOR-M3G Code for Waterford Unit 3." November 2017.
- 8. Westinghouse Letter LTR-REA-17-56, Rev. 2, "Documentation Demonstrating Adherence to Regulatory Guide 1.190 and Response to the NRC Request for Additional Information Regarding Neutron Fluence Determination in the Extended Beltline Region on the Waterford Unit 3 License Renewal Application." November 2017.
- 9. Westinghouse Report WCAP-17969-NP, Rev. 2, "Analysis of Capsule 83° from the Entergy Operations, Inc. Waterford Unit 3 Reactor Vessel Surveillance Program," November 2017.
- 10. Westinghouse Report WCAP-18002-NP, Rev. 0, "Waterford Unit 3 Time-Limited Aging Analysis on Reactor Vessel Integrity," July 2015.
- 11. W3F1-2017-0065, Attachment 2, Waterford 3 License Amendment Request for Use of RAPTOR-M3G Code for Neutron Fluence Calculations.
to W3F1-2017-0065 Page 31 of 31 12.SEP-RVl-007, "Waterford 3 Reactor Vessel Internals Inspection (RVII) Program Plan."
- 13. W3F1-2016-0012, "License Renewal Application Waterford Steam Electric Station, Unit 3." March 23, 2016.
- 14. "NRC Staff Discussion Topics for the October 19, 2017 Category 1 Public Meeting with Entergy Operations, Inc." ADAMS Accession Number ML17290A049.
- 15. Regulatory Guide 1.190, "Calculational and Dosimetry Methods for Determining Pressure Vessel Neutron Fluence," U.S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research, March 2001.
16.G146 Rev. 18, "General Arrangement Reactor Building-Section-Sh1."