Numerical Analysis of TIG Arc Heat and Mass Transfer Processes

Overview of the Study

This 1998 study by Fan Ding, Chen Jianhong, and Nio O. from Gansu University of Technology and Osaka University presents a comprehensive numerical model for the heat and mass transfer processes within a TIG welding arc. Published in the Journal of Mechanical Engineering (Vol. 34, No. 2, pp. 39-45), the work represents a significant milestone in computational welding arc research. The authors established a complete mathematical model of the TIG arc, incorporating Ohm's law for current density distribution and employing the K-ε turbulence model to handle turbulent flow phenomena. Numerical simulations were conducted using FORTRAN on a SUN workstation, yielding critical physical data including flow field distribution, temperature distribution, and current distribution across a current range of 100 A to 300 A.

Core Technical Framework

The mathematical model integrates several governing equations that describe the coupled electromagnetic, thermal, and fluid dynamic behavior of the TIG arc plasma. The key equations include the continuity equation for mass conservation, the momentum equation incorporating electromagnetic force terms, the energy equation accounting for Joule heating and thermal radiation, and the current continuity equation derived from Ohm's law. The K-ε turbulence model was selected to address the highly turbulent nature of the arc plasma flow, which is essential for accurately predicting convective heat transfer to the workpiece.

The numerical methodology employed finite volume discretization to solve the coupled partial differential equations. The following table summarizes the key modeling parameters and computational settings:

The coupling between the electromagnetic field and the fluid flow field was handled through an iterative approach, where the current density distribution obtained from Ohm's law was used to compute the Lorentz force acting on the plasma, which in turn influenced the flow field. This iterative coupling is critical because the electromagnetic force is the dominant body force driving arc plasma motion, and any decoupling would significantly degrade prediction accuracy.

Key Findings and Physical Insights

The numerical results revealed several important physical characteristics of the TIG arc that have direct implications for welding process design. At lower currents (100 A), the arc exhibits a relatively diffuse thermal profile with moderate plasma velocity, resulting in a broader but shallower heat input zone on the workpiece surface. As the current increases toward 300 A, the arc becomes more constricted, plasma velocities increase substantially, and the thermal energy becomes more concentrated at the arc root on the cathode side.

The temperature distribution within the arc shows a characteristic bell-shaped profile along the arc axis, with peak temperatures occurring near the cathode region. The maximum arc temperature decreases slightly with increasing current due to the increased volume of plasma participating in energy transport. The flow field analysis demonstrated that the plasma jets downward from the cathode toward the anode (workpiece), with the highest velocities occurring at the arc root where the electromagnetic compression effect is strongest.

The current density distribution revealed a non-uniform profile across the electrode cross-section, with higher densities at the electrode edge compared to the center. This non-uniformity is attributed to the constriction effect caused by the Lorentz force and has direct consequences for electrode wear patterns and arc stability. The agreement between numerical predictions and experimental measurements confirmed the validity of the model, particularly for predicting arc pressure and heat flux distribution on the workpiece surface.

Engineering Practice Implications

For steel pipe and pipe fitting welding operations, the numerical predictions from this study provide valuable guidance for process parameter selection. The arc pressure distribution data help in understanding the weld pool geometry and penetration characteristics, which are critical for achieving full penetration in thin-walled pipe joints. The temperature field data enable prediction of the heat-affected zone (HAZ) width and depth, directly impacting mechanical property gradients and residual stress distributions in welded structures.

In the context of pipeline welding, particularly for girth welds on large-diameter pipes, the understanding of arc behavior at different current levels informs decisions about travel speed, electrode diameter selection, and shielding gas flow rates. The study's findings support the practice of using higher currents for thicker pipe sections where greater penetration is required, while lower currents are preferred for thin-wall applications to minimize distortion and burn-through risks.

The numerical methodology established in this work laid the foundation for subsequent computational welding research, including more sophisticated three-dimensional models and coupled thermal-mechanical analyses. For quality control purposes, the predicted temperature distributions can be used to assess the risk of microstructural degradation in the HAZ, particularly for high-strength low-alloy (HSLA) steels commonly used in pipeline construction.

Study Insights and Reflections

After reviewing this foundational work, several observations emerge regarding the state of computational welding research in the late 1990s. The use of a two-dimensional axisymmetric model, while computationally efficient, inherently assumes perfect geometric symmetry that rarely exists in practice due to electrode tilt, workpiece curvature, and external disturbances. Nevertheless, the model's predictive accuracy for fundamental arc physics validates the underlying physics assumptions and provides a reliable baseline for understanding arc behavior.

The selection of the K-ε turbulence model warrants consideration. While widely used at the time, this model has known limitations in predicting separated flows and strong curvature effects, both of which are present in welding arcs. More recent research has explored more advanced turbulence models and even Large Eddy Simulation approaches for arc plasma modeling. However, for the purpose of establishing fundamental understanding and providing engineering-level predictions, the K-ε model proved adequate.

The practical significance of this work extends beyond pure academic interest. The ability to numerically predict arc heat flux distribution enabled more rational welding procedure specification (WPS) development, reducing reliance on extensive trial-and-error testing. For pipe welding applications governed by standards such as ASME B31.3 and API 5L, the quantitative understanding of arc behavior contributes to more reliable qualification testing and production quality assurance programs.

This study represents an important chapter in the evolution of computational welding science. The integration of electromagnetic, fluid dynamic, and thermal analyses into a unified framework demonstrated that numerical methods could provide physically meaningful predictions of welding arc phenomena. For practitioners in the steel pipe and fitting industry, the key takeaway is that arc behavior is fundamentally governed by the interplay of electromagnetic forces, plasma dynamics, and heat transfer, and that these phenomena can be modeled with sufficient accuracy to guide engineering decisions. The work by Fan Ding and colleagues established methodological foundations that continue to influence modern computational welding research and industrial process optimization.