Friction Stir Welding and Processing XII: Modeling & Validation
Sponsored by: TMS Materials Processing and Manufacturing Division, TMS: Shaping and Forming Committee
Program Organizers: Yuri Hovanski, Brigham Young University; Yutaka Sato, Tohoku University; Piyush Upadhyay, Pacific Northwest National Laboratory; Anton Naumov, Peter The Great St. Petersburg Polytechnic University; Nilesh Kumar, University of Alabama, Tuscaloosa

Tuesday 8:00 AM
March 21, 2023
Room: 29A
Location: SDCC

Session Chair: Ayoub Soulami, Pacific Northwest National Laboratory; Dwight Burford, University of North Texas


8:00 AM  Invited
Smoothed Particle Hydrodynamics Model for Friction Stir Processing of 316 L Stainless Steel: Process Modeling and Microstructure Evolution Analysis: Ayoub Soulami1; Lei Li1; Neil Henson1; Erin Barker1; Eric Smith1; 1Pacific Northwest National Laboratory
    Friction stir processing (FSP) is a solid-phase processing technique that provides localized modification and control of microstructures in the processed zones. Numerical models can help predict material deformation and temperature history during FSP that directly relate to microstructural refinement, densification, and homogeneity of the processed zone. This work presents a meshfree smoothed particle hydrodynamics (SPH) model for FSP of 316 L stainless steel using a thermo-elasto-plastic constitutive model and stick-slip tool-workpiece contact approach. The model’s predicted material flow, temperature distribution, and stress-strain state are presented and compared with experimental data. The strain rate and temperature histories obtained from the SPH model are used for predicting Zener-Hollomon parameter and average material grain size. Numerical results on the microscale are also found to agree with experimental observations. Based on the numerical results, optimized FSP parameters and tool designs are suggested to achieve desired grain size in the FSP processed zone.

8:20 AM  
Validation of Models for Predicting Bonding Behavior in Friction Stir Welding Processes: Christian Kocak1; Yanfei Gao1; Hyojin Park1; Hahn Choo1; Martin McDonnell2; Zhili Feng3; 1University of Tennessee; 2Ground Vehicle Systems Center; 3Oak Ridge National Laboratory
    Conventional process-structure-property development for friction stir welding has been largely based on trial and error experimentation. It is computationally demanding to conduct parametric studies that govern plastic flow, stress fields, and temperature gradients, sometimes even more laborious than actual laboratory tests. The motivation of this presentation is to develop models that can predict processing windows and produce reliable time dependent thermomechanical maps for a given process, with a particular focus on the extend of solid-state bonding upon the use of this unique processing technique. Using ABAQUS, the coupled Eulerian-Lagrangian method was employed to simulate contact responses in the workpiece form tool interactions. Two models are developed, with one focusing on an analytical relationship between the mechanical response to processing parameters, and with the other on the bonding analysis. It is hypothesized that the closure of interface cavities, governed by either diffusional or creep processes, dictates the extend of bonding.

8:40 AM  Invited
Analysis of Torque Data from Friction Stir Welds in Aluminum Alloys: Kevin Colligan1; 1Concurrent Technologies Corporation
    In the present work, procedures for friction stir welding (FSW) of aluminum alloys were collected from published and unpublished works, permitting analysis within a diverse collection of alloys, material thickness, tool designs and machine parameters. Spindle torque is a key variable in friction stir welding (FSW) since it is directly related to heat generation. The compiled data set permitted analysis of spindle torque from a wide variety of welding conditions. The data was analyzed with the assumption of a Tresca friction model to calculate the contact shear flow stress during steady-state welding. The results give insight into the effects of tool surface velocity, welding speed and initial temper on the average flow stress, with implications for machine control.

9:00 AM  Invited
The Influence of Flow Stress Data and Friction Models on 2D and 3D Simulations of Friction Stir Welding in AA 2219-T76: Kennen Brooks1; Bryan Ramos1; Michael Miles1; Tracy Nelson1; 1Brigham Young University
    The modeling of friction stir welding has long been challenged by a lack of accurate flow stress data over the large range of temperatures and strain rates that are typical of the process. Most often, hot compression or hot torsion tests are used to measure flow stresses, but the strain rates and deformation modes associated with these tests do not emulate the high strain rate shearing that occurs near the friction stir welding tool. An alternate method of measuring flow stresses for the modeling of FSW is employed, using a high-pressure shear approach. The influence of accurate flow stresses on model prediction, along with a study of different friction laws and their influence on predicted welding temperatures, are presented for plunge experiments, as well as for self-reacting friction stir welding of a linear joint. The influence of tool thermal properties on local flow stresses and material flow were also investigated.

9:20 AM  
Temperature Matching of Friction Stir Welding: 3D Simulation: Ryan Melander1; Matthew Goodson1; Michael Miles1; Troy Munro1; 1Brigham Young University
    Numerical modeling of friction stir welding (FSW) is challenging because of the complex interaction between components. Quality model inputs are critical to model accuracy. Two key items are the heat generated due to friction at the tool/workpiece interface and heat transfer boundary conditions. Tuning of the model relies on adjusting friction and/or heat transfer coefficients to match experimental measurements. Prior studies tend to validate models using measurements from either the workpiece or the tool, but not both. Validating model predictions for both ensures that the heat generation levels are correct. The current work validates a FSW plunge model in both the tool and the workpiece. Various levels of heat transfer and friction coefficients were studied to achieve agreement with the experiment. Uncertainty in the thermocouple positions and machine compliance were accounted for. It was found that tool temperatures were more sensitive to the heat transfer coefficient than the workpiece temperatures.

9:40 AM Break

10:00 AM  
A Coupled SPH-FEM Framework to Predict Residual Stresses during Friction Stir Processing: Ayoub Soulami1; Lei Li1; Kranthi Balusu1; Choi Kyoo Sil1; 1Pacific Northwest National Laboratory
    Residual stresses are often generated during processing methods such as Friction Stir Processing (FSP) and can lead to undesirable dimensional instabilities. A coupled smoothed particle hydrodynamics (SPH) – finite element method (FEM) framework is proposed to predict residual stresses during FSP. The meshfree SPH model is first developed to accurately capture the material deformation, temperature evolution, and steady-state heat generation rates in FSP accounting for actual processing conditions and complex tool geometry. Heat generation rates from SPH are then used as an input to a FEM-based residual stress model to predict the temperature evolution and residual stresses in the workpiece. The predicted temperature fields and residual stresses are compared with experimental measurements for several processing conditions. The proposed modeling approach retains the accuracy of the SPH method while significantly reducing the computational time. A parametric study on the impact of processing conditions on the residual stresses is presented and discussed.

10:20 AM  
3D Modeling and Experimental Validation of Linear Friction Welding Process: Srujan Rokkam1; Quang Truong1; Michael Eff2; Don Weaver3; 1Advanced Cooling Technologies, Inc.; 2Edison Welding Institute; 3Air Force Research Laboratory
    In this work, we developed a meshless approach that utilizes a combination of Smoothed Particle Hydrodynamics (SPH) and FEM to obtain a physics-based model capable of capturing the thermo-mechanical behavior of Linear Friction Welding (LFW) process in 3D. The meshless framework is implemented using a commercial FEA package using custom defined application programming interface (API). Subsequently, we employed the developed model to simulate and investigate flash formation and burn-off distance of LFW of Ti-6Al-4V parts. The simulated upset agreed well with FE simulation and experimental data, while the simulation’s weld time of 0.6-0.75 seconds did not match with the experimental trial. This work addresses the gap in modeling LFW using commercial software tools, which are limited to 2D due to large deformation. This work was funded by the U.S. Air Force, Phase II SBIR Contract FA8650-19-C-5050, awarded to ACT, Inc.