LR-N11-0008, Results of Westems Program Benchmarking Activities Associated with Response to NRC Request for Additional Information, Dated November 22, 2010, Related to License Renewal Application

Results of Westems Program Benchmarking Activities Associated with Response to NRC Request for Additional Information, Dated November 22, 2010, Related to License Renewal Application

ML110110428
Person / Time
Site:	Salem   (DPR-070, DPR-075)
Issue date:	01/07/2011
From:	Davison P Public Service Enterprise Group
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
LR-N11-0008
Download: ML110110428 (15)

Text

PSEG P.O. Box 236, Hancocks Bridge, NJ 08038-0236 0

PSEG Nuclear LLC JAN 0 7 201f 10 CFR 50 10 CFR 51 10 CFR 54 LR-N11-0008 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001 Salem Nuclear Generating Station, Unit No. 1 and Unit No. 2 Facility Operating License Nos. DPR-70 and DPR-75 NRC Docket Nos. 50-272 and 50-311

References:

Results of WESTEMSTM Program Benchmarking Activities Associated with the Response to NRC Request for Additional Information, dated November 22, 2010, Related to the Salem Nuclear Generating Station, Units 1 and 2 License Renewal Application

1. Letter from Ms. Bennett Brady (USNRC) to Mr. Thomas Joyce (PSEG Nuclear, LLC) "REQUEST FOR ADDITIONAL INFORMATION FOR SALEM NUCLEAR GENERATING STATION, UNITS 1 AND 2, LICENSE RENEWAL APPLICATION FOR USE OF WESTEMS PROGRAM IN METAL FATIGUE ANALYSIS (TAC NO. ME1 834 AND ME1 836)", dated November 22, 2010

2. Letter from Mr. Paul J. Davison (PSEG Nuclear, LLC) to the NRC, "Response to NRC Request for Additional Information, dated November 22, 2010, related to

1) The use of the WESTEMS TM Program in Metal Fatigue Analysis, and 2)

Confirmation of Environmental Fatigue Locations, associated with the Salem Nuclear Generating Station Units 1 and 2 License Renewal Application", dated December 21, 2010 As part of the NRC request for additional information RAI 4.3-07 transmitted to PSEG Nuclear, LLC (PSEG Nuclear) in Reference 1, the NRC requested PSEG Nuclear to perform a benchmarking evaluation for the WESTEMSTM computer software that is used to monitor metal fatigue at Salem. In Reference 2, PSEG Nuclear responded that the results of this benchmarking effort would be provided to the NRC by January 7, 2011.

The Enclosure to this letter provides the results of the requested benchmarking evaluation.

There are no new or revised regulatory commitments associated with this letter.

A(41 0'a

Document Control Desk LR-N11-0008 Page 2 If you have any questions, please contact Mr. Ali Fakhar, PSEG Manager - License Renewal, at 856-339-1646.

I declare under penalty of perjury that the foregoing is true and correct.

Executed on

("1 I,

Sincerely, IDjon Paul J. D i Vice President, Operations Support PSEG Nuclear LLC

Enclosure:

Results of Benchmarking Evaluation for the WESTEMSTM Metal Fatigue Monitoring Software associated with the Salem Nuclear Generating Station, Units 1 and 2 License Renewal Application cc:

William M. Dean, Regional Administrator - USNRC Region I B. Brady, Project Manager, License Renewal - USNRC R. Ennis, Project Manager - USNRC NRC Senior Resident Inspector - Salem P. Mulligan, Manager IV, NJBNE L. Marabella, Corporate Commitment Tracking Coordinator Howard Berrick, Salem Commitment Tracking Coordinator

Enclosure LR-N11-0008 Page 1 of 13 Enclosure Results of Benchmarking Evaluation for the WESTEMSTM Metal Fatigue Monitoring Software associated with the Salem Nuclear Generating Station, Units 1 and 2 License Renewal Application RAI 4.3-07 (Follow-up)

Enclosure LR-N11-0008 Page 2 of 13 As a follow-up to our response to RAI 4.3-07 dated December 21, 2010, PSEG Nuclear is providing the results of the benchmarking evaluations as requested in the original RAI dated November 22, 2010. For clarity, the applicable portion of RAI 4.3-07 is repeated below, followed by the PSEG Nuclear response.

RAI 4.3-07 (Follow-up):

In addition, the staff requests benchmarking evaluations for two of the limiting locations monitored in the Salem WESTEMS application using the same input parameters and assumptions as those used in traditional, ASME Code,Section III CUF calculations for each location. If such calculations do not exist for either of the selected locations, they should be developed using techniques that allow independent comparison with the WESTEMS results. The intent of this benchmarking evaluation is to confirm that the results of the WESTEMS models, including any analyst judgments, are acceptable and comparable to traditional ASME Code,Section III analyses for the selected monitored locations.

For the pressurizer surge nozzle and the 1.5" BIT line locations that Salem has indicated are monitored in WESTEMS, provide a summary of the benchmarking evaluation that includes the following information:

A comparison of the calculated stresses and CUF using WESTEMS to the same results from'the ASME Code,Section III CUF calculations for all transient pairs representing at least 75 percent of the total CUF from the ASME Code,Section III CUF calculations. One comparison for each unique stress model used in WESTEMS for each selected location is sufficient.

Describe the differences in the results between the WESTEMS evaluation and the ASME Code,Section III CUF calculations for each selected location, and provide a justification for acceptability of the differences.

PSEG Response:

Salem performed a benchmarking evaluation to confirm that the results of the WESTEMSTM models, including any analyst judgments, are acceptable and comparable to traditional ASME Code,Section III analyses for the selected monitored locations. The input parameters and assumptions used in the traditional ASME Code,Section III analyses (representative, hand calculation) and the benchmark of additional fatigue pairs with spreadsheet calculations are the same as those used in the WESTEMSTM design models.

In the environmentally-assisted fatigue (EAF) analyses that supported the Salem License Renewal Application (LRA), one stress model was prepared in WESTEMSTM to represent the Salem Units 1 and 2 components evaluated for EAF with the exception of the Units 1 and 2 Pressurizer Surge Nozzle, including the Safe End to Pipe Weld location. The stress models for the Units' respective selected locations are similar and can be addressed in this benchmarking evaluation as a unique model. The slight differences between the Units' Pressurizer Surge Nozzle Safe End to Pipe Weld stress models are due only to a nominal pipe schedule of 140 (Unit 1) or 160 (Unit 2) at the

Enclosure LR-N11-0008 Page 3 of 13 safe end weld, and would not result in any appreciable differences in the comparisons of stresses and CUFs. There are no differences between the Units' stress models for the BIT Nozzle [Coupling] to Cold Leg Weld locations. Therefore, one comparison was performed for each selected location. The Unit 2 components were selected over the Unit 1 components for the benchmarking due to their higher 60-year cumulative usage factor (CUF) values.

The WESTEMS TM models included in this benchmarking evaluation are the following controlling locations:

Unit 2 Pressurizer Surge Nozzle Safe End to Pipe Weld Unit 2 Safety Injection Boron Injection Tank (BIT) Nozzle [Coupling] to Cold Leg Weld (Note: The Salem LRA referred to this location as a nozzle, and the detailed calculations use the terms nozzle and coupling interchangeably. The physical connection is a coupling welded to two pipes; the 1.5-inch safety injection line, and the 27.5-inch cold leg.)

The benchmarking evaluation for each of the above locations consisted of the following analyses, which are described in more detail following the general overview of the WESTEMS TM fatigue calculation methodology:

1. Benchmarking of Calculated Stresses

2. Benchmarking of WESTEMS TM (Fatigue Usage) with a traditional ASME Code Section III Analysis (Representative Hand Calculation)

3. Benchmarking of Additional Fatigue Pairs with Spreadsheet Calculations

4. Benchmarking of the WESTEMS TM Online Monitoring Model Below is a general overview of the WESTEMSTM fatigue calculation methodology.

The WESTEMS TM computer program performs ASME Boiler & Pressure Vessel Code Section III design stress and fatigue evaluations using an approach that automates traditional analyses, with stress calculations performed using the stress transfer function method. A component model is analyzed by WESTEMS TM at different locations using Analysis Section Numbers (ASNs). A specific example is the Safety Injection BIT Nozzle component. To determine the controlling locations, this component was evaluated at an ASN at the upstream 1.5-inch piping, an ASN at the pipe at toe of socket weld, an ASN at the coupling to cold leg weld, and an ASN at the coupling to cold leg weld at a 90-degree angle to the previous ASN. Stress and fatigue are evaluated at each ASN according to the requirements described in ASME Code Section III, NB-3200.

The stress equations and 'limits in NB-3200 are general enough so that the inputs to the WESTEMSTM computer program can make it applicable to many editions of the Code.

The Salem fatigue analyses of record were performed to the 1986 edition of ASME Code Section II1. The WESTEMS TM computer program is applicable to the 1986 edition of ASME Code Section III.

Enclosure LR-N11-0008 Page 4 of 13 The structural stresses determined to NB-3200 are based on stress intensity and stress intensity range due to mechanical loads (i.e., pressure, moment) and thermal transient loads. The stress component histories for the ASN models are calculated by applying time history mechanical and thermal load scale factors to unit stresses using the transfer function method. The stress transfer functions for the ASN models are developed using finite element analysis (e.g., ANSYS), and are input into the WESTEMS TM models. The stress component histories obtained from the transfer function analysis of each input transfer history are linearized and classified according to ASME criteria; for example:

Primary local membrane stress, PL Primary bending stress, Pb Secondary expansion stress, Pe Secondary membrane plus bending stress, Q Peak Stress, F The stresses may also be adjusted by other input factors such as stress intensification factors or stress concentration factors as they are used to develop Code equation stress histories and ranges. The transient stress histories are evaluated to determine the stress peak and valley times for the ASME Total Stress Intensity and the Primary plus Secondary Stress Intensity. The stress components at the peak and valley times are used to calculate the applicable ranges of Total Stress Intensity (Sp) and Primary plus Secondary Stress Intensity (Sn) required for the Code stress equations in the fatigue calculations.

In performing design analysis, such as the Salem Environmentally-Assisted Fatigue (EAF) analyses supporting the Salem LRA, WESTEMTM uses the stress components for the peak and valley times identified for each input transient history to perform the fatigue usage factor calculations. The ranges of Sn and Sp are calculated for all possible stress cycle pairs between the identified peaks and valleys. The values of Sn are then compared to the allowable stress (3Sm).

The applicable K, (simplified elastic plastic penalty factor applied to alternating stress when Sn limit is exceeded [NB-3228.5]), Ky (elastic modulus correction factor applied to alternating stress [NB-3222.4(e)(4)]), and Kn, (Poisson's Ratio correction factor applied to local thermal stress [NB-3227.6]) adjustments are made according to'NB-3200 procedures to obtain the alternating stress (Sa) for each stress cycle pair. The available cycles (n) determined from the transient definitions are then used for each alternating stress value, along with the material allowable cycles (Naiow), based on Sa, to calculate the cumulative usage factor according to NB-3222.4(e)(5).

Unit 2 Pressurizer Surge Nozzle Safe End to Pipe Weld

1.

Benchmarking of Calculated Stresses The nozzle transfer function stress response from the WESTEMSTM model for this component was compared to an equivalent ANSYS finite element analysis of the same input loadings in a calculation supporting the EAF analyses performed for the Salem LRA. An arbitrary transient was imposed on the component to induce a severe thermal shock. This transient was analyzed by both

Enclosure LR-N11-0008 Page 5 of 13 WESTEMSTM and ANSYS. The time history stress responses of the two models at each ASN were compared based on the shapes of the curves and the numerical maxima and minima of the hoop and axial stress components, which are the controlling stress components in the evaluation. Based on the comparisons for all cases, the benchmark evaluation concluded that the transfer functions were acceptable to generate stress histories for all transients input to the WESTEMSTM analyses. The numerical method has been demonstrated to be applicable for any transient (

Reference:

Westinghouse Proprietary Class 2 WCAP-12315, Rev. 1, 'Transfer Function Method Thermal Stress and Fatigue Analysis: Technical Basis", May 1990). Therefore, benchmarking with one transient is sufficient. In addition, the stresses for the unit mechanical loads (pressure and moments) were also benchmarked by alternate hand calculations for use in the WESTEMS TM models.

2.

Benchmarking of WESTEMS TM (Fatigue Usage) with a traditional ASME Code Section III Analysis (Representative Hand Calculation)

A hand calculation was performed according to ASME Code Section III methodology using a traditional approach to calculate the fatigue usage for the controlling fatigue pair, which has the largest incremental usage factor and significant alternating stress. The controlling pair for this component was formed from stress states of a plant Heatup transient with a maximum system AT (difference between the pressurizer temperature and the reactor coolant system temperature) of 3200F (Heatup @ 320°F AT) at the corresponding peak and valley times.

Table 1 summarizes the results of the hand calculation.

Table 1: Summary of Representative Hand Calculation Results for WESTEMS TM Benchmarking for the Unit 2 Pressurizer Surge Nozzle Safe End to Pipe Weld 3Sm Sn Sp S.

Naiiow n

Case (ksi)

(ksi)

Ke (ksi)

Knu K,

(ksi)

(cycles) (cycles)

U, Hand Calc 49.6 30.46 1.00 129.56 0.37 1.042 72.25 6132 48 0.0078 WESTEMS TM 49.6 30.41 1.00 129.56 0.37 1.042 72.27 6120 48 0.0078 From the above comparison, the single hand calculation example indicates no difference in the incremental fatigue usage (Uj) for the selected stress pair derived from the hand calculation, using the traditional approach, and from WESTEMS TM.

3.

Benchmarking of Additional Fatigue Pairs with Spreadsheet Calculations To complete the benchmark, an Excel spreadsheet was created to calculate values of CUFs for each transient pair representing up to 75% of the total CUF, using the same technique as the hand calculation. Tables 2 and 3 summarize the results of the computations for the hand calculations, as performed by an Excel spreadsheet, to the output of WESTEMSTM, respectively.

Enclosure LR-N11-0008 Page 6 of 13 The definitions for the applicable transients listed in Tables 2 and 3 are as follows:

Transient 1:

Transient 2:

Transient 3:

Transient 5:

Transient 6:

Transient 7:

Transient 9:

Transient 25:

Heatup @ 320°F AT Heatup @ 300°F AT Heatup @ 270°F AT Cooldown @ 320°F AT Cooldown @ 3000F AT Cooldown @ 2700F AT Unit Unloading Feedwater Cycling

Enclosure LR-N1 1-0008 Page 7 of 13 Table 2: Summary of Results for Hand Calculated Fatigue Usage for All Fatigue Pairs (>75%) for the Unit 2 Pressurizer Surge Nozzle Safe End to Pipe Weld TimeA TimeB Pair (Peak or (Peak or 3 Sm Sn Sp Sa Naijow n

ID TranA Valley Time)

TranB Valley Time)

(ksi si)

Ke (ksi)

Knu K,

(ksi)

(cycles)

(cycles)

Uj 1

1 3:22:34 AM 1

3:39:32 AM 49.6 30.41 1

129.56 0.370 1.042 72.25 6132 48 0.0078 2

1 4:16:35 AM 1

5:10:33 AM 49.5 29.03 1

123.29 0.370 1.044 68.83 7457 48 0.0064 3

1 4:33:33 AM 1

4:52:33 AM 49.5 28.57 1

122.00 0.370 1.044 68.19 7743 48 0.0062 4

5 9:25:34 PM 5

9:42:33 PM 49.4 28.40 1

121.05 0.370 1.045 67.69 7976 49 0.0061 5

2 3:09:34 AM 2

3:26:33 AM 49.5 28.32 1

120.70 0.369 1.045 67.48 8077 51 0.0063 6

5 8:37:34 PM 5

8:55:37 PM 49.3 28.16 1

115.73 0.370 1.048 64.75 9541 49 0.0051 7

2 4:03:34 AM 2

4:57:32 AM 49.4 26.90 1

114.68 0.370 1.046 64.20 9875 51 0.0052 8

2 4:20:33 AM 2

4:39:34 AM 49.4 26.62 1

113.71 0.370 1.046 63.67 10211 51 0.0050 9

6 9:14:34 PM 6

9:31:32 PM 49.3 26.38 1

112.67 0.370 1.047 63.15 10554 51 0.0048 10 6

8:27:34 PM 6

8:45:36 PM 49.2 26.44 1

108.79 0.370 1.049 60.94 12691 51 0.0040 11 1

6:06:33 AM 1

6:23:34 AM 49.3 25.30 1

108.00 0.371 1.048 60.58 13062 48 0.0037 12 3

2:50:34 AM 3

3:07:35 AM 49.3 25.24 1

107.57 0.370 1.048 60.31 13349 52 0.0039 13 3

3:43:35 AM 3

4:37:33 AM 49.2 24.18 1

103.00 0.370 1.049 57.79 16429 52 0.0032 14 3

4:00:35 AM 3

4:19:32 AM 49.2 23.85 1

102.03 0.370 1.050 57.35 17051 52 0.0030 15 7

8:58:35 PM 7

9:15:33 PM 49.2 23.79 1

101.51 0.370 1.050 57.03 17522 52 0.0030 16 2

5:51:35 AM 2

6:08:35 AM 49.2 23.71 1

100.99 0.370 1.050 56.71 18008 51 0.0028 17 7

8:12:35 PM 7

8:30:35 PM 49.1 23.67 1

97.42 0.370 1.052 54.73 21464

.52 0.0024 18 5

3:48:36 PM 5

4:06:39 PM 49.0 23.11 1

94.79 0.371 1.052 53.21 24748 49 0.0020 19 9

12:04:10 AM 25 12:11:21 AM 40.6 7.73 1

33.23 0.267 1.120 19.54 3867245 17835 0.0046 Total CUF 0.0855

Enclosure LR-N1 1-0008 Page 8 of 13 Table 3: Summary of Results for WESTEMS TM Fatigue Usage for All Fatigue Pairs (>75%) for the Unit 2 Pressurizer Surge Nozzle Safe End to Pipe Weld for the Unit 2 Pressurizer Surge Nozzle Safe End to Pipe Weld TimeA TimeB Pair (Peak or (Peak or 3 Sm Sn SP Sa Naliow n

ID TranA Valley Time)

TranB Valley Time)

(ksi)

(ksi)

Ke (ksi)

Knu Ky (ksi)

(cycles)

(cycles)

UL 1

1 3:22:34 AM 1

3:39:32 AM 49.60 30.41 1

129ý56 0.370 1.042 72.27 6.12E+03 48 0.0078 2

1 4:16:35 AM 1

5:10:33 AM 49.50 29.03 1

123.29 0.370 1.044 68.85 7.45E+03 48 0.0064 3

1 4:33:33 AM 1

4:52:33 AM 49.50 28.57 1

122.00 0.370 1.044 68.22 7.73E+03 48 0.0062 4

5 9:25:34 PM 5-9:42:33 PM 49.43 28.40 1

121.05 0.370 1.045 67.71 7.97E+03 49 0.0062 5

2 3:09:34 AM 2

3:26:33 AM 49.48 28.32 1

120.70 0.369 1.045 67.47 8.08E+03 51 0.0063 6

5 8:37:34 PM 5

8:55:37 PM 49.30 28.16 1

115.73 0.370 1.048 64.73 9.55E+03 49 0.0051 7

2 4:03:34 AM 2

4:57:32 AM 49.41 26.90 1

114.68 0.369 1.046 64.22 9.86E+03 51 0.0052 8

2 4:20:33 AM 2

4:39:34 AM 49.39 26.62 1

113.71 0.369 1.046 63.70 1.02E+04 51 0.0050 9

6 9:14:34 PM 6

9:31:32 PM 49.34 26.38 1

112.67 0.369 1.047 63.17 1.07E+04 51 0.0048 10 6

8:27:34 PM 6

8:45:36 PM 49.21 26.44 1

108.79 0.370 1.049 60.96 1.27E+04 51 0.0040 11 1

6:06:33 AM 1

6:23:34 AM 49.29 25.30 1

108.00 0.370 1.048 60.60 1.30E+04 48 0.0037 12 3

2:50:34 AM 3

3:07:35 AM 49.29 25.24 1

107.57 0.369 1.048 60.31 1.33E+04 52 0.0039 13 3

3:43:35 AM 3

4:37:33 AM 49.24 24.18 1

103.00 0.370 1.049 57.81 1.64E+04 52 0.0032 14 3

4:00:35 AM 3

4:19:32 AM 49.21 23.85 1

102.03 0.370 1.050 57.33 1.71 E+04 52 0.0030 15 7

8:58:35 PM 7

9:15:33 PM 49.17 23.79 1

101.51 0.370 1.050 57.03 1.75E+04 52 0.0030 16 2

5:51:35 AM 2

6:08:35 AM 49.19 23.71 1

100.99 0.370 1.050 56.70 1.80E+04 51 0.0028 17 7

8:12:35 PM 7

8:30:35 PM 49.07 23.67 1

97.42 0.370 1.052 54.72 2.15E+04 52 0.0024 18 5

3:48:36 PM 5

4:06:39 PM 49.01 23.11 1

94.79 0.371 1.052 53.22 2.47E+04 49 0.0020 19 9

12:04:10 AM 25 12:11:21 AM 40.54 7.73 1

33.23 0.266 1.120 19.53 3.87E+06 17835 0.0046 Total CUF 0.0856 Total CUF 0.0856

Enclosure LR-N11-0008 Page 9 of 13 As seen from Tables 2 and 3, there is an insignificant difference in the CUF as calculated by both the traditional method and by WESTEMS TM. The slight difference in Total CUF values is attributed to rounding off when calculating stresses from the computer files and from interpolations within the material property tables.

4.

Benchmarking of the WESTEMSTM Online Monitoring Model To complete the benchmarking evaluation for the Unit 2 Pressurizer Nozzle Safe End to Pipe Weld location, a comparison was made between the results of the WESTEMSTM design analysis, and the online model used at Salem in support of the enhanced Metal Fatigue of Reactor Coolant Pressure Boundary aging management program (Salem LRA Appendix B, Section 3.1.1). This step was to demonstrate that the online monitoring model produces conservative fatigue usage.

The controlling location was analyzed in the WESTEMSTM online monitoring mode using the same input design transient loadings as those used in the design analysis, as opposed to the program using online plant information. This was done to provide a consistent basis for comparison.

Table 4 shows the comparison.

Table 4: Summary of Results for WESTEMS TM Design Analysis and Online Monitoring Modes for Fatigue Usage for the Unit 2 Pressurizer Surge Nozzle Safe End to Pipe Weld WESTEMSTM WESTEMSTM Design Analysis Online Monitoring Controlling Location Mode CUF Mode CUF Pressurizer Surge Nozzle Safe End to Pipe Weld 0.1121 0.8061 As seen from above table, the online monitoring CUF is conservative, and there is a significant difference between the values determined from the design analysis and from the online monitoring mode. The major contributing factors to the differences are as follows.

The stress peaks and valleys in the online monitoring mode are binned in 1.0 ksi intervals of Sp and Sn rounded up to the next 1.0 ksi in magnitude.

The binned stresses are assigned an appropriate sign (positive, "+" or negative, "-") for conservative contribution to a fatigue pair. For each peak or valley, stress components are used to determine the contribution to a fatigue pair's Sp and Sn values, and therefore it is necessary to assign a sign to these values to properly bin them for later range pairing in the CUF calculation.

The conservative approach used by WESTEMSTM online monitoring mode is to assign the sign of the controlling principal stress, determined from the peak or valley stress components, to the value of Sp or Sn stress intensity

Enclosure LR-N 11-0008 Page 10 of 13 contribution registered in the bins. This approach can result in more conservative stress intensity ranges when the binned stresses are paired, since the stress intensity range that would be produced if the stress components were retained could potentially be smaller if not all stress components had a sign reversal. The purpose of this approach is to minimize the amount of computer memory and data storage required during the life of the monitoring model. For this controlling location, due to the severity of the design transients, the conservative Sp or Sn stress intensity ranges produced a significantly greater, and more conservative CUF.

The design analysis mode also gives the user controls on the transient pairing and legitimate peaks and valleys used in the pairs, as typically done in a traditional ASME Code fatigue analysis. This is not used in the online monitoring mode, since there is no user interaction with the fatigue evaluation in online monitoring applications. This difference in approach also contributed to a significantly greater, and more conservative CUF in the online monitoring mode, as expected.

Overall, since the online monitoring result is greater than the design analysis result, as expected, it is demonstrated that Salem's use of online monitoring of fatigue usage at a controlling location produces a conservative result, as compared to a design analysis of that location.

However, it should be noted that differences this large are not expected for actual plant monitored transients, since they are typically not as severe as the design transients.

Unit 2 Safety Iniection Boron Iniection Tank (BIT) Nozzle [Couplinci] to Cold Leg Weld

1.

Benchmarking of Calculated Stresses The nozzle transfer function stress response from the WESTEMSTM model for this component was compared to an equivalent ANSYS finite element analysis of the same input loadings in a calculation supporting the EAF analyses performed for the Salem LRA. An arbitrary transient was imposed on the component to induce a severe thermal shock. The time history stress responses of the two models at each ASN were compared based on the shapes of the curves and the numerical maxima and minima of the hoop and axial stress components, which are the controlling stress components in the evaluation. Based on the comparisons for all cases, the benchmark evaluation concluded that the transfer functions were acceptable to generate stress histories for all transients input to the WESTEMSTM analyses. The numerical method has been demonstrated to be applicable for any transient (

Reference:

Westinghouse Proprietary Class 2 WCAP-12315, Rev. 1, 'Transfer Function Method Thermal Stress and Fatigue Analysis: Technical Basis", May 1990). Therefore, benchmarking with one transient is sufficient. In addition, the stresses for the unit mechanical loads (pressure and moments) were also benchmarked by alternate hand calculations for use in the WESTEMS TM models.

Enclosure LR-N1 1-0008 Page 11 of 13

2.

Benchmarking of WESTEMS TM (Fatigue Usage) with a traditional ASME Code Section III Analysis (Representative Hand Calculation)_

A hand calculation was performed according to ASME Code Section III methodology using a traditional approach to calculate the fatigue usage for the controlling fatigue pair, which has the largest incremental usage factor and significant alternating stress. The controlling pair for this component was formed from two stress states of the Inadvertent Safety Injection transient at the corresponding peak and valley times.

Table 5 summarizes the results of the hand calculation.

Table 5: Summary of Representative Hand Calculation Results for WESTEMSTM Benchmarking for the Unit 2 Safety Injection Boron Injection Tank (BIT)

Nozzle [Coupling] to Cold Leg Weld 3Sm Sn Sp Sa Nalow n

Case (ksi)

(ksi)

Ke (ksi)

Knu Ky (ksi)

(cycles) (cycles)

U1 Hand Calc 56.87 44.49 1.00 262.55 0.363 1.048 170.63 327 50 0.1529 WESTEMS TM 56.86 45.74 1.00 262.55 0.361 1.048 170.49 327 50 0.1527 From the above comparison, the single hand calculation example indicates an insignificant difference in the incremental fatigue usage (Ui) for the selected stress pair derived from the hand calculation using the traditional approach, and from WESTEMS TM.

3.

Benchmarking of Additional Fatigue Pairs with Spreadsheet Calculations From the above table, the hand calculation for the single transient pair produced a CUF of 0.1527, or 89% of the 60-Year Design CUF for this location reported in LRA Table 4.3.7-2, "Salem Unit 2 60-Year Environmentally-Assisted Fatigue Results".

As compared to the Pressurizer Surge Nozzle Safe End to Pipe Weld location, which required several transient pairs to add up to over 75% of the 60-Year Design CUF, the Safety Injection BIT Nozzle [Coupling] to Cold Leg Weld had only a single transient pair contributing to over 75% of the CUF, therefore, it was not required to generate additional calculations.

4.

Benchmarking of the WESTEMS TM Online Monitoring Model To complete the benchmarking evaluation for the Unit 2 Safety Injection BIT Nozzle [Coupling] to Cold Leg Weld location, a comparison was made between the results of the WESTEMSTM design analysis, and the online model used at Salem in support of the enhanced Metal Fatigue of Reactor Coolant Pressure Boundary aging management program. This step was to demonstrate that the online monitoring model produces conservative fatigue usage.

Enclosure LR-N1 1-0008 Page 12 of 13 The controlling ASN location in WESTEMSTM was analyzed in the computer program's online monitoring mode using the same input design transient loadings as those used in the design analysis, as opposed to the program using online plant information. This was done to provide a consistent basis for comparison.

Table 6 shows the comparison.

Table 6: Summary of Results for WESTEMS TM Design Analysis and Online Monitoring Modes for Fatigue Usage for the Unit 2 Safety Injection Boron Injection Tank (BIT) Nozzle [Coupling] to Cold Leg Weld WESTEMSTM WESTEMSTM Design Analysis Online Monitoring Controlling Location Mode CUF Mode CUF Safety Injection BIT Nozzle Coupling to Cold Leg Weld 0.1717 0.7078 As seen from above table, the online monitoring CUF is conservative, and there is a significant difference between the values determined from the design analysis mode and from the online monitoring mode. The major contributing factors to the differences are as follows.

The stress peaks and valleys in the online monitoring mode are binned in 1.0 ksi intervals of Sp and Sn rounded up to the next 1.0 ksi in magnitude.

0 The binned stresses are assigned an appropriate sign (positive, "+" or negative, "-") for conservative contribution to a fatigue pair. For each peak or valley, stress components are used to determine the contribution to a fatigue pair's Sp and Sn values, and therefore it is necessary to assign a sign to these values to properly bin them for later range pairing in the CUF calculation.

The conservative approach used by WESTEMSTM online monitoring mode is to assign the sign of the controlling principal stress, determined from the peak or valley stress components, to the value of Sp or Sn stress intensity contribution registered in the bins. This approach can result in more conservative stress intensity ranges when the binned stresses are paired, since the stress intensity range that would be produced if the stress components were retained could potentially be smaller if not all stress components had a sign reversal. The purpose of this approach is to minimize the amount of computer memory and data storage required during the life of the monitoring model. For this controlling location, due to the severity of the design transients, the conservative stress intensity ranges and associated K, penalty factors produced a significantly greater, and more conservative CUF.

The design analysis mode also gives the user controls on the transient pairing and legitimate peaks and valleys used in the pairs, as typically done in a traditional ASME Code fatigue analysis. This is not used in the "online monitoring" mode, since there is no user interaction with the fatigue evaluation in online monitoring applications. This difference in approach also

Enclosure LR-N1 1-0008 Page 13 of 13 contributed to a greater, and more conservative CUF in the online monitoring mode, as expected.

Overall, since the online monitoring result is greater than the design analysis result, as expected, it is demonstrated that Salem's use of online monitoring of fatigue usage at a controlling location produces a conservative result, as compared to a design analysis of that location.

However, it should be noted that differences this large are not expected for actual plant monitored transients, since they are typically not as severe as the design transients.

Summary Salem has provided reasonable assurance through the above benchmarking evaluations for two of the highest fatigue usage locations for Salem that the use of the WESTEMSTM computer program is comparable to a traditional ASME Section III approach for design regarding fatigue usage, and its online monitoring mode is conservative as compared to the design analysis mode, Salem will use WESTEMSTM as part of its enhanced Metal Fatigue of Reactor Coolant Pressure Boundary aging management program as discussed in the LRA.