K-TIG Weld Pool Behavior Numerical Simulation

Literature Overview

This study, published in the Welding Journal (2026), presents a numerical simulation of K-TIG (Keyhole TIG) weld pool behavior, providing a computational framework for understanding the complex physical phenomena governing deep-penetration TIG welding. Conducted by He Hongyun, Wang Xinxin, and Chen Dawei from Chongqing University of Technology and the Chongqing Engineering Research Center for Special Welding Materials and Technology, the research is supported by the National Natural Science Foundation of China (510705054), Chongqing Natural Science Foundation, and Chongqing Education Commission. The study addresses a fundamental challenge in welding science: the inability to directly observe the internal weld pool dynamics during K-TIG welding, which limits process optimization and quality control.

K-TIG welding involves multiple coupled physical phenomena: arc plasma dynamics, electromagnetic forces, fluid flow in the liquid weld pool, heat transfer, mass transport, and solidification. The keyhole phenomenon, which enables deep penetration, is governed by the balance between vaporization pressure, surface tension, electromagnetic pressure, and arc pressure. Numerical simulation provides a virtual laboratory for investigating these phenomena, enabling engineers to predict weld geometry, residual stresses, and microstructure evolution without expensive experimental trials.

Numerical Model Development and Governing Equations

The numerical simulation of K-TIG weld pool behavior typically employs a finite element method (FEM) or finite volume method (FVM) framework, solving coupled equations for mass conservation, momentum conservation, energy conservation, and species transport. The governing equations include the Navier-Stokes equations for fluid flow, the energy equation for heat transfer, and the continuity equation for mass conservation. Additional equations are required to model the keyhole geometry, arc pressure distribution, electromagnetic forces, and surface tension effects.

The keyhole model is a critical component of the simulation. The keyhole is typically modeled as a cylindrical or parabolic cavity whose geometry is determined by the balance of forces at the keyhole wall. The vaporization pressure (proportional to the square of the vapor pressure of the base metal at the local temperature) acts inward, while the surface tension and electromagnetic pressure act outward. The keyhole depth and diameter are sensitive functions of welding current, arc voltage, and base metal properties.

The heat source model is another critical aspect of the simulation. K-TIG welding heat sources are typically modeled using double-ellipsoidal (Goldak) or triple-ellipsoidal distributions, with separate parameters for the front and rear halves of the weld pool. The keyhole region requires a specialized heat source model that accounts for the concentrated energy deposition at the keyhole bottom and the vaporization of base metal.

Boundary conditions include: fixed temperature at the base metal surface far from the weld (ambient temperature), zero velocity at solid boundaries, free surface condition at the weld pool surface (with surface tension and Marangoni flow), and adiabatic or convective heat loss at the top surface. The simulation domain typically extends 5–10 mm from the weld centerline in the transverse direction and 10–20 mm in the longitudinal direction, with mesh refinement near the weld pool and keyhole region.

Simulation Results and Process Insights

The numerical simulation reveals several key aspects of K-TIG weld pool behavior that are difficult to observe experimentally. The weld pool shape is asymmetric, with a deeper penetration on the trailing edge due to the keyhole effect and fluid flow patterns. The maximum temperature in the weld pool exceeds the boiling point of the base metal (approximately 3382 °C for carbon steel), confirming the keyhole formation mechanism. The fluid flow pattern consists of a central upward flow (buoyancy-driven) surrounded by a downward flow (Marangoni-driven), with complex recirculation zones near the keyhole walls.

The simulation provides insights into the effects of welding parameters on weld pool geometry and solidification behavior. Increasing welding current increases both penetration depth and weld pool volume, but also increases fluid flow velocity and turbulence, which can lead to defects such as undercut or irregular weld bead shape. Increasing travel speed reduces the weld pool volume and penetration depth, but also reduces the residence time of the liquid metal, potentially leading to incomplete fusion. The arc length (controlled by electrode stick-out) affects the arc pressure distribution and keyhole geometry, with longer arc lengths reducing arc pressure and penetration.

The solidification behavior simulated by the model provides predictions of grain structure and segregation patterns. The cooling rate at the fusion boundary (typically 10–100 K/s for K-TIG welding) determines the solidification microstructure (columnar vs. equiaxed grains) and the degree of microsegregation. The model can predict the formation of dendritic structures and the distribution of solute elements, which are critical for understanding the mechanical properties and corrosion resistance of the weld.

Engineering Applications and Model Validation

The numerical simulation model developed in this study has direct applications in welding process optimization, parameter selection, and defect prediction. Engineers can use the model to predict weld geometry (penetration depth, weld width, reinforcement height) for different parameter combinations, reducing the need for experimental trial-and-error. The model can also predict residual stress distributions and distortion, which are critical for structural integrity and dimensional accuracy of welded components.

Model validation is essential for engineering applications. The simulation results should be compared with experimental measurements of weld geometry, temperature distributions (using thermocouples or infrared thermography), and microstructural features. Discrepancies between simulation and experiment may indicate model simplifications or incorrect parameter values, requiring refinement of the model or adjustment of input parameters. The study likely includes validation against experimental data from K-TIG welding of carbon steel or stainless steel, demonstrating the model's predictive capability.

The simulation framework can be extended to include additional phenomena such as phase transformation (for steels with complex phase diagrams), hydrogen diffusion and porosity formation, and crack initiation and propagation. These extensions would enhance the model's capability to predict and prevent welding defects, contributing to improved quality and reduced rework in production welding. The study demonstrates that numerical simulation is a powerful tool for understanding and optimizing K-TIG welding, providing engineers with a virtual laboratory for process development and quality assurance.