LR-N09-0027, Reactor Vessel Outlet Nozzle Dissimilar Metal Weld Flaw Evaluation (Fall 2008 Outage)

Reactor Vessel Outlet Nozzle Dissimilar Metal Weld Flaw Evaluation (Fall 2008 Outage)

ML090500369
Person / Time
Site:	Salem
Issue date:	12/31/2008
From:	Ching Ng Westinghouse
To:	Office of Nuclear Reactor Regulation
References
LR-N09-0027, LTR-PAFM-08-158
Download: ML090500369 (28)

Text

Westinghouse Non-Proprietary Class 3 LTR-PAFM-08-158 Salem Unit 1 Reactor Vessel Outlet Nozzle Dissimilar Metal Weld Flaw Evaluation (Fall 2008 Outage)

December 2008 Author: C. Ng*, Piping Analysis and Fracture Mechanics Verifier: A. Udyawar*, Piping Analysis and Fracture Mechanics Approved: S. A. Swamy*, Manager, Piping Analysis and Fracture Mechanics

• Electronicallyapproved records are authenticatedin the Electronic Document Management System

© 2008 Westinghouse Electric Company LLC All Rights Reserved O)Westinghouse

LTR-PAFM-08-158 1.0 Introduction An inside surface connected circumferential indication was detected in a Salem Unit 1 reactor vessel outlet nozzle dissimilar metal (Alloy 182) butt weld during the pre Mechanical Stress Improvement Process (MSIP) inspection in the Fall 2008 Refueling Outage. The detected indication exceeded the acceptance standards of ASME Boiler & Pressure Vessel Code Section XI, Subsection IWB Article IWB-3500 (Reference 1). MSIP was subsequently performed for the reactor vessel outlet nozzle of interest during the Fall 2008 outage. This process induces compressive residual stresses on the inside surface and inner wall of the nozzle where the indication was detected. The post-MSIP residual stresses, when combined with the stresses due to normal operating piping reaction loads would mitigate any future crack propagation due to Primary Water Stress Corrosion Cracking (PWSCC).

To demonstrate that such a mitigation is technically sound, a flaw evaluation was performed for the nozzle of interest, using the rules of ASME Code Section XI, Article IWB 3600 for the as-found indication. Fatigue and PWSCC crack growth analyses were also performed to demonstrate the effectiveness of MSIP to mitigate any future crack propagation.

2.0 Geometry, Piping Loads and Material Properties Nozzle dimensions at the weld location of interest (Reference 2),iand the relevant design parameters used in the flaw evaluation are summarized in Table 2-1. Table 2-2 provides the as-found flaw dimensions as reported in Reference 2. The piping reaction loads at the reactor vessel outlet nozzle dissimilar metal weld are summarized in Table 2-3. It should be noted that the loads tabulated in Table 2-3 are bounding Salem Units I and 2 loads and therefore are conservative and acceptable to use for Salem Unit 1.

Table 2-1: Geometry and Design Parameters Outside diameter 34.1 inch Wall thickness 2.62 inch Bounding Normal Operating Temperature 609 OF Normal Operating Pressure 2235 psig Table 2-2: As-found Flaw Parameters Flaw Depth 0.634 inch ID Flaw Length 2.06 inch Flaw Depth / ID Flaw Length 0.31 Flaw orientation Circumferential Page 2 of 7

LTR-PAFM-08-158 Table 2-3: Reactor Vessel Outlet Nozzle Piping Reaction Loads Forces Moments Loading (kips) (in-kips)

F, M), MY M, (Axial) (Torsion) (Bending) (Bending)

Deadweight -0.27 -83.5 -25.9 -1031 Thermal -48.4 289 -320 -17737 OBE 670.8 455 3324 7463 SSE 870 691 4625 14893 LOCA 938 2090 22011 8930 Note:

OBE - Operating Basis Earthquake SSE - Safe Shutdown Earthquake LOCA - Loss of Coolant Accident 3.0 Flaw Evaluation Since the as-found circumferential indication exceeded the acceptance standards of ASME Section XI, Subsection IWB Article IWB-3500, a flaw evaluation in accordance with the analysis procedures and acceptance criteria as specified in Subsection IWB-3600 was performed. The limit load approach in the ASME Code Section XI, Division 1 Appendix C was used in the flaw evaluation.

Using the equations in the ASME Code Appendix C, the maximum allowable end-of-evaluation period flaw depth to wall thickness ratios were calculated as a function of flaw depth to flaw length ratio for normal, upset, emergency and faulted conditions. The results from the limiting service condition are plotted in Figure 1 as well as the as-found flaw parameters. Since the as-found flaw depth is below the maximum allowable end-of-evaluation period flaw depth shown in Figure 1, the as-found indication was acceptable in accordance with the applicable ASME Section Xl acceptance criteria prior to the MSIP application at Salem Unit 1, thus demonstrating that the plant was in an operable condition with the detected circumferential indication in the reactor vessel outlet nozzle dissimilar metal weld.

Figure 2 shows that the post-MSIP axial through-wall stress is compressive up to about 40% of the wall thickness from the inside surface after superposing the post-MSIP residual stress with the normal operating axial stress due to piping reaction loads at the reactor vessel outlet nozzle dissimilar metal weld. The post-MSIP residual stress shown in Figure 2 is obtained from Reference

3. This compressive stress region far exceeds the as-found flaw depth to wall thickness ratio of 0.24, thereby indicating that the MSIP application will arrest any future crack growth due to PWSCC for the as-found indication. This is illustrated further by determining the crack tip stress intensity factor (SIF) for a circumferential indication based on a linearized post-MSIP axial through-wall stress distribution profile and the influence coefficients obtained from Reference 4. The crack tip stress intensity factor calculated for a circumferential indication with a conservative aspect ratio Page 3 of 7

LTR-PAFM-08-158 (length/depth) of 6 and a conservative flaw depth to wall thickness ratio of 0.25, which bounds the as-found indication, is shown in Table 3-1. The crack tip stress intensity factor calculated is clearly negative and reinforces the conclusion that MSIP will mitigate any future crack growth due to PWSCC for the as-found indication.

Table 3-1: Post-MSIP Crack Tip Stress Intensity Factor Flaw Length to Flaw Depth Ratio 6.0 Flaw Depth to Wall Thickness Ratio 0.25 Through-wall Membrane Stress 14.5 ksi Through-wall Bending Stress (Compressive at the Inside Surface)

Crack Tip Stress Intensity Factor (KI) -32.6 ksi i-In order to demonstrate that the as-found indication would remain in the compressive stress region in the post-MSIP configuration, fatigue crack growth for a circumferential indication with a conservative aspect ratio (length/depth) of 6 and a conservative flaw depth to wall thickness ratio of

.0.25 was evaluated in accordance with the fatigue crack growth rate for Alloy 182 weld material from NUREG/CR-6907 (Reference 5) based on the applicable Salem Unit 1 design thermal transients at the reactor vessel outlet nozzle dissimilar metal weld. The results of the fatigue crack growth analysis are shown in Table 3-2 and there is no fatigue crack growth. This is primarily due to the fact that the post-MSIP axial through-wall stress is compressive up to about 40% of the wall thickness from the inside surface and there are no severe thermal transients at the reactor vessel outlet nozzle dissimilar metal weld. The results of the fatigue crack growth analysis demonstrated that MSIP will also mitigate any future fatigue crack growth and the as-found circumferential indication would remain in the compressive stress region. Therefore, PWSCC is no longer a credible crack growth mechanism for the as-found indication.

Table 3-2: Post-MSIP Fatigue Crack Growth Results Flaw Depth to Wall Thickness Ratio After Initial Flaw Depth to Wall 10 years .20 years 30 years 40 years Thickness Ratio 0.25 0.25 0.25 0.25 0.25 Page 4 of 7

LTR-PAFM-08-158 4.0 Summary and Conclusions The circumferential indication detected in the Salem Unit 1 reactor vessel outlet nozzle dissimilar metal butt weld was found to meet the allowable limits of the ASME Section XI Article IWB-3600 and Appendix C requirements under normal, upset, emergency and faulted service conditions before MSIP application was performed during the Fall 2008 outage, thus demonstrating that Salem Unit 1 was in an operable condition at all times with the detected circumferential indication in the reactor vessel outlet nozzle dissimilar metal weld.

A review of the post-MSIP through-wall axial stress distribution shows that the indication of interest in the reactor vessel outlet nozzle dissimilar metal weld is well within the compressive stress region of the nozzle and that the resulting crack tip stress intensity factor calculated is negative. Fatigue crack growth evaluation was also performed for the post-MSIP configuration to demonstrate that there is no fatigue crack growth and that the as-found indication would remain in the compressive stress region of the nozzle. Therefore, it can be concluded that the MSIP application implemented at the Salem Unit 1 reactor vessel outlet nozzle of interest is effective and will mitigate any future crack growth due to both PWSCC and fatigue crack growth mechanisms.

5.0 References

1. Rules for Inservice Inspection of Nuclear Power Plant Components, ASME Boiler & Pressure Vessel C6de,Section XI, 1998 Edition, including 2000 Addenda.

2. WesDyne Letter WDI-PJF-1303981-TR-010, Rev. 0, "Salem Unit 1 - ASME Code Section XI IWB-3500 Evaluation of Flaw Indication #11 in Outlet Nozzle 29-RC-1140-1", dated October 26, 2008.

3. NuVision Engineering Report 4418-6-001, "MSIP Finite Element Stress Analysis Procedure -

Salem Unit 1 Outlet Nozzle To Safe End Weld 29-RC-1140-1 Flaw Evaluation", November 2008.

4. S. Chapuliot, M.H. Lacire and P. Le Delliou, "Stress Intensity Factors for Internal Circumferential Cracks in Tubes over a Wide Range of Radius over Thickness Ratios", ASME PVP Volume 365, 1998.

5. NUREG/CR-6907, ANL-04/3, ."Crack Growth Rates of Nickel Alloy Welds in a PWR Environment," May 2006.

Page 5 of 7

LTR-PAFM-08-158 0.80 0 0.70 0.60 U_

0.50 0

0.40 W

0.30

• 0

.2 0.20 E

E 0.10 X

0.00 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 Flaw Depth / Flaw Length Ratio Figure 1: Comparison of As-Found Flaw Size with Maximum Allowable End-of-Evaluation Period Flaw Size Page 6 of 7

LTR-PAFM-08-158 80.0 60.0 40.0 20.0

• J U) 0.0

-20.0

-40.0

-60.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Flaw Depth I Wall Thickness Ratio F* Post-MSIP Normal Operating Stress + Residual Stress ------ Post-MSIP Residual Stress J

Figure 2: Reactor Vessel Outlet Nozzle Dissimilar Metal Weld Post-MSIP Through-wall Axial Stress Distribution Page 7 of 7

ATTACHMENT 2 Westinghouse Non-Proprietary Class 3

• Westinghouse To: R. Walters Date: December 03, 2008 cc: S. A. Swamy R. L. Brice-Nash C. K. Ng From: M. Dowdell Your ref:

Ext: 724-722-6102 Our ref: LTR-PAFM-08-166 Fax: 724-722-5597

Subject:

Salem Units 1 and 2 - RCL Piping Stresses at the RPV Nozzles

References:

1. Westinghouse Calculation: CN-PAFM-06-1, Rev. 1, "Salem Unit 2 RCL Piping and Nozzle Qualifications for the Replacement Steam Generator," B. Carpenter.

[Westinghouse Proprietary Document]

2. ASME Boiler and Pressure Vessel Code,Section III, 1986 Edition.

3. ASME Boiler and Pressure Vessel Code, Section 111,1974 Edition.

4. NuVision Report: 44'18-4-001-00, "Analytical Verification of MSIP© for RV Cold Leg Nozzle to Safe End Weld Salem Unit 1," October 2007.

[NuVision Proprietary Document]

5. NuVision Report: 4418-4-002-00, "Analytical Verification of MSIP© for RV Hot Leg Nozzle to Safe End Weld Salem Unit 1 & 2," October 2007.

[NuVision Proprietary Document]

6. NuVision Report: 4418-4-003-00, "Analytical Verification of MSIP© for RV Cold Leg Nozzle to Safe End Weld Salem Unit 2," September 2008.

[NuVision Proprietary Document]

7. Westinghouse Calculation: CN-SMT-02-54, Rev. 1, "Salem 1 RCL SG Snubber Elimination Results," L. M. Valasek.

[Westinghouse Proprietary Document]

8. Westinghouse Calculation: CN-PAFM-05-91, Rev. 1; "Salem 2 RSG Deadweight, Thermal and LOCA WESTDYN Runs," B. Carpenter.

[Westinghouse Proprietary Document]

R. Walters, Please transmit the following to the Salem customer.

The normal operating condition stresses at the reactor pressure vessel (RPV) inlet and outlet nozzles for Salem Units 1 and 2 have been evaluated against the stress allowable, Sm. Based on the following evaluations, the normal operating condition stresses at the RPV inlet and outlet nozzles are less than 1 S,,,, as consistent with NUREG-0313.

The evaluation was performed using the following methodology. The normal operating condition stresses consisted of deadweight, 100% power thermal, and operating pressure stresses. The stresses were evaluated at the location of minimum wall thickness for the area between the RPV

©2008 Westinghouse Electric Company LLC All Rights Reserved

Westinghouse Non-Proprietary Class 3 Page 2 of 4 Our ref: LTR-PAFM-08-166 December 03, 2008 nozzle and the reactor coolant loop (RCL) piping. The normal operating condition stress was then evaluated against the minimum stress allowable, Smn, for the materials used.

The minimum wall thickness for the section from the RPV nozzle to the RCL piping was determined from a review of Figures 2.1-1 found on pages 19 of References 4, 5, and 6. From the review, the minimum wall thickness occurs at the location between the safe end and the safe end weld to the RCL piping. For the Salem Unit 1 and 2 hot legs, the minimum wall thickness was determined to be 2.4 inches; calculated as ((34.0" - 29.2") / 2), see Figure 1. For the Salem Unit 1 and 2 cold legs, the minimum wall thickness was also determined to be 2.4 inches; calculated as

((32.5" - 27.7") / 2), see Figure 1.

Hot LegI Cold Leg{

Safe-end 0.5" ISafe-end 005 34.0" 32.5" 29.0" 29.2" 27.5" 27.7" Figure 1: Hot Leg and Cold Leg Safe End Dimensions The following equations were used to calculate the deadweight (DW), thermal (TH), and pressure (P) stresses, as well as the section modulus (Z). The outer diameter (D.) and the inner diameter (D1) were taken from the location of the minimum wall thickness, as seen in Figure 1. For the following equations, a normal operation pressure of 2235 psi was used. The stress indices of 1.0 at the outlet nozzle and 1.62 at the inlet nozzle (References 1 and 7) were also used.

4 4 iM Dw iMTH PDi2 Z=- 32 D i (YDW -- p 2 2 Z 'TH

- z 32 D.

Do -Di The resultant moment loads (M) at the RPV nozzles for Salem Units I and 2 were taken from References 7 and 8. The deadweight and thermal resultant moment loads are summarized in Table 1 for the RPV inlet and outlet nozzles.

©2008 Westinghouse Electric Company LLC All Rights Reserved

Westinghouse Non-Proprietary Class 3 Page 3 of 4 Our ref: LTR-PAFM-08-166 December 03, 2008 Table 1: Resultant Moment Loads at the RPV Nozzles Unit Nozzle Case Load (in-k) Reference Unit 1 Outlet DW 802 Ref. 7 Unit 1 Outlet TH 17739 Ref. 7 Unit 1 Inlet DW 344 Ref. 7 Unit 1 Inlet TH 2552 Ref. 7 Unit 2 Outlet DW 1034.26 Ref. 8 Unit 2 Outlet TH 10813.4 Ref. 8 Unit 2 Inlet DW 350.23 Ref. 8 Unit 2 Inlet TH 2136.12 Ref. 8 The deadweight, thermal, and pressure stresses are calculated below in Table 2. The total stress listed in Table 2 is the summation of the deadweight, thermal, and pressure stress to represent the stress at the normal operating condition.

Table 2: Deadweight, Thermal, and Pressure Stresses Pressure Deadweight Thermal Total Unit Nozzle Stress (ksi) Stress (ksi) Stress (ksi) Stress (ksi)

Unit 1 Outlet 6.282 0.456 10.082 16.820 Unit 1 Inlet, - 5:935 0.350 2.597 8.882 Unit2 Outlet 6.282 0.588 6.146 13.015 Unit 2 Inlet 5.935 0.356 2.174 8.465 The stress allowable (Sm) was determined for the materials of the RPV nozzle, safe end, and RCL piping. The stress allowable for the weld materials is not listed as it is not the limiting allowable.

The RCL piping for Salem Units 1 and 2 are qualified according to the B3 1.1 piping code, which does not contain the stress allowable, Sm. However for this evaluation, the stress allowable, Sn, was determined using the ASME Code. Comparisons were made between the stress allowables in the 1974 and 1986 Editions (References 2 and 3) and there were no differences in value for the materials considered. The stress allowables are listed in Table 3. The stress allowables at 600°F and 650'F were determined from the ASME Code and the allowable at 620'F was interpolated.

Table 3: Material Stress Allowable (Sm)

Component Material Sm @ 600°F (ksi) Sm @ 620°F (ksi) Sm @ 650°F (ksi)

RPV Nozzle SA-508 Cl. 2 26.7 26.7 26.7 Safe End SA-182 F316 17.0 16.9 16.7 RCL Piping SA-376 TP316 17.0 16.9 16.7 Based on Table 3, the lowest allowable at 620°F (bounding temperature of the hot leg during normal operation) is 16.9 ksi. From Table 2, the maximum total normal operating condition stress is at the outlet nozzle and is 16.82 ksi. Therefore, the stresses at the RPV inlet and outlet nozzles for Salem Units I and 2 are less than 1 Sm.

©2008 Westinghouse Electric Company LLC All Rights Reserved

Westinghouse Non-Proprietary Class 3 Page 4 of 4 Our ref: LTR-PAFM-08-166 December 03, 2008 Should you have any questions, please contact the undersigned.

Regards, Author: M. Dowdell* Verifier: B. Carpenter*

Piping Analysis & Fracture Mechanics Piping Analysis & Fracture Mechanics

• Electronicallyapproved records are authenticatedin the Electronic Document ManagementSystem.

©2008 Westinghouse Electric Company LLC All Rights Reserved

Attachment 3 Response to Item 4 The description of the MSIP analysis procedure is given in report Number 4418-6-001, Outlet Nozzle to Safe End Weld 29-RC-1 140-1 Flaw Evaluation. The tool location and amount of squeeze is established such that the post- MSIP residual stresses are compressive in the inner weld region through about 50% of the wall thickness.

The limitations on the application of MSIPregarding size of the indications that may be found in the pre- MSIPinspection are 10% circumferential length and 30% through-wall depth for circumferential flaws. With the qualified NDE inspections, the 30% through-wall depth limitation is also used for axial flaws. These are consistent with what the NRC has approved in the past.

If an indication is found in the pre- MSIPinspection, then depending on its orientation, the appropriate through-wall stress distribution (axial or hoop) is determined for the actual squeeze at the location of the indication. This through-wall stress distribution includes the compressive residual stress generated by MSIP, plus the stresses due to the operating loads and is used in the fracture mechanics evaluation to confirm there will be no flaw growth.

Nu Ffly lan Ergf67se lr K. Arc.

MSIPR FINITE ELEMENT STRESS ANALYSIS PROCEDURE SALEM UNIT 1 OUTLET NOZZLE TO SAFE END WELD 29-RC-1140-I FLAW EVALUATION PROTECTED BY U.S. PATENT Nas. 4.683,014 & 4,612,D71 Prepared ftr WESTINGHOUSE ELECTRIC COMPANY, LLC PITTSBURGH, PENNSYLVANIA November 2008 1'repared by Dutp Qualilv A.isuranee Date VerifiedJ by 1542 AW kVl(_1 Date Apj'-fJivetl by Ive/-$r

- bN d bl ýle,76 NuVision Engineering, Inc.

River Park Common.

24W3 8idlnqy Street, Suite 700 Pittsburgh. PA 16203 -2181 (412) 586 - 1810 FAX: (412)588-1811 D L-Rey' APL- I

.LrJ' ~ rtt&avbw WQSC"ta.ZJWC US!!' krzbo ~ ~ ~.qM*j'tja'ii

.Yrikz *41~.#USsr.

44;&oy

Nu Vision Engineering,Inc.

MSIP FINITE ELEMENT STRESS ANALYSIS PROCEDURE SALEM UNIT 1 OUTLET NOZZLE TO SAFE END WELD 29-RC-1140-1 FLAW EVALUATION Table of Contents Page

1.0 INTRODUCTION

1 2.0 ANALYSIS METHODOLOGY 2 3.0 ANALYSIS RESULTS 5 Figure 1 - Finite Element Model 6 Figure 2 - Finite Element Model - Weld Detail 6 Figure 3 - Finite Element Model- Boundary Conditions w/Tool Pressure Loading , -"

'.' 7 Figure 4 - Finite Element Model - ANSYS Material Number 7 Figure 5 - NSE Weld - As-Designed Weld w/Local ID Weld Repair Residual Axial Stress 8 Figure 6 - NSE Weld - As-Designed Weld w/Local ID Weld Repair Residual Hoop Stress 8 Figure 7 - Salem Hot Leg, Path Along Flaw Location 9 Figure 8 - Salem Hot Leg, Stress Along Flaw Path, PRE MSIP 10 Figure 9 - Post MSIP Residual Axial Stress -

Nom. Position 11 Figure 10 - Post MSIP Residual Hoop Stress -

Nom. Position 11 Figure 11 - Salem Hot Leg, Stress Along Flaw Path, Post MSIP w/ Operating Pressure and Temperature 12

4.0 REFERENCES

13 4418-6-001 i

Nu Vision Engineering,Inc.

MSIP FINITE ELEMENT STRESS ANALYSIS PROCEDURE SALEM UNIT 1 OUTLET NOZZLE TO SAFE END WELD 29-RC-1140-1 FLAW EVALUATION

1.0 INTRODUCTION

MSIP@ is a mechanical process that replaces the large tensile as-welded residual stresses along the inner region of piping in the vicinity of a weld with a zone of compressive residual stress. In general terms, MSIP consists of squeezing a pipe plastically near a weld using a specifically designed set of rings that grip a short length of pipe. The squeezing is continued until the tensile residual stresses along the inner region of the weld are replaced by compressive stresses.

A finite element stress analysis is performed that provides the analytical basis for MSIP and verifies that the process does accomplish the desired stress improvement. The -general objective of the evaluation is to analytically simulate MSIP on the pipe-to-safe end-to-nozzle assembly and to predict the resulting redistribution of residual stresses. This is accomplished by performing a nonlinear finite element analysis of the process. Residual stresses are determined in the weldment for the as-welded condition with a local ID weld repair, post-hydrostatic test and post- MSIP process as well as steady state plant operating conditions both before and after the application of MSIP. A comparison of the analytical residual stress patterns shows that MSIP is successful in alleviating the substantial weld-induced tensile stresses in both the axial and circumferential (hoop) directions along the inner region of the nozzle to safe end weld.

A key objective of the evaluation is to determine the acceptable range of MSIP tool location and radial contractions necessary to achieve the desired stress redistribution. Ranges for the desired radial contraction and tool locations are used in development of the MSIP Parameters for field application of MSIP.

4418-6-001 I of 13

Nu Vision Engineering,Inc.

2.0 ANALYSIS METHODOLOGY The ANSYS finite element program is used for the MSIP analysis. The analysis of MSIP on the nozzle-to-safe end weldment is performed using an axisymmetric finite element computer model of the weldment assembly. The model represents dimensional details of the Nozzle, safe end, attached pipe and the circumferential welds that connect the pipe to safe end and the safe end to the nozzle.

Temperature dependent inelastic material properties are used in the evaluation to ensure that the nonlinear behavior of the materials is represented. The inelastic behavior of the material is simulated using a bilinear kinematic hardening (BKhIN) option within ANSYS.

The computer model is first used to simulate the residual stresses induced by the welding process in the weld regions. This is accomplished by imposing a temperature gradient in the weld region and then removing it, which results in residual stresses distributions similar in magnitude to the weld residual stresses (including ID weld repair) reported in the literature.

Nextpressure due to the hydrostatic test is applied to the model and thenwremoved.' Normal operating temperature and pressure are then applied and removed.

The application of MSIP is simulated in the model by first applying pressure along the outer surface of the safe end to simulate loading applied by the MSIP tool clamp ring assembly and then removing it. The maximum pressure generated by the MSIP tool is increased until acceptable residual stress redistribution is achieved in the nozzle to safe end weld region upon removal of the tool.

Steady state plant operating loading, represented by normal operating temperature and internal pressure, is then applied to the post- MSIP)residual stress pattern. A uniform temperature equivalent to the plant normal operating condition (Thor) is applied throughout the material, and a uniform pressure (Pop) equivalent to the plant normal operating condition is also applied to all inside surfaces of the nozzle, safe end, weld and pipe including blow-off pressure. The results of primary importance are the final axial and hoop stresses in the inner region of the nozzle-to-safe end weld.

Finally a series of finite element analyses are performed varying the amounts of radial contractions and MSIP tool locations to determine acceptable ranges for field implementation of MSIP.

4418-6-001 2 of 13

Niu Vision Engineering,Inc.

2.1 FINITE ELEMENT MODEL GEOMETRY AND BOUNDARY CONDITIONS A 2-D axisymmetric finite element computer model of the RV hot leg nozzle assembly was used in the analysis. The dimensions of the weldment assembly were based on drawings and walk-down information and were used as the foundation of the finite element model.

The model was constructed using the ANSYS PLANE42 axisymmetric isoparametric elements.

The overall finite element model and its detail in the weld region are shown in Figures 1 and 2 respectively.

The overall length of the model on the pipe side of the assembly is extended to be equal to or greater than 4(rt)(0 5)where "r" is the average radius and "t" is the wall thickness. The nodes at the end of the pipe are coupled in the axial direction to simulate the remaining piping not modeled. To ensure stability of the finite element solution, the nozzle end of the model is constrained normal to the nozzle cut plane. The model boundary conditions are shown in Figure 3.

The' oadiiig applied by the contour ring of the MSIP tool clamp ring assembly is simiilatedby a radial pressure along the outer surface of the pipe. The applied tool pressure is also illustrated in Figure 3.

2.2 MATERIAL PROPERTIES The different materials of construction for the RV hot leg nozzle assembly are shown in Figure 4 and tabulated in Table 1.

Table 1 - RV Nozzle Assembly Materials Component Description ANSYS Mat. Number Material Designation in Model Nozzle 1 SA-508 Cl. 2 Cladding 2 E308L Nozzle to Safe End Weld 4 Inconel 182 Safe End 5 SA-182-F316 Pipe to Safe End Weld 6 E308L Pipe 8 SA-376 TP316 The inelastic behavior of the material is simulated using the bilinear kinematic hardening (BKIN) option within ANSYS. The material properties used in the analysis are based on certified material test reports (CMTR) when available otherwise typical yield strength and ultimate tensile strength of the material are used.

4418-63001 3 of 13

Nu Vision Engineering,Inc.

2.3 SIMULATION OF AS-WELDED STRESSES The residual stresses due to welding are simulated in the model by applying a temperature distribution in each of the weld regions. The temperature distribution is then removed to achieve a weld residual stress profile. The temperature distributions are applied in the same sequence as they were when the weldment was assembled. The nozzle to safe end as-designed weld was first applied followed by an ID weld repair to the nozzle to safe end weld. This was then followed by pipe to safe end weld.

2.4 PRE-MSIP LOADING Prior to the application of MSIP the effect of the hydrostatic testing and normal operating conditions on the residual stress patterns is evaluated. This is accomplished by applying a uniform temperature to the entire model as well as a pressure loading to the inner surfaces of the nozzle assembly. The loading due to blow-off is applied to the end of the pipe furthest from the nozzle.

2.5,, SIMULATION OF TOOL APPLICATION After establishing the as-welded stresses with ID weld repair and the residual stress distribution after normal plant operations, the mechanical loading of MSIP was simulated. This is accomplished by applying a pressure along the outer surface of the pipe as shown in Figure 3.

The pressure load is incrementally applied to ensure convergence of the nonlinear solution.

After full loading is achieved, the pressure load is removed to simulate the removal of the tool.

There is a permanent change to the outside diameter of the pipe when the pressure is applied and removed from the model. All post- MSIPresults are for the case when the tool is removed.

2.6 POST- MSIPOPERATING LOAD After the desired MSIPstress redistribution is achieved, the effect of the plant normal operating conditions on the post- MSIP@ residual stress patterns is evaluated. This is accomplished by applying a uniform temperature to the entire model as well as a pressure loading to the inner surfaces of the nozzle assembly and the loading due to blow-off, as described in Section 2.4.

4418-6-001 4 of 13

Nu Vision Engineering, Inc.

3.0 ANALYSIS RESULTS The effectiveness of MSIP application on the nozzle assembly is determined through this analysis [Ref. 1]. As stated previously, a model with representative dimensions is generated followed by application of local ID weld repair residual stresses. Hydrostatic test pressure and operating conditions are then applied to the model. Once the residual stress condition in the nozzle to safe end weld is established, the MSIP application is then modeled to evaluate the redistribution of stresses in the weld region. The analysis results show that an unfavorable residual tensile stress resulting from ID weld repair exists along the inner region of the nozzle to safe end weld and is converted to a compressive stress field through application of MSIP.

Pre-MSIP ISI of the Outlet Nozzle to Safe End Weld 29-RC- 1140-1 (#14 Hot Leg nozzle/safe end weld) detected an indication at the buttering interface. The indication was characterized as an ID connected circumferential flaw that was about 2% in circumferential length and 24%

through-wall in depth [Ref. 2].

These results also verify that the application of MSIP would generate compressive stresses such that the flaw would *remainý in the compressive zone post- MSIP during operation.

3.1 AS-WELDED STRESSES The resulting residual stresses from an ID weld repair to the nozzle to safe end are illustrated in Figures 5 and 6. This stress pattern produces conservative high tensile residual stresses at the ID of the material. Figure 7 shows the path along which the through-wall stresses are plotted for the flaw evaluation. Though the indication is a circumferential flaw both the pre-MSIP hoop and axial through-wall stresses are shown in Figure 8 and tabulated in Table 2.

3.2 POST-MSIP STRESSES The post-MSIP stresses are shown in Figures 9 and 10. While these results correspond to the squeeze applied to the #14 hot leg nozzle/safe-end weld, they are typical for the range specified for the other Salem nozzle welds.

The most important observation to be gained from these figures is the general redistribution of stresses produced by MSIP. These results clearly show that the large residual tensile stresses along the inner region of the Inconel weld, induced by the welding process, are replaced by compressive stresses. These compressive hoop and axial residual stresses generated by MSIP will provide mitigation against PWSCC in the nozzle to safe end weld.

The post-MSIP hoop and axial through-wall stresses including operating pressure and temperature are shown in Figure 11 and tabulated in Table 3. These results confirm that the identified flaw will remain in the compressive zone at operation.

4418-6-001 5 ofl13

Nu Vision Engineering,Inc.

Salen Hot Leg (Outlet) Nozzle - Mid Pos, Nan Sqz Figure 1 - Finite Element Model AN Figure 2 - Finite Element Model - Weld Detail 4418-6-001 6 of 13

Nu Vision Engineering,Inc.

AN~ ~':

SalEn Hot Leg (Outlet) Nozzle - Mid Pos, NamxSqz Figure 3 - Finite Element Model - Boundary Conditions w/ Tool Pressure Loading AN*V MAT I

" ~ ~ MAT~4 MA8 x

SOM Salem Hot Leg (Outlet) Nozzle - Mid Pos, Homn Sqz Figure 4 - Finite Element Model - ANSYS Material Number 4418-6-001 7 ofl13

Nu Vision Engineering,Inc.

AN 1Y-T 19 2007 08:42:47 PLOT NO. 18 NODAL SOLUTIQN STEP=6 SUB =6 TIME=6 SY (AVG)

RSYS=0 PowerGraphics EFACET=I AVRES=Mat DMX=. 010018 SMN=-26477 SMX =63293

-80000

-60000

-40000

-20000 0

20000 40000 60000 80000 Saler Hot Leg (Outlet) Nozzle - Mid aos, NcmnSqz Figure 5 - NSE Weld - As-Designed Weld w/Local IDW eld Repair Residual Axial Stres 5S

,AN'>!-): CT 19 2007 08:42:48 PLOT NO. 19 NFAL S=TIC STEP=-6 SUB =6 SZ (AVG)

RSYS=0 PowerGrapic s EFACET=Il AVPZS=5Mat

-6X

ý.010018 SHN =-26638 SWX --80774

_-0000

-60000

-40000

-20000 0

20000 40000 60000 80000 Sale_ Hot Leg (Outlet) Nozzle - Mid Pos, Nam Sqz Figure 6 - NSE Weld - As-Designed Weld w/Local IDWeld Repair Residual Hoop Stress 4418-6-001 8 of 13

Nu Vision Engineering,Inc.

AN ~'

Path Start SalEm Hot Leg (Outlet) Nozzle - Mid Pos, Ncm Sqz Figure 7 - Salem Hot Leg, Path Along Flaw Location 4418-6-001 9 of 13

Nu Vision Engineering,Inc.

AN (x10**3) 80 40-a

-40r

-60

-80 01 2

.5 1.5 2.5 Distance Along Prep, from ID (in Salem Hot Leg (Outlet) Nozzle - Mid Pos, Nan Sqz Figure 8 - Salem Hot Leg, Stress Along Flaw Path, PRE MSIP Table 2 - Salem Hot Leg, Stress Along Flaw Path, PRE MSIP PRE MSIP Distance Axial Hoop (in) (psi) (psi) 0.000 36,840 63,680 0.137 35,507 63,979 0.274 30,249 64,860 0.411 -1,445 44,486 0.548 -11,735 13,586 0.685 -8,893 -100 0.822 -3,263 4,780 0.958 -1,772 6,932 1.095 -3,295 9,060 1.232 -5,512 11,412 1.369 -9,763 13,120 1.506 -17,675 13,978 1.643 -23,517 17,840 1.780 -19,326 28,342 1.917 -13,924 40,834 2.054 -8,911 48,676 2.191 -1,337 53,315 2.328 11,938 59,630 2.465 24,391 64,479 2.601 29,302 65,009 2.738 22,724 63,573 4418-6-001 10 of 13

Nu Vision Engineering,Inc.

AN .. COT 19 2007 08:42:53 PLOT NO. 55 N*DAL SOLU3TICN STEP=-32 SUB =6 TIME=32 SY (AVG)

RSYS=0 PowerGraphics EFACET=l1 AVRES=Mat DX =.318868 A' SWsT=-73127 SMX =63457

-80000

-60000

-40000

-20000 2000 20000 40000 80000 Salmn Hot Leg (Outlet) Nozzle Mid Pos, Nm Sqz Figure 9 - Post MSIP Residual Axial:Stress '- Nom. Position AN OCT 19 2007 08:42:54 PLOT NO. 56 N*DAL SCI..TICN STEXI=-32 SUB =6 TIM*=32 SZ (AVG)

RSYS=0 PowerGraphics EFACET:l AVRES=Mat DMX =. 318868 SW =-68552 SMX =50465

-80000

-60000

-40000

-20000 0

20000 m* 40000 60000 80000 Salem Hot Leg (Outlet) Nozzle - Mid Pos, Nan Sqz Figure 10 - Post MSIP Residual Hoop Stress - Nom. Position 4418-6-001 11 of 13

Nu Vision Engineering,Inc.

AN (x10**3) 80 60 406

-)

U)n -206

-6

-406

.5 Distance Along Prep, 1.5 from ID (in 2.5 Salem Hot Lea (Outlet) Nozzle - Mid Pos, Ncm Sqz Figure 11 - Salem Hot Leg, Stress Along Flaw Path, POST MsIP w/ Operating Pressure and Temperature Table 3 - Salem Hot Leg, Stress Along Flaw Path, POST MSIP w/ Operating Pressure and Temperature POST MSIP wI Operating Distance Axial Hoop (in) (psi) (psi) 0.000 -53,318 -11,915 0.137 -40,423 -8,181 0.274 -36,902 -2,443 0.411 -34,400 -4,026 0.548 -34,258 -13,268 0.685 -32,135 -19,569 0.822 -27,644 -22,323 0.958 -18,625 -19,284 1.095 -9,468 -12,683 1.232 -815 -5,942 1.369 5,843 20 1.506 9,730 4,405 1.643 18,503 9,269 1.780 35,023 20,609 1.917 49,771 33,688 2.054 55,587 42,425 2.191 54,204 45,436 2.328 50,385 44,880 2.465 44,842 41,465 2.601 38,896 37,931 2.738 34,774 35,286 4418-6-001 12 of 13

Nu Vision Engineering,Inc.

4.0 REFERENCES

1. Analytical Verification of MSIP for RV Hot Leg Nozzle to Safe End Weld, Salem Unit 1 & 2, NuVision Engineering, Inc. Report 4418-4-002-00

2. WESDYNE Letter, WDI-PJF-1303981-TR-010, Rev. 0, October 26, 2008 4418-6-001 13 of 13