CLADDING TECHNOLOGY SHANXI CO., LTDUnified Finite Element Model of TIG Welding Arc and Molten Pool
Literature Overview
Published in 2006 by Lu Fenggui, Tang Xinhua, Li Shaoqing, Yao Shun, and Lou Songnian from the Institute of Welding Engineering at Shanghai Jiao Tong University, this study presents a unified finite element (FE) model that simultaneously simulates the TIG welding arc and the molten pool dynamics. This work represents a significant advancement in welding simulation technology, moving beyond the conventional approach of treating the arc and molten pool as separate entities with simplified boundary conditions. By coupling the arc physics (current density distribution, electromagnetic forces, thermal flux) with the molten pool dynamics (fluid flow, heat transfer, solidification), the model provides a more accurate prediction of weld geometry, residual stress, and microstructure evolution.
Core Technical Content and Modeling Approach
The unified FE model integrates three coupled physical phenomena:
- Electromagnetic field: The current density distribution in the arc and the molten pool is governed by Maxwell's equations. The magnetic field generated by the current produces Lorentz forces (J × B) that drive the fluid flow in the molten pool. The model solves for the current density vector J and the magnetic field vector B using the finite element method.
- Thermal field: The heat transfer in the arc and molten pool is governed by the energy equation, which includes conduction, convection, and the volumetric heat source from the arc. The heat flux at the arc-molten pool interface is a critical boundary condition that determines the weld pool geometry and solidification behavior.
- Fluid flow: The fluid dynamics in the molten pool is governed by the Navier-Stokes equations, including the effects of Lorentz forces, surface tension (Marangoni effect), buoyancy, and electromagnetic stirring. The fluid flow significantly influences the heat transfer and the final weld geometry.
The unified model solves these coupled equations simultaneously, allowing for the accurate prediction of the arc current density distribution, the heat flux profile at the weld surface, and the resulting molten pool shape and fluid flow patterns. This is in contrast to conventional models that use simplified heat source models (such as the double-ellipse model or the Gaussian model) that do not account for the complex arc physics.
Key Technical Parameters and Model Validation
The model incorporates the following key parameters and assumptions:
| Parameter | Value/Range | Description |
|---|---|---|
| Arc current | 50–300 A | Typical TIG welding range |
| Arc voltage | 10–20 V | Depends on arc length and current |
| Tungsten electrode diameter | 2.4–4.0 mm | Determines arc spot size |
| Shielding gas | Argon | Affects arc stability and heat transfer |
| Base metal | Mild steel, stainless steel, aluminum | Different thermal properties |
| Mesh size | 0.1–0.5 mm | Near the arc-molten pool interface |
| Time step | 0.001–0.01 s | For transient simulation |
| Convergence criterion | 10^-6 | For coupled solver |
The model is validated against experimental data, including weld geometry (width, depth, profile), fluid flow patterns (measured by tracer techniques or particle image velocimetry), and residual stress distributions (measured by X-ray diffraction or hole-drilling methods). The agreement between the simulated and experimental results demonstrates the accuracy and reliability of the unified model.
Engineering Applications and Implications
The unified FE model has significant implications for welding process design and optimization:
- Process parameter optimization: The model can be used to predict the effect of welding parameters (current, voltage, travel speed, torch angle) on weld geometry, residual stress, and distortion. This allows for the optimization of parameters before actual welding, reducing trial-and-error costs.
- Welding procedure qualification: The model can assist in the qualification of welding procedures by predicting the weld properties and identifying potential defects (such as porosity, lack of fusion, or cracking) under different process conditions.
- Scale-up and automation: For automated welding systems, the model can be used to design welding sequences that minimize distortion and residual stress, particularly for large or complex geometries.
- Microstructure prediction: By coupling the thermal and fluid flow fields with solidification models, the unified FE model can predict the grain structure, phase distribution, and segregation patterns in the weld and HAZ. This is valuable for predicting the mechanical properties and service behavior of the welded joint.
Study Insights and Limitations
The unified FE model represents a significant step forward in welding simulation, but it also has limitations that must be acknowledged. The computational cost is high, requiring significant computational resources and expertise. The model assumes axisymmetric or 2D conditions, which may not accurately capture the complex 3D physics of actual welding. Additionally, the model does not account for all physical phenomena, such as arc instability, spatter, or the interaction between the arc and the shielding gas flow. Despite these limitations, the unified model provides a powerful tool for understanding and optimizing the TIG welding process, and its insights can be applied to improve welding quality and efficiency in engineering practice.