ML19226A263

Safety Evaluation by the Office of Nuclear Reactor Regulation Topical Report PWROG-17090-NP, Revision 0 Generic Rotterdam Forging and Weld Initial Upper-Shelf Energy Determination Pressurized Water Reactor Owners Group Westinghouse Electric

ML19226A263
Person / Time
Issue date:	08/14/2019
From:	Dennis Morey NRC/NRR/DLP/PLPB
To:
Drake J, NRR/DLP., 415-8378
Shared Package
ML19226A255 -Pkg.	List: ML19226A259 ML19226A263
References
EPID L-2018-TOP-0017, PWROG-17090-NP, Rev. 0
Download: ML19226A263 (24)

Text

1 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION 2

3 TOPICAL REPORT PWROG-17090-NP, REVISION 0, 4

5 GENERIC ROTTERDAM FORGING AND WELD INITIAL 6

7 UPPER-SHELF ENERGY DETERMINATION 8

9 PRESSURIZED WATER REACTOR OWNERS GROUP 10 11 EPID: L-2018-TOP0017 12 13 14

1.0 INTRODUCTION

15 16 By letter dated April 6, 2018 (Agencywide Documents Access and Management System 17 (ADAMS) Package Accession No. ML18114A173), the Pressurized Water Reactor Owners 18 Group (PWROG) submitted Topical Report (TR) PWROG-17090-NP, Revision (Rev.) 0, 19 Generic Rotterdam Forging and Weld Initial Upper-Shelf Energy Determination (Ref. 1) for 20 U.S. Nuclear Regulatory Commission (NRC) review and approval. Additional information 21 related to PWROG-17090-NP, Rev. 0 (also referred to herein as the TR) was submitted by letter 22 dated June 12, 2019 (Ref. 2) in response to a request for additional information (RAI) from the 23 NRC staff.

24 25 For reactor vessels (RVs) that were fabricated by the Rotterdam Dockyard Company 26 (Rotterdam), the TR provides generic values of the unirradiated Charpy Upper-Shelf Energy 27 (USE) for SA508, Class 2 RV forgings; and generic values of unirradiated Charpy USE, weight 28 percentage copper (Cu) content, and weight percentage nickel (Ni) content for RV Submerged 29 Arc Welds (SAWs) and Shielded Metal Arc Welds (SMAWs). The PWROGs submittal letter 30 identifies that licensees will reference the PWROG-17090-NP, Rev. 0 report as the basis for the 31 generic USE, Cu content, and Ni content values to demonstrate compliance with Title 10 of the 32 Code of Federal Regulations (10 CFR) Part 50, Appendix G requirements for extended 33 operating license terms when plant-specific RV material information is incomplete or not 34 available.

35 36 As addressed in the PWROGs April 6, 2018, submittal letter, this TR is for implementation by all 37 U.S. PWRs with RVs fabricated by Rotterdam in the late 1960s and early 1970s timeframe.

38 This statement also defines the limitation on the applicability of the TR, as per the PWROGs 39 April 6, 2018, submittal letter.

40 Enclosure

1

2.0 BACKGROUND

AND REGULATORY EVALAUTION 2

3 Background - Generic RV Properties, Application to RV Fracture Toughness Evaluations 4

5 Terms such as generic values, generic data, or best estimate values, etc. are often used in 6 industry reports and in NRC staff publications for addressing certain RV material properties that 7 are based on statistical evaluation of a set of original fabrication data for a class of RV 8 material1. Data sets such as those provided in the TR are developed from available RV 9 fabrication records for multiple plants and are often applicable to certain plant and/or RV 10 material categories. When used in licensing applications for meeting regulatory requirements 11 discussed below, generic RV material properties are subject to review and approval by the NRC 12 staff.

13 14 When approved by the NRC staff, generic RV material properties may be implemented in plant-15 specific licensing applications to demonstrate compliance with the requirements of 10 CFR 16 Part 50, Appendix G, Fracture Toughness Requirements. For PWR plants, generic RV 17 material properties may also be implemented in applications for addressing the requirements of 18 10 CFR 50.61, Fracture toughness requirements for protection against pressurized thermal 19 shock events, or the requirements of 10 CFR 50.61a, Alternate fracture toughness 20 requirements for protection against pressurized thermal shock events, as appropriate. In 21 applications for License Renewal (LR) and Subsequent License Renewal (SLR) under 10 CFR 22 Part 54, generic properties may be used in RV neutron embrittlement evaluations to meet the 23 technical information requirements for TLAAs, as set forth in 10 CFR 54.21(c)(1). TLAA 24 evaluations related to RV neutron embrittlement, pursuant to 10 CFR 54.21(c)(1), rely upon 25 demonstrations that the above 10 CFR Part 50 fracture toughness requirements are satisfied (or 26 will be satisfied) for proposed extended license terms. It should be noted that while generic 27 properties are used as inputs into time-dependent neutron embrittlement analyses, the 28 properties themselves are fixed based on the evaluation of available RV fabrication data for the 29 preservice (unirradiated) condition, and once approved they become incorporated into a plants 30 licensing basis documentation2.

31 32 Fracture Toughness Requirements and Guidance - Ferritic RCPB and RV Beltline Materials 33 34 Pursuant to Section IV.A of 10 CFR Part 50, Appendix G, pressure-retaining components of the 35 reactor coolant pressure boundary (RCPB) that are made of ferritic materials must meet the 36 requirements of the American Society of Mechanical Engineers (ASME) Boiler and Pressure 37 Vessel Code (Code),Section III, as supplemented by the additional requirements set forth in 38 paragraph IV.A.1, Reactor Vessel Charpy Upper-Shelf Energy Requirements, and 39 paragraph IV.A.2, Pressure-Temperature Limits and Minimum Temperature Requirements, of 40 the Rule. With respect to Charpy USE requirements, paragraph IV.A.1.a of 10 CFR Part 50 41 Appendix G states:

1 For generic values of the initial (unirradiated) reference temperature (RTNDT), 10 CFR 50.61, 50.61a; and NRC Regulatory Guide 1.99, Revision 2, state that the class of material is generally determined for welds by the type of welding flux (e.g., Linde 80 or other), and for base metal by the material specification. The material specification is generally the ASTM or ASME standard specification (e.g., ASME Section II, SA508, Class 2 or other).

2 NRC Branch Technical Position 5-3, Position 1.3, Reporting Requirements, states Fracture toughness information identified by the ASME Code and by Appendix G, 10 CFR Part 50, should be reported in the final safety analysis report (FSAR) to provide a basis for evaluating the adequacy of the operating limitations given in the [technical specifications] or [pressure-temperature limits report].

1 Reactor vessel beltline materials must have Charpy upper-shelf energy1 in the 2 transverse direction [i.e., weak direction] for base metal and along the weld for 3 weld material according to the ASME Code [Section III], of no less than 75 ft-lbs 4 (102 J) initially and must maintain Charpy upper-shelf energy throughout the life 5 of the vessel of no less than 50 ft-lbs, unless it is demonstrated in a manner 6 approved by the Director, [NRR] or Director, [NRO], as appropriate, that lower 7 values of Charpy upper-shelf energy will provide margins of safety against 8 fracture equivalent to those required by Appendix G of Section XI of the ASME 9 Code.

10 11 Note 1 of this paragraph states that Charpy USE is defined in American Society of Testing and 12 Materials (ASTM) Standard E185-82, Standard Practice for Conducting Surveillance Tests for 13 Light Water Cooled Nuclear Power Reactor Vessels (Ref. 3) which is incorporated by reference 14 in 10 CFR Part 50, Appendix H. Section 4 of ASTM E185-82 provides the following definitions 15 that are applicable to the determination of Charpy USE based on actual Charpy V-Notch Impact 16 Tests (also referred to in this SE as measured values of Charpy USE).

17 18

• Paragraph 4.17 defines the Charpy transition curve as a graphic presentation of 19 Charpy data, including absorbed energy, lateral expansion, and fracture appearance, 20 extending over a range including the lower shelf energy (less than (<) 5 percent shear),

21 the transition region, and the upper-shelf energy (greater than (>) 95 percent shear).

22 23

• Paragraph 4.18 defines the upper shelf energy level as the average energy value for all 24 Charpy specimens (normally three) whose test temperature is above the upper end of 25 the transition region. For specimens tested in sets of three at each test temperature, the 26 set having the highest average may be regarded as defining the upper-shelf energy.

27 28 Section IV.A of 10 CFR Part 50, Appendix G states that for ferritic materials that are part of the 29 RV Beltline Region, as defined3 in Section II of the Rule, the values of the reference 30 temperature (RTNDT, also defined in Section II of the Rule) and Charpy USE must account for 31 the effects of neutron radiation, including the results of the RV surveillance program required by 32 10 CFR Part 50, Appendix H. For protection of PWR RVs against pressurized thermal shock 33 (PTS) events, 10 CFR 50.61 also requires that the RTNDT for RV beltline materials account for 34 the effects of neutron radiation. The regulation at 10 CFR 50.61 defines RTPTS as the RTNDT 35 evaluated for the EOL Fluence for each of the RV beltline materials, using the procedures in 36 10 CFR 50.61(c). The regulation of 10 CFR 50.61 defines EOL Fluence as the best-estimate 37 neutron fluence projected for a specific RV beltline material at the clad-base-metal interface at 38 the location where the material receives the highest fluence on the expiration date of the 39 operating license. For PWR plants implementing the alternate PTS protection requirements of 40 10 CFR 50.61a, 10 CFR 50.61a(c)(1) requires that each licensee shall have projected values of 41 RTMAX-X, as defined in 10 CFR 50.61a(a)(6), for each RV beltline material for the EOL fluence of 42 the material.

43 44 The NRC Regulatory Issue Summary (RIS) 2014-11 (Ref. 4) identifies that the beltline definition 45 in 10 CFR Part 50, Appendix G is applicable to all RV ferritic materials with projected neutron 46 fluence values greater than 1.0 x 1017 n/cm2 (E > 1.0 MeV), and this fluence threshold remains 3

Section II of 10 CFR Part 50, Appendix G defines the RV beltline as the region of the RV (including welds, heat affected zones, and plates or forgings) that directly surrounds the effective height of the active core and adjacent regions of the RV that are predicted to experience sufficient neutron radiation damage to be considered in the selection of the most limiting material with regard to radiation damage.

1 applicable throughout the licensed operating period. Accordingly, RIS 2014-11 states that the 2 effects of neutron radiation must be considered for any RV locations that are predicted to 3 experience a neutron fluence exposure greater than 1.0 x 1017 n/cm2 (E > 1.0 MeV) at the end 4 of the licensed operating period; this includes periods of extended operation for LR and SLR.

5 6 In order to account for the effects of neutron embrittlement on RV beltline materials, Regulatory 7 Guide (RG) 1.99, Rev. 2 (Ref. 5) specifies methods for calculating projected values of Charpy 8 USE and adjusted RTNDT due to neutron fluence exposure. Procedures for calculating RTPTS 9 are specified directly in 10 CFR 50.61(c); and procedures for calculating RTMAX-X are specified in 10 paragraphs (f) and (g) of 10 CFR 50.61a. For RV beltline materials that are not represented in 11 the RV surveillance program, RG 1.99, Rev. 2 provides methods for direct calculation of 12 projected values of Charpy USE and adjusted RTNDT based on the weight percentage Cu and Ni 13 content and projected neutron fluence exposure of the RV beltline materials. The regulation at 14 10 CFR 50.61 specifies methods for direct calculation of RTPTS based on weight percentage Cu 15 and Ni content and EOL fluence. Per RG 1.99, Rev. 2, only the Cu content is needed to 16 determine the projected percentage decrease in USE as a function of projected neutron fluence 17 per Figure 2 of the RG. Per RG 1.99, Rev. 2 and 10 CFR 50.61, both Cu and Ni content are 18 needed in order to determine the chemistry factor (CF) for the material using CF Tables 19 provided therein. For RV beltline materials, the product of the CF and the neutron fluence factor 20 (FF) determines the projected mean value of the shift in RTNDT (RTNDT). These procedures 21 specify that the projected value of the adjusted RTNDT (or RTPTS under 10 CFR 50.61) is equal to 22 the sum of the values of unirradiated (initial) RTNDT, RTNDT, and a margin term (M). For PWR 23 plants implementing the alternate PTS protection requirements of 10 CFR 50.61a, calculation 24 procedures in Paragraphs (g) and (f) of this Rule specify more detailed inputs and equations for 25 calculating RTMAX-X; these inputs include, among other things, Cu content, Ni content, 26 phosphorus (P) content, manganese (Mn) content, and EOL neutron fluence.

27 28 If the RV beltline materials are represented in the RV surveillance program, RG 1.99, Rev. 2 29 and 10 CFR 50.61 specify methods for calculating projected USE, adjusted RTNDT, and RTPTS 30 that are based on measurements of percentage decrease in Charpy USE and RTNDT from 31 Charpy impact tests of irradiated surveillance materials. For PWR plants implementing the 32 alternate PTS protection requirements of 10 CFR 50.61a, paragraph (f)(6)(i) of this Rule 33 specifies that the licensee shall evaluate the results from a plant-specific or integrated 34 surveillance program if the surveillance data satisfy the criteria described in paragraphs 35 (f)(6)(i)(A) and (f)(6)(i)(B) of this section.

36 37 Requirements and Guidance, Preservice Fracture Toughness Tests, and Analysis of Test Data 38 39 Pursuant to 10 CFR Part 50, Appendix G, Section III, Fracture Toughness Tests, ferritic 40 materials for pressure-retaining components of the RCPB must be tested in accordance with the 41 ASME Code,Section III and, for RV beltline materials, the RV surveillance program test 42 requirements of 10 CFR Part 50, Appendix H, in order to demonstrate compliance with the 43 fracture toughness requirements in Section IV.A of the Rule. For an RV that was constructed to 44 an Edition and Addenda of the ASME Code,Section III earlier than the Summer 1972 Addenda 45 of the 1971 Edition,Section III of 10 CFR Part 50, Appendix G states the fracture toughness 46 data and data analyses must be supplemented in a manner approved by the NRC to 47 demonstrate equivalence with these fracture toughness requirements.

48 49 The NRC guidance in NUREG-0800, Branch Technical Position (BTP) 5-3 (Ref. 6), states that 50 the preservice fracture toughness test requirements for plants with construction permits issued 51 prior to August 15, 1973, may not comply with the later Codes and Regulations in all respects.

1 Accordingly, Section B.1, Preservice Fracture Toughness Test Requirements, of BTP 5-3 2 recommends that the preservice fracture toughness propertiesspecifically, unirradiated RTNDT 3 and unirradiated Charpy USEof the ferritic materials for these plants should be assessed by 4 using the available test data to estimate the preservice fracture toughness properties in the 5 same terms as the new requirements.

6 7 With respect to estimation of Charpy USE for the preservice (unirradiated) condition, 8 Position 1.2 of BTP 5-3 specifies that if Charpy impact tests were only conducted on longitudinal 9 specimens (i.e., Charpy V-Notch specimens oriented in the strong direction), the Charpy USE 10 values should be reduced to 65 percent of the measured longitudinal values to estimate the 11 transverse USE (i.e., USE for the weak direction).

12 13 For cases where there is insufficient test data in Certified Material Test Reports (CMTRs4) to 14 establish measured values of these properties for a plants own RV materials5 using BTP 5-3 15 methods, the implementation of NRC-approved generic estimates based on generic data for a 16 material class may be appropriate, especially for older plants. The regulations at 10 CFR 17 50.61, 10 CFR 50.61a, and RG 1.99, Rev. 2 have provisions that address the use of generic 18 data to demonstrate compliance with these fracture toughness requirements. The staffs 19 overview of criteria for use of generic data, as applied to the determination of generic values of 20 unirradiated Charpy USE, Cu content, and Ni content, is provided below.

21 22 Generic Values for Unirradiated Charpy USE, Cu Content, and Ni Content 23 24 The NRC regulations and guidance in 10 CFR Part 50, Appendix G and RG 1.99, Rev. 2 do not 25 provide explicit criteria regarding the implementation of generic unirradiated USE values for a 26 class of material for the pre-service condition. However, with respect to generic values of 27 initial (unirradiated) RTNDT that are used for P-T limits and PTS evaluations, 10 CFR 50.61 and 28 RG 1.99, Rev. 2 both state that if a measured value of initial RTNDT is not available, a generic 29 mean value of initial RTNDT for the class of material (as specified above) may be used if there 30 are sufficient test results to establish a mean and standard deviation () for the class. Per 31 RG 1.99, Rev. 2 and 10 CFR 50.61, the standard deviation on initial RTNDT is i (referred to as 32 U in 10 CFR 50.61) and is incorporated into the calculation of the adjusted RTNDT (or RTPTS) 33 due to the effects of neutron embrittlement by using a margin term, M. Equation (4) of the RG 34 and Equation (2) of 10 CFR 50.61 specify the same expression for the margin term, which is 35 shown below:

36 37 = x +

38 39 If only the preservice condition for the RV material is considered, the RTNDT term and the 40 standard deviation on RTNDT () are eliminated, and the expression for the margin term on 4

For later Editions and Addenda of the ASME Code,Section III, Certified Material Test Reports, as defined in NCA-9200 of ASME Section III, are required for Class 1 pressure-retaining materials, as specified in NB-2130. For impact testing of ferritic RPCB materials in accordance with NB-2300, the test results, test temperatures, specimen orientation and location, etc., as applicable, for all impact tests (Charpy V-Notch Tests and Drop Weight Tests) performed to meet the requirements of NB-2330 shall be reported in the CMTR, as specified in NB-2321 (later Editions and Addenda of ASME Section III).

5 For the regulatory applications addressed herein, plant-specific RV materials are often identified using a specific component identifier and an identifier for the specific heat (stated tonnage of metal obtained from a period of continuous melting) of material used to fabricate the plants RV weld, forging, or plate.

1 initial RTNDT reduces to twice the standard deviation on initial RTNDT (2i). Thus, the generic 2 value of initial RTNDT for the material class equals the mean value plus 2i. Equivalently, for 3 determinations of a generic value of unirradiated Charpy USE, where only the standard 4 deviation on the unirradiated property is considered, the appropriate value for a bounding 5 statistical representation of the property for a class of material is the mean value minus two 6 standard deviations (Mean - 2), since lower USE values are more bounding (as opposed to 7 RTNDT, where higher values are more bounding).

8 9 If measured values of Cu and Ni content for plant-specific RV beltline materials are not 10 available, Position 1.1 of RG 1.99, Rev. 2 and 10 CFR 50.61(c)(1)(iv)(A) provide equivalent 11 criteria regarding the use of generic values. Specifically, if measured Cu and Ni content are 12 unknown, the upper limiting values given in the material specifications may be used. If the 13 material specifications provide no upper limiting values, conservative estimates (Mean + 1) 14 based on generic data may be used if justification is provided. If there is no information 15 available based on measured content, material specifications, or conservative estimates 16 (Mean + 1) from generic data, 0.35 percent copper and 1.00 percent nickel must be 17 assumed. For calculations of RTMAX-X, 10 CFR 50.61a(f)(3) states that if measured values of 18 Cu, Ni, P, and Mn content are not available for the specific RV material, either the upper limiting 19 values given in the material specifications to which the RV material was fabricated, or 20 conservative estimates (i.e., mean plus one standard deviation) based on generic data must be 21 used. Table 4 of 10 CFR 50.61a provides the generic values for P and Mn content, which must 22 be used, if measured values are unknown for the specific RV material.

23 24 Finally, with respect to determination of generic values of Cu and Ni content based on 25 evaluation of Mean + 1 for generic data, Note 4 of 10 CFR 50.61(c)(1)(iv)(A) and Note 4 of 26 10 CFR 50.61a(f)(3) state: Data from reactor vessels fabricated to the same material 27 specification in the same shop as the vessel in question and in the same time period is an 28 example of generic data.

29 30 The NRC staff applied these criteria to determine whether the TR evaluations provide 31 acceptable generic estimates of unirradiated Charpy USE, Cu Content, and Ni content for use in 32 plant licensing applications to demonstrate compliance with the requirements of 10 CFR 33 Part 50, Appendix G; 10 CFR 50.61 or 50.61a; and 10 CFR 54.21(c)(1).

34 35 3.0 OVERVIEW OF PWROG-17090-NP, REVISION 0 36 37 For RVs fabricated by Rotterdam, the TR provides generic values of unirradiated Charpy USE 38 for SA508, Class 2 forgings and generic values of unirradiated Charpy USE, Cu content, and Ni 39 content for SAWs and SMAWs when no or limited plant-specific material information is 40 available. These generic values are developed using baseline (unirradiated) test data from RV 41 material surveillance program records and CMTRs that are available to Westinghouse. The TR 42 identifies that the need for these generic properties is prompted by the difficulty in identifying 43 plant-specific material information needed to establish measured values for these properties for 44 Rotterdam RVs fabricated in the late 1960s to early 1970s. The proposed generic values for 45 these material properties are as follows:

46

1

• For an SA508, Class 2 Rotterdam RV forging supplied by Rheinstahl Huttenwerke AG, 2 the TR provides a generic lower bound Charpy USE value of 56 ft-lbs, based on the 3 mean minus two standard deviations (Mean - 2) evaluation of measured USE values 4 for the Rheinstahl Huttenwerke AG data set.

5 6

• For an SA508, Class 2 Rotterdam RV forging supplied by Fried-Krupp Huttenwerke AG, 7 or an unknown Rotterdam RV forging supplier, the TR provides a generic lower bound 8 Charpy USE value of 52 ft-lbs based on the Mean - 2 evaluation of measured USE 9 values for the Fried-Krupp Huttenwerke AG data set, and consideration of additional 10 Charpy USE data for other forging suppliers addressed in Section 4.3 of the TR and 11 discussed below.

12 13

• For a Rotterdam RV SAW, the TR provides a generic lower bound Charpy USE value of 14 75 ft-lbs based on the Mean - 2 evaluation of measured USE values for the Rotterdam 15 RV SAW data set. The TR also provides a generic upper bound Cu content of 16 0.23 percent by weight, and a generic upper bound Ni content of 0.56 percent by weight, 17 both of which are based on the mean plus one standard deviation (Mean + 1) 18 evaluation for the Rotterdam RV SAW Cu and Ni chemistry data.

19 20

• For a Rotterdam RV SMAW, the TR provides a generic lower bound Charpy USE value 21 of 72 ft-lbs, which is the lowest of the lower bound USE values (as described in 22 Sections 3.2 and 5.2 for the Rotterdam SMAW evaluation) for the non-outlier SMAW 23 weld heats. The lower bound USE values are determined from measured absorbed 24 impact energies for the SMAW Charpy V-Notch tests, which are established to be below 25 an undetermined USE based on Charpy test temperatures at 10.4 °F or below and 26 available percent shear data. The TR recommends a generic Cu content of 0.35 percent 27 per RG 1.99, Rev. 2, and it provides a generic upper bound Ni content of 1.13 percent 28 (greater than the generic Ni content of 1.00 percent provided in the RG) based on the 29 Mean + 1 evaluation for the Rotterdam RV SMAW Ni chemistry data.

30 31 Section 3.0 of the TR describes the methodology for analyzing the material property data for the 32 subject Rotterdam RV forgings and welds. Section 4.0 of the TR provides the actual data sets 33 and statistical analyses for determining the generic values of unirradiated Charpy USE for 34 Rotterdam RV forging materials. Section 5.0 provides the data sets and analyses for 35 determining the generic values of unirradiated Charpy USE, Cu content, and Ni content for 36 Rotterdam RV weld materials. The NRC staffs independent evaluation of these methods and 37 data analyses is documented in Section 4.0 of this SE. The key aspects of the methodology 38 and data analyses, as reported in TR, are summarized below.

39 40 Generic Charpy USE Values for SA508, Class 2 Forgings in Rotterdam RVs 41 42 The TR identifies that its generic Charpy USE values are determined by calculating the 43 Mean - 2 for two independent sets of measured Charpy USE values in the weak direction.

44 The two independent data sets correspond to two Rotterdam RV forging suppliers: Rheinstahl 45 Huttenwerke AG and Fried-Krupp Huttenwerke AG. The TR indicates that measured USE 46 values were determined by reviewing Charpy impact test records from available CMTRs and 47 baseline (unirradiated) Charpy test data from RV material surveillance program records. The 48 total number of Charpy test records for forgings supplied by Rheinstahl Huttenwerke AG is 38, 49 and the total number of test records for Fried-Krupp Huttenwerke AG-supplied forgings is 67.

50 Each Charpy test record corresponds to a subset of Charpy impact tests on samples

1 (i.e., Charpy V-Notch specimens) of a specific forging material for an unnamed plant (e.g., Inlet 2 Nozzle 09 for Plant D, Upper Shell for Plant C, etc.). For each test record, the TR 3 reviewed measurements of absorbed Charpy impact energies, test temperatures, and 4 measurements of percentage shear fracture surface areas (percent shear) for the fractured 5 Charpy specimens (i.e., the broken pieces).

6 7 For each Charpy test record, the TR determined either a measured value of the USE for that 8 forging materialor where this is not possible due to less stringent impact testing criteria prior 9 to 1973a measured value of the absorbed Charpy impact energy, which is considered in the 10 TR to be a lower bound on actual USE for that forging. For these cases, the USE is identified in 11 the TR as being greater than or equal to () the reported absorbed impact energy.

12 13 Only measured USE values for Charpy tests on forging specimens oriented in the weak 14 direction (transverse specimens) are used to calculate the recommended Mean - 2 values for 15 the two forging supplier data sets. For cases where measured USE values are available only in 16 the strong direction (longitudinal specimens), or where the strong direction must be assumed 17 because the Charpy V-Notch specimen orientation was not reported in the CMTR, the TR 18 estimates USE in the weak direction using Position 1.2 of BTP 5-3; specifically, the weak 19 direction USE is estimated to be 65 percent of the measured strong direction USE. These 20 reduced values of measured strong direction USE are reported in the data sets as the BTP 5-3 21 USE values. Separate Mean - 2 calculations are also reported for both the measured 22 BTP 5-3 USE values and the full complement of measured USE values and estimated lower 23 bound USE values to contextualize and justify the recommended generic values. However, it 24 must be emphasized that the TRs recommended generic USE values are set equal to the 25 calculated Mean - 2 values only for Charpy USE data that is measured in the weak direction.

26 27 The TR states that all measured USE values for Charpy tests in both the strong and weak 28 directions are determined as per the following criteria:

29 30

• For a given forging Charpy test record, the TR attempts to determine a value for USE 31 based on available percent shear measurements in a manner that is consistent with 32 ASTM E185-82. Specifically, if the Charpy test data for the forging material contains 33 at least one impact energy measurement with greater than 95 percent shear 34 (i.e., > 95 percent shear), but some of the impact energy measurements report no 35 percent shear values, all impact energies approximately greater than or equal to those 36 that are known to exhibit greater than 95 percent shear are assumed to have 37 > 95 percent shear. All non-outlier impact energy measurements with > 95 percent 38 shear are averaged to determine the measured USE, which is reported in the TR for the 39 forging test record.

40 41

• TR states that if the Charpy test record contains limited or no percent shear data, 42 however the upper-shelf region of the Charpy curve can be clearly determined from the 43 data provided, the USE is identified by an approximately constant energy versus 44 temperature region. As an example, the TR identifies cases where data points at four 45 temperatures over a 50 °F range exhibited impact energy values within a scatter of 10 °F 46 or less. TR states that the existence of the upper-shelf region is confirmed by plotting 47 the impact energy data and identifying if the plot levels off at higher temperatures. The 48 reported USE represents an average of all Charpy energy values considered to be in the 49 upper-shelf region.

50

1 In addition to measurements of actual USE, the TR also reports measured values of the 2 absorbed Charpy impact energy in the strong direction for which actual USE is undetermined.

3 For these cases, the reported Charpy impact energy is considered to represent a lower bound 4 on actual USE for the forging, and the unknown USE is therefore reported to be greater than or 5 equal to the reported Charpy impact energy. Further, since these impact energies are all 6 measured in the strong direction, the TR provides BTP 5-3 estimates of absorbed impact 7 energies in the weak direction, which are 65 percent of the measured impact energies in the 8 strong direction. These cases are all designated in the TR as USE XX, where the value of 9 XX is a number that equals the reported Charpy impact energy, and they are also reported as 10 BTP 5-3 values. The TRs criteria for determining that USE XX, where XX is a number 11 that equals the reported impact energy are as follows:

12 13

• The TR states that if the test record reports percent shear values, but all data indicates a 14 percent shear less than 95 percent, the USE is reported to be greater than or equal to 15 the maximum Charpy impact energy. The reported impact energy is not incorporated 16 into the calculation of the recommended Mean - 2 value since this recommended 17 generic value is based exclusively on the measured weak direction USE values.

18 19

• The TR states that if the test record included limited shear data or did not include shear 20 data, and Charpy impact energies are increasing throughout the temperature range 21 available, it is unknown if the upper-shelf has been reached. The TR states that the 22 USE is reported to be greater or equal to the highest Charpy impact energy value 23 available; or if the highest data point is determined to be a potential 'outlier' or a 24 non-representative data point, the USE is reported as greater than or equal to a value 25 less than the highest energy value based on the average of the comparable preceding 26 data points. In these instances, the reported impact energy is not incorporated into the 27 calculation of the generic USE.

28 29 In addition to data sets for SA508, Class 2 forgings supplied by Rheinstahl Huttenwerke AG and 30 Fried-Krupp Huttenwerke AG, the TR also reports Charpy impact energies and USE data for 31 several other firms who supplied SA508, Class 2 forgings to Rotterdam; these include 32 Klckner-Werke AG, Terni, Marrl-Freres, and an unknown supplier. The Charpy test data from 33 these other suppliers is independently evaluated in Section 4.3 of the TR, but due to more 34 limited data sets for each of the other suppliers, a statistical evaluation to determine generic 35 USE values for the other suppliers is not performed. Instead the TR shows that all measured 36 USE values and Charpy impact energies (where USE is reported as greater than or equal to the 37 reported Charpy impact energy, as per the above), are greater than the recommended 38 Mean - 2 values from the Rheinstahl Huttenwerke AG and Fried-Krupp Huttenwerke AG data 39 sets. Based on this comparison, this section of the TR concludes that an SA508, Class 2 40 forging from an unknown supplier in a Rotterdam-fabricated RV would be expected to have an 41 unirradiated Charpy USE value of at least 52 ft-lbs. Therefore, the TR recommends 52 ft-lbs as 42 the generic unirradiated Charpy USE value to be used for SA508, Class 2 forgings in Rotterdam 43 RVs if the forging supplier is unknown.

44 45 Generic Charpy USE, Copper Content, and Nickel Content for Rotterdam RV Welds 46 47 The TR states that the Rotterdam CMTRs identify two types of welds used in the fabrication of 48 the RVs: SMAWs and SAWs. Each type is evaluated separately. The TR states that the 49 industry practice at the time of Rotterdam RV fabrication was to perform Charpy tests at a 50 limited number of temperatures to show 30 ft-lbs or more of absorbed energy at 10 °F. These

1 tests were considered sufficient to satisfy ASME Code requirements at that time; however, the 2 CMTRs often contain insufficient Charpy impact data to determine measured values of USE.

3 The TR recommends a generic unirradiated Charpy USE value of 75 ft-lbs for Rotterdam RV 4 SAWs; this is the Mean - 2 value for the set of seven measured unirradiated USE values from 5 RV material surveillance programs. The TR states that the Rotterdam RV SAW data set 6 represents every SAW material vendor and every flux type, with the exception of Linde 80 flux 7 type. The TR emphasizes that the SAW welds of the Linde 80 flux type are specifically 8 excluded from the Rotterdam weld analyses. The TR identifies that outside of the baseline USE 9 measurements from the RV material surveillance programs, there is no meaningful USE 10 information available in the CMTRs for Rotterdam SAWs. Therefore, only the seven measured 11 USE values for SAWs from RV surveillance programs are reported in the TR.

12 13 The TR states that out of 38 SMAW Charpy test records, actual measured USE values are 14 available for only three heats of SMAW material. The three measured USE values are 116, 15 130, and 134 ft-lbs. The TR does not determine a Mean - 2 value for these three due to the 16 statistically insignificant size of the data set. For the remaining 35 Charpy test records, the TR 17 determined a lower bound on the USE for each SMAW based on the available Charpy impact 18 energy data using methods similar to those described above for SA508, Class 2 forgings. If no 19 percent shear data is available, the USE is reported as greater than or equal to the average of 20 the Charpy impact energies at the test temperature, typically around 10 °F. When percent 21 shear values are reported, and each is less than 95 percent, then the TR reports maximum 22 Charpy impact energy for the weld test. Based on these methods, the TR determined that its 23 recommended generic unirradiated Charpy USE value for Rotterdam SMAWs is 72 ft-lbs, which 24 is based on the non-outlier weld heat showing the lowest of the lower bound USE values.

25 26 In addition to the generic USE, the TR determines generic Cu and Ni weight percentages for 27 both SAWs and SMAWs based on the calculation of Mean + 1 for the data sets. The TR 28 identifies this method as consistent with RG 1.99, Rev. 2, which states that conservative 29 estimates of Cu and Ni content based on generic data may be used if justification is provided; 30 the TR notes that for Cu and Ni content, the RG identifies conservative estimates as mean 31 plus one standard deviation. The TR further states that if a common weld metal heat and flux 32 type combination is shared between multiple welds, the average chemistry value for the heat is 33 considered as one data point when determining the generic weld chemistry values so as not to 34 assign undue weight to multiple samples of weld material of the same heat. The chemistry data 35 used in the evaluation consists of measurements from RV surveillance programs, supplemented 36 with all available chemistry data for heats outside the surveillance programs. The TR states that 37 the data is limited to deposited weld chemistry results, unless otherwise noted. The TR notes 38 that the Cu content for one SAW material (Smit-Weld Heat No. 25006) is based on the weld 39 wire analysis since deposited weld chemistry is not available.

40 41 4.0 STAFF EVALUATION 42 43 The NRC staffs review of PWROG-17090-NP, Rev. 0 addressed whether the TR evaluations 44 for determining the generic properties for Rotterdam RVs are acceptable as a basis for 45 implementation of these properties in plant licensing applications for demonstrating compliance 46 with RV fracture toughness requirements in 10 CFR Part 50, Appendix G; 10 CFR 50.61 or 47 50.61a; and TLAAs related to RV fracture toughness per 10 CFR 54.21(c)(1). The staff applied 48 the regulatory guidance regarding the use of generic data, as set forth in Section 2.0 of this SE, 49 to determine whether the TR evaluations provide reasonably conservative generic estimates of 50 these properties for use in plant licensing applications that address these requirements.

51

1 4.1 Generic Unirradiated Charpy USE for SA508, Class 2 Forgings in Rotterdam RVs 2

3 The TR determines generic unirradiated Charpy USE values for SA508, Class 2 RV forgings 4 based on calculating the Mean - 2 value for the set of measurements of Charpy USE in the 5 weak direction, consistent with the criteria addressed in Section 2.0 of this SE. The staff 6 identified that the TRs evaluation of generic data sets for determining generic USE values for 7 the ASME SA508, Class 2 forging specification is consistent with the definition of the material 8 class provided in 10 CFR 50.61, 10 CFR 50.61a, and RG 1.99, Rev. 2.

9 10 The TR identifies that several firms manufactured and supplied SA508, Class 2 RV forging 11 components to Rotterdam; the TR indicates that Rotterdam procured the forgings to fabricate 12 the welded RVs in the late 1960s and early 1970s timeframe. The Charpy test data sets used 13 to establish the TRs generic USE values are based on the forging suppliers. Note 4 of 10 CFR 14 Sections 50.61 and 50.61a states: Data from reactor vessels fabricated to the same material 15 specification in the same shop as the vessel in question and in the same time period is an 16 example of generic data. The staff determined that for multiple suppliers of SA508, Class 2 17 forgings to the RV fabricator (Rotterdam), the same shop is appropriately considered in the TR 18 to be the same firm responsible for manufacturing the forging component. Therefore, the staff 19 determined that these generic data sets and corresponding generic USE values are 20 appropriately delineated for plant-specific application in a manner that is consistent with Note 4 21 of 10 CFR 50.61 and 50.61a.

22 23 Although this Note 4 is specifically cited for the use of generic Cu and Ni content data to 24 demonstrate compliance with applicable PTS requirements, the NRC staff finds that the criteria 25 in Note 4 are also relevant to the TRs application of generic data for determining generic values 26 of unirradiated Charpy USE. The NRC staff also finds there are no criteria in 10 CFR Part 50, 27 Appendix G or NRC guidance related to Charpy USE that would prohibit or otherwise preclude 28 the application of Note 4 to the determination of generic unirradiated Charpy USE based on 29 forging manufacturer. Therefore, the staff finds that the TRs classification of the generic data 30 sets for SA508, Class 2 forgings based on the forging manufacturer, consistent with Note 4 of 31 the PTS requirements, is acceptable.

32 33 The RV forging suppliers and the number of Charpy test records for each forging supplier are 34 listed below:

35 Forging Supplier Number of Charpy Test Records For Forging Components Supplied to Rotterdam Rheinstahl Huttenwerke AG 38 Fried-Krupp Huttenwerke AG 67 Klckner-Werke AG 8 Terni 6 Marrl-Freres 2 Unknown 1

1 The staff reviewed the TR methods for evaluating Charpy impact test data to determine either a 2 measured value of the USE for the forgingor where a measured USE value could not be 3 determinedthe methods for determining a lower bound on the USE for the forging based on 4 available Charpy test data. The staffs review of these methods is documented below.

5 6 For cases where Charpy impact energy data is accompanied by at least 1 percent shear 7 measurement greater than 95 percent shear, the staff reviewed the TR methods for determining 8 USE based on the definitions in ASTM E185-82. Specifically, upper-shelf energy is defined as 9 the region in the Charpy transition curve where the broken specimens exhibit greater than 10 95 percent shear; and upper-shelf energy level is defined as the average of absorbed impact 11 energy values for Charpy specimens whose test temperature is above the upper end of the 12 transition region, which is below the USE region. This definition also states that for specimens 13 tested in sets of three at each temperature, the set having the highest average impact energy 14 may be regarded as defining the USE. The staff determined that the TRs statement that the 15 reported USE is the average of all non-outlier impact energy values greater than or equal to 16 the value(s) with greater than 95 percent shear is sufficiently consistent with this definition, 17 provided that the staff could confirm, based on review of examples, that the elimination of the 18 outlier data point is reasonable. Therefore, in RAI correspondence, the staff requested that the 19 PWROG provide examples of both high and low outliers that were eliminated from this 20 calculation of the average.

21 22 In its June 12, 2019, RAI-3 response (Ref. 2), the PWROG provided an example of an 23 uncharacteristically low impact energy and an example of an uncharacteristically high impact 24 energy, both of which are considered to be outliers and eliminated from the calculation of the 25 average, which is the USE reported in TR. For both the high and low outlier impact energies, 26 the RAI response identified all the other impact energies that went into the calculation of the 27 average, as well as the test temperatures and the available percent shear measurements. The 28 PWROG compared the outlier impact energies with the non-outlier data that was used to 29 determine applicable USE values for these forging components, as reported in the TR. As 30 described below, based on its review of the high and low outlier impact energies, and its review 31 of PWROGs comparison of the outliers to the other data that was used to determine USE, the 32 NRC staff was able to verify that appropriate engineering judgement was used in the elimination 33 of the high and low outlier impact energies to determine the reported USE for these 34 components. Therefore, the staff finds that this method for determining measured USE is 35 acceptable.

36 37 If a Charpy test record includes no percent shear data for identifying USE, the staff reviewed the 38 TRs reported method of determining USE by identifying an approximately constant energy 39 versus temperature region in the Charpy data. The staff noted that the TRs example of four 40 data points over a 50 °F temperature range exhibiting impact energy values within a scatter of 41 10 °F or less is appropriate for identifying the upper-shelf region because at temperatures above 42 the transition region, impact energy values become approximately constant at or near the USE 43 level. The staff noted that ASTM E185-82 defines the upper-shelf energy level as the average 44 energy value for Charpy specimens whose test temperature is above the upper end of the 45 transition region. Therefore, the staff confirmed that the USE value can be determined as the 46 average of impact energies that are determined to be in the upper-shelf region based on low 47 scatter over a large temperature range. The staff finds that this method for determining 48 measured USE is acceptable.

49

1 When measured USE for the forging test record cannot be determined based on the above 2 methods, the staff reviewed the two methods described in the TR for establishing a lower bound 3 on the USE based on the available absorbed impact energy data.

4 5

• Based on the definitions in ASTM E185-82, the staff noted that if the test record includes 6 percent shear values, and all are less than 95 percent, then the corresponding impact 7 energies are not in the upper-shelf region; therefore, the staff identified that USE would 8 be greater than or equal to the maximum absorbed impact energy with percent shear 9 less than 95 percent. Based on the definitions in the ASTM standard, the staff finds this 10 method for determining a lower bound on USE for the available test data to be 11 acceptable.

12 13

• If shear data are not available, and it is seen, based on examination of impact energy 14 data, that energies are increasing throughout the temperature range available, the staff 15 confirmed that it would be unknown whether the USE region has been reached. The 16 staff noted that for RV materials that were fabricated to earlier ASME Code editions, 17 Charpy impact testing may not have occurred at temperatures above the transition 18 region. Thus, it is reasonable for the TR to determine that USE for the material would be 19 greater than or equal to absorbed impact energy in the transition region. The staffs 20 review of specific cases for this situation is documented below based on its audit of 21 Charpy test data for one of the forging suppliers.

22 23 The TR performed independent evaluations of Charpy test data for SA508, Class 2 forgings 24 supplied by Rheinstahl Huttenwerke AG and Fried-Krupp Huttenwerke AG. The TR performed 25 a third evaluation that addressed the other forging suppliers, which collectively includes 26 Klckner-Werke AG, Terni, Marrl-Freres, and the unknown supplier. The staffs review 27 addressed the three TR evaluations for establishing the recommended generic USE values.

28 29 Unirradiated USE Evaluation of RV Forgings Supplied by Rheinstahl Huttenwerke 30 31 The NRC staff confirmed that of the 38 Charpy test records that are available for Rheinstahl 32 Huttenwerke AG forgings, there are 11 test records where measured values for unirradiated 33 Charpy USE were able to be determined based on the methodologies described above. The 34 staff noted that 8 of the 11 forgings have measured USE values for both the strong and the 35 weak directions; one forging has a measured USE value only in the strong direction; and two 36 forgings have measured USE values only in the weak direction. For the 9 measured USE 37 values in the strong direction, the staff confirmed that the TR correctly used Position 1.2 of 38 BTP 5-3 to estimate the USE values in the weak direction; specifically, BTP 5-3 estimates of 39 weak direction USE are equal to 65 percent of the measured USE values in the strong direction.

40 41 For the 11 forgings with measured values for Charpy USE, the staffs independent calculations 42 showed the following.

43 44

• For the 10 measured USE values in the weak direction, the staff confirmed that the 45 Mean - 2 value for this data set is 56 ft-lbs, which is the recommended generic USE 46 value for forgings supplied by Rheinstahl Huttenwerke AG. The staff noted that this 47 bounds (i.e., is more conservative than) the lowest measured weak direction USE value 48 of 64 ft-lbs.

49

1

• For the 9 BTP 5-3 estimates of USE in the weak direction, the staff confirmed that the 2 Mean - 2 value is 70 ft-lbs, which bounds the lowest BTP 5-3 USE value of 75 ft-lbs.

3 The staff noted that the TRs recommended generic USE value of 56 ft-lbs bounds these 4 BTP 5-3 USE estimates. Therefore, based on review of all available USE data for 5 Rheinstahl Huttenwerke AG forgings, the staff verified that the recommended generic 6 USE value of 56 ft-lbs is the most conservative.

7 8 BTP 5-3 Estimates of Lower Bound USE for Rheinstahl Huttenwerke Forgings 9

10 The NRC staff also confirmed that out of the 38 Charpy test records for the Rheinstahl 11 Huttenwerke AG data set, there are 27 for which measured USE could not be established, but 12 the available impact energy data were used to determine a lower bound on USE in the strong 13 direction using the methods summarized above. Since all of the absorbed impact energies 14 were measured in the strong direction, the TR determined estimates of absorbed impact energy 15 in the weak direction to be 65 percent of the measured impact energies in the strong direction 16 by applying BTP 5-3. The staff noted that 23 of these 27 estimates of lower bound USE are 17 considered along with measured weak direction USE values and BTP 5-3 USE values in a 18 separate Mean - 2 calculation, which is 40 ft-lbs. The TR does not recommend 40 ft-lbs as 19 the generic USE value for Rheinstahl Huttenwerke AG forgings because the 23 BTP 5-3 20 estimates of lower bound USE included in this Mean - 2 calculation do not represent actual 21 USE for that test record. Specifically, Note b of the data set states that it is unknown whether 22 the upper-shelf was reached during the test since the Charpy impact energies are increasing 23 throughout the temperature range available, and the actual USE value is likely higher.

24 25 The staff noted that the four lowest of the 23 lower bound USE estimates that were included in 26 the Mean - 2 calculation of 40 ft-lbs are lower than the TRs recommended generic USE value 27 of 56 ft-lbs. These values, which are annotated with Note b in the Rheinstahl Huttenwerke AG 28 data set are 53 ft-lbs, 52 ft-lbs, and two values at 47 ft-lbs. The staff also noted that the 29 calculation of 40 ft-lbs excludes the four lowest of the 27 available lower bound USE estimates, 30 which are 44 ft-lbs, 42 ft-lbs, and two values at 39 ft-lbs. Note f of the Rheinstahl Huttenwerke 31 AG data set explains that the four lowest values of 44 ft-lbs, 42 ft-lbs, and two values at 39 ft-lbs 32 are excluded from statistical analysis because the values do not provide accurate 33 representation of USE, and the actual USE is likely much higher since a Charpy test with a 34 similar absorbed impact energy has a shear value much less than 95 percent.

35 36 The NRC staff identified that the Charpy test data used to determine the eight lowest estimates 37 of lower bound USE needed to be reviewed to assess whether the reported impact energies are 38 below the upper-shelf region. Therefore, as part of its TR review, the staff audited Charpy test 39 records for the following Rheinstahl Huttenwerke AG forgings, which had the eight lowest 40 reported absorbed impact energies:

1 Four Lowest Impact Energies, Rheinstahl Huttenwerke AG Forgings with Note b (Included in Mean - 2 of 40 ft-lbs; 40 ft-lbs is not the recommended USE)

Component Measured Absorbed Impact BTP 5-3 Estimate of Absorbed Impact Identification Energy, Strong Direction Energy, Weak Direction (i.e., Lower Bound USE Estimates)

Plant D, 82 ft-lbs 82 ft-lbs X 65% = 53 ft-lbs Intermediate Shell Plant E, 80 ft-lbs 80 ft-lbs X 65% = 52 ft-lbs Inlet Nozzle 11 Plant D, 72 ft-lbs 72 ft-lbs X 65% = 47 ft-lbs Inlet Nozzle 09 Plant F, 72 ft-lbs 72 ft-lbs X 65% = 47 ft-lbs Inlet Nozzle 09 2

3 4

Four Lowest Impact Energies, Rheinstahl Huttenwerke AG Forgings with Note f (Excluded from All Statistical Evaluations)

Component Measured Absorbed Impact BTP 5-3 Estimate of Absorbed Impact Identification Energy, Strong Direction Energy, Weak Direction (i.e., Lower Bound USE Estimates)

Plant E, 68 ft-lbs 68 ft-lbs X 65% = 44 ft-lbs Upper Shell Plant F, 64 ft-lbs 64 ft-lbs X 65% = 42 ft-lbs Outlet Nozzle 13 Plant D, 60 ft-lbs 60 ft-lbs X 65% = 39 ft-lbs Inlet Nozzle 11 Plant D, 60 ft-lbs 60 ft-lbs X 65% = 39 ft-lbs Outlet Nozzle 14 5

6 The staffs audit of the Charpy test records for the eight forging materials included review of test 7 temperatures, absorbed impact energies, and available percent shear measurements. The 8 staffs review identified that the maximum absorbed impact energies in the eight Charpy test 9 records correspond to the measured impact energy values for the strong direction that are 10 reported in the TR. The staff's audit generally confirmed TR statements that the eight lowest 11 impact energies are not representative of the USE for those forgings since the impact energies 12 were increasing throughout the temperature range shown in the test record. Based on the 13 increasing energy trends, available test temperatures, and the limited amount of shear data, the 14 staff found that there is sufficient evidence that the materials were likely in the transition region 15 at the highest test temperatures documented in the records. Thus, the staff confirmed that the 16 actual USE values for these forgings, while unknown (because it was likely not reached during 17 the test evolution), can reasonably be expected to be higher than the measured values of 18 absorbed impact energy for the strong direction, as reported in the TR for these eight forgings.

19 Based on its audit of the test records, the staff found that the eight lowest impact energies do 20 not need to be considered in the statistical evaluation of the measured USE values for 21 determining the TRs recommended generic USE value.

1 The staff also noted that the other 19 BTP 5-3 estimates of lower bound USE are all greater 2 than the TRs recommended generic value of 56 ft-lbs; this provides additional evidence that the 3 Mean - 2 value for the ten measurements of Charpy USE in the weak direction is a 4 conservative generic estimate of unirradiated Charpy USE for this forging supplier. Therefore, 5 the staff confirmed that 40 ft-lbs does not warrant implementation as a generic estimate of USE 6 for Rheinstahl Huttenwerke AG forgings. Accordingly, the staff finds that 56 ft-lbs is acceptable 7 for implementation as a generic unirradiated Charpy USE value for SA508, Class 2 RV forgings 8 supplied by Rheinstahl Huttenwerke AG for Rotterdam RVs.

9 10 Unirradiated USE Evaluation of RV Forgings Supplied by Fried-Krupp Huttenwerke AG 11 12 The NRC staff confirmed that of the 67 Charpy test records that are available for Fried-Krupp 13 Huttenwerke AG forgings, there are 38 test records where measured values for unirradiated 14 Charpy USE were able to be determined based on the methodologies described above. For the 15 38 forgings with measured USE values, the staff noted that 5 of the 38 forgings have measured 16 USE values for both the strong and the weak directions, and the other 33 have measured USE 17 values only in the strong direction. For all 38 measured USE values in the strong direction, the 18 staff confirmed that the TR correctly used Position 1.2 of BTP 5-3 to estimate USE values in the 19 weak direction. For the 38 forgings with measured values for Charpy USE, the staffs 20 independent calculations showed the following.

21 22

• For the 5 measured USE values in the weak direction, the staff confirmed that the 23 Mean - 2 value for this data set is 52 ft-lbs, which is the recommended generic USE 24 value for forgings supplied by Fried-Krupp Huttenwerke AG. The staff noted that this 25 bounds (i.e., is more conservative than) the lowest measured weak direction USE value 26 of 62 ft-lbs.

27 28

• For the 38 BTP 5-3 estimates of USE in the weak direction, the staff confirmed that the 29 Mean - 2 value is 61 ft-lbs, which is equal to the lowest of the 38 BTP 5-3 USE 30 estimates. The staff noted that the TRs recommended generic USE value of 52 ft-lbs 31 bounds this BTP 5-3 USE estimate. Therefore, based on review of all available USE 32 data for Fried-Krupp Huttenwerke AG forgings, the staff verified that the recommended 33 generic USE value of 52 ft-lbs is the most conservative.

34 35 BTP 5-3 Estimates of Lower Bound USE for Fried-Krupp Huttenwerke AG Forgings 36 37 The NRC staff also confirmed that out of the 67 Charpy test records for the Fried-Krupp 38 Huttenwerke AG data set, there are 29 for which measured USE could not be determined, but 39 BTP 5-3 estimates of the lower bound on USE could be determined, as per the criteria above.

40 All of the 29 estimates of lower bound USE are considered along with the measured USE 41 values in a separate evaluation that calculates a Mean - 2 value of 51 ft-lbs. The TR does not 42 recommend 51 ft-lbs as the generic USE value for these forgings because the 29 BTP 5-3 43 estimates of lower bound USE included in this Mean - 2 calculation do not represent actual 44 USE for that test record. The 29 BTP 5-3 estimates of lower bound USE are annotated with 45 either Note b or Note f. Note b states that it is unknown whether the upper-shelf was 46 reached during the test since the Charpy impact energies are increasing throughout the 47 temperature range available, and the actual USE value is likely higher. Note f states that 48 reported shear values are less than 95 percent shear, and the reported [impact energy] value is 49 less than or equal to the maximum energy value of a Charpy specimen with less than 50 95 percent shear, and as a result, the USE is higher than the Charpy data reported.

1 The staff identified that there are no lower bound USE estimates that are excluded from the 2 Mean - 2 calculation of 51 ft-lbs. Further, just two of the 29 lower bound USE estimates 3 (51 ft-lbs and 50 ft-lbs) are lower than the TRs recommended generic USE value of 52 ft-lbs.

4 The staff determined that actual USE values for these two forgings, while unknown, would likely 5 be higher than their reported lower bound values given that these impact energies are 6 annotated with Note b identifying that impact energies are increasing throughout the 7 temperature range available. The staff noted that all of the other 27 lower bound USE estimates 8 are greater than the TRs recommended generic USE value of 52 ft-lbs for this forging supplier.

9 Therefore, the staff finds that 52 ft-lbs, based on the Mean - 2 for the 5 measured USE values 10 in the weak direction, is acceptable for implementation as a generic unirradiated Charpy USE 11 value for SA508, Class 2 RV forgings supplied by Fried-Krupp Huttenwerke AG for Rotterdam 12 RVs.

13 14 Unirradiated USE Evaluation of RV Forgings from Other Suppliers 15 16 The staff noted that there are 17 Charpy test records represented in a single data set provided 17 in TR Table 7 for the other forging suppliers, which includes Klckner-Werke AG, Terni, Marrl-18 Freres, and an unknown company. In RAI correspondence, the NRC requested that the 19 PWROG resolve the apparent inconsistency in the TR regarding the number of Charpy test 20 records from an unknown company because in Table 2 the TR identifies that there are two 21 forging components from an unknown supplier, whereas Table 7 of the TR lists one impact 22 energy measurement for the unknown supplier. In its June 12, 2019, RAI-2 response (Ref. 2),

23 the PWROG indicated that there are two SA508, Class 2 forging materials with an unknown 24 supplier, as shown in TR Table 2 - however, only one such material is listed in TR Table 7 25 because the Charpy test record is not available for the other unknown supplier material.

26 Accordingly, Charpy data for the other SA508, Class 2 forging component from an unknown 27 supplier could not be included Table 7.

28 29 Out of these 17 test records, the staff noted there are two with measured USE values in the 30 weak direction; these USE values are 134 ft-lbs and 141 ft-lbs. There are 10 measured USE 31 values in the strong direction, for which the TR applied BTP 5-3 to estimate weak direction USE; 32 the staff identified that the lowest of the BTP 5-3 USE estimates is 94 ft lbs. The USE is 33 unknown for 7 test records, but a lower bound on USE in the strong direction was established 34 based on evaluation of available absorbed impact energy data that is established to be in the 35 transition region (i.e., the energies are increasing throughout the temperature range available);

36 this is same method as that used for the Rheinstahl Huttenwerke AG and Fried-Krupp 37 Huttenwerke AG data sets, as identified in Note b for these data sets. As with the other 38 SA508, Class 2 data sets, the lower bound USE values are reduced per BTP 5-3 to determine 39 estimates of lower bound USE in the weak direction. The lowest of these BTP 5-3 estimates of 40 lower bound USE for the other forging suppliers is 75 ft-lbs.

41 42 Considering the smaller number of records covering the three known suppliers and the unknown 43 supplier, the staff confirmed that it is reasonable for the TR to select a lower generic USE value 44 that could be used for SA508, Class 2 forgings in Rotterdam RVs if the forging manufacturer is 45 unknown. For this purpose, the TR recommended that a generic USE value of 52 ft-lbs be used 46 for SA508, Class 2 forgings in Rotterdam RVs if the forging manufacturer is unknown. In its 47 June 12, 2019, RAI-7 response (Ref. 2) the PWROG clarified that the generic value of 52 ft-lbs 48 is not intended for the suppliers Klckner-Werke AG, Terni, or Marrl-Freres because data for 49 all known and applicable Rotterdam RV forgings from these suppliers are provided in Table 7 of 50 the TR for this data set. The PWROG stated that there is sufficient data in Table 7 for these 51 components to justify a component-specific USE value that is higher than 52 ft-lbs. The staff

1 reviewed this RAI response and confirmed that the applicable component-specific USE value 2 (or lower bound USE estimate, as applicable) should be used for plants that can identify their 3 forgings from among the components listed in Table 7.

4 5 With respect to a generic USE value of 52 ft-lbs for an unknown forging supplier in a Rotterdam 6 RV, the NRC staff considered all Charpy test data for all SA508, Class 2 forging manufacturers 7 evaluated in the TR and noted the following:

8 9

• There are 122 Charpy impact test records evaluated in the TR.

10

• 52 ft-lbs is the most bounding of the two Mean - 2 values for the two largest suppliers, 11 Rheinstahl Huttenwerke AG and Fried-Krupp Huttenwerke AG.

12

• 52 ft-lbs bounds all available Charpy impact test data (measured weak direction USE 13 data, BTP 5-3 USE data, and lower bound USE estimates) from the three other known 14 suppliers and the unknown supplier.

15

• For all forging suppliers, 52 ft-lbs bounds all measured USE values in the weak direction 16 and all BTP 5-3 USE estimates for the weak direction (i.e., USE estimates based on 17 application of BTP 5-3 to USE measurements in the strong direction).

18

• 52 ft-lbs bounds 21 of the 27 BTP 5-3 estimates of lower bound USE in the Rheinstahl 19 Huttenwerke AG data set and 27 of the 29 lower bound USE estimates in the Fried-20 Krupp Huttenwerke AG data set.

21

• For the those several BTP 5-3 estimates of lower bound USE in the two largest data sets 22 that are less than 52 ft-lbs, the staff determined based on review TR methods and audit 23 of Charpy test records that actual USE values for these forgings, while unknown, would 24 likely be higher than their reported lower bound values given that these impact energies 25 are increasing throughout the temperature range available.

26 27 Therefore, based on its review of all the SA508, Class 2 Charpy test data, the staff determined 28 that for a Rotterdam RV with SA508, Class 2 forging(s) from an unknown supplier, there is 29 reasonable assurance the USE value for that forging would be at least 52 ft-lbs. Accordingly, 30 the staff finds that 52 ft-lbs is acceptable for implementation as a generic unirradiated Charpy 31 USE value for SA508, Class 2 RV forgings from an unknown supplier in Rotterdam RVs.

32 33 4.2 Generic Charpy USE, Cu Content, and Ni Content for Rotterdam RV Submerged Arc 34 Welds and Shielded Metal Arc Welds 35 36 The TR determined generic values of unirradiated Charpy USE, Cu content, and Ni content for 37 Rotterdam SAWs and SMAWs. Data sets for Charpy USE, Cu content, and Ni content were 38 separately evaluated in the TR to determine the recommended generic properties for these two 39 weld types. The NRC staffs review of these data sets and data analyses follows below.

40 41 Rotterdam RV Submerged Arc Welds 42 43 The NRC staff reviewed Charpy USE values and chemistry data for Rotterdam SAWs, which 44 are provided in TR Tables 9 and 10, respectively. These tables also identify the flux types and 45 weld wire vendors. The TR indicates that the two flux types (SAF89 and LW320) and six 46 weld wire vendors identified in these tables are generically applicable to SAW materials for 47 Rotterdam RVs, except for welds with Linde 80 flux type. SAWs with Linde 80 flux type are 48 excluded from these generic analyses since welds with Linde 80 flux have been generically 49 analyzed previously. The staff found that the TR adequately defined the material class for

1 Rotterdam SAWs based on its identification of the two flux types and six wire vendors used to 2 fabricate these welds.

3 4 The TR determines the generic unirradiated Charpy USE value for Rotterdam SAWs by 5 calculating the Mean - 2 for the set of seven measured values of unirradiated Charpy USE for 6 the Non-Linde 80 SAWs included in Rotterdam RV surveillance programs. The seven USE 7 values are listed in TR Table 9. Note a of Table 9 identifies that these USE values are 8 determined as the average of all available absorbed energy values with percent shear greater 9 than or equal 95 percent, as per the ASTM E185-82 method that was used to determine USE 10 for the RV forgings. As with the forgings, the staff determined that the TRs application of the 11 ASTM E185-82 definitions for determining the seven USE values is acceptable.

12 13 The staff noted that for six of the seven Charpy USE values in Table 9, the reported USE value 14 corresponds to a SAW for a specific unnamed plant (e.g., Plant B). One USE value in Table 9 15 corresponds to SAWs at four plants (Plants A, G, H, and I). In its June 12, 2019, RAI-5 16 response (Ref. 2), the PWROG identified that each of the seven unirradiated Charpy USE 17 values in Table 9 represents a unique heat of weld material.

18 19 The staff confirmed that the Mean - 2 value for the set of seven baseline Charpy USE values 20 for SAWs in Rotterdam RV surveillance programs is 75 ft-lbs. The staff noted that this 21 recommended generic USE value bounds (i.e., is more conservative than) the lowest of the 22 seven measured USE values, which is 82 ft-lbs. Considering that surveillance welds were 23 selected from the amongst the core region welds (i.e., welds located in the original 40-year 24 beltline region), the staff noted that the Mean - 2 value of 75 ft-lbs may be reasonable for 25 generic application to the non-Linde 80 core region SAWs in Rotterdam RVs. Additional 26 considerations regarding the applicability of this data to all Rotterdam SAWs are discussed 27 below.

28 29 The TR identifies that outside of the baseline USE measurements from RV surveillance 30 programs, no USE information is available for Rotterdam SAWs. The TR states that the 31 industry practice at the time of Rotterdam RV fabrication was to test Charpy specimens at 10 °F 32 or lower to show 30 ft-lbs or more of absorbed energy, and the reporting of this test information 33 in the Rotterdam CMTRs was considered sufficient to satisfy the fracture toughness 34 requirements of the ASME Code at that time. The TR also states that the core region welds had 35 the same specification requirements as the other RV welds, but for core region welds, 36 Rotterdam was required to aim for both a Charpy V-Notch Transition Temperature (TTCV) and 37 a Nil-Ductility Transition Temperature (NDTT) based on drop weight testing less than 10 °F, and 38 to furnish additional test results relevant to TTCV and NDTT. The TR also states the TTCV 39 and NDTT do not occur near the upper-shelf region, and thus, the surveillance capsule program 40 test results are generically representative of the SAWs produced at Rotterdam for USE 41 calculations. Considering this information, the staff surmised that CMTRs for non-surveillance 42 welds would show low test temperatures and correspondingly low impact energies, which could 43 not be used to further support the generic USE value of 75 ft-lbs based on the surveillance weld 44 USE data in Table 9.

45 46 The NRC staff identified that additional information was needed to confirm that the seven USE 47 values for Rotterdam SAWs in Table 9 are representative for Rotterdam SAWs in general. In its 48 RAI, the staff requested that the PWROG address how specification requirements and test 49 criteria for Rotterdam SAWs support the TR determination that the baseline USE data for 50 surveillance program SAWs in Table 9 is representative of Rotterdam SAWs in general.

1 In its June 12, 2019, RAI-6 response (Ref. 2), the PWROG stated that based on acceptance 2 criteria for Rotterdam RV welds, the flux/wire welds (SAWs) did not have chemistry 3 requirements, but all welds had the same mechanical requirements, which included a minimum 4 tensile strength of 80 ksi, a minimum absorbed Charpy impact energy of 30 ft-lbs for the 5 average of three specimens, and a minimum absorbed Charpy impact energy 25 ft-lb for one 6 individual specimen. The PWROG emphasized that these requirements were identical for the 7 beltline and non-beltline SAWs; however, the core region welds had additional requirements to 8 establish both the TTCV based on Charpy testing and NDTT based on drop weight testing, and 9 to aim for a transition temperature of 10° F. The PWROG emphasized that these additional 10 testing requirements for core region welds would not affect the USE, since these properties are 11 associated with ductile-to-brittle transition, not the ductile region. Since the requirements for 12 tensile strength and absorbed Charpy impact energy were equivalent, the PWROG stated that it 13 is expected that all Rotterdam RV SAWs were taken from the same set of available weld 14 metals. The PWROG identified that this is further supported by the known instances where a 15 core region weld and a non-core region weld share the same heat number. Therefore, the 16 PWROG concluded that the statistical analysis of the core region surveillance program SAW 17 materials are also applicable to the non-core region SAWs.

18 19 The staff considered the RAI response statement that the same set of requirements for 20 minimum tensile strength and absorbed Charpy impact energy were applicable to all Rotterdam 21 RV SAW materials (core region welds and welds outside of the region), and the TR information 22 indicating that the six SAW weld wire vendors and two flux types identified in Tables 9 and 10 23 are applicable to all Rotterdam SAWs (excluding Linde 80 flux type). The staff also took into 24 consideration the RAI response statement that applicability of SAW generic USE data in Table 9 25 is supported by known instances where a core region weld and a non-core region weld share 26 the same heat number. Based on consideration of this information, the staff determined that the 27 Mean - 2 value of 75 ft-lbs, for the seven heat-specific unirradiated USE measurements for 28 SAWs in Rotterdam RV surveillance programs is reasonable as a conservative estimate of the 29 unirradiated Charpy USE for Rotterdam SAWs. Therefore, the staff finds that 75 ft-lbs is 30 acceptable for implementation as a generic unirradiated Charpy USE value for non-Linde 80 31 SAWs in Rotterdam RVs.

32 33 Based on its review of chemistry data provided in Table 10, the staff confirmed that measured 34 values of Cu content are specified for 14 SAW heats. Where multiple Cu measurements 35 (corresponding to specific flux lot numbers) are listed for a specific SAW heat and flux type, the 36 staff confirmed that the Cu content value used in the Mean + 1 calculation is the average of the 37 measured Cu content values for that heat and flux type. One of the 14 Cu content values 38 (0.17 percent) is identified as being based on the weld wire analysis, whereas all the other 39 13 values (as well as all Ni values addressed below) are based on the deposited content. The 40 staff confirmed that the Mean + 1 for the set of 14 Cu content values for Rotterdam SAW 41 materials listed in Table 10 is 0.23 percent. Therefore, the staff determined that a generic Cu 42 content of 0.23 percent may be used for non-Linde 80 Rotterdam SAWs if the measured value 43 is unknown for the specific SAW material.

44 45 The staff confirmed that measured values of Ni content are available for 10 SAW heats. All 46 10 Ni content measurements are identified as based on the deposited weld content. The staff 47 confirmed that the Mean + 1 for the set of 10 Ni content values for Rotterdam SAW materials 48 listed in Table 10 is 0.56 percent. Therefore, the staff determined that a generic Ni content of 49 0.56 percent may be used for non-Linde 80 Rotterdam SAWs if the measured value is unknown 50 for the specific SAW material.

1 Rotterdam RV Shielded Metal Arc Welds 2

3 The NRC staff reviewed the available Charpy impact energy data, percent shear data, chemistry 4 data, and weld identification information for Rotterdam SMAWs, which are provided in TR 5 Table 12. The staff noted that there are 38 Rotterdam SMAWs addressed in TR Table 12, each 6 of which has a unique heat number associated with it. Table 12 identifies four weld types 7 corresponding to four vendors for the 38 heats of SMAW material listed therein. The TR states 8 that all SMAW heats that were used in RV fabrication by Rotterdam, and that are available in 9 the Westinghouse records, are included in Table 12. The staff determined that this information 10 is sufficient to define the material class for Rotterdam SMAWs.

11 12 Of the 38 absorbed impact energy values reported in Table 12, the staff confirmed that there are 13 only three measured values for Charpy USE for Rotterdam SMAWs. These values are:

14 130 ft-lbs, 116 ft-lbs, and 134 ft-lbs. All three USE values have percent shear values of 15 100 percent. The staff agreed with the TR determination that the three measured USE values 16 do not constitute a data set of sufficient size to define a generic USE value based on 17 Mean - 2. Therefore, the TR used a different approach to determine its recommended generic 18 Charpy USE value of 72 ft-lbs for Rotterdam SMAWs.

19 20 As addressed in Section 2.0 of this SE, RG 1.99, Rev. 2, does not explicitly specify Mean - 2 21 as a recommended method for determining a generic unirradiated USE based on evaluation of 22 generic data for a class of material, and the USE requirements of 10 CFR Part 50, Appendix G 23 are generally silent on this issue. Further, Position 1.3 of BTP 5-3 states that in the case of 24 older plants, the [preservice fracture toughness] data may be estimated using procedures listed 25 above [Position 1.2 for plant-specific USE measurements] or other methods that can be shown 26 to be conservative [emphasis added]

27 28 For Rotterdam RV SMAWs, the staff reviewed the TR evaluation for determining a generic USE 29 value of 72 ft-lbs by assessing whether the evaluation has been shown to be conservative, as 30 specified in Position 1.3 of BTP 5-3. While only 3 of the 38 impact energies in Table 12 are 31 determined to be the actual USE, the other 35 absorbed impact energies in Table 12 are 32 reported to represent a lower bound on the USE for the test record. As with the lower bound 33 USE values for SA508, Class 2 forgings, the actual USE values for these 35 SMAW materials 34 are unknown, and they are identified as being greater than or equal to the reported impact 35 energy. Of these 35 lower bound USE values, four of them have percent shear measurements 36 that are less than 95 percent, and 31 do not have percent shear measurements. The TR 37 identifies that all 35 lower bound USE values are based on Charpy tests completed at 10.4 °F or 38 lower and shear values which are either unknown or less than 95 percent. The TR states that 39 USE is typically reached at a temperature greater than 10 °F, as demonstrated by the welds 40 with actual measured USE values, which reached the upper-shelf at temperatures of 41 approximately 70 °F or higher.

42 43 Considering all 38 impact energies reported in Table 12 (35 lower bound USE values plus 44 3 actual USE values), the staff noted 72 ft-lbs for Heat No. 9092 is the second lowest value.

45 The staff also noted that 72 ft-lbs is the lowest of the 31 lower bound USE values that do not 46 have percent shear measurements reported in Table 12. The staff confirmed that the lowest 47 impact energy value in Table 12 is 63 ft-lbs for Heat No. 7359.6708; this impact energy has a 48 corresponding percent shear value of 52 percent. Considering this percent shear 49 measurement, and the low temperature range reported for the Charpy impact test (10.4 °F or 50 lower), the staff found that it is reasonable to expect that actual USE for Heat No. 7359.6708 51 would be significantly greater than 63 ft-lbs if testing of this SMAW material had continued at

1 higher temperatures into the upper-shelf region. On this basis, the staff determined that 2 63 ft-lbs for Heat No. 7359.6708 does not warrant consideration for determining the generic 3 USE value for Rotterdam SMAWs based on the lower bound USE values listed in Table 12.

4 Therefore, the staff finds that the lower bound USE value of 72 ft-lbs is acceptable for 5 implementation as a generic unirradiated Charpy USE value for SMAWs in Rotterdam RVs.

6 7 Based on its review of chemistry data provided in Table 12, the staff confirmed that measured 8 values of Cu content are available for only two SMAW heats. The staff agreed with the 9 PWROG determination that this is insufficient data to determine a generic Cu content value.

10 Accordingly, the staff confirmed that the default Cu content of 0.35 percent, as specified in 11 RG 1.99, Rev. 2, would be the correct value to use if no other information is available (i.e., no 12 heat-specific measurements, no Cu content requirements in material specifications, and no 13 conservative estimates (Mean + 1) based on generic data). The staff noted that the default Cu 14 content of 0.35 percent is conservative relative to the measured values, 0.01 percent and 15 0.023 percent, for Rotterdam SMAWs. Therefore, the staff finds that the RG 1.99, Rev. 2, 16 default Cu content of 0.35 percent is acceptable for Rotterdam SMAWs if the measured value is 17 unknown for the specific SMAW material.

18 19 The staff noted that 32 of the 38 SMAW heats listed in Table 12 have measured values of Ni 20 content. The staff determined that this constitutes a sufficient set of measurements to 21 determine a generic value based on Mean + 1 for Rotterdam SMAWs, as per RG 1.99, Rev. 2.

22 The staff confirmed that the Mean + 1 for the set of 32 Ni content values for Rotterdam SMAW 23 materials listed in Table 12 is 1.13 percent. Therefore, the staff finds that a generic Ni content 24 1.13 percent is acceptable for Rotterdam SMAWs if the measured value is unknown for the 25 specific SMAW material.

26 27 The TR also recommends that if insufficient data exists to determine whether a Rotterdam RV 28 weld is a SAW or a SMAW, then the generic values for unirradiated Charpy USE, Cu content, 29 and Ni content for Rotterdam SMAWs can be utilized. The staff confirmed that the above 30 generic values for unirradiated Charpy USE, Cu content, and Ni content for Rotterdam SMAWs 31 are bounding relative to those for Rotterdam SAWs. The staff also noted that the TR 32 affirmatively states that Rotterdam CMTRs specify only these two weld types for Rotterdam 33 RVs. Therefore, the staff finds that the generic properties for Rotterdam SMAWs are 34 acceptable for Rotterdam RV welds if the weld type (SAW or SMAW) is unknown and if 35 measured values of the applicable properties are unknown for the specific weld materials.

36 37 5.0 CONDITIONS AND LIMITATIONS 38 39 There is no NRC staff-imposed condition or limitation on the use of this TR in licensing 40 applications for addressing regulatory requirements in 10 CFR Part 50, Appendix G; 10 CFR 41 50.61 or 50.61a; and/or TLAA requirements in 10 CFR 54.21(c)(1). However, PWR plants 42 referencing the TR as the basis for the generic Rotterdam RV material properties provided 43 therein must ensure that their RV materials meet the criteria specified in the TR, as set forth 44 below.

45 46 The generic properties provided in the TR are for implementation as conservative generic 47 estimates for the material classes identified below only if no measured values are available for 48 the specific RV material under consideration. PWR plants that implement these generic 49 estimates must identify their RV materials as follows:

50

1

• A PWR plant with a Rotterdam RV proposing to use the generic unirradiated Charpy 2 USE value of 56 ft-lbs for its RV forging(s) must identify that its forging(s) are of the 3 SA508, Class 2 specification and that the forging(s) were supplied by Rheinstahl 4 Huttenwerke AG.

5 6

• A PWR plant with a Rotterdam RV proposing to use the generic unirradiated Charpy 7 USE value 52 ft-lbs for its RV forging(s) must identify that its forging(s) are of the SA508, 8 Class 2 specification. This generic unirradiated Charpy USE value may be used if the 9 Rotterdam RV forging supplier is identified as Fried-Krupp Huttenwerke AG or if the 10 forging supplier is unknown.

11 12

• A PWR plant with a Rotterdam RV proposing to use the generic unirradiated Charpy 13 USE value of 75 ft-lbs for its RV weld(s) must identify that the weld(s) are of the SAW 14 type, that the SAWs are not of Linde 80 flux type, and that its SAW(s) were fabricated by 15 Rotterdam.

16 17

• A PWR plant with a Rotterdam RV proposing to use the generic Cu content of 18 0.23 percent and generic Ni content of 0.56 percent for its RV weld(s) must identify that 19 the weld(s) are of the SAW type, that the SAWs are not of Linde 80 flux type, and that its 20 SAW(s) were fabricated by Rotterdam.

21 22

• A PWR plant with a Rotterdam RV proposing to use the generic unirradiated Charpy 23 USE value of 72 ft-lbs for its RV weld(s) must identify that the weld(s) were fabricated by 24 Rotterdam. This generic unirradiated Charpy USE value may be used if the Rotterdam 25 RV weld is identified as a SMAW or if the Rotterdam RV weld type is unknown.

26 27

• A PWR plant with a Rotterdam RV proposing to use the RG 1.99, Rev. 2, default Cu 28 content of 0.35 percent and generic Ni content of 1.13 percent for its RV weld(s) must 29 identify that the weld(s) were fabricated by Rotterdam. These values may be used if the 30 Rotterdam RV weld is identified as a SMAW or if the Rotterdam RV weld type is 31 unknown.

32 33

6.0 CONCLUSION

34 35 As set forth above, the NRC staff has reviewed the PWROG-17090-NP, Rev. 0, TR and has 36 determined that the TR is acceptable for providing conservative estimates of generic 37 unirradiated Charpy USE for SA508, Class 2 forgings in Rotterdam RVs; and conservative 38 estimates of generic unirradiated Charpy USE, Cu content, and Ni content for Rotterdam SAWs 39 and SMAWs based on the material classes specified in the TR. When measured values of 40 unirradiated Charpy USE, Cu content, and/or Ni content are available for the specific RV 41 materials under consideration, the measured values should be used.

42 43 The NRC staffs review has concluded that when measured values of unirradiated Charpy USE, 44 Cu content, and/or Ni content are not available for the plant-specific Rotterdam RV materials 45 under consideration, the generic values for the Rotterdam RV material classes identified in the 46 TR may be used in PWR plant licensing applications for addressing regulatory requirements in 47 10 CFR Part 50, Appendix G; 10 CFR 50.61 or 50.61a; and/or TLAA requirements in 48 10 CFR 54.21(c)(1).

49

1

7.0 REFERENCES

2 3 1. Letter from Ken Schrader, Pressurized Water Reactor Owners Group, to USNRC 4 Document Control Desk, April 6, 2018, Transmittal of Westinghouse Electric Company 5 Report PWROG-17090-NP, Revision 0, Generic Rotterdam Forging and Weld Initial 6 Upper-Shelf Energy Determination, Westinghouse Non-Proprietary Class 3, March 7 2018 (ADAMS Package Accession No. ML18114A173).

8 9 2. Letter from Ken Schrader, Pressurized Water Reactor Owners Group, to USNRC 10 Document Control Desk, June 12, 2019, Transmittal of the Response to Request for 11 Additional Information, RAI 1-3 and 5-7 Associated with PWROG-17090, Revision 0, 12 Generic Rotterdam Forging and Weld Initial Upper-Shelf Energy Determination, 13 Westinghouse Non-Proprietary Class 3, June 2019 (ADAMS Accession 14 No. ML19170A106).

15 16 3. American Society of Testing and Materials (ASTM) Standard E185-82, Standard 17 Practice for Conducting Surveillance Tests for Light Water Cooled Nuclear Power 18 Reactor Vessels, 1982.

19 20 4. USNRC Regulatory Issue Summary 2014-11, Information on Licensing Applications for 21 Fracture Toughness Requirements for Ferritic Reactor Coolant Pressure Boundary 22 Components, October 14, 2014 (ADAMS Accession No. ML14149A165).

23 24 5. USNRC Regulatory Guide 1.99, Revision 2, Radiation Embrittlement of Reactor Vessel 25 Materials, May 1988 (ADAMS Accession No. ML003740284).

26 27 6. USNRC NUREG-0800, Standard Review Plan for the Review of Safety Analysis Reports 28 for Nuclear Power Plants, Branch Technical Position 5-3, Draft Revision 3, "Fracture 29 Toughness Requirements," September 2018 (ADAMS Accession No. ML18254A090).

30 31 Principle Contributor: Christopher R. Sydnor, NRR/DMLR 32 33 Date:

34 35