Numerical Simulation of TIG Arc Using Fluent Software A Study Note

Literature Overview

This paper, published in the journal Electrician Welder in 2011 (Volume 41, Issue 3, pages 11-14), was authored by Guo Zhaobo, Shi Yu, Huang Jiankang, Kang Xiaoxue, and Fan Ding from Lanzhou University of Technology. The research was supported by the National Natural Science Foundation of China (Grant No. 50805073). The work addresses a fundamental challenge in welding engineering: the quantitative characterization of the free-burning TIG (Tungsten Inert Gas) arc through computational fluid dynamics (CFD) simulation. The authors employed ANSYS Fluent software to establish a two-dimensional steady-state axisymmetric model of the TIG arc, performing UDF (User-Defined Function) secondary development to implement the governing equations and source terms. The resulting simulation produced temperature field, flow field, pressure field, and workpiece surface pressure distribution data, which were validated against published experimental measurements.

Core Technical Content and Governing Equations

The TIG arc is a complex plasma phenomenon governed by coupled electromagnetic, thermodynamic, and fluid dynamic processes. The authors identified and solved the following control equations: the Navier-Stokes momentum equation, the energy equation, the continuity equation, and the electric current continuity equation. For a free-burning TIG arc, the axisymmetric assumption is justified because the cathode (tungsten electrode) and anode (workpiece) are typically aligned along a common axis, and the arc geometry exhibits rotational symmetry under steady-state conditions.

The UDF development in Fluent was critical to this work. Standard CFD solvers do not natively include the electromagnetic coupling terms that characterize arc plasma behavior. The authors developed custom source terms for the Lorentz force (rho*J cross B), the Joule heating term (J^2/sigma), and the arc current density distribution. These source terms were coupled with the fluid dynamic equations to achieve a fully coupled electro-hydrodynamic simulation.

Process Parameters and Simulation Configuration

The simulation adopted a two-dimensional axisymmetric mesh to reduce computational cost while preserving the essential physics of the arc column. The arc was modeled as a free-burning arc, meaning the current flows through the gas between the tungsten cathode tip and the workpiece anode without external electrode confinement.

The boundary conditions were carefully defined: the cathode surface was assigned a constant current density (or constant potential, depending on the arc model used), the anode surface was assigned a fixed potential (ground), and the outer boundary was set as a pressure outlet with appropriate temperature conditions. The authors also considered the effect of gas flow on arc shape and heat transfer, which is a critical practical consideration since the shielding gas flow directly affects the arc geometry and the resulting heat input distribution on the workpiece.

Results and Validation

The simulation yielded three primary outputs: the temperature field within the arc plasma, the velocity field of the gas flow, and the pressure distribution on the workpiece surface. The temperature at the arc axis near the cathode tip reached values exceeding 10,000 K, consistent with the known thermal characteristics of a free-burning TIG arc. The maximum temperature gradually decreased along the arc length toward the anode, with a significant temperature gradient near the cathode due to the high current density and the cathode fall region.

The workpiece surface pressure distribution was particularly noteworthy. The authors found that the arc plasma exerts a non-uniform pressure on the workpiece surface, with the maximum pressure occurring at the arc attachment point. This pressure contributes to weld pool depression and influences weld geometry, particularly in the context of penetration welding where the weld pool surface morphology directly determines the final weld profile.

Validation against published experimental data showed good agreement, with deviations typically within 10-15% for temperature profiles and within 20% for pressure distributions. The authors attributed the remaining discrepancies to simplifications inherent in the model, including the assumption of local thermodynamic equilibrium (LTE) in the arc plasma, the neglect of radiation heat transfer within the arc column, and the use of a simplified arc conductivity model.

Integration with Engineering Practice

From a practical engineering standpoint, this type of numerical simulation serves several important purposes in welding technology development and quality assurance. First, it provides a means to predict weld pool geometry and heat input distribution without conducting extensive physical experiments, which is particularly valuable when welding exotic materials or novel joint configurations where experimental data are scarce. Second, the simulation can be used to optimize welding parameters by systematically varying current, arc length, and gas flow rate in a virtual environment and observing the resulting thermal and fluid dynamic responses.

For pipe and fitting manufacturing, the relevance is significant. TIG welding is extensively used for welding stainless steel pipes, alloy pipes, and corrosion-resistant alloy (CRA) pipes where metallurgical control is critical. The ability to predict arc heat input and weld pool behavior through simulation directly supports the development of welding procedure specifications (WPS) that meet the stringent requirements of standards such as ASME B31.3, API 5L, and EN 10216. In particular, for thin-wall pipe welding where the risk of burn-through and excessive distortion is high, numerical simulation of arc behavior can guide the selection of parameters that achieve full penetration while minimizing the heat-affected zone (HAZ) width.

However, engineers must recognize the limitations of such simulations. The model assumes steady-state conditions, whereas actual welding involves transient phenomena including arc instability, spatter, and dynamic weld pool evolution. Furthermore, the model does not account for material-specific phenomena such as phase transformations in the HAZ, solidification cracking susceptibility, or the effects of alloy segregation on weld metal properties. These factors must be addressed through complementary experimental validation and metallurgical analysis.

Key Questions and Reflections

A critical question arising from this work is the applicability of steady-state models to practical welding scenarios. In production welding, the arc moves along the joint, creating a transient thermal field that governs microstructure evolution and residual stress development. The steady-state assumption is valid for characterizing the arc itself under fixed conditions, but it cannot capture the dynamic coupling between arc motion and weld pool evolution. Future work should explore transient CFD simulations coupled with solidification models to bridge this gap.

Another important consideration is the model's predictive capability for different electrode materials and gas compositions. The simulation was performed for a standard tungsten-copper electrode with pure argon shielding gas. In practice, ceramic-containing tungsten electrodes (such as tungsten-thoria, tungsten-lanthana, or tungsten-cerium) are widely used, and their different cathode emission characteristics significantly alter arc properties. Similarly, the addition of small amounts of hydrogen or helium to the shielding gas mixture changes arc conductivity and heat transfer characteristics. The model framework developed in this paper can be extended to these cases, but it requires appropriate modifications to the arc conductivity and cathode boundary condition formulations.

Study Insights and Implications

This paper represents a significant contribution to the computational modeling of TIG welding processes. The successful implementation of UDF development in Fluent demonstrates that commercial CFD software, when properly customized, can serve as a powerful tool for welding process analysis. The agreement between simulation results and experimental measurements validates the approach and provides confidence in its use for engineering applications.

For practicing engineers, the key takeaway is that numerical simulation should be viewed as a complement to, not a replacement for, experimental welding. The simulation provides mechanistic insight and parameter sensitivity analysis that are difficult to obtain from experiments alone, but the final welding procedure must always be validated through physical trials and non-destructive testing. The integration of computational modeling with experimental verification represents the most effective approach to welding process development and quality assurance in modern manufacturing.