Numerical Simulation and Thermal Modeling of Bobbin Tool Friction Stir Welding: A Technical Guide to Moving Geometry and Material Flow

Evolution of Solid-State Joining: The Rise of Bobbin Tool Friction Stir Welding (BT-FSW)

Friction Stir Welding (FSW), since its inception by TWI in 1991, has revolutionized the joining of high-strength aluminum alloys that were previously considered difficult to weld using traditional fusion techniques. While conventional FSW relies on a single shoulder and a backing plate (anvil) to sustain the high axial forces generated during the process, Bobbin Tool Friction Stir Welding (BT-FSW) introduces a self-reacting tool geometry. This configuration utilizes two shoulders—one on the top surface and one on the bottom—connected by a central pin that penetrates the entire thickness of the workpiece.

The elimination of the backing plate and the reduction of net axial forces on the welding machine make BT-FSW an ideal candidate for complex structures in aerospace, marine, and automotive industries. However, the unique tool design introduces significant complexity in terms of thermal modeling and material flow analysis. To optimize the process and ensure weld integrity, engineers must rely on advanced numerical simulations, specifically moving geometry models and computational fluid dynamics (CFD) to predict heat distribution, tool torque, and potential defect formation.

The Theoretical Framework of Moving Geometry Models

In the context of BT-FSW, a "Moving Geometry Model" is essential for capturing the transient nature of the welding process. Unlike steady-state models that assume a static heat source, a moving geometry model accounts for the physical translation and rotation of the tool relative to the workpiece. This is typically achieved through two primary numerical approaches: Lagrangian and Eulerian frameworks, or a hybrid Arbitrary Lagrangian-Eulerian (ALE) method.

1. Lagrangian Framework

In a pure Lagrangian approach, the mesh moves and deforms with the material. This is highly effective for capturing the boundary changes of the workpiece but fails when the material undergoes the extreme plastic deformation characteristic of the stir zone in FSW. The mesh becomes so distorted that the simulation loses accuracy or crashes.

2. Eulerian Framework

The Eulerian approach keeps the mesh fixed in space while the material flows through it. This is the preferred method for Computational Fluid Dynamics (CFD) models of BT-FSW. It excels at simulating the "liquid-like" flow of the plasticized aluminum around the pin, allowing researchers to evaluate material streamlines and pressure gradients without mesh distortion issues.

3. Arbitrary Lagrangian-Eulerian (ALE) Method

The ALE method provides a compromise, where the mesh is neither fixed to the material nor fixed in space. This allows the tool to move (moving geometry) while maintaining high mesh quality around the tool-workpiece interface. Many researchers using COMSOL Multiphysics or Abaqus employ ALE to simulate the interaction between the revolving bobbin tool and the surrounding material.

Thermal Modeling and Heat Generation Mechanisms

The thermal history of a BT-FSW joint determines the microstructural evolution and the final mechanical properties of the weld. In a bobbin tool configuration, heat is generated at three primary interfaces: the top shoulder, the bottom shoulder, and the pin surface. The total heat input ($Q_{total}$) can be mathematically represented as the sum of frictional heat ($Q_{fric}$) and plastic deformation heat ($Q_{plas}$):

$Q_{total} = Q_{fric} + Q_{plas}$

Frictional Heat Generation

Frictional heat is governed by the contact condition between the tool and the workpiece. The local heat flux ($q$) at a point on the shoulder can be expressed as:

$q = \mu \cdot p \cdot \omega \cdot r$

Where:

• $\mu$: The friction coefficient (often modeled as a function of temperature).
• $p$: The contact pressure (which, in BT-FSW, is balanced between the two shoulders).
• $\omega$: The angular velocity of the tool.
• $r$: The radial distance from the tool axis.

Volumetric Plastic Heat

As the material enters the stir zone, it undergoes intense shear, generating heat volumetrically. This is defined by the plastic work dissipation: $Q_{plas} = \eta \cdot \sigma \cdot \dot{\epsilon}$, where $\eta$ is the Taylor-Quinney coefficient (typically 0.9), $\sigma$ is the flow stress, and $\dot{\epsilon}$ is the plastic strain rate.

Comparison of Conventional FSW vs. Bobbin Tool FSW

To understand the necessity of specialized models for BT-FSW, we must compare its operational parameters with conventional FSW.

Material Flow Around the Bobbin Tool

Material flow in BT-FSW is inherently 3D and highly non-linear. Numerical simulations, such as those performed using CFD in Matlab or COMSOL, have revealed that the material undergoes a complex path around the pin. Unlike conventional FSW where material is often pushed downward by the shoulder tilt, BT-FSW material flow is constrained by the two shoulders.

Pin Shape Effects

Research by Asadi (2022) and others emphasizes that the pin shape is the primary driver of material transport. Common geometries include:

• Cylindrical: Provides basic stirring but may lead to stagnant zones.
• Threaded: Enhances vertical material mixing, reducing the likelihood of "lazy" material at the mid-thickness.
• Conical: Changes the material volume displacement, with outward conical pins moving more material and increasing heat generation.
• Fluted/Tri-fluted: Increases the "pulsing" effect of the stir zone, improving grain refinement.

The Role of Finite Element Analysis (FEA) and CFD in Simulation

Developing a robust BT-FSW model requires the integration of multiple physics. Finite Element Analysis (FEA) is typically used for structural evaluations—calculating tool torque and the stress experienced by the pin. Computational Fluid Dynamics (CFD) is used to model the workpiece as a high-viscosity fluid.

Step-by-Step Numerical Implementation Workflow

1. Geometry Definition: Modeling the tool (pin diameter, shoulder diameter, gap) and the workpiece.
2. Material Property Definition: Assigning temperature-dependent properties (thermal conductivity, specific heat, and flow stress based on the Johnson-Cook or Zener-Hollomon models).
3. Boundary Conditions: Setting the rotational speed, advancing speed, and convective heat loss to the environment. In BT-FSW, heat loss to the ambient air is higher because there is no backing plate to act as a heat sink.
4. Mesh Generation: Using a dense mesh near the stir zone and a coarser mesh toward the boundaries. Boundary layer meshes are critical at the tool-material interface.
5. Solver Configuration: Choosing between a coupled thermal-pseudo-steady-state solver (for speed) or a fully transient solver (for capturing startup and dwell phases).

Force and Torque Evaluation

One of the primary outputs of a numerical simulation is the calculation of tool torque ($M$) and advancing force ($F_x$). High torque indicates high resistance to flow, which can lead to tool breakage. The simulation helps in predicting the "Process Window"—the range of RPM and welding speeds where the torque is manageable and the heat is sufficient for a defect-free joint.

Practical Implementation and Knowledge-Based Process Development

Transitioning from a simulation model to a physical weld requires understanding the relationship between tool geometry and process parameters. The shoulder gap is the most critical parameter in BT-FSW. If the gap is too narrow, the tool will cause excessive flash and material thinning. If the gap is too wide, there will be insufficient forging pressure, leading to internal voids or "tunnel" defects.

Technical Troubleshooting Matrix

Mathematical Modeling of Material Viscosity

In BT-FSW simulations, the aluminum alloy is often modeled as a non-Newtonian fluid. The viscosity ($\mu$) is a function of the flow stress ($\sigma$) and the effective strain rate ($\dot{\epsilon}$):

$\mu = \frac{\sigma}{3\dot{\epsilon}}$

The flow stress is typically calculated using the Zener-Hollomon parameter ($Z$), which correlates strain rate and temperature: $Z = \dot{\epsilon} \exp(\frac{Q}{RT})$ Where $Q$ is the activation energy, $R$ is the gas constant, and $T$ is the absolute temperature. This mathematical relationship is the backbone of moving geometry models in platforms like COMSOL, allowing the software to adjust the material's resistance to flow as the tool moves through it.

Strategic Implications of Advanced Simulation

The use of thermal and material flow models in BT-FSW is not merely an academic exercise; it is a critical component of Knowledge-Based Process Development. By utilizing moving geometry models, manufacturers can:

• Reduce Trial-and-Error: Simulation narrows down the process window before a single piece of metal is cut.
• Tool Design Optimization: Engineers can test the effects of various pin features (flutes, threads) in a virtual environment.
• Scale-Up Confidence: Moving from laboratory-scale 3mm plates to industrial-scale 25mm sections requires accurate scaling of heat input, which only a 3D thermal model can provide.

The integration of moving geometry models and numerical simulations represents the pinnacle of modern welding engineering. By accurately capturing the thermal profiles and material flow patterns unique to bobbin tool friction stir welding, researchers and industrial practitioners can overcome the inherent challenges of this innovative joining technique. As computational power increases, the fidelity of these models—incorporating microstructural evolution and grain size prediction—will continue to advance, further solidifying BT-FSW's role in the future of lightweight, high-performance manufacturing.

The balance of heat generation across the top and bottom shoulders, the precise control of material volume transport around the pin, and the mitigation of stresses within the tool itself are all areas where simulation provides indispensable insights. As the industry moves toward Industry 4.0, these digital twins of the welding process will be essential for real-time monitoring and autonomous quality control in the production of the next generation of transport vehicles and infrastructure.