Weld Residual Stress Finite Element Analysis Validation Introduction and Overview June 14-15, 2011, Rockville, MD Paul Crooker EPRI
Howard Rathbun U.S. Nuclear Regulatory Commission Introduction
• Welcome
• Introduction by meeting attendees – Names and affiliations
• Agenda – Revised since Public Meeting Announcement – Hardcopy available
• Review revised agenda
© 2010 Electric Power Research Institute, Inc. All rights reserved. 2 This is a Category 2 Public Meeting
• Category 1 – Discussion of one particular facility or site • Category 2 – Issues that could affect multiple licensees • Category 3 – Held with representatives of non-government organizations, private citizens or interested parties, or various businesses or industries (other than those covered under Category 2) to fully engage them in a discussion on regulatory issues
© 2010 Electric Power Research Institute, Inc. All rights reserved. 3 Program Overview
•Scientific Weld Specimens •Fabricated Prototypic Nozzles •Phase 1A: Restrained Plates (QTY 4) •Type 8 Surge Nozzles (QTY 2) •Phase 1B: Small Cylinders (QTY 4) •Purpose: Prototypic scale under controlled NRC EPRI - - •Purpose: Develop FE models. conditions. Validate FE models. Phase 2 Phase Phase 1 Phase
•Plant Components •Plant Components •WNP-3 S&R PZR Nozzles (QTY 3) •WNP-3 CL Nozzle (QTY 1) •Purpose: Validate FE models. •RS Measurements funded by NRC EPRI EPRI -
- •Purpose: Effect of overlay on ID. Phase 3 Phase Phase 4 Phase
© 2010 Electric Power Research Institute, Inc. All rights reserved. 4 Goals of the Meeting
• Focus on finite element modeling techniques
– What works well, what doesn’t • Allow meeting participants to express their views • Day 1: Present modeling and measurement results • Day 2: Discuss the implications of the findings • Present plans for documentation • Future work opportunities
© 2010 Electric Power Research Institute, Inc. All rights reserved. 5 Welding Residual Stress Validation Program Phase 1 Summary
Presented To: Welding Residual Stress Validation Experts Meeting
Presented By: John Broussard Dominion Engineering, Inc.
June 14, 2011
12100 Sunrise Valley Dr. #220 Reston, VA 20191 703.657.7300 www.domeng.com Phase 1 Weld Specimen Design Phase 1A Phase 1B Restrained Plates Cylinders . Fabricate and measure RS in simple Geometry 0.6" Thk, 0.4" Groove, 14" Long 6.51" OD x 0.47" Thick experimental specimens in order to Simplest Geometry Increasing Specimen Complexity develop and refine FE models Base Metal: 304L SS Base Metal: 304L SS, CS . Scientific Design Approach: Material Weld Metal: Alloy 82 Weld Metal: Alloy 82 – Investigate how weld parameters and Buttering: Alloy 82 Fix Geometry Fix Weld Parameters geometry affect RS distribution Variables – Representative weld configurations Vary Weld Parameters Vary Configuration – Controlled fabrication P-3: Base Case - 11 passes C-1: SS to SS – Maximize region of fully-developed stress P-4: Incr. Heat Input - 7 passes C-3: Buttered CS to SS w/ PWHT Specimens – Facilitate multiple RS measurements P-5: Incr. Heat Input - 7 passes C-4: Buttered CS to SS w/o PWHT . Investigate influence on RS state of: P-6: Decr. Heat Input - 23 passes C-5: Add SE and Repair to C-4 Aluminum backing plate is fairly Machine beam windows allow – Phase 1A: weld parameter variation Allow ND transparent to neutrons measurement of hoop strain • Current and wire feed rate Fixture design preserves original C-6 has manual 90° 75% TW ID • Torch speed Misc. – Phase 1B: weld configuration stress state. Repair Weld. Groove machined. • Buttering • Post-Weld Heat Treatment (PWHT) • Safe-End and SS weld • Repair Weld
2 Phase 1 Analysis Summary Measurements Performed Phase 1A Plates
Measured Specimens RS Measurement Method Vendor Location Directions Measured P-3 P-4 P-5 P-6 Longitudinal Neutron Diffraction: ORNL 45 Point Grid on Cross-section Plane Transverse XXXX Basic Measurements Normal
Neutron Diffraction: 7 Depths down Weld Centerline ORNL 6 Directions X Full Strain Tensor 2 Depths in Base Metal
Longitudinal Neutron Diffraction: 8 Longitudinal Locations, ORNL Transverse X Longitudinal Traverse 3 Depth in WM & 3 Depths in BM Normal Longitudinal U of Deep Hole Drilling 1 Hole through Centerline of Weld Transverse X Bristol In-plane Shear
7 Surface Points Across Weld Longitudinal X-ray Diffraction TEC XXXX On Topside of Specimen Transverse
Longitudinal 7 Surface Points Across Weld Surface Hole Drilling LTI Transverse XX X On Topside of Specimen In-plane Shear 2 Longitudinal Positions, Longitudinal Ring-Core LTI Both down Weld Centerline Transverse XX on Topside of Specimen In-plane Shear
Slotting Hill Eng. 4 Transverse Measurement Slots Transverse X
1 Longitudinal Measurement Slice Longitudinal Contour Hill Eng. X 1 Transverse Measurement Slice Transverse
ORNL: Oak Ridge National Laboratory TEC: Technology for Energy Corporation LTI: Lambda Technology, Inc. Hill Eng: Hill Engineering, LLC 3 Phase 1 Analysis Summary Plate RS Measurement Locations General Plate Arrangement (P-4 Shown)
Contour Transverse - one plate
Contour Longitudinal - full cross section - one plate
ND Locations - all plate specimens - longitudinal, transverse, normal directions
4 Phase 1 Analysis Summary Measurements Performed Phase 1B Cylinders
Measured Specimens RS Measurement Method Vendor Location Directions Measured C-1 C-3 C-4 C-5 C-5R Hoop Neutron Diffraction: ORNL 80 Point Grid on Cross-section Plane Axial XXXXX Basic Measurements Radial
Neutron Diffraction: Stepped every 5° for 90° in Normal Weld ORNL Axial Strain X Axisymmetry Study Stepped every 5° for 70° in Repair Weld
Hoop Deep Hole Drilling VEQTER 1 Hole through Centerline of Weld Axial XX XX In-plane Shear EPRI-CLT 10-11 Surface Points Across Weld Hoop X X X X-ray Diffraction & on OD & ID of Specimen Axial (EPRI) (EPRI) (TEC) TEC
2 Longitudinal (Hoop) Meas. Slices Hoop Contour Hill Eng. X 2 Transverse (Axial) Meas. Slices Axial
ORNL: Oak Ridge National Laboratory TEC: Technology for Energy Corporation LTI: Lambda Technology, Inc. EPRI-CLT: EPRI Charlotte Hill Eng: Hill Engineering, LLC
5 Phase 1 Analysis Summary Cylinder RS Measurement Locations General Cylinder Arrangement (C-3 Shown)
Contour Hoop Contour Axial Contour Axial - full cross section - one cylinder - one cylinder - one cylinder OD
ID
ND Locations DHD and iDHD - all cylinder specimens - all cylinder specimens - hoop, axial, radial directions 6 Phase 1 Analysis Summary Surface Stress Measurement Results Plate P-4, Weld Centerline
500 200 XRD
400 HD 100
300 0 RC Slitting 200 -100 100 RC -200 HD 0 Transverse Stresses (MPa) Stresses Transverse
Longitudinal Stresses (MPa) Stresses Longitudinal XRD X-ray Diffraction -300 Ring-Core -100 X-ray Diffraction Ring-Core Surface Hole Drilling Surface Hole Drilling Slitting -200 -400 -2.00 -1.50 -1.00 -0.50 0.00 0.50 1.00 -2.00 -1.00 0.00 1.00 2.00 3.00 4.00 Depth from Plate Top Surface (mm) Depth from Plate Top Surface (mm)
7 Phase 1 Analysis Summary Stress Measurements and FEA Predictions
. Results from Plate P-4 and Cylinder C-3 presented . Four sets of model results compared identified “A” through “D” – All models two dimensional plane strain (plate) or axisymmetric (cylinder) – Modeler’s best judgment for mesh, thermal inputs based on fabrication data – All models apply power generation as a function of time for thermal model • Roughly same amount of total energy input consistent with weld process – Model A: ANSYS, elastic perfectly-plastic hardening – Model B: ABAQUS, isotropic hardening law – Model C: ABAQUS, isotropic hardening law – Model D: ABAQUS, kinematic hardening law . Measurements performed at facilities identified “A” through “C” – No correlation between modelers “A” through “D” and facilities “A” through “C” – Facility A: contour method only – Facility B: neutron diffraction only – Facility C: contour method and neutron diffraction 8 Phase 1 Analysis Summary Plate P-4 Measurement and FEA Results
FEA Model A FEA Model B FEA Model A FEA Model B FEA Model C FEA Model D FEA Model C FEA Model D Contour, Facility A Contour, Facility C Contour, Facility A Neutron Diff, Facility B Neutron Diff, Facility B Neutron Diff, Facility C Neutron Diff, Facility C
600.0 300.0
500.0 200.0
400.0 100.0
300.0 0.0 200.0 -100.0 100.0
Longitudinal Mpa Stress, Longitudinal -200.0 0.0 Stress, Mpa Transverse
-100.0 -300.0
-200.0 -400.0 -4.00 -2.00 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 -4.00 1.00 6.00 11.00 16.00 Distance from Plate Top Surface, mm Distance from Plate Top Surface, mm
9 Phase 1 Analysis Summary Cylinder C-3 Measurement and FEA Results
FEA Model A FEA Model B FEA Model C FEA Model A FEA Model B FEA Model C Contour - A DHD / iDHD Neutron Diff, Facility B Contour - A DHD / iDHD Neutron Diff, Facility B 400.0
600.0 300.0
500.0 200.0
400.0 100.0
300.0 0.0 200.0 -100.0
100.0 Mpa Stress, Axial Hoop Stress, Stress, Mpa Hoop -200.0 0.0
-100.0 -300.0
-200.0 -400.0 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 Distance from OD, mm Distance from OD, mm
10 Phase 1 Analysis Summary Three Dimensional Plate Model
. Three dimensional model prepared for Plate P-4 . Progressive block dump for thermal model, five blocks lengthwise for each weld pass . Results are identical for the 2D/3D in the weld, with some differences far field in the longitudinal stresses
11 Phase 1 Analysis Summary Conclusions
. Surface stress measurements from the plates demonstrate that additional investigation is needed into surface measurements of welded locations . Bead geometry arrangements played a role in scatter among cylinder model stress results – Plate weld cavity did not change shape during welding consistent beads among modelers – Cylinder weld cavity changed shape during welding modelers assumptions for bead shape varied . Smaller, simpler models emphasize individual bead contributions – In some cases, such as C-3, lead to greater disparity in modeling results than for thicker cross sections
12 Phase 1 Analysis Summary
EPRI - NRC Welding Residual Stress Model Validation Program – Phase IV – Cold Leg Nozzle Optimized Weld Overlay Doug Killian AREVA NRC/Industry WRS FEA Validation Meeting Rockville, MD June 14-15, 2011 Outline
Description of OWOL Mockup for Phase IV of EPRI/NRC Residual Stress Model Validation Program Welding Residual Stress Simulation Model Stress Results Sensitivity Studies Comparison with Measured Data Influence of Operating Conditions Summary
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 3 Description of OWOL Mockup – As-Received Reactor Vessel Nozzle
36-inch Reactor Vessel Cold Leg Nozzle with Cladding, Butter, DM Weld, Safe End, and Portion of Reactor Vessel
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 4 Description of OWOL Mockup – After Weld Overlay
Weld Overlay Applied to Nozzle, Modified Safe End, and Pipe Section
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 5 Description of OWOL Mockup
Process Steps DM weld included in as-received nozzle prior to preparation of mockup Assumed shop hydrostatic leak test performed prior to sectioning Mockup preparation included ID weld repair in DM weld and butter • 25% through-wall • 30-degree arc length Pipe section added to safe end by SMAW process with 308L weld metal Pre-weld overlay residual stress measurements by deep hole drilling* 0.8" (20mm) of weld overlay added in eight 0.1" (2.5mm) layers Final layer removed by machining Final residual stress measurements by deep hole drilling*
*Also incremental hole drilling and X-ray diffraction
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 6 Description of OWOL Mockup
Stainless Steel Pipe Carbon Steel Nozzle
Cladding Alloy 82 Barrier Alloy 52 ER309L Overlay Barrier Alloy Stainless 182 Alloy Steel DMW 182 PWSCC Weld Stainless Butter Susceptible Steel Material Safe End
Alloy 182 Repair Weld NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 7 Description of OWOL Mockup
58.3 in. (1481 mm) 20.1 in. (511 mm)
34.97 in. (888 mm) Diam. 29.13 in. (740 mm) Diam.
0.7 in. (17.8 mm) Thk. 2.92 in. (74.2 mm) Thk.
1.70 in.
3.92 in. (99.7 mm)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 8 Description of OWOL Mockup – Fabrication Sequence
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 9 Actual Fabrication Sequence
Layers 1, 2, and 3 could not be completed initially because the uphill slope was to great for the flood cup. Flood cup was re-designed after 3rd partial layer and first three layers were completed.
t = .940”
(Layers 1-8) (final)
Extra layer of 309L needed due to slag producing safe end buildup using SMAW or SAW
Lesson learned:
t = .165” after grinding Document welding process prior to numerical simulation.
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 10 Outline
Description of OWOL Mockup Welding Residual Stress Simulation Model Stress Results Sensitivity Studies Comparison with Measured Data Influence of Operating Conditions Summary
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 11 Welding Residual Stress Simulation Model
Two-dimensional Welding Parameters for Groove Welds axisymmetric, Welding Process SMAW uncoupled thermal and Welding Efficiency, η 0.75 mechanical finite Interpass Temperature 350 °F element analysis using ANSYS Current, I 100 amps Multi-pass welding Voltage, E 20 volts simulations with Heat Power, ηEI 1.5 kJ/sec volumetric heat source Travel Speed 3 in/min Heat-treated butter Welding Speed, v 0.05 in/sec added as stress-free material Heat Input 30 kJ/in DMW Passes 95* Typical welding parameters used for DMW Repair Passes 20* groove welds SSW Passes 150* * Assumed
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 12 Welding Residual Stress Simulation Model
Welding Parameters for Weld Overlay Dual Wire, High Deposition Rate Welding Parameters Welding Process GTAW Welding Efficiency, η 0.75 Interpass Temperature 350 °F Specified Modeled Layer / Material Passes Heat Input Passes Heat Input Barrier / ER309L 45 28.4 kJ/in 17 34.5 kJ/in Barrier / Alloy 82 4 28.4 kJ/in 1 34.5 kJ/in Barrier / Alloy 52 28 38.9 kJ/in 17 40.0 kJ/in 2nd Layer / Alloy 52 60 40.1 kJ/in 35 40.0 kJ/in 3rd Layer / Alloy 52 60 42.4 kJ/in 34 40.0 kJ/in Remaining / Alloy 52 60 44.3 kJ/in 31/32 45.0 kJ/in
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 13 Two-Dimensional Finite Element Model
Nozzle Pipe
~20,000 Nodes
Overlay
Cladding
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 14 Two-Dimensional Finite Element Model Without Repair Weld
Typical Bead in Overlay: 0.625 in. x 0.1 in. ( 15.9mm x 2.54mm )
DMW Stainless Safe Butter Steel End Weld
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 15 Two-Dimensional Finite Element Model With Repair Weld Dissimilar Metal Typical Bead: 0.18 in. x 0.15 in. Weld (4.6mm x 3.8mm)
25% Through- Wall Repair Weld
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 16 Outline
Description of OWOL Mockup Welding Residual Stress Simulation Model Stress Results Sensitivity Studies Comparison with Measured Data Influence of Operating Conditions Summary
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 17 Development of Axial Stresses
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 3 Development of Hoop Stresses
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 3 Through-Wall Axial Stresses After DMW After DMW Repair After Hydro After SSW After Layer 1 After Layer 2 After Layer 3 After Layer 4 After Layer 5 After Layer 6 After Layer 7 After Layer 8 After Machining
80000 60000 Before OWOL 40000 20000 0 -20000 After -40000 Axial Stress Through Through Stress Axial Center of Repair (psi) Repair of Center OWOL -60000 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 20 Through-Wall Hoop Stresses After DMW After DMW Repair After Hydro After SSW After Layer 1 After Layer 2 After Layer 3 After Layer 4 After Layer 5 After Layer 6 After Layer 7 After Layer 8 After Machining
80000 Before OWOL 60000 40000 20000 0 -20000 -40000 After OWOL Center of Repair (psi) Repair of Center Hoop Stress Through Through Stress Hoop -60000 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 21 Pre- & Post -Overlay Stresses Distance from Inside Surface (mm) 0.0 6.4 12.7 19.1 25.4 31.8 38.1 44.5 50.8 57.2 63.5 69.9 76.2 82.6 88.9 95.3 80000 551.6
60000 413.7
40000 275.8
20000 137.9
Stress (psi) 0 0.0 Stress (MPa)
Post-WOL Hoop Stress -20000 -137.9 Pre-WOL Hoop Stress Post-WOL Axial Stress -40000 Pre-WOL Axial Stress -275.8
-60000 -413.7 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 22 Inside Surface Axial Stresses After DMW After DMW Repair After Hydro After SSW After Layer 1 After Layer 2 After Layer 3 After Layer 4 After Layer 5 After Layer 6 After Layer 7 After Layer 8 After Machining Safe End / DMW DMW / Butter Butter / Nozzle 80000 60000 40000 20000 0 -20000 -40000 -60000 Inside Surface Axial Stress (psi) Stress Axial Surface Inside -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 Distance from Center of DMW (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 23 Inside Surface Hoop Stresses After DMW After DMW Repair After Hydro After SSW After Layer 1 After Layer 2 After Layer 3 After Layer 4 After Layer 5 After Layer 6 After Layer 7 After Layer 8 After Machining Safe End / DMW DMW / Butter Butter / Nozzle 80000 60000 40000 20000 0 -20000 -40000 -60000 Inside Surface Hoop Stress (psi) Stress Inside Hoop Surface -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 Distance from Center of DMW (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 24 Outline
Description of OWOL Mockup Welding Residual Stress Simulation Model Stress Results Sensitivity Studies Comparison with Measured Data Influence of Operating Conditions Summary
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 25 Sensitivity Studies
Look at Sensitivity of Results to Changes in: Yield strength • Standard values • Values for Alloy 182 and Alloy 52 welds increased by 10% Hardening model • Kinematic • Isotropic Mesh refinement • Fine model with 20,000 nodes - DMW passes = 95 - Repair passes = 20 - SSW passes = 150 - Overlay passes = 261 • Coarse model with 6600 nodes - DMW passes = 88 - Repair passes = 19 - SSW passes = 103 - Overlay passes = 241
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 26 Pre-Overlay Stresses – Sensitivity to Yield Strength Distance from Inside Surface (mm) 0.0 6.4 12.7 19.1 25.4 31.8 38.1 44.5 50.8 57.2 63.5 69.9 76.2 80000 551.6
60000 413.7
40000 275.8
20000 137.9
Stress (psi) 0 0.0 Stress (MPa)
-20000 FEA Axial Stress -137.9 FEA Hoop Stress FEA Axial Stress with +10% Yield -40000 -275.8 FEA Hoop Stress with +10% Yield
-60000 -413.7 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 27 Pre-Overlay Stresses – Sensitivity to Hardening Model Distance from Inside Surface (mm) 0.0 6.4 12.7 19.1 25.4 31.8 38.1 44.5 50.8 57.2 63.5 69.9 76.2 80000 551.6
60000 413.7
40000 275.8
20000 137.9
Stress (psi) 0 0.0 Stress (MPa)
-20000 Axial Stress w/ Kinematic Hardening -137.9 Hoop Stress w/ Kinematic Hardening Axial Stress w/ Isotropic Hardening -40000 -275.8 Hoop Stress w/ Isotropic Hardening
-60000 -413.7 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 28 Pre-Overlay Stresses – Sensitivity to Mesh Refinement Distance from Inside Surface (mm) 0 6.35 12.7 19.05 25.4 31.75 38.1 44.45 50.8 57.15 63.5 69.85 76.2 80000 551.6
60000 413.7
40000 275.8
20000 137.9
Stress (psi) 0 0 Stress (MPa)
-20000 Axial Stress w/ Fine Mesh -137.9 Hoop Stress w/ Fine Mesh Axial Stress w/ Coarse Mesh -40000 -275.8 Hoop Stress w/ Coarse Mesh
-60000 -413.7 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 29 Outline
Description of OWOL Mockup Welding Residual Stress Simulation Model Stress Results Sensitivity Studies Comparison with Measured Data Influence of Operating Conditions Summary
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 30 Residual Stresses Before OWOL Distance from Inside Surface (mm) 0.0 6.4 12.7 19.1 25.4 31.8 38.1 44.5 50.8 57.2 63.5 69.9 76.2 80000 551.6
60000 413.7
40000 275.8
20000 137.9
Stress (psi) 0 0.0 Stress (MPa)
-20000 DHD Axial Data at 180 Deg. -137.9 DHD Hoop Data at 180 Deg. FEA Axial Stress FEA Hoop Stress -40000 IHD Axial Data at 180 Deg. -275.8 IHD Hoop Data at 180 Deg.
-60000 -413.7 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 31 Residual Stresses After OWOL Distance from Inside Surface (mm) 0.0 6.4 12.7 19.1 25.4 31.8 38.1 44.5 50.8 57.2 63.5 69.9 76.2 82.6 88.9 95.3 80000 551.6
60000 413.7
40000 275.8
20000 137.9
Stress (psi) 0 0.0 Stress (MPa)
-20000 DHD Axial Data at 180 Deg. -137.9 DHD Hoop Data at 180 Deg. FEA Axial Stress FEA Hoop Stress -40000 IHD Axial Data at 180 Deg. -275.8 IHD Hoop Data at 180 Deg.
-60000 -413.7 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 32 Outline
Description of OWOL Mockup Welding Residual Stress Simulation Model Stress Results Sensitivity Studies Comparison with Measured Data Influence of Operating Conditions Summary
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 33 Post-Overlay Stresses at Operation Distance from Inside Surface (mm) 0.0 6.4 12.7 19.1 25.4 31.8 38.1 44.5 50.8 57.2 63.5 69.9 76.2 82.6 88.9 95.3 80000 551.6
60000 413.7
40000 275.8
20000 137.9
Stress (psi) 0 0.0 Stress (MPa) Stress
Post-WOL Hoop Stress -20000 -137.9 Operating Hoop Stress Post-WOL Axial Stress -40000 Operating Axial Stress -275.8
-60000 -413.7 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 34 Post-Overlay Stresses at Operation (2) Distance from Inside Surface (mm) 0.0 6.4 12.7 19.1 25.4 31.8 38.1 44.5 50.8 57.2 63.5 69.9 76.2 82.6 88.9 95.3 80000 551.6
60000 413.7
40000 275.8
20000 137.9
Stress (psi) 0 0.0 Stress (MPa) Stress Post-WOL Hoop Stress Oper Temp+Pres Hoop -20000 -137.9 Post-WOL Axial Stress Oper Temp+Pres Axial -40000 Oper Temp Hoop -275.8 Oper Temp Axial
-60000 -413.7 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 Distance from Inside Surface (in.)
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 35 Outline
Description of OWOL Mockup Welding Residual Stress Simulation Model Stress Results Sensitivity Studies Comparison with Measured Data Influence of Operating Conditions Summary
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 36 Summary
Demonstrated effectiveness of optimized weld overlays in reducing inside surface stresses in large bore pipe nozzles Completed two-dimensional simulation of welding residual stresses in reactor vessel nozzle optimized weld overlay mockup for Phase IV of EPRI/NRC Residual Stress Model Validation Program Achieved reasonable agreement with measured data Three-dimensional analysis could be performed to more realistically simulate 30° repair weld
NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 37 NRC/Industry WRS FEA Validation Meeting – Rockville, MD – June 14-15, 2011 Page 38 NRC Weld Residual Stress Finite Element Analysis Validation
Phase IV Mockup FEA Experience
Francis H. Ku June 14, 2011 Outline • Phase IV OWOL Mockup Geometry • Initial FEA Using Standard SIA Procedures • Revised FEA Following Actual Weld Process and Sequence • Comparisons Between the Two FEA Evaluations • Observations and Lessons Learned
NRC WRS FEA Validation, Phase IV Experience Slide 2 www.structint.com 877-4SI-POWER EPRI/NRC Phase IV Mockup • RPV outlet nozzle mockup fabricated by EPRI – Nozzle from cancelled plant w/DMW to Safe End – ID repair performed (25% thru-wall, 30° arc) – SS spool piece added – OWOL installed by AREVA – Extensive RS measurements performed – sponsored by NRC
NRC WRS FEA Validation, Phase IV Experience Slide 3 www.structint.com 877-4SI-POWER Initial Finite Element Model and Analysis • 2D axisymmetric FE model • Assumed standard SIA practice at the time: – Continuous welding for each weld (no intermediate shutdown) – WOL performed in complete layers – No significant lag time between layers – Small weld beads from WSI tooling (~1/4” wide) • Isotropic hardening material model (cap at UTS)
NRC WRS FEA Validation, Phase IV Experience Slide 4 www.structint.com 877-4SI-POWER Actual WOL Bead Pattern and Weld Sequence
21.3"
6.0" 15 8 14 7 4.5" 9 5 3 2 13 12 11 10 6 0.775" 4 1
3.0" 2.92" 3.425"
RV Nozzle Safe End Attached Pipe
Sequence: 1) ER309L- Double Down 9) ALLOY 52M- Double Up- Completed Layer Three 2) ALLOY 82- Double Up Layer One: t=0.06" 10) ALLOY 52M- Double Up- Layer Four: t=0.05" 3) ALLOY 52M- Double Up 11) ALLOY 52M- Double Up- Layer Five: t=0.075" 4) ER309L- Double Down 12) ALLOY 52M- Double Up- Layer Six: t=0.075" 5) ALLOY 52M- Double Up Layer Two: t=0.065" 13) ALLOY 52M- Double Up- Layer Seven: t=0.1" 6) ALLOY 52M- Double Up- Layer Three: t=0.175" 14) ALLOY 52M- Double Up- Layer Eight: t=0.175" 7) ALLOY 52M- Double Up- Completed Layer One 15) ALLOY 52M- Double Up- Layer Nine 8) ALLOY 52M- Double Up- Completed Layer Two
NRC WRS FEA Validation, Phase IV Experience Slide 5 www.structint.com 877-4SI-POWER Revised Analysis to Reflect Actual Configuration • 2D axisymmetric FE model • Used actual mockup process by AREVA: – One layer per day (complete cool down after each layer) – First three layers performed in segments due to tooling issue – High deposition with large weld beads (~5/8” wide) • Hybrid isotropic hardening material model (plateau at flow stress)
NRC WRS FEA Validation, Phase IV Experience Slide 6 www.structint.com 877-4SI-POWER Axial Residual Stress Results Comparison • Both initial and revised FEA results compare OK to measurements • Revised FEA yields more conservative predictions • Indicates bead sequencing and heat inputs have small effects on axial stress predictions • Material model has marginal influence on axial results
NRC WRS FEA Validation, Phase IV Experience Slide 7 www.structint.com 877-4SI-POWER Hoop Residual Stress Results Comparison • Initial FEA predictions exceed the measurement bands • Revised FEA predictions match well with measurements • Accurate descriptions of weld settings are important • Material model a crucial role in hoop results
NRC WRS FEA Validation, Phase IV Experience Slide 8 www.structint.com 877-4SI-POWER ID Surface Residual Stress Results Comparison • FEA under-predicts post-WOL axial weld residual stress benefits Similar observations on other WOL geometries • FEA under-predicts post-WOL hoop weld residual stress benefits This geometry only Good FEA hoop predictions on other WOL geometries
NRC WRS FEA Validation, Phase IV Experience Slide 9 www.structint.com 877-4SI-POWER Observations and Lessons Learned • Typical 2-D FEA tend to under-predict axial weld residual stress benefits, especially near ID surface • Strain hardening law has significant effects on WRS predictions Pure isotropic hardening tend to over-predict WRS benefits Pure kinematic hardening tend to under-predict WRS benefits The hybrid isotropic hardening model (isotropic hardening capped at flow stress) seems to offer improvements over either hardening law • Bead sequencing and welding parameters have significant impacts on hoop WRS predictions Revised FEA following actual mockup process agrees well with measurements
NRC WRS FEA Validation, Phase IV Experience Slide 10 www.structint.com 877-4SI-POWER Cold Leg Nozzle Mock-Up Evaluation of Stress Reduction caused by Optimized Weld Overlay
6/14/11
Lee Fredette PhD, PE Research Leader Battelle, Systems Development 505 King Ave Columbus, OH 43201 e:[email protected] w: www.battelle.org p: 614.424.4462
1 Cold Leg Nozzle • Cold Leg Nozzle Stress Improvement Analysis – Residual Stresses with ID weld repair 25% through wall - Model Geometry - No Weld Overlay - After OWOL - After Operating Pressure and Temperature - Modeling Conclusions – Stress Measurements Comparison - With 25% ID Repair - Without ID Repair - Measurement Conclusions – OWOL Thickness Study - With 25% ID Repair - Without ID Repair - OWOL Thickness Conclusions – Overall Conclusions
2 Cold Leg Nozzle • Purpose: – Evaluate the effect of Optimized Weld Overlay (OWOL) in the Mock- Up Cold Leg Nozzle.
INCO 52M OWOL Dissimilar Metal Weld Area where PWSCC is a concern
3 Cold Leg Nozzle
• Cold Leg Nozzle Mock-Up Before OWOL
Cold Leg Nozzle
Image from EPRI Large Bore Pipe OWOL Mockup - Residual Stress Phase IV presentation at meeting with NRC 9/1/10
4 Cold Leg Nozzle
• Pipe and Weld Profile:
INCO 52M OWOL OD = 34.97’’
SS Buffer Layer INCO A508 INCO 82/182 Class 2 82/182 Weld SS Pipe Butter SS Cladding SS Safe End SS Weld 25% ID ID = 28.12’’ Repair • Boundary Conditions: • Convection equal on OD and ID surfaces in thermal analysis • Fixed on left end and free on right end during welding • Right end loaded with end cap load when operating pressure applied • Free to expand radial when operating temperature applied
58.34’’ 5 Cold Leg Nozzle
• Pipe and Weld Profile: 21.31’’
INCO 52M Optimized Weld Overlay in 239 passes right-to-left
0.765’’
Weld Bead Size (roughly) Isotropic hardening used in all analyses Butter, INCO, Repair = 0.1” x 0.2” Annealing temperature of 2,500oF used. SS Weld = 0.1” x 0.25” Buffer, OWOL = 0.1” x 0.625”
6 Cold Leg Nozzle • Cold Leg Nozzle Stress Improvement Analysis – Residual Stresses with ID weld repair 25% through wall - Model Geometry - No Weld Overlay - After OWOL - After Operating Pressure and Temperature - Modeling Conclusions – Stress Measurements Comparison - With 25% ID Repair - Without ID Repair - Measurement Conclusions – OWOL Thickness Study - With 25% ID Repair - Without ID Repair - OWOL Thickness Conclusions – Overall Conclusions
7 Cold Leg Nozzle
• Cold Leg Nozzle Stress Improvement Analysis – Axial Stress (ksi)
After INCO Weld
After SS Weld
After OWOL
Models: t_25_phase4_1.inp s_25_phase4_1.inp
8 Cold Leg Nozzle – Cold Leg Nozzle Weld Overlay – Axial Stress Improvement
9 Cold Leg Nozzle
• Cold Leg Nozzle Stress Improvement Analysis – Hoop Stress (ksi)
After INCO Weld
After SS Weld
After OWOL
Models: t_25_phase4_1.inp s_25_phase4_1.inp
10 Cold Leg Nozzle
• Cold Leg Nozzle Weld Overlay – Hoop Stress Improvement
11 Cold Leg Nozzle • Cold Leg Nozzle Stress Improvement Analysis – Residual Stresses with ID weld repair 25% through wall - Model Geometry - No Weld Overlay - After OWOL - After Operating Pressure and Temperature - Modeling Conclusions – Stress Measurements Comparison - With 25% ID Repair - Without ID Repair - Measurement Conclusions – OWOL Thickness Study - With 25% ID Repair - Without ID Repair - OWOL Thickness Conclusions – Overall Conclusions
12 Cold Leg Nozzle • Results –Axial Stresses operating pressure of 2,250 psi and temperature of 572oF applied
Operating
Improvement pressure and temperature causes little change before OWOL, but after OWOL reduces stresses to be all in compression in DMW area
13 Cold Leg Nozzle • Results – Hoop Stresses operating pressure of 2,250 psi and temperature of 572oF applied
Operating pressure and temperature Improvement increase hoop stresses both before and after OWOL application. OWOL stresses remain in compression in the DMW area
14 Cold Leg Nozzle • Cold Leg Nozzle Stress Improvement Analysis – Residual Stresses with ID weld repair 25% through wall - Model Geometry - No Weld Overlay - After OWOL - After Operating Pressure and Temperature - Modeling Conclusions – Stress Measurements Comparison - With 25% ID Repair - Without ID Repair - Measurement Conclusions – OWOL Thickness Study - With 25% ID Repair - Without ID Repair - OWOL Thickness Conclusions – Overall Conclusions
15 Cold Leg Nozzle • Modeling Conclusions:
• The OWOL reduces axial stress in DM weld when operating pressure and temperature are applied.
• The OWOL reduces hoop stress in DM weld both at room temperature and with operating pressure and temperature.
16 Cold Leg Nozzle • Cold Leg Nozzle Stress Improvement Analysis – Residual Stresses with ID weld repair 25% through wall - Model Geometry - No Weld Overlay - After OWOL - After Operating Pressure and Temperature - Modeling Conclusions – Stress Measurements Comparison - With 25% ID Repair - Without ID Repair - Measurement Conclusions – OWOL Thickness Study - With 25% ID Repair - Without ID Repair - OWOL Thickness Conclusions – Overall Conclusions
17 Cold Leg Nozzle
• Analysis results obtained with the finite element models are compared with measured results obtained by the deep hole drilling (DHD) method. – The mock-up had a partial arc repair covering a 30o arc on the ID of the dissimilar metal weld and 25% of the thickness deep. – Measurements were taken at one location in the repaired region and three places in un- repaired areas at 90o circumferential spaced locations. – Incremental deep hole drilling (iDHD - thought to be more accurate) was used in some locations DHD and iDHD performed by VEQTER Ltd, Bristol, UK
18 Cold Leg Nozzle
• Cold Leg Nozzle Stress Improvement Analysis – Axial Stress (ksi)
After SS Weld
After OWOL
Models: t_25_phase4_1.inp s_25_phase4_1.inp
19 Cold Leg Nozzle
• Cold Leg Nozzle Stress Improvement Analysis – Axial Stress with 25% ID repair
Good match between FEA results and Deep Hole Drilling results Not much of a change before and after OWOL in this cross section
20 Cold Leg Nozzle
• Cold Leg Nozzle Stress Improvement Analysis – Hoop Stress (ksi)
After SS Weld
After OWOL
Models: t_25_phase4_1.inp s_25_phase4_1.inp
21 Cold Leg Nozzle
• Cold Leg Nozzle Stress Improvement Analysis – Hoop Stress with 25% ID repair
Measurements show less improvement than FEA model DHD shows increase in stress with OWOL and iDHD shows reduction in stress with OWOL
22 Cold Leg Nozzle
• Cold Leg Nozzle Stress Improvement Analysis – Axial Stress without ID repair
Good match between FEA results and Deep Hole Drilling results Trends are similar, FEA initial spike in stress caused by root- pass grind and re-weld probably not done in actual nozzle.
23 Cold Leg Nozzle
• Cold Leg Nozzle Stress Improvement Analysis – Hoop Stress without ID repair
Good match between FEA results and Deep Hole Drilling results Trends are similar, FEA initial spike in stress caused by root- pass grind and re-weld probably not done in actual nozzle.
24 Cold Leg Nozzle • Measurement Conclusions:
• There is a good comparison between the deep hole drilling measurement, the incremental deep hole drilling measurements and the finite element model results.
• The axial stress measurements match the FEA predictions better than the hoop stress measurements.
25 Cold Leg Nozzle • Cold Leg Nozzle Stress Improvement Analysis – Residual Stresses with ID weld repair 25% through wall - Model Geometry - No Weld Overlay - After OWOL - After Operating Pressure and Temperature - Modeling Conclusions – Stress Measurements Comparison - With 25% ID Repair - Without ID Repair - Measurement Conclusions – OWOL Thickness Study - With 25% ID Repair - Without ID Repair - OWOL Thickness Conclusions – Overall Conclusions
26 Cold Leg Nozzle • Overall Conclusions:
• FEA predictions showed OWOL caused both axial and hoop stress improvement at operating pressure and temperature conditions. • Deep hole drilling stress measurements compare well with FEA predictions and lend believability to further sensitivity studies with the FEA model. • OWOL thickness studies showed a thickness at which greater weld overlay thickness did not further change the room temperature stress profile. • Similar studies should be performed when designing an OWOL for a specific geometry.
27 Cold Leg Nozzle
Questions?
28 Residual Stress Analysis for Welding Process with ABAQUS
Model Building: Geometry and Mesh for Single Pass or Multiple Passes User subroutine for heat flux generation In weld ABAQUS - DFLUX
Transient Thermal Model to Calculate Temperature Histories
Apply BC, Materials, hardening Rules, etc for weld Deposition
Calculate Stress using temperature histories as transient thermal load
1 • Goldak Exponential Heat Source
3ηEI −− 22 q = e 0 /)(3 Ttt π w )( TvA
q --- Energy Density; E --- Welding Voltage η --- Arc Efficiency; I --- Welding Current
Aw --- Area of Welding Beads; v --- Welding Speed T --- Welding Duration; t --- Welding Time
• Thermal Properties – Thermal Conductivity Usually Temperature – Specific Heat Dependent – Mass Density
2 ANALYSIS MODEL
Predicted fusion zone
• Butter – 27 passes • Axi-symmetric model • DMW – 29 passes • 4-node linear element • SS – 51 passes • 25% repair • Overlay – 103 passes 3 Analysis Procedure
• Software: ABAQUS • Thermal Model: DFLUX (1-D Goldak). Also did case using • Structural: ABAQUS Library (Chaboche Models: isotropic, nonlinear kinematic, mixed Chaboche – with ABAQUS abrupt annealing • (Note: No phase transformation or gradual annealing here (requires VFT UMAT)
Steps: 1. Model Butter (About 4-5 Beads Lumped into Each Pass) 2. PWHT (via Norton Creep) 3. Main Weld (About 3-4 Beads Lumped into Each Pass) 4. Weld Repair (25%) 5. SS Weld (Lumped Beads) 6. OWOL Weld (Three Layers)
4 MECHANICAL STRESS ANALYSIS
Isotropic
Mix
Room Temperature No Service Load
5 Non-Linear Kinematic MECHANICAL STRESS ANALYSIS
Isotropic
Mix
Room Temperature 6 Non-Linear Kinematic No Service Load Line Plots (Path 1) After Repair
7 Line Plots (Path 2) After Repair
8 Line Plots (Path 3) After Repair
9 Line Plots (Path 1) After SS Weld
10 Line Plots (Path 2) After SS Weld
11 Line Plots (Path 3) After SS Weld
12 Line Plots (Path 1) After OWOL
13 Line Plots (Path 2) After OWOL
14 Line Plots (Path 3) After OWOL
15 Back-up
16 A508 nozzle stress- strain curve data and thermal data
17 A182 nozzle stress- strain curve data and thermal data 700 A182 600
500
400
22 300 316 538 True Stress (MPa) Stress True 200 760 982 100 1300
0 0 0.05 0.1 0.15 0.2 0.25 Plastic strain
18 A182 nozzle stress- strain curve data and thermal data
19 Materials – Nozzle A508 Nonlinear Kinematic Lemaitre-Chaboche
700.0
600.0
500.0
400.0 20C (Multi NLKIN)
True Stress (MPa) Stress True 316C (Multi NLKIN)
538C (Multi NLKIN) 300.0 760C (BE-Multi NLKIN)
760C (EMC2-MultiNLKIN)
200.0 982C (Multi-NLKIN)
100.0
0.0 0 0.02 0.04 0.06 0.08 0.1 0.12
Plastic Strain
Yield C1 γ1 C2 γ2 Temp. Stress (MPa) (oC) 421.00 136101.7 5130.9 7308.9 53.2 20 341.16 255001.9 5130.9 10671.5 53.2 316 235.33 508928.3 5130.9 5589.7 53.2 538 8.47 215895.3 5130.9 2053.8 53.2 760 21.97 14497.0 5130.9 1034.3 53.2 982 20 Materials – Safe-End and Pipe: SS316 Mixed Lemaitre-Chaboche
450.0
400.0
350.0
300.0
250.0 True Stress (MPa) Stress True
200.0
150.0 20C (Multi NLKIN)
275C (Multi NLKIN) 100.0 550C (Multi NLKIN)
750C (Multi NLKIN) 50.0
0.0 0 0.05 0.1 0.15 0.2
Plastic Strain
Yield C1 γ1 C2 γ2 Q∞ b Temp. Stress (MPa) (oC) 149.01 156435.0 1410.85 6134 47.19 130.0 6.9 20 91.18 100631.0 1410.85 5568 47.19 161.1 6.9 275 103.87 64341.0 1410.85 5227 47.19 137.6 6.9 550 68.82 56232.0 1410.85 4108 47.19 60.5 6.9 750
21 Materials – Weld/Butter A182 Mixed Lemaitre-Chaboche
700.0 22C (Multi NLKIN)
316C (Multi NLKIN)
600.0 538C (Multi NLKIN)
760C (BE-Multi NLKIN)
500.0 982C (Multi-NLKIN)
400.0 True Stress (MPa) Stress True
300.0
200.0
100.0
0.0 0 0.05 0.1 0.15 0.2
Plastic Strain
Yield C1 γ1 C2 γ2 Q∞ b Temp. Stress (MPa) (oC) 206.80 254865.1 2496.2 1731 3.956 83.5 20.0 22 161.00 192627.6 2496.2 2304.56 3.956 75.3 20.0 316 171.00 104132.1 2496.2 2571.2 3.956 83.2 20.0 538 144.70 131679.3 2496.2 2668.4 3.956 0.0 20.0 760 34.80 179992.6 2496.2 623.6 3.956 0.0 20.0 982 22 Phase II International WRS Round Robin Data Analysis
Matthew Kerr, PhD Component Integrity Branch Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission
Lee Fredette, PhD, PE Equipment Development & Mechanical Systems Battelle Memorial Institute Outline
• Brief overview of WRS Validation Program
• Typical steps in FE element analysis • Geometry/Fabrication • International FE WRS Round Robin analysis requirements
• Results from FE Round Robin
• Industry Flaw Evaluation Guidelines WRS Model Validation Program Joint NRC/EPRI WRS Validation Program • Background: – Component integrity analyses for PWSCC in DM welds showed that the results were highly dependent upon WRS profiles – ACRS letter dated 10/19/07 supported further WRS research • Purpose: – Refine WRS FE model development for 82/182 DM welds through sequential development from Phase 1 to 4 – Develop reasonable assurance that WRS FE models are defensible through a blind validation using well controlled mockups to various WRS measurement testing techniques • Expected Outcome: – Validation of WRS FE models using well controlled mockups focusing on through-wall axial & hoop stresses – Develop uncertainty distributions in WRS modeling
3 WRS Model Validation Program Joint NRC/EPRI WRS Validation Program • Joint NRC/EPRI WRS Validation Program: – Approved MOU Addenda (2009) – Phases 1-4 with specific goals and progressively more in-plant representative conditions – Validate WRS models to experimental measurements – Inform and improve WRS predictions for component integrity calculations • Joint NRC/EPRI Development of xLPR Code: – Evaluate piping integrity with validated WRS profiles – Assess variability of WRS (mean, scatter, and distribution)
4 WRS Model Validation Program Joint NRC/EPRI WRS Validation Program
• Phase 2: PZR Surge Mockups (NRC-led, International WRS)
• Validate WRS FE analysis methods with • Prototypic components and fabrication history • International FE Round Robin • Residual stress measurments Phase 2 Participants International WRS Round Robin
• ANSTO (Australia) • AREVA (USA and EU) • Battelle (USA) • Dominion Engineering (USA) • Goldak Technologies (Canada) • ESI Group (USA) • EMC2 (USA) • Inspecta Technology (EU) • Institute of Nuclear Safety System (Japan) • Osaka University (Japan) • Rolls Royce (UK) • Structural Integrity Associates (USA) WRS Model Validation Program International WRS Round Robin - Analyses • Analysis 1: Prior to Safe-End Weld • Analysis 1a: User defined material properties, no Thermal Couple (TC) data • Analysis 1b: 1a + TC data • Analysis 1c: 1b + NRC Material Properties
• Analysis 2: Post Safe-End Weld WRS Model Validation Program International WRS Round Robin - Geometry • Pipe and Weld Geometry
Alloy 82 Butter
Alloy 82 Weld SS Safe End
SS Weld
SA-105
SS Cladding
SS Pipe
Fill-In Weld WRS Model Validation Program International WRS Round Robin - Geometry • Pipe and Weld Cross-Section
Alloy 82 Alloy 82 Butter Weld SA-105 SS Safe End SS Weld
SS Cladding SS Pipe
Fill-In Weld
Boundary Conditions: •Fixed axially on left end and free on right end •Equivalent convective cooling on both outer and inner diameter surfaces WRS Model Validation Program International WRS Round Robin - Geometry • Step 1: Butter, post weld heat treat, and machine
Alloy 82 Butter (137 Passes) WRS Model Validation Program International WRS Round Robin - Geometry • Step 2: Main DM Weld
Alloy 82 Weld (40 Passes) WRS Model Validation Program International WRS Round Robin - Geometry • Step 3: Machining followed by Fill-In Weld
Fill-In Weld Groove Machined
Alloy 82 Fill- In Weld (27 Passes) WRS Model Validation Program International WRS Round Robin - Geometry • Step 4: Machine Crown and Fill-In Welds WRS Model Validation Program International WRS Round Robin - Geometry • Step 5: Safe-End Weld
SS Safe End Weld (52 Passes) Phase 2 Surge Nozzle Data Analysis - Centerline • Comparison of Axial and Hoop Stress FE results to
iDHD (incremental Deep Hole Drilling)Stainless Steel Safe residual End Weld Completion stress measurements CG482478-211, Tie-In Weld Prep CG482478-212, Pipe Weld Prep CG482478-213, Tie-In Weld
Axial Stress
Hoop Stress Phase 2 Surge Nozzle Data Analysis Steps • Raw data provided by participants – Data provided with different number of divisions through the thickness for FE and Deep Hole Drilling (DHD) measurements – Calculate linearly interpolated data point for uniform 50 divisions through the normalized thickness – Calculate the average and standard deviation based on all the submitted data for each of the uniform divisions • student t-distribution for small sample size:
• N = number of items in the division. • x is the value at that division • x with a bar over it is the average value for that division • Plot curves for +/- 3 standard deviations Phase 2 Analysis 1a Axial Stress, pre safe -end
1000
A - MIXED 800 B - KIN 3 standard deviations C - ISO 600 C - KIN D - KIN* 400 E - ISO 200 E - MIXED E - KIN 0 F - ISO
Stress (MPa) Stress G - ISO -200 H - ISO I - ISO -400 I - KIN J - ISO -600 iDHD #1 residual stress measurements iDHD #2 -800 Average 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 3x Distance from ID (x/t)
* Elastic/plastic hardening with flow stress greater than annealed yield stress Phase 2 Analysis 1c Axial Stress, pre safe -end
1000
800 Some sharpening in dataset when compared to 1a, but B - KIN C - ISO 600 similar in average stress C - KIN D - ISO 400 E - ISO 200 E - MIXED E - KIN 0 F - ISO
Stress (MPa) Stress G - ISO -200 H - ISO I - ISO -400 I - KIN iDHD #1 -600 iDHD #2 Average -800 3x 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1c Axial Stress (ISO)
1000
800 ISO results develop greater max and min stress B - KIN 600 C - KIN E - KIN 400 I - KIN 200 E - MIXED C - ISO 0 D - ISO
Stress (MPa) Stress E - ISO -200 F - ISO G - ISO -400 H - ISO I - ISO -600 iDHD #1 iDHD #2 -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1c Axial Stress (KIN)
1000
800 Lower stresses developed C - ISO 600 D - ISO in the KIN results E - ISO 400 F - ISO G - ISO 200 H - ISO I - ISO 0 J - ISO Stress (MPa) Stress B - KIN -200 C - KIN E - KIN -400 I - KIN E - MIXED -600 iDHD #1 iDHD #2 -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1c Axial Stress (ISO vs KIN )
1000
800 ISO and KIN results for B - KIN 600 a given analysis do not bound the dataset D - ISO E - ISO 400 E - MIXED 200 E - KIN F - ISO 0 G - ISO
Stress (MPa) Stress H - ISO -200 I - ISO I - KIN -400 C - ISO C - KIN -600 iDHD #1 iDHD #2 -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1c Axial Stress (ISO vs KIN )
1000
800 MIXED hardening falls between B - KIN 600 ISO and KIN results C - ISO C - KIN 400 D - ISO 200 F - ISO G - ISO 0 H - ISO
Stress (MPa) Stress I - ISO -200 I - KIN E - ISO -400 E - MIXED E - KIN -600 iDHD #1 iDHD #2 -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1 Axial Stress Difference (1c – 1a)
1000 Outliers can be explained by stress differences resulting from 800 changes in thermal model (1b) 600 and material properties (1c) B - KIN 400 C - ISO C - KIN 200 E - ISO E - MIXED 0 E - KIN Stress (MPa) Stress F - ISO -200 G - ISO I - ISO -400 I - KIN Average -600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1 Axial Stress Difference (1b – 1a), thermal model
1000 Only the thermal models have 800 been changed based on TC data 600
B - KIN 400 D - ISO 200 E - ISO E - MIXED 0 E - KIN
Stress (MPa) Stress F - ISO -200 G - ISO I - ISO -400 I - KIN Average -600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1 Axial Stress Difference (1b – 1a), thermal model
1000 Heat input significantly 800 underestimated
600
400 B - KIN D - ISO 200 E - ISO E - MIXED 0 E - KIN Stress (MPa) Stress F - ISO -200 G - ISO I - ISO -400 I - KIN
-600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1 Axial Stress Difference (1b – 1a), thermal model
1000 Outlier for 1a analysis, material 800 property differences also 600 show a large effect for this analysis
400 B - KIN 200 D - ISO E - ISO 0 E - MIXED
Stress (MPa) Stress E - KIN -200 F - ISO G - ISO -400 ID stresses are stable J - ISO
-600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1 Axial Stress Difference (1b – 1a), thermal model
1000
800 Feature indicates that zero stress
600 is crossed at a different location
ID400 stresses are stable B - KIN 200 D - ISO E - MIXED 0 F - ISO Stress (MPa) Stress G - ISO -200 E - ISO E - KIN -400
-600 Kinematic results appear less sensitive to changes in heat input -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1 Axial Stress Difference (1c – 1b), material prop.
1000 Updated thermal model, only 800 material properties changed 600 B - KIN 400 C - ISO C - KIN 200 E - ISO E - MIXED 0 E - KIN Stress (MPa) Stress F - ISO -200 G - ISO I - ISO -400 I - KIN Average -600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1a Hoop Stress, pre safe -end
1000 A - MIXED 800 B - KIN C - ISO 600 C - KIN D - KIN* 400 E - ISO E - MIXED 200 E - KIN F - ISO 0 G - ISO Stress (MPa) Stress H - ISO -200 I - ISO I - KIN -400 Hoop dataset is more scattered J - ISO than axial dataset iDHD #1 -600 iDHD #2 Average -800 3x 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 3x Distance from ID (x/t)
* Elastic/plastic hardening with flow stress greater than annealed yield stress Phase 2 Analysis 1c Hoop Stress, pre safe -end
1000
800 B - KIN C - ISO 600 C - KIN D - ISO 400 E - ISO E - MIXED 200 E - KIN F - ISO 0 G - ISO Stress (MPa) Stress H - ISO -200 I - ISO I - KIN -400 iDHD #1 Some sharpening of the data, iDHD #2 -600 but similar scatter to 1a Average 3x -800 3x 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1c Hoop Stress (ISO)
1000
800
B - KIN 600 C - KIN E - KIN 400 I - KIN E - MIXED 200 C - ISO D - ISO 0
Stress (MPa) Stress E - ISO ISO results develop higher stresses F - ISO -200 G - ISO -400 H - ISO I - ISO -600 iDHD #1 iDHD #2 -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1c Hoop Stress (KIN) g , y 1000
800
600 C - ISO D - ISO 400 E - ISO F - ISO 200 G - ISO H - ISO 0 I - ISO
Stress (MPa) Stress B - KIN -200 KIN stresses are lower, more C - KIN pronounced than for axial results E - KIN -400 I - KIN E - MIXED -600 iDHD #1 iDHD #2 -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1c Hoop Stress (ISO vs KIN ) g , y 1000
800
B - KIN 600 D - ISO E - ISO 400 E - MIXED E - KIN 200 F - ISO G - ISO 0
Stress (MPa) Stress H - ISO ISO and KIN results for I - ISO -200 a given analysis do not I - KIN -400 bound the dataset iDHD #1 iDHD #2 -600 C - ISO C - KIN -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1c Hoop Stress (ISO vs KIN )
1000
800
B - KIN 600 C - ISO C - KIN 400 D - ISO F - ISO 200 G - ISO H - ISO 0
Stress (MPa) Stress I - ISO I - KIN -200 MIXED data falls between ISO and KIN results E - ISO -400 E - MIXED E - KIN -600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1 Hoop Stress Difference (1c – 1a) 1000
800 Hoop stress shows similar 600 trends to axial stress B - KIN 400 C - ISO C - KIN 200 E - ISO E - MIXED 0 E - KIN Stress (MPa) Stress F - ISO -200 G - ISO I - ISO -400 I - KIN Average -600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1 Hoop Stress Difference (1b – 1a), thermal model
1000
800
600
B - KIN 400 D - ISO 200 E - ISO E - MIXED 0 E - KIN
Stress (MPa) Stress F - ISO -200 G - ISO I - ISO -400 Hoop stress distributions appear I - KIN less sensitive to heat input Average -600 than axial stresses …
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 1 Hoop Stress Difference (1c – 1b), material prop.
1000
800
600 B - KIN 400 C - ISO C - KIN 200 E - ISO E - MIXED 0 E - KIN Stress (MPa) Stress F - ISO -200 G - ISO I - ISO -400 … but more sensitive to changes I - KIN in material properties. Average -600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 2 Axial Stress, post safe -end
1000
800 B - ISO B - KIN 600 C - ISO C - KIN 400 D - ISO E - ISO 200 E - MIXED E - KIN 0 Stress (MPa) Stress F - ISO G - ISO -200 H - ISO I - ISO -400 I - KIN iDHD #1 -600 iDHD #2 Average -800 3x 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 2
Axial Stress Comparison (2 and 1c)
1000
800 Stress reduction at ID observed as a result of safe-end weld 600
400
200
0 2 Average
Stress (MPa) Stress 1c Average -200
-400
-600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 2 Axial Stress Difference (2 – 1c)
1000 Through wall bending
800 reduce stress at ID, the effect is predominately linear B - KIN 600 C - ISO C - KIN 400 E - ISO E - MIXED 200 E - KIN F - ISO 0 G - ISO Stress (MPa) Stress H - ISO -200 I - ISO I - KIN -400 iDHD #1 iDHD #2 -600 Average 3x -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 2 Axial Stress, post safe -end
1000
800 B - ISO B - KIN 600 C - ISO C - KIN 400 D - ISO E - ISO 200 E - MIXED E - KIN 0 F - ISO Stress (MPa) Stress G - ISO -200 H - ISO I - ISO -400 I - KIN
-600 iDHD #2 Average -800 3x 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 2
Hoop Stress Comparison (2 and 1c)
1000 Stress reduction at ID observed as a result of safe-end weld 800
600
400
200
0 2 Average
Stress (MPa) Stress 1c Average -200
-400
-600
-800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Phase 2 Analysis 2 Hoop Stress Difference (2 – 1c)
1000
800 Through wall bending B - KIN 600 C - ISO reduces stress at ID C - KIN 400 E - ISO E - MIXED 200 E - KIN F - ISO 0 G - ISO Stress (MPa) Stress H - ISO -200 I - ISO I - KIN -400 iDHD #1 iDHD #2 -600 Average 3x -800 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Distance from ID (x/t) Industry Guidelines Flaw Evaluation • Residual Stress Analysis Assumptions: Geometry and Materials – Simulate actual weld configuration and fabrication sequence – Include ID weld repair: • If no documented repairs found, 50% through-wall, 360°repair • If documented repair found, demonstrate bounding by the 50%, 360° repair assumption • A less conservative repair assumption may be used, but technically justified – Adjacent safe-end to pipe weld • Safe-end length critical – Benchmarking and Validation • Validation approach suggested, based on K – Operating Temperature and Pressure – Compare with Inspection Results – Sensitivity Studies • informational, not to impose additional constraints • to demonstrate robustness Industry Guidelines Flaw Evaluation
• Analysis technique benchmarked and validated with respect to stress measurements on one or more well-characterized mockups
• Mockup not required for every analysis – once the analytical method is validated on one geometry, can be extended to other geometries and welding conditions – evaluator should justify applicability of mockup to the subject flaw evaluation NRC/EPRI Welding Residual Stress Validation Program - Phase III Details and Findings
6-14-11
Lee Fredette PhD, PE Research Leader Battelle, Systems Development 505 King Ave Columbus, OH 43201 e:[email protected] w: www.battelle.org p: 614.424.4462 f: 614.424.3228
1 Phase III Safety Nozzle • Phase III Safety Nozzle Weld Residual Stress Analysis • Purpose: – Simulate the weld residual stresses in a PWR Pressurizer Safety Nozzle geometry using the geometry and weld pass information provided – Results - 9 FEA results produced by 4 modelers - Battelle - Fredette - DEI - Broussard - EMC2 – Brust/Zhang - NRC - Kerr - 4 measurements produced by 2 methods - Contour Method – Hill Engineering - DHD and iDHD - Veqter – The results of the simulations compared to each other and to measurements as a method to quantify uncertainties in weld residual stress predictions and measurements.
2 Phase III Safety Nozzle • Phase III Safety Nozzle Mock-Up Stress Analysis – Model Geometry – Pipe and Weld Profile – Material Properties – Results - FEA Predictions - Measurements - Comparison – Conclusions
3 Phase III Safety Nozzle • Phase III Safety Nozzle Weld Residual Stress Analysis – Nozzle #2 • Geometry: Provided by EPRI 174mm outer diameter
132mm inner DHD2 (270°) diameter
113mm axial length of 316 stainless steel 24mm assumed width 39mm safe-end of DMW 19mm 36mm total width of Inconel alloy 182 buttering and DMW 12mm assumed width approximately of buttering 117mm axial length of ferritic steel nozzle
Ferritic steel nozzle cladded on the ID with stainless steel 215mm outer diameter
4 Phase III Safety Nozzle • Phase III Safety Nozzle Weld Residual Stress Analysis – Nozzle #3 • Geometry:
• Some SS Pipe length assumptions determined to made in the have no effect model due to on DM weld for this case inconsistencies between measured 711mm mock-up 432mm nozzles Diameter from previous sketch.
Sensitivity study done on nozzle end boundary 210mm 215mm conditions 5 Phase III Safety Nozzle • Phase III Safety Nozzle Weld Residual Stress Analysis • Geometry: – Metallography
A82/182 Weld Low Alloy Steel
SS Safe-End
SS Cladding
6 Phase III Safety Nozzle Weld Bead Size (roughly) • Geometry: Butter = 8mm x 3mm (0.3” x 0.1”) INCO = 3mm x 10mm (0.1” x 0.4”) – Weld Passes (roughly) SS Weld = 3mm x 6mm (0.1” x 0.25”)
8 14 1 15 21 27 13 2 9 12 11 22 28 3 16 10 10 9
23 29 8 17 4 11 7
24 6 5 18 12 5
25 4 6 13 19
3
7 26 2 14 20 1
7 Phase III Safety Nozzle • Geometry: – Axi-symmetric model used in analysis, but revolved model shown here for illustration
Alloy 82 Alloy 82 Butter Weld SS Safe End SA-105 SS Weld
SS Cladding
SS Pipe
8 Phase III Safety Nozzle • Geometry: – Axi-symmetric model used in analysis, but revolved model shown here for illustration SS Pipe
SS Weld Nozzle #3 SS Safe End Configuration
Alloy 82 Weld
Alloy 82 Nozzle #2 Butter Configuration SA-105
9 Phase III Safety Nozzle • Phase III Safety Nozzle Mock-Up Stress Analysis – Model Geometry – Pipe and Weld Profile – Material Properties – Results - FEA Predictions - Measurements - Comparison – Conclusions
10 Phase III Safety Nozzle Typical • Pipe and Weld Profile:
Alloy 82 Alloy 82 Butter Weld SS Safe End SA-105
SS Weld
SS Pipe SS Cladding
Boundary Conditions: •Fixed axially on left end and free on right end •Equivalent convective cooling on both outer and inner diameter surfaces (44 BTU/hr ft2 oF)
11 Phase III Safety Nozzle Typical • Pipe and Weld Profile: Step 1, Butter and Machine
Butter modeled with 30 passes to match weld pass profiles per Alloy 82 photo, and Post Weld Heat Treated. Butter (29 Passes)
Weld Bead Size (roughly) Butter = 8mm x 3mm (0.3” x 0.1”)
12 Phase III Safety Nozzle Typical • Pipe and Weld Profile: Step 2, DM Weld
Alloy 82 DM Weld constructed of with 14 passes modeled to match Weld provided weld pass profiles per photo. (14 Passes) Weld Bead Size (roughly) INCO = 3mm x 10mm (0.1” x 0.4”)
13 Phase III Safety Nozzle Typical • Pipe and Weld Profile: Step 3, SS Safe End Weld
SS Safe End Weld constructed of with 32 passes.
Weld Bead Size (roughly) SS Weld = 3mm x 6mm (0.1” x 0.25”) SS Safe End Weld (32 Passes)
14 Phase III Safety Nozzle • Phase III Safety Nozzle Mock-Up Stress Analysis – Model Geometry – Pipe and Weld Profile – Material Properties – Results - FEA Predictions - Measurements - Comparison – Conclusions
15 Phase III Safety Nozzle • Material Properties: – Each analyst used their own material property database, but NRC provided material properties from the Phase II mock-up program were readily available. - SA-105 carbon steel nozzle - F316 stainless steel safe end - TP316 stainless steel pipe - Alloy 82 weld material for the Butter and DM Weld - TP308 stainless steel for the stainless steel safe end to pipe weld
16 Phase III Safety Nozzle • Phase III Safety Nozzle Mock-Up Stress Analysis – Model Geometry – Pipe and Weld Profile – Material Properties – Results - FEA Predictions - Measurements - Comparison – Conclusions
17 Phase III Safety Nozzle • Results: – Typical model results stress contour plots are shown for reference. – Graphs show stresses through the centerline of the DM weld – Linear interpolation of original data used to compare 50 steps through the thickness for all cases. – Averages and Standard Deviations calculated and plotted – Comparisons made using the average distance of FEA predictions from both the contour method measurements and the DHD/iDHD measurements
18 Phase III Safety Nozzle • Phase III Safety Nozzle Mock-Up Stress Analysis – Model Geometry – Pipe and Weld Profile – Material Properties – Results - FEA Predictions - Measurements - Comparison – Conclusions
19 Phase III Safety Nozzle
• Phase III Safety Nozzle Mock Up Analysis – Isotropic Hardening (MPa) Axial Stress
413
Hoop Stress
-413
Radial Stress Typical
20 Phase III Safety Nozzle Typical • Phase III Safety Nozzle Mock Up Analysis – Isotropic Hardening
21 Phase III Safety Nozzle
• Phase III Safety Nozzle Mock Up Analysis – Isotropic Hardening (MPa) – Axial Stress before and after SS Safe End Weld
Axial Stress Before SS Weld 413
Axial Stress After SS Weld -413
Typical Very little difference due to small pipe thickness and safe end length
22 Phase III Safety Nozzle Typical • Phase III Safety Nozzle Mock Up Analysis – Isotropic Hardening
Very little difference due to small pipe thickness and safe end length
23 Phase III Safety Nozzle
• Phase III Safety Nozzle Mock Up Analysis – Axial Stress
Average 3σ = 188 MPa (27.3 kpsi)
± 3σ Envelope
24 Phase III Safety Nozzle
• Phase III Safety Nozzle Mock Up Analysis – Hoop Stress
Average 3σ = 235 MPa (34.1 kpsi)
± 3σ Envelope
25 Phase III Safety Nozzle • Phase III Safety Nozzle Mock-Up Stress Analysis – Model Geometry – Pipe and Weld Profile – Material Properties – Results - FEA Predictions - Measurements - Comparison – Conclusions
26 Phase III Safety Nozzle
Average 3σ = 243 MPa (35.2 kpsi)
± 3σ Envelope
27 Phase III Safety Nozzle
Average 3σ = 311 MPa (45.1 kpsi)
± 3σ Envelope
28 Phase III Safety Nozzle • Phase III Safety Nozzle Mock-Up Stress Analysis – Model Geometry – Pipe and Weld Profile – Material Properties – Results - FEA Predictions - Measurements - Comparison – Conclusions
29 Phase III Safety Nozzle • Average Distance from Measurement Data
Axial Stress Hoop Stress Compared to Compared to Contour Contour
Axial Stress Hoop Stress Compared to Compared to iDHD iDHD
30 Phase III Safety Nozzle • Average Distance from Axial Measurement Data
Axial Stress Compared to Contour
Kinematic Hardening Matches Best
ISO Avg. = 92 MPa KIN Avg. = 47 MPa Meas. FEA Meas.
31 Phase III Safety Nozzle • Average Distance from Hoop Measurement Data
Hoop Stress Compared to Contour
Kinematic Hardening Matches Best
ISO Avg. = 169 MPa KIN Avg. = 110 MPa Meas. FEA Meas.
32 Phase III Safety Nozzle • Average Distance from Axial Measurement Data
Axial Stress Compared to iDHD
Little Difference Between Kinematic and Isotropic Hardening Match
ISO Avg. = 131 MPa KIN Avg. = 127 MPa
20 MPa (2.9 kpsi) Meas. FEA Meas.
33 Phase III Safety Nozzle • Average Distance from Hoop Measurement Data
Hoop Stress Compared to iDHD
Kinematic Hardening Matches Best
ISO Avg. = 114 MPa KIN Avg. = 79 MPa Meas. FEA Meas.
34 Phase III Safety Nozzle • Phase III Safety Nozzle Mock-Up Stress Analysis – Model Geometry – Pipe and Weld Profile – Material Properties – Results - FEA Predictions - Measurements - Comparison – Conclusions
35 Phase III Safety Nozzle • Conclusions – Predictions and measurements have shown key aspects of FEA prediction methods which affect results - Material hardening law behavior clearly has the most effect - Treatment of the annealing process in the welded material and material in the heat affected zone has a large effect - Weld sequence can have an effect - Safe end design can change the usually beneficial effect of the safe end stainless steel weld. - The cross section weld macrograph appears to provide sufficient constraint to tune thermal models, as reasonable agreement with experimental results is achieved. – The spread of data provided from measurements and predictions will be used to quantify the uncertainly in claimed weld residual stress values. This statistical information will be valuable in probabilistic predictions used in the future.
36 Phase III Safety Nozzle
Questions?
37 Welding Residual Stresses Frédéric Hasnaoui – EDF R&D
2011 June 14th &1 5th summary
Welding Road map
Multiphysics Weld simulation Plasma weld pool Hot cracking
Thermomechanical approach for Weld simulation Photos : EDF© R&D
2011, june 14th & 15th – WRS meeting - Washington Welding Road map
EDF R&D main project
SPAR WELDING PROJECT Study mechanical and metallurgical welding effects (residuals stress, craking,etc) on PWR pressure vessel by testing and simulation. Ageing Management for Nuclear Power Plants 2010 – 2013 : multiphysics analysis (modelling) 2012 – 2015 : thermo-mechanical analysis by heat equivalent source
Base Metal and welding process : TIG Stainless Steels (304L,316L, 347, etc) Nickel-Base Alloys (Inc 600-690) Ferritic Steels
2011, june 14th & 15th – WRS meeting - Washington Weldability Weldability Welding Road map : overview Study Study Arc Weld pool Weld bead Ex electrode- Plasma Shape Shape
Testing equivalent heat T-HAZ source
Modelling Thermo- Solidification méchanical Hot cracking
Mass transfert Metallurgy (diffusion H )
residual stresses Ductility dip Cold cracking cracking
2011, june 14th & 15th – WRS meeting - Washington Welding Road map : aims and tools
Weldability Study Electrode shape-base metal vapor- Weld bead shape
Thermo-méchanical and metallurgy
Arc Plasma Code Aster Metallurgy & Weld Pool Weldability. Code Saturne « solidification » Hot, Cold AND Dip Ductility Code Aster Cracking
Salomé
Residual Stresses
2011, june 14th & 15th – WRS meeting - Washington Multiphysic Weld simulation Multiphysics weld simulation
Road Map
Arc plasma : 2010-2013 PhD
Weld Pool : 2010 -2013 Ph D
Hot Cracking : 2010-2013 Ph D
Cold Cracking : 2011-2014 Phd 2011 working on « diffusion H », « Martensitic transformation » (bilbiography and numerical analysis) and design testing
DDC Cracking: 2011-2012 2011 bibliography EDF R&D 2012 first test : modeling and comparaison between simulation and testing results (STF Test)
2011, june 14th & 15th – WRS meeting - Washington Multiphysics weld simulation : SPAR project
Arc plasma
2010-2013 PhD (with B CHERON, CORIA - ROUEN) First step : modeling the transition zone between the cathode/plasma comparaison between simulation and testing results (impact parameter : electrode shape, W, W+Ce)
Second Step : modeling the transition zone between the anode/plasma comparaison between simulation and testing results (impact parameter : base metal Fe ,Ni, Cr)
2011, june 14th & 15th – WRS meeting - Washington Multiphysics weld simulation : SPAR project
Weld Pool
2010-2013 PhD (with M MEDALE, IUSTI MARSEILLE) First step : modeling thermocapillary forces(heat and mass transfert) comparaison between simulation and testing results in conduction laser welding (impact parameter : % S)
Second Step : modeling thermocapillary forces and electromagnetic forces comparaison between simulation and testing results in TIG welding (impact parameter : % S)
2011, june 14th & 15th – WRS meeting - Washington Multiphysics weld simulation : SPAR project
Hot Cracking
2010-2013 PhD (Y BRECHET-M SUERY SIMAP GRENOBLE)
2011, june 14th & 15th – WRS meeting - Washington 1-JWRI testing
3- TENSIL TEST 2-Modele RDG mesoscopic in mushy zone
4-Code_Aster : Critérion implementation
5-Industrial testing
2011, june 14th & 15th – WRS meeting - Washington Thermomechanical approach Thermomechanical approach for Weld simulation
An old topic for EDF R&D…
2011, june 14th & 15th – WRS meeting - Washington Thermomechanical approach for Weld simulation
…but on wich we still have work
In order to : Improve numerical methodology : Thermal source prediction / determination Reduce time calculation Introduce statsistical approach
© EDF/EADS/PHIMECA
Improve precision on residual stress prediction… …But also on residual deformation prediction To improve Knowledge on material modelisation (Constitutive Low, matallurgical Phase transformation etc.)
2011, june 14th & 15th – WRS meeting - Washington Thermomechanical approach for Weld simulation
…but on wich we still have work
In order to : Improve exeprimental measurement : Residual stress measurement Residual deformation measurement ; Temperature measurement To improve our knowledge on structure behaviour Residual stress ine partition plate Residual stress in BMI Residual stresses in main cooling pump Weld overlay ….
2011, june 14th & 15th – WRS meeting - Washington Other information
Organization and other activities on Residuals stress
Organization Thermomechanical approach really introduce in a new research programm since 2011 Final R&D programm will be defined in October 2011 Budget share between modelisation and simulation development and experimental validation We intend share efforts & developpments with Industrials parteners Laboratories NeT network (european net Work on residual stress) for ex. Other activities / topics Mechanical mitigation Shot peening modelisation Water Jet Peening, Laser Peening … Heat treatment simulation Machining effects Effects of welding parameters on Non Destructive Evaluation
2011, june 14th & 15th – WRS meeting - Washington Thank you for your attention !
Contact : [email protected] +336 66 62 19 60
2011, june 14th & 15th – WRS meeting - Washington Numerical Simulation in Solid Mechanics
EDF – R&D
1 1
SCOPE
2 Numerical simulation
R&D Research in Solid Mechanics e.g. fracture mechanics
Computational Mechanics e.g. contact – friction – dynamics
Analysis methodologies
CODE_ASTER e.g. reinforced concrete Finite element sovler element Finite
Analyses of components e.g. shutdown cooling system
ENGINEERING
3 Topics
• Welding residual stresses • Ductile fracture • Brittle fracture • Crack propagation arrest • 3D crack propagation • Fatigue analyses • Code_Aster summary
4 2
WELDING RESIDUAL STRESSES
5 Main objectives and strategy
Main objectives : residual stress estimation (fatigue, fracture) • defects under (austenitic) coating in the nuclear reactor • bimetallic welded bond (austenitic / ferritic steel)
Strategy • modelling of phase transformation (metallurgy) • nonlinear thermal solution (enthalpy formulation) • advanced viscoplastic constitutive laws • welding process simulation
6 Available functionalities
HEAT FLOW METALLURGY THERMAL LOADING and EXCHANGES SOLID / SOLID PHASE CHANGE NON LINEAR BEHAVIOUR ( HARDNESS ) LIQUID / SOLID STATE CHANGE Austenitic Grain Size WELD POOL MOTION Precipitation - Tempering Surface tension (Marangoni effect) Carbon and Hydrogen Diffusion Electromagnetic effect Base and Deposit Metals dilution MECANICS PHASE CHANGE STRAIN MULTIPHASED BEHAVIOUR TRANSFORMATION PLASTICITY METALLURGICAL HARDENING RECOVERY CREEP and VISCOPLASTICITY WELD POOL BEHAVIOUR
7 Experimental validation : moving heat source
Temperature (upper wall) 370 / 400 width MZ 1370 / 1400 410 / 430 750 / 600
depth MZ 1370 / 1340
Temperature (inner wall) 1030 / 400
Comparison experiment / simulation (η = 0.67)
8 Experimental validation : multi layer welding
Q Geometry heat source w0=19 mm
t d0=9.9 mm e0=12.9 mm
gap
micromètre L=560 mm
gauges
Maquette tubulaire pour soudage TIG
opening displacement 13 layers residual stresses
9 Modelling assumptions
Temperature evolution • material characteristics depend on T° • convection + radiation on the upper wall • transition heat solid – liquid (latent heat)
Mechanics • elasto-visco-plastic constitutive law • nonlinear isotropic hardening • finite strain • constraint effect
10 Experimental validation
Axial stress Gap closing [mm]4 350 3.5 300 EDF 250 FRA 3 exp-moy-init 200 2.5 experiment 150 2 with constraint 100 1.5 Sizz(MPa) without constraint 50 1 0 0 100 200 300 -50 0.5
-100 0
d (mm) 0 1 2 3 4 5 n° de passe
11 In progress
Partnership • EDF / Framatome / CEA / EADS / ESI • french universities (INSA Lyon, …) / Shangaï university • 3 PhD students : robustness, damage, multi-scale
Remaining difficulty • high sensitivity to temperature spatial distribution • not predictive with regard to the heat flow • data fitting is required
12 3
DUCTILE FRACTURE
13 Main objective and strategy
Main objective : the reactor • defects under coating results in small ductile crack propagation • knowledge of the resulting stress field for brittle fracture
Strategy • local approach : ductile constitutive behaviour (the Rousselier law) • issues related to incompressible finite strain, strain localisation, adaptive remeshing • first step : simulation of a notched specimen
14 Reasonable computational objective
« cup – cone »
global response
AE4 specimen
15 Results
plasticity porosity 16 Results
1. localisation 2. branching m µ 300
3. bifurcation 4. ultimate fracture
17 Ductile fracture : a limiting challenge (2D computation)
CPU time • Alpha Server 1 processor ~ 50 hours • Cluster IBM 1 processor ~ 30 hours • Cluster IBM 8 processors 20 hours
Where is time spent ? assembling • MPI parallelization linear solver • linear solver • speed up < 2
element level CPU time with 1 processor 18 4
BRITTLE FRACTURE
1 – Nuclear reactor 2 – Containment building 3 – Nuclear fuel pellets
19 Main objective and strategy
Main objective • compute the probability of brittle fracture of the reactor • taking into account plastic strain (initial ductile phase)
Approaches • Global approach : extension of energy release rate • local approach : the Beremin model (Weibull law) • Access to irradiated material data : the Charpy experiment
20 Global approach for brittle fracture
Assumption • the dissipated energy is proportional to the fractured area
Advantages • Post-treatment compution • Relation with traditional approaches (J – integrals) • No need for an initial crack • Information about the crack propagation (stable or unstable)
Drawbacks • Require a locally very refined mesh • No probabilistic bases
21 Local approach for brittle fracture
Assumption • stress-driven mechanism • weakest link assumption
Advantages • Post-treatment computation • Microscopic physical bases • Natural probabilistic bases
Drawbacks • No relation with traditional approaches • Mesh-dependent as soon as an initial crack exists
22 Validation : warm pre-stress effect (CT specimen)
Fracture Cooling
F=2082 N
Loading
23 Validation : warm pre-stress effect
energy release rate Experimental results
Beremin
24 From global fracture energy to local toughness
Numerical simulation of the Charpy experiment
25 5
CRACK PROPAGATION ARREST
26 Main objective and strategy
Main objective : the reactor • demonstrate that crack initiation does not result in global fracture • the thermal effect : toughness increase + stress containment
Strategy • Dynamic crack propagation with « arrest SIF » • Cohesive zone models (in progress)
27 Cohesive zone models : ingredients
1. A cohesive law 2. An interface finite element σ U4 U3 []u = UB u][ U1 U2
3. A path-following technique (for initial quasi-static analyses)
potential crack path
displacement 28 Application in dynamics
Uimp instability A Force
2 H R Displacement
B L Uimp Force
instability
Time 29 Validation : cracked cylinder
Internal impact load
Initial crack
30 In progress
Dynamics • A better theoretical understanding • Effects of wave propagation
Numerical simulation • 3D and higher order cohesive elements • Choice of time-stepping algorithms (explicit / implicit ?)
Experimental validation • impact loading on a cracked cylinder • thermal shock on a cracked cylinder (representative of a reactor)
31 6
3D CRACK PROPAGATION
32 Objectives and strategy
Objectives • Easy introduction of a 3D crack in a finite element mesh crack opening in steam generator tubes (residual stresses + contact + crack) automatic tool for cracked pipes and elbows
• 3D crack propagation non plane fatigue crack propagation inside a turbine shaft
Strategies • multi-scale – multi-modelling : ARLEQUIN • X-FEM
33 The ARLEQUIN approach : multi-modelling / multi scale
multi-modelling multi-scale shell + pipe elements crack block introduction
34 ARLEQUIN : application to a turbine blade
sound mesh + crack block
35 The extended finite element method
Extension of usual finite elements – jump shape function (crack length) – singular shape function (crack tip)
The crack description is independent of the mesh
36 X-FEM application to a cracked elbow
crack opening
elliptic crack (cross section view)
37 X-FEM : objective
crack in turbine shaft
38 7
FATIGUE ANALYSES
39 Main objective and strategy
Main objective : thermal fatigue (thermal fluctuations) • qualitative understanding of thermal fatigue • predict crack initiation • assert crack propagation kinetics (maintenance programme)
Strategy • A theory to explain experimental evidences • Initiation criteria : e.g. Dang Van criterion • Crack propagation (Paris law + X-FEM or CZM : in progress) • Propabilistic approach in the case of a crack network • Paramount effect of residual stress and pre-hardening
40 The thermal fatigue phenomenon
Main assumption • « thermal fatigue = mechanical fatigue » (Taheri)
Specificities • Memory effect in cyclic strain hardening (stainless steel) ⇒ hardening has a benefic effect on stress cycle loading hardening has a detrimental effect on strain cycle loading
• Cyclic thermal loading is close to strain cycle loading
• Main difference with car and airplane constructors
41 The thermal fatigue phenomenon (2)
Conditions for a crack network to appear • crack propagation must stop (sufficient stress decrease in depth)
• loading has to be biaxial (e.g. thermal loading)
Stability of a crack network • parametric computations • statistical analyses
42 On-going research : main leads
Better description of the structure behaviour • polycristal laws • agregate computations
Energy criteria (for plastic applications) CPU time • computation of stabilized cycle consumming !
Computation of the initial state • Residual stress • Pre-hardening (detrimental effect for strain cycles)
43 8
CODE_ASTER SUMMARY
44 Why a single finite element software ? Why developping it at EDF R&D ?
Bring together developers and end-users • A common language • A shared ambition
Provide an open architecture to encompass various scientific fields • Multi-physics ; static / dynamics ; … • Industrial applications (completeness, shell and beam elements, …)
Preserve R&D advances
Focus all available skills on a single project • 3 MEUR / year / 200 EDF users = 15 000 EUR / year / user
45 The open source experiment (2003 – … )
www.code-aster.org • On-line documentation (12 000 pages) • Download sources (1 200 000 lines + 1700 tests + SALOME)
Four years later • more than 1000 identified steady accesses • Download : 25% universities, 25% companies 50% anonymous • 150 visits / day ; 1850 downloads for the six past months
A success ? • Up-to-date approach (e.g. Linux, Python, …) • Increasing use for pedagogical purpose (schools, universities) • Next step : aggregating developments from an emerging community
46 9
OTHER TOPICS
47 OTHER TOPICS
Waste deposit • coupling temperature – gaz and liquid flows – mechanics
Containment building • see Ghavamian’s lecture
Seismic analyses • dam : interaction soil / structure • powerplant : networks of pipes and vessels (1D modelling with plasticity)
Nuclear fuel pellets • deformation, cracking, … • objective : contact interaction with tube skin
48 Structural mechanics simulation at EDF
Needs and consequences on software policy
Christophe Durand EDF R&D Outline
Engineering challenges induce simulation challenges
A typical study as a guiding thread
Software policy for researches and studies
2 Challenge : operating decades life-time systems
3 … which implies simulation challenges
Physics and numerics for complex phenomenons
knowledge HPC for bigger and more accurate models
performance
High level supervision for multi-disciplinary, optimization, sensitivity, stochastics confidence in results
PLM and qualification confidence in process
4 A typical study
A big, complex, multi-scale and sollicited structure
5 A typical study
Cracks on screw-heads : Why ? When ? How long will it last ?
… with very local, hard to simulate pathologies
6 A typical study : complex scheme
Neutronics
Constitutive law
Thermic / fluids load
Mechanics
behaviour load
7 A typical study : numerous and multi-scale results
Tresca strain in the 1088 screws fillets
8 8 meters A typical study : needs and consequences on simulation assets Physics and numerics
ability to code and qualify a specific constitutive law ( IRRAD3M, elastoplasticity with creep and swelling induced by irradiation)
15.000 contact nodes : « frontier » study which leads to direct improvements in the code algorithms Need of « open » codes HPC
26 M tetraedrons for solid thermics / 7 M dof for mechanics. 75 days of cpu (11 days elapse thanks to //ism) Need of HPC codes High level supervision
Multi-disciplinarity : importance of standardization, inter-operability
Use of best-in-class tools for each physic (and associated competences …) Need of a platform as a company standard
9 EDF’s choice for software policy
In-house development for :
programming and capitalizing innovative researches, dedicated numerical models
continuous process for qualification and quick transfer from upstream researches to engineering studies
complete mastering (tools and competences) : sharing expertise with reactivity, validation and verification in a nuclear context, quality assurance
10 50 years of structural mechanics simulation
11 20 years of Code_Aster development
An all-purpose FEA software for structural mechanics :
Plugged in a user-friendly inter-operable environment : Salome_Meca
12 20 years of Code_Aster development
60 releases each year 1.250 documents freely available (17.000 pages) 15 PHD for the 8 last years 250 new features in the code in 2008 2.300 tests runned for each release
CODE_ASTER as a GPL free software since 2001 : 50.000 downloads, more than 10. 000 hits each week
13 Free software distribution : intents and motivations
Recognition by the use > 50.000 downloads, living community structured by the forum : industry, research, teaching. Multiplication of uses by number and variety. Qualification Benchmarks about performances and abilities. Feedbacks in a bug- tracker. Contributions Modular architecture making constitutive laws and finite elements easy to code and plug Spreading of competences Academic and industrial partners, providers, students and PhD
14 Free software distribution : intents and motivations
But most of all :
An open-source base for cooperations Developments at shared costs. Easier capitalization for Phd works and researches. Software base for industrial and scientific partnerships
15 Thanks for your attention
For more information and download : www.code-aster.org
2006 french award : best open-source project for industry 16 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Westinghouse Method of Residual Weld Stress Calculation and Verification using Mockup Measurements
WAAP-7545
Stephen Marlette
1 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Overview of Presentation ● Background/History of FEA Simulation Development at Westinghouse ● Brief Overview of Westinghouse FEA Method ● Residual Weld Stress FEA Results for Mockup Simulation ● Comparison to Measurements
2 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Structural Weld Overlay (SWOL)
Repair or mitigation for cracking in Alloy 600 dissimilar metal (DM) welds (also stainless welds in BWRs) – Provides structural support assuming a fully cracked nozzle – Drives the residual weld stresses in the DM weld into a more compressive state (slows crack growth) – Typically the nozzle is more inspectable after SWOL
3 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Background/History ● A FEA methodology was developed to support FCG calculations for SWOL mitigation ● At the time the method was developed, there was very limited stress measurement data available for verification of results ● Results evaluated using sensitivity studies – Heat input, material models, weld pass number, etc.
4 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Overview of FEA Method
2D axisymmetric model
ANSYS birth-and-death feature used to bring the weld elements into solution
Weld passes lumped into larger areas to reduce run time
Weld pass geometry approximated with rectangular areas
Stress-strain curves (up to melt temperature) are approximately bilinear
5 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Overview of FEA Method (Cont)
ANSYS kinematic strain hardening (KINH) rule used for SWOL simulation – Generally more conservative for this application
Westinghouse method uses temperature constraints as opposed to heat source modeling – Near melt temperature of 2100oF(1149oC) used
Weld passes are brought into the solution at the near melt temperature and held for a period of time
6 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
FEA Model of Safety Nozzle
Stainless Steel Safe End Alloy 52/152 SWOL
Alloy 82/182 Weld Path (Through center of alloy Alloy 82/182 Butt Weld 82/182 Weld) Alloy 82/182 Build-up Low Alloy Steel Nozzle
7 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
DM Weld Bead Sequence in FEA Simulation
Notes: •Buttering and post-weld heat treatment not simulated •The weld sequence based on original nozzle weld not mockup •Weld capping beads not modeled •Machining processes not modeled
8 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Heat Input Cases
5 heat input cases were run due to uncertainty of welding process – 0.1, 0.5, 1.0, 5.0, and10.0 seconds Hold times chosen to cover a large heat input range and demonstrate FEA sensitivity
9 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
FEA Results before SWOL (Hoop)
DM Weld Hoop Stress before SWOL (Ambient Condtions)
72.0 496 60.0 414 48.0 331 36.0 248 24.0 166 0.1 sec 0.5 sec 12.0 83 1.0 sec 0.0 0 5.0 sec Stress (ksi)
-12.0 -83 Stress (MPa) 10.0 sec -24.0 -165 -36.0 -248 -48.0 -331 -60.0 -413 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% ID of Weld OD of Weld % Through Wall CN-MRCDA-09-46
10 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
FEA Results before SWOL (Axial)
DM Weld Axial Stress before SWOL (Ambient Condtions)
60.0 414 50.0 345 40.0 276 30.0 207 20.0 138 0.1 sec 0.5 sec 10.0 69 1.0 sec 0.0 0 5.0 sec Stress (ksi) -10.0 -69 Stress (MPa) 10.0 sec -20.0 -138 -30.0 -207 -40.0 -276 -50.0 -345 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% ID of Weld OD of Weld % Through Wall CN-MRCDA-09-46
11 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved. Mockup Fabrication and Residual Weld Stress Measurement
Two identical safety nozzle mockups – One without SWOL (Mockup A – Only these results are discussed here) – One with SWOL (Mockup B) – Materials similar to actual DHD and IDHD Measurements were performed by VEQTER Ltd. in Bristol England Measurements taken from 3 locations around the circumference of the DMW (0o, 150o, 210o)
12 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Welded Mockups A and B
Mockup A (without SWOL) Mockup B (with SWOL)
13 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved. Comparison to Measurement Data Before SWOL (Measured Hoop)
14 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved. Comparison to Measurement Data Before SWOL (Measured Axial)
15 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
2010 PVP Papers Based on this Work
PVP2010-25736 “SIMULATION AND MEASUREMENT OF THROUGH-WALL RESIDUAL STRESSES IN A STRUCTURAL WELD OVERLAID PRESSURIZER NOZZLE PVP2010-26023 “THE IMPACT OF KEY SIMULATION VARIABLES ON PREDICTED RESIDUAL STRESSES IN PRESSURISER NOZZLE DISSIMILAR METAL WELD MOCK-UPS. PART 1 – SIMULATION” PVP2010-26025 “THE IMPACT OF KEY SIMULATION VARIABLES ON PREDICTED RESIDUAL STRESSES IN PRESSURISER NOZZLE DISSIMILAR METAL WELD MOCK-UPS. PART 2 – COMPARISON OF SIMULATION AND MEASUREMENTS
16 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Westinghouse FEA Results Compared to Others
The results of 3 vendor simulations in PVP2010-26025 – Engineering Mechanics Corporation of Columbus (EMC2) – British Energy with Australian Nuclear Science and Technology Organization (ANSTO) – Westinghouse Electric Company LLC DM Weld Simulation Only
17 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Compared Results of Vendors (Hoop)
18 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Compared Results of Vendors (Axial)
19 Westinghouse Non-Proprietary Class 3 © 2011 Westinghouse Electric Company LLC. All Rights Reserved.
Questions ?
20